diff options
Diffstat (limited to 'arch/ppc64')
106 files changed, 418 insertions, 27789 deletions
diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig index c658650af429..8cbac7f32092 100644 --- a/arch/ppc64/Kconfig +++ b/arch/ppc64/Kconfig @@ -10,6 +10,9 @@ config MMU bool default y +config PPC_STD_MMU + def_bool y + config UID16 bool @@ -186,6 +189,9 @@ config BOOTX_TEXT Say Y here to see progress messages from the boot firmware in text mode. Requires an Open Firmware compatible video card. +config POWER4 + def_bool y + config POWER4_ONLY bool "Optimize for POWER4" default n @@ -357,7 +363,6 @@ config HOTPLUG_CPU config PROC_DEVICETREE bool "Support for Open Firmware device tree in /proc" - depends on !PPC_ISERIES help This option adds a device-tree directory under /proc which contains an image of the device tree that the kernel copies from Open @@ -461,7 +466,7 @@ config VIOPATH depends on VIOCONS || VIODASD || VIOCD || VIOTAPE || VETH default y -source "arch/ppc64/oprofile/Kconfig" +source "arch/powerpc/oprofile/Kconfig" source "arch/ppc64/Kconfig.debug" diff --git a/arch/ppc64/Makefile b/arch/ppc64/Makefile index 521c2a5a2862..4d18bdb680f0 100644 --- a/arch/ppc64/Makefile +++ b/arch/ppc64/Makefile @@ -82,10 +82,12 @@ CFLAGS += $(call cc-option,-funit-at-a-time) head-y := arch/ppc64/kernel/head.o libs-y += arch/ppc64/lib/ -core-y += arch/ppc64/kernel/ -core-y += arch/ppc64/mm/ +core-y += arch/ppc64/kernel/ arch/powerpc/kernel/ +core-y += arch/powerpc/mm/ +core-y += arch/powerpc/sysdev/ +core-y += arch/powerpc/platforms/ core-$(CONFIG_XMON) += arch/ppc64/xmon/ -drivers-$(CONFIG_OPROFILE) += arch/ppc64/oprofile/ +drivers-$(CONFIG_OPROFILE) += arch/powerpc/oprofile/ boot := arch/ppc64/boot diff --git a/arch/ppc64/kernel/HvCall.c b/arch/ppc64/kernel/HvCall.c deleted file mode 100644 index b772e65b57a2..000000000000 --- a/arch/ppc64/kernel/HvCall.c +++ /dev/null @@ -1,36 +0,0 @@ -/* - * HvCall.c - * Copyright (C) 2001 Mike Corrigan IBM Corporation - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - */ - -#include <asm/page.h> -#include <asm/abs_addr.h> -#include <asm/iSeries/HvCall.h> -#include <asm/iSeries/HvCallSc.h> -#include <asm/iSeries/HvTypes.h> - - -void HvCall_writeLogBuffer(const void *buffer, u64 len) -{ - struct HvLpBufferList hv_buf; - u64 left_this_page; - u64 cur = virt_to_abs(buffer); - - while (len) { - hv_buf.addr = cur; - left_this_page = ((cur & PAGE_MASK) + PAGE_SIZE) - cur; - if (left_this_page > len) - left_this_page = len; - hv_buf.len = left_this_page; - len -= left_this_page; - HvCall2(HvCallBaseWriteLogBuffer, - virt_to_abs(&hv_buf), - left_this_page); - cur = (cur & PAGE_MASK) + PAGE_SIZE; - } -} diff --git a/arch/ppc64/kernel/HvLpConfig.c b/arch/ppc64/kernel/HvLpConfig.c deleted file mode 100644 index cb1d6473203c..000000000000 --- a/arch/ppc64/kernel/HvLpConfig.c +++ /dev/null @@ -1,27 +0,0 @@ -/* - * HvLpConfig.c - * Copyright (C) 2001 Kyle A. Lucke, IBM Corporation - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - */ - -#include <linux/module.h> -#include <asm/iSeries/HvLpConfig.h> - -HvLpIndex HvLpConfig_getLpIndex_outline(void) -{ - return HvLpConfig_getLpIndex(); -} -EXPORT_SYMBOL(HvLpConfig_getLpIndex_outline); diff --git a/arch/ppc64/kernel/HvLpEvent.c b/arch/ppc64/kernel/HvLpEvent.c deleted file mode 100644 index 90032b138902..000000000000 --- a/arch/ppc64/kernel/HvLpEvent.c +++ /dev/null @@ -1,88 +0,0 @@ -/* - * Copyright 2001 Mike Corrigan IBM Corp - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#include <linux/stddef.h> -#include <linux/kernel.h> -#include <linux/module.h> -#include <asm/system.h> -#include <asm/iSeries/HvLpEvent.h> -#include <asm/iSeries/HvCallEvent.h> -#include <asm/iSeries/ItLpNaca.h> - -/* Array of LpEvent handler functions */ -LpEventHandler lpEventHandler[HvLpEvent_Type_NumTypes]; -unsigned lpEventHandlerPaths[HvLpEvent_Type_NumTypes]; - -/* Register a handler for an LpEvent type */ - -int HvLpEvent_registerHandler( HvLpEvent_Type eventType, LpEventHandler handler ) -{ - int rc = 1; - if ( eventType < HvLpEvent_Type_NumTypes ) { - lpEventHandler[eventType] = handler; - rc = 0; - } - return rc; - -} - -int HvLpEvent_unregisterHandler( HvLpEvent_Type eventType ) -{ - int rc = 1; - - might_sleep(); - - if ( eventType < HvLpEvent_Type_NumTypes ) { - if ( !lpEventHandlerPaths[eventType] ) { - lpEventHandler[eventType] = NULL; - rc = 0; - - /* We now sleep until all other CPUs have scheduled. This ensures that - * the deletion is seen by all other CPUs, and that the deleted handler - * isn't still running on another CPU when we return. */ - synchronize_rcu(); - } - } - return rc; -} -EXPORT_SYMBOL(HvLpEvent_registerHandler); -EXPORT_SYMBOL(HvLpEvent_unregisterHandler); - -/* (lpIndex is the partition index of the target partition. - * needed only for VirtualIo, VirtualLan and SessionMgr. Zero - * indicates to use our partition index - for the other types) - */ -int HvLpEvent_openPath( HvLpEvent_Type eventType, HvLpIndex lpIndex ) -{ - int rc = 1; - if ( eventType < HvLpEvent_Type_NumTypes && - lpEventHandler[eventType] ) { - if ( lpIndex == 0 ) - lpIndex = itLpNaca.xLpIndex; - HvCallEvent_openLpEventPath( lpIndex, eventType ); - ++lpEventHandlerPaths[eventType]; - rc = 0; - } - return rc; -} - -int HvLpEvent_closePath( HvLpEvent_Type eventType, HvLpIndex lpIndex ) -{ - int rc = 1; - if ( eventType < HvLpEvent_Type_NumTypes && - lpEventHandler[eventType] && - lpEventHandlerPaths[eventType] ) { - if ( lpIndex == 0 ) - lpIndex = itLpNaca.xLpIndex; - HvCallEvent_closeLpEventPath( lpIndex, eventType ); - --lpEventHandlerPaths[eventType]; - rc = 0; - } - return rc; -} - diff --git a/arch/ppc64/kernel/ItLpQueue.c b/arch/ppc64/kernel/ItLpQueue.c deleted file mode 100644 index 4231861288a3..000000000000 --- a/arch/ppc64/kernel/ItLpQueue.c +++ /dev/null @@ -1,262 +0,0 @@ -/* - * ItLpQueue.c - * Copyright (C) 2001 Mike Corrigan IBM Corporation - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - */ - -#include <linux/stddef.h> -#include <linux/kernel.h> -#include <linux/sched.h> -#include <linux/bootmem.h> -#include <linux/seq_file.h> -#include <linux/proc_fs.h> -#include <asm/system.h> -#include <asm/paca.h> -#include <asm/iSeries/ItLpQueue.h> -#include <asm/iSeries/HvLpEvent.h> -#include <asm/iSeries/HvCallEvent.h> - -/* - * The LpQueue is used to pass event data from the hypervisor to - * the partition. This is where I/O interrupt events are communicated. - * - * It is written to by the hypervisor so cannot end up in the BSS. - */ -struct hvlpevent_queue hvlpevent_queue __attribute__((__section__(".data"))); - -DEFINE_PER_CPU(unsigned long[HvLpEvent_Type_NumTypes], hvlpevent_counts); - -static char *event_types[HvLpEvent_Type_NumTypes] = { - "Hypervisor", - "Machine Facilities", - "Session Manager", - "SPD I/O", - "Virtual Bus", - "PCI I/O", - "RIO I/O", - "Virtual Lan", - "Virtual I/O" -}; - -/* Array of LpEvent handler functions */ -extern LpEventHandler lpEventHandler[HvLpEvent_Type_NumTypes]; - -static struct HvLpEvent * get_next_hvlpevent(void) -{ - struct HvLpEvent * event; - event = (struct HvLpEvent *)hvlpevent_queue.xSlicCurEventPtr; - - if (event->xFlags.xValid) { - /* rmb() needed only for weakly consistent machines (regatta) */ - rmb(); - /* Set pointer to next potential event */ - hvlpevent_queue.xSlicCurEventPtr += ((event->xSizeMinus1 + - LpEventAlign) / LpEventAlign) * LpEventAlign; - - /* Wrap to beginning if no room at end */ - if (hvlpevent_queue.xSlicCurEventPtr > - hvlpevent_queue.xSlicLastValidEventPtr) { - hvlpevent_queue.xSlicCurEventPtr = - hvlpevent_queue.xSlicEventStackPtr; - } - } else { - event = NULL; - } - - return event; -} - -static unsigned long spread_lpevents = NR_CPUS; - -int hvlpevent_is_pending(void) -{ - struct HvLpEvent *next_event; - - if (smp_processor_id() >= spread_lpevents) - return 0; - - next_event = (struct HvLpEvent *)hvlpevent_queue.xSlicCurEventPtr; - - return next_event->xFlags.xValid | - hvlpevent_queue.xPlicOverflowIntPending; -} - -static void hvlpevent_clear_valid(struct HvLpEvent * event) -{ - /* Tell the Hypervisor that we're done with this event. - * Also clear bits within this event that might look like valid bits. - * ie. on 64-byte boundaries. - */ - struct HvLpEvent *tmp; - unsigned extra = ((event->xSizeMinus1 + LpEventAlign) / - LpEventAlign) - 1; - - switch (extra) { - case 3: - tmp = (struct HvLpEvent*)((char*)event + 3 * LpEventAlign); - tmp->xFlags.xValid = 0; - case 2: - tmp = (struct HvLpEvent*)((char*)event + 2 * LpEventAlign); - tmp->xFlags.xValid = 0; - case 1: - tmp = (struct HvLpEvent*)((char*)event + 1 * LpEventAlign); - tmp->xFlags.xValid = 0; - } - - mb(); - - event->xFlags.xValid = 0; -} - -void process_hvlpevents(struct pt_regs *regs) -{ - struct HvLpEvent * event; - - /* If we have recursed, just return */ - if (!spin_trylock(&hvlpevent_queue.lock)) - return; - - for (;;) { - event = get_next_hvlpevent(); - if (event) { - /* Call appropriate handler here, passing - * a pointer to the LpEvent. The handler - * must make a copy of the LpEvent if it - * needs it in a bottom half. (perhaps for - * an ACK) - * - * Handlers are responsible for ACK processing - * - * The Hypervisor guarantees that LpEvents will - * only be delivered with types that we have - * registered for, so no type check is necessary - * here! - */ - if (event->xType < HvLpEvent_Type_NumTypes) - __get_cpu_var(hvlpevent_counts)[event->xType]++; - if (event->xType < HvLpEvent_Type_NumTypes && - lpEventHandler[event->xType]) - lpEventHandler[event->xType](event, regs); - else - printk(KERN_INFO "Unexpected Lp Event type=%d\n", event->xType ); - - hvlpevent_clear_valid(event); - } else if (hvlpevent_queue.xPlicOverflowIntPending) - /* - * No more valid events. If overflow events are - * pending process them - */ - HvCallEvent_getOverflowLpEvents(hvlpevent_queue.xIndex); - else - break; - } - - spin_unlock(&hvlpevent_queue.lock); -} - -static int set_spread_lpevents(char *str) -{ - unsigned long val = simple_strtoul(str, NULL, 0); - - /* - * The parameter is the number of processors to share in processing - * lp events. - */ - if (( val > 0) && (val <= NR_CPUS)) { - spread_lpevents = val; - printk("lpevent processing spread over %ld processors\n", val); - } else { - printk("invalid spread_lpevents %ld\n", val); - } - - return 1; -} -__setup("spread_lpevents=", set_spread_lpevents); - -void setup_hvlpevent_queue(void) -{ - void *eventStack; - - /* - * Allocate a page for the Event Stack. The Hypervisor needs the - * absolute real address, so we subtract out the KERNELBASE and add - * in the absolute real address of the kernel load area. - */ - eventStack = alloc_bootmem_pages(LpEventStackSize); - memset(eventStack, 0, LpEventStackSize); - - /* Invoke the hypervisor to initialize the event stack */ - HvCallEvent_setLpEventStack(0, eventStack, LpEventStackSize); - - hvlpevent_queue.xSlicEventStackPtr = (char *)eventStack; - hvlpevent_queue.xSlicCurEventPtr = (char *)eventStack; - hvlpevent_queue.xSlicLastValidEventPtr = (char *)eventStack + - (LpEventStackSize - LpEventMaxSize); - hvlpevent_queue.xIndex = 0; -} - -static int proc_lpevents_show(struct seq_file *m, void *v) -{ - int cpu, i; - unsigned long sum; - static unsigned long cpu_totals[NR_CPUS]; - - /* FIXME: do we care that there's no locking here? */ - sum = 0; - for_each_online_cpu(cpu) { - cpu_totals[cpu] = 0; - for (i = 0; i < HvLpEvent_Type_NumTypes; i++) { - cpu_totals[cpu] += per_cpu(hvlpevent_counts, cpu)[i]; - } - sum += cpu_totals[cpu]; - } - - seq_printf(m, "LpEventQueue 0\n"); - seq_printf(m, " events processed:\t%lu\n", sum); - - for (i = 0; i < HvLpEvent_Type_NumTypes; ++i) { - sum = 0; - for_each_online_cpu(cpu) { - sum += per_cpu(hvlpevent_counts, cpu)[i]; - } - - seq_printf(m, " %-20s %10lu\n", event_types[i], sum); - } - - seq_printf(m, "\n events processed by processor:\n"); - - for_each_online_cpu(cpu) { - seq_printf(m, " CPU%02d %10lu\n", cpu, cpu_totals[cpu]); - } - - return 0; -} - -static int proc_lpevents_open(struct inode *inode, struct file *file) -{ - return single_open(file, proc_lpevents_show, NULL); -} - -static struct file_operations proc_lpevents_operations = { - .open = proc_lpevents_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -static int __init proc_lpevents_init(void) -{ - struct proc_dir_entry *e; - - e = create_proc_entry("iSeries/lpevents", S_IFREG|S_IRUGO, NULL); - if (e) - e->proc_fops = &proc_lpevents_operations; - - return 0; -} -__initcall(proc_lpevents_init); - diff --git a/arch/ppc64/kernel/LparData.c b/arch/ppc64/kernel/LparData.c deleted file mode 100644 index 0a9c23ca2f0c..000000000000 --- a/arch/ppc64/kernel/LparData.c +++ /dev/null @@ -1,227 +0,0 @@ -/* - * Copyright 2001 Mike Corrigan, IBM Corp - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#include <linux/config.h> -#include <linux/types.h> -#include <linux/threads.h> -#include <linux/module.h> -#include <linux/bitops.h> -#include <asm/processor.h> -#include <asm/ptrace.h> -#include <asm/naca.h> -#include <asm/abs_addr.h> -#include <asm/iSeries/ItLpNaca.h> -#include <asm/lppaca.h> -#include <asm/iSeries/ItLpRegSave.h> -#include <asm/paca.h> -#include <asm/iSeries/HvReleaseData.h> -#include <asm/iSeries/LparMap.h> -#include <asm/iSeries/ItVpdAreas.h> -#include <asm/iSeries/ItIplParmsReal.h> -#include <asm/iSeries/ItExtVpdPanel.h> -#include <asm/iSeries/ItLpQueue.h> -#include <asm/iSeries/IoHriProcessorVpd.h> -#include <asm/iSeries/ItSpCommArea.h> - - -/* The HvReleaseData is the root of the information shared between - * the hypervisor and Linux. - */ -struct HvReleaseData hvReleaseData = { - .xDesc = 0xc8a5d9c4, /* "HvRD" ebcdic */ - .xSize = sizeof(struct HvReleaseData), - .xVpdAreasPtrOffset = offsetof(struct naca_struct, xItVpdAreas), - .xSlicNacaAddr = &naca, /* 64-bit Naca address */ - .xMsNucDataOffset = LPARMAP_PHYS, - .xFlags = HVREL_TAGSINACTIVE /* tags inactive */ - /* 64 bit */ - /* shared processors */ - /* HMT allowed */ - | 6, /* TEMP: This allows non-GA driver */ - .xVrmIndex = 4, /* We are v5r2m0 */ - .xMinSupportedPlicVrmIndex = 3, /* v5r1m0 */ - .xMinCompatablePlicVrmIndex = 3, /* v5r1m0 */ - .xVrmName = { 0xd3, 0x89, 0x95, 0xa4, /* "Linux 2.4.64" ebcdic */ - 0xa7, 0x40, 0xf2, 0x4b, - 0xf4, 0x4b, 0xf6, 0xf4 }, -}; - -/* - * The NACA. The first dword of the naca is required by the iSeries - * hypervisor to point to itVpdAreas. The hypervisor finds the NACA - * through the pointer in hvReleaseData. - */ -struct naca_struct naca = { - .xItVpdAreas = &itVpdAreas, - .xRamDisk = 0, - .xRamDiskSize = 0, -}; - -extern void system_reset_iSeries(void); -extern void machine_check_iSeries(void); -extern void data_access_iSeries(void); -extern void instruction_access_iSeries(void); -extern void hardware_interrupt_iSeries(void); -extern void alignment_iSeries(void); -extern void program_check_iSeries(void); -extern void fp_unavailable_iSeries(void); -extern void decrementer_iSeries(void); -extern void trap_0a_iSeries(void); -extern void trap_0b_iSeries(void); -extern void system_call_iSeries(void); -extern void single_step_iSeries(void); -extern void trap_0e_iSeries(void); -extern void performance_monitor_iSeries(void); -extern void data_access_slb_iSeries(void); -extern void instruction_access_slb_iSeries(void); - -struct ItLpNaca itLpNaca = { - .xDesc = 0xd397d581, /* "LpNa" ebcdic */ - .xSize = 0x0400, /* size of ItLpNaca */ - .xIntHdlrOffset = 0x0300, /* offset to int array */ - .xMaxIntHdlrEntries = 19, /* # ents */ - .xPrimaryLpIndex = 0, /* Part # of primary */ - .xServiceLpIndex = 0, /* Part # of serv */ - .xLpIndex = 0, /* Part # of me */ - .xMaxLpQueues = 0, /* # of LP queues */ - .xLpQueueOffset = 0x100, /* offset of start of LP queues */ - .xPirEnvironMode = 0, /* Piranha stuff */ - .xPirConsoleMode = 0, - .xPirDasdMode = 0, - .xLparInstalled = 0, - .xSysPartitioned = 0, - .xHwSyncedTBs = 0, - .xIntProcUtilHmt = 0, - .xSpVpdFormat = 0, - .xIntProcRatio = 0, - .xPlicVrmIndex = 0, /* VRM index of PLIC */ - .xMinSupportedSlicVrmInd = 0, /* min supported SLIC */ - .xMinCompatableSlicVrmInd = 0, /* min compat SLIC */ - .xLoadAreaAddr = 0, /* 64-bit addr of load area */ - .xLoadAreaChunks = 0, /* chunks for load area */ - .xPaseSysCallCRMask = 0, /* PASE mask */ - .xSlicSegmentTablePtr = 0, /* seg table */ - .xOldLpQueue = { 0 }, /* Old LP Queue */ - .xInterruptHdlr = { - (u64)system_reset_iSeries, /* 0x100 System Reset */ - (u64)machine_check_iSeries, /* 0x200 Machine Check */ - (u64)data_access_iSeries, /* 0x300 Data Access */ - (u64)instruction_access_iSeries, /* 0x400 Instruction Access */ - (u64)hardware_interrupt_iSeries, /* 0x500 External */ - (u64)alignment_iSeries, /* 0x600 Alignment */ - (u64)program_check_iSeries, /* 0x700 Program Check */ - (u64)fp_unavailable_iSeries, /* 0x800 FP Unavailable */ - (u64)decrementer_iSeries, /* 0x900 Decrementer */ - (u64)trap_0a_iSeries, /* 0xa00 Trap 0A */ - (u64)trap_0b_iSeries, /* 0xb00 Trap 0B */ - (u64)system_call_iSeries, /* 0xc00 System Call */ - (u64)single_step_iSeries, /* 0xd00 Single Step */ - (u64)trap_0e_iSeries, /* 0xe00 Trap 0E */ - (u64)performance_monitor_iSeries,/* 0xf00 Performance Monitor */ - 0, /* int 0x1000 */ - 0, /* int 0x1010 */ - 0, /* int 0x1020 CPU ctls */ - (u64)hardware_interrupt_iSeries, /* SC Ret Hdlr */ - (u64)data_access_slb_iSeries, /* 0x380 D-SLB */ - (u64)instruction_access_slb_iSeries /* 0x480 I-SLB */ - } -}; -EXPORT_SYMBOL(itLpNaca); - -/* May be filled in by the hypervisor so cannot end up in the BSS */ -struct ItIplParmsReal xItIplParmsReal __attribute__((__section__(".data"))); - -/* May be filled in by the hypervisor so cannot end up in the BSS */ -struct ItExtVpdPanel xItExtVpdPanel __attribute__((__section__(".data"))); -EXPORT_SYMBOL(xItExtVpdPanel); - -#define maxPhysicalProcessors 32 - -struct IoHriProcessorVpd xIoHriProcessorVpd[maxPhysicalProcessors] = { - { - .xInstCacheOperandSize = 32, - .xDataCacheOperandSize = 32, - .xProcFreq = 50000000, - .xTimeBaseFreq = 50000000, - .xPVR = 0x3600 - } -}; - -/* Space for Main Store Vpd 27,200 bytes */ -/* May be filled in by the hypervisor so cannot end up in the BSS */ -u64 xMsVpd[3400] __attribute__((__section__(".data"))); - -/* Space for Recovery Log Buffer */ -/* May be filled in by the hypervisor so cannot end up in the BSS */ -u64 xRecoveryLogBuffer[32] __attribute__((__section__(".data"))); - -struct SpCommArea xSpCommArea = { - .xDesc = 0xE2D7C3C2, - .xFormat = 1, -}; - -/* The LparMap data is now located at offset 0x6000 in head.S - * It was put there so that the HvReleaseData could address it - * with a 32-bit offset as required by the iSeries hypervisor - * - * The Naca has a pointer to the ItVpdAreas. The hypervisor finds - * the Naca via the HvReleaseData area. The HvReleaseData has the - * offset into the Naca of the pointer to the ItVpdAreas. - */ -struct ItVpdAreas itVpdAreas = { - .xSlicDesc = 0xc9a3e5c1, /* "ItVA" */ - .xSlicSize = sizeof(struct ItVpdAreas), - .xSlicVpdEntries = ItVpdMaxEntries, /* # VPD array entries */ - .xSlicDmaEntries = ItDmaMaxEntries, /* # DMA array entries */ - .xSlicMaxLogicalProcs = NR_CPUS * 2, /* Max logical procs */ - .xSlicMaxPhysicalProcs = maxPhysicalProcessors, /* Max physical procs */ - .xSlicDmaToksOffset = offsetof(struct ItVpdAreas, xPlicDmaToks), - .xSlicVpdAdrsOffset = offsetof(struct ItVpdAreas, xSlicVpdAdrs), - .xSlicDmaLensOffset = offsetof(struct ItVpdAreas, xPlicDmaLens), - .xSlicVpdLensOffset = offsetof(struct ItVpdAreas, xSlicVpdLens), - .xSlicMaxSlotLabels = 0, /* max slot labels */ - .xSlicMaxLpQueues = 1, /* max LP queues */ - .xPlicDmaLens = { 0 }, /* DMA lengths */ - .xPlicDmaToks = { 0 }, /* DMA tokens */ - .xSlicVpdLens = { /* VPD lengths */ - 0,0,0, /* 0 - 2 */ - sizeof(xItExtVpdPanel), /* 3 Extended VPD */ - sizeof(struct paca_struct), /* 4 length of Paca */ - 0, /* 5 */ - sizeof(struct ItIplParmsReal),/* 6 length of IPL parms */ - 26992, /* 7 length of MS VPD */ - 0, /* 8 */ - sizeof(struct ItLpNaca),/* 9 length of LP Naca */ - 0, /* 10 */ - 256, /* 11 length of Recovery Log Buf */ - sizeof(struct SpCommArea), /* 12 length of SP Comm Area */ - 0,0,0, /* 13 - 15 */ - sizeof(struct IoHriProcessorVpd),/* 16 length of Proc Vpd */ - 0,0,0,0,0,0, /* 17 - 22 */ - sizeof(struct hvlpevent_queue), /* 23 length of Lp Queue */ - 0,0 /* 24 - 25 */ - }, - .xSlicVpdAdrs = { /* VPD addresses */ - 0,0,0, /* 0 - 2 */ - &xItExtVpdPanel, /* 3 Extended VPD */ - &paca[0], /* 4 first Paca */ - 0, /* 5 */ - &xItIplParmsReal, /* 6 IPL parms */ - &xMsVpd, /* 7 MS Vpd */ - 0, /* 8 */ - &itLpNaca, /* 9 LpNaca */ - 0, /* 10 */ - &xRecoveryLogBuffer, /* 11 Recovery Log Buffer */ - &xSpCommArea, /* 12 SP Comm Area */ - 0,0,0, /* 13 - 15 */ - &xIoHriProcessorVpd, /* 16 Proc Vpd */ - 0,0,0,0,0,0, /* 17 - 22 */ - &hvlpevent_queue, /* 23 Lp Queue */ - 0,0 - } -}; diff --git a/arch/ppc64/kernel/Makefile b/arch/ppc64/kernel/Makefile index ae60eb1193c6..424dd250cd87 100644 --- a/arch/ppc64/kernel/Makefile +++ b/arch/ppc64/kernel/Makefile @@ -2,36 +2,34 @@ # Makefile for the linux ppc64 kernel. # +ifneq ($(CONFIG_PPC_MERGE),y) + EXTRA_CFLAGS += -mno-minimal-toc extra-y := head.o vmlinux.lds -obj-y := setup.o entry.o traps.o irq.o idle.o dma.o \ - time.o process.o signal.o syscalls.o misc.o ptrace.o \ - align.o semaphore.o bitops.o pacaData.o \ - udbg.o binfmt_elf32.o sys_ppc32.o ioctl32.o \ - ptrace32.o signal32.o rtc.o init_task.o \ - lmb.o cputable.o cpu_setup_power4.o idle_power4.o \ - iommu.o sysfs.o vdso.o pmc.o firmware.o -obj-y += vdso32/ vdso64/ +obj-y := setup.o entry.o misc.o prom.o -obj-$(CONFIG_PPC_OF) += of_device.o +endif + +obj-y += irq.o idle.o dma.o \ + signal.o \ + align.o bitops.o pacaData.o \ + udbg.o ioctl32.o \ + rtc.o \ + cpu_setup_power4.o \ + iommu.o sysfs.o vdso.o firmware.o +obj-y += vdso32/ vdso64/ -pci-obj-$(CONFIG_PPC_ISERIES) += iSeries_pci.o iSeries_irq.o \ - iSeries_VpdInfo.o pci-obj-$(CONFIG_PPC_MULTIPLATFORM) += pci_dn.o pci_direct_iommu.o obj-$(CONFIG_PCI) += pci.o pci_iommu.o iomap.o $(pci-obj-y) -obj-$(CONFIG_PPC_ISERIES) += HvCall.o HvLpConfig.o LparData.o \ - iSeries_setup.o ItLpQueue.o hvCall.o \ - mf.o HvLpEvent.o iSeries_proc.o iSeries_htab.o \ - iSeries_iommu.o - -obj-$(CONFIG_PPC_MULTIPLATFORM) += nvram.o i8259.o prom_init.o prom.o +obj-$(CONFIG_PPC_MULTIPLATFORM) += nvram.o i8259.o +ifneq ($(CONFIG_PPC_MERGE),y) +obj-$(CONFIG_PPC_MULTIPLATFORM) += prom_init.o +endif -obj-$(CONFIG_PPC_PSERIES) += pSeries_pci.o pSeries_lpar.o pSeries_hvCall.o \ - pSeries_nvram.o rtasd.o ras.o pSeries_reconfig.o \ - pSeries_setup.o pSeries_iommu.o udbg_16550.o +obj-$(CONFIG_PPC_PSERIES) += rtasd.o ras.o udbg_16550.o obj-$(CONFIG_PPC_BPA) += bpa_setup.o bpa_iommu.o bpa_nvram.o \ bpa_iic.o spider-pic.o @@ -41,25 +39,24 @@ obj-$(CONFIG_EEH) += eeh.o obj-$(CONFIG_PROC_FS) += proc_ppc64.o obj-$(CONFIG_RTAS_FLASH) += rtas_flash.o obj-$(CONFIG_SMP) += smp.o -obj-$(CONFIG_MODULES) += module.o ppc_ksyms.o +obj-$(CONFIG_MODULES) += module.o +ifneq ($(CONFIG_PPC_MERGE),y) +obj-$(CONFIG_MODULES) += ppc_ksyms.o +endif obj-$(CONFIG_PPC_RTAS) += rtas.o rtas_pci.o obj-$(CONFIG_RTAS_PROC) += rtas-proc.o obj-$(CONFIG_SCANLOG) += scanlog.o -obj-$(CONFIG_VIOPATH) += viopath.o obj-$(CONFIG_LPARCFG) += lparcfg.o obj-$(CONFIG_HVC_CONSOLE) += hvconsole.o +ifneq ($(CONFIG_PPC_MERGE),y) obj-$(CONFIG_BOOTX_TEXT) += btext.o +endif obj-$(CONFIG_HVCS) += hvcserver.o -vio-obj-$(CONFIG_PPC_PSERIES) += pSeries_vio.o -vio-obj-$(CONFIG_PPC_ISERIES) += iSeries_vio.o -obj-$(CONFIG_IBMVIO) += vio.o $(vio-obj-y) +obj-$(CONFIG_IBMVIO) += vio.o obj-$(CONFIG_XICS) += xics.o -obj-$(CONFIG_MPIC) += mpic.o -obj-$(CONFIG_PPC_PMAC) += pmac_setup.o pmac_feature.o pmac_pci.o \ - pmac_time.o pmac_nvram.o pmac_low_i2c.o \ - udbg_scc.o +obj-$(CONFIG_PPC_PMAC) += udbg_scc.o obj-$(CONFIG_PPC_MAPLE) += maple_setup.o maple_pci.o maple_time.o \ udbg_16550.o @@ -67,19 +64,17 @@ obj-$(CONFIG_PPC_MAPLE) += maple_setup.o maple_pci.o maple_time.o \ obj-$(CONFIG_U3_DART) += u3_iommu.o ifdef CONFIG_SMP -obj-$(CONFIG_PPC_PMAC) += pmac_smp.o smp-tbsync.o -obj-$(CONFIG_PPC_ISERIES) += iSeries_smp.o -obj-$(CONFIG_PPC_PSERIES) += pSeries_smp.o -obj-$(CONFIG_PPC_BPA) += pSeries_smp.o +obj-$(CONFIG_PPC_PMAC) += smp-tbsync.o obj-$(CONFIG_PPC_MAPLE) += smp-tbsync.o endif -obj-$(CONFIG_ALTIVEC) += vecemu.o vector.o obj-$(CONFIG_KPROBES) += kprobes.o CFLAGS_ioctl32.o += -Ifs/ +ifneq ($(CONFIG_PPC_MERGE),y) ifeq ($(CONFIG_PPC_ISERIES),y) -arch/ppc64/kernel/head.o: arch/ppc64/kernel/lparmap.s -AFLAGS_head.o += -Iarch/ppc64/kernel +arch/ppc64/kernel/head.o: arch/powerpc/kernel/lparmap.s +AFLAGS_head.o += -Iarch/powerpc/kernel +endif endif diff --git a/arch/ppc64/kernel/asm-offsets.c b/arch/ppc64/kernel/asm-offsets.c index 1ff4fa05a973..5e6046cb414e 100644 --- a/arch/ppc64/kernel/asm-offsets.c +++ b/arch/ppc64/kernel/asm-offsets.c @@ -46,8 +46,6 @@ int main(void) { /* thread struct on stack */ - DEFINE(THREAD_SHIFT, THREAD_SHIFT); - DEFINE(THREAD_SIZE, THREAD_SIZE); DEFINE(TI_FLAGS, offsetof(struct thread_info, flags)); DEFINE(TI_PREEMPT, offsetof(struct thread_info, preempt_count)); DEFINE(TI_SC_NOERR, offsetof(struct thread_info, syscall_noerror)); @@ -77,6 +75,7 @@ int main(void) DEFINE(ICACHEL1LOGLINESIZE, offsetof(struct ppc64_caches, log_iline_size)); DEFINE(ICACHEL1LINESPERPAGE, offsetof(struct ppc64_caches, ilines_per_page)); DEFINE(PLATFORM, offsetof(struct systemcfg, platform)); + DEFINE(PLATFORM_LPAR, PLATFORM_LPAR); /* paca */ DEFINE(PACA_SIZE, sizeof(struct paca_struct)); diff --git a/arch/ppc64/kernel/binfmt_elf32.c b/arch/ppc64/kernel/binfmt_elf32.c deleted file mode 100644 index fadc699a0497..000000000000 --- a/arch/ppc64/kernel/binfmt_elf32.c +++ /dev/null @@ -1,78 +0,0 @@ -/* - * binfmt_elf32.c: Support 32-bit PPC ELF binaries on Power3 and followons. - * based on the SPARC64 version. - * Copyright (C) 1995, 1996, 1997, 1998 David S. Miller (davem@redhat.com) - * Copyright (C) 1995, 1996, 1997, 1998 Jakub Jelinek (jj@ultra.linux.cz) - * - * Copyright (C) 2000,2001 Ken Aaker (kdaaker@rchland.vnet.ibm.com), IBM Corp - * Copyright (C) 2001 Anton Blanchard (anton@au.ibm.com), IBM - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#define ELF_ARCH EM_PPC -#define ELF_CLASS ELFCLASS32 -#define ELF_DATA ELFDATA2MSB; - -#include <asm/processor.h> -#include <linux/module.h> -#include <linux/config.h> -#include <linux/elfcore.h> -#include <linux/compat.h> - -#define elf_prstatus elf_prstatus32 -struct elf_prstatus32 -{ - struct elf_siginfo pr_info; /* Info associated with signal */ - short pr_cursig; /* Current signal */ - unsigned int pr_sigpend; /* Set of pending signals */ - unsigned int pr_sighold; /* Set of held signals */ - pid_t pr_pid; - pid_t pr_ppid; - pid_t pr_pgrp; - pid_t pr_sid; - struct compat_timeval pr_utime; /* User time */ - struct compat_timeval pr_stime; /* System time */ - struct compat_timeval pr_cutime; /* Cumulative user time */ - struct compat_timeval pr_cstime; /* Cumulative system time */ - elf_gregset_t pr_reg; /* General purpose registers. */ - int pr_fpvalid; /* True if math co-processor being used. */ -}; - -#define elf_prpsinfo elf_prpsinfo32 -struct elf_prpsinfo32 -{ - char pr_state; /* numeric process state */ - char pr_sname; /* char for pr_state */ - char pr_zomb; /* zombie */ - char pr_nice; /* nice val */ - unsigned int pr_flag; /* flags */ - u32 pr_uid; - u32 pr_gid; - pid_t pr_pid, pr_ppid, pr_pgrp, pr_sid; - /* Lots missing */ - char pr_fname[16]; /* filename of executable */ - char pr_psargs[ELF_PRARGSZ]; /* initial part of arg list */ -}; - -#include <linux/time.h> - -#undef cputime_to_timeval -#define cputime_to_timeval cputime_to_compat_timeval -static __inline__ void -cputime_to_compat_timeval(const cputime_t cputime, struct compat_timeval *value) -{ - unsigned long jiffies = cputime_to_jiffies(cputime); - value->tv_usec = (jiffies % HZ) * (1000000L / HZ); - value->tv_sec = jiffies / HZ; -} - -extern void start_thread32(struct pt_regs *, unsigned long, unsigned long); -#undef start_thread -#define start_thread start_thread32 -#define init_elf_binfmt init_elf32_binfmt - -#include "../../../fs/binfmt_elf.c" diff --git a/arch/ppc64/kernel/bpa_iommu.c b/arch/ppc64/kernel/bpa_iommu.c index 5f2460090e03..da1b4b7a3269 100644 --- a/arch/ppc64/kernel/bpa_iommu.c +++ b/arch/ppc64/kernel/bpa_iommu.c @@ -39,8 +39,8 @@ #include <asm/pmac_feature.h> #include <asm/abs_addr.h> #include <asm/system.h> +#include <asm/ppc-pci.h> -#include "pci.h" #include "bpa_iommu.h" static inline unsigned long diff --git a/arch/ppc64/kernel/bpa_setup.c b/arch/ppc64/kernel/bpa_setup.c index 57b3db66f458..c2dc8f282eb8 100644 --- a/arch/ppc64/kernel/bpa_setup.c +++ b/arch/ppc64/kernel/bpa_setup.c @@ -43,8 +43,9 @@ #include <asm/time.h> #include <asm/nvram.h> #include <asm/cputable.h> +#include <asm/ppc-pci.h> +#include <asm/irq.h> -#include "pci.h" #include "bpa_iic.h" #include "bpa_iommu.h" @@ -54,7 +55,7 @@ #define DBG(fmt...) #endif -void bpa_get_cpuinfo(struct seq_file *m) +void bpa_show_cpuinfo(struct seq_file *m) { struct device_node *root; const char *model = ""; @@ -128,7 +129,7 @@ struct machdep_calls __initdata bpa_md = { .probe = bpa_probe, .setup_arch = bpa_setup_arch, .init_early = bpa_init_early, - .get_cpuinfo = bpa_get_cpuinfo, + .show_cpuinfo = bpa_show_cpuinfo, .restart = rtas_restart, .power_off = rtas_power_off, .halt = rtas_halt, diff --git a/arch/ppc64/kernel/btext.c b/arch/ppc64/kernel/btext.c index b6fbfbe9032d..506a37885c5c 100644 --- a/arch/ppc64/kernel/btext.c +++ b/arch/ppc64/kernel/btext.c @@ -18,6 +18,7 @@ #include <asm/io.h> #include <asm/lmb.h> #include <asm/processor.h> +#include <asm/udbg.h> #undef NO_SCROLL @@ -131,6 +132,47 @@ int btext_initialize(struct device_node *np) return 0; } +static void btext_putc(unsigned char c) +{ + btext_drawchar(c); +} + +void __init init_boot_display(void) +{ + char *name; + struct device_node *np = NULL; + int rc = -ENODEV; + + printk("trying to initialize btext ...\n"); + + name = (char *)get_property(of_chosen, "linux,stdout-path", NULL); + if (name != NULL) { + np = of_find_node_by_path(name); + if (np != NULL) { + if (strcmp(np->type, "display") != 0) { + printk("boot stdout isn't a display !\n"); + of_node_put(np); + np = NULL; + } + } + } + if (np) + rc = btext_initialize(np); + if (rc) { + for (np = NULL; (np = of_find_node_by_type(np, "display"));) { + if (get_property(np, "linux,opened", NULL)) { + printk("trying %s ...\n", np->full_name); + rc = btext_initialize(np); + printk("result: %d\n", rc); + } + if (rc == 0) + break; + } + } + if (rc == 0 && udbg_putc == NULL) + udbg_putc = btext_putc; +} + /* Calc the base address of a given point (x,y) */ static unsigned char * calc_base(int x, int y) diff --git a/arch/ppc64/kernel/cputable.c b/arch/ppc64/kernel/cputable.c deleted file mode 100644 index 8831a28c3c4e..000000000000 --- a/arch/ppc64/kernel/cputable.c +++ /dev/null @@ -1,308 +0,0 @@ -/* - * arch/ppc64/kernel/cputable.c - * - * Copyright (C) 2001 Ben. Herrenschmidt (benh@kernel.crashing.org) - * - * Modifications for ppc64: - * Copyright (C) 2003 Dave Engebretsen <engebret@us.ibm.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/config.h> -#include <linux/string.h> -#include <linux/sched.h> -#include <linux/threads.h> -#include <linux/init.h> -#include <linux/module.h> - -#include <asm/oprofile_impl.h> -#include <asm/cputable.h> - -struct cpu_spec* cur_cpu_spec = NULL; -EXPORT_SYMBOL(cur_cpu_spec); - -/* NOTE: - * Unlike ppc32, ppc64 will only call this once for the boot CPU, it's - * the responsibility of the appropriate CPU save/restore functions to - * eventually copy these settings over. Those save/restore aren't yet - * part of the cputable though. That has to be fixed for both ppc32 - * and ppc64 - */ -extern void __setup_cpu_power3(unsigned long offset, struct cpu_spec* spec); -extern void __setup_cpu_power4(unsigned long offset, struct cpu_spec* spec); -extern void __setup_cpu_ppc970(unsigned long offset, struct cpu_spec* spec); -extern void __setup_cpu_be(unsigned long offset, struct cpu_spec* spec); - - -/* We only set the altivec features if the kernel was compiled with altivec - * support - */ -#ifdef CONFIG_ALTIVEC -#define CPU_FTR_ALTIVEC_COMP CPU_FTR_ALTIVEC -#define PPC_FEATURE_HAS_ALTIVEC_COMP PPC_FEATURE_HAS_ALTIVEC -#else -#define CPU_FTR_ALTIVEC_COMP 0 -#define PPC_FEATURE_HAS_ALTIVEC_COMP 0 -#endif - -struct cpu_spec cpu_specs[] = { - { /* Power3 */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x00400000, - .cpu_name = "POWER3 (630)", - .cpu_features = CPU_FTR_SPLIT_ID_CACHE | - CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_IABR, - .cpu_user_features = COMMON_USER_PPC64, - .icache_bsize = 128, - .dcache_bsize = 128, - .num_pmcs = 8, - .cpu_setup = __setup_cpu_power3, -#ifdef CONFIG_OPROFILE - .oprofile_cpu_type = "ppc64/power3", - .oprofile_model = &op_model_rs64, -#endif - }, - { /* Power3+ */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x00410000, - .cpu_name = "POWER3 (630+)", - .cpu_features = CPU_FTR_SPLIT_ID_CACHE | - CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_IABR, - .cpu_user_features = COMMON_USER_PPC64, - .icache_bsize = 128, - .dcache_bsize = 128, - .num_pmcs = 8, - .cpu_setup = __setup_cpu_power3, -#ifdef CONFIG_OPROFILE - .oprofile_cpu_type = "ppc64/power3", - .oprofile_model = &op_model_rs64, -#endif - }, - { /* Northstar */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x00330000, - .cpu_name = "RS64-II (northstar)", - .cpu_features = CPU_FTR_SPLIT_ID_CACHE | - CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_IABR | - CPU_FTR_MMCRA | CPU_FTR_CTRL, - .cpu_user_features = COMMON_USER_PPC64, - .icache_bsize = 128, - .dcache_bsize = 128, - .num_pmcs = 8, - .cpu_setup = __setup_cpu_power3, -#ifdef CONFIG_OPROFILE - .oprofile_cpu_type = "ppc64/rs64", - .oprofile_model = &op_model_rs64, -#endif - }, - { /* Pulsar */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x00340000, - .cpu_name = "RS64-III (pulsar)", - .cpu_features = CPU_FTR_SPLIT_ID_CACHE | - CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_IABR | - CPU_FTR_MMCRA | CPU_FTR_CTRL, - .cpu_user_features = COMMON_USER_PPC64, - .icache_bsize = 128, - .dcache_bsize = 128, - .num_pmcs = 8, - .cpu_setup = __setup_cpu_power3, -#ifdef CONFIG_OPROFILE - .oprofile_cpu_type = "ppc64/rs64", - .oprofile_model = &op_model_rs64, -#endif - }, - { /* I-star */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x00360000, - .cpu_name = "RS64-III (icestar)", - .cpu_features = CPU_FTR_SPLIT_ID_CACHE | - CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_IABR | - CPU_FTR_MMCRA | CPU_FTR_CTRL, - .cpu_user_features = COMMON_USER_PPC64, - .icache_bsize = 128, - .dcache_bsize = 128, - .num_pmcs = 8, - .cpu_setup = __setup_cpu_power3, -#ifdef CONFIG_OPROFILE - .oprofile_cpu_type = "ppc64/rs64", - .oprofile_model = &op_model_rs64, -#endif - }, - { /* S-star */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x00370000, - .cpu_name = "RS64-IV (sstar)", - .cpu_features = CPU_FTR_SPLIT_ID_CACHE | - CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_IABR | - CPU_FTR_MMCRA | CPU_FTR_CTRL, - .cpu_user_features = COMMON_USER_PPC64, - .icache_bsize = 128, - .dcache_bsize = 128, - .num_pmcs = 8, - .cpu_setup = __setup_cpu_power3, -#ifdef CONFIG_OPROFILE - .oprofile_cpu_type = "ppc64/rs64", - .oprofile_model = &op_model_rs64, -#endif - }, - { /* Power4 */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x00350000, - .cpu_name = "POWER4 (gp)", - .cpu_features = CPU_FTR_SPLIT_ID_CACHE | - CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | - CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_MMCRA, - .cpu_user_features = COMMON_USER_PPC64, - .icache_bsize = 128, - .dcache_bsize = 128, - .num_pmcs = 8, - .cpu_setup = __setup_cpu_power4, -#ifdef CONFIG_OPROFILE - .oprofile_cpu_type = "ppc64/power4", - .oprofile_model = &op_model_rs64, -#endif - }, - { /* Power4+ */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x00380000, - .cpu_name = "POWER4+ (gq)", - .cpu_features = CPU_FTR_SPLIT_ID_CACHE | - CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | - CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_MMCRA, - .cpu_user_features = COMMON_USER_PPC64, - .icache_bsize = 128, - .dcache_bsize = 128, - .num_pmcs = 8, - .cpu_setup = __setup_cpu_power4, -#ifdef CONFIG_OPROFILE - .oprofile_cpu_type = "ppc64/power4", - .oprofile_model = &op_model_power4, -#endif - }, - { /* PPC970 */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x00390000, - .cpu_name = "PPC970", - .cpu_features = CPU_FTR_SPLIT_ID_CACHE | - CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | - CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_ALTIVEC_COMP | - CPU_FTR_CAN_NAP | CPU_FTR_MMCRA, - .cpu_user_features = COMMON_USER_PPC64 | - PPC_FEATURE_HAS_ALTIVEC_COMP, - .icache_bsize = 128, - .dcache_bsize = 128, - .num_pmcs = 8, - .cpu_setup = __setup_cpu_ppc970, -#ifdef CONFIG_OPROFILE - .oprofile_cpu_type = "ppc64/970", - .oprofile_model = &op_model_power4, -#endif - }, - { /* PPC970FX */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x003c0000, - .cpu_name = "PPC970FX", - .cpu_features = CPU_FTR_SPLIT_ID_CACHE | - CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | - CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_ALTIVEC_COMP | - CPU_FTR_CAN_NAP | CPU_FTR_MMCRA, - .cpu_user_features = COMMON_USER_PPC64 | - PPC_FEATURE_HAS_ALTIVEC_COMP, - .icache_bsize = 128, - .dcache_bsize = 128, - .num_pmcs = 8, - .cpu_setup = __setup_cpu_ppc970, -#ifdef CONFIG_OPROFILE - .oprofile_cpu_type = "ppc64/970", - .oprofile_model = &op_model_power4, -#endif - }, - { /* PPC970MP */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x00440000, - .cpu_name = "PPC970MP", - .cpu_features = CPU_FTR_SPLIT_ID_CACHE | - CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | - CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_ALTIVEC_COMP | - CPU_FTR_CAN_NAP | CPU_FTR_MMCRA, - .cpu_user_features = COMMON_USER_PPC64 | - PPC_FEATURE_HAS_ALTIVEC_COMP, - .icache_bsize = 128, - .dcache_bsize = 128, - .cpu_setup = __setup_cpu_ppc970, -#ifdef CONFIG_OPROFILE - .oprofile_cpu_type = "ppc64/970", - .oprofile_model = &op_model_power4, -#endif - }, - { /* Power5 */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x003a0000, - .cpu_name = "POWER5 (gr)", - .cpu_features = CPU_FTR_SPLIT_ID_CACHE | - CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | - CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_MMCRA | CPU_FTR_SMT | - CPU_FTR_COHERENT_ICACHE | CPU_FTR_LOCKLESS_TLBIE | - CPU_FTR_MMCRA_SIHV, - .cpu_user_features = COMMON_USER_PPC64, - .icache_bsize = 128, - .dcache_bsize = 128, - .num_pmcs = 6, - .cpu_setup = __setup_cpu_power4, -#ifdef CONFIG_OPROFILE - .oprofile_cpu_type = "ppc64/power5", - .oprofile_model = &op_model_power4, -#endif - }, - { /* Power5 */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x003b0000, - .cpu_name = "POWER5 (gs)", - .cpu_features = CPU_FTR_SPLIT_ID_CACHE | - CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | - CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_MMCRA | CPU_FTR_SMT | - CPU_FTR_COHERENT_ICACHE | CPU_FTR_LOCKLESS_TLBIE | - CPU_FTR_MMCRA_SIHV, - .cpu_user_features = COMMON_USER_PPC64, - .icache_bsize = 128, - .dcache_bsize = 128, - .num_pmcs = 6, - .cpu_setup = __setup_cpu_power4, -#ifdef CONFIG_OPROFILE - .oprofile_cpu_type = "ppc64/power5", - .oprofile_model = &op_model_power4, -#endif - }, - { /* BE DD1.x */ - .pvr_mask = 0xffff0000, - .pvr_value = 0x00700000, - .cpu_name = "Broadband Engine", - .cpu_features = CPU_FTR_SPLIT_ID_CACHE | - CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | - CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_ALTIVEC_COMP | - CPU_FTR_SMT, - .cpu_user_features = COMMON_USER_PPC64 | - PPC_FEATURE_HAS_ALTIVEC_COMP, - .icache_bsize = 128, - .dcache_bsize = 128, - .cpu_setup = __setup_cpu_be, - }, - { /* default match */ - .pvr_mask = 0x00000000, - .pvr_value = 0x00000000, - .cpu_name = "POWER4 (compatible)", - .cpu_features = CPU_FTR_SPLIT_ID_CACHE | - CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | - CPU_FTR_PPCAS_ARCH_V2, - .cpu_user_features = COMMON_USER_PPC64, - .icache_bsize = 128, - .dcache_bsize = 128, - .num_pmcs = 6, - .cpu_setup = __setup_cpu_power4, - } -}; diff --git a/arch/ppc64/kernel/eeh.c b/arch/ppc64/kernel/eeh.c index ba93fd731222..035d1b14a207 100644 --- a/arch/ppc64/kernel/eeh.c +++ b/arch/ppc64/kernel/eeh.c @@ -33,7 +33,7 @@ #include <asm/rtas.h> #include <asm/atomic.h> #include <asm/systemcfg.h> -#include "pci.h" +#include <asm/ppc-pci.h> #undef DEBUG diff --git a/arch/ppc64/kernel/entry.S b/arch/ppc64/kernel/entry.S index e8c0bbf4d000..5d2fcbe384c1 100644 --- a/arch/ppc64/kernel/entry.S +++ b/arch/ppc64/kernel/entry.S @@ -191,8 +191,8 @@ syscall_exit_trace_cont: ld r1,GPR1(r1) mtlr r4 mtcr r5 - mtspr SRR0,r7 - mtspr SRR1,r8 + mtspr SPRN_SRR0,r7 + mtspr SPRN_SRR1,r8 rfid b . /* prevent speculative execution */ @@ -265,7 +265,7 @@ _GLOBAL(save_nvgprs) */ _GLOBAL(ppc32_sigsuspend) bl .save_nvgprs - bl .sys32_sigsuspend + bl .compat_sys_sigsuspend b 70f _GLOBAL(ppc64_rt_sigsuspend) @@ -275,7 +275,7 @@ _GLOBAL(ppc64_rt_sigsuspend) _GLOBAL(ppc32_rt_sigsuspend) bl .save_nvgprs - bl .sys32_rt_sigsuspend + bl .compat_sys_rt_sigsuspend 70: cmpdi 0,r3,0 /* If it returned an error, we need to return via syscall_exit to set the SO bit in cr0 and potentially stop for ptrace. */ @@ -310,7 +310,7 @@ _GLOBAL(ppc_clone) _GLOBAL(ppc32_swapcontext) bl .save_nvgprs - bl .sys32_swapcontext + bl .compat_sys_swapcontext b 80f _GLOBAL(ppc64_swapcontext) @@ -319,11 +319,11 @@ _GLOBAL(ppc64_swapcontext) b 80f _GLOBAL(ppc32_sigreturn) - bl .sys32_sigreturn + bl .compat_sys_sigreturn b 80f _GLOBAL(ppc32_rt_sigreturn) - bl .sys32_rt_sigreturn + bl .compat_sys_rt_sigreturn b 80f _GLOBAL(ppc64_rt_sigreturn) @@ -531,7 +531,7 @@ restore: mtctr r3 mtlr r0 ld r3,_XER(r1) - mtspr XER,r3 + mtspr SPRN_XER,r3 REST_8GPRS(5, r1) @@ -543,12 +543,12 @@ restore: mtmsrd r0,1 ld r0,_MSR(r1) - mtspr SRR1,r0 + mtspr SPRN_SRR1,r0 ld r2,_CCR(r1) mtcrf 0xFF,r2 ld r2,_NIP(r1) - mtspr SRR0,r2 + mtspr SPRN_SRR0,r2 ld r0,GPR0(r1) ld r2,GPR2(r1) @@ -643,7 +643,7 @@ _GLOBAL(enter_rtas) std r4,_CCR(r1) mfctr r5 std r5,_CTR(r1) - mfspr r6,XER + mfspr r6,SPRN_XER std r6,_XER(r1) mfdar r7 std r7,_DAR(r1) @@ -697,14 +697,14 @@ _GLOBAL(enter_rtas) ld r5,RTASENTRY(r4) /* get the rtas->entry value */ ld r4,RTASBASE(r4) /* get the rtas->base value */ - mtspr SRR0,r5 - mtspr SRR1,r6 + mtspr SPRN_SRR0,r5 + mtspr SPRN_SRR1,r6 rfid b . /* prevent speculative execution */ _STATIC(rtas_return_loc) /* relocation is off at this point */ - mfspr r4,SPRG3 /* Get PACA */ + mfspr r4,SPRN_SPRG3 /* Get PACA */ SET_REG_TO_CONST(r5, KERNELBASE) sub r4,r4,r5 /* RELOC the PACA base pointer */ @@ -718,8 +718,8 @@ _STATIC(rtas_return_loc) LOADADDR(r3,.rtas_restore_regs) ld r4,PACASAVEDMSR(r4) /* Restore our MSR */ - mtspr SRR0,r3 - mtspr SRR1,r4 + mtspr SPRN_SRR0,r3 + mtspr SPRN_SRR1,r4 rfid b . /* prevent speculative execution */ @@ -730,14 +730,14 @@ _STATIC(rtas_restore_regs) REST_8GPRS(14, r1) /* Restore the non-volatiles */ REST_10GPRS(22, r1) /* ditto */ - mfspr r13,SPRG3 + mfspr r13,SPRN_SPRG3 ld r4,_CCR(r1) mtcr r4 ld r5,_CTR(r1) mtctr r5 ld r6,_XER(r1) - mtspr XER,r6 + mtspr SPRN_XER,r6 ld r7,_DAR(r1) mtdar r7 ld r8,_DSISR(r1) @@ -774,7 +774,7 @@ _GLOBAL(enter_prom) std r4,_CCR(r1) mfctr r5 std r5,_CTR(r1) - mfspr r6,XER + mfspr r6,SPRN_XER std r6,_XER(r1) mfdar r7 std r7,_DAR(r1) @@ -827,7 +827,7 @@ _GLOBAL(enter_prom) ld r5,_CTR(r1) mtctr r5 ld r6,_XER(r1) - mtspr XER,r6 + mtspr SPRN_XER,r6 ld r7,_DAR(r1) mtdar r7 ld r8,_DSISR(r1) diff --git a/arch/ppc64/kernel/head.S b/arch/ppc64/kernel/head.S index 72c61041151a..f58af9c246cb 100644 --- a/arch/ppc64/kernel/head.S +++ b/arch/ppc64/kernel/head.S @@ -36,6 +36,7 @@ #include <asm/setup.h> #include <asm/hvcall.h> #include <asm/iSeries/LparMap.h> +#include <asm/thread_info.h> #ifdef CONFIG_PPC_ISERIES #define DO_SOFT_DISABLE @@ -201,22 +202,22 @@ exception_marker: #define EX_CCR 60 #define EXCEPTION_PROLOG_PSERIES(area, label) \ - mfspr r13,SPRG3; /* get paca address into r13 */ \ + mfspr r13,SPRN_SPRG3; /* get paca address into r13 */ \ std r9,area+EX_R9(r13); /* save r9 - r12 */ \ std r10,area+EX_R10(r13); \ std r11,area+EX_R11(r13); \ std r12,area+EX_R12(r13); \ - mfspr r9,SPRG1; \ + mfspr r9,SPRN_SPRG1; \ std r9,area+EX_R13(r13); \ mfcr r9; \ clrrdi r12,r13,32; /* get high part of &label */ \ mfmsr r10; \ - mfspr r11,SRR0; /* save SRR0 */ \ + mfspr r11,SPRN_SRR0; /* save SRR0 */ \ ori r12,r12,(label)@l; /* virt addr of handler */ \ ori r10,r10,MSR_IR|MSR_DR|MSR_RI; \ - mtspr SRR0,r12; \ - mfspr r12,SRR1; /* and SRR1 */ \ - mtspr SRR1,r10; \ + mtspr SPRN_SRR0,r12; \ + mfspr r12,SPRN_SRR1; /* and SRR1 */ \ + mtspr SPRN_SRR1,r10; \ rfid; \ b . /* prevent speculative execution */ @@ -225,12 +226,12 @@ exception_marker: * This code runs with relocation on. */ #define EXCEPTION_PROLOG_ISERIES_1(area) \ - mfspr r13,SPRG3; /* get paca address into r13 */ \ + mfspr r13,SPRN_SPRG3; /* get paca address into r13 */ \ std r9,area+EX_R9(r13); /* save r9 - r12 */ \ std r10,area+EX_R10(r13); \ std r11,area+EX_R11(r13); \ std r12,area+EX_R12(r13); \ - mfspr r9,SPRG1; \ + mfspr r9,SPRN_SPRG1; \ std r9,area+EX_R13(r13); \ mfcr r9 @@ -283,7 +284,7 @@ exception_marker: std r9,_LINK(r1); \ mfctr r10; /* save CTR in stackframe */ \ std r10,_CTR(r1); \ - mfspr r11,XER; /* save XER in stackframe */ \ + mfspr r11,SPRN_XER; /* save XER in stackframe */ \ std r11,_XER(r1); \ li r9,(n)+1; \ std r9,_TRAP(r1); /* set trap number */ \ @@ -300,7 +301,7 @@ exception_marker: .globl label##_pSeries; \ label##_pSeries: \ HMT_MEDIUM; \ - mtspr SPRG1,r13; /* save r13 */ \ + mtspr SPRN_SPRG1,r13; /* save r13 */ \ RUNLATCH_ON(r13); \ EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, label##_common) @@ -308,7 +309,7 @@ label##_pSeries: \ .globl label##_iSeries; \ label##_iSeries: \ HMT_MEDIUM; \ - mtspr SPRG1,r13; /* save r13 */ \ + mtspr SPRN_SPRG1,r13; /* save r13 */ \ RUNLATCH_ON(r13); \ EXCEPTION_PROLOG_ISERIES_1(area); \ EXCEPTION_PROLOG_ISERIES_2; \ @@ -318,7 +319,7 @@ label##_iSeries: \ .globl label##_iSeries; \ label##_iSeries: \ HMT_MEDIUM; \ - mtspr SPRG1,r13; /* save r13 */ \ + mtspr SPRN_SPRG1,r13; /* save r13 */ \ RUNLATCH_ON(r13); \ EXCEPTION_PROLOG_ISERIES_1(PACA_EXGEN); \ lbz r10,PACAPROCENABLED(r13); \ @@ -388,7 +389,7 @@ __start_interrupts: . = 0x200 _machine_check_pSeries: HMT_MEDIUM - mtspr SPRG1,r13 /* save r13 */ + mtspr SPRN_SPRG1,r13 /* save r13 */ RUNLATCH_ON(r13) EXCEPTION_PROLOG_PSERIES(PACA_EXMC, machine_check_common) @@ -396,18 +397,18 @@ _machine_check_pSeries: .globl data_access_pSeries data_access_pSeries: HMT_MEDIUM - mtspr SPRG1,r13 + mtspr SPRN_SPRG1,r13 BEGIN_FTR_SECTION - mtspr SPRG2,r12 - mfspr r13,DAR - mfspr r12,DSISR + mtspr SPRN_SPRG2,r12 + mfspr r13,SPRN_DAR + mfspr r12,SPRN_DSISR srdi r13,r13,60 rlwimi r13,r12,16,0x20 mfcr r12 cmpwi r13,0x2c beq .do_stab_bolted_pSeries mtcrf 0x80,r12 - mfspr r12,SPRG2 + mfspr r12,SPRN_SPRG2 END_FTR_SECTION_IFCLR(CPU_FTR_SLB) EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, data_access_common) @@ -415,19 +416,19 @@ END_FTR_SECTION_IFCLR(CPU_FTR_SLB) .globl data_access_slb_pSeries data_access_slb_pSeries: HMT_MEDIUM - mtspr SPRG1,r13 + mtspr SPRN_SPRG1,r13 RUNLATCH_ON(r13) - mfspr r13,SPRG3 /* get paca address into r13 */ + mfspr r13,SPRN_SPRG3 /* get paca address into r13 */ std r9,PACA_EXSLB+EX_R9(r13) /* save r9 - r12 */ std r10,PACA_EXSLB+EX_R10(r13) std r11,PACA_EXSLB+EX_R11(r13) std r12,PACA_EXSLB+EX_R12(r13) std r3,PACA_EXSLB+EX_R3(r13) - mfspr r9,SPRG1 + mfspr r9,SPRN_SPRG1 std r9,PACA_EXSLB+EX_R13(r13) mfcr r9 - mfspr r12,SRR1 /* and SRR1 */ - mfspr r3,DAR + mfspr r12,SPRN_SRR1 /* and SRR1 */ + mfspr r3,SPRN_DAR b .do_slb_miss /* Rel. branch works in real mode */ STD_EXCEPTION_PSERIES(0x400, instruction_access) @@ -436,19 +437,19 @@ data_access_slb_pSeries: .globl instruction_access_slb_pSeries instruction_access_slb_pSeries: HMT_MEDIUM - mtspr SPRG1,r13 + mtspr SPRN_SPRG1,r13 RUNLATCH_ON(r13) - mfspr r13,SPRG3 /* get paca address into r13 */ + mfspr r13,SPRN_SPRG3 /* get paca address into r13 */ std r9,PACA_EXSLB+EX_R9(r13) /* save r9 - r12 */ std r10,PACA_EXSLB+EX_R10(r13) std r11,PACA_EXSLB+EX_R11(r13) std r12,PACA_EXSLB+EX_R12(r13) std r3,PACA_EXSLB+EX_R3(r13) - mfspr r9,SPRG1 + mfspr r9,SPRN_SPRG1 std r9,PACA_EXSLB+EX_R13(r13) mfcr r9 - mfspr r12,SRR1 /* and SRR1 */ - mfspr r3,SRR0 /* SRR0 is faulting address */ + mfspr r12,SPRN_SRR1 /* and SRR1 */ + mfspr r3,SPRN_SRR0 /* SRR0 is faulting address */ b .do_slb_miss /* Rel. branch works in real mode */ STD_EXCEPTION_PSERIES(0x500, hardware_interrupt) @@ -466,15 +467,15 @@ system_call_pSeries: RUNLATCH_ON(r9) mr r9,r13 mfmsr r10 - mfspr r13,SPRG3 - mfspr r11,SRR0 + mfspr r13,SPRN_SPRG3 + mfspr r11,SPRN_SRR0 clrrdi r12,r13,32 oris r12,r12,system_call_common@h ori r12,r12,system_call_common@l - mtspr SRR0,r12 + mtspr SPRN_SRR0,r12 ori r10,r10,MSR_IR|MSR_DR|MSR_RI - mfspr r12,SRR1 - mtspr SRR1,r10 + mfspr r12,SPRN_SRR1 + mtspr SPRN_SRR1,r10 rfid b . /* prevent speculative execution */ @@ -504,25 +505,25 @@ system_call_pSeries: .align 7 _GLOBAL(do_stab_bolted_pSeries) mtcrf 0x80,r12 - mfspr r12,SPRG2 + mfspr r12,SPRN_SPRG2 EXCEPTION_PROLOG_PSERIES(PACA_EXSLB, .do_stab_bolted) /* * Vectors for the FWNMI option. Share common code. */ - .globl system_reset_fwnmi + .globl system_reset_fwnmi system_reset_fwnmi: - HMT_MEDIUM - mtspr SPRG1,r13 /* save r13 */ - RUNLATCH_ON(r13) - EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, system_reset_common) + HMT_MEDIUM + mtspr SPRN_SPRG1,r13 /* save r13 */ + RUNLATCH_ON(r13) + EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, system_reset_common) - .globl machine_check_fwnmi + .globl machine_check_fwnmi machine_check_fwnmi: - HMT_MEDIUM - mtspr SPRG1,r13 /* save r13 */ - RUNLATCH_ON(r13) - EXCEPTION_PROLOG_PSERIES(PACA_EXMC, machine_check_common) + HMT_MEDIUM + mtspr SPRN_SPRG1,r13 /* save r13 */ + RUNLATCH_ON(r13) + EXCEPTION_PROLOG_PSERIES(PACA_EXMC, machine_check_common) #ifdef CONFIG_PPC_ISERIES /*** ISeries-LPAR interrupt handlers ***/ @@ -531,18 +532,18 @@ machine_check_fwnmi: .globl data_access_iSeries data_access_iSeries: - mtspr SPRG1,r13 + mtspr SPRN_SPRG1,r13 BEGIN_FTR_SECTION - mtspr SPRG2,r12 - mfspr r13,DAR - mfspr r12,DSISR + mtspr SPRN_SPRG2,r12 + mfspr r13,SPRN_DAR + mfspr r12,SPRN_DSISR srdi r13,r13,60 rlwimi r13,r12,16,0x20 mfcr r12 cmpwi r13,0x2c beq .do_stab_bolted_iSeries mtcrf 0x80,r12 - mfspr r12,SPRG2 + mfspr r12,SPRN_SPRG2 END_FTR_SECTION_IFCLR(CPU_FTR_SLB) EXCEPTION_PROLOG_ISERIES_1(PACA_EXGEN) EXCEPTION_PROLOG_ISERIES_2 @@ -550,25 +551,25 @@ END_FTR_SECTION_IFCLR(CPU_FTR_SLB) .do_stab_bolted_iSeries: mtcrf 0x80,r12 - mfspr r12,SPRG2 + mfspr r12,SPRN_SPRG2 EXCEPTION_PROLOG_ISERIES_1(PACA_EXSLB) EXCEPTION_PROLOG_ISERIES_2 b .do_stab_bolted .globl data_access_slb_iSeries data_access_slb_iSeries: - mtspr SPRG1,r13 /* save r13 */ + mtspr SPRN_SPRG1,r13 /* save r13 */ EXCEPTION_PROLOG_ISERIES_1(PACA_EXSLB) std r3,PACA_EXSLB+EX_R3(r13) ld r12,PACALPPACA+LPPACASRR1(r13) - mfspr r3,DAR + mfspr r3,SPRN_DAR b .do_slb_miss STD_EXCEPTION_ISERIES(0x400, instruction_access, PACA_EXGEN) .globl instruction_access_slb_iSeries instruction_access_slb_iSeries: - mtspr SPRG1,r13 /* save r13 */ + mtspr SPRN_SPRG1,r13 /* save r13 */ EXCEPTION_PROLOG_ISERIES_1(PACA_EXSLB) std r3,PACA_EXSLB+EX_R3(r13) ld r12,PACALPPACA+LPPACASRR1(r13) @@ -586,7 +587,7 @@ instruction_access_slb_iSeries: .globl system_call_iSeries system_call_iSeries: mr r9,r13 - mfspr r13,SPRG3 + mfspr r13,SPRN_SPRG3 EXCEPTION_PROLOG_ISERIES_2 b system_call_common @@ -596,7 +597,7 @@ system_call_iSeries: .globl system_reset_iSeries system_reset_iSeries: - mfspr r13,SPRG3 /* Get paca address */ + mfspr r13,SPRN_SPRG3 /* Get paca address */ mfmsr r24 ori r24,r24,MSR_RI mtmsrd r24 /* RI on */ @@ -639,7 +640,7 @@ iSeries_secondary_smp_loop: #endif /* CONFIG_SMP */ li r0,-1 /* r0=-1 indicates a Hypervisor call */ sc /* Invoke the hypervisor via a system call */ - mfspr r13,SPRG3 /* Put r13 back ???? */ + mfspr r13,SPRN_SPRG3 /* Put r13 back ???? */ b 1b /* If SMP not configured, secondaries * loop forever */ @@ -656,8 +657,8 @@ hardware_interrupt_iSeries_masked: mtcrf 0x80,r9 /* Restore regs */ ld r11,PACALPPACA+LPPACASRR0(r13) ld r12,PACALPPACA+LPPACASRR1(r13) - mtspr SRR0,r11 - mtspr SRR1,r12 + mtspr SPRN_SRR0,r11 + mtspr SPRN_SRR1,r12 ld r9,PACA_EXGEN+EX_R9(r13) ld r10,PACA_EXGEN+EX_R10(r13) ld r11,PACA_EXGEN+EX_R11(r13) @@ -713,8 +714,8 @@ bad_stack: std r10,GPR1(r1) std r11,_NIP(r1) std r12,_MSR(r1) - mfspr r11,DAR - mfspr r12,DSISR + mfspr r11,SPRN_DAR + mfspr r12,SPRN_DSISR std r11,_DAR(r1) std r12,_DSISR(r1) mflr r10 @@ -766,8 +767,8 @@ fast_exception_return: clrrdi r10,r10,2 /* clear RI (LE is 0 already) */ mtmsrd r10,1 - mtspr SRR1,r12 - mtspr SRR0,r11 + mtspr SPRN_SRR1,r12 + mtspr SPRN_SRR0,r11 REST_4GPRS(10, r1) ld r1,GPR1(r1) rfid @@ -788,9 +789,9 @@ unrecov_fer: .globl data_access_common data_access_common: RUNLATCH_ON(r10) /* It wont fit in the 0x300 handler */ - mfspr r10,DAR + mfspr r10,SPRN_DAR std r10,PACA_EXGEN+EX_DAR(r13) - mfspr r10,DSISR + mfspr r10,SPRN_DSISR stw r10,PACA_EXGEN+EX_DSISR(r13) EXCEPTION_PROLOG_COMMON(0x300, PACA_EXGEN) ld r3,PACA_EXGEN+EX_DAR(r13) @@ -821,9 +822,9 @@ hardware_interrupt_entry: .align 7 .globl alignment_common alignment_common: - mfspr r10,DAR + mfspr r10,SPRN_DAR std r10,PACA_EXGEN+EX_DAR(r13) - mfspr r10,DSISR + mfspr r10,SPRN_DSISR stw r10,PACA_EXGEN+EX_DSISR(r13) EXCEPTION_PROLOG_COMMON(0x600, PACA_EXGEN) ld r3,PACA_EXGEN+EX_DAR(r13) @@ -1120,7 +1121,7 @@ _GLOBAL(do_stab_bolted) /* Hash to the primary group */ ld r10,PACASTABVIRT(r13) - mfspr r11,DAR + mfspr r11,SPRN_DAR srdi r11,r11,28 rldimi r10,r11,7,52 /* r10 = first ste of the group */ @@ -1162,7 +1163,7 @@ _GLOBAL(do_stab_bolted) 2: std r9,8(r10) /* Store the vsid part of the ste */ eieio - mfspr r11,DAR /* Get the new esid */ + mfspr r11,SPRN_DAR /* Get the new esid */ clrrdi r11,r11,28 /* Permits a full 32b of ESID */ ori r11,r11,0x90 /* Turn on valid and kp */ std r11,0(r10) /* Put new entry back into the stab */ @@ -1182,8 +1183,8 @@ _GLOBAL(do_stab_bolted) clrrdi r10,r10,2 mtmsrd r10,1 - mtspr SRR0,r11 - mtspr SRR1,r12 + mtspr SPRN_SRR0,r11 + mtspr SPRN_SRR1,r12 ld r9,PACA_EXSLB+EX_R9(r13) ld r10,PACA_EXSLB+EX_R10(r13) ld r11,PACA_EXSLB+EX_R11(r13) @@ -1229,8 +1230,8 @@ _GLOBAL(do_slb_miss) .machine pop #ifdef CONFIG_PPC_ISERIES - mtspr SRR0,r11 - mtspr SRR1,r12 + mtspr SPRN_SRR0,r11 + mtspr SPRN_SRR1,r12 #endif /* CONFIG_PPC_ISERIES */ ld r9,PACA_EXSLB+EX_R9(r13) ld r10,PACA_EXSLB+EX_R10(r13) @@ -1253,7 +1254,7 @@ unrecov_slb: * * On iSeries, the hypervisor must fill in at least one entry before * we get control (with relocate on). The address is give to the hv - * as a page number (see xLparMap in LparData.c), so this must be at a + * as a page number (see xLparMap in lpardata.c), so this must be at a * fixed address (the linker can't compute (u64)&initial_stab >> * PAGE_SHIFT). */ @@ -1316,7 +1317,7 @@ _GLOBAL(pSeries_secondary_smp_init) mr r3,r24 /* not found, copy phys to r3 */ b .kexec_wait /* next kernel might do better */ -2: mtspr SPRG3,r13 /* Save vaddr of paca in SPRG3 */ +2: mtspr SPRN_SPRG3,r13 /* Save vaddr of paca in SPRG3 */ /* From now on, r24 is expected to be logical cpuid */ mr r24,r5 3: HMT_LOW @@ -1364,6 +1365,7 @@ _STATIC(__start_initialization_iSeries) addi r2,r2,0x4000 bl .iSeries_early_setup + bl .early_setup /* relocation is on at this point */ @@ -1554,20 +1556,17 @@ copy_to_here: .section ".text"; .align 2 ; - .globl pmac_secondary_start_1 -pmac_secondary_start_1: - li r24, 1 - b .pmac_secondary_start - - .globl pmac_secondary_start_2 -pmac_secondary_start_2: - li r24, 2 - b .pmac_secondary_start - - .globl pmac_secondary_start_3 -pmac_secondary_start_3: - li r24, 3 - b .pmac_secondary_start + .globl __secondary_start_pmac_0 +__secondary_start_pmac_0: + /* NB the entries for cpus 0, 1, 2 must each occupy 8 bytes. */ + li r24,0 + b 1f + li r24,1 + b 1f + li r24,2 + b 1f + li r24,3 +1: _GLOBAL(pmac_secondary_start) /* turn on 64-bit mode */ @@ -1586,7 +1585,7 @@ _GLOBAL(pmac_secondary_start) LOADADDR(r4, paca) /* Get base vaddr of paca array */ mulli r13,r24,PACA_SIZE /* Calculate vaddr of right paca */ add r13,r13,r4 /* for this processor. */ - mtspr SPRG3,r13 /* Save vaddr of paca in SPRG3 */ + mtspr SPRN_SPRG3,r13 /* Save vaddr of paca in SPRG3 */ /* Create a temp kernel stack for use before relocation is on. */ ld r1,PACAEMERGSP(r13) @@ -1621,7 +1620,7 @@ _GLOBAL(__secondary_start) /* Initialize the page table pointer register. */ LOADADDR(r6,_SDR1) ld r6,0(r6) /* get the value of _SDR1 */ - mtspr SDR1,r6 /* set the htab location */ + mtspr SPRN_SDR1,r6 /* set the htab location */ #endif /* Initialize the first segment table (or SLB) entry */ ld r3,PACASTABVIRT(r13) /* get addr of segment table */ @@ -1650,7 +1649,7 @@ _GLOBAL(__secondary_start) lwz r3,PLATFORM(r3) /* r3 = platform flags */ andi. r3,r3,PLATFORM_LPAR /* Test if bit 0 is set (LPAR bit) */ beq 98f /* branch if result is 0 */ - mfspr r3,PVR + mfspr r3,SPRN_PVR srwi r3,r3,16 cmpwi r3,0x37 /* SStar */ beq 97f @@ -1674,8 +1673,8 @@ _GLOBAL(__secondary_start) #ifdef DO_SOFT_DISABLE ori r4,r4,MSR_EE #endif - mtspr SRR0,r3 - mtspr SRR1,r4 + mtspr SPRN_SRR0,r3 + mtspr SPRN_SRR1,r4 rfid b . /* prevent speculative execution */ @@ -1737,7 +1736,7 @@ _STATIC(start_here_multiplatform) #ifdef CONFIG_HMT /* Start up the second thread on cpu 0 */ - mfspr r3,PVR + mfspr r3,SPRN_PVR srwi r3,r3,16 cmpwi r3,0x34 /* Pulsar */ beq 90f @@ -1797,7 +1796,7 @@ _STATIC(start_here_multiplatform) mulli r13,r27,PACA_SIZE /* Calculate vaddr of right paca */ add r13,r13,r24 /* for this processor. */ sub r13,r13,r26 /* convert to physical addr */ - mtspr SPRG3,r13 /* PPPBBB: Temp... -Peter */ + mtspr SPRN_SPRG3,r13 /* PPPBBB: Temp... -Peter */ /* Do very early kernel initializations, including initial hash table, * stab and slb setup before we turn on relocation. */ @@ -1814,7 +1813,7 @@ _STATIC(start_here_multiplatform) lwz r3,PLATFORM(r3) /* r3 = platform flags */ andi. r3,r3,PLATFORM_LPAR /* Test if bit 0 is set (LPAR bit) */ beq 98f /* branch if result is 0 */ - mfspr r3,PVR + mfspr r3,SPRN_PVR srwi r3,r3,16 cmpwi r3,0x37 /* SStar */ beq 97f @@ -1838,12 +1837,12 @@ _STATIC(start_here_multiplatform) LOADADDR(r6,_SDR1) /* Only if NOT LPAR */ sub r6,r6,r26 ld r6,0(r6) /* get the value of _SDR1 */ - mtspr SDR1,r6 /* set the htab location */ + mtspr SPRN_SDR1,r6 /* set the htab location */ 98: LOADADDR(r3,.start_here_common) SET_REG_TO_CONST(r4, MSR_KERNEL) - mtspr SRR0,r3 - mtspr SRR1,r4 + mtspr SPRN_SRR0,r3 + mtspr SPRN_SRR1,r4 rfid b . /* prevent speculative execution */ #endif /* CONFIG_PPC_MULTIPLATFORM */ @@ -1874,7 +1873,7 @@ _STATIC(start_here_common) LOADADDR(r24, paca) /* Get base vaddr of paca array */ mulli r13,r26,PACA_SIZE /* Calculate vaddr of right paca */ add r13,r13,r24 /* for this processor. */ - mtspr SPRG3,r13 + mtspr SPRN_SPRG3,r13 /* ptr to current */ LOADADDR(r4,init_task) @@ -1901,7 +1900,7 @@ _STATIC(start_here_common) _GLOBAL(hmt_init) #ifdef CONFIG_HMT LOADADDR(r5, hmt_thread_data) - mfspr r7,PVR + mfspr r7,SPRN_PVR srwi r7,r7,16 cmpwi r7,0x34 /* Pulsar */ beq 90f @@ -1910,10 +1909,10 @@ _GLOBAL(hmt_init) cmpwi r7,0x37 /* SStar */ beq 91f b 101f -90: mfspr r6,PIR +90: mfspr r6,SPRN_PIR andi. r6,r6,0x1f b 92f -91: mfspr r6,PIR +91: mfspr r6,SPRN_PIR andi. r6,r6,0x3ff 92: sldi r4,r24,3 stwx r6,r5,r4 @@ -1924,8 +1923,8 @@ __hmt_secondary_hold: LOADADDR(r5, hmt_thread_data) clrldi r5,r5,4 li r7,0 - mfspr r6,PIR - mfspr r8,PVR + mfspr r6,SPRN_PIR + mfspr r8,SPRN_PVR srwi r8,r8,16 cmpwi r8,0x34 bne 93f @@ -1951,39 +1950,41 @@ __hmt_secondary_hold: _GLOBAL(hmt_start_secondary) LOADADDR(r4,__hmt_secondary_hold) clrldi r4,r4,4 - mtspr NIADORM, r4 - mfspr r4, MSRDORM + mtspr SPRN_NIADORM, r4 + mfspr r4, SPRN_MSRDORM li r5, -65 and r4, r4, r5 - mtspr MSRDORM, r4 + mtspr SPRN_MSRDORM, r4 lis r4,0xffef ori r4,r4,0x7403 - mtspr TSC, r4 + mtspr SPRN_TSC, r4 li r4,0x1f4 - mtspr TST, r4 - mfspr r4, HID0 + mtspr SPRN_TST, r4 + mfspr r4, SPRN_HID0 ori r4, r4, 0x1 - mtspr HID0, r4 + mtspr SPRN_HID0, r4 mfspr r4, SPRN_CTRLF oris r4, r4, 0x40 mtspr SPRN_CTRLT, r4 blr #endif -#if defined(CONFIG_KEXEC) || (defined(CONFIG_SMP) && !defined(CONFIG_PPC_ISERIES)) +#if defined(CONFIG_KEXEC) || defined(CONFIG_SMP) _GLOBAL(smp_release_cpus) /* All secondary cpus are spinning on a common * spinloop, release them all now so they can start * to spin on their individual paca spinloops. * For non SMP kernels, the secondary cpus never * get out of the common spinloop. + * XXX This does nothing useful on iSeries, secondaries are + * already waiting on their paca. */ li r3,1 LOADADDR(r5,__secondary_hold_spinloop) std r3,0(r5) sync blr -#endif /* CONFIG_SMP && !CONFIG_PPC_ISERIES */ +#endif /* CONFIG_SMP */ /* @@ -1992,7 +1993,7 @@ _GLOBAL(smp_release_cpus) */ .section ".bss" - .align 12 + .align PAGE_SHIFT .globl empty_zero_page empty_zero_page: diff --git a/arch/ppc64/kernel/hvCall.S b/arch/ppc64/kernel/hvCall.S deleted file mode 100644 index 4c699eab1b95..000000000000 --- a/arch/ppc64/kernel/hvCall.S +++ /dev/null @@ -1,98 +0,0 @@ -/* - * arch/ppc64/kernel/hvCall.S - * - * - * This file contains the code to perform calls to the - * iSeries LPAR hypervisor - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <asm/ppc_asm.h> -#include <asm/processor.h> - - .text - -/* - * Hypervisor call - * - * Invoke the iSeries hypervisor via the System Call instruction - * Parameters are passed to this routine in registers r3 - r10 - * - * r3 contains the HV function to be called - * r4-r10 contain the operands to the hypervisor function - * - */ - -_GLOBAL(HvCall) -_GLOBAL(HvCall0) -_GLOBAL(HvCall1) -_GLOBAL(HvCall2) -_GLOBAL(HvCall3) -_GLOBAL(HvCall4) -_GLOBAL(HvCall5) -_GLOBAL(HvCall6) -_GLOBAL(HvCall7) - - - mfcr r0 - std r0,-8(r1) - stdu r1,-(STACK_FRAME_OVERHEAD+16)(r1) - - /* r0 = 0xffffffffffffffff indicates a hypervisor call */ - - li r0,-1 - - /* Invoke the hypervisor */ - - sc - - ld r1,0(r1) - ld r0,-8(r1) - mtcrf 0xff,r0 - - /* return to caller, return value in r3 */ - - blr - -_GLOBAL(HvCall0Ret16) -_GLOBAL(HvCall1Ret16) -_GLOBAL(HvCall2Ret16) -_GLOBAL(HvCall3Ret16) -_GLOBAL(HvCall4Ret16) -_GLOBAL(HvCall5Ret16) -_GLOBAL(HvCall6Ret16) -_GLOBAL(HvCall7Ret16) - - mfcr r0 - std r0,-8(r1) - std r31,-16(r1) - stdu r1,-(STACK_FRAME_OVERHEAD+32)(r1) - - mr r31,r4 - li r0,-1 - mr r4,r5 - mr r5,r6 - mr r6,r7 - mr r7,r8 - mr r8,r9 - mr r9,r10 - - sc - - std r3,0(r31) - std r4,8(r31) - - mr r3,r5 - - ld r1,0(r1) - ld r0,-8(r1) - mtcrf 0xff,r0 - ld r31,-16(r1) - - blr - - diff --git a/arch/ppc64/kernel/iSeries_VpdInfo.c b/arch/ppc64/kernel/iSeries_VpdInfo.c deleted file mode 100644 index 5d921792571f..000000000000 --- a/arch/ppc64/kernel/iSeries_VpdInfo.c +++ /dev/null @@ -1,268 +0,0 @@ -/* - * File iSeries_vpdInfo.c created by Allan Trautman on Fri Feb 2 2001. - * - * This code gets the card location of the hardware - * Copyright (C) 2001 <Allan H Trautman> <IBM Corp> - * Copyright (C) 2005 Stephen Rothwel, IBM Corp - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the: - * Free Software Foundation, Inc., - * 59 Temple Place, Suite 330, - * Boston, MA 02111-1307 USA - * - * Change Activity: - * Created, Feb 2, 2001 - * Ported to ppc64, August 20, 2001 - * End Change Activity - */ -#include <linux/init.h> -#include <linux/module.h> -#include <linux/pci.h> -#include <asm/types.h> -#include <asm/resource.h> - -#include <asm/iSeries/HvCallPci.h> -#include <asm/iSeries/HvTypes.h> -#include <asm/iSeries/iSeries_pci.h> - -/* - * Size of Bus VPD data - */ -#define BUS_VPDSIZE 1024 - -/* - * Bus Vpd Tags - */ -#define VpdEndOfAreaTag 0x79 -#define VpdIdStringTag 0x82 -#define VpdVendorAreaTag 0x84 - -/* - * Mfg Area Tags - */ -#define VpdFruFrameId 0x4649 // "FI" -#define VpdSlotMapFormat 0x4D46 // "MF" -#define VpdSlotMap 0x534D // "SM" - -/* - * Structures of the areas - */ -struct MfgVpdAreaStruct { - u16 Tag; - u8 TagLength; - u8 AreaData1; - u8 AreaData2; -}; -typedef struct MfgVpdAreaStruct MfgArea; -#define MFG_ENTRY_SIZE 3 - -struct SlotMapStruct { - u8 AgentId; - u8 SecondaryAgentId; - u8 PhbId; - char CardLocation[3]; - char Parms[8]; - char Reserved[2]; -}; -typedef struct SlotMapStruct SlotMap; -#define SLOT_ENTRY_SIZE 16 - -/* - * Parse the Slot Area - */ -static void __init iSeries_Parse_SlotArea(SlotMap *MapPtr, int MapLen, - HvAgentId agent, u8 *PhbId, char card[4]) -{ - int SlotMapLen = MapLen; - SlotMap *SlotMapPtr = MapPtr; - - /* - * Parse Slot label until we find the one requested - */ - while (SlotMapLen > 0) { - if (SlotMapPtr->AgentId == agent) { - /* - * If Phb wasn't found, grab the entry first one found. - */ - if (*PhbId == 0xff) - *PhbId = SlotMapPtr->PhbId; - /* Found it, extract the data. */ - if (SlotMapPtr->PhbId == *PhbId) { - memcpy(card, &SlotMapPtr->CardLocation, 3); - card[3] = 0; - break; - } - } - /* Point to the next Slot */ - SlotMapPtr = (SlotMap *)((char *)SlotMapPtr + SLOT_ENTRY_SIZE); - SlotMapLen -= SLOT_ENTRY_SIZE; - } -} - -/* - * Parse the Mfg Area - */ -static void __init iSeries_Parse_MfgArea(u8 *AreaData, int AreaLen, - HvAgentId agent, u8 *PhbId, - u8 *frame, char card[4]) -{ - MfgArea *MfgAreaPtr = (MfgArea *)AreaData; - int MfgAreaLen = AreaLen; - u16 SlotMapFmt = 0; - - /* Parse Mfg Data */ - while (MfgAreaLen > 0) { - int MfgTagLen = MfgAreaPtr->TagLength; - /* Frame ID (FI 4649020310 ) */ - if (MfgAreaPtr->Tag == VpdFruFrameId) /* FI */ - *frame = MfgAreaPtr->AreaData1; - /* Slot Map Format (MF 4D46020004 ) */ - else if (MfgAreaPtr->Tag == VpdSlotMapFormat) /* MF */ - SlotMapFmt = (MfgAreaPtr->AreaData1 * 256) - + MfgAreaPtr->AreaData2; - /* Slot Map (SM 534D90 */ - else if (MfgAreaPtr->Tag == VpdSlotMap) { /* SM */ - SlotMap *SlotMapPtr; - - if (SlotMapFmt == 0x1004) - SlotMapPtr = (SlotMap *)((char *)MfgAreaPtr - + MFG_ENTRY_SIZE + 1); - else - SlotMapPtr = (SlotMap *)((char *)MfgAreaPtr - + MFG_ENTRY_SIZE); - iSeries_Parse_SlotArea(SlotMapPtr, MfgTagLen, - agent, PhbId, card); - } - /* - * Point to the next Mfg Area - * Use defined size, sizeof give wrong answer - */ - MfgAreaPtr = (MfgArea *)((char *)MfgAreaPtr + MfgTagLen - + MFG_ENTRY_SIZE); - MfgAreaLen -= (MfgTagLen + MFG_ENTRY_SIZE); - } -} - -/* - * Look for "BUS".. Data is not Null terminated. - * PHBID of 0xFF indicates PHB was not found in VPD Data. - */ -static int __init iSeries_Parse_PhbId(u8 *AreaPtr, int AreaLength) -{ - u8 *PhbPtr = AreaPtr; - int DataLen = AreaLength; - char PhbId = 0xFF; - - while (DataLen > 0) { - if ((*PhbPtr == 'B') && (*(PhbPtr + 1) == 'U') - && (*(PhbPtr + 2) == 'S')) { - PhbPtr += 3; - while (*PhbPtr == ' ') - ++PhbPtr; - PhbId = (*PhbPtr & 0x0F); - break; - } - ++PhbPtr; - --DataLen; - } - return PhbId; -} - -/* - * Parse out the VPD Areas - */ -static void __init iSeries_Parse_Vpd(u8 *VpdData, int VpdDataLen, - HvAgentId agent, u8 *frame, char card[4]) -{ - u8 *TagPtr = VpdData; - int DataLen = VpdDataLen - 3; - u8 PhbId; - - while ((*TagPtr != VpdEndOfAreaTag) && (DataLen > 0)) { - int AreaLen = *(TagPtr + 1) + (*(TagPtr + 2) * 256); - u8 *AreaData = TagPtr + 3; - - if (*TagPtr == VpdIdStringTag) - PhbId = iSeries_Parse_PhbId(AreaData, AreaLen); - else if (*TagPtr == VpdVendorAreaTag) - iSeries_Parse_MfgArea(AreaData, AreaLen, - agent, &PhbId, frame, card); - /* Point to next Area. */ - TagPtr = AreaData + AreaLen; - DataLen -= AreaLen; - } -} - -static void __init iSeries_Get_Location_Code(u16 bus, HvAgentId agent, - u8 *frame, char card[4]) -{ - int BusVpdLen = 0; - u8 *BusVpdPtr = kmalloc(BUS_VPDSIZE, GFP_KERNEL); - - if (BusVpdPtr == NULL) { - printk("PCI: Bus VPD Buffer allocation failure.\n"); - return; - } - BusVpdLen = HvCallPci_getBusVpd(bus, ISERIES_HV_ADDR(BusVpdPtr), - BUS_VPDSIZE); - if (BusVpdLen == 0) { - printk("PCI: Bus VPD Buffer zero length.\n"); - goto out_free; - } - /* printk("PCI: BusVpdPtr: %p, %d\n",BusVpdPtr, BusVpdLen); */ - /* Make sure this is what I think it is */ - if (*BusVpdPtr != VpdIdStringTag) { /* 0x82 */ - printk("PCI: Bus VPD Buffer missing starting tag.\n"); - goto out_free; - } - iSeries_Parse_Vpd(BusVpdPtr, BusVpdLen, agent, frame, card); -out_free: - kfree(BusVpdPtr); -} - -/* - * Prints the device information. - * - Pass in pci_dev* pointer to the device. - * - Pass in the device count - * - * Format: - * PCI: Bus 0, Device 26, Vendor 0x12AE Frame 1, Card C10 Ethernet - * controller - */ -void __init iSeries_Device_Information(struct pci_dev *PciDev, int count) -{ - struct iSeries_Device_Node *DevNode = PciDev->sysdata; - u16 bus; - u8 frame; - char card[4]; - HvSubBusNumber subbus; - HvAgentId agent; - - if (DevNode == NULL) { - printk("%d. PCI: iSeries_Device_Information DevNode is NULL\n", - count); - return; - } - - bus = ISERIES_BUS(DevNode); - subbus = ISERIES_SUBBUS(DevNode); - agent = ISERIES_PCI_AGENTID(ISERIES_GET_DEVICE_FROM_SUBBUS(subbus), - ISERIES_GET_FUNCTION_FROM_SUBBUS(subbus)); - iSeries_Get_Location_Code(bus, agent, &frame, card); - - printk("%d. PCI: Bus%3d, Device%3d, Vendor %04X Frame%3d, Card %4s ", - count, bus, PCI_SLOT(PciDev->devfn), PciDev->vendor, - frame, card); - printk("0x%04X\n", (int)(PciDev->class >> 8)); -} diff --git a/arch/ppc64/kernel/iSeries_htab.c b/arch/ppc64/kernel/iSeries_htab.c deleted file mode 100644 index 073b76661747..000000000000 --- a/arch/ppc64/kernel/iSeries_htab.c +++ /dev/null @@ -1,236 +0,0 @@ -/* - * iSeries hashtable management. - * Derived from pSeries_htab.c - * - * SMP scalability work: - * Copyright (C) 2001 Anton Blanchard <anton@au.ibm.com>, IBM - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#include <asm/machdep.h> -#include <asm/pgtable.h> -#include <asm/mmu.h> -#include <asm/mmu_context.h> -#include <asm/iSeries/HvCallHpt.h> -#include <asm/abs_addr.h> -#include <linux/spinlock.h> - -static spinlock_t iSeries_hlocks[64] __cacheline_aligned_in_smp = { [0 ... 63] = SPIN_LOCK_UNLOCKED}; - -/* - * Very primitive algorithm for picking up a lock - */ -static inline void iSeries_hlock(unsigned long slot) -{ - if (slot & 0x8) - slot = ~slot; - spin_lock(&iSeries_hlocks[(slot >> 4) & 0x3f]); -} - -static inline void iSeries_hunlock(unsigned long slot) -{ - if (slot & 0x8) - slot = ~slot; - spin_unlock(&iSeries_hlocks[(slot >> 4) & 0x3f]); -} - -static long iSeries_hpte_insert(unsigned long hpte_group, unsigned long va, - unsigned long prpn, unsigned long vflags, - unsigned long rflags) -{ - unsigned long arpn; - long slot; - hpte_t lhpte; - int secondary = 0; - - /* - * The hypervisor tries both primary and secondary. - * If we are being called to insert in the secondary, - * it means we have already tried both primary and secondary, - * so we return failure immediately. - */ - if (vflags & HPTE_V_SECONDARY) - return -1; - - iSeries_hlock(hpte_group); - - slot = HvCallHpt_findValid(&lhpte, va >> PAGE_SHIFT); - BUG_ON(lhpte.v & HPTE_V_VALID); - - if (slot == -1) { /* No available entry found in either group */ - iSeries_hunlock(hpte_group); - return -1; - } - - if (slot < 0) { /* MSB set means secondary group */ - vflags |= HPTE_V_SECONDARY; - secondary = 1; - slot &= 0x7fffffffffffffff; - } - - arpn = phys_to_abs(prpn << PAGE_SHIFT) >> PAGE_SHIFT; - - lhpte.v = (va >> 23) << HPTE_V_AVPN_SHIFT | vflags | HPTE_V_VALID; - lhpte.r = (arpn << HPTE_R_RPN_SHIFT) | rflags; - - /* Now fill in the actual HPTE */ - HvCallHpt_addValidate(slot, secondary, &lhpte); - - iSeries_hunlock(hpte_group); - - return (secondary << 3) | (slot & 7); -} - -static unsigned long iSeries_hpte_getword0(unsigned long slot) -{ - hpte_t hpte; - - HvCallHpt_get(&hpte, slot); - return hpte.v; -} - -static long iSeries_hpte_remove(unsigned long hpte_group) -{ - unsigned long slot_offset; - int i; - unsigned long hpte_v; - - /* Pick a random slot to start at */ - slot_offset = mftb() & 0x7; - - iSeries_hlock(hpte_group); - - for (i = 0; i < HPTES_PER_GROUP; i++) { - hpte_v = iSeries_hpte_getword0(hpte_group + slot_offset); - - if (! (hpte_v & HPTE_V_BOLTED)) { - HvCallHpt_invalidateSetSwBitsGet(hpte_group + - slot_offset, 0, 0); - iSeries_hunlock(hpte_group); - return i; - } - - slot_offset++; - slot_offset &= 0x7; - } - - iSeries_hunlock(hpte_group); - - return -1; -} - -/* - * The HyperVisor expects the "flags" argument in this form: - * bits 0..59 : reserved - * bit 60 : N - * bits 61..63 : PP2,PP1,PP0 - */ -static long iSeries_hpte_updatepp(unsigned long slot, unsigned long newpp, - unsigned long va, int large, int local) -{ - hpte_t hpte; - unsigned long avpn = va >> 23; - - iSeries_hlock(slot); - - HvCallHpt_get(&hpte, slot); - if ((HPTE_V_AVPN_VAL(hpte.v) == avpn) && (hpte.v & HPTE_V_VALID)) { - /* - * Hypervisor expects bits as NPPP, which is - * different from how they are mapped in our PP. - */ - HvCallHpt_setPp(slot, (newpp & 0x3) | ((newpp & 0x4) << 1)); - iSeries_hunlock(slot); - return 0; - } - iSeries_hunlock(slot); - - return -1; -} - -/* - * Functions used to find the PTE for a particular virtual address. - * Only used during boot when bolting pages. - * - * Input : vpn : virtual page number - * Output: PTE index within the page table of the entry - * -1 on failure - */ -static long iSeries_hpte_find(unsigned long vpn) -{ - hpte_t hpte; - long slot; - - /* - * The HvCallHpt_findValid interface is as follows: - * 0xffffffffffffffff : No entry found. - * 0x00000000xxxxxxxx : Entry found in primary group, slot x - * 0x80000000xxxxxxxx : Entry found in secondary group, slot x - */ - slot = HvCallHpt_findValid(&hpte, vpn); - if (hpte.v & HPTE_V_VALID) { - if (slot < 0) { - slot &= 0x7fffffffffffffff; - slot = -slot; - } - } else - slot = -1; - return slot; -} - -/* - * Update the page protection bits. Intended to be used to create - * guard pages for kernel data structures on pages which are bolted - * in the HPT. Assumes pages being operated on will not be stolen. - * Does not work on large pages. - * - * No need to lock here because we should be the only user. - */ -static void iSeries_hpte_updateboltedpp(unsigned long newpp, unsigned long ea) -{ - unsigned long vsid,va,vpn; - long slot; - - vsid = get_kernel_vsid(ea); - va = (vsid << 28) | (ea & 0x0fffffff); - vpn = va >> PAGE_SHIFT; - slot = iSeries_hpte_find(vpn); - if (slot == -1) - panic("updateboltedpp: Could not find page to bolt\n"); - HvCallHpt_setPp(slot, newpp); -} - -static void iSeries_hpte_invalidate(unsigned long slot, unsigned long va, - int large, int local) -{ - unsigned long hpte_v; - unsigned long avpn = va >> 23; - unsigned long flags; - - local_irq_save(flags); - - iSeries_hlock(slot); - - hpte_v = iSeries_hpte_getword0(slot); - - if ((HPTE_V_AVPN_VAL(hpte_v) == avpn) && (hpte_v & HPTE_V_VALID)) - HvCallHpt_invalidateSetSwBitsGet(slot, 0, 0); - - iSeries_hunlock(slot); - - local_irq_restore(flags); -} - -void hpte_init_iSeries(void) -{ - ppc_md.hpte_invalidate = iSeries_hpte_invalidate; - ppc_md.hpte_updatepp = iSeries_hpte_updatepp; - ppc_md.hpte_updateboltedpp = iSeries_hpte_updateboltedpp; - ppc_md.hpte_insert = iSeries_hpte_insert; - ppc_md.hpte_remove = iSeries_hpte_remove; - - htab_finish_init(); -} diff --git a/arch/ppc64/kernel/iSeries_iommu.c b/arch/ppc64/kernel/iSeries_iommu.c deleted file mode 100644 index f8ff1bb054dc..000000000000 --- a/arch/ppc64/kernel/iSeries_iommu.c +++ /dev/null @@ -1,176 +0,0 @@ -/* - * arch/ppc64/kernel/iSeries_iommu.c - * - * Copyright (C) 2001 Mike Corrigan & Dave Engebretsen, IBM Corporation - * - * Rewrite, cleanup: - * - * Copyright (C) 2004 Olof Johansson <olof@austin.ibm.com>, IBM Corporation - * - * Dynamic DMA mapping support, iSeries-specific parts. - * - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - */ - -#include <linux/types.h> -#include <linux/dma-mapping.h> -#include <linux/list.h> - -#include <asm/iommu.h> -#include <asm/machdep.h> -#include <asm/iSeries/HvCallXm.h> -#include <asm/iSeries/iSeries_pci.h> - -extern struct list_head iSeries_Global_Device_List; - - -static void tce_build_iSeries(struct iommu_table *tbl, long index, long npages, - unsigned long uaddr, enum dma_data_direction direction) -{ - u64 rc; - union tce_entry tce; - - while (npages--) { - tce.te_word = 0; - tce.te_bits.tb_rpn = virt_to_abs(uaddr) >> PAGE_SHIFT; - - if (tbl->it_type == TCE_VB) { - /* Virtual Bus */ - tce.te_bits.tb_valid = 1; - tce.te_bits.tb_allio = 1; - if (direction != DMA_TO_DEVICE) - tce.te_bits.tb_rdwr = 1; - } else { - /* PCI Bus */ - tce.te_bits.tb_rdwr = 1; /* Read allowed */ - if (direction != DMA_TO_DEVICE) - tce.te_bits.tb_pciwr = 1; - } - - rc = HvCallXm_setTce((u64)tbl->it_index, (u64)index, - tce.te_word); - if (rc) - panic("PCI_DMA: HvCallXm_setTce failed, Rc: 0x%lx\n", - rc); - index++; - uaddr += PAGE_SIZE; - } -} - -static void tce_free_iSeries(struct iommu_table *tbl, long index, long npages) -{ - u64 rc; - - while (npages--) { - rc = HvCallXm_setTce((u64)tbl->it_index, (u64)index, 0); - if (rc) - panic("PCI_DMA: HvCallXm_setTce failed, Rc: 0x%lx\n", - rc); - index++; - } -} - -#ifdef CONFIG_PCI -/* - * This function compares the known tables to find an iommu_table - * that has already been built for hardware TCEs. - */ -static struct iommu_table *iommu_table_find(struct iommu_table * tbl) -{ - struct iSeries_Device_Node *dp; - - list_for_each_entry(dp, &iSeries_Global_Device_List, Device_List) { - if ((dp->iommu_table != NULL) && - (dp->iommu_table->it_type == TCE_PCI) && - (dp->iommu_table->it_offset == tbl->it_offset) && - (dp->iommu_table->it_index == tbl->it_index) && - (dp->iommu_table->it_size == tbl->it_size)) - return dp->iommu_table; - } - return NULL; -} - -/* - * Call Hv with the architected data structure to get TCE table info. - * info. Put the returned data into the Linux representation of the - * TCE table data. - * The Hardware Tce table comes in three flavors. - * 1. TCE table shared between Buses. - * 2. TCE table per Bus. - * 3. TCE Table per IOA. - */ -static void iommu_table_getparms(struct iSeries_Device_Node* dn, - struct iommu_table* tbl) -{ - struct iommu_table_cb *parms; - - parms = kmalloc(sizeof(*parms), GFP_KERNEL); - if (parms == NULL) - panic("PCI_DMA: TCE Table Allocation failed."); - - memset(parms, 0, sizeof(*parms)); - - parms->itc_busno = ISERIES_BUS(dn); - parms->itc_slotno = dn->LogicalSlot; - parms->itc_virtbus = 0; - - HvCallXm_getTceTableParms(ISERIES_HV_ADDR(parms)); - - if (parms->itc_size == 0) - panic("PCI_DMA: parms->size is zero, parms is 0x%p", parms); - - /* itc_size is in pages worth of table, it_size is in # of entries */ - tbl->it_size = (parms->itc_size * PAGE_SIZE) / sizeof(union tce_entry); - tbl->it_busno = parms->itc_busno; - tbl->it_offset = parms->itc_offset; - tbl->it_index = parms->itc_index; - tbl->it_blocksize = 1; - tbl->it_type = TCE_PCI; - - kfree(parms); -} - - -void iommu_devnode_init_iSeries(struct iSeries_Device_Node *dn) -{ - struct iommu_table *tbl; - - tbl = kmalloc(sizeof(struct iommu_table), GFP_KERNEL); - - iommu_table_getparms(dn, tbl); - - /* Look for existing tce table */ - dn->iommu_table = iommu_table_find(tbl); - if (dn->iommu_table == NULL) - dn->iommu_table = iommu_init_table(tbl); - else - kfree(tbl); -} -#endif - -static void iommu_dev_setup_iSeries(struct pci_dev *dev) { } -static void iommu_bus_setup_iSeries(struct pci_bus *bus) { } - -void iommu_init_early_iSeries(void) -{ - ppc_md.tce_build = tce_build_iSeries; - ppc_md.tce_free = tce_free_iSeries; - - ppc_md.iommu_dev_setup = iommu_dev_setup_iSeries; - ppc_md.iommu_bus_setup = iommu_bus_setup_iSeries; - - pci_iommu_init(); -} diff --git a/arch/ppc64/kernel/iSeries_irq.c b/arch/ppc64/kernel/iSeries_irq.c deleted file mode 100644 index 77376c1bd611..000000000000 --- a/arch/ppc64/kernel/iSeries_irq.c +++ /dev/null @@ -1,353 +0,0 @@ -/* - * This module supports the iSeries PCI bus interrupt handling - * Copyright (C) 20yy <Robert L Holtorf> <IBM Corp> - * Copyright (C) 2004-2005 IBM Corporation - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the: - * Free Software Foundation, Inc., - * 59 Temple Place, Suite 330, - * Boston, MA 02111-1307 USA - * - * Change Activity: - * Created, December 13, 2000 by Wayne Holm - * End Change Activity - */ -#include <linux/config.h> -#include <linux/pci.h> -#include <linux/init.h> -#include <linux/threads.h> -#include <linux/smp.h> -#include <linux/param.h> -#include <linux/string.h> -#include <linux/bootmem.h> -#include <linux/ide.h> -#include <linux/irq.h> -#include <linux/spinlock.h> - -#include <asm/ppcdebug.h> -#include <asm/iSeries/HvTypes.h> -#include <asm/iSeries/HvLpEvent.h> -#include <asm/iSeries/HvCallPci.h> -#include <asm/iSeries/HvCallXm.h> -#include <asm/iSeries/iSeries_irq.h> - -/* This maps virtual irq numbers to real irqs */ -unsigned int virt_irq_to_real_map[NR_IRQS]; - -/* The next available virtual irq number */ -/* Note: the pcnet32 driver assumes irq numbers < 2 aren't valid. :( */ -static int next_virtual_irq = 2; - -static long Pci_Interrupt_Count; -static long Pci_Event_Count; - -enum XmPciLpEvent_Subtype { - XmPciLpEvent_BusCreated = 0, // PHB has been created - XmPciLpEvent_BusError = 1, // PHB has failed - XmPciLpEvent_BusFailed = 2, // Msg to Secondary, Primary failed bus - XmPciLpEvent_NodeFailed = 4, // Multi-adapter bridge has failed - XmPciLpEvent_NodeRecovered = 5, // Multi-adapter bridge has recovered - XmPciLpEvent_BusRecovered = 12, // PHB has been recovered - XmPciLpEvent_UnQuiesceBus = 18, // Secondary bus unqiescing - XmPciLpEvent_BridgeError = 21, // Bridge Error - XmPciLpEvent_SlotInterrupt = 22 // Slot interrupt -}; - -struct XmPciLpEvent_BusInterrupt { - HvBusNumber busNumber; - HvSubBusNumber subBusNumber; -}; - -struct XmPciLpEvent_NodeInterrupt { - HvBusNumber busNumber; - HvSubBusNumber subBusNumber; - HvAgentId deviceId; -}; - -struct XmPciLpEvent { - struct HvLpEvent hvLpEvent; - - union { - u64 alignData; // Align on an 8-byte boundary - - struct { - u32 fisr; - HvBusNumber busNumber; - HvSubBusNumber subBusNumber; - HvAgentId deviceId; - } slotInterrupt; - - struct XmPciLpEvent_BusInterrupt busFailed; - struct XmPciLpEvent_BusInterrupt busRecovered; - struct XmPciLpEvent_BusInterrupt busCreated; - - struct XmPciLpEvent_NodeInterrupt nodeFailed; - struct XmPciLpEvent_NodeInterrupt nodeRecovered; - - } eventData; - -}; - -static void intReceived(struct XmPciLpEvent *eventParm, - struct pt_regs *regsParm) -{ - int irq; - - ++Pci_Interrupt_Count; - - switch (eventParm->hvLpEvent.xSubtype) { - case XmPciLpEvent_SlotInterrupt: - irq = eventParm->hvLpEvent.xCorrelationToken; - /* Dispatch the interrupt handlers for this irq */ - ppc_irq_dispatch_handler(regsParm, irq); - HvCallPci_eoi(eventParm->eventData.slotInterrupt.busNumber, - eventParm->eventData.slotInterrupt.subBusNumber, - eventParm->eventData.slotInterrupt.deviceId); - break; - /* Ignore error recovery events for now */ - case XmPciLpEvent_BusCreated: - printk(KERN_INFO "intReceived: system bus %d created\n", - eventParm->eventData.busCreated.busNumber); - break; - case XmPciLpEvent_BusError: - case XmPciLpEvent_BusFailed: - printk(KERN_INFO "intReceived: system bus %d failed\n", - eventParm->eventData.busFailed.busNumber); - break; - case XmPciLpEvent_BusRecovered: - case XmPciLpEvent_UnQuiesceBus: - printk(KERN_INFO "intReceived: system bus %d recovered\n", - eventParm->eventData.busRecovered.busNumber); - break; - case XmPciLpEvent_NodeFailed: - case XmPciLpEvent_BridgeError: - printk(KERN_INFO - "intReceived: multi-adapter bridge %d/%d/%d failed\n", - eventParm->eventData.nodeFailed.busNumber, - eventParm->eventData.nodeFailed.subBusNumber, - eventParm->eventData.nodeFailed.deviceId); - break; - case XmPciLpEvent_NodeRecovered: - printk(KERN_INFO - "intReceived: multi-adapter bridge %d/%d/%d recovered\n", - eventParm->eventData.nodeRecovered.busNumber, - eventParm->eventData.nodeRecovered.subBusNumber, - eventParm->eventData.nodeRecovered.deviceId); - break; - default: - printk(KERN_ERR - "intReceived: unrecognized event subtype 0x%x\n", - eventParm->hvLpEvent.xSubtype); - break; - } -} - -static void XmPciLpEvent_handler(struct HvLpEvent *eventParm, - struct pt_regs *regsParm) -{ -#ifdef CONFIG_PCI - ++Pci_Event_Count; - - if (eventParm && (eventParm->xType == HvLpEvent_Type_PciIo)) { - switch (eventParm->xFlags.xFunction) { - case HvLpEvent_Function_Int: - intReceived((struct XmPciLpEvent *)eventParm, regsParm); - break; - case HvLpEvent_Function_Ack: - printk(KERN_ERR - "XmPciLpEvent_handler: unexpected ack received\n"); - break; - default: - printk(KERN_ERR - "XmPciLpEvent_handler: unexpected event function %d\n", - (int)eventParm->xFlags.xFunction); - break; - } - } else if (eventParm) - printk(KERN_ERR - "XmPciLpEvent_handler: Unrecognized PCI event type 0x%x\n", - (int)eventParm->xType); - else - printk(KERN_ERR "XmPciLpEvent_handler: NULL event received\n"); -#endif -} - -/* - * This is called by init_IRQ. set in ppc_md.init_IRQ by iSeries_setup.c - * It must be called before the bus walk. - */ -void __init iSeries_init_IRQ(void) -{ - /* Register PCI event handler and open an event path */ - int xRc; - - xRc = HvLpEvent_registerHandler(HvLpEvent_Type_PciIo, - &XmPciLpEvent_handler); - if (xRc == 0) { - xRc = HvLpEvent_openPath(HvLpEvent_Type_PciIo, 0); - if (xRc != 0) - printk(KERN_ERR "iSeries_init_IRQ: open event path " - "failed with rc 0x%x\n", xRc); - } else - printk(KERN_ERR "iSeries_init_IRQ: register handler " - "failed with rc 0x%x\n", xRc); -} - -#define REAL_IRQ_TO_BUS(irq) ((((irq) >> 6) & 0xff) + 1) -#define REAL_IRQ_TO_IDSEL(irq) ((((irq) >> 3) & 7) + 1) -#define REAL_IRQ_TO_FUNC(irq) ((irq) & 7) - -/* - * This will be called by device drivers (via enable_IRQ) - * to enable INTA in the bridge interrupt status register. - */ -static void iSeries_enable_IRQ(unsigned int irq) -{ - u32 bus, deviceId, function, mask; - const u32 subBus = 0; - unsigned int rirq = virt_irq_to_real_map[irq]; - - /* The IRQ has already been locked by the caller */ - bus = REAL_IRQ_TO_BUS(rirq); - function = REAL_IRQ_TO_FUNC(rirq); - deviceId = (REAL_IRQ_TO_IDSEL(rirq) << 4) + function; - - /* Unmask secondary INTA */ - mask = 0x80000000; - HvCallPci_unmaskInterrupts(bus, subBus, deviceId, mask); - PPCDBG(PPCDBG_BUSWALK, "iSeries_enable_IRQ 0x%02X.%02X.%02X 0x%04X\n", - bus, subBus, deviceId, irq); -} - -/* This is called by iSeries_activate_IRQs */ -static unsigned int iSeries_startup_IRQ(unsigned int irq) -{ - u32 bus, deviceId, function, mask; - const u32 subBus = 0; - unsigned int rirq = virt_irq_to_real_map[irq]; - - bus = REAL_IRQ_TO_BUS(rirq); - function = REAL_IRQ_TO_FUNC(rirq); - deviceId = (REAL_IRQ_TO_IDSEL(rirq) << 4) + function; - - /* Link the IRQ number to the bridge */ - HvCallXm_connectBusUnit(bus, subBus, deviceId, irq); - - /* Unmask bridge interrupts in the FISR */ - mask = 0x01010000 << function; - HvCallPci_unmaskFisr(bus, subBus, deviceId, mask); - iSeries_enable_IRQ(irq); - return 0; -} - -/* - * This is called out of iSeries_fixup to activate interrupt - * generation for usable slots - */ -void __init iSeries_activate_IRQs() -{ - int irq; - unsigned long flags; - - for_each_irq (irq) { - irq_desc_t *desc = get_irq_desc(irq); - - if (desc && desc->handler && desc->handler->startup) { - spin_lock_irqsave(&desc->lock, flags); - desc->handler->startup(irq); - spin_unlock_irqrestore(&desc->lock, flags); - } - } -} - -/* this is not called anywhere currently */ -static void iSeries_shutdown_IRQ(unsigned int irq) -{ - u32 bus, deviceId, function, mask; - const u32 subBus = 0; - unsigned int rirq = virt_irq_to_real_map[irq]; - - /* irq should be locked by the caller */ - bus = REAL_IRQ_TO_BUS(rirq); - function = REAL_IRQ_TO_FUNC(rirq); - deviceId = (REAL_IRQ_TO_IDSEL(rirq) << 4) + function; - - /* Invalidate the IRQ number in the bridge */ - HvCallXm_connectBusUnit(bus, subBus, deviceId, 0); - - /* Mask bridge interrupts in the FISR */ - mask = 0x01010000 << function; - HvCallPci_maskFisr(bus, subBus, deviceId, mask); -} - -/* - * This will be called by device drivers (via disable_IRQ) - * to disable INTA in the bridge interrupt status register. - */ -static void iSeries_disable_IRQ(unsigned int irq) -{ - u32 bus, deviceId, function, mask; - const u32 subBus = 0; - unsigned int rirq = virt_irq_to_real_map[irq]; - - /* The IRQ has already been locked by the caller */ - bus = REAL_IRQ_TO_BUS(rirq); - function = REAL_IRQ_TO_FUNC(rirq); - deviceId = (REAL_IRQ_TO_IDSEL(rirq) << 4) + function; - - /* Mask secondary INTA */ - mask = 0x80000000; - HvCallPci_maskInterrupts(bus, subBus, deviceId, mask); - PPCDBG(PPCDBG_BUSWALK, "iSeries_disable_IRQ 0x%02X.%02X.%02X 0x%04X\n", - bus, subBus, deviceId, irq); -} - -/* - * Need to define this so ppc_irq_dispatch_handler will NOT call - * enable_IRQ at the end of interrupt handling. However, this does - * nothing because there is not enough information provided to do - * the EOI HvCall. This is done by XmPciLpEvent.c - */ -static void iSeries_end_IRQ(unsigned int irq) -{ -} - -static hw_irq_controller iSeries_IRQ_handler = { - .typename = "iSeries irq controller", - .startup = iSeries_startup_IRQ, - .shutdown = iSeries_shutdown_IRQ, - .enable = iSeries_enable_IRQ, - .disable = iSeries_disable_IRQ, - .end = iSeries_end_IRQ -}; - -/* - * This is called out of iSeries_scan_slot to allocate an IRQ for an EADS slot - * It calculates the irq value for the slot. - * Note that subBusNumber is always 0 (at the moment at least). - */ -int __init iSeries_allocate_IRQ(HvBusNumber busNumber, - HvSubBusNumber subBusNumber, HvAgentId deviceId) -{ - unsigned int realirq, virtirq; - u8 idsel = (deviceId >> 4); - u8 function = deviceId & 7; - - virtirq = next_virtual_irq++; - realirq = ((busNumber - 1) << 6) + ((idsel - 1) << 3) + function; - virt_irq_to_real_map[virtirq] = realirq; - - irq_desc[virtirq].handler = &iSeries_IRQ_handler; - return virtirq; -} diff --git a/arch/ppc64/kernel/iSeries_pci.c b/arch/ppc64/kernel/iSeries_pci.c deleted file mode 100644 index fbc273c32bcc..000000000000 --- a/arch/ppc64/kernel/iSeries_pci.c +++ /dev/null @@ -1,905 +0,0 @@ -/* - * iSeries_pci.c - * - * Copyright (C) 2001 Allan Trautman, IBM Corporation - * - * iSeries specific routines for PCI. - * - * Based on code from pci.c and iSeries_pci.c 32bit - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - */ -#include <linux/kernel.h> -#include <linux/list.h> -#include <linux/string.h> -#include <linux/init.h> -#include <linux/module.h> -#include <linux/ide.h> -#include <linux/pci.h> - -#include <asm/io.h> -#include <asm/irq.h> -#include <asm/prom.h> -#include <asm/machdep.h> -#include <asm/pci-bridge.h> -#include <asm/ppcdebug.h> -#include <asm/iommu.h> - -#include <asm/iSeries/HvCallPci.h> -#include <asm/iSeries/HvCallXm.h> -#include <asm/iSeries/iSeries_irq.h> -#include <asm/iSeries/iSeries_pci.h> -#include <asm/iSeries/mf.h> - -#include "pci.h" - -extern unsigned long io_page_mask; - -/* - * Forward declares of prototypes. - */ -static struct iSeries_Device_Node *find_Device_Node(int bus, int devfn); -static void scan_PHB_slots(struct pci_controller *Phb); -static void scan_EADS_bridge(HvBusNumber Bus, HvSubBusNumber SubBus, int IdSel); -static int scan_bridge_slot(HvBusNumber Bus, struct HvCallPci_BridgeInfo *Info); - -LIST_HEAD(iSeries_Global_Device_List); - -static int DeviceCount; - -/* Counters and control flags. */ -static long Pci_Io_Read_Count; -static long Pci_Io_Write_Count; -#if 0 -static long Pci_Cfg_Read_Count; -static long Pci_Cfg_Write_Count; -#endif -static long Pci_Error_Count; - -static int Pci_Retry_Max = 3; /* Only retry 3 times */ -static int Pci_Error_Flag = 1; /* Set Retry Error on. */ - -static struct pci_ops iSeries_pci_ops; - -/* - * Table defines - * Each Entry size is 4 MB * 1024 Entries = 4GB I/O address space. - */ -#define IOMM_TABLE_MAX_ENTRIES 1024 -#define IOMM_TABLE_ENTRY_SIZE 0x0000000000400000UL -#define BASE_IO_MEMORY 0xE000000000000000UL - -static unsigned long max_io_memory = 0xE000000000000000UL; -static long current_iomm_table_entry; - -/* - * Lookup Tables. - */ -static struct iSeries_Device_Node **iomm_table; -static u8 *iobar_table; - -/* - * Static and Global variables - */ -static char *pci_io_text = "iSeries PCI I/O"; -static DEFINE_SPINLOCK(iomm_table_lock); - -/* - * iomm_table_initialize - * - * Allocates and initalizes the Address Translation Table and Bar - * Tables to get them ready for use. Must be called before any - * I/O space is handed out to the device BARs. - */ -static void iomm_table_initialize(void) -{ - spin_lock(&iomm_table_lock); - iomm_table = kmalloc(sizeof(*iomm_table) * IOMM_TABLE_MAX_ENTRIES, - GFP_KERNEL); - iobar_table = kmalloc(sizeof(*iobar_table) * IOMM_TABLE_MAX_ENTRIES, - GFP_KERNEL); - spin_unlock(&iomm_table_lock); - if ((iomm_table == NULL) || (iobar_table == NULL)) - panic("PCI: I/O tables allocation failed.\n"); -} - -/* - * iomm_table_allocate_entry - * - * Adds pci_dev entry in address translation table - * - * - Allocates the number of entries required in table base on BAR - * size. - * - Allocates starting at BASE_IO_MEMORY and increases. - * - The size is round up to be a multiple of entry size. - * - CurrentIndex is incremented to keep track of the last entry. - * - Builds the resource entry for allocated BARs. - */ -static void iomm_table_allocate_entry(struct pci_dev *dev, int bar_num) -{ - struct resource *bar_res = &dev->resource[bar_num]; - long bar_size = pci_resource_len(dev, bar_num); - - /* - * No space to allocate, quick exit, skip Allocation. - */ - if (bar_size == 0) - return; - /* - * Set Resource values. - */ - spin_lock(&iomm_table_lock); - bar_res->name = pci_io_text; - bar_res->start = - IOMM_TABLE_ENTRY_SIZE * current_iomm_table_entry; - bar_res->start += BASE_IO_MEMORY; - bar_res->end = bar_res->start + bar_size - 1; - /* - * Allocate the number of table entries needed for BAR. - */ - while (bar_size > 0 ) { - iomm_table[current_iomm_table_entry] = dev->sysdata; - iobar_table[current_iomm_table_entry] = bar_num; - bar_size -= IOMM_TABLE_ENTRY_SIZE; - ++current_iomm_table_entry; - } - max_io_memory = BASE_IO_MEMORY + - (IOMM_TABLE_ENTRY_SIZE * current_iomm_table_entry); - spin_unlock(&iomm_table_lock); -} - -/* - * allocate_device_bars - * - * - Allocates ALL pci_dev BAR's and updates the resources with the - * BAR value. BARS with zero length will have the resources - * The HvCallPci_getBarParms is used to get the size of the BAR - * space. It calls iomm_table_allocate_entry to allocate - * each entry. - * - Loops through The Bar resources(0 - 5) including the ROM - * is resource(6). - */ -static void allocate_device_bars(struct pci_dev *dev) -{ - struct resource *bar_res; - int bar_num; - - for (bar_num = 0; bar_num <= PCI_ROM_RESOURCE; ++bar_num) { - bar_res = &dev->resource[bar_num]; - iomm_table_allocate_entry(dev, bar_num); - } -} - -/* - * Log error information to system console. - * Filter out the device not there errors. - * PCI: EADs Connect Failed 0x18.58.10 Rc: 0x00xx - * PCI: Read Vendor Failed 0x18.58.10 Rc: 0x00xx - * PCI: Connect Bus Unit Failed 0x18.58.10 Rc: 0x00xx - */ -static void pci_Log_Error(char *Error_Text, int Bus, int SubBus, - int AgentId, int HvRc) -{ - if (HvRc == 0x0302) - return; - printk(KERN_ERR "PCI: %s Failed: 0x%02X.%02X.%02X Rc: 0x%04X", - Error_Text, Bus, SubBus, AgentId, HvRc); -} - -/* - * build_device_node(u16 Bus, int SubBus, u8 DevFn) - */ -static struct iSeries_Device_Node *build_device_node(HvBusNumber Bus, - HvSubBusNumber SubBus, int AgentId, int Function) -{ - struct iSeries_Device_Node *node; - - PPCDBG(PPCDBG_BUSWALK, - "-build_device_node 0x%02X.%02X.%02X Function: %02X\n", - Bus, SubBus, AgentId, Function); - - node = kmalloc(sizeof(struct iSeries_Device_Node), GFP_KERNEL); - if (node == NULL) - return NULL; - - memset(node, 0, sizeof(struct iSeries_Device_Node)); - list_add_tail(&node->Device_List, &iSeries_Global_Device_List); -#if 0 - node->DsaAddr = ((u64)Bus << 48) + ((u64)SubBus << 40) + ((u64)0x10 << 32); -#endif - node->DsaAddr.DsaAddr = 0; - node->DsaAddr.Dsa.busNumber = Bus; - node->DsaAddr.Dsa.subBusNumber = SubBus; - node->DsaAddr.Dsa.deviceId = 0x10; - node->DevFn = PCI_DEVFN(ISERIES_ENCODE_DEVICE(AgentId), Function); - return node; -} - -/* - * unsigned long __init find_and_init_phbs(void) - * - * Description: - * This function checks for all possible system PCI host bridges that connect - * PCI buses. The system hypervisor is queried as to the guest partition - * ownership status. A pci_controller is built for any bus which is partially - * owned or fully owned by this guest partition. - */ -unsigned long __init find_and_init_phbs(void) -{ - struct pci_controller *phb; - HvBusNumber bus; - - PPCDBG(PPCDBG_BUSWALK, "find_and_init_phbs Entry\n"); - - /* Check all possible buses. */ - for (bus = 0; bus < 256; bus++) { - int ret = HvCallXm_testBus(bus); - if (ret == 0) { - printk("bus %d appears to exist\n", bus); - - phb = (struct pci_controller *)kmalloc(sizeof(struct pci_controller), GFP_KERNEL); - if (phb == NULL) - return -ENOMEM; - pci_setup_pci_controller(phb); - - phb->pci_mem_offset = phb->local_number = bus; - phb->first_busno = bus; - phb->last_busno = bus; - phb->ops = &iSeries_pci_ops; - - PPCDBG(PPCDBG_BUSWALK, "PCI:Create iSeries pci_controller(%p), Bus: %04X\n", - phb, bus); - - /* Find and connect the devices. */ - scan_PHB_slots(phb); - } - /* - * Check for Unexpected Return code, a clue that something - * has gone wrong. - */ - else if (ret != 0x0301) - printk(KERN_ERR "Unexpected Return on Probe(0x%04X): 0x%04X", - bus, ret); - } - return 0; -} - -/* - * iSeries_pcibios_init - * - * Chance to initialize and structures or variable before PCI Bus walk. - */ -void iSeries_pcibios_init(void) -{ - PPCDBG(PPCDBG_BUSWALK, "iSeries_pcibios_init Entry.\n"); - iomm_table_initialize(); - find_and_init_phbs(); - io_page_mask = -1; - PPCDBG(PPCDBG_BUSWALK, "iSeries_pcibios_init Exit.\n"); -} - -/* - * iSeries_pci_final_fixup(void) - */ -void __init iSeries_pci_final_fixup(void) -{ - struct pci_dev *pdev = NULL; - struct iSeries_Device_Node *node; - int DeviceCount = 0; - - PPCDBG(PPCDBG_BUSWALK, "iSeries_pcibios_fixup Entry.\n"); - - /* Fix up at the device node and pci_dev relationship */ - mf_display_src(0xC9000100); - - printk("pcibios_final_fixup\n"); - for_each_pci_dev(pdev) { - node = find_Device_Node(pdev->bus->number, pdev->devfn); - printk("pci dev %p (%x.%x), node %p\n", pdev, - pdev->bus->number, pdev->devfn, node); - - if (node != NULL) { - ++DeviceCount; - pdev->sysdata = (void *)node; - node->PciDev = pdev; - PPCDBG(PPCDBG_BUSWALK, - "pdev 0x%p <==> DevNode 0x%p\n", - pdev, node); - allocate_device_bars(pdev); - iSeries_Device_Information(pdev, DeviceCount); - iommu_devnode_init_iSeries(node); - } else - printk("PCI: Device Tree not found for 0x%016lX\n", - (unsigned long)pdev); - pdev->irq = node->Irq; - } - iSeries_activate_IRQs(); - mf_display_src(0xC9000200); -} - -void pcibios_fixup_bus(struct pci_bus *PciBus) -{ - PPCDBG(PPCDBG_BUSWALK, "iSeries_pcibios_fixup_bus(0x%04X) Entry.\n", - PciBus->number); -} - -void pcibios_fixup_resources(struct pci_dev *pdev) -{ - PPCDBG(PPCDBG_BUSWALK, "fixup_resources pdev %p\n", pdev); -} - -/* - * Loop through each node function to find usable EADs bridges. - */ -static void scan_PHB_slots(struct pci_controller *Phb) -{ - struct HvCallPci_DeviceInfo *DevInfo; - HvBusNumber bus = Phb->local_number; /* System Bus */ - const HvSubBusNumber SubBus = 0; /* EADs is always 0. */ - int HvRc = 0; - int IdSel; - const int MaxAgents = 8; - - DevInfo = (struct HvCallPci_DeviceInfo*) - kmalloc(sizeof(struct HvCallPci_DeviceInfo), GFP_KERNEL); - if (DevInfo == NULL) - return; - - /* - * Probe for EADs Bridges - */ - for (IdSel = 1; IdSel < MaxAgents; ++IdSel) { - HvRc = HvCallPci_getDeviceInfo(bus, SubBus, IdSel, - ISERIES_HV_ADDR(DevInfo), - sizeof(struct HvCallPci_DeviceInfo)); - if (HvRc == 0) { - if (DevInfo->deviceType == HvCallPci_NodeDevice) - scan_EADS_bridge(bus, SubBus, IdSel); - else - printk("PCI: Invalid System Configuration(0x%02X)" - " for bus 0x%02x id 0x%02x.\n", - DevInfo->deviceType, bus, IdSel); - } - else - pci_Log_Error("getDeviceInfo", bus, SubBus, IdSel, HvRc); - } - kfree(DevInfo); -} - -static void scan_EADS_bridge(HvBusNumber bus, HvSubBusNumber SubBus, - int IdSel) -{ - struct HvCallPci_BridgeInfo *BridgeInfo; - HvAgentId AgentId; - int Function; - int HvRc; - - BridgeInfo = (struct HvCallPci_BridgeInfo *) - kmalloc(sizeof(struct HvCallPci_BridgeInfo), GFP_KERNEL); - if (BridgeInfo == NULL) - return; - - /* Note: hvSubBus and irq is always be 0 at this level! */ - for (Function = 0; Function < 8; ++Function) { - AgentId = ISERIES_PCI_AGENTID(IdSel, Function); - HvRc = HvCallXm_connectBusUnit(bus, SubBus, AgentId, 0); - if (HvRc == 0) { - printk("found device at bus %d idsel %d func %d (AgentId %x)\n", - bus, IdSel, Function, AgentId); - /* Connect EADs: 0x18.00.12 = 0x00 */ - PPCDBG(PPCDBG_BUSWALK, - "PCI:Connect EADs: 0x%02X.%02X.%02X\n", - bus, SubBus, AgentId); - HvRc = HvCallPci_getBusUnitInfo(bus, SubBus, AgentId, - ISERIES_HV_ADDR(BridgeInfo), - sizeof(struct HvCallPci_BridgeInfo)); - if (HvRc == 0) { - printk("bridge info: type %x subbus %x maxAgents %x maxsubbus %x logslot %x\n", - BridgeInfo->busUnitInfo.deviceType, - BridgeInfo->subBusNumber, - BridgeInfo->maxAgents, - BridgeInfo->maxSubBusNumber, - BridgeInfo->logicalSlotNumber); - PPCDBG(PPCDBG_BUSWALK, - "PCI: BridgeInfo, Type:0x%02X, SubBus:0x%02X, MaxAgents:0x%02X, MaxSubBus: 0x%02X, LSlot: 0x%02X\n", - BridgeInfo->busUnitInfo.deviceType, - BridgeInfo->subBusNumber, - BridgeInfo->maxAgents, - BridgeInfo->maxSubBusNumber, - BridgeInfo->logicalSlotNumber); - - if (BridgeInfo->busUnitInfo.deviceType == - HvCallPci_BridgeDevice) { - /* Scan_Bridge_Slot...: 0x18.00.12 */ - scan_bridge_slot(bus, BridgeInfo); - } else - printk("PCI: Invalid Bridge Configuration(0x%02X)", - BridgeInfo->busUnitInfo.deviceType); - } - } else if (HvRc != 0x000B) - pci_Log_Error("EADs Connect", - bus, SubBus, AgentId, HvRc); - } - kfree(BridgeInfo); -} - -/* - * This assumes that the node slot is always on the primary bus! - */ -static int scan_bridge_slot(HvBusNumber Bus, - struct HvCallPci_BridgeInfo *BridgeInfo) -{ - struct iSeries_Device_Node *node; - HvSubBusNumber SubBus = BridgeInfo->subBusNumber; - u16 VendorId = 0; - int HvRc = 0; - u8 Irq = 0; - int IdSel = ISERIES_GET_DEVICE_FROM_SUBBUS(SubBus); - int Function = ISERIES_GET_FUNCTION_FROM_SUBBUS(SubBus); - HvAgentId EADsIdSel = ISERIES_PCI_AGENTID(IdSel, Function); - - /* iSeries_allocate_IRQ.: 0x18.00.12(0xA3) */ - Irq = iSeries_allocate_IRQ(Bus, 0, EADsIdSel); - PPCDBG(PPCDBG_BUSWALK, - "PCI:- allocate and assign IRQ 0x%02X.%02X.%02X = 0x%02X\n", - Bus, 0, EADsIdSel, Irq); - - /* - * Connect all functions of any device found. - */ - for (IdSel = 1; IdSel <= BridgeInfo->maxAgents; ++IdSel) { - for (Function = 0; Function < 8; ++Function) { - HvAgentId AgentId = ISERIES_PCI_AGENTID(IdSel, Function); - HvRc = HvCallXm_connectBusUnit(Bus, SubBus, - AgentId, Irq); - if (HvRc != 0) { - pci_Log_Error("Connect Bus Unit", - Bus, SubBus, AgentId, HvRc); - continue; - } - - HvRc = HvCallPci_configLoad16(Bus, SubBus, AgentId, - PCI_VENDOR_ID, &VendorId); - if (HvRc != 0) { - pci_Log_Error("Read Vendor", - Bus, SubBus, AgentId, HvRc); - continue; - } - printk("read vendor ID: %x\n", VendorId); - - /* FoundDevice: 0x18.28.10 = 0x12AE */ - PPCDBG(PPCDBG_BUSWALK, - "PCI:- FoundDevice: 0x%02X.%02X.%02X = 0x%04X, irq %d\n", - Bus, SubBus, AgentId, VendorId, Irq); - HvRc = HvCallPci_configStore8(Bus, SubBus, AgentId, - PCI_INTERRUPT_LINE, Irq); - if (HvRc != 0) - pci_Log_Error("PciCfgStore Irq Failed!", - Bus, SubBus, AgentId, HvRc); - - ++DeviceCount; - node = build_device_node(Bus, SubBus, EADsIdSel, Function); - node->Irq = Irq; - node->LogicalSlot = BridgeInfo->logicalSlotNumber; - - } /* for (Function = 0; Function < 8; ++Function) */ - } /* for (IdSel = 1; IdSel <= MaxAgents; ++IdSel) */ - return HvRc; -} - -/* - * I/0 Memory copy MUST use mmio commands on iSeries - * To do; For performance, include the hv call directly - */ -void iSeries_memset_io(volatile void __iomem *dest, char c, size_t Count) -{ - u8 ByteValue = c; - long NumberOfBytes = Count; - - while (NumberOfBytes > 0) { - iSeries_Write_Byte(ByteValue, dest++); - -- NumberOfBytes; - } -} -EXPORT_SYMBOL(iSeries_memset_io); - -void iSeries_memcpy_toio(volatile void __iomem *dest, void *source, size_t count) -{ - char *src = source; - long NumberOfBytes = count; - - while (NumberOfBytes > 0) { - iSeries_Write_Byte(*src++, dest++); - -- NumberOfBytes; - } -} -EXPORT_SYMBOL(iSeries_memcpy_toio); - -void iSeries_memcpy_fromio(void *dest, const volatile void __iomem *src, size_t count) -{ - char *dst = dest; - long NumberOfBytes = count; - - while (NumberOfBytes > 0) { - *dst++ = iSeries_Read_Byte(src++); - -- NumberOfBytes; - } -} -EXPORT_SYMBOL(iSeries_memcpy_fromio); - -/* - * Look down the chain to find the matching Device Device - */ -static struct iSeries_Device_Node *find_Device_Node(int bus, int devfn) -{ - struct list_head *pos; - - list_for_each(pos, &iSeries_Global_Device_List) { - struct iSeries_Device_Node *node = - list_entry(pos, struct iSeries_Device_Node, Device_List); - - if ((bus == ISERIES_BUS(node)) && (devfn == node->DevFn)) - return node; - } - return NULL; -} - -#if 0 -/* - * Returns the device node for the passed pci_dev - * Sanity Check Node PciDev to passed pci_dev - * If none is found, returns a NULL which the client must handle. - */ -static struct iSeries_Device_Node *get_Device_Node(struct pci_dev *pdev) -{ - struct iSeries_Device_Node *node; - - node = pdev->sysdata; - if (node == NULL || node->PciDev != pdev) - node = find_Device_Node(pdev->bus->number, pdev->devfn); - return node; -} -#endif - -/* - * Config space read and write functions. - * For now at least, we look for the device node for the bus and devfn - * that we are asked to access. It may be possible to translate the devfn - * to a subbus and deviceid more directly. - */ -static u64 hv_cfg_read_func[4] = { - HvCallPciConfigLoad8, HvCallPciConfigLoad16, - HvCallPciConfigLoad32, HvCallPciConfigLoad32 -}; - -static u64 hv_cfg_write_func[4] = { - HvCallPciConfigStore8, HvCallPciConfigStore16, - HvCallPciConfigStore32, HvCallPciConfigStore32 -}; - -/* - * Read PCI config space - */ -static int iSeries_pci_read_config(struct pci_bus *bus, unsigned int devfn, - int offset, int size, u32 *val) -{ - struct iSeries_Device_Node *node = find_Device_Node(bus->number, devfn); - u64 fn; - struct HvCallPci_LoadReturn ret; - - if (node == NULL) - return PCIBIOS_DEVICE_NOT_FOUND; - if (offset > 255) { - *val = ~0; - return PCIBIOS_BAD_REGISTER_NUMBER; - } - - fn = hv_cfg_read_func[(size - 1) & 3]; - HvCall3Ret16(fn, &ret, node->DsaAddr.DsaAddr, offset, 0); - - if (ret.rc != 0) { - *val = ~0; - return PCIBIOS_DEVICE_NOT_FOUND; /* or something */ - } - - *val = ret.value; - return 0; -} - -/* - * Write PCI config space - */ - -static int iSeries_pci_write_config(struct pci_bus *bus, unsigned int devfn, - int offset, int size, u32 val) -{ - struct iSeries_Device_Node *node = find_Device_Node(bus->number, devfn); - u64 fn; - u64 ret; - - if (node == NULL) - return PCIBIOS_DEVICE_NOT_FOUND; - if (offset > 255) - return PCIBIOS_BAD_REGISTER_NUMBER; - - fn = hv_cfg_write_func[(size - 1) & 3]; - ret = HvCall4(fn, node->DsaAddr.DsaAddr, offset, val, 0); - - if (ret != 0) - return PCIBIOS_DEVICE_NOT_FOUND; - - return 0; -} - -static struct pci_ops iSeries_pci_ops = { - .read = iSeries_pci_read_config, - .write = iSeries_pci_write_config -}; - -/* - * Check Return Code - * -> On Failure, print and log information. - * Increment Retry Count, if exceeds max, panic partition. - * - * PCI: Device 23.90 ReadL I/O Error( 0): 0x1234 - * PCI: Device 23.90 ReadL Retry( 1) - * PCI: Device 23.90 ReadL Retry Successful(1) - */ -static int CheckReturnCode(char *TextHdr, struct iSeries_Device_Node *DevNode, - int *retry, u64 ret) -{ - if (ret != 0) { - ++Pci_Error_Count; - (*retry)++; - printk("PCI: %s: Device 0x%04X:%02X I/O Error(%2d): 0x%04X\n", - TextHdr, DevNode->DsaAddr.Dsa.busNumber, DevNode->DevFn, - *retry, (int)ret); - /* - * Bump the retry and check for retry count exceeded. - * If, Exceeded, panic the system. - */ - if (((*retry) > Pci_Retry_Max) && - (Pci_Error_Flag > 0)) { - mf_display_src(0xB6000103); - panic_timeout = 0; - panic("PCI: Hardware I/O Error, SRC B6000103, " - "Automatic Reboot Disabled.\n"); - } - return -1; /* Retry Try */ - } - return 0; -} - -/* - * Translate the I/O Address into a device node, bar, and bar offset. - * Note: Make sure the passed variable end up on the stack to avoid - * the exposure of being device global. - */ -static inline struct iSeries_Device_Node *xlate_iomm_address( - const volatile void __iomem *IoAddress, - u64 *dsaptr, u64 *BarOffsetPtr) -{ - unsigned long OrigIoAddr; - unsigned long BaseIoAddr; - unsigned long TableIndex; - struct iSeries_Device_Node *DevNode; - - OrigIoAddr = (unsigned long __force)IoAddress; - if ((OrigIoAddr < BASE_IO_MEMORY) || (OrigIoAddr >= max_io_memory)) - return NULL; - BaseIoAddr = OrigIoAddr - BASE_IO_MEMORY; - TableIndex = BaseIoAddr / IOMM_TABLE_ENTRY_SIZE; - DevNode = iomm_table[TableIndex]; - - if (DevNode != NULL) { - int barnum = iobar_table[TableIndex]; - *dsaptr = DevNode->DsaAddr.DsaAddr | (barnum << 24); - *BarOffsetPtr = BaseIoAddr % IOMM_TABLE_ENTRY_SIZE; - } else - panic("PCI: Invalid PCI IoAddress detected!\n"); - return DevNode; -} - -/* - * Read MM I/O Instructions for the iSeries - * On MM I/O error, all ones are returned and iSeries_pci_IoError is cal - * else, data is returned in big Endian format. - * - * iSeries_Read_Byte = Read Byte ( 8 bit) - * iSeries_Read_Word = Read Word (16 bit) - * iSeries_Read_Long = Read Long (32 bit) - */ -u8 iSeries_Read_Byte(const volatile void __iomem *IoAddress) -{ - u64 BarOffset; - u64 dsa; - int retry = 0; - struct HvCallPci_LoadReturn ret; - struct iSeries_Device_Node *DevNode = - xlate_iomm_address(IoAddress, &dsa, &BarOffset); - - if (DevNode == NULL) { - static unsigned long last_jiffies; - static int num_printed; - - if ((jiffies - last_jiffies) > 60 * HZ) { - last_jiffies = jiffies; - num_printed = 0; - } - if (num_printed++ < 10) - printk(KERN_ERR "iSeries_Read_Byte: invalid access at IO address %p\n", IoAddress); - return 0xff; - } - do { - ++Pci_Io_Read_Count; - HvCall3Ret16(HvCallPciBarLoad8, &ret, dsa, BarOffset, 0); - } while (CheckReturnCode("RDB", DevNode, &retry, ret.rc) != 0); - - return (u8)ret.value; -} -EXPORT_SYMBOL(iSeries_Read_Byte); - -u16 iSeries_Read_Word(const volatile void __iomem *IoAddress) -{ - u64 BarOffset; - u64 dsa; - int retry = 0; - struct HvCallPci_LoadReturn ret; - struct iSeries_Device_Node *DevNode = - xlate_iomm_address(IoAddress, &dsa, &BarOffset); - - if (DevNode == NULL) { - static unsigned long last_jiffies; - static int num_printed; - - if ((jiffies - last_jiffies) > 60 * HZ) { - last_jiffies = jiffies; - num_printed = 0; - } - if (num_printed++ < 10) - printk(KERN_ERR "iSeries_Read_Word: invalid access at IO address %p\n", IoAddress); - return 0xffff; - } - do { - ++Pci_Io_Read_Count; - HvCall3Ret16(HvCallPciBarLoad16, &ret, dsa, - BarOffset, 0); - } while (CheckReturnCode("RDW", DevNode, &retry, ret.rc) != 0); - - return swab16((u16)ret.value); -} -EXPORT_SYMBOL(iSeries_Read_Word); - -u32 iSeries_Read_Long(const volatile void __iomem *IoAddress) -{ - u64 BarOffset; - u64 dsa; - int retry = 0; - struct HvCallPci_LoadReturn ret; - struct iSeries_Device_Node *DevNode = - xlate_iomm_address(IoAddress, &dsa, &BarOffset); - - if (DevNode == NULL) { - static unsigned long last_jiffies; - static int num_printed; - - if ((jiffies - last_jiffies) > 60 * HZ) { - last_jiffies = jiffies; - num_printed = 0; - } - if (num_printed++ < 10) - printk(KERN_ERR "iSeries_Read_Long: invalid access at IO address %p\n", IoAddress); - return 0xffffffff; - } - do { - ++Pci_Io_Read_Count; - HvCall3Ret16(HvCallPciBarLoad32, &ret, dsa, - BarOffset, 0); - } while (CheckReturnCode("RDL", DevNode, &retry, ret.rc) != 0); - - return swab32((u32)ret.value); -} -EXPORT_SYMBOL(iSeries_Read_Long); - -/* - * Write MM I/O Instructions for the iSeries - * - * iSeries_Write_Byte = Write Byte (8 bit) - * iSeries_Write_Word = Write Word(16 bit) - * iSeries_Write_Long = Write Long(32 bit) - */ -void iSeries_Write_Byte(u8 data, volatile void __iomem *IoAddress) -{ - u64 BarOffset; - u64 dsa; - int retry = 0; - u64 rc; - struct iSeries_Device_Node *DevNode = - xlate_iomm_address(IoAddress, &dsa, &BarOffset); - - if (DevNode == NULL) { - static unsigned long last_jiffies; - static int num_printed; - - if ((jiffies - last_jiffies) > 60 * HZ) { - last_jiffies = jiffies; - num_printed = 0; - } - if (num_printed++ < 10) - printk(KERN_ERR "iSeries_Write_Byte: invalid access at IO address %p\n", IoAddress); - return; - } - do { - ++Pci_Io_Write_Count; - rc = HvCall4(HvCallPciBarStore8, dsa, BarOffset, data, 0); - } while (CheckReturnCode("WWB", DevNode, &retry, rc) != 0); -} -EXPORT_SYMBOL(iSeries_Write_Byte); - -void iSeries_Write_Word(u16 data, volatile void __iomem *IoAddress) -{ - u64 BarOffset; - u64 dsa; - int retry = 0; - u64 rc; - struct iSeries_Device_Node *DevNode = - xlate_iomm_address(IoAddress, &dsa, &BarOffset); - - if (DevNode == NULL) { - static unsigned long last_jiffies; - static int num_printed; - - if ((jiffies - last_jiffies) > 60 * HZ) { - last_jiffies = jiffies; - num_printed = 0; - } - if (num_printed++ < 10) - printk(KERN_ERR "iSeries_Write_Word: invalid access at IO address %p\n", IoAddress); - return; - } - do { - ++Pci_Io_Write_Count; - rc = HvCall4(HvCallPciBarStore16, dsa, BarOffset, swab16(data), 0); - } while (CheckReturnCode("WWW", DevNode, &retry, rc) != 0); -} -EXPORT_SYMBOL(iSeries_Write_Word); - -void iSeries_Write_Long(u32 data, volatile void __iomem *IoAddress) -{ - u64 BarOffset; - u64 dsa; - int retry = 0; - u64 rc; - struct iSeries_Device_Node *DevNode = - xlate_iomm_address(IoAddress, &dsa, &BarOffset); - - if (DevNode == NULL) { - static unsigned long last_jiffies; - static int num_printed; - - if ((jiffies - last_jiffies) > 60 * HZ) { - last_jiffies = jiffies; - num_printed = 0; - } - if (num_printed++ < 10) - printk(KERN_ERR "iSeries_Write_Long: invalid access at IO address %p\n", IoAddress); - return; - } - do { - ++Pci_Io_Write_Count; - rc = HvCall4(HvCallPciBarStore32, dsa, BarOffset, swab32(data), 0); - } while (CheckReturnCode("WWL", DevNode, &retry, rc) != 0); -} -EXPORT_SYMBOL(iSeries_Write_Long); diff --git a/arch/ppc64/kernel/iSeries_proc.c b/arch/ppc64/kernel/iSeries_proc.c deleted file mode 100644 index 0fe3116eba29..000000000000 --- a/arch/ppc64/kernel/iSeries_proc.c +++ /dev/null @@ -1,113 +0,0 @@ -/* - * iSeries_proc.c - * Copyright (C) 2001 Kyle A. Lucke IBM Corporation - * Copyright (C) 2001 Mike Corrigan & Dave Engebretsen IBM Corporation - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - */ -#include <linux/init.h> -#include <linux/proc_fs.h> -#include <linux/seq_file.h> -#include <linux/param.h> /* for HZ */ -#include <asm/paca.h> -#include <asm/processor.h> -#include <asm/time.h> -#include <asm/lppaca.h> -#include <asm/iSeries/ItLpQueue.h> -#include <asm/iSeries/HvCallXm.h> -#include <asm/iSeries/IoHriMainStore.h> -#include <asm/iSeries/IoHriProcessorVpd.h> - -static int __init iseries_proc_create(void) -{ - struct proc_dir_entry *e = proc_mkdir("iSeries", 0); - if (!e) - return 1; - - return 0; -} -core_initcall(iseries_proc_create); - -static unsigned long startTitan = 0; -static unsigned long startTb = 0; - -static int proc_titantod_show(struct seq_file *m, void *v) -{ - unsigned long tb0, titan_tod; - - tb0 = get_tb(); - titan_tod = HvCallXm_loadTod(); - - seq_printf(m, "Titan\n" ); - seq_printf(m, " time base = %016lx\n", tb0); - seq_printf(m, " titan tod = %016lx\n", titan_tod); - seq_printf(m, " xProcFreq = %016x\n", - xIoHriProcessorVpd[0].xProcFreq); - seq_printf(m, " xTimeBaseFreq = %016x\n", - xIoHriProcessorVpd[0].xTimeBaseFreq); - seq_printf(m, " tb_ticks_per_jiffy = %lu\n", tb_ticks_per_jiffy); - seq_printf(m, " tb_ticks_per_usec = %lu\n", tb_ticks_per_usec); - - if (!startTitan) { - startTitan = titan_tod; - startTb = tb0; - } else { - unsigned long titan_usec = (titan_tod - startTitan) >> 12; - unsigned long tb_ticks = (tb0 - startTb); - unsigned long titan_jiffies = titan_usec / (1000000/HZ); - unsigned long titan_jiff_usec = titan_jiffies * (1000000/HZ); - unsigned long titan_jiff_rem_usec = titan_usec - titan_jiff_usec; - unsigned long tb_jiffies = tb_ticks / tb_ticks_per_jiffy; - unsigned long tb_jiff_ticks = tb_jiffies * tb_ticks_per_jiffy; - unsigned long tb_jiff_rem_ticks = tb_ticks - tb_jiff_ticks; - unsigned long tb_jiff_rem_usec = tb_jiff_rem_ticks / tb_ticks_per_usec; - unsigned long new_tb_ticks_per_jiffy = (tb_ticks * (1000000/HZ))/titan_usec; - - seq_printf(m, " titan elapsed = %lu uSec\n", titan_usec); - seq_printf(m, " tb elapsed = %lu ticks\n", tb_ticks); - seq_printf(m, " titan jiffies = %lu.%04lu \n", titan_jiffies, - titan_jiff_rem_usec); - seq_printf(m, " tb jiffies = %lu.%04lu\n", tb_jiffies, - tb_jiff_rem_usec); - seq_printf(m, " new tb_ticks_per_jiffy = %lu\n", - new_tb_ticks_per_jiffy); - } - - return 0; -} - -static int proc_titantod_open(struct inode *inode, struct file *file) -{ - return single_open(file, proc_titantod_show, NULL); -} - -static struct file_operations proc_titantod_operations = { - .open = proc_titantod_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -static int __init iseries_proc_init(void) -{ - struct proc_dir_entry *e; - - e = create_proc_entry("iSeries/titanTod", S_IFREG|S_IRUGO, NULL); - if (e) - e->proc_fops = &proc_titantod_operations; - - return 0; -} -__initcall(iseries_proc_init); diff --git a/arch/ppc64/kernel/iSeries_setup.c b/arch/ppc64/kernel/iSeries_setup.c deleted file mode 100644 index 3ffefbbc6623..000000000000 --- a/arch/ppc64/kernel/iSeries_setup.c +++ /dev/null @@ -1,977 +0,0 @@ -/* - * Copyright (c) 2000 Mike Corrigan <mikejc@us.ibm.com> - * Copyright (c) 1999-2000 Grant Erickson <grant@lcse.umn.edu> - * - * Module name: iSeries_setup.c - * - * Description: - * Architecture- / platform-specific boot-time initialization code for - * the IBM iSeries LPAR. Adapted from original code by Grant Erickson and - * code by Gary Thomas, Cort Dougan <cort@fsmlabs.com>, and Dan Malek - * <dan@net4x.com>. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#undef DEBUG - -#include <linux/config.h> -#include <linux/init.h> -#include <linux/threads.h> -#include <linux/smp.h> -#include <linux/param.h> -#include <linux/string.h> -#include <linux/initrd.h> -#include <linux/seq_file.h> -#include <linux/kdev_t.h> -#include <linux/major.h> -#include <linux/root_dev.h> - -#include <asm/processor.h> -#include <asm/machdep.h> -#include <asm/page.h> -#include <asm/mmu.h> -#include <asm/pgtable.h> -#include <asm/mmu_context.h> -#include <asm/cputable.h> -#include <asm/sections.h> -#include <asm/iommu.h> -#include <asm/firmware.h> - -#include <asm/time.h> -#include "iSeries_setup.h" -#include <asm/naca.h> -#include <asm/paca.h> -#include <asm/cache.h> -#include <asm/sections.h> -#include <asm/abs_addr.h> -#include <asm/iSeries/HvCallHpt.h> -#include <asm/iSeries/HvLpConfig.h> -#include <asm/iSeries/HvCallEvent.h> -#include <asm/iSeries/HvCallSm.h> -#include <asm/iSeries/HvCallXm.h> -#include <asm/iSeries/ItLpQueue.h> -#include <asm/iSeries/IoHriMainStore.h> -#include <asm/iSeries/mf.h> -#include <asm/iSeries/HvLpEvent.h> -#include <asm/iSeries/iSeries_irq.h> -#include <asm/iSeries/IoHriProcessorVpd.h> -#include <asm/iSeries/ItVpdAreas.h> -#include <asm/iSeries/LparMap.h> - -extern void hvlog(char *fmt, ...); - -#ifdef DEBUG -#define DBG(fmt...) hvlog(fmt) -#else -#define DBG(fmt...) -#endif - -/* Function Prototypes */ -extern void ppcdbg_initialize(void); - -static void build_iSeries_Memory_Map(void); -static void setup_iSeries_cache_sizes(void); -static void iSeries_bolt_kernel(unsigned long saddr, unsigned long eaddr); -#ifdef CONFIG_PCI -extern void iSeries_pci_final_fixup(void); -#else -static void iSeries_pci_final_fixup(void) { } -#endif - -/* Global Variables */ -static unsigned long procFreqHz; -static unsigned long procFreqMhz; -static unsigned long procFreqMhzHundreths; - -static unsigned long tbFreqHz; -static unsigned long tbFreqMhz; -static unsigned long tbFreqMhzHundreths; - -int piranha_simulator; - -extern int rd_size; /* Defined in drivers/block/rd.c */ -extern unsigned long klimit; -extern unsigned long embedded_sysmap_start; -extern unsigned long embedded_sysmap_end; - -extern unsigned long iSeries_recal_tb; -extern unsigned long iSeries_recal_titan; - -static int mf_initialized; - -struct MemoryBlock { - unsigned long absStart; - unsigned long absEnd; - unsigned long logicalStart; - unsigned long logicalEnd; -}; - -/* - * Process the main store vpd to determine where the holes in memory are - * and return the number of physical blocks and fill in the array of - * block data. - */ -static unsigned long iSeries_process_Condor_mainstore_vpd( - struct MemoryBlock *mb_array, unsigned long max_entries) -{ - unsigned long holeFirstChunk, holeSizeChunks; - unsigned long numMemoryBlocks = 1; - struct IoHriMainStoreSegment4 *msVpd = - (struct IoHriMainStoreSegment4 *)xMsVpd; - unsigned long holeStart = msVpd->nonInterleavedBlocksStartAdr; - unsigned long holeEnd = msVpd->nonInterleavedBlocksEndAdr; - unsigned long holeSize = holeEnd - holeStart; - - printk("Mainstore_VPD: Condor\n"); - /* - * Determine if absolute memory has any - * holes so that we can interpret the - * access map we get back from the hypervisor - * correctly. - */ - mb_array[0].logicalStart = 0; - mb_array[0].logicalEnd = 0x100000000; - mb_array[0].absStart = 0; - mb_array[0].absEnd = 0x100000000; - - if (holeSize) { - numMemoryBlocks = 2; - holeStart = holeStart & 0x000fffffffffffff; - holeStart = addr_to_chunk(holeStart); - holeFirstChunk = holeStart; - holeSize = addr_to_chunk(holeSize); - holeSizeChunks = holeSize; - printk( "Main store hole: start chunk = %0lx, size = %0lx chunks\n", - holeFirstChunk, holeSizeChunks ); - mb_array[0].logicalEnd = holeFirstChunk; - mb_array[0].absEnd = holeFirstChunk; - mb_array[1].logicalStart = holeFirstChunk; - mb_array[1].logicalEnd = 0x100000000 - holeSizeChunks; - mb_array[1].absStart = holeFirstChunk + holeSizeChunks; - mb_array[1].absEnd = 0x100000000; - } - return numMemoryBlocks; -} - -#define MaxSegmentAreas 32 -#define MaxSegmentAdrRangeBlocks 128 -#define MaxAreaRangeBlocks 4 - -static unsigned long iSeries_process_Regatta_mainstore_vpd( - struct MemoryBlock *mb_array, unsigned long max_entries) -{ - struct IoHriMainStoreSegment5 *msVpdP = - (struct IoHriMainStoreSegment5 *)xMsVpd; - unsigned long numSegmentBlocks = 0; - u32 existsBits = msVpdP->msAreaExists; - unsigned long area_num; - - printk("Mainstore_VPD: Regatta\n"); - - for (area_num = 0; area_num < MaxSegmentAreas; ++area_num ) { - unsigned long numAreaBlocks; - struct IoHriMainStoreArea4 *currentArea; - - if (existsBits & 0x80000000) { - unsigned long block_num; - - currentArea = &msVpdP->msAreaArray[area_num]; - numAreaBlocks = currentArea->numAdrRangeBlocks; - printk("ms_vpd: processing area %2ld blocks=%ld", - area_num, numAreaBlocks); - for (block_num = 0; block_num < numAreaBlocks; - ++block_num ) { - /* Process an address range block */ - struct MemoryBlock tempBlock; - unsigned long i; - - tempBlock.absStart = - (unsigned long)currentArea->xAdrRangeBlock[block_num].blockStart; - tempBlock.absEnd = - (unsigned long)currentArea->xAdrRangeBlock[block_num].blockEnd; - tempBlock.logicalStart = 0; - tempBlock.logicalEnd = 0; - printk("\n block %ld absStart=%016lx absEnd=%016lx", - block_num, tempBlock.absStart, - tempBlock.absEnd); - - for (i = 0; i < numSegmentBlocks; ++i) { - if (mb_array[i].absStart == - tempBlock.absStart) - break; - } - if (i == numSegmentBlocks) { - if (numSegmentBlocks == max_entries) - panic("iSeries_process_mainstore_vpd: too many memory blocks"); - mb_array[numSegmentBlocks] = tempBlock; - ++numSegmentBlocks; - } else - printk(" (duplicate)"); - } - printk("\n"); - } - existsBits <<= 1; - } - /* Now sort the blocks found into ascending sequence */ - if (numSegmentBlocks > 1) { - unsigned long m, n; - - for (m = 0; m < numSegmentBlocks - 1; ++m) { - for (n = numSegmentBlocks - 1; m < n; --n) { - if (mb_array[n].absStart < - mb_array[n-1].absStart) { - struct MemoryBlock tempBlock; - - tempBlock = mb_array[n]; - mb_array[n] = mb_array[n-1]; - mb_array[n-1] = tempBlock; - } - } - } - } - /* - * Assign "logical" addresses to each block. These - * addresses correspond to the hypervisor "bitmap" space. - * Convert all addresses into units of 256K chunks. - */ - { - unsigned long i, nextBitmapAddress; - - printk("ms_vpd: %ld sorted memory blocks\n", numSegmentBlocks); - nextBitmapAddress = 0; - for (i = 0; i < numSegmentBlocks; ++i) { - unsigned long length = mb_array[i].absEnd - - mb_array[i].absStart; - - mb_array[i].logicalStart = nextBitmapAddress; - mb_array[i].logicalEnd = nextBitmapAddress + length; - nextBitmapAddress += length; - printk(" Bitmap range: %016lx - %016lx\n" - " Absolute range: %016lx - %016lx\n", - mb_array[i].logicalStart, - mb_array[i].logicalEnd, - mb_array[i].absStart, mb_array[i].absEnd); - mb_array[i].absStart = addr_to_chunk(mb_array[i].absStart & - 0x000fffffffffffff); - mb_array[i].absEnd = addr_to_chunk(mb_array[i].absEnd & - 0x000fffffffffffff); - mb_array[i].logicalStart = - addr_to_chunk(mb_array[i].logicalStart); - mb_array[i].logicalEnd = addr_to_chunk(mb_array[i].logicalEnd); - } - } - - return numSegmentBlocks; -} - -static unsigned long iSeries_process_mainstore_vpd(struct MemoryBlock *mb_array, - unsigned long max_entries) -{ - unsigned long i; - unsigned long mem_blocks = 0; - - if (cpu_has_feature(CPU_FTR_SLB)) - mem_blocks = iSeries_process_Regatta_mainstore_vpd(mb_array, - max_entries); - else - mem_blocks = iSeries_process_Condor_mainstore_vpd(mb_array, - max_entries); - - printk("Mainstore_VPD: numMemoryBlocks = %ld \n", mem_blocks); - for (i = 0; i < mem_blocks; ++i) { - printk("Mainstore_VPD: block %3ld logical chunks %016lx - %016lx\n" - " abs chunks %016lx - %016lx\n", - i, mb_array[i].logicalStart, mb_array[i].logicalEnd, - mb_array[i].absStart, mb_array[i].absEnd); - } - return mem_blocks; -} - -static void __init iSeries_get_cmdline(void) -{ - char *p, *q; - - /* copy the command line parameter from the primary VSP */ - HvCallEvent_dmaToSp(cmd_line, 2 * 64* 1024, 256, - HvLpDma_Direction_RemoteToLocal); - - p = cmd_line; - q = cmd_line + 255; - while(p < q) { - if (!*p || *p == '\n') - break; - ++p; - } - *p = 0; -} - -static void __init iSeries_init_early(void) -{ - extern unsigned long memory_limit; - - DBG(" -> iSeries_init_early()\n"); - - ppc64_firmware_features = FW_FEATURE_ISERIES; - - ppcdbg_initialize(); - -#if defined(CONFIG_BLK_DEV_INITRD) - /* - * If the init RAM disk has been configured and there is - * a non-zero starting address for it, set it up - */ - if (naca.xRamDisk) { - initrd_start = (unsigned long)__va(naca.xRamDisk); - initrd_end = initrd_start + naca.xRamDiskSize * PAGE_SIZE; - initrd_below_start_ok = 1; // ramdisk in kernel space - ROOT_DEV = Root_RAM0; - if (((rd_size * 1024) / PAGE_SIZE) < naca.xRamDiskSize) - rd_size = (naca.xRamDiskSize * PAGE_SIZE) / 1024; - } else -#endif /* CONFIG_BLK_DEV_INITRD */ - { - /* ROOT_DEV = MKDEV(VIODASD_MAJOR, 1); */ - } - - iSeries_recal_tb = get_tb(); - iSeries_recal_titan = HvCallXm_loadTod(); - - /* - * Cache sizes must be initialized before hpte_init_iSeries is called - * as the later need them for flush_icache_range() - */ - setup_iSeries_cache_sizes(); - - /* - * Initialize the hash table management pointers - */ - hpte_init_iSeries(); - - /* - * Initialize the DMA/TCE management - */ - iommu_init_early_iSeries(); - - /* - * Initialize the table which translate Linux physical addresses to - * AS/400 absolute addresses - */ - build_iSeries_Memory_Map(); - - iSeries_get_cmdline(); - - /* Save unparsed command line copy for /proc/cmdline */ - strlcpy(saved_command_line, cmd_line, COMMAND_LINE_SIZE); - - /* Parse early parameters, in particular mem=x */ - parse_early_param(); - - if (memory_limit) { - if (memory_limit < systemcfg->physicalMemorySize) - systemcfg->physicalMemorySize = memory_limit; - else { - printk("Ignoring mem=%lu >= ram_top.\n", memory_limit); - memory_limit = 0; - } - } - - /* Bolt kernel mappings for all of memory (or just a bit if we've got a limit) */ - iSeries_bolt_kernel(0, systemcfg->physicalMemorySize); - - lmb_init(); - lmb_add(0, systemcfg->physicalMemorySize); - lmb_analyze(); - lmb_reserve(0, __pa(klimit)); - - /* Initialize machine-dependency vectors */ -#ifdef CONFIG_SMP - smp_init_iSeries(); -#endif - if (itLpNaca.xPirEnvironMode == 0) - piranha_simulator = 1; - - /* Associate Lp Event Queue 0 with processor 0 */ - HvCallEvent_setLpEventQueueInterruptProc(0, 0); - - mf_init(); - mf_initialized = 1; - mb(); - - /* If we were passed an initrd, set the ROOT_DEV properly if the values - * look sensible. If not, clear initrd reference. - */ -#ifdef CONFIG_BLK_DEV_INITRD - if (initrd_start >= KERNELBASE && initrd_end >= KERNELBASE && - initrd_end > initrd_start) - ROOT_DEV = Root_RAM0; - else - initrd_start = initrd_end = 0; -#endif /* CONFIG_BLK_DEV_INITRD */ - - DBG(" <- iSeries_init_early()\n"); -} - -struct mschunks_map mschunks_map = { - /* XXX We don't use these, but Piranha might need them. */ - .chunk_size = MSCHUNKS_CHUNK_SIZE, - .chunk_shift = MSCHUNKS_CHUNK_SHIFT, - .chunk_mask = MSCHUNKS_OFFSET_MASK, -}; -EXPORT_SYMBOL(mschunks_map); - -void mschunks_alloc(unsigned long num_chunks) -{ - klimit = _ALIGN(klimit, sizeof(u32)); - mschunks_map.mapping = (u32 *)klimit; - klimit += num_chunks * sizeof(u32); - mschunks_map.num_chunks = num_chunks; -} - -/* - * The iSeries may have very large memories ( > 128 GB ) and a partition - * may get memory in "chunks" that may be anywhere in the 2**52 real - * address space. The chunks are 256K in size. To map this to the - * memory model Linux expects, the AS/400 specific code builds a - * translation table to translate what Linux thinks are "physical" - * addresses to the actual real addresses. This allows us to make - * it appear to Linux that we have contiguous memory starting at - * physical address zero while in fact this could be far from the truth. - * To avoid confusion, I'll let the words physical and/or real address - * apply to the Linux addresses while I'll use "absolute address" to - * refer to the actual hardware real address. - * - * build_iSeries_Memory_Map gets information from the Hypervisor and - * looks at the Main Store VPD to determine the absolute addresses - * of the memory that has been assigned to our partition and builds - * a table used to translate Linux's physical addresses to these - * absolute addresses. Absolute addresses are needed when - * communicating with the hypervisor (e.g. to build HPT entries) - */ - -static void __init build_iSeries_Memory_Map(void) -{ - u32 loadAreaFirstChunk, loadAreaLastChunk, loadAreaSize; - u32 nextPhysChunk; - u32 hptFirstChunk, hptLastChunk, hptSizeChunks, hptSizePages; - u32 num_ptegs; - u32 totalChunks,moreChunks; - u32 currChunk, thisChunk, absChunk; - u32 currDword; - u32 chunkBit; - u64 map; - struct MemoryBlock mb[32]; - unsigned long numMemoryBlocks, curBlock; - - /* Chunk size on iSeries is 256K bytes */ - totalChunks = (u32)HvLpConfig_getMsChunks(); - mschunks_alloc(totalChunks); - - /* - * Get absolute address of our load area - * and map it to physical address 0 - * This guarantees that the loadarea ends up at physical 0 - * otherwise, it might not be returned by PLIC as the first - * chunks - */ - - loadAreaFirstChunk = (u32)addr_to_chunk(itLpNaca.xLoadAreaAddr); - loadAreaSize = itLpNaca.xLoadAreaChunks; - - /* - * Only add the pages already mapped here. - * Otherwise we might add the hpt pages - * The rest of the pages of the load area - * aren't in the HPT yet and can still - * be assigned an arbitrary physical address - */ - if ((loadAreaSize * 64) > HvPagesToMap) - loadAreaSize = HvPagesToMap / 64; - - loadAreaLastChunk = loadAreaFirstChunk + loadAreaSize - 1; - - /* - * TODO Do we need to do something if the HPT is in the 64MB load area? - * This would be required if the itLpNaca.xLoadAreaChunks includes - * the HPT size - */ - - printk("Mapping load area - physical addr = 0000000000000000\n" - " absolute addr = %016lx\n", - chunk_to_addr(loadAreaFirstChunk)); - printk("Load area size %dK\n", loadAreaSize * 256); - - for (nextPhysChunk = 0; nextPhysChunk < loadAreaSize; ++nextPhysChunk) - mschunks_map.mapping[nextPhysChunk] = - loadAreaFirstChunk + nextPhysChunk; - - /* - * Get absolute address of our HPT and remember it so - * we won't map it to any physical address - */ - hptFirstChunk = (u32)addr_to_chunk(HvCallHpt_getHptAddress()); - hptSizePages = (u32)HvCallHpt_getHptPages(); - hptSizeChunks = hptSizePages >> (MSCHUNKS_CHUNK_SHIFT - PAGE_SHIFT); - hptLastChunk = hptFirstChunk + hptSizeChunks - 1; - - printk("HPT absolute addr = %016lx, size = %dK\n", - chunk_to_addr(hptFirstChunk), hptSizeChunks * 256); - - /* Fill in the hashed page table hash mask */ - num_ptegs = hptSizePages * - (PAGE_SIZE / (sizeof(hpte_t) * HPTES_PER_GROUP)); - htab_hash_mask = num_ptegs - 1; - - /* - * The actual hashed page table is in the hypervisor, - * we have no direct access - */ - htab_address = NULL; - - /* - * Determine if absolute memory has any - * holes so that we can interpret the - * access map we get back from the hypervisor - * correctly. - */ - numMemoryBlocks = iSeries_process_mainstore_vpd(mb, 32); - - /* - * Process the main store access map from the hypervisor - * to build up our physical -> absolute translation table - */ - curBlock = 0; - currChunk = 0; - currDword = 0; - moreChunks = totalChunks; - - while (moreChunks) { - map = HvCallSm_get64BitsOfAccessMap(itLpNaca.xLpIndex, - currDword); - thisChunk = currChunk; - while (map) { - chunkBit = map >> 63; - map <<= 1; - if (chunkBit) { - --moreChunks; - while (thisChunk >= mb[curBlock].logicalEnd) { - ++curBlock; - if (curBlock >= numMemoryBlocks) - panic("out of memory blocks"); - } - if (thisChunk < mb[curBlock].logicalStart) - panic("memory block error"); - - absChunk = mb[curBlock].absStart + - (thisChunk - mb[curBlock].logicalStart); - if (((absChunk < hptFirstChunk) || - (absChunk > hptLastChunk)) && - ((absChunk < loadAreaFirstChunk) || - (absChunk > loadAreaLastChunk))) { - mschunks_map.mapping[nextPhysChunk] = - absChunk; - ++nextPhysChunk; - } - } - ++thisChunk; - } - ++currDword; - currChunk += 64; - } - - /* - * main store size (in chunks) is - * totalChunks - hptSizeChunks - * which should be equal to - * nextPhysChunk - */ - systemcfg->physicalMemorySize = chunk_to_addr(nextPhysChunk); -} - -/* - * Set up the variables that describe the cache line sizes - * for this machine. - */ -static void __init setup_iSeries_cache_sizes(void) -{ - unsigned int i, n; - unsigned int procIx = get_paca()->lppaca.dyn_hv_phys_proc_index; - - systemcfg->icache_size = - ppc64_caches.isize = xIoHriProcessorVpd[procIx].xInstCacheSize * 1024; - systemcfg->icache_line_size = - ppc64_caches.iline_size = - xIoHriProcessorVpd[procIx].xInstCacheOperandSize; - systemcfg->dcache_size = - ppc64_caches.dsize = - xIoHriProcessorVpd[procIx].xDataL1CacheSizeKB * 1024; - systemcfg->dcache_line_size = - ppc64_caches.dline_size = - xIoHriProcessorVpd[procIx].xDataCacheOperandSize; - ppc64_caches.ilines_per_page = PAGE_SIZE / ppc64_caches.iline_size; - ppc64_caches.dlines_per_page = PAGE_SIZE / ppc64_caches.dline_size; - - i = ppc64_caches.iline_size; - n = 0; - while ((i = (i / 2))) - ++n; - ppc64_caches.log_iline_size = n; - - i = ppc64_caches.dline_size; - n = 0; - while ((i = (i / 2))) - ++n; - ppc64_caches.log_dline_size = n; - - printk("D-cache line size = %d\n", - (unsigned int)ppc64_caches.dline_size); - printk("I-cache line size = %d\n", - (unsigned int)ppc64_caches.iline_size); -} - -/* - * Create a pte. Used during initialization only. - */ -static void iSeries_make_pte(unsigned long va, unsigned long pa, - int mode) -{ - hpte_t local_hpte, rhpte; - unsigned long hash, vpn; - long slot; - - vpn = va >> PAGE_SHIFT; - hash = hpt_hash(vpn, 0); - - local_hpte.r = pa | mode; - local_hpte.v = ((va >> 23) << HPTE_V_AVPN_SHIFT) - | HPTE_V_BOLTED | HPTE_V_VALID; - - slot = HvCallHpt_findValid(&rhpte, vpn); - if (slot < 0) { - /* Must find space in primary group */ - panic("hash_page: hpte already exists\n"); - } - HvCallHpt_addValidate(slot, 0, &local_hpte); -} - -/* - * Bolt the kernel addr space into the HPT - */ -static void __init iSeries_bolt_kernel(unsigned long saddr, unsigned long eaddr) -{ - unsigned long pa; - unsigned long mode_rw = _PAGE_ACCESSED | _PAGE_COHERENT | PP_RWXX; - hpte_t hpte; - - for (pa = saddr; pa < eaddr ;pa += PAGE_SIZE) { - unsigned long ea = (unsigned long)__va(pa); - unsigned long vsid = get_kernel_vsid(ea); - unsigned long va = (vsid << 28) | (pa & 0xfffffff); - unsigned long vpn = va >> PAGE_SHIFT; - unsigned long slot = HvCallHpt_findValid(&hpte, vpn); - - /* Make non-kernel text non-executable */ - if (!in_kernel_text(ea)) - mode_rw |= HW_NO_EXEC; - - if (hpte.v & HPTE_V_VALID) { - /* HPTE exists, so just bolt it */ - HvCallHpt_setSwBits(slot, 0x10, 0); - /* And make sure the pp bits are correct */ - HvCallHpt_setPp(slot, PP_RWXX); - } else - /* No HPTE exists, so create a new bolted one */ - iSeries_make_pte(va, phys_to_abs(pa), mode_rw); - } -} - -/* - * Document me. - */ -static void __init iSeries_setup_arch(void) -{ - unsigned procIx = get_paca()->lppaca.dyn_hv_phys_proc_index; - - /* Add an eye catcher and the systemcfg layout version number */ - strcpy(systemcfg->eye_catcher, "SYSTEMCFG:PPC64"); - systemcfg->version.major = SYSTEMCFG_MAJOR; - systemcfg->version.minor = SYSTEMCFG_MINOR; - - /* Setup the Lp Event Queue */ - setup_hvlpevent_queue(); - - /* Compute processor frequency */ - procFreqHz = ((1UL << 34) * 1000000) / - xIoHriProcessorVpd[procIx].xProcFreq; - procFreqMhz = procFreqHz / 1000000; - procFreqMhzHundreths = (procFreqHz / 10000) - (procFreqMhz * 100); - ppc_proc_freq = procFreqHz; - - /* Compute time base frequency */ - tbFreqHz = ((1UL << 32) * 1000000) / - xIoHriProcessorVpd[procIx].xTimeBaseFreq; - tbFreqMhz = tbFreqHz / 1000000; - tbFreqMhzHundreths = (tbFreqHz / 10000) - (tbFreqMhz * 100); - ppc_tb_freq = tbFreqHz; - - printk("Max logical processors = %d\n", - itVpdAreas.xSlicMaxLogicalProcs); - printk("Max physical processors = %d\n", - itVpdAreas.xSlicMaxPhysicalProcs); - printk("Processor frequency = %lu.%02lu\n", procFreqMhz, - procFreqMhzHundreths); - printk("Time base frequency = %lu.%02lu\n", tbFreqMhz, - tbFreqMhzHundreths); - systemcfg->processor = xIoHriProcessorVpd[procIx].xPVR; - printk("Processor version = %x\n", systemcfg->processor); -} - -static void iSeries_get_cpuinfo(struct seq_file *m) -{ - seq_printf(m, "machine\t\t: 64-bit iSeries Logical Partition\n"); -} - -/* - * Document me. - * and Implement me. - */ -static int iSeries_get_irq(struct pt_regs *regs) -{ - /* -2 means ignore this interrupt */ - return -2; -} - -/* - * Document me. - */ -static void iSeries_restart(char *cmd) -{ - mf_reboot(); -} - -/* - * Document me. - */ -static void iSeries_power_off(void) -{ - mf_power_off(); -} - -/* - * Document me. - */ -static void iSeries_halt(void) -{ - mf_power_off(); -} - -/* - * void __init iSeries_calibrate_decr() - * - * Description: - * This routine retrieves the internal processor frequency from the VPD, - * and sets up the kernel timer decrementer based on that value. - * - */ -static void __init iSeries_calibrate_decr(void) -{ - unsigned long cyclesPerUsec; - struct div_result divres; - - /* Compute decrementer (and TB) frequency in cycles/sec */ - cyclesPerUsec = ppc_tb_freq / 1000000; - - /* - * Set the amount to refresh the decrementer by. This - * is the number of decrementer ticks it takes for - * 1/HZ seconds. - */ - tb_ticks_per_jiffy = ppc_tb_freq / HZ; - -#if 0 - /* TEST CODE FOR ADJTIME */ - tb_ticks_per_jiffy += tb_ticks_per_jiffy / 5000; - /* END OF TEST CODE */ -#endif - - /* - * tb_ticks_per_sec = freq; would give better accuracy - * but tb_ticks_per_sec = tb_ticks_per_jiffy*HZ; assures - * that jiffies (and xtime) will match the time returned - * by do_gettimeofday. - */ - tb_ticks_per_sec = tb_ticks_per_jiffy * HZ; - tb_ticks_per_usec = cyclesPerUsec; - tb_to_us = mulhwu_scale_factor(ppc_tb_freq, 1000000); - div128_by_32(1024 * 1024, 0, tb_ticks_per_sec, &divres); - tb_to_xs = divres.result_low; - setup_default_decr(); -} - -static void __init iSeries_progress(char * st, unsigned short code) -{ - printk("Progress: [%04x] - %s\n", (unsigned)code, st); - if (!piranha_simulator && mf_initialized) { - if (code != 0xffff) - mf_display_progress(code); - else - mf_clear_src(); - } -} - -static void __init iSeries_fixup_klimit(void) -{ - /* - * Change klimit to take into account any ram disk - * that may be included - */ - if (naca.xRamDisk) - klimit = KERNELBASE + (u64)naca.xRamDisk + - (naca.xRamDiskSize * PAGE_SIZE); - else { - /* - * No ram disk was included - check and see if there - * was an embedded system map. Change klimit to take - * into account any embedded system map - */ - if (embedded_sysmap_end) - klimit = KERNELBASE + ((embedded_sysmap_end + 4095) & - 0xfffffffffffff000); - } -} - -static int __init iSeries_src_init(void) -{ - /* clear the progress line */ - ppc_md.progress(" ", 0xffff); - return 0; -} - -late_initcall(iSeries_src_init); - -static inline void process_iSeries_events(void) -{ - asm volatile ("li 0,0x5555; sc" : : : "r0", "r3"); -} - -static void yield_shared_processor(void) -{ - unsigned long tb; - - HvCall_setEnabledInterrupts(HvCall_MaskIPI | - HvCall_MaskLpEvent | - HvCall_MaskLpProd | - HvCall_MaskTimeout); - - tb = get_tb(); - /* Compute future tb value when yield should expire */ - HvCall_yieldProcessor(HvCall_YieldTimed, tb+tb_ticks_per_jiffy); - - /* - * The decrementer stops during the yield. Force a fake decrementer - * here and let the timer_interrupt code sort out the actual time. - */ - get_paca()->lppaca.int_dword.fields.decr_int = 1; - process_iSeries_events(); -} - -static int iseries_shared_idle(void) -{ - while (1) { - while (!need_resched() && !hvlpevent_is_pending()) { - local_irq_disable(); - ppc64_runlatch_off(); - - /* Recheck with irqs off */ - if (!need_resched() && !hvlpevent_is_pending()) - yield_shared_processor(); - - HMT_medium(); - local_irq_enable(); - } - - ppc64_runlatch_on(); - - if (hvlpevent_is_pending()) - process_iSeries_events(); - - schedule(); - } - - return 0; -} - -static int iseries_dedicated_idle(void) -{ - long oldval; - - while (1) { - oldval = test_and_clear_thread_flag(TIF_NEED_RESCHED); - - if (!oldval) { - set_thread_flag(TIF_POLLING_NRFLAG); - - while (!need_resched()) { - ppc64_runlatch_off(); - HMT_low(); - - if (hvlpevent_is_pending()) { - HMT_medium(); - ppc64_runlatch_on(); - process_iSeries_events(); - } - } - - HMT_medium(); - clear_thread_flag(TIF_POLLING_NRFLAG); - } else { - set_need_resched(); - } - - ppc64_runlatch_on(); - schedule(); - } - - return 0; -} - -#ifndef CONFIG_PCI -void __init iSeries_init_IRQ(void) { } -#endif - -void __init iSeries_early_setup(void) -{ - iSeries_fixup_klimit(); - - ppc_md.setup_arch = iSeries_setup_arch; - ppc_md.get_cpuinfo = iSeries_get_cpuinfo; - ppc_md.init_IRQ = iSeries_init_IRQ; - ppc_md.get_irq = iSeries_get_irq; - ppc_md.init_early = iSeries_init_early, - - ppc_md.pcibios_fixup = iSeries_pci_final_fixup; - - ppc_md.restart = iSeries_restart; - ppc_md.power_off = iSeries_power_off; - ppc_md.halt = iSeries_halt; - - ppc_md.get_boot_time = iSeries_get_boot_time; - ppc_md.set_rtc_time = iSeries_set_rtc_time; - ppc_md.get_rtc_time = iSeries_get_rtc_time; - ppc_md.calibrate_decr = iSeries_calibrate_decr; - ppc_md.progress = iSeries_progress; - - /* XXX Implement enable_pmcs for iSeries */ - - if (get_paca()->lppaca.shared_proc) { - ppc_md.idle_loop = iseries_shared_idle; - printk(KERN_INFO "Using shared processor idle loop\n"); - } else { - ppc_md.idle_loop = iseries_dedicated_idle; - printk(KERN_INFO "Using dedicated idle loop\n"); - } -} - diff --git a/arch/ppc64/kernel/iSeries_setup.h b/arch/ppc64/kernel/iSeries_setup.h deleted file mode 100644 index c6eb29a245ac..000000000000 --- a/arch/ppc64/kernel/iSeries_setup.h +++ /dev/null @@ -1,26 +0,0 @@ -/* - * Copyright (c) 2000 Mike Corrigan <mikejc@us.ibm.com> - * Copyright (c) 1999-2000 Grant Erickson <grant@lcse.umn.edu> - * - * Module name: as400_setup.h - * - * Description: - * Architecture- / platform-specific boot-time initialization code for - * the IBM AS/400 LPAR. Adapted from original code by Grant Erickson and - * code by Gary Thomas, Cort Dougan <cort@cs.nmt.edu>, and Dan Malek - * <dan@netx4.com>. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#ifndef __ISERIES_SETUP_H__ -#define __ISERIES_SETUP_H__ - -extern void iSeries_get_boot_time(struct rtc_time *tm); -extern int iSeries_set_rtc_time(struct rtc_time *tm); -extern void iSeries_get_rtc_time(struct rtc_time *tm); - -#endif /* __ISERIES_SETUP_H__ */ diff --git a/arch/ppc64/kernel/iSeries_smp.c b/arch/ppc64/kernel/iSeries_smp.c deleted file mode 100644 index f74386e31638..000000000000 --- a/arch/ppc64/kernel/iSeries_smp.c +++ /dev/null @@ -1,149 +0,0 @@ -/* - * SMP support for iSeries machines. - * - * Dave Engebretsen, Peter Bergner, and - * Mike Corrigan {engebret|bergner|mikec}@us.ibm.com - * - * Plus various changes from other IBM teams... - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#undef DEBUG - -#include <linux/config.h> -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/sched.h> -#include <linux/smp.h> -#include <linux/smp_lock.h> -#include <linux/interrupt.h> -#include <linux/kernel_stat.h> -#include <linux/delay.h> -#include <linux/init.h> -#include <linux/spinlock.h> -#include <linux/cache.h> -#include <linux/err.h> -#include <linux/sysdev.h> -#include <linux/cpu.h> - -#include <asm/ptrace.h> -#include <asm/atomic.h> -#include <asm/irq.h> -#include <asm/page.h> -#include <asm/pgtable.h> -#include <asm/io.h> -#include <asm/smp.h> -#include <asm/paca.h> -#include <asm/iSeries/HvCall.h> -#include <asm/time.h> -#include <asm/ppcdebug.h> -#include <asm/machdep.h> -#include <asm/cputable.h> -#include <asm/system.h> - -static unsigned long iSeries_smp_message[NR_CPUS]; - -void iSeries_smp_message_recv( struct pt_regs * regs ) -{ - int cpu = smp_processor_id(); - int msg; - - if ( num_online_cpus() < 2 ) - return; - - for ( msg = 0; msg < 4; ++msg ) - if ( test_and_clear_bit( msg, &iSeries_smp_message[cpu] ) ) - smp_message_recv( msg, regs ); -} - -static inline void smp_iSeries_do_message(int cpu, int msg) -{ - set_bit(msg, &iSeries_smp_message[cpu]); - HvCall_sendIPI(&(paca[cpu])); -} - -static void smp_iSeries_message_pass(int target, int msg) -{ - int i; - - if (target < NR_CPUS) - smp_iSeries_do_message(target, msg); - else { - for_each_online_cpu(i) { - if (target == MSG_ALL_BUT_SELF - && i == smp_processor_id()) - continue; - smp_iSeries_do_message(i, msg); - } - } -} - -static int smp_iSeries_numProcs(void) -{ - unsigned np, i; - - np = 0; - for (i=0; i < NR_CPUS; ++i) { - if (paca[i].lppaca.dyn_proc_status < 2) { - cpu_set(i, cpu_possible_map); - cpu_set(i, cpu_present_map); - cpu_set(i, cpu_sibling_map[i]); - ++np; - } - } - return np; -} - -static int smp_iSeries_probe(void) -{ - unsigned i; - unsigned np = 0; - - for (i=0; i < NR_CPUS; ++i) { - if (paca[i].lppaca.dyn_proc_status < 2) { - /*paca[i].active = 1;*/ - ++np; - } - } - - return np; -} - -static void smp_iSeries_kick_cpu(int nr) -{ - BUG_ON(nr < 0 || nr >= NR_CPUS); - - /* Verify that our partition has a processor nr */ - if (paca[nr].lppaca.dyn_proc_status >= 2) - return; - - /* The processor is currently spinning, waiting - * for the cpu_start field to become non-zero - * After we set cpu_start, the processor will - * continue on to secondary_start in iSeries_head.S - */ - paca[nr].cpu_start = 1; -} - -static void __devinit smp_iSeries_setup_cpu(int nr) -{ -} - -static struct smp_ops_t iSeries_smp_ops = { - .message_pass = smp_iSeries_message_pass, - .probe = smp_iSeries_probe, - .kick_cpu = smp_iSeries_kick_cpu, - .setup_cpu = smp_iSeries_setup_cpu, -}; - -/* This is called very early. */ -void __init smp_init_iSeries(void) -{ - smp_ops = &iSeries_smp_ops; - systemcfg->processorCount = smp_iSeries_numProcs(); -} - diff --git a/arch/ppc64/kernel/iSeries_vio.c b/arch/ppc64/kernel/iSeries_vio.c deleted file mode 100644 index 6b754b0c8344..000000000000 --- a/arch/ppc64/kernel/iSeries_vio.c +++ /dev/null @@ -1,155 +0,0 @@ -/* - * IBM PowerPC iSeries Virtual I/O Infrastructure Support. - * - * Copyright (c) 2005 Stephen Rothwell, IBM Corp. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#include <linux/types.h> -#include <linux/device.h> -#include <linux/init.h> - -#include <asm/vio.h> -#include <asm/iommu.h> -#include <asm/abs_addr.h> -#include <asm/page.h> -#include <asm/iSeries/vio.h> -#include <asm/iSeries/HvTypes.h> -#include <asm/iSeries/HvLpConfig.h> -#include <asm/iSeries/HvCallXm.h> - -struct device *iSeries_vio_dev = &vio_bus_device.dev; -EXPORT_SYMBOL(iSeries_vio_dev); - -static struct iommu_table veth_iommu_table; -static struct iommu_table vio_iommu_table; - -static void __init iommu_vio_init(void) -{ - struct iommu_table *t; - struct iommu_table_cb cb; - unsigned long cbp; - unsigned long itc_entries; - - cb.itc_busno = 255; /* Bus 255 is the virtual bus */ - cb.itc_virtbus = 0xff; /* Ask for virtual bus */ - - cbp = virt_to_abs(&cb); - HvCallXm_getTceTableParms(cbp); - - itc_entries = cb.itc_size * PAGE_SIZE / sizeof(union tce_entry); - veth_iommu_table.it_size = itc_entries / 2; - veth_iommu_table.it_busno = cb.itc_busno; - veth_iommu_table.it_offset = cb.itc_offset; - veth_iommu_table.it_index = cb.itc_index; - veth_iommu_table.it_type = TCE_VB; - veth_iommu_table.it_blocksize = 1; - - t = iommu_init_table(&veth_iommu_table); - - if (!t) - printk("Virtual Bus VETH TCE table failed.\n"); - - vio_iommu_table.it_size = itc_entries - veth_iommu_table.it_size; - vio_iommu_table.it_busno = cb.itc_busno; - vio_iommu_table.it_offset = cb.itc_offset + - veth_iommu_table.it_size; - vio_iommu_table.it_index = cb.itc_index; - vio_iommu_table.it_type = TCE_VB; - vio_iommu_table.it_blocksize = 1; - - t = iommu_init_table(&vio_iommu_table); - - if (!t) - printk("Virtual Bus VIO TCE table failed.\n"); -} - -/** - * vio_register_device_iseries: - Register a new iSeries vio device. - * @voidev: The device to register. - */ -static struct vio_dev *__init vio_register_device_iseries(char *type, - uint32_t unit_num) -{ - struct vio_dev *viodev; - - /* allocate a vio_dev for this device */ - viodev = kmalloc(sizeof(struct vio_dev), GFP_KERNEL); - if (!viodev) - return NULL; - memset(viodev, 0, sizeof(struct vio_dev)); - - snprintf(viodev->dev.bus_id, BUS_ID_SIZE, "%s%d", type, unit_num); - - viodev->name = viodev->dev.bus_id; - viodev->type = type; - viodev->unit_address = unit_num; - viodev->iommu_table = &vio_iommu_table; - if (vio_register_device(viodev) == NULL) { - kfree(viodev); - return NULL; - } - return viodev; -} - -void __init probe_bus_iseries(void) -{ - HvLpIndexMap vlan_map; - struct vio_dev *viodev; - int i; - - /* there is only one of each of these */ - vio_register_device_iseries("viocons", 0); - vio_register_device_iseries("vscsi", 0); - - vlan_map = HvLpConfig_getVirtualLanIndexMap(); - for (i = 0; i < HVMAXARCHITECTEDVIRTUALLANS; i++) { - if ((vlan_map & (0x8000 >> i)) == 0) - continue; - viodev = vio_register_device_iseries("vlan", i); - /* veth is special and has it own iommu_table */ - viodev->iommu_table = &veth_iommu_table; - } - for (i = 0; i < HVMAXARCHITECTEDVIRTUALDISKS; i++) - vio_register_device_iseries("viodasd", i); - for (i = 0; i < HVMAXARCHITECTEDVIRTUALCDROMS; i++) - vio_register_device_iseries("viocd", i); - for (i = 0; i < HVMAXARCHITECTEDVIRTUALTAPES; i++) - vio_register_device_iseries("viotape", i); -} - -/** - * vio_match_device_iseries: - Tell if a iSeries VIO device matches a - * vio_device_id - */ -static int vio_match_device_iseries(const struct vio_device_id *id, - const struct vio_dev *dev) -{ - return strncmp(dev->type, id->type, strlen(id->type)) == 0; -} - -static struct vio_bus_ops vio_bus_ops_iseries = { - .match = vio_match_device_iseries, -}; - -/** - * vio_bus_init_iseries: - Initialize the iSeries virtual IO bus - */ -static int __init vio_bus_init_iseries(void) -{ - int err; - - err = vio_bus_init(&vio_bus_ops_iseries); - if (err == 0) { - iommu_vio_init(); - vio_bus_device.iommu_table = &vio_iommu_table; - iSeries_vio_dev = &vio_bus_device.dev; - probe_bus_iseries(); - } - return err; -} - -__initcall(vio_bus_init_iseries); diff --git a/arch/ppc64/kernel/idle.c b/arch/ppc64/kernel/idle.c index 954395d42636..8abd2ad92832 100644 --- a/arch/ppc64/kernel/idle.c +++ b/arch/ppc64/kernel/idle.c @@ -31,7 +31,7 @@ extern void power4_idle(void); -int default_idle(void) +void default_idle(void) { long oldval; unsigned int cpu = smp_processor_id(); @@ -64,11 +64,9 @@ int default_idle(void) if (cpu_is_offline(cpu) && system_state == SYSTEM_RUNNING) cpu_die(); } - - return 0; } -int native_idle(void) +void native_idle(void) { while (1) { ppc64_runlatch_off(); @@ -85,8 +83,6 @@ int native_idle(void) system_state == SYSTEM_RUNNING) cpu_die(); } - - return 0; } void cpu_idle(void) diff --git a/arch/ppc64/kernel/idle_power4.S b/arch/ppc64/kernel/idle_power4.S deleted file mode 100644 index ca02afe2a795..000000000000 --- a/arch/ppc64/kernel/idle_power4.S +++ /dev/null @@ -1,79 +0,0 @@ -/* - * This file contains the power_save function for 6xx & 7xxx CPUs - * rewritten in assembler - * - * Warning ! This code assumes that if your machine has a 750fx - * it will have PLL 1 set to low speed mode (used during NAP/DOZE). - * if this is not the case some additional changes will have to - * be done to check a runtime var (a bit like powersave-nap) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/config.h> -#include <linux/threads.h> -#include <asm/processor.h> -#include <asm/page.h> -#include <asm/cputable.h> -#include <asm/thread_info.h> -#include <asm/ppc_asm.h> -#include <asm/asm-offsets.h> - -#undef DEBUG - - .text - -/* - * Here is the power_save_6xx function. This could eventually be - * split into several functions & changing the function pointer - * depending on the various features. - */ -_GLOBAL(power4_idle) -BEGIN_FTR_SECTION - blr -END_FTR_SECTION_IFCLR(CPU_FTR_CAN_NAP) - /* We must dynamically check for the NAP feature as it - * can be cleared by CPU init after the fixups are done - */ - LOADBASE(r3,cur_cpu_spec) - ld r4,cur_cpu_spec@l(r3) - ld r4,CPU_SPEC_FEATURES(r4) - andi. r0,r4,CPU_FTR_CAN_NAP - beqlr - /* Now check if user or arch enabled NAP mode */ - LOADBASE(r3,powersave_nap) - lwz r4,powersave_nap@l(r3) - cmpwi 0,r4,0 - beqlr - - /* Clear MSR:EE */ - mfmsr r7 - li r4,0 - ori r4,r4,MSR_EE - andc r0,r7,r4 - mtmsrd r0 - - /* Check current_thread_info()->flags */ - clrrdi r4,r1,THREAD_SHIFT - ld r4,TI_FLAGS(r4) - andi. r0,r4,_TIF_NEED_RESCHED - beq 1f - mtmsrd r7 /* out of line this ? */ - blr -1: - /* Go to NAP now */ -BEGIN_FTR_SECTION - DSSALL - sync -END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) - oris r7,r7,MSR_POW@h - sync - isync - mtmsrd r7 - isync - sync - blr - diff --git a/arch/ppc64/kernel/init_task.c b/arch/ppc64/kernel/init_task.c deleted file mode 100644 index 941043ae040f..000000000000 --- a/arch/ppc64/kernel/init_task.c +++ /dev/null @@ -1,36 +0,0 @@ -#include <linux/mm.h> -#include <linux/module.h> -#include <linux/sched.h> -#include <linux/init.h> -#include <linux/init_task.h> -#include <linux/fs.h> -#include <linux/mqueue.h> -#include <asm/uaccess.h> - -static struct fs_struct init_fs = INIT_FS; -static struct files_struct init_files = INIT_FILES; -static struct signal_struct init_signals = INIT_SIGNALS(init_signals); -static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand); -struct mm_struct init_mm = INIT_MM(init_mm); - -EXPORT_SYMBOL(init_mm); - -/* - * Initial thread structure. - * - * We need to make sure that this is 16384-byte aligned due to the - * way process stacks are handled. This is done by having a special - * "init_task" linker map entry.. - */ -union thread_union init_thread_union - __attribute__((__section__(".data.init_task"))) = - { INIT_THREAD_INFO(init_task) }; - -/* - * Initial task structure. - * - * All other task structs will be allocated on slabs in fork.c - */ -struct task_struct init_task = INIT_TASK(init_task); - -EXPORT_SYMBOL(init_task); diff --git a/arch/ppc64/kernel/kprobes.c b/arch/ppc64/kernel/kprobes.c index 9c6facc24f70..ed876a5178ae 100644 --- a/arch/ppc64/kernel/kprobes.c +++ b/arch/ppc64/kernel/kprobes.c @@ -395,7 +395,6 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self, if (post_kprobe_handler(args->regs)) ret = NOTIFY_STOP; break; - case DIE_GPF: case DIE_PAGE_FAULT: if (kprobe_running() && kprobe_fault_handler(args->regs, args->trapnr)) diff --git a/arch/ppc64/kernel/lmb.c b/arch/ppc64/kernel/lmb.c deleted file mode 100644 index 5adaca2ddc9d..000000000000 --- a/arch/ppc64/kernel/lmb.c +++ /dev/null @@ -1,299 +0,0 @@ -/* - * Procedures for interfacing to Open Firmware. - * - * Peter Bergner, IBM Corp. June 2001. - * Copyright (C) 2001 Peter Bergner. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/config.h> -#include <linux/kernel.h> -#include <linux/init.h> -#include <linux/bitops.h> -#include <asm/types.h> -#include <asm/page.h> -#include <asm/prom.h> -#include <asm/lmb.h> -#include <asm/abs_addr.h> - -struct lmb lmb; - -#undef DEBUG - -void lmb_dump_all(void) -{ -#ifdef DEBUG - unsigned long i; - - udbg_printf("lmb_dump_all:\n"); - udbg_printf(" memory.cnt = 0x%lx\n", - lmb.memory.cnt); - udbg_printf(" memory.size = 0x%lx\n", - lmb.memory.size); - for (i=0; i < lmb.memory.cnt ;i++) { - udbg_printf(" memory.region[0x%x].base = 0x%lx\n", - i, lmb.memory.region[i].base); - udbg_printf(" .size = 0x%lx\n", - lmb.memory.region[i].size); - } - - udbg_printf("\n reserved.cnt = 0x%lx\n", - lmb.reserved.cnt); - udbg_printf(" reserved.size = 0x%lx\n", - lmb.reserved.size); - for (i=0; i < lmb.reserved.cnt ;i++) { - udbg_printf(" reserved.region[0x%x].base = 0x%lx\n", - i, lmb.reserved.region[i].base); - udbg_printf(" .size = 0x%lx\n", - lmb.reserved.region[i].size); - } -#endif /* DEBUG */ -} - -static unsigned long __init -lmb_addrs_overlap(unsigned long base1, unsigned long size1, - unsigned long base2, unsigned long size2) -{ - return ((base1 < (base2+size2)) && (base2 < (base1+size1))); -} - -static long __init -lmb_addrs_adjacent(unsigned long base1, unsigned long size1, - unsigned long base2, unsigned long size2) -{ - if (base2 == base1 + size1) - return 1; - else if (base1 == base2 + size2) - return -1; - - return 0; -} - -static long __init -lmb_regions_adjacent(struct lmb_region *rgn, unsigned long r1, unsigned long r2) -{ - unsigned long base1 = rgn->region[r1].base; - unsigned long size1 = rgn->region[r1].size; - unsigned long base2 = rgn->region[r2].base; - unsigned long size2 = rgn->region[r2].size; - - return lmb_addrs_adjacent(base1, size1, base2, size2); -} - -/* Assumption: base addr of region 1 < base addr of region 2 */ -static void __init -lmb_coalesce_regions(struct lmb_region *rgn, unsigned long r1, unsigned long r2) -{ - unsigned long i; - - rgn->region[r1].size += rgn->region[r2].size; - for (i=r2; i < rgn->cnt-1; i++) { - rgn->region[i].base = rgn->region[i+1].base; - rgn->region[i].size = rgn->region[i+1].size; - } - rgn->cnt--; -} - -/* This routine called with relocation disabled. */ -void __init -lmb_init(void) -{ - /* Create a dummy zero size LMB which will get coalesced away later. - * This simplifies the lmb_add() code below... - */ - lmb.memory.region[0].base = 0; - lmb.memory.region[0].size = 0; - lmb.memory.cnt = 1; - - /* Ditto. */ - lmb.reserved.region[0].base = 0; - lmb.reserved.region[0].size = 0; - lmb.reserved.cnt = 1; -} - -/* This routine called with relocation disabled. */ -void __init -lmb_analyze(void) -{ - int i; - - lmb.memory.size = 0; - - for (i = 0; i < lmb.memory.cnt; i++) - lmb.memory.size += lmb.memory.region[i].size; -} - -/* This routine called with relocation disabled. */ -static long __init -lmb_add_region(struct lmb_region *rgn, unsigned long base, unsigned long size) -{ - unsigned long i, coalesced = 0; - long adjacent; - - /* First try and coalesce this LMB with another. */ - for (i=0; i < rgn->cnt; i++) { - unsigned long rgnbase = rgn->region[i].base; - unsigned long rgnsize = rgn->region[i].size; - - adjacent = lmb_addrs_adjacent(base,size,rgnbase,rgnsize); - if ( adjacent > 0 ) { - rgn->region[i].base -= size; - rgn->region[i].size += size; - coalesced++; - break; - } - else if ( adjacent < 0 ) { - rgn->region[i].size += size; - coalesced++; - break; - } - } - - if ((i < rgn->cnt-1) && lmb_regions_adjacent(rgn, i, i+1) ) { - lmb_coalesce_regions(rgn, i, i+1); - coalesced++; - } - - if ( coalesced ) { - return coalesced; - } else if ( rgn->cnt >= MAX_LMB_REGIONS ) { - return -1; - } - - /* Couldn't coalesce the LMB, so add it to the sorted table. */ - for (i=rgn->cnt-1; i >= 0; i--) { - if (base < rgn->region[i].base) { - rgn->region[i+1].base = rgn->region[i].base; - rgn->region[i+1].size = rgn->region[i].size; - } else { - rgn->region[i+1].base = base; - rgn->region[i+1].size = size; - break; - } - } - rgn->cnt++; - - return 0; -} - -/* This routine called with relocation disabled. */ -long __init -lmb_add(unsigned long base, unsigned long size) -{ - struct lmb_region *_rgn = &(lmb.memory); - - /* On pSeries LPAR systems, the first LMB is our RMO region. */ - if ( base == 0 ) - lmb.rmo_size = size; - - return lmb_add_region(_rgn, base, size); - -} - -long __init -lmb_reserve(unsigned long base, unsigned long size) -{ - struct lmb_region *_rgn = &(lmb.reserved); - - return lmb_add_region(_rgn, base, size); -} - -long __init -lmb_overlaps_region(struct lmb_region *rgn, unsigned long base, unsigned long size) -{ - unsigned long i; - - for (i=0; i < rgn->cnt; i++) { - unsigned long rgnbase = rgn->region[i].base; - unsigned long rgnsize = rgn->region[i].size; - if ( lmb_addrs_overlap(base,size,rgnbase,rgnsize) ) { - break; - } - } - - return (i < rgn->cnt) ? i : -1; -} - -unsigned long __init -lmb_alloc(unsigned long size, unsigned long align) -{ - return lmb_alloc_base(size, align, LMB_ALLOC_ANYWHERE); -} - -unsigned long __init -lmb_alloc_base(unsigned long size, unsigned long align, unsigned long max_addr) -{ - long i, j; - unsigned long base = 0; - - for (i=lmb.memory.cnt-1; i >= 0; i--) { - unsigned long lmbbase = lmb.memory.region[i].base; - unsigned long lmbsize = lmb.memory.region[i].size; - - if ( max_addr == LMB_ALLOC_ANYWHERE ) - base = _ALIGN_DOWN(lmbbase+lmbsize-size, align); - else if ( lmbbase < max_addr ) - base = _ALIGN_DOWN(min(lmbbase+lmbsize,max_addr)-size, align); - else - continue; - - while ( (lmbbase <= base) && - ((j = lmb_overlaps_region(&lmb.reserved,base,size)) >= 0) ) { - base = _ALIGN_DOWN(lmb.reserved.region[j].base-size, align); - } - - if ( (base != 0) && (lmbbase <= base) ) - break; - } - - if ( i < 0 ) - return 0; - - lmb_add_region(&lmb.reserved, base, size); - - return base; -} - -/* You must call lmb_analyze() before this. */ -unsigned long __init -lmb_phys_mem_size(void) -{ - return lmb.memory.size; -} - -unsigned long __init -lmb_end_of_DRAM(void) -{ - int idx = lmb.memory.cnt - 1; - - return (lmb.memory.region[idx].base + lmb.memory.region[idx].size); -} - -/* - * Truncate the lmb list to memory_limit if it's set - * You must call lmb_analyze() after this. - */ -void __init lmb_enforce_memory_limit(void) -{ - extern unsigned long memory_limit; - unsigned long i, limit; - - if (! memory_limit) - return; - - limit = memory_limit; - for (i = 0; i < lmb.memory.cnt; i++) { - if (limit > lmb.memory.region[i].size) { - limit -= lmb.memory.region[i].size; - continue; - } - - lmb.memory.region[i].size = limit; - lmb.memory.cnt = i + 1; - break; - } -} diff --git a/arch/ppc64/kernel/lparmap.c b/arch/ppc64/kernel/lparmap.c deleted file mode 100644 index b81de286df5e..000000000000 --- a/arch/ppc64/kernel/lparmap.c +++ /dev/null @@ -1,31 +0,0 @@ -/* - * Copyright (C) 2005 Stephen Rothwell IBM Corp. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#include <asm/mmu.h> -#include <asm/page.h> -#include <asm/iSeries/LparMap.h> - -const struct LparMap __attribute__((__section__(".text"))) xLparMap = { - .xNumberEsids = HvEsidsToMap, - .xNumberRanges = HvRangesToMap, - .xSegmentTableOffs = STAB0_PAGE, - - .xEsids = { - { .xKernelEsid = GET_ESID(KERNELBASE), - .xKernelVsid = KERNEL_VSID(KERNELBASE), }, - { .xKernelEsid = GET_ESID(VMALLOCBASE), - .xKernelVsid = KERNEL_VSID(VMALLOCBASE), }, - }, - - .xRanges = { - { .xPages = HvPagesToMap, - .xOffset = 0, - .xVPN = KERNEL_VSID(KERNELBASE) << (SID_SHIFT - PAGE_SHIFT), - }, - }, -}; diff --git a/arch/ppc64/kernel/maple_pci.c b/arch/ppc64/kernel/maple_pci.c index 1d297e0edfc0..633324b5e61b 100644 --- a/arch/ppc64/kernel/maple_pci.c +++ b/arch/ppc64/kernel/maple_pci.c @@ -23,8 +23,7 @@ #include <asm/pci-bridge.h> #include <asm/machdep.h> #include <asm/iommu.h> - -#include "pci.h" +#include <asm/ppc-pci.h> #ifdef DEBUG #define DBG(x...) printk(x) @@ -276,7 +275,7 @@ static void __init setup_u3_agp(struct pci_controller* hose) { /* On G5, we move AGP up to high bus number so we don't need * to reassign bus numbers for HT. If we ever have P2P bridges - * on AGP, we'll have to move pci_assign_all_busses to the + * on AGP, we'll have to move pci_assign_all_buses to the * pci_controller structure so we enable it for AGP and not for * HT childs. * We hard code the address because of the different size of @@ -360,7 +359,7 @@ static int __init add_bridge(struct device_node *dev) /* Interpret the "ranges" property */ /* This also maps the I/O region and sets isa_io/mem_base */ - pci_process_bridge_OF_ranges(hose, dev); + pci_process_bridge_OF_ranges(hose, dev, primary); pci_setup_phb_io(hose, primary); /* Fixup "bus-range" OF property */ diff --git a/arch/ppc64/kernel/maple_setup.c b/arch/ppc64/kernel/maple_setup.c index fc0567498a3a..a107ed69a355 100644 --- a/arch/ppc64/kernel/maple_setup.c +++ b/arch/ppc64/kernel/maple_setup.c @@ -59,8 +59,8 @@ #include <asm/time.h> #include <asm/of_device.h> #include <asm/lmb.h> - -#include "mpic.h" +#include <asm/mpic.h> +#include <asm/udbg.h> #ifdef DEBUG #define DBG(fmt...) udbg_printf(fmt) @@ -70,7 +70,7 @@ extern int maple_set_rtc_time(struct rtc_time *tm); extern void maple_get_rtc_time(struct rtc_time *tm); -extern void maple_get_boot_time(struct rtc_time *tm); +extern unsigned long maple_get_boot_time(void); extern void maple_calibrate_decr(void); extern void maple_pci_init(void); extern void maple_pcibios_fixup(void); diff --git a/arch/ppc64/kernel/maple_time.c b/arch/ppc64/kernel/maple_time.c index d65210abcd03..445cb7470bf5 100644 --- a/arch/ppc64/kernel/maple_time.c +++ b/arch/ppc64/kernel/maple_time.c @@ -156,8 +156,9 @@ int maple_set_rtc_time(struct rtc_time *tm) return 0; } -void __init maple_get_boot_time(struct rtc_time *tm) +unsigned long __init maple_get_boot_time(void) { + struct rtc_time tm; struct device_node *rtcs; rtcs = find_compatible_devices("rtc", "pnpPNP,b00"); @@ -170,6 +171,8 @@ void __init maple_get_boot_time(struct rtc_time *tm) "legacy address (0x%x)\n", maple_rtc_addr); } - maple_get_rtc_time(tm); + maple_get_rtc_time(&tm); + return mktime(tm.tm_year+1900, tm.tm_mon+1, tm.tm_mday, + tm.tm_hour, tm.tm_min, tm.tm_sec); } diff --git a/arch/ppc64/kernel/mf.c b/arch/ppc64/kernel/mf.c deleted file mode 100644 index ef4a338ebd01..000000000000 --- a/arch/ppc64/kernel/mf.c +++ /dev/null @@ -1,1281 +0,0 @@ -/* - * mf.c - * Copyright (C) 2001 Troy D. Armstrong IBM Corporation - * Copyright (C) 2004-2005 Stephen Rothwell IBM Corporation - * - * This modules exists as an interface between a Linux secondary partition - * running on an iSeries and the primary partition's Virtual Service - * Processor (VSP) object. The VSP has final authority over powering on/off - * all partitions in the iSeries. It also provides miscellaneous low-level - * machine facility type operations. - * - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - */ - -#include <linux/types.h> -#include <linux/errno.h> -#include <linux/kernel.h> -#include <linux/init.h> -#include <linux/completion.h> -#include <linux/delay.h> -#include <linux/dma-mapping.h> -#include <linux/bcd.h> - -#include <asm/time.h> -#include <asm/uaccess.h> -#include <asm/paca.h> -#include <asm/iSeries/vio.h> -#include <asm/iSeries/mf.h> -#include <asm/iSeries/HvLpConfig.h> -#include <asm/iSeries/ItLpQueue.h> - -/* - * This is the structure layout for the Machine Facilites LPAR event - * flows. - */ -struct vsp_cmd_data { - u64 token; - u16 cmd; - HvLpIndex lp_index; - u8 result_code; - u32 reserved; - union { - u64 state; /* GetStateOut */ - u64 ipl_type; /* GetIplTypeOut, Function02SelectIplTypeIn */ - u64 ipl_mode; /* GetIplModeOut, Function02SelectIplModeIn */ - u64 page[4]; /* GetSrcHistoryIn */ - u64 flag; /* GetAutoIplWhenPrimaryIplsOut, - SetAutoIplWhenPrimaryIplsIn, - WhiteButtonPowerOffIn, - Function08FastPowerOffIn, - IsSpcnRackPowerIncompleteOut */ - struct { - u64 token; - u64 address_type; - u64 side; - u32 length; - u32 offset; - } kern; /* SetKernelImageIn, GetKernelImageIn, - SetKernelCmdLineIn, GetKernelCmdLineIn */ - u32 length_out; /* GetKernelImageOut, GetKernelCmdLineOut */ - u8 reserved[80]; - } sub_data; -}; - -struct vsp_rsp_data { - struct completion com; - struct vsp_cmd_data *response; -}; - -struct alloc_data { - u16 size; - u16 type; - u32 count; - u16 reserved1; - u8 reserved2; - HvLpIndex target_lp; -}; - -struct ce_msg_data; - -typedef void (*ce_msg_comp_hdlr)(void *token, struct ce_msg_data *vsp_cmd_rsp); - -struct ce_msg_comp_data { - ce_msg_comp_hdlr handler; - void *token; -}; - -struct ce_msg_data { - u8 ce_msg[12]; - char reserved[4]; - struct ce_msg_comp_data *completion; -}; - -struct io_mf_lp_event { - struct HvLpEvent hp_lp_event; - u16 subtype_result_code; - u16 reserved1; - u32 reserved2; - union { - struct alloc_data alloc; - struct ce_msg_data ce_msg; - struct vsp_cmd_data vsp_cmd; - } data; -}; - -#define subtype_data(a, b, c, d) \ - (((a) << 24) + ((b) << 16) + ((c) << 8) + (d)) - -/* - * All outgoing event traffic is kept on a FIFO queue. The first - * pointer points to the one that is outstanding, and all new - * requests get stuck on the end. Also, we keep a certain number of - * preallocated pending events so that we can operate very early in - * the boot up sequence (before kmalloc is ready). - */ -struct pending_event { - struct pending_event *next; - struct io_mf_lp_event event; - MFCompleteHandler hdlr; - char dma_data[72]; - unsigned dma_data_length; - unsigned remote_address; -}; -static spinlock_t pending_event_spinlock; -static struct pending_event *pending_event_head; -static struct pending_event *pending_event_tail; -static struct pending_event *pending_event_avail; -static struct pending_event pending_event_prealloc[16]; - -/* - * Put a pending event onto the available queue, so it can get reused. - * Attention! You must have the pending_event_spinlock before calling! - */ -static void free_pending_event(struct pending_event *ev) -{ - if (ev != NULL) { - ev->next = pending_event_avail; - pending_event_avail = ev; - } -} - -/* - * Enqueue the outbound event onto the stack. If the queue was - * empty to begin with, we must also issue it via the Hypervisor - * interface. There is a section of code below that will touch - * the first stack pointer without the protection of the pending_event_spinlock. - * This is OK, because we know that nobody else will be modifying - * the first pointer when we do this. - */ -static int signal_event(struct pending_event *ev) -{ - int rc = 0; - unsigned long flags; - int go = 1; - struct pending_event *ev1; - HvLpEvent_Rc hv_rc; - - /* enqueue the event */ - if (ev != NULL) { - ev->next = NULL; - spin_lock_irqsave(&pending_event_spinlock, flags); - if (pending_event_head == NULL) - pending_event_head = ev; - else { - go = 0; - pending_event_tail->next = ev; - } - pending_event_tail = ev; - spin_unlock_irqrestore(&pending_event_spinlock, flags); - } - - /* send the event */ - while (go) { - go = 0; - - /* any DMA data to send beforehand? */ - if (pending_event_head->dma_data_length > 0) - HvCallEvent_dmaToSp(pending_event_head->dma_data, - pending_event_head->remote_address, - pending_event_head->dma_data_length, - HvLpDma_Direction_LocalToRemote); - - hv_rc = HvCallEvent_signalLpEvent( - &pending_event_head->event.hp_lp_event); - if (hv_rc != HvLpEvent_Rc_Good) { - printk(KERN_ERR "mf.c: HvCallEvent_signalLpEvent() " - "failed with %d\n", (int)hv_rc); - - spin_lock_irqsave(&pending_event_spinlock, flags); - ev1 = pending_event_head; - pending_event_head = pending_event_head->next; - if (pending_event_head != NULL) - go = 1; - spin_unlock_irqrestore(&pending_event_spinlock, flags); - - if (ev1 == ev) - rc = -EIO; - else if (ev1->hdlr != NULL) - (*ev1->hdlr)((void *)ev1->event.hp_lp_event.xCorrelationToken, -EIO); - - spin_lock_irqsave(&pending_event_spinlock, flags); - free_pending_event(ev1); - spin_unlock_irqrestore(&pending_event_spinlock, flags); - } - } - - return rc; -} - -/* - * Allocate a new pending_event structure, and initialize it. - */ -static struct pending_event *new_pending_event(void) -{ - struct pending_event *ev = NULL; - HvLpIndex primary_lp = HvLpConfig_getPrimaryLpIndex(); - unsigned long flags; - struct HvLpEvent *hev; - - spin_lock_irqsave(&pending_event_spinlock, flags); - if (pending_event_avail != NULL) { - ev = pending_event_avail; - pending_event_avail = pending_event_avail->next; - } - spin_unlock_irqrestore(&pending_event_spinlock, flags); - if (ev == NULL) { - ev = kmalloc(sizeof(struct pending_event), GFP_ATOMIC); - if (ev == NULL) { - printk(KERN_ERR "mf.c: unable to kmalloc %ld bytes\n", - sizeof(struct pending_event)); - return NULL; - } - } - memset(ev, 0, sizeof(struct pending_event)); - hev = &ev->event.hp_lp_event; - hev->xFlags.xValid = 1; - hev->xFlags.xAckType = HvLpEvent_AckType_ImmediateAck; - hev->xFlags.xAckInd = HvLpEvent_AckInd_DoAck; - hev->xFlags.xFunction = HvLpEvent_Function_Int; - hev->xType = HvLpEvent_Type_MachineFac; - hev->xSourceLp = HvLpConfig_getLpIndex(); - hev->xTargetLp = primary_lp; - hev->xSizeMinus1 = sizeof(ev->event) - 1; - hev->xRc = HvLpEvent_Rc_Good; - hev->xSourceInstanceId = HvCallEvent_getSourceLpInstanceId(primary_lp, - HvLpEvent_Type_MachineFac); - hev->xTargetInstanceId = HvCallEvent_getTargetLpInstanceId(primary_lp, - HvLpEvent_Type_MachineFac); - - return ev; -} - -static int signal_vsp_instruction(struct vsp_cmd_data *vsp_cmd) -{ - struct pending_event *ev = new_pending_event(); - int rc; - struct vsp_rsp_data response; - - if (ev == NULL) - return -ENOMEM; - - init_completion(&response.com); - response.response = vsp_cmd; - ev->event.hp_lp_event.xSubtype = 6; - ev->event.hp_lp_event.x.xSubtypeData = - subtype_data('M', 'F', 'V', 'I'); - ev->event.data.vsp_cmd.token = (u64)&response; - ev->event.data.vsp_cmd.cmd = vsp_cmd->cmd; - ev->event.data.vsp_cmd.lp_index = HvLpConfig_getLpIndex(); - ev->event.data.vsp_cmd.result_code = 0xFF; - ev->event.data.vsp_cmd.reserved = 0; - memcpy(&(ev->event.data.vsp_cmd.sub_data), - &(vsp_cmd->sub_data), sizeof(vsp_cmd->sub_data)); - mb(); - - rc = signal_event(ev); - if (rc == 0) - wait_for_completion(&response.com); - return rc; -} - - -/* - * Send a 12-byte CE message to the primary partition VSP object - */ -static int signal_ce_msg(char *ce_msg, struct ce_msg_comp_data *completion) -{ - struct pending_event *ev = new_pending_event(); - - if (ev == NULL) - return -ENOMEM; - - ev->event.hp_lp_event.xSubtype = 0; - ev->event.hp_lp_event.x.xSubtypeData = - subtype_data('M', 'F', 'C', 'E'); - memcpy(ev->event.data.ce_msg.ce_msg, ce_msg, 12); - ev->event.data.ce_msg.completion = completion; - return signal_event(ev); -} - -/* - * Send a 12-byte CE message (with no data) to the primary partition VSP object - */ -static int signal_ce_msg_simple(u8 ce_op, struct ce_msg_comp_data *completion) -{ - u8 ce_msg[12]; - - memset(ce_msg, 0, sizeof(ce_msg)); - ce_msg[3] = ce_op; - return signal_ce_msg(ce_msg, completion); -} - -/* - * Send a 12-byte CE message and DMA data to the primary partition VSP object - */ -static int dma_and_signal_ce_msg(char *ce_msg, - struct ce_msg_comp_data *completion, void *dma_data, - unsigned dma_data_length, unsigned remote_address) -{ - struct pending_event *ev = new_pending_event(); - - if (ev == NULL) - return -ENOMEM; - - ev->event.hp_lp_event.xSubtype = 0; - ev->event.hp_lp_event.x.xSubtypeData = - subtype_data('M', 'F', 'C', 'E'); - memcpy(ev->event.data.ce_msg.ce_msg, ce_msg, 12); - ev->event.data.ce_msg.completion = completion; - memcpy(ev->dma_data, dma_data, dma_data_length); - ev->dma_data_length = dma_data_length; - ev->remote_address = remote_address; - return signal_event(ev); -} - -/* - * Initiate a nice (hopefully) shutdown of Linux. We simply are - * going to try and send the init process a SIGINT signal. If - * this fails (why?), we'll simply force it off in a not-so-nice - * manner. - */ -static int shutdown(void) -{ - int rc = kill_proc(1, SIGINT, 1); - - if (rc) { - printk(KERN_ALERT "mf.c: SIGINT to init failed (%d), " - "hard shutdown commencing\n", rc); - mf_power_off(); - } else - printk(KERN_INFO "mf.c: init has been successfully notified " - "to proceed with shutdown\n"); - return rc; -} - -/* - * The primary partition VSP object is sending us a new - * event flow. Handle it... - */ -static void handle_int(struct io_mf_lp_event *event) -{ - struct ce_msg_data *ce_msg_data; - struct ce_msg_data *pce_msg_data; - unsigned long flags; - struct pending_event *pev; - - /* ack the interrupt */ - event->hp_lp_event.xRc = HvLpEvent_Rc_Good; - HvCallEvent_ackLpEvent(&event->hp_lp_event); - - /* process interrupt */ - switch (event->hp_lp_event.xSubtype) { - case 0: /* CE message */ - ce_msg_data = &event->data.ce_msg; - switch (ce_msg_data->ce_msg[3]) { - case 0x5B: /* power control notification */ - if ((ce_msg_data->ce_msg[5] & 0x20) != 0) { - printk(KERN_INFO "mf.c: Commencing partition shutdown\n"); - if (shutdown() == 0) - signal_ce_msg_simple(0xDB, NULL); - } - break; - case 0xC0: /* get time */ - spin_lock_irqsave(&pending_event_spinlock, flags); - pev = pending_event_head; - if (pev != NULL) - pending_event_head = pending_event_head->next; - spin_unlock_irqrestore(&pending_event_spinlock, flags); - if (pev == NULL) - break; - pce_msg_data = &pev->event.data.ce_msg; - if (pce_msg_data->ce_msg[3] != 0x40) - break; - if (pce_msg_data->completion != NULL) { - ce_msg_comp_hdlr handler = - pce_msg_data->completion->handler; - void *token = pce_msg_data->completion->token; - - if (handler != NULL) - (*handler)(token, ce_msg_data); - } - spin_lock_irqsave(&pending_event_spinlock, flags); - free_pending_event(pev); - spin_unlock_irqrestore(&pending_event_spinlock, flags); - /* send next waiting event */ - if (pending_event_head != NULL) - signal_event(NULL); - break; - } - break; - case 1: /* IT sys shutdown */ - printk(KERN_INFO "mf.c: Commencing system shutdown\n"); - shutdown(); - break; - } -} - -/* - * The primary partition VSP object is acknowledging the receipt - * of a flow we sent to them. If there are other flows queued - * up, we must send another one now... - */ -static void handle_ack(struct io_mf_lp_event *event) -{ - unsigned long flags; - struct pending_event *two = NULL; - unsigned long free_it = 0; - struct ce_msg_data *ce_msg_data; - struct ce_msg_data *pce_msg_data; - struct vsp_rsp_data *rsp; - - /* handle current event */ - if (pending_event_head == NULL) { - printk(KERN_ERR "mf.c: stack empty for receiving ack\n"); - return; - } - - switch (event->hp_lp_event.xSubtype) { - case 0: /* CE msg */ - ce_msg_data = &event->data.ce_msg; - if (ce_msg_data->ce_msg[3] != 0x40) { - free_it = 1; - break; - } - if (ce_msg_data->ce_msg[2] == 0) - break; - free_it = 1; - pce_msg_data = &pending_event_head->event.data.ce_msg; - if (pce_msg_data->completion != NULL) { - ce_msg_comp_hdlr handler = - pce_msg_data->completion->handler; - void *token = pce_msg_data->completion->token; - - if (handler != NULL) - (*handler)(token, ce_msg_data); - } - break; - case 4: /* allocate */ - case 5: /* deallocate */ - if (pending_event_head->hdlr != NULL) - (*pending_event_head->hdlr)((void *)event->hp_lp_event.xCorrelationToken, event->data.alloc.count); - free_it = 1; - break; - case 6: - free_it = 1; - rsp = (struct vsp_rsp_data *)event->data.vsp_cmd.token; - if (rsp == NULL) { - printk(KERN_ERR "mf.c: no rsp\n"); - break; - } - if (rsp->response != NULL) - memcpy(rsp->response, &event->data.vsp_cmd, - sizeof(event->data.vsp_cmd)); - complete(&rsp->com); - break; - } - - /* remove from queue */ - spin_lock_irqsave(&pending_event_spinlock, flags); - if ((pending_event_head != NULL) && (free_it == 1)) { - struct pending_event *oldHead = pending_event_head; - - pending_event_head = pending_event_head->next; - two = pending_event_head; - free_pending_event(oldHead); - } - spin_unlock_irqrestore(&pending_event_spinlock, flags); - - /* send next waiting event */ - if (two != NULL) - signal_event(NULL); -} - -/* - * This is the generic event handler we are registering with - * the Hypervisor. Ensure the flows are for us, and then - * parse it enough to know if it is an interrupt or an - * acknowledge. - */ -static void hv_handler(struct HvLpEvent *event, struct pt_regs *regs) -{ - if ((event != NULL) && (event->xType == HvLpEvent_Type_MachineFac)) { - switch(event->xFlags.xFunction) { - case HvLpEvent_Function_Ack: - handle_ack((struct io_mf_lp_event *)event); - break; - case HvLpEvent_Function_Int: - handle_int((struct io_mf_lp_event *)event); - break; - default: - printk(KERN_ERR "mf.c: non ack/int event received\n"); - break; - } - } else - printk(KERN_ERR "mf.c: alien event received\n"); -} - -/* - * Global kernel interface to allocate and seed events into the - * Hypervisor. - */ -void mf_allocate_lp_events(HvLpIndex target_lp, HvLpEvent_Type type, - unsigned size, unsigned count, MFCompleteHandler hdlr, - void *user_token) -{ - struct pending_event *ev = new_pending_event(); - int rc; - - if (ev == NULL) { - rc = -ENOMEM; - } else { - ev->event.hp_lp_event.xSubtype = 4; - ev->event.hp_lp_event.xCorrelationToken = (u64)user_token; - ev->event.hp_lp_event.x.xSubtypeData = - subtype_data('M', 'F', 'M', 'A'); - ev->event.data.alloc.target_lp = target_lp; - ev->event.data.alloc.type = type; - ev->event.data.alloc.size = size; - ev->event.data.alloc.count = count; - ev->hdlr = hdlr; - rc = signal_event(ev); - } - if ((rc != 0) && (hdlr != NULL)) - (*hdlr)(user_token, rc); -} -EXPORT_SYMBOL(mf_allocate_lp_events); - -/* - * Global kernel interface to unseed and deallocate events already in - * Hypervisor. - */ -void mf_deallocate_lp_events(HvLpIndex target_lp, HvLpEvent_Type type, - unsigned count, MFCompleteHandler hdlr, void *user_token) -{ - struct pending_event *ev = new_pending_event(); - int rc; - - if (ev == NULL) - rc = -ENOMEM; - else { - ev->event.hp_lp_event.xSubtype = 5; - ev->event.hp_lp_event.xCorrelationToken = (u64)user_token; - ev->event.hp_lp_event.x.xSubtypeData = - subtype_data('M', 'F', 'M', 'D'); - ev->event.data.alloc.target_lp = target_lp; - ev->event.data.alloc.type = type; - ev->event.data.alloc.count = count; - ev->hdlr = hdlr; - rc = signal_event(ev); - } - if ((rc != 0) && (hdlr != NULL)) - (*hdlr)(user_token, rc); -} -EXPORT_SYMBOL(mf_deallocate_lp_events); - -/* - * Global kernel interface to tell the VSP object in the primary - * partition to power this partition off. - */ -void mf_power_off(void) -{ - printk(KERN_INFO "mf.c: Down it goes...\n"); - signal_ce_msg_simple(0x4d, NULL); - for (;;) - ; -} - -/* - * Global kernel interface to tell the VSP object in the primary - * partition to reboot this partition. - */ -void mf_reboot(void) -{ - printk(KERN_INFO "mf.c: Preparing to bounce...\n"); - signal_ce_msg_simple(0x4e, NULL); - for (;;) - ; -} - -/* - * Display a single word SRC onto the VSP control panel. - */ -void mf_display_src(u32 word) -{ - u8 ce[12]; - - memset(ce, 0, sizeof(ce)); - ce[3] = 0x4a; - ce[7] = 0x01; - ce[8] = word >> 24; - ce[9] = word >> 16; - ce[10] = word >> 8; - ce[11] = word; - signal_ce_msg(ce, NULL); -} - -/* - * Display a single word SRC of the form "PROGXXXX" on the VSP control panel. - */ -void mf_display_progress(u16 value) -{ - u8 ce[12]; - u8 src[72]; - - memcpy(ce, "\x00\x00\x04\x4A\x00\x00\x00\x48\x00\x00\x00\x00", 12); - memcpy(src, "\x01\x00\x00\x01\x00\x00\x00\x00\x00\x00\x00\x00" - "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00" - "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00" - "\x00\x00\x00\x00PROGxxxx ", - 72); - src[6] = value >> 8; - src[7] = value & 255; - src[44] = "0123456789ABCDEF"[(value >> 12) & 15]; - src[45] = "0123456789ABCDEF"[(value >> 8) & 15]; - src[46] = "0123456789ABCDEF"[(value >> 4) & 15]; - src[47] = "0123456789ABCDEF"[value & 15]; - dma_and_signal_ce_msg(ce, NULL, src, sizeof(src), 9 * 64 * 1024); -} - -/* - * Clear the VSP control panel. Used to "erase" an SRC that was - * previously displayed. - */ -void mf_clear_src(void) -{ - signal_ce_msg_simple(0x4b, NULL); -} - -/* - * Initialization code here. - */ -void mf_init(void) -{ - int i; - - /* initialize */ - spin_lock_init(&pending_event_spinlock); - for (i = 0; - i < sizeof(pending_event_prealloc) / sizeof(*pending_event_prealloc); - ++i) - free_pending_event(&pending_event_prealloc[i]); - HvLpEvent_registerHandler(HvLpEvent_Type_MachineFac, &hv_handler); - - /* virtual continue ack */ - signal_ce_msg_simple(0x57, NULL); - - /* initialization complete */ - printk(KERN_NOTICE "mf.c: iSeries Linux LPAR Machine Facilities " - "initialized\n"); -} - -struct rtc_time_data { - struct completion com; - struct ce_msg_data ce_msg; - int rc; -}; - -static void get_rtc_time_complete(void *token, struct ce_msg_data *ce_msg) -{ - struct rtc_time_data *rtc = token; - - memcpy(&rtc->ce_msg, ce_msg, sizeof(rtc->ce_msg)); - rtc->rc = 0; - complete(&rtc->com); -} - -static int rtc_set_tm(int rc, u8 *ce_msg, struct rtc_time *tm) -{ - tm->tm_wday = 0; - tm->tm_yday = 0; - tm->tm_isdst = 0; - if (rc) { - tm->tm_sec = 0; - tm->tm_min = 0; - tm->tm_hour = 0; - tm->tm_mday = 15; - tm->tm_mon = 5; - tm->tm_year = 52; - return rc; - } - - if ((ce_msg[2] == 0xa9) || - (ce_msg[2] == 0xaf)) { - /* TOD clock is not set */ - tm->tm_sec = 1; - tm->tm_min = 1; - tm->tm_hour = 1; - tm->tm_mday = 10; - tm->tm_mon = 8; - tm->tm_year = 71; - mf_set_rtc(tm); - } - { - u8 year = ce_msg[5]; - u8 sec = ce_msg[6]; - u8 min = ce_msg[7]; - u8 hour = ce_msg[8]; - u8 day = ce_msg[10]; - u8 mon = ce_msg[11]; - - BCD_TO_BIN(sec); - BCD_TO_BIN(min); - BCD_TO_BIN(hour); - BCD_TO_BIN(day); - BCD_TO_BIN(mon); - BCD_TO_BIN(year); - - if (year <= 69) - year += 100; - - tm->tm_sec = sec; - tm->tm_min = min; - tm->tm_hour = hour; - tm->tm_mday = day; - tm->tm_mon = mon; - tm->tm_year = year; - } - - return 0; -} - -int mf_get_rtc(struct rtc_time *tm) -{ - struct ce_msg_comp_data ce_complete; - struct rtc_time_data rtc_data; - int rc; - - memset(&ce_complete, 0, sizeof(ce_complete)); - memset(&rtc_data, 0, sizeof(rtc_data)); - init_completion(&rtc_data.com); - ce_complete.handler = &get_rtc_time_complete; - ce_complete.token = &rtc_data; - rc = signal_ce_msg_simple(0x40, &ce_complete); - if (rc) - return rc; - wait_for_completion(&rtc_data.com); - return rtc_set_tm(rtc_data.rc, rtc_data.ce_msg.ce_msg, tm); -} - -struct boot_rtc_time_data { - int busy; - struct ce_msg_data ce_msg; - int rc; -}; - -static void get_boot_rtc_time_complete(void *token, struct ce_msg_data *ce_msg) -{ - struct boot_rtc_time_data *rtc = token; - - memcpy(&rtc->ce_msg, ce_msg, sizeof(rtc->ce_msg)); - rtc->rc = 0; - rtc->busy = 0; -} - -int mf_get_boot_rtc(struct rtc_time *tm) -{ - struct ce_msg_comp_data ce_complete; - struct boot_rtc_time_data rtc_data; - int rc; - - memset(&ce_complete, 0, sizeof(ce_complete)); - memset(&rtc_data, 0, sizeof(rtc_data)); - rtc_data.busy = 1; - ce_complete.handler = &get_boot_rtc_time_complete; - ce_complete.token = &rtc_data; - rc = signal_ce_msg_simple(0x40, &ce_complete); - if (rc) - return rc; - /* We need to poll here as we are not yet taking interrupts */ - while (rtc_data.busy) { - if (hvlpevent_is_pending()) - process_hvlpevents(NULL); - } - return rtc_set_tm(rtc_data.rc, rtc_data.ce_msg.ce_msg, tm); -} - -int mf_set_rtc(struct rtc_time *tm) -{ - char ce_time[12]; - u8 day, mon, hour, min, sec, y1, y2; - unsigned year; - - year = 1900 + tm->tm_year; - y1 = year / 100; - y2 = year % 100; - - sec = tm->tm_sec; - min = tm->tm_min; - hour = tm->tm_hour; - day = tm->tm_mday; - mon = tm->tm_mon + 1; - - BIN_TO_BCD(sec); - BIN_TO_BCD(min); - BIN_TO_BCD(hour); - BIN_TO_BCD(mon); - BIN_TO_BCD(day); - BIN_TO_BCD(y1); - BIN_TO_BCD(y2); - - memset(ce_time, 0, sizeof(ce_time)); - ce_time[3] = 0x41; - ce_time[4] = y1; - ce_time[5] = y2; - ce_time[6] = sec; - ce_time[7] = min; - ce_time[8] = hour; - ce_time[10] = day; - ce_time[11] = mon; - - return signal_ce_msg(ce_time, NULL); -} - -#ifdef CONFIG_PROC_FS - -static int proc_mf_dump_cmdline(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - int len; - char *p; - struct vsp_cmd_data vsp_cmd; - int rc; - dma_addr_t dma_addr; - - /* The HV appears to return no more than 256 bytes of command line */ - if (off >= 256) - return 0; - if ((off + count) > 256) - count = 256 - off; - - dma_addr = dma_map_single(iSeries_vio_dev, page, off + count, - DMA_FROM_DEVICE); - if (dma_mapping_error(dma_addr)) - return -ENOMEM; - memset(page, 0, off + count); - memset(&vsp_cmd, 0, sizeof(vsp_cmd)); - vsp_cmd.cmd = 33; - vsp_cmd.sub_data.kern.token = dma_addr; - vsp_cmd.sub_data.kern.address_type = HvLpDma_AddressType_TceIndex; - vsp_cmd.sub_data.kern.side = (u64)data; - vsp_cmd.sub_data.kern.length = off + count; - mb(); - rc = signal_vsp_instruction(&vsp_cmd); - dma_unmap_single(iSeries_vio_dev, dma_addr, off + count, - DMA_FROM_DEVICE); - if (rc) - return rc; - if (vsp_cmd.result_code != 0) - return -ENOMEM; - p = page; - len = 0; - while (len < (off + count)) { - if ((*p == '\0') || (*p == '\n')) { - if (*p == '\0') - *p = '\n'; - p++; - len++; - *eof = 1; - break; - } - p++; - len++; - } - - if (len < off) { - *eof = 1; - len = 0; - } - return len; -} - -#if 0 -static int mf_getVmlinuxChunk(char *buffer, int *size, int offset, u64 side) -{ - struct vsp_cmd_data vsp_cmd; - int rc; - int len = *size; - dma_addr_t dma_addr; - - dma_addr = dma_map_single(iSeries_vio_dev, buffer, len, - DMA_FROM_DEVICE); - memset(buffer, 0, len); - memset(&vsp_cmd, 0, sizeof(vsp_cmd)); - vsp_cmd.cmd = 32; - vsp_cmd.sub_data.kern.token = dma_addr; - vsp_cmd.sub_data.kern.address_type = HvLpDma_AddressType_TceIndex; - vsp_cmd.sub_data.kern.side = side; - vsp_cmd.sub_data.kern.offset = offset; - vsp_cmd.sub_data.kern.length = len; - mb(); - rc = signal_vsp_instruction(&vsp_cmd); - if (rc == 0) { - if (vsp_cmd.result_code == 0) - *size = vsp_cmd.sub_data.length_out; - else - rc = -ENOMEM; - } - - dma_unmap_single(iSeries_vio_dev, dma_addr, len, DMA_FROM_DEVICE); - - return rc; -} - -static int proc_mf_dump_vmlinux(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - int sizeToGet = count; - - if (!capable(CAP_SYS_ADMIN)) - return -EACCES; - - if (mf_getVmlinuxChunk(page, &sizeToGet, off, (u64)data) == 0) { - if (sizeToGet != 0) { - *start = page + off; - return sizeToGet; - } - *eof = 1; - return 0; - } - *eof = 1; - return 0; -} -#endif - -static int proc_mf_dump_side(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - int len; - char mf_current_side = ' '; - struct vsp_cmd_data vsp_cmd; - - memset(&vsp_cmd, 0, sizeof(vsp_cmd)); - vsp_cmd.cmd = 2; - vsp_cmd.sub_data.ipl_type = 0; - mb(); - - if (signal_vsp_instruction(&vsp_cmd) == 0) { - if (vsp_cmd.result_code == 0) { - switch (vsp_cmd.sub_data.ipl_type) { - case 0: mf_current_side = 'A'; - break; - case 1: mf_current_side = 'B'; - break; - case 2: mf_current_side = 'C'; - break; - default: mf_current_side = 'D'; - break; - } - } - } - - len = sprintf(page, "%c\n", mf_current_side); - - if (len <= (off + count)) - *eof = 1; - *start = page + off; - len -= off; - if (len > count) - len = count; - if (len < 0) - len = 0; - return len; -} - -static int proc_mf_change_side(struct file *file, const char __user *buffer, - unsigned long count, void *data) -{ - char side; - u64 newSide; - struct vsp_cmd_data vsp_cmd; - - if (!capable(CAP_SYS_ADMIN)) - return -EACCES; - - if (count == 0) - return 0; - - if (get_user(side, buffer)) - return -EFAULT; - - switch (side) { - case 'A': newSide = 0; - break; - case 'B': newSide = 1; - break; - case 'C': newSide = 2; - break; - case 'D': newSide = 3; - break; - default: - printk(KERN_ERR "mf_proc.c: proc_mf_change_side: invalid side\n"); - return -EINVAL; - } - - memset(&vsp_cmd, 0, sizeof(vsp_cmd)); - vsp_cmd.sub_data.ipl_type = newSide; - vsp_cmd.cmd = 10; - - (void)signal_vsp_instruction(&vsp_cmd); - - return count; -} - -#if 0 -static void mf_getSrcHistory(char *buffer, int size) -{ - struct IplTypeReturnStuff return_stuff; - struct pending_event *ev = new_pending_event(); - int rc = 0; - char *pages[4]; - - pages[0] = kmalloc(4096, GFP_ATOMIC); - pages[1] = kmalloc(4096, GFP_ATOMIC); - pages[2] = kmalloc(4096, GFP_ATOMIC); - pages[3] = kmalloc(4096, GFP_ATOMIC); - if ((ev == NULL) || (pages[0] == NULL) || (pages[1] == NULL) - || (pages[2] == NULL) || (pages[3] == NULL)) - return -ENOMEM; - - return_stuff.xType = 0; - return_stuff.xRc = 0; - return_stuff.xDone = 0; - ev->event.hp_lp_event.xSubtype = 6; - ev->event.hp_lp_event.x.xSubtypeData = - subtype_data('M', 'F', 'V', 'I'); - ev->event.data.vsp_cmd.xEvent = &return_stuff; - ev->event.data.vsp_cmd.cmd = 4; - ev->event.data.vsp_cmd.lp_index = HvLpConfig_getLpIndex(); - ev->event.data.vsp_cmd.result_code = 0xFF; - ev->event.data.vsp_cmd.reserved = 0; - ev->event.data.vsp_cmd.sub_data.page[0] = ISERIES_HV_ADDR(pages[0]); - ev->event.data.vsp_cmd.sub_data.page[1] = ISERIES_HV_ADDR(pages[1]); - ev->event.data.vsp_cmd.sub_data.page[2] = ISERIES_HV_ADDR(pages[2]); - ev->event.data.vsp_cmd.sub_data.page[3] = ISERIES_HV_ADDR(pages[3]); - mb(); - if (signal_event(ev) != 0) - return; - - while (return_stuff.xDone != 1) - udelay(10); - if (return_stuff.xRc == 0) - memcpy(buffer, pages[0], size); - kfree(pages[0]); - kfree(pages[1]); - kfree(pages[2]); - kfree(pages[3]); -} -#endif - -static int proc_mf_dump_src(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ -#if 0 - int len; - - mf_getSrcHistory(page, count); - len = count; - len -= off; - if (len < count) { - *eof = 1; - if (len <= 0) - return 0; - } else - len = count; - *start = page + off; - return len; -#else - return 0; -#endif -} - -static int proc_mf_change_src(struct file *file, const char __user *buffer, - unsigned long count, void *data) -{ - char stkbuf[10]; - - if (!capable(CAP_SYS_ADMIN)) - return -EACCES; - - if ((count < 4) && (count != 1)) { - printk(KERN_ERR "mf_proc: invalid src\n"); - return -EINVAL; - } - - if (count > (sizeof(stkbuf) - 1)) - count = sizeof(stkbuf) - 1; - if (copy_from_user(stkbuf, buffer, count)) - return -EFAULT; - - if ((count == 1) && (*stkbuf == '\0')) - mf_clear_src(); - else - mf_display_src(*(u32 *)stkbuf); - - return count; -} - -static int proc_mf_change_cmdline(struct file *file, const char __user *buffer, - unsigned long count, void *data) -{ - struct vsp_cmd_data vsp_cmd; - dma_addr_t dma_addr; - char *page; - int ret = -EACCES; - - if (!capable(CAP_SYS_ADMIN)) - goto out; - - dma_addr = 0; - page = dma_alloc_coherent(iSeries_vio_dev, count, &dma_addr, - GFP_ATOMIC); - ret = -ENOMEM; - if (page == NULL) - goto out; - - ret = -EFAULT; - if (copy_from_user(page, buffer, count)) - goto out_free; - - memset(&vsp_cmd, 0, sizeof(vsp_cmd)); - vsp_cmd.cmd = 31; - vsp_cmd.sub_data.kern.token = dma_addr; - vsp_cmd.sub_data.kern.address_type = HvLpDma_AddressType_TceIndex; - vsp_cmd.sub_data.kern.side = (u64)data; - vsp_cmd.sub_data.kern.length = count; - mb(); - (void)signal_vsp_instruction(&vsp_cmd); - ret = count; - -out_free: - dma_free_coherent(iSeries_vio_dev, count, page, dma_addr); -out: - return ret; -} - -static ssize_t proc_mf_change_vmlinux(struct file *file, - const char __user *buf, - size_t count, loff_t *ppos) -{ - struct proc_dir_entry *dp = PDE(file->f_dentry->d_inode); - ssize_t rc; - dma_addr_t dma_addr; - char *page; - struct vsp_cmd_data vsp_cmd; - - rc = -EACCES; - if (!capable(CAP_SYS_ADMIN)) - goto out; - - dma_addr = 0; - page = dma_alloc_coherent(iSeries_vio_dev, count, &dma_addr, - GFP_ATOMIC); - rc = -ENOMEM; - if (page == NULL) { - printk(KERN_ERR "mf.c: couldn't allocate memory to set vmlinux chunk\n"); - goto out; - } - rc = -EFAULT; - if (copy_from_user(page, buf, count)) - goto out_free; - - memset(&vsp_cmd, 0, sizeof(vsp_cmd)); - vsp_cmd.cmd = 30; - vsp_cmd.sub_data.kern.token = dma_addr; - vsp_cmd.sub_data.kern.address_type = HvLpDma_AddressType_TceIndex; - vsp_cmd.sub_data.kern.side = (u64)dp->data; - vsp_cmd.sub_data.kern.offset = *ppos; - vsp_cmd.sub_data.kern.length = count; - mb(); - rc = signal_vsp_instruction(&vsp_cmd); - if (rc) - goto out_free; - rc = -ENOMEM; - if (vsp_cmd.result_code != 0) - goto out_free; - - *ppos += count; - rc = count; -out_free: - dma_free_coherent(iSeries_vio_dev, count, page, dma_addr); -out: - return rc; -} - -static struct file_operations proc_vmlinux_operations = { - .write = proc_mf_change_vmlinux, -}; - -static int __init mf_proc_init(void) -{ - struct proc_dir_entry *mf_proc_root; - struct proc_dir_entry *ent; - struct proc_dir_entry *mf; - char name[2]; - int i; - - mf_proc_root = proc_mkdir("iSeries/mf", NULL); - if (!mf_proc_root) - return 1; - - name[1] = '\0'; - for (i = 0; i < 4; i++) { - name[0] = 'A' + i; - mf = proc_mkdir(name, mf_proc_root); - if (!mf) - return 1; - - ent = create_proc_entry("cmdline", S_IFREG|S_IRUSR|S_IWUSR, mf); - if (!ent) - return 1; - ent->nlink = 1; - ent->data = (void *)(long)i; - ent->read_proc = proc_mf_dump_cmdline; - ent->write_proc = proc_mf_change_cmdline; - - if (i == 3) /* no vmlinux entry for 'D' */ - continue; - - ent = create_proc_entry("vmlinux", S_IFREG|S_IWUSR, mf); - if (!ent) - return 1; - ent->nlink = 1; - ent->data = (void *)(long)i; - ent->proc_fops = &proc_vmlinux_operations; - } - - ent = create_proc_entry("side", S_IFREG|S_IRUSR|S_IWUSR, mf_proc_root); - if (!ent) - return 1; - ent->nlink = 1; - ent->data = (void *)0; - ent->read_proc = proc_mf_dump_side; - ent->write_proc = proc_mf_change_side; - - ent = create_proc_entry("src", S_IFREG|S_IRUSR|S_IWUSR, mf_proc_root); - if (!ent) - return 1; - ent->nlink = 1; - ent->data = (void *)0; - ent->read_proc = proc_mf_dump_src; - ent->write_proc = proc_mf_change_src; - - return 0; -} - -__initcall(mf_proc_init); - -#endif /* CONFIG_PROC_FS */ diff --git a/arch/ppc64/kernel/misc.S b/arch/ppc64/kernel/misc.S index e7241ad80a08..a33448c2bd91 100644 --- a/arch/ppc64/kernel/misc.S +++ b/arch/ppc64/kernel/misc.S @@ -28,6 +28,7 @@ #include <asm/ppc_asm.h> #include <asm/asm-offsets.h> #include <asm/cputable.h> +#include <asm/thread_info.h> .text @@ -64,44 +65,6 @@ _GLOBAL(get_srr1) _GLOBAL(get_sp) mr r3,r1 blr - -#ifdef CONFIG_PPC_ISERIES -/* unsigned long local_save_flags(void) */ -_GLOBAL(local_get_flags) - lbz r3,PACAPROCENABLED(r13) - blr - -/* unsigned long local_irq_disable(void) */ -_GLOBAL(local_irq_disable) - lbz r3,PACAPROCENABLED(r13) - li r4,0 - stb r4,PACAPROCENABLED(r13) - blr /* Done */ - -/* void local_irq_restore(unsigned long flags) */ -_GLOBAL(local_irq_restore) - lbz r5,PACAPROCENABLED(r13) - /* Check if things are setup the way we want _already_. */ - cmpw 0,r3,r5 - beqlr - /* are we enabling interrupts? */ - cmpdi 0,r3,0 - stb r3,PACAPROCENABLED(r13) - beqlr - /* Check pending interrupts */ - /* A decrementer, IPI or PMC interrupt may have occurred - * while we were in the hypervisor (which enables) */ - ld r4,PACALPPACA+LPPACAANYINT(r13) - cmpdi r4,0 - beqlr - - /* - * Handle pending interrupts in interrupt context - */ - li r0,0x5555 - sc - blr -#endif /* CONFIG_PPC_ISERIES */ #ifdef CONFIG_IRQSTACKS _GLOBAL(call_do_softirq) @@ -329,7 +292,7 @@ _GLOBAL(__flush_dcache_icache) /* Flush the dcache */ ld r7,PPC64_CACHES@toc(r2) - clrrdi r3,r3,12 /* Page align */ + clrrdi r3,r3,PAGE_SHIFT /* Page align */ lwz r4,DCACHEL1LINESPERPAGE(r7) /* Get # dcache lines per page */ lwz r5,DCACHEL1LINESIZE(r7) /* Get dcache line size */ mr r6,r3 @@ -778,6 +741,13 @@ _GLOBAL(giveup_altivec) _GLOBAL(__setup_cpu_power3) blr +_GLOBAL(execve) + li r0,__NR_execve + sc + bnslr + neg r3,r3 + blr + /* kexec_wait(phys_cpu) * * wait for the flag to change, indicating this kernel is going away but @@ -959,11 +929,11 @@ _GLOBAL(sys_call_table32) .llong .sys_write .llong .compat_sys_open /* 5 */ .llong .sys_close - .llong .sys32_waitpid - .llong .sys32_creat + .llong .compat_sys_waitpid + .llong .compat_sys_creat .llong .sys_link .llong .sys_unlink /* 10 */ - .llong .sys32_execve + .llong .compat_sys_execve .llong .sys_chdir .llong .compat_sys_time .llong .sys_mknod @@ -978,20 +948,20 @@ _GLOBAL(sys_call_table32) .llong .sys_setuid .llong .sys_getuid .llong .compat_sys_stime /* 25 */ - .llong .sys32_ptrace + .llong .compat_sys_ptrace .llong .sys_alarm .llong .sys_ni_syscall /* old fstat syscall */ - .llong .sys32_pause - .llong .compat_sys_utime /* 30 */ + .llong .compat_sys_pause + .llong .compat_sys_utime /* 30 */ .llong .sys_ni_syscall /* old stty syscall */ .llong .sys_ni_syscall /* old gtty syscall */ - .llong .sys32_access - .llong .sys32_nice + .llong .compat_sys_access + .llong .compat_sys_nice .llong .sys_ni_syscall /* 35 - old ftime syscall */ .llong .sys_sync - .llong .sys32_kill + .llong .compat_sys_kill .llong .sys_rename - .llong .sys32_mkdir + .llong .compat_sys_mkdir .llong .sys_rmdir /* 40 */ .llong .sys_dup .llong .sys_pipe @@ -1009,35 +979,35 @@ _GLOBAL(sys_call_table32) .llong .compat_sys_ioctl .llong .compat_sys_fcntl /* 55 */ .llong .sys_ni_syscall /* old mpx syscall */ - .llong .sys32_setpgid + .llong .compat_sys_setpgid .llong .sys_ni_syscall /* old ulimit syscall */ - .llong .sys32_olduname - .llong .sys32_umask /* 60 */ + .llong .sys_olduname + .llong .compat_sys_umask /* 60 */ .llong .sys_chroot .llong .sys_ustat .llong .sys_dup2 .llong .sys_getppid .llong .sys_getpgrp /* 65 */ .llong .sys_setsid - .llong .sys32_sigaction + .llong .compat_sys_sigaction .llong .sys_sgetmask - .llong .sys32_ssetmask + .llong .compat_sys_ssetmask .llong .sys_setreuid /* 70 */ .llong .sys_setregid .llong .ppc32_sigsuspend .llong .compat_sys_sigpending - .llong .sys32_sethostname - .llong .compat_sys_setrlimit /* 75 */ + .llong .compat_sys_sethostname + .llong .compat_sys_setrlimit /* 75 */ .llong .compat_sys_old_getrlimit .llong .compat_sys_getrusage - .llong .sys32_gettimeofday - .llong .sys32_settimeofday - .llong .sys32_getgroups /* 80 */ - .llong .sys32_setgroups + .llong .compat_sys_gettimeofday + .llong .compat_sys_settimeofday + .llong .compat_sys_getgroups /* 80 */ + .llong .compat_sys_setgroups .llong .sys_ni_syscall /* old select syscall */ .llong .sys_symlink .llong .sys_ni_syscall /* old lstat syscall */ - .llong .sys32_readlink /* 85 */ + .llong .compat_sys_readlink /* 85 */ .llong .sys_uselib .llong .sys_swapon .llong .sys_reboot @@ -1048,35 +1018,35 @@ _GLOBAL(sys_call_table32) .llong .sys_ftruncate .llong .sys_fchmod .llong .sys_fchown /* 95 */ - .llong .sys32_getpriority - .llong .sys32_setpriority + .llong .compat_sys_getpriority + .llong .compat_sys_setpriority .llong .sys_ni_syscall /* old profil syscall */ .llong .compat_sys_statfs - .llong .compat_sys_fstatfs /* 100 */ + .llong .compat_sys_fstatfs /* 100 */ .llong .sys_ni_syscall /* old ioperm syscall */ .llong .compat_sys_socketcall - .llong .sys32_syslog + .llong .compat_sys_syslog .llong .compat_sys_setitimer - .llong .compat_sys_getitimer /* 105 */ + .llong .compat_sys_getitimer /* 105 */ .llong .compat_sys_newstat .llong .compat_sys_newlstat .llong .compat_sys_newfstat - .llong .sys32_uname + .llong .sys_uname .llong .sys_ni_syscall /* 110 old iopl syscall */ .llong .sys_vhangup .llong .sys_ni_syscall /* old idle syscall */ .llong .sys_ni_syscall /* old vm86 syscall */ .llong .compat_sys_wait4 .llong .sys_swapoff /* 115 */ - .llong .sys32_sysinfo + .llong .compat_sys_sysinfo .llong .sys32_ipc .llong .sys_fsync .llong .ppc32_sigreturn .llong .ppc_clone /* 120 */ - .llong .sys32_setdomainname - .llong .ppc64_newuname + .llong .compat_sys_setdomainname + .llong .ppc_newuname .llong .sys_ni_syscall /* old modify_ldt syscall */ - .llong .sys32_adjtimex + .llong .compat_sys_adjtimex .llong .sys_mprotect /* 125 */ .llong .compat_sys_sigprocmask .llong .sys_ni_syscall /* old create_module syscall */ @@ -1084,36 +1054,36 @@ _GLOBAL(sys_call_table32) .llong .sys_delete_module .llong .sys_ni_syscall /* 130 old get_kernel_syms syscall */ .llong .sys_quotactl - .llong .sys32_getpgid + .llong .compat_sys_getpgid .llong .sys_fchdir .llong .sys_bdflush - .llong .sys32_sysfs /* 135 */ + .llong .compat_sys_sysfs /* 135 */ .llong .ppc64_personality .llong .sys_ni_syscall /* for afs_syscall */ .llong .sys_setfsuid .llong .sys_setfsgid .llong .sys_llseek /* 140 */ - .llong .sys32_getdents + .llong .compat_sys_getdents .llong .ppc32_select .llong .sys_flock .llong .sys_msync .llong .compat_sys_readv /* 145 */ .llong .compat_sys_writev - .llong .sys32_getsid + .llong .compat_sys_getsid .llong .sys_fdatasync - .llong .sys32_sysctl + .llong .compat_sys_sysctl .llong .sys_mlock /* 150 */ .llong .sys_munlock .llong .sys_mlockall .llong .sys_munlockall - .llong .sys32_sched_setparam - .llong .sys32_sched_getparam /* 155 */ - .llong .sys32_sched_setscheduler - .llong .sys32_sched_getscheduler + .llong .compat_sys_sched_setparam + .llong .compat_sys_sched_getparam /* 155 */ + .llong .compat_sys_sched_setscheduler + .llong .compat_sys_sched_getscheduler .llong .sys_sched_yield - .llong .sys32_sched_get_priority_max - .llong .sys32_sched_get_priority_min /* 160 */ - .llong .sys32_sched_rr_get_interval + .llong .compat_sys_sched_get_priority_max + .llong .compat_sys_sched_get_priority_min /* 160 */ + .llong .compat_sys_sched_rr_get_interval .llong .compat_sys_nanosleep .llong .sys_mremap .llong .sys_setresuid @@ -1123,36 +1093,36 @@ _GLOBAL(sys_call_table32) .llong .compat_sys_nfsservctl .llong .sys_setresgid .llong .sys_getresgid /* 170 */ - .llong .sys32_prctl + .llong .compat_sys_prctl .llong .ppc32_rt_sigreturn - .llong .sys32_rt_sigaction - .llong .sys32_rt_sigprocmask - .llong .sys32_rt_sigpending /* 175 */ + .llong .compat_sys_rt_sigaction + .llong .compat_sys_rt_sigprocmask + .llong .compat_sys_rt_sigpending /* 175 */ .llong .compat_sys_rt_sigtimedwait - .llong .sys32_rt_sigqueueinfo + .llong .compat_sys_rt_sigqueueinfo .llong .ppc32_rt_sigsuspend - .llong .sys32_pread64 - .llong .sys32_pwrite64 /* 180 */ + .llong .compat_sys_pread64 + .llong .compat_sys_pwrite64 /* 180 */ .llong .sys_chown .llong .sys_getcwd .llong .sys_capget .llong .sys_capset - .llong .sys32_sigaltstack /* 185 */ - .llong .sys32_sendfile + .llong .compat_sys_sigaltstack /* 185 */ + .llong .compat_sys_sendfile .llong .sys_ni_syscall /* reserved for streams1 */ .llong .sys_ni_syscall /* reserved for streams2 */ .llong .ppc_vfork .llong .compat_sys_getrlimit /* 190 */ - .llong .sys32_readahead - .llong .sys32_mmap2 - .llong .sys32_truncate64 - .llong .sys32_ftruncate64 + .llong .compat_sys_readahead + .llong .compat_sys_mmap2 + .llong .compat_sys_truncate64 + .llong .compat_sys_ftruncate64 .llong .sys_stat64 /* 195 */ .llong .sys_lstat64 .llong .sys_fstat64 - .llong .sys32_pciconfig_read - .llong .sys32_pciconfig_write - .llong .sys32_pciconfig_iobase /* 200 - pciconfig_iobase */ + .llong .compat_sys_pciconfig_read + .llong .compat_sys_pciconfig_write + .llong .compat_sys_pciconfig_iobase /* 200 - pciconfig_iobase */ .llong .sys_ni_syscall /* reserved for MacOnLinux */ .llong .sys_getdents64 .llong .sys_pivot_root @@ -1178,7 +1148,7 @@ _GLOBAL(sys_call_table32) .llong .compat_sys_sched_getaffinity .llong .sys_ni_syscall .llong .sys_ni_syscall /* 225 - reserved for tux */ - .llong .sys32_sendfile64 + .llong .compat_sys_sendfile64 .llong .compat_sys_io_setup .llong .sys_io_destroy .llong .compat_sys_io_getevents @@ -1197,16 +1167,16 @@ _GLOBAL(sys_call_table32) .llong .compat_sys_timer_gettime .llong .sys_timer_getoverrun .llong .sys_timer_delete - .llong .compat_sys_clock_settime /* 245 */ + .llong .compat_sys_clock_settime/* 245 */ .llong .compat_sys_clock_gettime .llong .compat_sys_clock_getres .llong .compat_sys_clock_nanosleep .llong .ppc32_swapcontext - .llong .sys32_tgkill /* 250 */ - .llong .sys32_utimes + .llong .compat_sys_tgkill /* 250 */ + .llong .compat_sys_utimes .llong .compat_sys_statfs64 .llong .compat_sys_fstatfs64 - .llong .ppc32_fadvise64_64 /* 32bit only fadvise64_64 */ + .llong .ppc_fadvise64_64 /* 32bit only fadvise64_64 */ .llong .ppc_rtas /* 255 */ .llong .sys_ni_syscall /* 256 reserved for sys_debug_setcontext */ .llong .sys_ni_syscall /* 257 reserved for vserver */ @@ -1221,12 +1191,12 @@ _GLOBAL(sys_call_table32) .llong .compat_sys_mq_notify .llong .compat_sys_mq_getsetattr .llong .compat_sys_kexec_load - .llong .sys32_add_key - .llong .sys32_request_key /* 270 */ + .llong .compat_sys_add_key + .llong .compat_sys_request_key /* 270 */ .llong .compat_sys_keyctl .llong .compat_sys_waitid - .llong .sys32_ioprio_set - .llong .sys32_ioprio_get + .llong .compat_sys_ioprio_set + .llong .compat_sys_ioprio_get .llong .sys_inotify_init /* 275 */ .llong .sys_inotify_add_watch .llong .sys_inotify_rm_watch @@ -1355,7 +1325,7 @@ _GLOBAL(sys_call_table) .llong .sys_ni_syscall .llong .ppc_clone /* 120 */ .llong .sys_setdomainname - .llong .ppc64_newuname + .llong .ppc_newuname .llong .sys_ni_syscall /* old modify_ldt syscall */ .llong .sys_adjtimex .llong .sys_mprotect /* 125 */ diff --git a/arch/ppc64/kernel/mpic.c b/arch/ppc64/kernel/mpic.c deleted file mode 100644 index cc262a05ddb4..000000000000 --- a/arch/ppc64/kernel/mpic.c +++ /dev/null @@ -1,888 +0,0 @@ -/* - * arch/ppc64/kernel/mpic.c - * - * Driver for interrupt controllers following the OpenPIC standard, the - * common implementation beeing IBM's MPIC. This driver also can deal - * with various broken implementations of this HW. - * - * Copyright (C) 2004 Benjamin Herrenschmidt, IBM Corp. - * - * This file is subject to the terms and conditions of the GNU General Public - * License. See the file COPYING in the main directory of this archive - * for more details. - */ - -#undef DEBUG - -#include <linux/config.h> -#include <linux/types.h> -#include <linux/kernel.h> -#include <linux/init.h> -#include <linux/irq.h> -#include <linux/smp.h> -#include <linux/interrupt.h> -#include <linux/bootmem.h> -#include <linux/spinlock.h> -#include <linux/pci.h> - -#include <asm/ptrace.h> -#include <asm/signal.h> -#include <asm/io.h> -#include <asm/pgtable.h> -#include <asm/irq.h> -#include <asm/machdep.h> - -#include "mpic.h" - -#ifdef DEBUG -#define DBG(fmt...) printk(fmt) -#else -#define DBG(fmt...) -#endif - -static struct mpic *mpics; -static struct mpic *mpic_primary; -static DEFINE_SPINLOCK(mpic_lock); - - -/* - * Register accessor functions - */ - - -static inline u32 _mpic_read(unsigned int be, volatile u32 __iomem *base, - unsigned int reg) -{ - if (be) - return in_be32(base + (reg >> 2)); - else - return in_le32(base + (reg >> 2)); -} - -static inline void _mpic_write(unsigned int be, volatile u32 __iomem *base, - unsigned int reg, u32 value) -{ - if (be) - out_be32(base + (reg >> 2), value); - else - out_le32(base + (reg >> 2), value); -} - -static inline u32 _mpic_ipi_read(struct mpic *mpic, unsigned int ipi) -{ - unsigned int be = (mpic->flags & MPIC_BIG_ENDIAN) != 0; - unsigned int offset = MPIC_GREG_IPI_VECTOR_PRI_0 + (ipi * 0x10); - - if (mpic->flags & MPIC_BROKEN_IPI) - be = !be; - return _mpic_read(be, mpic->gregs, offset); -} - -static inline void _mpic_ipi_write(struct mpic *mpic, unsigned int ipi, u32 value) -{ - unsigned int offset = MPIC_GREG_IPI_VECTOR_PRI_0 + (ipi * 0x10); - - _mpic_write(mpic->flags & MPIC_BIG_ENDIAN, mpic->gregs, offset, value); -} - -static inline u32 _mpic_cpu_read(struct mpic *mpic, unsigned int reg) -{ - unsigned int cpu = 0; - - if (mpic->flags & MPIC_PRIMARY) - cpu = hard_smp_processor_id(); - - return _mpic_read(mpic->flags & MPIC_BIG_ENDIAN, mpic->cpuregs[cpu], reg); -} - -static inline void _mpic_cpu_write(struct mpic *mpic, unsigned int reg, u32 value) -{ - unsigned int cpu = 0; - - if (mpic->flags & MPIC_PRIMARY) - cpu = hard_smp_processor_id(); - - _mpic_write(mpic->flags & MPIC_BIG_ENDIAN, mpic->cpuregs[cpu], reg, value); -} - -static inline u32 _mpic_irq_read(struct mpic *mpic, unsigned int src_no, unsigned int reg) -{ - unsigned int isu = src_no >> mpic->isu_shift; - unsigned int idx = src_no & mpic->isu_mask; - - return _mpic_read(mpic->flags & MPIC_BIG_ENDIAN, mpic->isus[isu], - reg + (idx * MPIC_IRQ_STRIDE)); -} - -static inline void _mpic_irq_write(struct mpic *mpic, unsigned int src_no, - unsigned int reg, u32 value) -{ - unsigned int isu = src_no >> mpic->isu_shift; - unsigned int idx = src_no & mpic->isu_mask; - - _mpic_write(mpic->flags & MPIC_BIG_ENDIAN, mpic->isus[isu], - reg + (idx * MPIC_IRQ_STRIDE), value); -} - -#define mpic_read(b,r) _mpic_read(mpic->flags & MPIC_BIG_ENDIAN,(b),(r)) -#define mpic_write(b,r,v) _mpic_write(mpic->flags & MPIC_BIG_ENDIAN,(b),(r),(v)) -#define mpic_ipi_read(i) _mpic_ipi_read(mpic,(i)) -#define mpic_ipi_write(i,v) _mpic_ipi_write(mpic,(i),(v)) -#define mpic_cpu_read(i) _mpic_cpu_read(mpic,(i)) -#define mpic_cpu_write(i,v) _mpic_cpu_write(mpic,(i),(v)) -#define mpic_irq_read(s,r) _mpic_irq_read(mpic,(s),(r)) -#define mpic_irq_write(s,r,v) _mpic_irq_write(mpic,(s),(r),(v)) - - -/* - * Low level utility functions - */ - - - -/* Check if we have one of those nice broken MPICs with a flipped endian on - * reads from IPI registers - */ -static void __init mpic_test_broken_ipi(struct mpic *mpic) -{ - u32 r; - - mpic_write(mpic->gregs, MPIC_GREG_IPI_VECTOR_PRI_0, MPIC_VECPRI_MASK); - r = mpic_read(mpic->gregs, MPIC_GREG_IPI_VECTOR_PRI_0); - - if (r == le32_to_cpu(MPIC_VECPRI_MASK)) { - printk(KERN_INFO "mpic: Detected reversed IPI registers\n"); - mpic->flags |= MPIC_BROKEN_IPI; - } -} - -#ifdef CONFIG_MPIC_BROKEN_U3 - -/* Test if an interrupt is sourced from HyperTransport (used on broken U3s) - * to force the edge setting on the MPIC and do the ack workaround. - */ -static inline int mpic_is_ht_interrupt(struct mpic *mpic, unsigned int source_no) -{ - if (source_no >= 128 || !mpic->fixups) - return 0; - return mpic->fixups[source_no].base != NULL; -} - -static inline void mpic_apic_end_irq(struct mpic *mpic, unsigned int source_no) -{ - struct mpic_irq_fixup *fixup = &mpic->fixups[source_no]; - u32 tmp; - - spin_lock(&mpic->fixup_lock); - writeb(0x11 + 2 * fixup->irq, fixup->base); - tmp = readl(fixup->base + 2); - writel(tmp | 0x80000000ul, fixup->base + 2); - /* config writes shouldn't be posted but let's be safe ... */ - (void)readl(fixup->base + 2); - spin_unlock(&mpic->fixup_lock); -} - - -static void __init mpic_amd8111_read_irq(struct mpic *mpic, u8 __iomem *devbase) -{ - int i, irq; - u32 tmp; - - printk(KERN_INFO "mpic: - Workarounds on AMD 8111 @ %p\n", devbase); - - for (i=0; i < 24; i++) { - writeb(0x10 + 2*i, devbase + 0xf2); - tmp = readl(devbase + 0xf4); - if ((tmp & 0x1) || !(tmp & 0x20)) - continue; - irq = (tmp >> 16) & 0xff; - mpic->fixups[irq].irq = i; - mpic->fixups[irq].base = devbase + 0xf2; - } -} - -static void __init mpic_amd8131_read_irq(struct mpic *mpic, u8 __iomem *devbase) -{ - int i, irq; - u32 tmp; - - printk(KERN_INFO "mpic: - Workarounds on AMD 8131 @ %p\n", devbase); - - for (i=0; i < 4; i++) { - writeb(0x10 + 2*i, devbase + 0xba); - tmp = readl(devbase + 0xbc); - if ((tmp & 0x1) || !(tmp & 0x20)) - continue; - irq = (tmp >> 16) & 0xff; - mpic->fixups[irq].irq = i; - mpic->fixups[irq].base = devbase + 0xba; - } -} - -static void __init mpic_scan_ioapics(struct mpic *mpic) -{ - unsigned int devfn; - u8 __iomem *cfgspace; - - printk(KERN_INFO "mpic: Setting up IO-APICs workarounds for U3\n"); - - /* Allocate fixups array */ - mpic->fixups = alloc_bootmem(128 * sizeof(struct mpic_irq_fixup)); - BUG_ON(mpic->fixups == NULL); - memset(mpic->fixups, 0, 128 * sizeof(struct mpic_irq_fixup)); - - /* Init spinlock */ - spin_lock_init(&mpic->fixup_lock); - - /* Map u3 config space. We assume all IO-APICs are on the primary bus - * and slot will never be above "0xf" so we only need to map 32k - */ - cfgspace = (unsigned char __iomem *)ioremap(0xf2000000, 0x8000); - BUG_ON(cfgspace == NULL); - - /* Now we scan all slots. We do a very quick scan, we read the header type, - * vendor ID and device ID only, that's plenty enough - */ - for (devfn = 0; devfn < PCI_DEVFN(0x10,0); devfn ++) { - u8 __iomem *devbase = cfgspace + (devfn << 8); - u8 hdr_type = readb(devbase + PCI_HEADER_TYPE); - u32 l = readl(devbase + PCI_VENDOR_ID); - u16 vendor_id, device_id; - int multifunc = 0; - - DBG("devfn %x, l: %x\n", devfn, l); - - /* If no device, skip */ - if (l == 0xffffffff || l == 0x00000000 || - l == 0x0000ffff || l == 0xffff0000) - goto next; - - /* Check if it's a multifunction device (only really used - * to function 0 though - */ - multifunc = !!(hdr_type & 0x80); - vendor_id = l & 0xffff; - device_id = (l >> 16) & 0xffff; - - /* If a known device, go to fixup setup code */ - if (vendor_id == PCI_VENDOR_ID_AMD && device_id == 0x7460) - mpic_amd8111_read_irq(mpic, devbase); - if (vendor_id == PCI_VENDOR_ID_AMD && device_id == 0x7450) - mpic_amd8131_read_irq(mpic, devbase); - next: - /* next device, if function 0 */ - if ((PCI_FUNC(devfn) == 0) && !multifunc) - devfn += 7; - } -} - -#endif /* CONFIG_MPIC_BROKEN_U3 */ - - -/* Find an mpic associated with a given linux interrupt */ -static struct mpic *mpic_find(unsigned int irq, unsigned int *is_ipi) -{ - struct mpic *mpic = mpics; - - while(mpic) { - /* search IPIs first since they may override the main interrupts */ - if (irq >= mpic->ipi_offset && irq < (mpic->ipi_offset + 4)) { - if (is_ipi) - *is_ipi = 1; - return mpic; - } - if (irq >= mpic->irq_offset && - irq < (mpic->irq_offset + mpic->irq_count)) { - if (is_ipi) - *is_ipi = 0; - return mpic; - } - mpic = mpic -> next; - } - return NULL; -} - -/* Convert a cpu mask from logical to physical cpu numbers. */ -static inline u32 mpic_physmask(u32 cpumask) -{ - int i; - u32 mask = 0; - - for (i = 0; i < NR_CPUS; ++i, cpumask >>= 1) - mask |= (cpumask & 1) << get_hard_smp_processor_id(i); - return mask; -} - -#ifdef CONFIG_SMP -/* Get the mpic structure from the IPI number */ -static inline struct mpic * mpic_from_ipi(unsigned int ipi) -{ - return container_of(irq_desc[ipi].handler, struct mpic, hc_ipi); -} -#endif - -/* Get the mpic structure from the irq number */ -static inline struct mpic * mpic_from_irq(unsigned int irq) -{ - return container_of(irq_desc[irq].handler, struct mpic, hc_irq); -} - -/* Send an EOI */ -static inline void mpic_eoi(struct mpic *mpic) -{ - mpic_cpu_write(MPIC_CPU_EOI, 0); - (void)mpic_cpu_read(MPIC_CPU_WHOAMI); -} - -#ifdef CONFIG_SMP -static irqreturn_t mpic_ipi_action(int irq, void *dev_id, struct pt_regs *regs) -{ - struct mpic *mpic = dev_id; - - smp_message_recv(irq - mpic->ipi_offset, regs); - return IRQ_HANDLED; -} -#endif /* CONFIG_SMP */ - -/* - * Linux descriptor level callbacks - */ - - -static void mpic_enable_irq(unsigned int irq) -{ - unsigned int loops = 100000; - struct mpic *mpic = mpic_from_irq(irq); - unsigned int src = irq - mpic->irq_offset; - - DBG("%s: enable_irq: %d (src %d)\n", mpic->name, irq, src); - - mpic_irq_write(src, MPIC_IRQ_VECTOR_PRI, - mpic_irq_read(src, MPIC_IRQ_VECTOR_PRI) & ~MPIC_VECPRI_MASK); - - /* make sure mask gets to controller before we return to user */ - do { - if (!loops--) { - printk(KERN_ERR "mpic_enable_irq timeout\n"); - break; - } - } while(mpic_irq_read(src, MPIC_IRQ_VECTOR_PRI) & MPIC_VECPRI_MASK); -} - -static void mpic_disable_irq(unsigned int irq) -{ - unsigned int loops = 100000; - struct mpic *mpic = mpic_from_irq(irq); - unsigned int src = irq - mpic->irq_offset; - - DBG("%s: disable_irq: %d (src %d)\n", mpic->name, irq, src); - - mpic_irq_write(src, MPIC_IRQ_VECTOR_PRI, - mpic_irq_read(src, MPIC_IRQ_VECTOR_PRI) | MPIC_VECPRI_MASK); - - /* make sure mask gets to controller before we return to user */ - do { - if (!loops--) { - printk(KERN_ERR "mpic_enable_irq timeout\n"); - break; - } - } while(!(mpic_irq_read(src, MPIC_IRQ_VECTOR_PRI) & MPIC_VECPRI_MASK)); -} - -static void mpic_end_irq(unsigned int irq) -{ - struct mpic *mpic = mpic_from_irq(irq); - - DBG("%s: end_irq: %d\n", mpic->name, irq); - - /* We always EOI on end_irq() even for edge interrupts since that - * should only lower the priority, the MPIC should have properly - * latched another edge interrupt coming in anyway - */ - -#ifdef CONFIG_MPIC_BROKEN_U3 - if (mpic->flags & MPIC_BROKEN_U3) { - unsigned int src = irq - mpic->irq_offset; - if (mpic_is_ht_interrupt(mpic, src)) - mpic_apic_end_irq(mpic, src); - } -#endif /* CONFIG_MPIC_BROKEN_U3 */ - - mpic_eoi(mpic); -} - -#ifdef CONFIG_SMP - -static void mpic_enable_ipi(unsigned int irq) -{ - struct mpic *mpic = mpic_from_ipi(irq); - unsigned int src = irq - mpic->ipi_offset; - - DBG("%s: enable_ipi: %d (ipi %d)\n", mpic->name, irq, src); - mpic_ipi_write(src, mpic_ipi_read(src) & ~MPIC_VECPRI_MASK); -} - -static void mpic_disable_ipi(unsigned int irq) -{ - /* NEVER disable an IPI... that's just plain wrong! */ -} - -static void mpic_end_ipi(unsigned int irq) -{ - struct mpic *mpic = mpic_from_ipi(irq); - - /* - * IPIs are marked IRQ_PER_CPU. This has the side effect of - * preventing the IRQ_PENDING/IRQ_INPROGRESS logic from - * applying to them. We EOI them late to avoid re-entering. - * We mark IPI's with SA_INTERRUPT as they must run with - * irqs disabled. - */ - mpic_eoi(mpic); -} - -#endif /* CONFIG_SMP */ - -static void mpic_set_affinity(unsigned int irq, cpumask_t cpumask) -{ - struct mpic *mpic = mpic_from_irq(irq); - - cpumask_t tmp; - - cpus_and(tmp, cpumask, cpu_online_map); - - mpic_irq_write(irq - mpic->irq_offset, MPIC_IRQ_DESTINATION, - mpic_physmask(cpus_addr(tmp)[0])); -} - - -/* - * Exported functions - */ - - -struct mpic * __init mpic_alloc(unsigned long phys_addr, - unsigned int flags, - unsigned int isu_size, - unsigned int irq_offset, - unsigned int irq_count, - unsigned int ipi_offset, - unsigned char *senses, - unsigned int senses_count, - const char *name) -{ - struct mpic *mpic; - u32 reg; - const char *vers; - int i; - - mpic = alloc_bootmem(sizeof(struct mpic)); - if (mpic == NULL) - return NULL; - - memset(mpic, 0, sizeof(struct mpic)); - mpic->name = name; - - mpic->hc_irq.typename = name; - mpic->hc_irq.enable = mpic_enable_irq; - mpic->hc_irq.disable = mpic_disable_irq; - mpic->hc_irq.end = mpic_end_irq; - if (flags & MPIC_PRIMARY) - mpic->hc_irq.set_affinity = mpic_set_affinity; -#ifdef CONFIG_SMP - mpic->hc_ipi.typename = name; - mpic->hc_ipi.enable = mpic_enable_ipi; - mpic->hc_ipi.disable = mpic_disable_ipi; - mpic->hc_ipi.end = mpic_end_ipi; -#endif /* CONFIG_SMP */ - - mpic->flags = flags; - mpic->isu_size = isu_size; - mpic->irq_offset = irq_offset; - mpic->irq_count = irq_count; - mpic->ipi_offset = ipi_offset; - mpic->num_sources = 0; /* so far */ - mpic->senses = senses; - mpic->senses_count = senses_count; - - /* Map the global registers */ - mpic->gregs = ioremap(phys_addr + MPIC_GREG_BASE, 0x1000); - mpic->tmregs = mpic->gregs + (MPIC_TIMER_BASE >> 2); - BUG_ON(mpic->gregs == NULL); - - /* Reset */ - if (flags & MPIC_WANTS_RESET) { - mpic_write(mpic->gregs, MPIC_GREG_GLOBAL_CONF_0, - mpic_read(mpic->gregs, MPIC_GREG_GLOBAL_CONF_0) - | MPIC_GREG_GCONF_RESET); - while( mpic_read(mpic->gregs, MPIC_GREG_GLOBAL_CONF_0) - & MPIC_GREG_GCONF_RESET) - mb(); - } - - /* Read feature register, calculate num CPUs and, for non-ISU - * MPICs, num sources as well. On ISU MPICs, sources are counted - * as ISUs are added - */ - reg = mpic_read(mpic->gregs, MPIC_GREG_FEATURE_0); - mpic->num_cpus = ((reg & MPIC_GREG_FEATURE_LAST_CPU_MASK) - >> MPIC_GREG_FEATURE_LAST_CPU_SHIFT) + 1; - if (isu_size == 0) - mpic->num_sources = ((reg & MPIC_GREG_FEATURE_LAST_SRC_MASK) - >> MPIC_GREG_FEATURE_LAST_SRC_SHIFT) + 1; - - /* Map the per-CPU registers */ - for (i = 0; i < mpic->num_cpus; i++) { - mpic->cpuregs[i] = ioremap(phys_addr + MPIC_CPU_BASE + - i * MPIC_CPU_STRIDE, 0x1000); - BUG_ON(mpic->cpuregs[i] == NULL); - } - - /* Initialize main ISU if none provided */ - if (mpic->isu_size == 0) { - mpic->isu_size = mpic->num_sources; - mpic->isus[0] = ioremap(phys_addr + MPIC_IRQ_BASE, - MPIC_IRQ_STRIDE * mpic->isu_size); - BUG_ON(mpic->isus[0] == NULL); - } - mpic->isu_shift = 1 + __ilog2(mpic->isu_size - 1); - mpic->isu_mask = (1 << mpic->isu_shift) - 1; - - /* Display version */ - switch (reg & MPIC_GREG_FEATURE_VERSION_MASK) { - case 1: - vers = "1.0"; - break; - case 2: - vers = "1.2"; - break; - case 3: - vers = "1.3"; - break; - default: - vers = "<unknown>"; - break; - } - printk(KERN_INFO "mpic: Setting up MPIC \"%s\" version %s at %lx, max %d CPUs\n", - name, vers, phys_addr, mpic->num_cpus); - printk(KERN_INFO "mpic: ISU size: %d, shift: %d, mask: %x\n", mpic->isu_size, - mpic->isu_shift, mpic->isu_mask); - - mpic->next = mpics; - mpics = mpic; - - if (flags & MPIC_PRIMARY) - mpic_primary = mpic; - - return mpic; -} - -void __init mpic_assign_isu(struct mpic *mpic, unsigned int isu_num, - unsigned long phys_addr) -{ - unsigned int isu_first = isu_num * mpic->isu_size; - - BUG_ON(isu_num >= MPIC_MAX_ISU); - - mpic->isus[isu_num] = ioremap(phys_addr, MPIC_IRQ_STRIDE * mpic->isu_size); - if ((isu_first + mpic->isu_size) > mpic->num_sources) - mpic->num_sources = isu_first + mpic->isu_size; -} - -void __init mpic_setup_cascade(unsigned int irq, mpic_cascade_t handler, - void *data) -{ - struct mpic *mpic = mpic_find(irq, NULL); - unsigned long flags; - - /* Synchronization here is a bit dodgy, so don't try to replace cascade - * interrupts on the fly too often ... but normally it's set up at boot. - */ - spin_lock_irqsave(&mpic_lock, flags); - if (mpic->cascade) - mpic_disable_irq(mpic->cascade_vec + mpic->irq_offset); - mpic->cascade = NULL; - wmb(); - mpic->cascade_vec = irq - mpic->irq_offset; - mpic->cascade_data = data; - wmb(); - mpic->cascade = handler; - mpic_enable_irq(irq); - spin_unlock_irqrestore(&mpic_lock, flags); -} - -void __init mpic_init(struct mpic *mpic) -{ - int i; - - BUG_ON(mpic->num_sources == 0); - - printk(KERN_INFO "mpic: Initializing for %d sources\n", mpic->num_sources); - - /* Set current processor priority to max */ - mpic_cpu_write(MPIC_CPU_CURRENT_TASK_PRI, 0xf); - - /* Initialize timers: just disable them all */ - for (i = 0; i < 4; i++) { - mpic_write(mpic->tmregs, - i * MPIC_TIMER_STRIDE + MPIC_TIMER_DESTINATION, 0); - mpic_write(mpic->tmregs, - i * MPIC_TIMER_STRIDE + MPIC_TIMER_VECTOR_PRI, - MPIC_VECPRI_MASK | - (MPIC_VEC_TIMER_0 + i)); - } - - /* Initialize IPIs to our reserved vectors and mark them disabled for now */ - mpic_test_broken_ipi(mpic); - for (i = 0; i < 4; i++) { - mpic_ipi_write(i, - MPIC_VECPRI_MASK | - (10 << MPIC_VECPRI_PRIORITY_SHIFT) | - (MPIC_VEC_IPI_0 + i)); -#ifdef CONFIG_SMP - if (!(mpic->flags & MPIC_PRIMARY)) - continue; - irq_desc[mpic->ipi_offset+i].status |= IRQ_PER_CPU; - irq_desc[mpic->ipi_offset+i].handler = &mpic->hc_ipi; - -#endif /* CONFIG_SMP */ - } - - /* Initialize interrupt sources */ - if (mpic->irq_count == 0) - mpic->irq_count = mpic->num_sources; - -#ifdef CONFIG_MPIC_BROKEN_U3 - /* Do the ioapic fixups on U3 broken mpic */ - DBG("MPIC flags: %x\n", mpic->flags); - if ((mpic->flags & MPIC_BROKEN_U3) && (mpic->flags & MPIC_PRIMARY)) - mpic_scan_ioapics(mpic); -#endif /* CONFIG_MPIC_BROKEN_U3 */ - - for (i = 0; i < mpic->num_sources; i++) { - /* start with vector = source number, and masked */ - u32 vecpri = MPIC_VECPRI_MASK | i | (8 << MPIC_VECPRI_PRIORITY_SHIFT); - int level = 0; - - /* if it's an IPI, we skip it */ - if ((mpic->irq_offset + i) >= (mpic->ipi_offset + i) && - (mpic->irq_offset + i) < (mpic->ipi_offset + i + 4)) - continue; - - /* do senses munging */ - if (mpic->senses && i < mpic->senses_count) { - if (mpic->senses[i] & IRQ_SENSE_LEVEL) - vecpri |= MPIC_VECPRI_SENSE_LEVEL; - if (mpic->senses[i] & IRQ_POLARITY_POSITIVE) - vecpri |= MPIC_VECPRI_POLARITY_POSITIVE; - } else - vecpri |= MPIC_VECPRI_SENSE_LEVEL; - - /* remember if it was a level interrupts */ - level = (vecpri & MPIC_VECPRI_SENSE_LEVEL); - - /* deal with broken U3 */ - if (mpic->flags & MPIC_BROKEN_U3) { -#ifdef CONFIG_MPIC_BROKEN_U3 - if (mpic_is_ht_interrupt(mpic, i)) { - vecpri &= ~(MPIC_VECPRI_SENSE_MASK | - MPIC_VECPRI_POLARITY_MASK); - vecpri |= MPIC_VECPRI_POLARITY_POSITIVE; - } -#else - printk(KERN_ERR "mpic: BROKEN_U3 set, but CONFIG doesn't match\n"); -#endif - } - - DBG("setup source %d, vecpri: %08x, level: %d\n", i, vecpri, - (level != 0)); - - /* init hw */ - mpic_irq_write(i, MPIC_IRQ_VECTOR_PRI, vecpri); - mpic_irq_write(i, MPIC_IRQ_DESTINATION, - 1 << get_hard_smp_processor_id(boot_cpuid)); - - /* init linux descriptors */ - if (i < mpic->irq_count) { - irq_desc[mpic->irq_offset+i].status = level ? IRQ_LEVEL : 0; - irq_desc[mpic->irq_offset+i].handler = &mpic->hc_irq; - } - } - - /* Init spurrious vector */ - mpic_write(mpic->gregs, MPIC_GREG_SPURIOUS, MPIC_VEC_SPURRIOUS); - - /* Disable 8259 passthrough */ - mpic_write(mpic->gregs, MPIC_GREG_GLOBAL_CONF_0, - mpic_read(mpic->gregs, MPIC_GREG_GLOBAL_CONF_0) - | MPIC_GREG_GCONF_8259_PTHROU_DIS); - - /* Set current processor priority to 0 */ - mpic_cpu_write(MPIC_CPU_CURRENT_TASK_PRI, 0); -} - - - -void mpic_irq_set_priority(unsigned int irq, unsigned int pri) -{ - int is_ipi; - struct mpic *mpic = mpic_find(irq, &is_ipi); - unsigned long flags; - u32 reg; - - spin_lock_irqsave(&mpic_lock, flags); - if (is_ipi) { - reg = mpic_ipi_read(irq - mpic->ipi_offset) & MPIC_VECPRI_PRIORITY_MASK; - mpic_ipi_write(irq - mpic->ipi_offset, - reg | (pri << MPIC_VECPRI_PRIORITY_SHIFT)); - } else { - reg = mpic_irq_read(irq - mpic->irq_offset, MPIC_IRQ_VECTOR_PRI) - & MPIC_VECPRI_PRIORITY_MASK; - mpic_irq_write(irq - mpic->irq_offset, MPIC_IRQ_VECTOR_PRI, - reg | (pri << MPIC_VECPRI_PRIORITY_SHIFT)); - } - spin_unlock_irqrestore(&mpic_lock, flags); -} - -unsigned int mpic_irq_get_priority(unsigned int irq) -{ - int is_ipi; - struct mpic *mpic = mpic_find(irq, &is_ipi); - unsigned long flags; - u32 reg; - - spin_lock_irqsave(&mpic_lock, flags); - if (is_ipi) - reg = mpic_ipi_read(irq - mpic->ipi_offset); - else - reg = mpic_irq_read(irq - mpic->irq_offset, MPIC_IRQ_VECTOR_PRI); - spin_unlock_irqrestore(&mpic_lock, flags); - return (reg & MPIC_VECPRI_PRIORITY_MASK) >> MPIC_VECPRI_PRIORITY_SHIFT; -} - -void mpic_setup_this_cpu(void) -{ -#ifdef CONFIG_SMP - struct mpic *mpic = mpic_primary; - unsigned long flags; - u32 msk = 1 << hard_smp_processor_id(); - unsigned int i; - - BUG_ON(mpic == NULL); - - DBG("%s: setup_this_cpu(%d)\n", mpic->name, hard_smp_processor_id()); - - spin_lock_irqsave(&mpic_lock, flags); - - /* let the mpic know we want intrs. default affinity is 0xffffffff - * until changed via /proc. That's how it's done on x86. If we want - * it differently, then we should make sure we also change the default - * values of irq_affinity in irq.c. - */ - if (distribute_irqs) { - for (i = 0; i < mpic->num_sources ; i++) - mpic_irq_write(i, MPIC_IRQ_DESTINATION, - mpic_irq_read(i, MPIC_IRQ_DESTINATION) | msk); - } - - /* Set current processor priority to 0 */ - mpic_cpu_write(MPIC_CPU_CURRENT_TASK_PRI, 0); - - spin_unlock_irqrestore(&mpic_lock, flags); -#endif /* CONFIG_SMP */ -} - -/* - * XXX: someone who knows mpic should check this. - * do we need to eoi the ipi including for kexec cpu here (see xics comments)? - * or can we reset the mpic in the new kernel? - */ -void mpic_teardown_this_cpu(int secondary) -{ - struct mpic *mpic = mpic_primary; - unsigned long flags; - u32 msk = 1 << hard_smp_processor_id(); - unsigned int i; - - BUG_ON(mpic == NULL); - - DBG("%s: teardown_this_cpu(%d)\n", mpic->name, hard_smp_processor_id()); - spin_lock_irqsave(&mpic_lock, flags); - - /* let the mpic know we don't want intrs. */ - for (i = 0; i < mpic->num_sources ; i++) - mpic_irq_write(i, MPIC_IRQ_DESTINATION, - mpic_irq_read(i, MPIC_IRQ_DESTINATION) & ~msk); - - /* Set current processor priority to max */ - mpic_cpu_write(MPIC_CPU_CURRENT_TASK_PRI, 0xf); - - spin_unlock_irqrestore(&mpic_lock, flags); -} - - -void mpic_send_ipi(unsigned int ipi_no, unsigned int cpu_mask) -{ - struct mpic *mpic = mpic_primary; - - BUG_ON(mpic == NULL); - - DBG("%s: send_ipi(ipi_no: %d)\n", mpic->name, ipi_no); - - mpic_cpu_write(MPIC_CPU_IPI_DISPATCH_0 + ipi_no * 0x10, - mpic_physmask(cpu_mask & cpus_addr(cpu_online_map)[0])); -} - -int mpic_get_one_irq(struct mpic *mpic, struct pt_regs *regs) -{ - u32 irq; - - irq = mpic_cpu_read(MPIC_CPU_INTACK) & MPIC_VECPRI_VECTOR_MASK; - DBG("%s: get_one_irq(): %d\n", mpic->name, irq); - - if (mpic->cascade && irq == mpic->cascade_vec) { - DBG("%s: cascading ...\n", mpic->name); - irq = mpic->cascade(regs, mpic->cascade_data); - mpic_eoi(mpic); - return irq; - } - if (unlikely(irq == MPIC_VEC_SPURRIOUS)) - return -1; - if (irq < MPIC_VEC_IPI_0) - return irq + mpic->irq_offset; - DBG("%s: ipi %d !\n", mpic->name, irq - MPIC_VEC_IPI_0); - return irq - MPIC_VEC_IPI_0 + mpic->ipi_offset; -} - -int mpic_get_irq(struct pt_regs *regs) -{ - struct mpic *mpic = mpic_primary; - - BUG_ON(mpic == NULL); - - return mpic_get_one_irq(mpic, regs); -} - - -#ifdef CONFIG_SMP -void mpic_request_ipis(void) -{ - struct mpic *mpic = mpic_primary; - - BUG_ON(mpic == NULL); - - printk("requesting IPIs ... \n"); - - /* IPIs are marked SA_INTERRUPT as they must run with irqs disabled */ - request_irq(mpic->ipi_offset+0, mpic_ipi_action, SA_INTERRUPT, - "IPI0 (call function)", mpic); - request_irq(mpic->ipi_offset+1, mpic_ipi_action, SA_INTERRUPT, - "IPI1 (reschedule)", mpic); - request_irq(mpic->ipi_offset+2, mpic_ipi_action, SA_INTERRUPT, - "IPI2 (unused)", mpic); - request_irq(mpic->ipi_offset+3, mpic_ipi_action, SA_INTERRUPT, - "IPI3 (debugger break)", mpic); - - printk("IPIs requested... \n"); -} -#endif /* CONFIG_SMP */ diff --git a/arch/ppc64/kernel/mpic.h b/arch/ppc64/kernel/mpic.h deleted file mode 100644 index ca78a7f10528..000000000000 --- a/arch/ppc64/kernel/mpic.h +++ /dev/null @@ -1,273 +0,0 @@ -#include <linux/irq.h> - -/* - * Global registers - */ - -#define MPIC_GREG_BASE 0x01000 - -#define MPIC_GREG_FEATURE_0 0x00000 -#define MPIC_GREG_FEATURE_LAST_SRC_MASK 0x07ff0000 -#define MPIC_GREG_FEATURE_LAST_SRC_SHIFT 16 -#define MPIC_GREG_FEATURE_LAST_CPU_MASK 0x00001f00 -#define MPIC_GREG_FEATURE_LAST_CPU_SHIFT 8 -#define MPIC_GREG_FEATURE_VERSION_MASK 0xff -#define MPIC_GREG_FEATURE_1 0x00010 -#define MPIC_GREG_GLOBAL_CONF_0 0x00020 -#define MPIC_GREG_GCONF_RESET 0x80000000 -#define MPIC_GREG_GCONF_8259_PTHROU_DIS 0x20000000 -#define MPIC_GREG_GCONF_BASE_MASK 0x000fffff -#define MPIC_GREG_GLOBAL_CONF_1 0x00030 -#define MPIC_GREG_VENDOR_0 0x00040 -#define MPIC_GREG_VENDOR_1 0x00050 -#define MPIC_GREG_VENDOR_2 0x00060 -#define MPIC_GREG_VENDOR_3 0x00070 -#define MPIC_GREG_VENDOR_ID 0x00080 -#define MPIC_GREG_VENDOR_ID_STEPPING_MASK 0x00ff0000 -#define MPIC_GREG_VENDOR_ID_STEPPING_SHIFT 16 -#define MPIC_GREG_VENDOR_ID_DEVICE_ID_MASK 0x0000ff00 -#define MPIC_GREG_VENDOR_ID_DEVICE_ID_SHIFT 8 -#define MPIC_GREG_VENDOR_ID_VENDOR_ID_MASK 0x000000ff -#define MPIC_GREG_PROCESSOR_INIT 0x00090 -#define MPIC_GREG_IPI_VECTOR_PRI_0 0x000a0 -#define MPIC_GREG_IPI_VECTOR_PRI_1 0x000b0 -#define MPIC_GREG_IPI_VECTOR_PRI_2 0x000c0 -#define MPIC_GREG_IPI_VECTOR_PRI_3 0x000d0 -#define MPIC_GREG_SPURIOUS 0x000e0 -#define MPIC_GREG_TIMER_FREQ 0x000f0 - -/* - * - * Timer registers - */ -#define MPIC_TIMER_BASE 0x01100 -#define MPIC_TIMER_STRIDE 0x40 - -#define MPIC_TIMER_CURRENT_CNT 0x00000 -#define MPIC_TIMER_BASE_CNT 0x00010 -#define MPIC_TIMER_VECTOR_PRI 0x00020 -#define MPIC_TIMER_DESTINATION 0x00030 - -/* - * Per-Processor registers - */ - -#define MPIC_CPU_THISBASE 0x00000 -#define MPIC_CPU_BASE 0x20000 -#define MPIC_CPU_STRIDE 0x01000 - -#define MPIC_CPU_IPI_DISPATCH_0 0x00040 -#define MPIC_CPU_IPI_DISPATCH_1 0x00050 -#define MPIC_CPU_IPI_DISPATCH_2 0x00060 -#define MPIC_CPU_IPI_DISPATCH_3 0x00070 -#define MPIC_CPU_CURRENT_TASK_PRI 0x00080 -#define MPIC_CPU_TASKPRI_MASK 0x0000000f -#define MPIC_CPU_WHOAMI 0x00090 -#define MPIC_CPU_WHOAMI_MASK 0x0000001f -#define MPIC_CPU_INTACK 0x000a0 -#define MPIC_CPU_EOI 0x000b0 - -/* - * Per-source registers - */ - -#define MPIC_IRQ_BASE 0x10000 -#define MPIC_IRQ_STRIDE 0x00020 -#define MPIC_IRQ_VECTOR_PRI 0x00000 -#define MPIC_VECPRI_MASK 0x80000000 -#define MPIC_VECPRI_ACTIVITY 0x40000000 /* Read Only */ -#define MPIC_VECPRI_PRIORITY_MASK 0x000f0000 -#define MPIC_VECPRI_PRIORITY_SHIFT 16 -#define MPIC_VECPRI_VECTOR_MASK 0x000007ff -#define MPIC_VECPRI_POLARITY_POSITIVE 0x00800000 -#define MPIC_VECPRI_POLARITY_NEGATIVE 0x00000000 -#define MPIC_VECPRI_POLARITY_MASK 0x00800000 -#define MPIC_VECPRI_SENSE_LEVEL 0x00400000 -#define MPIC_VECPRI_SENSE_EDGE 0x00000000 -#define MPIC_VECPRI_SENSE_MASK 0x00400000 -#define MPIC_IRQ_DESTINATION 0x00010 - -#define MPIC_MAX_IRQ_SOURCES 2048 -#define MPIC_MAX_CPUS 32 -#define MPIC_MAX_ISU 32 - -/* - * Special vector numbers (internal use only) - */ -#define MPIC_VEC_SPURRIOUS 255 -#define MPIC_VEC_IPI_3 254 -#define MPIC_VEC_IPI_2 253 -#define MPIC_VEC_IPI_1 252 -#define MPIC_VEC_IPI_0 251 - -/* unused */ -#define MPIC_VEC_TIMER_3 250 -#define MPIC_VEC_TIMER_2 249 -#define MPIC_VEC_TIMER_1 248 -#define MPIC_VEC_TIMER_0 247 - -/* Type definition of the cascade handler */ -typedef int (*mpic_cascade_t)(struct pt_regs *regs, void *data); - -#ifdef CONFIG_MPIC_BROKEN_U3 -/* Fixup table entry */ -struct mpic_irq_fixup -{ - u8 __iomem *base; - unsigned int irq; -}; -#endif /* CONFIG_MPIC_BROKEN_U3 */ - - -/* The instance data of a given MPIC */ -struct mpic -{ - /* The "linux" controller struct */ - hw_irq_controller hc_irq; -#ifdef CONFIG_SMP - hw_irq_controller hc_ipi; -#endif - const char *name; - /* Flags */ - unsigned int flags; - /* How many irq sources in a given ISU */ - unsigned int isu_size; - unsigned int isu_shift; - unsigned int isu_mask; - /* Offset of irq vector numbers */ - unsigned int irq_offset; - unsigned int irq_count; - /* Offset of ipi vector numbers */ - unsigned int ipi_offset; - /* Number of sources */ - unsigned int num_sources; - /* Number of CPUs */ - unsigned int num_cpus; - /* cascade handler */ - mpic_cascade_t cascade; - void *cascade_data; - unsigned int cascade_vec; - /* senses array */ - unsigned char *senses; - unsigned int senses_count; - -#ifdef CONFIG_MPIC_BROKEN_U3 - /* The fixup table */ - struct mpic_irq_fixup *fixups; - spinlock_t fixup_lock; -#endif - - /* The various ioremap'ed bases */ - volatile u32 __iomem *gregs; - volatile u32 __iomem *tmregs; - volatile u32 __iomem *cpuregs[MPIC_MAX_CPUS]; - volatile u32 __iomem *isus[MPIC_MAX_ISU]; - - /* link */ - struct mpic *next; -}; - -/* This is the primary controller, only that one has IPIs and - * has afinity control. A non-primary MPIC always uses CPU0 - * registers only - */ -#define MPIC_PRIMARY 0x00000001 -/* Set this for a big-endian MPIC */ -#define MPIC_BIG_ENDIAN 0x00000002 -/* Broken U3 MPIC */ -#define MPIC_BROKEN_U3 0x00000004 -/* Broken IPI registers (autodetected) */ -#define MPIC_BROKEN_IPI 0x00000008 -/* MPIC wants a reset */ -#define MPIC_WANTS_RESET 0x00000010 - -/* Allocate the controller structure and setup the linux irq descs - * for the range if interrupts passed in. No HW initialization is - * actually performed. - * - * @phys_addr: physial base address of the MPIC - * @flags: flags, see constants above - * @isu_size: number of interrupts in an ISU. Use 0 to use a - * standard ISU-less setup (aka powermac) - * @irq_offset: first irq number to assign to this mpic - * @irq_count: number of irqs to use with this mpic IRQ sources. Pass 0 - * to match the number of sources - * @ipi_offset: first irq number to assign to this mpic IPI sources, - * used only on primary mpic - * @senses: array of sense values - * @senses_num: number of entries in the array - * - * Note about the sense array. If none is passed, all interrupts are - * setup to be level negative unless MPIC_BROKEN_U3 is set in which - * case they are edge positive (and the array is ignored anyway). - * The values in the array start at the first source of the MPIC, - * that is senses[0] correspond to linux irq "irq_offset". - */ -extern struct mpic *mpic_alloc(unsigned long phys_addr, - unsigned int flags, - unsigned int isu_size, - unsigned int irq_offset, - unsigned int irq_count, - unsigned int ipi_offset, - unsigned char *senses, - unsigned int senses_num, - const char *name); - -/* Assign ISUs, to call before mpic_init() - * - * @mpic: controller structure as returned by mpic_alloc() - * @isu_num: ISU number - * @phys_addr: physical address of the ISU - */ -extern void mpic_assign_isu(struct mpic *mpic, unsigned int isu_num, - unsigned long phys_addr); - -/* Initialize the controller. After this has been called, none of the above - * should be called again for this mpic - */ -extern void mpic_init(struct mpic *mpic); - -/* Setup a cascade. Currently, only one cascade is supported this - * way, though you can always do a normal request_irq() and add - * other cascades this way. You should call this _after_ having - * added all the ISUs - * - * @irq_no: "linux" irq number of the cascade (that is offset'ed vector) - * @handler: cascade handler function - */ -extern void mpic_setup_cascade(unsigned int irq_no, mpic_cascade_t hanlder, - void *data); - -/* - * All of the following functions must only be used after the - * ISUs have been assigned and the controller fully initialized - * with mpic_init() - */ - - -/* Change/Read the priority of an interrupt. Default is 8 for irqs and - * 10 for IPIs. You can call this on both IPIs and IRQ numbers, but the - * IPI number is then the offset'ed (linux irq number mapped to the IPI) - */ -extern void mpic_irq_set_priority(unsigned int irq, unsigned int pri); -extern unsigned int mpic_irq_get_priority(unsigned int irq); - -/* Setup a non-boot CPU */ -extern void mpic_setup_this_cpu(void); - -/* Clean up for kexec (or cpu offline or ...) */ -extern void mpic_teardown_this_cpu(int secondary); - -/* Request IPIs on primary mpic */ -extern void mpic_request_ipis(void); - -/* Send an IPI (non offseted number 0..3) */ -extern void mpic_send_ipi(unsigned int ipi_no, unsigned int cpu_mask); - -/* Fetch interrupt from a given mpic */ -extern int mpic_get_one_irq(struct mpic *mpic, struct pt_regs *regs); -/* This one gets to the primary mpic */ -extern int mpic_get_irq(struct pt_regs *regs); - -/* global mpic for pSeries */ -extern struct mpic *pSeries_mpic; diff --git a/arch/ppc64/kernel/of_device.c b/arch/ppc64/kernel/of_device.c deleted file mode 100644 index 9f200f0f2ad5..000000000000 --- a/arch/ppc64/kernel/of_device.c +++ /dev/null @@ -1,274 +0,0 @@ -#include <linux/config.h> -#include <linux/string.h> -#include <linux/kernel.h> -#include <linux/init.h> -#include <linux/module.h> -#include <linux/mod_devicetable.h> -#include <asm/errno.h> -#include <asm/of_device.h> - -/** - * of_match_device - Tell if an of_device structure has a matching - * of_match structure - * @ids: array of of device match structures to search in - * @dev: the of device structure to match against - * - * Used by a driver to check whether an of_device present in the - * system is in its list of supported devices. - */ -const struct of_device_id *of_match_device(const struct of_device_id *matches, - const struct of_device *dev) -{ - if (!dev->node) - return NULL; - while (matches->name[0] || matches->type[0] || matches->compatible[0]) { - int match = 1; - if (matches->name[0]) - match &= dev->node->name - && !strcmp(matches->name, dev->node->name); - if (matches->type[0]) - match &= dev->node->type - && !strcmp(matches->type, dev->node->type); - if (matches->compatible[0]) - match &= device_is_compatible(dev->node, - matches->compatible); - if (match) - return matches; - matches++; - } - return NULL; -} - -static int of_platform_bus_match(struct device *dev, struct device_driver *drv) -{ - struct of_device * of_dev = to_of_device(dev); - struct of_platform_driver * of_drv = to_of_platform_driver(drv); - const struct of_device_id * matches = of_drv->match_table; - - if (!matches) - return 0; - - return of_match_device(matches, of_dev) != NULL; -} - -struct of_device *of_dev_get(struct of_device *dev) -{ - struct device *tmp; - - if (!dev) - return NULL; - tmp = get_device(&dev->dev); - if (tmp) - return to_of_device(tmp); - else - return NULL; -} - -void of_dev_put(struct of_device *dev) -{ - if (dev) - put_device(&dev->dev); -} - - -static int of_device_probe(struct device *dev) -{ - int error = -ENODEV; - struct of_platform_driver *drv; - struct of_device *of_dev; - const struct of_device_id *match; - - drv = to_of_platform_driver(dev->driver); - of_dev = to_of_device(dev); - - if (!drv->probe) - return error; - - of_dev_get(of_dev); - - match = of_match_device(drv->match_table, of_dev); - if (match) - error = drv->probe(of_dev, match); - if (error) - of_dev_put(of_dev); - - return error; -} - -static int of_device_remove(struct device *dev) -{ - struct of_device * of_dev = to_of_device(dev); - struct of_platform_driver * drv = to_of_platform_driver(dev->driver); - - if (dev->driver && drv->remove) - drv->remove(of_dev); - return 0; -} - -static int of_device_suspend(struct device *dev, pm_message_t state) -{ - struct of_device * of_dev = to_of_device(dev); - struct of_platform_driver * drv = to_of_platform_driver(dev->driver); - int error = 0; - - if (dev->driver && drv->suspend) - error = drv->suspend(of_dev, state); - return error; -} - -static int of_device_resume(struct device * dev) -{ - struct of_device * of_dev = to_of_device(dev); - struct of_platform_driver * drv = to_of_platform_driver(dev->driver); - int error = 0; - - if (dev->driver && drv->resume) - error = drv->resume(of_dev); - return error; -} - -struct bus_type of_platform_bus_type = { - .name = "of_platform", - .match = of_platform_bus_match, - .suspend = of_device_suspend, - .resume = of_device_resume, -}; - -static int __init of_bus_driver_init(void) -{ - return bus_register(&of_platform_bus_type); -} - -postcore_initcall(of_bus_driver_init); - -int of_register_driver(struct of_platform_driver *drv) -{ - int count = 0; - - /* initialize common driver fields */ - drv->driver.name = drv->name; - drv->driver.bus = &of_platform_bus_type; - drv->driver.probe = of_device_probe; - drv->driver.remove = of_device_remove; - - /* register with core */ - count = driver_register(&drv->driver); - return count ? count : 1; -} - -void of_unregister_driver(struct of_platform_driver *drv) -{ - driver_unregister(&drv->driver); -} - - -static ssize_t dev_show_devspec(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct of_device *ofdev; - - ofdev = to_of_device(dev); - return sprintf(buf, "%s", ofdev->node->full_name); -} - -static DEVICE_ATTR(devspec, S_IRUGO, dev_show_devspec, NULL); - -/** - * of_release_dev - free an of device structure when all users of it are finished. - * @dev: device that's been disconnected - * - * Will be called only by the device core when all users of this of device are - * done. - */ -void of_release_dev(struct device *dev) -{ - struct of_device *ofdev; - - ofdev = to_of_device(dev); - kfree(ofdev); -} - -int of_device_register(struct of_device *ofdev) -{ - int rc; - struct of_device **odprop; - - BUG_ON(ofdev->node == NULL); - - odprop = (struct of_device **)get_property(ofdev->node, "linux,device", NULL); - if (!odprop) { - struct property *new_prop; - - new_prop = kmalloc(sizeof(struct property) + sizeof(struct of_device *), - GFP_KERNEL); - if (new_prop == NULL) - return -ENOMEM; - new_prop->name = "linux,device"; - new_prop->length = sizeof(sizeof(struct of_device *)); - new_prop->value = (unsigned char *)&new_prop[1]; - odprop = (struct of_device **)new_prop->value; - *odprop = NULL; - prom_add_property(ofdev->node, new_prop); - } - *odprop = ofdev; - - rc = device_register(&ofdev->dev); - if (rc) - return rc; - - device_create_file(&ofdev->dev, &dev_attr_devspec); - - return 0; -} - -void of_device_unregister(struct of_device *ofdev) -{ - struct of_device **odprop; - - device_remove_file(&ofdev->dev, &dev_attr_devspec); - - odprop = (struct of_device **)get_property(ofdev->node, "linux,device", NULL); - if (odprop) - *odprop = NULL; - - device_unregister(&ofdev->dev); -} - -struct of_device* of_platform_device_create(struct device_node *np, - const char *bus_id, - struct device *parent) -{ - struct of_device *dev; - - dev = kmalloc(sizeof(*dev), GFP_KERNEL); - if (!dev) - return NULL; - memset(dev, 0, sizeof(*dev)); - - dev->node = np; - dev->dma_mask = 0xffffffffUL; - dev->dev.dma_mask = &dev->dma_mask; - dev->dev.parent = parent; - dev->dev.bus = &of_platform_bus_type; - dev->dev.release = of_release_dev; - - strlcpy(dev->dev.bus_id, bus_id, BUS_ID_SIZE); - - if (of_device_register(dev) != 0) { - kfree(dev); - return NULL; - } - - return dev; -} - - -EXPORT_SYMBOL(of_match_device); -EXPORT_SYMBOL(of_platform_bus_type); -EXPORT_SYMBOL(of_register_driver); -EXPORT_SYMBOL(of_unregister_driver); -EXPORT_SYMBOL(of_device_register); -EXPORT_SYMBOL(of_device_unregister); -EXPORT_SYMBOL(of_dev_get); -EXPORT_SYMBOL(of_dev_put); -EXPORT_SYMBOL(of_platform_device_create); -EXPORT_SYMBOL(of_release_dev); diff --git a/arch/ppc64/kernel/pSeries_hvCall.S b/arch/ppc64/kernel/pSeries_hvCall.S deleted file mode 100644 index 176e8da76466..000000000000 --- a/arch/ppc64/kernel/pSeries_hvCall.S +++ /dev/null @@ -1,131 +0,0 @@ -/* - * arch/ppc64/kernel/pSeries_hvCall.S - * - * This file contains the generic code to perform a call to the - * pSeries LPAR hypervisor. - * NOTE: this file will go away when we move to inline this work. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#include <asm/hvcall.h> -#include <asm/processor.h> -#include <asm/ppc_asm.h> - -#define STK_PARM(i) (48 + ((i)-3)*8) - - .text - -/* long plpar_hcall(unsigned long opcode, R3 - unsigned long arg1, R4 - unsigned long arg2, R5 - unsigned long arg3, R6 - unsigned long arg4, R7 - unsigned long *out1, R8 - unsigned long *out2, R9 - unsigned long *out3); R10 - */ -_GLOBAL(plpar_hcall) - HMT_MEDIUM - - mfcr r0 - - std r8,STK_PARM(r8)(r1) /* Save out ptrs */ - std r9,STK_PARM(r9)(r1) - std r10,STK_PARM(r10)(r1) - - stw r0,8(r1) - - HVSC /* invoke the hypervisor */ - - lwz r0,8(r1) - - ld r8,STK_PARM(r8)(r1) /* Fetch r4-r6 ret args */ - ld r9,STK_PARM(r9)(r1) - ld r10,STK_PARM(r10)(r1) - std r4,0(r8) - std r5,0(r9) - std r6,0(r10) - - mtcrf 0xff,r0 - blr /* return r3 = status */ - - -/* Simple interface with no output values (other than status) */ -_GLOBAL(plpar_hcall_norets) - HMT_MEDIUM - - mfcr r0 - stw r0,8(r1) - - HVSC /* invoke the hypervisor */ - - lwz r0,8(r1) - mtcrf 0xff,r0 - blr /* return r3 = status */ - - -/* long plpar_hcall_8arg_2ret(unsigned long opcode, R3 - unsigned long arg1, R4 - unsigned long arg2, R5 - unsigned long arg3, R6 - unsigned long arg4, R7 - unsigned long arg5, R8 - unsigned long arg6, R9 - unsigned long arg7, R10 - unsigned long arg8, 112(R1) - unsigned long *out1); 120(R1) - */ -_GLOBAL(plpar_hcall_8arg_2ret) - HMT_MEDIUM - - mfcr r0 - ld r11,STK_PARM(r11)(r1) /* put arg8 in R11 */ - stw r0,8(r1) - - HVSC /* invoke the hypervisor */ - - lwz r0,8(r1) - ld r10,STK_PARM(r12)(r1) /* Fetch r4 ret arg */ - std r4,0(r10) - mtcrf 0xff,r0 - blr /* return r3 = status */ - - -/* long plpar_hcall_4out(unsigned long opcode, R3 - unsigned long arg1, R4 - unsigned long arg2, R5 - unsigned long arg3, R6 - unsigned long arg4, R7 - unsigned long *out1, R8 - unsigned long *out2, R9 - unsigned long *out3, R10 - unsigned long *out4); 112(R1) - */ -_GLOBAL(plpar_hcall_4out) - HMT_MEDIUM - - mfcr r0 - stw r0,8(r1) - - std r8,STK_PARM(r8)(r1) /* Save out ptrs */ - std r9,STK_PARM(r9)(r1) - std r10,STK_PARM(r10)(r1) - - HVSC /* invoke the hypervisor */ - - lwz r0,8(r1) - - ld r8,STK_PARM(r8)(r1) /* Fetch r4-r7 ret args */ - ld r9,STK_PARM(r9)(r1) - ld r10,STK_PARM(r10)(r1) - ld r11,STK_PARM(r11)(r1) - std r4,0(r8) - std r5,0(r9) - std r6,0(r10) - std r7,0(r11) - - mtcrf 0xff,r0 - blr /* return r3 = status */ diff --git a/arch/ppc64/kernel/pSeries_iommu.c b/arch/ppc64/kernel/pSeries_iommu.c deleted file mode 100644 index d17f0108a032..000000000000 --- a/arch/ppc64/kernel/pSeries_iommu.c +++ /dev/null @@ -1,590 +0,0 @@ -/* - * arch/ppc64/kernel/pSeries_iommu.c - * - * Copyright (C) 2001 Mike Corrigan & Dave Engebretsen, IBM Corporation - * - * Rewrite, cleanup: - * - * Copyright (C) 2004 Olof Johansson <olof@austin.ibm.com>, IBM Corporation - * - * Dynamic DMA mapping support, pSeries-specific parts, both SMP and LPAR. - * - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - */ - -#include <linux/config.h> -#include <linux/init.h> -#include <linux/types.h> -#include <linux/slab.h> -#include <linux/mm.h> -#include <linux/spinlock.h> -#include <linux/string.h> -#include <linux/pci.h> -#include <linux/dma-mapping.h> -#include <asm/io.h> -#include <asm/prom.h> -#include <asm/rtas.h> -#include <asm/ppcdebug.h> -#include <asm/iommu.h> -#include <asm/pci-bridge.h> -#include <asm/machdep.h> -#include <asm/abs_addr.h> -#include <asm/plpar_wrappers.h> -#include <asm/pSeries_reconfig.h> -#include <asm/systemcfg.h> -#include <asm/firmware.h> -#include "pci.h" - -#define DBG(fmt...) - -extern int is_python(struct device_node *); - -static void tce_build_pSeries(struct iommu_table *tbl, long index, - long npages, unsigned long uaddr, - enum dma_data_direction direction) -{ - union tce_entry t; - union tce_entry *tp; - - t.te_word = 0; - t.te_rdwr = 1; // Read allowed - - if (direction != DMA_TO_DEVICE) - t.te_pciwr = 1; - - tp = ((union tce_entry *)tbl->it_base) + index; - - while (npages--) { - /* can't move this out since we might cross LMB boundary */ - t.te_rpn = (virt_to_abs(uaddr)) >> PAGE_SHIFT; - - tp->te_word = t.te_word; - - uaddr += PAGE_SIZE; - tp++; - } -} - - -static void tce_free_pSeries(struct iommu_table *tbl, long index, long npages) -{ - union tce_entry t; - union tce_entry *tp; - - t.te_word = 0; - tp = ((union tce_entry *)tbl->it_base) + index; - - while (npages--) { - tp->te_word = t.te_word; - - tp++; - } -} - - -static void tce_build_pSeriesLP(struct iommu_table *tbl, long tcenum, - long npages, unsigned long uaddr, - enum dma_data_direction direction) -{ - u64 rc; - union tce_entry tce; - - tce.te_word = 0; - tce.te_rpn = (virt_to_abs(uaddr)) >> PAGE_SHIFT; - tce.te_rdwr = 1; - if (direction != DMA_TO_DEVICE) - tce.te_pciwr = 1; - - while (npages--) { - rc = plpar_tce_put((u64)tbl->it_index, - (u64)tcenum << 12, - tce.te_word ); - - if (rc && printk_ratelimit()) { - printk("tce_build_pSeriesLP: plpar_tce_put failed. rc=%ld\n", rc); - printk("\tindex = 0x%lx\n", (u64)tbl->it_index); - printk("\ttcenum = 0x%lx\n", (u64)tcenum); - printk("\ttce val = 0x%lx\n", tce.te_word ); - show_stack(current, (unsigned long *)__get_SP()); - } - - tcenum++; - tce.te_rpn++; - } -} - -static DEFINE_PER_CPU(void *, tce_page) = NULL; - -static void tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum, - long npages, unsigned long uaddr, - enum dma_data_direction direction) -{ - u64 rc; - union tce_entry tce, *tcep; - long l, limit; - - if (npages == 1) - return tce_build_pSeriesLP(tbl, tcenum, npages, uaddr, - direction); - - tcep = __get_cpu_var(tce_page); - - /* This is safe to do since interrupts are off when we're called - * from iommu_alloc{,_sg}() - */ - if (!tcep) { - tcep = (void *)__get_free_page(GFP_ATOMIC); - /* If allocation fails, fall back to the loop implementation */ - if (!tcep) - return tce_build_pSeriesLP(tbl, tcenum, npages, - uaddr, direction); - __get_cpu_var(tce_page) = tcep; - } - - tce.te_word = 0; - tce.te_rpn = (virt_to_abs(uaddr)) >> PAGE_SHIFT; - tce.te_rdwr = 1; - if (direction != DMA_TO_DEVICE) - tce.te_pciwr = 1; - - /* We can map max one pageful of TCEs at a time */ - do { - /* - * Set up the page with TCE data, looping through and setting - * the values. - */ - limit = min_t(long, npages, PAGE_SIZE/sizeof(union tce_entry)); - - for (l = 0; l < limit; l++) { - tcep[l] = tce; - tce.te_rpn++; - } - - rc = plpar_tce_put_indirect((u64)tbl->it_index, - (u64)tcenum << 12, - (u64)virt_to_abs(tcep), - limit); - - npages -= limit; - tcenum += limit; - } while (npages > 0 && !rc); - - if (rc && printk_ratelimit()) { - printk("tce_buildmulti_pSeriesLP: plpar_tce_put failed. rc=%ld\n", rc); - printk("\tindex = 0x%lx\n", (u64)tbl->it_index); - printk("\tnpages = 0x%lx\n", (u64)npages); - printk("\ttce[0] val = 0x%lx\n", tcep[0].te_word); - show_stack(current, (unsigned long *)__get_SP()); - } -} - -static void tce_free_pSeriesLP(struct iommu_table *tbl, long tcenum, long npages) -{ - u64 rc; - union tce_entry tce; - - tce.te_word = 0; - - while (npages--) { - rc = plpar_tce_put((u64)tbl->it_index, - (u64)tcenum << 12, - tce.te_word); - - if (rc && printk_ratelimit()) { - printk("tce_free_pSeriesLP: plpar_tce_put failed. rc=%ld\n", rc); - printk("\tindex = 0x%lx\n", (u64)tbl->it_index); - printk("\ttcenum = 0x%lx\n", (u64)tcenum); - printk("\ttce val = 0x%lx\n", tce.te_word ); - show_stack(current, (unsigned long *)__get_SP()); - } - - tcenum++; - } -} - - -static void tce_freemulti_pSeriesLP(struct iommu_table *tbl, long tcenum, long npages) -{ - u64 rc; - union tce_entry tce; - - tce.te_word = 0; - - rc = plpar_tce_stuff((u64)tbl->it_index, - (u64)tcenum << 12, - tce.te_word, - npages); - - if (rc && printk_ratelimit()) { - printk("tce_freemulti_pSeriesLP: plpar_tce_stuff failed\n"); - printk("\trc = %ld\n", rc); - printk("\tindex = 0x%lx\n", (u64)tbl->it_index); - printk("\tnpages = 0x%lx\n", (u64)npages); - printk("\ttce val = 0x%lx\n", tce.te_word ); - show_stack(current, (unsigned long *)__get_SP()); - } -} - -static void iommu_table_setparms(struct pci_controller *phb, - struct device_node *dn, - struct iommu_table *tbl) -{ - struct device_node *node; - unsigned long *basep; - unsigned int *sizep; - - node = (struct device_node *)phb->arch_data; - - basep = (unsigned long *)get_property(node, "linux,tce-base", NULL); - sizep = (unsigned int *)get_property(node, "linux,tce-size", NULL); - if (basep == NULL || sizep == NULL) { - printk(KERN_ERR "PCI_DMA: iommu_table_setparms: %s has " - "missing tce entries !\n", dn->full_name); - return; - } - - tbl->it_base = (unsigned long)__va(*basep); - memset((void *)tbl->it_base, 0, *sizep); - - tbl->it_busno = phb->bus->number; - - /* Units of tce entries */ - tbl->it_offset = phb->dma_window_base_cur >> PAGE_SHIFT; - - /* Test if we are going over 2GB of DMA space */ - if (phb->dma_window_base_cur + phb->dma_window_size > 0x80000000ul) { - udbg_printf("PCI_DMA: Unexpected number of IOAs under this PHB.\n"); - panic("PCI_DMA: Unexpected number of IOAs under this PHB.\n"); - } - - phb->dma_window_base_cur += phb->dma_window_size; - - /* Set the tce table size - measured in entries */ - tbl->it_size = phb->dma_window_size >> PAGE_SHIFT; - - tbl->it_index = 0; - tbl->it_blocksize = 16; - tbl->it_type = TCE_PCI; -} - -/* - * iommu_table_setparms_lpar - * - * Function: On pSeries LPAR systems, return TCE table info, given a pci bus. - * - * ToDo: properly interpret the ibm,dma-window property. The definition is: - * logical-bus-number (1 word) - * phys-address (#address-cells words) - * size (#cell-size words) - * - * Currently we hard code these sizes (more or less). - */ -static void iommu_table_setparms_lpar(struct pci_controller *phb, - struct device_node *dn, - struct iommu_table *tbl, - unsigned int *dma_window) -{ - tbl->it_busno = PCI_DN(dn)->bussubno; - - /* TODO: Parse field size properties properly. */ - tbl->it_size = (((unsigned long)dma_window[4] << 32) | - (unsigned long)dma_window[5]) >> PAGE_SHIFT; - tbl->it_offset = (((unsigned long)dma_window[2] << 32) | - (unsigned long)dma_window[3]) >> PAGE_SHIFT; - tbl->it_base = 0; - tbl->it_index = dma_window[0]; - tbl->it_blocksize = 16; - tbl->it_type = TCE_PCI; -} - -static void iommu_bus_setup_pSeries(struct pci_bus *bus) -{ - struct device_node *dn; - struct iommu_table *tbl; - struct device_node *isa_dn, *isa_dn_orig; - struct device_node *tmp; - struct pci_dn *pci; - int children; - - DBG("iommu_bus_setup_pSeries, bus %p, bus->self %p\n", bus, bus->self); - - dn = pci_bus_to_OF_node(bus); - pci = PCI_DN(dn); - - if (bus->self) { - /* This is not a root bus, any setup will be done for the - * device-side of the bridge in iommu_dev_setup_pSeries(). - */ - return; - } - - /* Check if the ISA bus on the system is under - * this PHB. - */ - isa_dn = isa_dn_orig = of_find_node_by_type(NULL, "isa"); - - while (isa_dn && isa_dn != dn) - isa_dn = isa_dn->parent; - - if (isa_dn_orig) - of_node_put(isa_dn_orig); - - /* Count number of direct PCI children of the PHB. - * All PCI device nodes have class-code property, so it's - * an easy way to find them. - */ - for (children = 0, tmp = dn->child; tmp; tmp = tmp->sibling) - if (get_property(tmp, "class-code", NULL)) - children++; - - DBG("Children: %d\n", children); - - /* Calculate amount of DMA window per slot. Each window must be - * a power of two (due to pci_alloc_consistent requirements). - * - * Keep 256MB aside for PHBs with ISA. - */ - - if (!isa_dn) { - /* No ISA/IDE - just set window size and return */ - pci->phb->dma_window_size = 0x80000000ul; /* To be divided */ - - while (pci->phb->dma_window_size * children > 0x80000000ul) - pci->phb->dma_window_size >>= 1; - DBG("No ISA/IDE, window size is 0x%lx\n", - pci->phb->dma_window_size); - pci->phb->dma_window_base_cur = 0; - - return; - } - - /* If we have ISA, then we probably have an IDE - * controller too. Allocate a 128MB table but - * skip the first 128MB to avoid stepping on ISA - * space. - */ - pci->phb->dma_window_size = 0x8000000ul; - pci->phb->dma_window_base_cur = 0x8000000ul; - - tbl = kmalloc(sizeof(struct iommu_table), GFP_KERNEL); - - iommu_table_setparms(pci->phb, dn, tbl); - pci->iommu_table = iommu_init_table(tbl); - - /* Divide the rest (1.75GB) among the children */ - pci->phb->dma_window_size = 0x80000000ul; - while (pci->phb->dma_window_size * children > 0x70000000ul) - pci->phb->dma_window_size >>= 1; - - DBG("ISA/IDE, window size is 0x%lx\n", pci->phb->dma_window_size); - -} - - -static void iommu_bus_setup_pSeriesLP(struct pci_bus *bus) -{ - struct iommu_table *tbl; - struct device_node *dn, *pdn; - struct pci_dn *ppci; - unsigned int *dma_window = NULL; - - DBG("iommu_bus_setup_pSeriesLP, bus %p, bus->self %p\n", bus, bus->self); - - dn = pci_bus_to_OF_node(bus); - - /* Find nearest ibm,dma-window, walking up the device tree */ - for (pdn = dn; pdn != NULL; pdn = pdn->parent) { - dma_window = (unsigned int *)get_property(pdn, "ibm,dma-window", NULL); - if (dma_window != NULL) - break; - } - - if (dma_window == NULL) { - DBG("iommu_bus_setup_pSeriesLP: bus %s seems to have no ibm,dma-window property\n", dn->full_name); - return; - } - - ppci = pdn->data; - if (!ppci->iommu_table) { - /* Bussubno hasn't been copied yet. - * Do it now because iommu_table_setparms_lpar needs it. - */ - - ppci->bussubno = bus->number; - - tbl = (struct iommu_table *)kmalloc(sizeof(struct iommu_table), - GFP_KERNEL); - - iommu_table_setparms_lpar(ppci->phb, pdn, tbl, dma_window); - - ppci->iommu_table = iommu_init_table(tbl); - } - - if (pdn != dn) - PCI_DN(dn)->iommu_table = ppci->iommu_table; -} - - -static void iommu_dev_setup_pSeries(struct pci_dev *dev) -{ - struct device_node *dn, *mydn; - struct iommu_table *tbl; - - DBG("iommu_dev_setup_pSeries, dev %p (%s)\n", dev, pci_name(dev)); - - mydn = dn = pci_device_to_OF_node(dev); - - /* If we're the direct child of a root bus, then we need to allocate - * an iommu table ourselves. The bus setup code should have setup - * the window sizes already. - */ - if (!dev->bus->self) { - DBG(" --> first child, no bridge. Allocating iommu table.\n"); - tbl = kmalloc(sizeof(struct iommu_table), GFP_KERNEL); - iommu_table_setparms(PCI_DN(dn)->phb, dn, tbl); - PCI_DN(mydn)->iommu_table = iommu_init_table(tbl); - - return; - } - - /* If this device is further down the bus tree, search upwards until - * an already allocated iommu table is found and use that. - */ - - while (dn && dn->data && PCI_DN(dn)->iommu_table == NULL) - dn = dn->parent; - - if (dn && dn->data) { - PCI_DN(mydn)->iommu_table = PCI_DN(dn)->iommu_table; - } else { - DBG("iommu_dev_setup_pSeries, dev %p (%s) has no iommu table\n", dev, pci_name(dev)); - } -} - -static int iommu_reconfig_notifier(struct notifier_block *nb, unsigned long action, void *node) -{ - int err = NOTIFY_OK; - struct device_node *np = node; - struct pci_dn *pci = np->data; - - switch (action) { - case PSERIES_RECONFIG_REMOVE: - if (pci->iommu_table && - get_property(np, "ibm,dma-window", NULL)) - iommu_free_table(np); - break; - default: - err = NOTIFY_DONE; - break; - } - return err; -} - -static struct notifier_block iommu_reconfig_nb = { - .notifier_call = iommu_reconfig_notifier, -}; - -static void iommu_dev_setup_pSeriesLP(struct pci_dev *dev) -{ - struct device_node *pdn, *dn; - struct iommu_table *tbl; - int *dma_window = NULL; - struct pci_dn *pci; - - DBG("iommu_dev_setup_pSeriesLP, dev %p (%s)\n", dev, pci_name(dev)); - - /* dev setup for LPAR is a little tricky, since the device tree might - * contain the dma-window properties per-device and not neccesarily - * for the bus. So we need to search upwards in the tree until we - * either hit a dma-window property, OR find a parent with a table - * already allocated. - */ - dn = pci_device_to_OF_node(dev); - - for (pdn = dn; pdn && pdn->data && !PCI_DN(pdn)->iommu_table; - pdn = pdn->parent) { - dma_window = (unsigned int *) - get_property(pdn, "ibm,dma-window", NULL); - if (dma_window) - break; - } - - /* Check for parent == NULL so we don't try to setup the empty EADS - * slots on POWER4 machines. - */ - if (dma_window == NULL || pdn->parent == NULL) { - DBG("No dma window for device, linking to parent\n"); - PCI_DN(dn)->iommu_table = PCI_DN(pdn)->iommu_table; - return; - } else { - DBG("Found DMA window, allocating table\n"); - } - - pci = pdn->data; - if (!pci->iommu_table) { - /* iommu_table_setparms_lpar needs bussubno. */ - pci->bussubno = pci->phb->bus->number; - - tbl = (struct iommu_table *)kmalloc(sizeof(struct iommu_table), - GFP_KERNEL); - - iommu_table_setparms_lpar(pci->phb, pdn, tbl, dma_window); - - pci->iommu_table = iommu_init_table(tbl); - } - - if (pdn != dn) - PCI_DN(dn)->iommu_table = pci->iommu_table; -} - -static void iommu_bus_setup_null(struct pci_bus *b) { } -static void iommu_dev_setup_null(struct pci_dev *d) { } - -/* These are called very early. */ -void iommu_init_early_pSeries(void) -{ - if (of_chosen && get_property(of_chosen, "linux,iommu-off", NULL)) { - /* Direct I/O, IOMMU off */ - ppc_md.iommu_dev_setup = iommu_dev_setup_null; - ppc_md.iommu_bus_setup = iommu_bus_setup_null; - pci_direct_iommu_init(); - - return; - } - - if (systemcfg->platform & PLATFORM_LPAR) { - if (firmware_has_feature(FW_FEATURE_MULTITCE)) { - ppc_md.tce_build = tce_buildmulti_pSeriesLP; - ppc_md.tce_free = tce_freemulti_pSeriesLP; - } else { - ppc_md.tce_build = tce_build_pSeriesLP; - ppc_md.tce_free = tce_free_pSeriesLP; - } - ppc_md.iommu_bus_setup = iommu_bus_setup_pSeriesLP; - ppc_md.iommu_dev_setup = iommu_dev_setup_pSeriesLP; - } else { - ppc_md.tce_build = tce_build_pSeries; - ppc_md.tce_free = tce_free_pSeries; - ppc_md.iommu_bus_setup = iommu_bus_setup_pSeries; - ppc_md.iommu_dev_setup = iommu_dev_setup_pSeries; - } - - - pSeries_reconfig_notifier_register(&iommu_reconfig_nb); - - pci_iommu_init(); -} - diff --git a/arch/ppc64/kernel/pSeries_lpar.c b/arch/ppc64/kernel/pSeries_lpar.c deleted file mode 100644 index a6de83f2078f..000000000000 --- a/arch/ppc64/kernel/pSeries_lpar.c +++ /dev/null @@ -1,518 +0,0 @@ -/* - * pSeries_lpar.c - * Copyright (C) 2001 Todd Inglett, IBM Corporation - * - * pSeries LPAR support. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - */ - -#define DEBUG - -#include <linux/config.h> -#include <linux/kernel.h> -#include <linux/dma-mapping.h> -#include <asm/processor.h> -#include <asm/mmu.h> -#include <asm/page.h> -#include <asm/pgtable.h> -#include <asm/machdep.h> -#include <asm/abs_addr.h> -#include <asm/mmu_context.h> -#include <asm/ppcdebug.h> -#include <asm/iommu.h> -#include <asm/tlbflush.h> -#include <asm/tlb.h> -#include <asm/prom.h> -#include <asm/abs_addr.h> -#include <asm/cputable.h> -#include <asm/plpar_wrappers.h> - -#ifdef DEBUG -#define DBG(fmt...) udbg_printf(fmt) -#else -#define DBG(fmt...) -#endif - -/* in pSeries_hvCall.S */ -EXPORT_SYMBOL(plpar_hcall); -EXPORT_SYMBOL(plpar_hcall_4out); -EXPORT_SYMBOL(plpar_hcall_norets); -EXPORT_SYMBOL(plpar_hcall_8arg_2ret); - -extern void pSeries_find_serial_port(void); - - -int vtermno; /* virtual terminal# for udbg */ - -#define __ALIGNED__ __attribute__((__aligned__(sizeof(long)))) -static void udbg_hvsi_putc(unsigned char c) -{ - /* packet's seqno isn't used anyways */ - uint8_t packet[] __ALIGNED__ = { 0xff, 5, 0, 0, c }; - int rc; - - if (c == '\n') - udbg_hvsi_putc('\r'); - - do { - rc = plpar_put_term_char(vtermno, sizeof(packet), packet); - } while (rc == H_Busy); -} - -static long hvsi_udbg_buf_len; -static uint8_t hvsi_udbg_buf[256]; - -static int udbg_hvsi_getc_poll(void) -{ - unsigned char ch; - int rc, i; - - if (hvsi_udbg_buf_len == 0) { - rc = plpar_get_term_char(vtermno, &hvsi_udbg_buf_len, hvsi_udbg_buf); - if (rc != H_Success || hvsi_udbg_buf[0] != 0xff) { - /* bad read or non-data packet */ - hvsi_udbg_buf_len = 0; - } else { - /* remove the packet header */ - for (i = 4; i < hvsi_udbg_buf_len; i++) - hvsi_udbg_buf[i-4] = hvsi_udbg_buf[i]; - hvsi_udbg_buf_len -= 4; - } - } - - if (hvsi_udbg_buf_len <= 0 || hvsi_udbg_buf_len > 256) { - /* no data ready */ - hvsi_udbg_buf_len = 0; - return -1; - } - - ch = hvsi_udbg_buf[0]; - /* shift remaining data down */ - for (i = 1; i < hvsi_udbg_buf_len; i++) { - hvsi_udbg_buf[i-1] = hvsi_udbg_buf[i]; - } - hvsi_udbg_buf_len--; - - return ch; -} - -static unsigned char udbg_hvsi_getc(void) -{ - int ch; - for (;;) { - ch = udbg_hvsi_getc_poll(); - if (ch == -1) { - /* This shouldn't be needed...but... */ - volatile unsigned long delay; - for (delay=0; delay < 2000000; delay++) - ; - } else { - return ch; - } - } -} - -static void udbg_putcLP(unsigned char c) -{ - char buf[16]; - unsigned long rc; - - if (c == '\n') - udbg_putcLP('\r'); - - buf[0] = c; - do { - rc = plpar_put_term_char(vtermno, 1, buf); - } while(rc == H_Busy); -} - -/* Buffered chars getc */ -static long inbuflen; -static long inbuf[2]; /* must be 2 longs */ - -static int udbg_getc_pollLP(void) -{ - /* The interface is tricky because it may return up to 16 chars. - * We save them statically for future calls to udbg_getc(). - */ - char ch, *buf = (char *)inbuf; - int i; - long rc; - if (inbuflen == 0) { - /* get some more chars. */ - inbuflen = 0; - rc = plpar_get_term_char(vtermno, &inbuflen, buf); - if (rc != H_Success) - inbuflen = 0; /* otherwise inbuflen is garbage */ - } - if (inbuflen <= 0 || inbuflen > 16) { - /* Catch error case as well as other oddities (corruption) */ - inbuflen = 0; - return -1; - } - ch = buf[0]; - for (i = 1; i < inbuflen; i++) /* shuffle them down. */ - buf[i-1] = buf[i]; - inbuflen--; - return ch; -} - -static unsigned char udbg_getcLP(void) -{ - int ch; - for (;;) { - ch = udbg_getc_pollLP(); - if (ch == -1) { - /* This shouldn't be needed...but... */ - volatile unsigned long delay; - for (delay=0; delay < 2000000; delay++) - ; - } else { - return ch; - } - } -} - -/* call this from early_init() for a working debug console on - * vterm capable LPAR machines - */ -void udbg_init_debug_lpar(void) -{ - vtermno = 0; - udbg_putc = udbg_putcLP; - udbg_getc = udbg_getcLP; - udbg_getc_poll = udbg_getc_pollLP; -} - -/* returns 0 if couldn't find or use /chosen/stdout as console */ -int find_udbg_vterm(void) -{ - struct device_node *stdout_node; - u32 *termno; - char *name; - int found = 0; - - /* find the boot console from /chosen/stdout */ - if (!of_chosen) - return 0; - name = (char *)get_property(of_chosen, "linux,stdout-path", NULL); - if (name == NULL) - return 0; - stdout_node = of_find_node_by_path(name); - if (!stdout_node) - return 0; - - /* now we have the stdout node; figure out what type of device it is. */ - name = (char *)get_property(stdout_node, "name", NULL); - if (!name) { - printk(KERN_WARNING "stdout node missing 'name' property!\n"); - goto out; - } - - if (strncmp(name, "vty", 3) == 0) { - if (device_is_compatible(stdout_node, "hvterm1")) { - termno = (u32 *)get_property(stdout_node, "reg", NULL); - if (termno) { - vtermno = termno[0]; - udbg_putc = udbg_putcLP; - udbg_getc = udbg_getcLP; - udbg_getc_poll = udbg_getc_pollLP; - found = 1; - } - } else if (device_is_compatible(stdout_node, "hvterm-protocol")) { - termno = (u32 *)get_property(stdout_node, "reg", NULL); - if (termno) { - vtermno = termno[0]; - udbg_putc = udbg_hvsi_putc; - udbg_getc = udbg_hvsi_getc; - udbg_getc_poll = udbg_hvsi_getc_poll; - found = 1; - } - } - } else if (strncmp(name, "serial", 6)) { - /* XXX fix ISA serial console */ - printk(KERN_WARNING "serial stdout on LPAR ('%s')! " - "can't print udbg messages\n", - stdout_node->full_name); - } else { - printk(KERN_WARNING "don't know how to print to stdout '%s'\n", - stdout_node->full_name); - } - -out: - of_node_put(stdout_node); - return found; -} - -void vpa_init(int cpu) -{ - int hwcpu = get_hard_smp_processor_id(cpu); - unsigned long vpa = (unsigned long)&(paca[cpu].lppaca); - long ret; - unsigned long flags; - - /* Register the Virtual Processor Area (VPA) */ - flags = 1UL << (63 - 18); - - if (cpu_has_feature(CPU_FTR_ALTIVEC)) - paca[cpu].lppaca.vmxregs_in_use = 1; - - ret = register_vpa(flags, hwcpu, __pa(vpa)); - - if (ret) - printk(KERN_ERR "WARNING: vpa_init: VPA registration for " - "cpu %d (hw %d) of area %lx returns %ld\n", - cpu, hwcpu, __pa(vpa), ret); -} - -long pSeries_lpar_hpte_insert(unsigned long hpte_group, - unsigned long va, unsigned long prpn, - unsigned long vflags, unsigned long rflags) -{ - unsigned long lpar_rc; - unsigned long flags; - unsigned long slot; - unsigned long hpte_v, hpte_r; - unsigned long dummy0, dummy1; - - hpte_v = ((va >> 23) << HPTE_V_AVPN_SHIFT) | vflags | HPTE_V_VALID; - if (vflags & HPTE_V_LARGE) - hpte_v &= ~(1UL << HPTE_V_AVPN_SHIFT); - - hpte_r = (prpn << HPTE_R_RPN_SHIFT) | rflags; - - /* Now fill in the actual HPTE */ - /* Set CEC cookie to 0 */ - /* Zero page = 0 */ - /* I-cache Invalidate = 0 */ - /* I-cache synchronize = 0 */ - /* Exact = 0 */ - flags = 0; - - /* XXX why is this here? - Anton */ - if (rflags & (_PAGE_GUARDED|_PAGE_NO_CACHE)) - hpte_r &= ~_PAGE_COHERENT; - - lpar_rc = plpar_hcall(H_ENTER, flags, hpte_group, hpte_v, - hpte_r, &slot, &dummy0, &dummy1); - - if (unlikely(lpar_rc == H_PTEG_Full)) - return -1; - - /* - * Since we try and ioremap PHBs we don't own, the pte insert - * will fail. However we must catch the failure in hash_page - * or we will loop forever, so return -2 in this case. - */ - if (unlikely(lpar_rc != H_Success)) - return -2; - - /* Because of iSeries, we have to pass down the secondary - * bucket bit here as well - */ - return (slot & 7) | (!!(vflags & HPTE_V_SECONDARY) << 3); -} - -static DEFINE_SPINLOCK(pSeries_lpar_tlbie_lock); - -static long pSeries_lpar_hpte_remove(unsigned long hpte_group) -{ - unsigned long slot_offset; - unsigned long lpar_rc; - int i; - unsigned long dummy1, dummy2; - - /* pick a random slot to start at */ - slot_offset = mftb() & 0x7; - - for (i = 0; i < HPTES_PER_GROUP; i++) { - - /* don't remove a bolted entry */ - lpar_rc = plpar_pte_remove(H_ANDCOND, hpte_group + slot_offset, - (0x1UL << 4), &dummy1, &dummy2); - - if (lpar_rc == H_Success) - return i; - - BUG_ON(lpar_rc != H_Not_Found); - - slot_offset++; - slot_offset &= 0x7; - } - - return -1; -} - -static void pSeries_lpar_hptab_clear(void) -{ - unsigned long size_bytes = 1UL << ppc64_pft_size; - unsigned long hpte_count = size_bytes >> 4; - unsigned long dummy1, dummy2; - int i; - - /* TODO: Use bulk call */ - for (i = 0; i < hpte_count; i++) - plpar_pte_remove(0, i, 0, &dummy1, &dummy2); -} - -/* - * NOTE: for updatepp ops we are fortunate that the linux "newpp" bits and - * the low 3 bits of flags happen to line up. So no transform is needed. - * We can probably optimize here and assume the high bits of newpp are - * already zero. For now I am paranoid. - */ -static long pSeries_lpar_hpte_updatepp(unsigned long slot, unsigned long newpp, - unsigned long va, int large, int local) -{ - unsigned long lpar_rc; - unsigned long flags = (newpp & 7) | H_AVPN; - unsigned long avpn = va >> 23; - - if (large) - avpn &= ~0x1UL; - - lpar_rc = plpar_pte_protect(flags, slot, (avpn << 7)); - - if (lpar_rc == H_Not_Found) - return -1; - - BUG_ON(lpar_rc != H_Success); - - return 0; -} - -static unsigned long pSeries_lpar_hpte_getword0(unsigned long slot) -{ - unsigned long dword0; - unsigned long lpar_rc; - unsigned long dummy_word1; - unsigned long flags; - - /* Read 1 pte at a time */ - /* Do not need RPN to logical page translation */ - /* No cross CEC PFT access */ - flags = 0; - - lpar_rc = plpar_pte_read(flags, slot, &dword0, &dummy_word1); - - BUG_ON(lpar_rc != H_Success); - - return dword0; -} - -static long pSeries_lpar_hpte_find(unsigned long vpn) -{ - unsigned long hash; - unsigned long i, j; - long slot; - unsigned long hpte_v; - - hash = hpt_hash(vpn, 0); - - for (j = 0; j < 2; j++) { - slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; - for (i = 0; i < HPTES_PER_GROUP; i++) { - hpte_v = pSeries_lpar_hpte_getword0(slot); - - if ((HPTE_V_AVPN_VAL(hpte_v) == (vpn >> 11)) - && (hpte_v & HPTE_V_VALID) - && (!!(hpte_v & HPTE_V_SECONDARY) == j)) { - /* HPTE matches */ - if (j) - slot = -slot; - return slot; - } - ++slot; - } - hash = ~hash; - } - - return -1; -} - -static void pSeries_lpar_hpte_updateboltedpp(unsigned long newpp, - unsigned long ea) -{ - unsigned long lpar_rc; - unsigned long vsid, va, vpn, flags; - long slot; - - vsid = get_kernel_vsid(ea); - va = (vsid << 28) | (ea & 0x0fffffff); - vpn = va >> PAGE_SHIFT; - - slot = pSeries_lpar_hpte_find(vpn); - BUG_ON(slot == -1); - - flags = newpp & 7; - lpar_rc = plpar_pte_protect(flags, slot, 0); - - BUG_ON(lpar_rc != H_Success); -} - -static void pSeries_lpar_hpte_invalidate(unsigned long slot, unsigned long va, - int large, int local) -{ - unsigned long avpn = va >> 23; - unsigned long lpar_rc; - unsigned long dummy1, dummy2; - - if (large) - avpn &= ~0x1UL; - - lpar_rc = plpar_pte_remove(H_AVPN, slot, (avpn << 7), &dummy1, - &dummy2); - - if (lpar_rc == H_Not_Found) - return; - - BUG_ON(lpar_rc != H_Success); -} - -/* - * Take a spinlock around flushes to avoid bouncing the hypervisor tlbie - * lock. - */ -void pSeries_lpar_flush_hash_range(unsigned long context, unsigned long number, - int local) -{ - int i; - unsigned long flags = 0; - struct ppc64_tlb_batch *batch = &__get_cpu_var(ppc64_tlb_batch); - int lock_tlbie = !cpu_has_feature(CPU_FTR_LOCKLESS_TLBIE); - - if (lock_tlbie) - spin_lock_irqsave(&pSeries_lpar_tlbie_lock, flags); - - for (i = 0; i < number; i++) - flush_hash_page(context, batch->addr[i], batch->pte[i], local); - - if (lock_tlbie) - spin_unlock_irqrestore(&pSeries_lpar_tlbie_lock, flags); -} - -void hpte_init_lpar(void) -{ - ppc_md.hpte_invalidate = pSeries_lpar_hpte_invalidate; - ppc_md.hpte_updatepp = pSeries_lpar_hpte_updatepp; - ppc_md.hpte_updateboltedpp = pSeries_lpar_hpte_updateboltedpp; - ppc_md.hpte_insert = pSeries_lpar_hpte_insert; - ppc_md.hpte_remove = pSeries_lpar_hpte_remove; - ppc_md.flush_hash_range = pSeries_lpar_flush_hash_range; - ppc_md.hpte_clear_all = pSeries_lpar_hptab_clear; - - htab_finish_init(); -} diff --git a/arch/ppc64/kernel/pSeries_nvram.c b/arch/ppc64/kernel/pSeries_nvram.c deleted file mode 100644 index 18abfb1f4e24..000000000000 --- a/arch/ppc64/kernel/pSeries_nvram.c +++ /dev/null @@ -1,148 +0,0 @@ -/* - * c 2001 PPC 64 Team, IBM Corp - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * /dev/nvram driver for PPC64 - * - * This perhaps should live in drivers/char - */ - - -#include <linux/types.h> -#include <linux/errno.h> -#include <linux/init.h> -#include <linux/slab.h> -#include <linux/spinlock.h> -#include <asm/uaccess.h> -#include <asm/nvram.h> -#include <asm/rtas.h> -#include <asm/prom.h> -#include <asm/machdep.h> - -static unsigned int nvram_size; -static int nvram_fetch, nvram_store; -static char nvram_buf[NVRW_CNT]; /* assume this is in the first 4GB */ -static DEFINE_SPINLOCK(nvram_lock); - - -static ssize_t pSeries_nvram_read(char *buf, size_t count, loff_t *index) -{ - unsigned int i; - unsigned long len; - int done; - unsigned long flags; - char *p = buf; - - - if (nvram_size == 0 || nvram_fetch == RTAS_UNKNOWN_SERVICE) - return -ENODEV; - - if (*index >= nvram_size) - return 0; - - i = *index; - if (i + count > nvram_size) - count = nvram_size - i; - - spin_lock_irqsave(&nvram_lock, flags); - - for (; count != 0; count -= len) { - len = count; - if (len > NVRW_CNT) - len = NVRW_CNT; - - if ((rtas_call(nvram_fetch, 3, 2, &done, i, __pa(nvram_buf), - len) != 0) || len != done) { - spin_unlock_irqrestore(&nvram_lock, flags); - return -EIO; - } - - memcpy(p, nvram_buf, len); - - p += len; - i += len; - } - - spin_unlock_irqrestore(&nvram_lock, flags); - - *index = i; - return p - buf; -} - -static ssize_t pSeries_nvram_write(char *buf, size_t count, loff_t *index) -{ - unsigned int i; - unsigned long len; - int done; - unsigned long flags; - const char *p = buf; - - if (nvram_size == 0 || nvram_store == RTAS_UNKNOWN_SERVICE) - return -ENODEV; - - if (*index >= nvram_size) - return 0; - - i = *index; - if (i + count > nvram_size) - count = nvram_size - i; - - spin_lock_irqsave(&nvram_lock, flags); - - for (; count != 0; count -= len) { - len = count; - if (len > NVRW_CNT) - len = NVRW_CNT; - - memcpy(nvram_buf, p, len); - - if ((rtas_call(nvram_store, 3, 2, &done, i, __pa(nvram_buf), - len) != 0) || len != done) { - spin_unlock_irqrestore(&nvram_lock, flags); - return -EIO; - } - - p += len; - i += len; - } - spin_unlock_irqrestore(&nvram_lock, flags); - - *index = i; - return p - buf; -} - -static ssize_t pSeries_nvram_get_size(void) -{ - return nvram_size ? nvram_size : -ENODEV; -} - -int __init pSeries_nvram_init(void) -{ - struct device_node *nvram; - unsigned int *nbytes_p, proplen; - - nvram = of_find_node_by_type(NULL, "nvram"); - if (nvram == NULL) - return -ENODEV; - - nbytes_p = (unsigned int *)get_property(nvram, "#bytes", &proplen); - if (nbytes_p == NULL || proplen != sizeof(unsigned int)) - return -EIO; - - nvram_size = *nbytes_p; - - nvram_fetch = rtas_token("nvram-fetch"); - nvram_store = rtas_token("nvram-store"); - printk(KERN_INFO "PPC64 nvram contains %d bytes\n", nvram_size); - of_node_put(nvram); - - ppc_md.nvram_read = pSeries_nvram_read; - ppc_md.nvram_write = pSeries_nvram_write; - ppc_md.nvram_size = pSeries_nvram_get_size; - - return 0; -} diff --git a/arch/ppc64/kernel/pSeries_pci.c b/arch/ppc64/kernel/pSeries_pci.c deleted file mode 100644 index 928f8febdb3b..000000000000 --- a/arch/ppc64/kernel/pSeries_pci.c +++ /dev/null @@ -1,143 +0,0 @@ -/* - * arch/ppc64/kernel/pSeries_pci.c - * - * Copyright (C) 2001 Dave Engebretsen, IBM Corporation - * Copyright (C) 2003 Anton Blanchard <anton@au.ibm.com>, IBM - * - * pSeries specific routines for PCI. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - */ - -#include <linux/init.h> -#include <linux/ioport.h> -#include <linux/kernel.h> -#include <linux/pci.h> -#include <linux/string.h> - -#include <asm/pci-bridge.h> -#include <asm/prom.h> - -#include "pci.h" - -static int __devinitdata s7a_workaround = -1; - -#if 0 -void pcibios_name_device(struct pci_dev *dev) -{ - struct device_node *dn; - - /* - * Add IBM loc code (slot) as a prefix to the device names for service - */ - dn = pci_device_to_OF_node(dev); - if (dn) { - char *loc_code = get_property(dn, "ibm,loc-code", 0); - if (loc_code) { - int loc_len = strlen(loc_code); - if (loc_len < sizeof(dev->dev.name)) { - memmove(dev->dev.name+loc_len+1, dev->dev.name, - sizeof(dev->dev.name)-loc_len-1); - memcpy(dev->dev.name, loc_code, loc_len); - dev->dev.name[loc_len] = ' '; - dev->dev.name[sizeof(dev->dev.name)-1] = '\0'; - } - } - } -} -DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pcibios_name_device); -#endif - -static void __devinit check_s7a(void) -{ - struct device_node *root; - char *model; - - s7a_workaround = 0; - root = of_find_node_by_path("/"); - if (root) { - model = get_property(root, "model", NULL); - if (model && !strcmp(model, "IBM,7013-S7A")) - s7a_workaround = 1; - of_node_put(root); - } -} - -void __devinit pSeries_irq_bus_setup(struct pci_bus *bus) -{ - struct pci_dev *dev; - - if (s7a_workaround < 0) - check_s7a(); - list_for_each_entry(dev, &bus->devices, bus_list) { - pci_read_irq_line(dev); - if (s7a_workaround) { - if (dev->irq > 16) { - dev->irq -= 3; - pci_write_config_byte(dev, PCI_INTERRUPT_LINE, - dev->irq); - } - } - } -} - -static void __init pSeries_request_regions(void) -{ - if (!isa_io_base) - return; - - request_region(0x20,0x20,"pic1"); - request_region(0xa0,0x20,"pic2"); - request_region(0x00,0x20,"dma1"); - request_region(0x40,0x20,"timer"); - request_region(0x80,0x10,"dma page reg"); - request_region(0xc0,0x20,"dma2"); -} - -void __init pSeries_final_fixup(void) -{ - phbs_remap_io(); - pSeries_request_regions(); - - pci_addr_cache_build(); -} - -/* - * Assume the winbond 82c105 is the IDE controller on a - * p610. We should probably be more careful in case - * someone tries to plug in a similar adapter. - */ -static void fixup_winbond_82c105(struct pci_dev* dev) -{ - int i; - unsigned int reg; - - if (!(systemcfg->platform & PLATFORM_PSERIES)) - return; - - printk("Using INTC for W82c105 IDE controller.\n"); - pci_read_config_dword(dev, 0x40, ®); - /* Enable LEGIRQ to use INTC instead of ISA interrupts */ - pci_write_config_dword(dev, 0x40, reg | (1<<11)); - - for (i = 0; i < DEVICE_COUNT_RESOURCE; ++i) { - /* zap the 2nd function of the winbond chip */ - if (dev->resource[i].flags & IORESOURCE_IO - && dev->bus->number == 0 && dev->devfn == 0x81) - dev->resource[i].flags &= ~IORESOURCE_IO; - } -} -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_WINBOND, PCI_DEVICE_ID_WINBOND_82C105, - fixup_winbond_82c105); diff --git a/arch/ppc64/kernel/pSeries_reconfig.c b/arch/ppc64/kernel/pSeries_reconfig.c deleted file mode 100644 index 58c61219d08e..000000000000 --- a/arch/ppc64/kernel/pSeries_reconfig.c +++ /dev/null @@ -1,426 +0,0 @@ -/* - * pSeries_reconfig.c - support for dynamic reconfiguration (including PCI - * Hotplug and Dynamic Logical Partitioning on RPA platforms). - * - * Copyright (C) 2005 Nathan Lynch - * Copyright (C) 2005 IBM Corporation - * - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License version - * 2 as published by the Free Software Foundation. - */ - -#include <linux/kernel.h> -#include <linux/kref.h> -#include <linux/notifier.h> -#include <linux/proc_fs.h> - -#include <asm/prom.h> -#include <asm/pSeries_reconfig.h> -#include <asm/uaccess.h> - - - -/* - * Routines for "runtime" addition and removal of device tree nodes. - */ -#ifdef CONFIG_PROC_DEVICETREE -/* - * Add a node to /proc/device-tree. - */ -static void add_node_proc_entries(struct device_node *np) -{ - struct proc_dir_entry *ent; - - ent = proc_mkdir(strrchr(np->full_name, '/') + 1, np->parent->pde); - if (ent) - proc_device_tree_add_node(np, ent); -} - -static void remove_node_proc_entries(struct device_node *np) -{ - struct property *pp = np->properties; - struct device_node *parent = np->parent; - - while (pp) { - remove_proc_entry(pp->name, np->pde); - pp = pp->next; - } - if (np->pde) - remove_proc_entry(np->pde->name, parent->pde); -} -#else /* !CONFIG_PROC_DEVICETREE */ -static void add_node_proc_entries(struct device_node *np) -{ - return; -} - -static void remove_node_proc_entries(struct device_node *np) -{ - return; -} -#endif /* CONFIG_PROC_DEVICETREE */ - -/** - * derive_parent - basically like dirname(1) - * @path: the full_name of a node to be added to the tree - * - * Returns the node which should be the parent of the node - * described by path. E.g., for path = "/foo/bar", returns - * the node with full_name = "/foo". - */ -static struct device_node *derive_parent(const char *path) -{ - struct device_node *parent = NULL; - char *parent_path = "/"; - size_t parent_path_len = strrchr(path, '/') - path + 1; - - /* reject if path is "/" */ - if (!strcmp(path, "/")) - return ERR_PTR(-EINVAL); - - if (strrchr(path, '/') != path) { - parent_path = kmalloc(parent_path_len, GFP_KERNEL); - if (!parent_path) - return ERR_PTR(-ENOMEM); - strlcpy(parent_path, path, parent_path_len); - } - parent = of_find_node_by_path(parent_path); - if (!parent) - return ERR_PTR(-EINVAL); - if (strcmp(parent_path, "/")) - kfree(parent_path); - return parent; -} - -static struct notifier_block *pSeries_reconfig_chain; - -int pSeries_reconfig_notifier_register(struct notifier_block *nb) -{ - return notifier_chain_register(&pSeries_reconfig_chain, nb); -} - -void pSeries_reconfig_notifier_unregister(struct notifier_block *nb) -{ - notifier_chain_unregister(&pSeries_reconfig_chain, nb); -} - -static int pSeries_reconfig_add_node(const char *path, struct property *proplist) -{ - struct device_node *np; - int err = -ENOMEM; - - np = kzalloc(sizeof(*np), GFP_KERNEL); - if (!np) - goto out_err; - - np->full_name = kmalloc(strlen(path) + 1, GFP_KERNEL); - if (!np->full_name) - goto out_err; - - strcpy(np->full_name, path); - - np->properties = proplist; - OF_MARK_DYNAMIC(np); - kref_init(&np->kref); - - np->parent = derive_parent(path); - if (IS_ERR(np->parent)) { - err = PTR_ERR(np->parent); - goto out_err; - } - - err = notifier_call_chain(&pSeries_reconfig_chain, - PSERIES_RECONFIG_ADD, np); - if (err == NOTIFY_BAD) { - printk(KERN_ERR "Failed to add device node %s\n", path); - err = -ENOMEM; /* For now, safe to assume kmalloc failure */ - goto out_err; - } - - of_attach_node(np); - - add_node_proc_entries(np); - - of_node_put(np->parent); - - return 0; - -out_err: - if (np) { - of_node_put(np->parent); - kfree(np->full_name); - kfree(np); - } - return err; -} - -static int pSeries_reconfig_remove_node(struct device_node *np) -{ - struct device_node *parent, *child; - - parent = of_get_parent(np); - if (!parent) - return -EINVAL; - - if ((child = of_get_next_child(np, NULL))) { - of_node_put(child); - return -EBUSY; - } - - remove_node_proc_entries(np); - - notifier_call_chain(&pSeries_reconfig_chain, - PSERIES_RECONFIG_REMOVE, np); - of_detach_node(np); - - of_node_put(parent); - of_node_put(np); /* Must decrement the refcount */ - return 0; -} - -/* - * /proc/ppc64/ofdt - yucky binary interface for adding and removing - * OF device nodes. Should be deprecated as soon as we get an - * in-kernel wrapper for the RTAS ibm,configure-connector call. - */ - -static void release_prop_list(const struct property *prop) -{ - struct property *next; - for (; prop; prop = next) { - next = prop->next; - kfree(prop->name); - kfree(prop->value); - kfree(prop); - } - -} - -/** - * parse_next_property - process the next property from raw input buffer - * @buf: input buffer, must be nul-terminated - * @end: end of the input buffer + 1, for validation - * @name: return value; set to property name in buf - * @length: return value; set to length of value - * @value: return value; set to the property value in buf - * - * Note that the caller must make copies of the name and value returned, - * this function does no allocation or copying of the data. Return value - * is set to the next name in buf, or NULL on error. - */ -static char * parse_next_property(char *buf, char *end, char **name, int *length, - unsigned char **value) -{ - char *tmp; - - *name = buf; - - tmp = strchr(buf, ' '); - if (!tmp) { - printk(KERN_ERR "property parse failed in %s at line %d\n", - __FUNCTION__, __LINE__); - return NULL; - } - *tmp = '\0'; - - if (++tmp >= end) { - printk(KERN_ERR "property parse failed in %s at line %d\n", - __FUNCTION__, __LINE__); - return NULL; - } - - /* now we're on the length */ - *length = -1; - *length = simple_strtoul(tmp, &tmp, 10); - if (*length == -1) { - printk(KERN_ERR "property parse failed in %s at line %d\n", - __FUNCTION__, __LINE__); - return NULL; - } - if (*tmp != ' ' || ++tmp >= end) { - printk(KERN_ERR "property parse failed in %s at line %d\n", - __FUNCTION__, __LINE__); - return NULL; - } - - /* now we're on the value */ - *value = tmp; - tmp += *length; - if (tmp > end) { - printk(KERN_ERR "property parse failed in %s at line %d\n", - __FUNCTION__, __LINE__); - return NULL; - } - else if (tmp < end && *tmp != ' ' && *tmp != '\0') { - printk(KERN_ERR "property parse failed in %s at line %d\n", - __FUNCTION__, __LINE__); - return NULL; - } - tmp++; - - /* and now we should be on the next name, or the end */ - return tmp; -} - -static struct property *new_property(const char *name, const int length, - const unsigned char *value, struct property *last) -{ - struct property *new = kmalloc(sizeof(*new), GFP_KERNEL); - - if (!new) - return NULL; - memset(new, 0, sizeof(*new)); - - if (!(new->name = kmalloc(strlen(name) + 1, GFP_KERNEL))) - goto cleanup; - if (!(new->value = kmalloc(length + 1, GFP_KERNEL))) - goto cleanup; - - strcpy(new->name, name); - memcpy(new->value, value, length); - *(((char *)new->value) + length) = 0; - new->length = length; - new->next = last; - return new; - -cleanup: - if (new->name) - kfree(new->name); - if (new->value) - kfree(new->value); - kfree(new); - return NULL; -} - -static int do_add_node(char *buf, size_t bufsize) -{ - char *path, *end, *name; - struct device_node *np; - struct property *prop = NULL; - unsigned char* value; - int length, rv = 0; - - end = buf + bufsize; - path = buf; - buf = strchr(buf, ' '); - if (!buf) - return -EINVAL; - *buf = '\0'; - buf++; - - if ((np = of_find_node_by_path(path))) { - of_node_put(np); - return -EINVAL; - } - - /* rv = build_prop_list(tmp, bufsize - (tmp - buf), &proplist); */ - while (buf < end && - (buf = parse_next_property(buf, end, &name, &length, &value))) { - struct property *last = prop; - - prop = new_property(name, length, value, last); - if (!prop) { - rv = -ENOMEM; - prop = last; - goto out; - } - } - if (!buf) { - rv = -EINVAL; - goto out; - } - - rv = pSeries_reconfig_add_node(path, prop); - -out: - if (rv) - release_prop_list(prop); - return rv; -} - -static int do_remove_node(char *buf) -{ - struct device_node *node; - int rv = -ENODEV; - - if ((node = of_find_node_by_path(buf))) - rv = pSeries_reconfig_remove_node(node); - - of_node_put(node); - return rv; -} - -/** - * ofdt_write - perform operations on the Open Firmware device tree - * - * @file: not used - * @buf: command and arguments - * @count: size of the command buffer - * @off: not used - * - * Operations supported at this time are addition and removal of - * whole nodes along with their properties. Operations on individual - * properties are not implemented (yet). - */ -static ssize_t ofdt_write(struct file *file, const char __user *buf, size_t count, - loff_t *off) -{ - int rv = 0; - char *kbuf; - char *tmp; - - if (!(kbuf = kmalloc(count + 1, GFP_KERNEL))) { - rv = -ENOMEM; - goto out; - } - if (copy_from_user(kbuf, buf, count)) { - rv = -EFAULT; - goto out; - } - - kbuf[count] = '\0'; - - tmp = strchr(kbuf, ' '); - if (!tmp) { - rv = -EINVAL; - goto out; - } - *tmp = '\0'; - tmp++; - - if (!strcmp(kbuf, "add_node")) - rv = do_add_node(tmp, count - (tmp - kbuf)); - else if (!strcmp(kbuf, "remove_node")) - rv = do_remove_node(tmp); - else - rv = -EINVAL; -out: - kfree(kbuf); - return rv ? rv : count; -} - -static struct file_operations ofdt_fops = { - .write = ofdt_write -}; - -/* create /proc/ppc64/ofdt write-only by root */ -static int proc_ppc64_create_ofdt(void) -{ - struct proc_dir_entry *ent; - - if (!(systemcfg->platform & PLATFORM_PSERIES)) - return 0; - - ent = create_proc_entry("ppc64/ofdt", S_IWUSR, NULL); - if (ent) { - ent->nlink = 1; - ent->data = NULL; - ent->size = 0; - ent->proc_fops = &ofdt_fops; - } - - return 0; -} -__initcall(proc_ppc64_create_ofdt); diff --git a/arch/ppc64/kernel/pSeries_setup.c b/arch/ppc64/kernel/pSeries_setup.c deleted file mode 100644 index 3009701eb90d..000000000000 --- a/arch/ppc64/kernel/pSeries_setup.c +++ /dev/null @@ -1,622 +0,0 @@ -/* - * linux/arch/ppc/kernel/setup.c - * - * Copyright (C) 1995 Linus Torvalds - * Adapted from 'alpha' version by Gary Thomas - * Modified by Cort Dougan (cort@cs.nmt.edu) - * Modified by PPC64 Team, IBM Corp - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -/* - * bootup setup stuff.. - */ - -#undef DEBUG - -#include <linux/config.h> -#include <linux/cpu.h> -#include <linux/errno.h> -#include <linux/sched.h> -#include <linux/kernel.h> -#include <linux/mm.h> -#include <linux/stddef.h> -#include <linux/unistd.h> -#include <linux/slab.h> -#include <linux/user.h> -#include <linux/a.out.h> -#include <linux/tty.h> -#include <linux/major.h> -#include <linux/interrupt.h> -#include <linux/reboot.h> -#include <linux/init.h> -#include <linux/ioport.h> -#include <linux/console.h> -#include <linux/pci.h> -#include <linux/utsname.h> -#include <linux/adb.h> -#include <linux/module.h> -#include <linux/delay.h> -#include <linux/irq.h> -#include <linux/seq_file.h> -#include <linux/root_dev.h> - -#include <asm/mmu.h> -#include <asm/processor.h> -#include <asm/io.h> -#include <asm/pgtable.h> -#include <asm/prom.h> -#include <asm/rtas.h> -#include <asm/pci-bridge.h> -#include <asm/iommu.h> -#include <asm/dma.h> -#include <asm/machdep.h> -#include <asm/irq.h> -#include <asm/time.h> -#include <asm/nvram.h> -#include <asm/plpar_wrappers.h> -#include <asm/xics.h> -#include <asm/firmware.h> -#include <asm/pmc.h> - -#include "i8259.h" -#include "mpic.h" -#include "pci.h" - -#ifdef DEBUG -#define DBG(fmt...) udbg_printf(fmt) -#else -#define DBG(fmt...) -#endif - -extern void find_udbg_vterm(void); -extern void system_reset_fwnmi(void); /* from head.S */ -extern void machine_check_fwnmi(void); /* from head.S */ -extern void generic_find_legacy_serial_ports(u64 *physport, - unsigned int *default_speed); - -int fwnmi_active; /* TRUE if an FWNMI handler is present */ - -extern void pSeries_system_reset_exception(struct pt_regs *regs); -extern int pSeries_machine_check_exception(struct pt_regs *regs); - -static int pseries_shared_idle(void); -static int pseries_dedicated_idle(void); - -static volatile void __iomem * chrp_int_ack_special; -struct mpic *pSeries_mpic; - -void pSeries_get_cpuinfo(struct seq_file *m) -{ - struct device_node *root; - const char *model = ""; - - root = of_find_node_by_path("/"); - if (root) - model = get_property(root, "model", NULL); - seq_printf(m, "machine\t\t: CHRP %s\n", model); - of_node_put(root); -} - -/* Initialize firmware assisted non-maskable interrupts if - * the firmware supports this feature. - * - */ -static void __init fwnmi_init(void) -{ - int ret; - int ibm_nmi_register = rtas_token("ibm,nmi-register"); - if (ibm_nmi_register == RTAS_UNKNOWN_SERVICE) - return; - ret = rtas_call(ibm_nmi_register, 2, 1, NULL, - __pa((unsigned long)system_reset_fwnmi), - __pa((unsigned long)machine_check_fwnmi)); - if (ret == 0) - fwnmi_active = 1; -} - -static int pSeries_irq_cascade(struct pt_regs *regs, void *data) -{ - if (chrp_int_ack_special) - return readb(chrp_int_ack_special); - else - return i8259_irq(smp_processor_id()); -} - -static void __init pSeries_init_mpic(void) -{ - unsigned int *addrp; - struct device_node *np; - int i; - - /* All ISUs are setup, complete initialization */ - mpic_init(pSeries_mpic); - - /* Check what kind of cascade ACK we have */ - if (!(np = of_find_node_by_name(NULL, "pci")) - || !(addrp = (unsigned int *) - get_property(np, "8259-interrupt-acknowledge", NULL))) - printk(KERN_ERR "Cannot find pci to get ack address\n"); - else - chrp_int_ack_special = ioremap(addrp[prom_n_addr_cells(np)-1], 1); - of_node_put(np); - - /* Setup the legacy interrupts & controller */ - for (i = 0; i < NUM_ISA_INTERRUPTS; i++) - irq_desc[i].handler = &i8259_pic; - i8259_init(0); - - /* Hook cascade to mpic */ - mpic_setup_cascade(NUM_ISA_INTERRUPTS, pSeries_irq_cascade, NULL); -} - -static void __init pSeries_setup_mpic(void) -{ - unsigned int *opprop; - unsigned long openpic_addr = 0; - unsigned char senses[NR_IRQS - NUM_ISA_INTERRUPTS]; - struct device_node *root; - int irq_count; - - /* Find the Open PIC if present */ - root = of_find_node_by_path("/"); - opprop = (unsigned int *) get_property(root, "platform-open-pic", NULL); - if (opprop != 0) { - int n = prom_n_addr_cells(root); - - for (openpic_addr = 0; n > 0; --n) - openpic_addr = (openpic_addr << 32) + *opprop++; - printk(KERN_DEBUG "OpenPIC addr: %lx\n", openpic_addr); - } - of_node_put(root); - - BUG_ON(openpic_addr == 0); - - /* Get the sense values from OF */ - prom_get_irq_senses(senses, NUM_ISA_INTERRUPTS, NR_IRQS); - - /* Setup the openpic driver */ - irq_count = NR_IRQS - NUM_ISA_INTERRUPTS - 4; /* leave room for IPIs */ - pSeries_mpic = mpic_alloc(openpic_addr, MPIC_PRIMARY, - 16, 16, irq_count, /* isu size, irq offset, irq count */ - NR_IRQS - 4, /* ipi offset */ - senses, irq_count, /* sense & sense size */ - " MPIC "); -} - -static void pseries_lpar_enable_pmcs(void) -{ - unsigned long set, reset; - - power4_enable_pmcs(); - - set = 1UL << 63; - reset = 0; - plpar_hcall_norets(H_PERFMON, set, reset); - - /* instruct hypervisor to maintain PMCs */ - if (firmware_has_feature(FW_FEATURE_SPLPAR)) - get_paca()->lppaca.pmcregs_in_use = 1; -} - -static void __init pSeries_setup_arch(void) -{ - /* Fixup ppc_md depending on the type of interrupt controller */ - if (ppc64_interrupt_controller == IC_OPEN_PIC) { - ppc_md.init_IRQ = pSeries_init_mpic; - ppc_md.get_irq = mpic_get_irq; - ppc_md.cpu_irq_down = mpic_teardown_this_cpu; - /* Allocate the mpic now, so that find_and_init_phbs() can - * fill the ISUs */ - pSeries_setup_mpic(); - } else { - ppc_md.init_IRQ = xics_init_IRQ; - ppc_md.get_irq = xics_get_irq; - ppc_md.cpu_irq_down = xics_teardown_cpu; - } - -#ifdef CONFIG_SMP - smp_init_pSeries(); -#endif - /* openpic global configuration register (64-bit format). */ - /* openpic Interrupt Source Unit pointer (64-bit format). */ - /* python0 facility area (mmio) (64-bit format) REAL address. */ - - /* init to some ~sane value until calibrate_delay() runs */ - loops_per_jiffy = 50000000; - - if (ROOT_DEV == 0) { - printk("No ramdisk, default root is /dev/sda2\n"); - ROOT_DEV = Root_SDA2; - } - - fwnmi_init(); - - /* Find and initialize PCI host bridges */ - init_pci_config_tokens(); - find_and_init_phbs(); - eeh_init(); - -#ifdef CONFIG_DUMMY_CONSOLE - conswitchp = &dummy_con; -#endif - - pSeries_nvram_init(); - - /* Choose an idle loop */ - if (firmware_has_feature(FW_FEATURE_SPLPAR)) { - vpa_init(boot_cpuid); - if (get_paca()->lppaca.shared_proc) { - printk(KERN_INFO "Using shared processor idle loop\n"); - ppc_md.idle_loop = pseries_shared_idle; - } else { - printk(KERN_INFO "Using dedicated idle loop\n"); - ppc_md.idle_loop = pseries_dedicated_idle; - } - } else { - printk(KERN_INFO "Using default idle loop\n"); - ppc_md.idle_loop = default_idle; - } - - if (systemcfg->platform & PLATFORM_LPAR) - ppc_md.enable_pmcs = pseries_lpar_enable_pmcs; - else - ppc_md.enable_pmcs = power4_enable_pmcs; -} - -static int __init pSeries_init_panel(void) -{ - /* Manually leave the kernel version on the panel. */ - ppc_md.progress("Linux ppc64\n", 0); - ppc_md.progress(system_utsname.version, 0); - - return 0; -} -arch_initcall(pSeries_init_panel); - - -/* Build up the ppc64_firmware_features bitmask field - * using contents of device-tree/ibm,hypertas-functions. - * Ultimately this functionality may be moved into prom.c prom_init(). - */ -static void __init fw_feature_init(void) -{ - struct device_node * dn; - char * hypertas; - unsigned int len; - - DBG(" -> fw_feature_init()\n"); - - ppc64_firmware_features = 0; - dn = of_find_node_by_path("/rtas"); - if (dn == NULL) { - printk(KERN_ERR "WARNING ! Cannot find RTAS in device-tree !\n"); - goto no_rtas; - } - - hypertas = get_property(dn, "ibm,hypertas-functions", &len); - if (hypertas) { - while (len > 0){ - int i, hypertas_len; - /* check value against table of strings */ - for(i=0; i < FIRMWARE_MAX_FEATURES ;i++) { - if ((firmware_features_table[i].name) && - (strcmp(firmware_features_table[i].name,hypertas))==0) { - /* we have a match */ - ppc64_firmware_features |= - (firmware_features_table[i].val); - break; - } - } - hypertas_len = strlen(hypertas); - len -= hypertas_len +1; - hypertas+= hypertas_len +1; - } - } - - of_node_put(dn); - no_rtas: - printk(KERN_INFO "firmware_features = 0x%lx\n", - ppc64_firmware_features); - - DBG(" <- fw_feature_init()\n"); -} - - -static void __init pSeries_discover_pic(void) -{ - struct device_node *np; - char *typep; - - /* - * Setup interrupt mapping options that are needed for finish_device_tree - * to properly parse the OF interrupt tree & do the virtual irq mapping - */ - __irq_offset_value = NUM_ISA_INTERRUPTS; - ppc64_interrupt_controller = IC_INVALID; - for (np = NULL; (np = of_find_node_by_name(np, "interrupt-controller"));) { - typep = (char *)get_property(np, "compatible", NULL); - if (strstr(typep, "open-pic")) - ppc64_interrupt_controller = IC_OPEN_PIC; - else if (strstr(typep, "ppc-xicp")) - ppc64_interrupt_controller = IC_PPC_XIC; - else - printk("pSeries_discover_pic: failed to recognize" - " interrupt-controller\n"); - break; - } -} - -static void pSeries_mach_cpu_die(void) -{ - local_irq_disable(); - idle_task_exit(); - /* Some hardware requires clearing the CPPR, while other hardware does not - * it is safe either way - */ - pSeriesLP_cppr_info(0, 0); - rtas_stop_self(); - /* Should never get here... */ - BUG(); - for(;;); -} - - -/* - * Early initialization. Relocation is on but do not reference unbolted pages - */ -static void __init pSeries_init_early(void) -{ - void *comport; - int iommu_off = 0; - unsigned int default_speed; - u64 physport; - - DBG(" -> pSeries_init_early()\n"); - - fw_feature_init(); - - if (systemcfg->platform & PLATFORM_LPAR) - hpte_init_lpar(); - else { - hpte_init_native(); - iommu_off = (of_chosen && - get_property(of_chosen, "linux,iommu-off", NULL)); - } - - generic_find_legacy_serial_ports(&physport, &default_speed); - - if (systemcfg->platform & PLATFORM_LPAR) - find_udbg_vterm(); - else if (physport) { - /* Map the uart for udbg. */ - comport = (void *)ioremap(physport, 16); - udbg_init_uart(comport, default_speed); - - DBG("Hello World !\n"); - } - - - iommu_init_early_pSeries(); - - pSeries_discover_pic(); - - DBG(" <- pSeries_init_early()\n"); -} - - -static int pSeries_check_legacy_ioport(unsigned int baseport) -{ - struct device_node *np; - -#define I8042_DATA_REG 0x60 -#define FDC_BASE 0x3f0 - - - switch(baseport) { - case I8042_DATA_REG: - np = of_find_node_by_type(NULL, "8042"); - if (np == NULL) - return -ENODEV; - of_node_put(np); - break; - case FDC_BASE: - np = of_find_node_by_type(NULL, "fdc"); - if (np == NULL) - return -ENODEV; - of_node_put(np); - break; - } - return 0; -} - -/* - * Called very early, MMU is off, device-tree isn't unflattened - */ -extern struct machdep_calls pSeries_md; - -static int __init pSeries_probe(int platform) -{ - if (platform != PLATFORM_PSERIES && - platform != PLATFORM_PSERIES_LPAR) - return 0; - - /* if we have some ppc_md fixups for LPAR to do, do - * it here ... - */ - - return 1; -} - -DECLARE_PER_CPU(unsigned long, smt_snooze_delay); - -static inline void dedicated_idle_sleep(unsigned int cpu) -{ - struct paca_struct *ppaca = &paca[cpu ^ 1]; - - /* Only sleep if the other thread is not idle */ - if (!(ppaca->lppaca.idle)) { - local_irq_disable(); - - /* - * We are about to sleep the thread and so wont be polling any - * more. - */ - clear_thread_flag(TIF_POLLING_NRFLAG); - - /* - * SMT dynamic mode. Cede will result in this thread going - * dormant, if the partner thread is still doing work. Thread - * wakes up if partner goes idle, an interrupt is presented, or - * a prod occurs. Returning from the cede enables external - * interrupts. - */ - if (!need_resched()) - cede_processor(); - else - local_irq_enable(); - } else { - /* - * Give the HV an opportunity at the processor, since we are - * not doing any work. - */ - poll_pending(); - } -} - -static int pseries_dedicated_idle(void) -{ - long oldval; - struct paca_struct *lpaca = get_paca(); - unsigned int cpu = smp_processor_id(); - unsigned long start_snooze; - unsigned long *smt_snooze_delay = &__get_cpu_var(smt_snooze_delay); - - while (1) { - /* - * Indicate to the HV that we are idle. Now would be - * a good time to find other work to dispatch. - */ - lpaca->lppaca.idle = 1; - - oldval = test_and_clear_thread_flag(TIF_NEED_RESCHED); - if (!oldval) { - set_thread_flag(TIF_POLLING_NRFLAG); - - start_snooze = __get_tb() + - *smt_snooze_delay * tb_ticks_per_usec; - - while (!need_resched() && !cpu_is_offline(cpu)) { - ppc64_runlatch_off(); - - /* - * Go into low thread priority and possibly - * low power mode. - */ - HMT_low(); - HMT_very_low(); - - if (*smt_snooze_delay != 0 && - __get_tb() > start_snooze) { - HMT_medium(); - dedicated_idle_sleep(cpu); - } - - } - - HMT_medium(); - clear_thread_flag(TIF_POLLING_NRFLAG); - } else { - set_need_resched(); - } - - lpaca->lppaca.idle = 0; - ppc64_runlatch_on(); - - schedule(); - - if (cpu_is_offline(cpu) && system_state == SYSTEM_RUNNING) - cpu_die(); - } -} - -static int pseries_shared_idle(void) -{ - struct paca_struct *lpaca = get_paca(); - unsigned int cpu = smp_processor_id(); - - while (1) { - /* - * Indicate to the HV that we are idle. Now would be - * a good time to find other work to dispatch. - */ - lpaca->lppaca.idle = 1; - - while (!need_resched() && !cpu_is_offline(cpu)) { - local_irq_disable(); - ppc64_runlatch_off(); - - /* - * Yield the processor to the hypervisor. We return if - * an external interrupt occurs (which are driven prior - * to returning here) or if a prod occurs from another - * processor. When returning here, external interrupts - * are enabled. - * - * Check need_resched() again with interrupts disabled - * to avoid a race. - */ - if (!need_resched()) - cede_processor(); - else - local_irq_enable(); - - HMT_medium(); - } - - lpaca->lppaca.idle = 0; - ppc64_runlatch_on(); - - schedule(); - - if (cpu_is_offline(cpu) && system_state == SYSTEM_RUNNING) - cpu_die(); - } - - return 0; -} - -static int pSeries_pci_probe_mode(struct pci_bus *bus) -{ - if (systemcfg->platform & PLATFORM_LPAR) - return PCI_PROBE_DEVTREE; - return PCI_PROBE_NORMAL; -} - -struct machdep_calls __initdata pSeries_md = { - .probe = pSeries_probe, - .setup_arch = pSeries_setup_arch, - .init_early = pSeries_init_early, - .get_cpuinfo = pSeries_get_cpuinfo, - .log_error = pSeries_log_error, - .pcibios_fixup = pSeries_final_fixup, - .pci_probe_mode = pSeries_pci_probe_mode, - .irq_bus_setup = pSeries_irq_bus_setup, - .restart = rtas_restart, - .power_off = rtas_power_off, - .halt = rtas_halt, - .panic = rtas_os_term, - .cpu_die = pSeries_mach_cpu_die, - .get_boot_time = rtas_get_boot_time, - .get_rtc_time = rtas_get_rtc_time, - .set_rtc_time = rtas_set_rtc_time, - .calibrate_decr = generic_calibrate_decr, - .progress = rtas_progress, - .check_legacy_ioport = pSeries_check_legacy_ioport, - .system_reset_exception = pSeries_system_reset_exception, - .machine_check_exception = pSeries_machine_check_exception, -}; diff --git a/arch/ppc64/kernel/pSeries_smp.c b/arch/ppc64/kernel/pSeries_smp.c deleted file mode 100644 index d2c7e2c4733b..000000000000 --- a/arch/ppc64/kernel/pSeries_smp.c +++ /dev/null @@ -1,517 +0,0 @@ -/* - * SMP support for pSeries and BPA machines. - * - * Dave Engebretsen, Peter Bergner, and - * Mike Corrigan {engebret|bergner|mikec}@us.ibm.com - * - * Plus various changes from other IBM teams... - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#undef DEBUG - -#include <linux/config.h> -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/sched.h> -#include <linux/smp.h> -#include <linux/interrupt.h> -#include <linux/delay.h> -#include <linux/init.h> -#include <linux/spinlock.h> -#include <linux/cache.h> -#include <linux/err.h> -#include <linux/sysdev.h> -#include <linux/cpu.h> - -#include <asm/ptrace.h> -#include <asm/atomic.h> -#include <asm/irq.h> -#include <asm/page.h> -#include <asm/pgtable.h> -#include <asm/io.h> -#include <asm/prom.h> -#include <asm/smp.h> -#include <asm/paca.h> -#include <asm/time.h> -#include <asm/machdep.h> -#include <asm/xics.h> -#include <asm/cputable.h> -#include <asm/firmware.h> -#include <asm/system.h> -#include <asm/rtas.h> -#include <asm/plpar_wrappers.h> -#include <asm/pSeries_reconfig.h> - -#include "mpic.h" -#include "bpa_iic.h" - -#ifdef DEBUG -#define DBG(fmt...) udbg_printf(fmt) -#else -#define DBG(fmt...) -#endif - -/* - * The primary thread of each non-boot processor is recorded here before - * smp init. - */ -static cpumask_t of_spin_map; - -extern void pSeries_secondary_smp_init(unsigned long); - -#ifdef CONFIG_HOTPLUG_CPU - -/* Get state of physical CPU. - * Return codes: - * 0 - The processor is in the RTAS stopped state - * 1 - stop-self is in progress - * 2 - The processor is not in the RTAS stopped state - * -1 - Hardware Error - * -2 - Hardware Busy, Try again later. - */ -static int query_cpu_stopped(unsigned int pcpu) -{ - int cpu_status; - int status, qcss_tok; - - qcss_tok = rtas_token("query-cpu-stopped-state"); - if (qcss_tok == RTAS_UNKNOWN_SERVICE) - return -1; - status = rtas_call(qcss_tok, 1, 2, &cpu_status, pcpu); - if (status != 0) { - printk(KERN_ERR - "RTAS query-cpu-stopped-state failed: %i\n", status); - return status; - } - - return cpu_status; -} - -int pSeries_cpu_disable(void) -{ - int cpu = smp_processor_id(); - - cpu_clear(cpu, cpu_online_map); - systemcfg->processorCount--; - - /*fix boot_cpuid here*/ - if (cpu == boot_cpuid) - boot_cpuid = any_online_cpu(cpu_online_map); - - /* FIXME: abstract this to not be platform specific later on */ - xics_migrate_irqs_away(); - return 0; -} - -void pSeries_cpu_die(unsigned int cpu) -{ - int tries; - int cpu_status; - unsigned int pcpu = get_hard_smp_processor_id(cpu); - - for (tries = 0; tries < 25; tries++) { - cpu_status = query_cpu_stopped(pcpu); - if (cpu_status == 0 || cpu_status == -1) - break; - msleep(200); - } - if (cpu_status != 0) { - printk("Querying DEAD? cpu %i (%i) shows %i\n", - cpu, pcpu, cpu_status); - } - - /* Isolation and deallocation are definatly done by - * drslot_chrp_cpu. If they were not they would be - * done here. Change isolate state to Isolate and - * change allocation-state to Unusable. - */ - paca[cpu].cpu_start = 0; -} - -/* - * Update cpu_present_map and paca(s) for a new cpu node. The wrinkle - * here is that a cpu device node may represent up to two logical cpus - * in the SMT case. We must honor the assumption in other code that - * the logical ids for sibling SMT threads x and y are adjacent, such - * that x^1 == y and y^1 == x. - */ -static int pSeries_add_processor(struct device_node *np) -{ - unsigned int cpu; - cpumask_t candidate_map, tmp = CPU_MASK_NONE; - int err = -ENOSPC, len, nthreads, i; - u32 *intserv; - - intserv = (u32 *)get_property(np, "ibm,ppc-interrupt-server#s", &len); - if (!intserv) - return 0; - - nthreads = len / sizeof(u32); - for (i = 0; i < nthreads; i++) - cpu_set(i, tmp); - - lock_cpu_hotplug(); - - BUG_ON(!cpus_subset(cpu_present_map, cpu_possible_map)); - - /* Get a bitmap of unoccupied slots. */ - cpus_xor(candidate_map, cpu_possible_map, cpu_present_map); - if (cpus_empty(candidate_map)) { - /* If we get here, it most likely means that NR_CPUS is - * less than the partition's max processors setting. - */ - printk(KERN_ERR "Cannot add cpu %s; this system configuration" - " supports %d logical cpus.\n", np->full_name, - cpus_weight(cpu_possible_map)); - goto out_unlock; - } - - while (!cpus_empty(tmp)) - if (cpus_subset(tmp, candidate_map)) - /* Found a range where we can insert the new cpu(s) */ - break; - else - cpus_shift_left(tmp, tmp, nthreads); - - if (cpus_empty(tmp)) { - printk(KERN_ERR "Unable to find space in cpu_present_map for" - " processor %s with %d thread(s)\n", np->name, - nthreads); - goto out_unlock; - } - - for_each_cpu_mask(cpu, tmp) { - BUG_ON(cpu_isset(cpu, cpu_present_map)); - cpu_set(cpu, cpu_present_map); - set_hard_smp_processor_id(cpu, *intserv++); - } - err = 0; -out_unlock: - unlock_cpu_hotplug(); - return err; -} - -/* - * Update the present map for a cpu node which is going away, and set - * the hard id in the paca(s) to -1 to be consistent with boot time - * convention for non-present cpus. - */ -static void pSeries_remove_processor(struct device_node *np) -{ - unsigned int cpu; - int len, nthreads, i; - u32 *intserv; - - intserv = (u32 *)get_property(np, "ibm,ppc-interrupt-server#s", &len); - if (!intserv) - return; - - nthreads = len / sizeof(u32); - - lock_cpu_hotplug(); - for (i = 0; i < nthreads; i++) { - for_each_present_cpu(cpu) { - if (get_hard_smp_processor_id(cpu) != intserv[i]) - continue; - BUG_ON(cpu_online(cpu)); - cpu_clear(cpu, cpu_present_map); - set_hard_smp_processor_id(cpu, -1); - break; - } - if (cpu == NR_CPUS) - printk(KERN_WARNING "Could not find cpu to remove " - "with physical id 0x%x\n", intserv[i]); - } - unlock_cpu_hotplug(); -} - -static int pSeries_smp_notifier(struct notifier_block *nb, unsigned long action, void *node) -{ - int err = NOTIFY_OK; - - switch (action) { - case PSERIES_RECONFIG_ADD: - if (pSeries_add_processor(node)) - err = NOTIFY_BAD; - break; - case PSERIES_RECONFIG_REMOVE: - pSeries_remove_processor(node); - break; - default: - err = NOTIFY_DONE; - break; - } - return err; -} - -static struct notifier_block pSeries_smp_nb = { - .notifier_call = pSeries_smp_notifier, -}; - -#endif /* CONFIG_HOTPLUG_CPU */ - -/** - * smp_startup_cpu() - start the given cpu - * - * At boot time, there is nothing to do for primary threads which were - * started from Open Firmware. For anything else, call RTAS with the - * appropriate start location. - * - * Returns: - * 0 - failure - * 1 - success - */ -static inline int __devinit smp_startup_cpu(unsigned int lcpu) -{ - int status; - unsigned long start_here = __pa((u32)*((unsigned long *) - pSeries_secondary_smp_init)); - unsigned int pcpu; - int start_cpu; - - if (cpu_isset(lcpu, of_spin_map)) - /* Already started by OF and sitting in spin loop */ - return 1; - - pcpu = get_hard_smp_processor_id(lcpu); - - /* Fixup atomic count: it exited inside IRQ handler. */ - paca[lcpu].__current->thread_info->preempt_count = 0; - - /* - * If the RTAS start-cpu token does not exist then presume the - * cpu is already spinning. - */ - start_cpu = rtas_token("start-cpu"); - if (start_cpu == RTAS_UNKNOWN_SERVICE) - return 1; - - status = rtas_call(start_cpu, 3, 1, NULL, pcpu, start_here, lcpu); - if (status != 0) { - printk(KERN_ERR "start-cpu failed: %i\n", status); - return 0; - } - - return 1; -} - -#ifdef CONFIG_XICS -static inline void smp_xics_do_message(int cpu, int msg) -{ - set_bit(msg, &xics_ipi_message[cpu].value); - mb(); - xics_cause_IPI(cpu); -} - -static void smp_xics_message_pass(int target, int msg) -{ - unsigned int i; - - if (target < NR_CPUS) { - smp_xics_do_message(target, msg); - } else { - for_each_online_cpu(i) { - if (target == MSG_ALL_BUT_SELF - && i == smp_processor_id()) - continue; - smp_xics_do_message(i, msg); - } - } -} - -static int __init smp_xics_probe(void) -{ - xics_request_IPIs(); - - return cpus_weight(cpu_possible_map); -} - -static void __devinit smp_xics_setup_cpu(int cpu) -{ - if (cpu != boot_cpuid) - xics_setup_cpu(); - - if (firmware_has_feature(FW_FEATURE_SPLPAR)) - vpa_init(cpu); - - cpu_clear(cpu, of_spin_map); - -} -#endif /* CONFIG_XICS */ -#ifdef CONFIG_BPA_IIC -static void smp_iic_message_pass(int target, int msg) -{ - unsigned int i; - - if (target < NR_CPUS) { - iic_cause_IPI(target, msg); - } else { - for_each_online_cpu(i) { - if (target == MSG_ALL_BUT_SELF - && i == smp_processor_id()) - continue; - iic_cause_IPI(i, msg); - } - } -} - -static int __init smp_iic_probe(void) -{ - iic_request_IPIs(); - - return cpus_weight(cpu_possible_map); -} - -static void __devinit smp_iic_setup_cpu(int cpu) -{ - if (cpu != boot_cpuid) - iic_setup_cpu(); -} -#endif /* CONFIG_BPA_IIC */ - -static DEFINE_SPINLOCK(timebase_lock); -static unsigned long timebase = 0; - -static void __devinit pSeries_give_timebase(void) -{ - spin_lock(&timebase_lock); - rtas_call(rtas_token("freeze-time-base"), 0, 1, NULL); - timebase = get_tb(); - spin_unlock(&timebase_lock); - - while (timebase) - barrier(); - rtas_call(rtas_token("thaw-time-base"), 0, 1, NULL); -} - -static void __devinit pSeries_take_timebase(void) -{ - while (!timebase) - barrier(); - spin_lock(&timebase_lock); - set_tb(timebase >> 32, timebase & 0xffffffff); - timebase = 0; - spin_unlock(&timebase_lock); -} - -static void __devinit smp_pSeries_kick_cpu(int nr) -{ - BUG_ON(nr < 0 || nr >= NR_CPUS); - - if (!smp_startup_cpu(nr)) - return; - - /* - * The processor is currently spinning, waiting for the - * cpu_start field to become non-zero After we set cpu_start, - * the processor will continue on to secondary_start - */ - paca[nr].cpu_start = 1; -} - -static int smp_pSeries_cpu_bootable(unsigned int nr) -{ - /* Special case - we inhibit secondary thread startup - * during boot if the user requests it. Odd-numbered - * cpus are assumed to be secondary threads. - */ - if (system_state < SYSTEM_RUNNING && - cpu_has_feature(CPU_FTR_SMT) && - !smt_enabled_at_boot && nr % 2 != 0) - return 0; - - return 1; -} -#ifdef CONFIG_MPIC -static struct smp_ops_t pSeries_mpic_smp_ops = { - .message_pass = smp_mpic_message_pass, - .probe = smp_mpic_probe, - .kick_cpu = smp_pSeries_kick_cpu, - .setup_cpu = smp_mpic_setup_cpu, -}; -#endif -#ifdef CONFIG_XICS -static struct smp_ops_t pSeries_xics_smp_ops = { - .message_pass = smp_xics_message_pass, - .probe = smp_xics_probe, - .kick_cpu = smp_pSeries_kick_cpu, - .setup_cpu = smp_xics_setup_cpu, - .cpu_bootable = smp_pSeries_cpu_bootable, -}; -#endif -#ifdef CONFIG_BPA_IIC -static struct smp_ops_t bpa_iic_smp_ops = { - .message_pass = smp_iic_message_pass, - .probe = smp_iic_probe, - .kick_cpu = smp_pSeries_kick_cpu, - .setup_cpu = smp_iic_setup_cpu, - .cpu_bootable = smp_pSeries_cpu_bootable, -}; -#endif - -/* This is called very early */ -void __init smp_init_pSeries(void) -{ - int i; - - DBG(" -> smp_init_pSeries()\n"); - - switch (ppc64_interrupt_controller) { -#ifdef CONFIG_MPIC - case IC_OPEN_PIC: - smp_ops = &pSeries_mpic_smp_ops; - break; -#endif -#ifdef CONFIG_XICS - case IC_PPC_XIC: - smp_ops = &pSeries_xics_smp_ops; - break; -#endif -#ifdef CONFIG_BPA_IIC - case IC_BPA_IIC: - smp_ops = &bpa_iic_smp_ops; - break; -#endif - default: - panic("Invalid interrupt controller"); - } - -#ifdef CONFIG_HOTPLUG_CPU - smp_ops->cpu_disable = pSeries_cpu_disable; - smp_ops->cpu_die = pSeries_cpu_die; - - /* Processors can be added/removed only on LPAR */ - if (systemcfg->platform == PLATFORM_PSERIES_LPAR) - pSeries_reconfig_notifier_register(&pSeries_smp_nb); -#endif - - /* Mark threads which are still spinning in hold loops. */ - if (cpu_has_feature(CPU_FTR_SMT)) { - for_each_present_cpu(i) { - if (i % 2 == 0) - /* - * Even-numbered logical cpus correspond to - * primary threads. - */ - cpu_set(i, of_spin_map); - } - } else { - of_spin_map = cpu_present_map; - } - - cpu_clear(boot_cpuid, of_spin_map); - - /* Non-lpar has additional take/give timebase */ - if (rtas_token("freeze-time-base") != RTAS_UNKNOWN_SERVICE) { - smp_ops->give_timebase = pSeries_give_timebase; - smp_ops->take_timebase = pSeries_take_timebase; - } - - DBG(" <- smp_init_pSeries()\n"); -} - diff --git a/arch/ppc64/kernel/pSeries_vio.c b/arch/ppc64/kernel/pSeries_vio.c deleted file mode 100644 index e0ae06f58f86..000000000000 --- a/arch/ppc64/kernel/pSeries_vio.c +++ /dev/null @@ -1,273 +0,0 @@ -/* - * IBM PowerPC pSeries Virtual I/O Infrastructure Support. - * - * Copyright (c) 2003-2005 IBM Corp. - * Dave Engebretsen engebret@us.ibm.com - * Santiago Leon santil@us.ibm.com - * Hollis Blanchard <hollisb@us.ibm.com> - * Stephen Rothwell - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/init.h> -#include <linux/module.h> -#include <linux/mm.h> -#include <linux/kobject.h> -#include <asm/iommu.h> -#include <asm/dma.h> -#include <asm/prom.h> -#include <asm/vio.h> -#include <asm/hvcall.h> - -extern struct subsystem devices_subsys; /* needed for vio_find_name() */ - -static void probe_bus_pseries(void) -{ - struct device_node *node_vroot, *of_node; - - node_vroot = find_devices("vdevice"); - if ((node_vroot == NULL) || (node_vroot->child == NULL)) - /* this machine doesn't do virtual IO, and that's ok */ - return; - - /* - * Create struct vio_devices for each virtual device in the device tree. - * Drivers will associate with them later. - */ - for (of_node = node_vroot->child; of_node != NULL; - of_node = of_node->sibling) { - printk(KERN_DEBUG "%s: processing %p\n", __FUNCTION__, of_node); - vio_register_device_node(of_node); - } -} - -/** - * vio_match_device_pseries: - Tell if a pSeries VIO device matches a - * vio_device_id - */ -static int vio_match_device_pseries(const struct vio_device_id *id, - const struct vio_dev *dev) -{ - return (strncmp(dev->type, id->type, strlen(id->type)) == 0) && - device_is_compatible(dev->dev.platform_data, id->compat); -} - -static void vio_release_device_pseries(struct device *dev) -{ - /* XXX free TCE table */ - of_node_put(dev->platform_data); -} - -static ssize_t viodev_show_devspec(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct device_node *of_node = dev->platform_data; - - return sprintf(buf, "%s\n", of_node->full_name); -} -DEVICE_ATTR(devspec, S_IRUSR | S_IRGRP | S_IROTH, viodev_show_devspec, NULL); - -static void vio_unregister_device_pseries(struct vio_dev *viodev) -{ - device_remove_file(&viodev->dev, &dev_attr_devspec); -} - -static struct vio_bus_ops vio_bus_ops_pseries = { - .match = vio_match_device_pseries, - .unregister_device = vio_unregister_device_pseries, - .release_device = vio_release_device_pseries, -}; - -/** - * vio_bus_init_pseries: - Initialize the pSeries virtual IO bus - */ -static int __init vio_bus_init_pseries(void) -{ - int err; - - err = vio_bus_init(&vio_bus_ops_pseries); - if (err == 0) - probe_bus_pseries(); - return err; -} - -__initcall(vio_bus_init_pseries); - -/** - * vio_build_iommu_table: - gets the dma information from OF and - * builds the TCE tree. - * @dev: the virtual device. - * - * Returns a pointer to the built tce tree, or NULL if it can't - * find property. -*/ -static struct iommu_table *vio_build_iommu_table(struct vio_dev *dev) -{ - unsigned int *dma_window; - struct iommu_table *newTceTable; - unsigned long offset; - int dma_window_property_size; - - dma_window = (unsigned int *) get_property(dev->dev.platform_data, "ibm,my-dma-window", &dma_window_property_size); - if(!dma_window) { - return NULL; - } - - newTceTable = (struct iommu_table *) kmalloc(sizeof(struct iommu_table), GFP_KERNEL); - - /* There should be some code to extract the phys-encoded offset - using prom_n_addr_cells(). However, according to a comment - on earlier versions, it's always zero, so we don't bother */ - offset = dma_window[1] >> PAGE_SHIFT; - - /* TCE table size - measured in tce entries */ - newTceTable->it_size = dma_window[4] >> PAGE_SHIFT; - /* offset for VIO should always be 0 */ - newTceTable->it_offset = offset; - newTceTable->it_busno = 0; - newTceTable->it_index = (unsigned long)dma_window[0]; - newTceTable->it_type = TCE_VB; - - return iommu_init_table(newTceTable); -} - -/** - * vio_register_device_node: - Register a new vio device. - * @of_node: The OF node for this device. - * - * Creates and initializes a vio_dev structure from the data in - * of_node (dev.platform_data) and adds it to the list of virtual devices. - * Returns a pointer to the created vio_dev or NULL if node has - * NULL device_type or compatible fields. - */ -struct vio_dev * __devinit vio_register_device_node(struct device_node *of_node) -{ - struct vio_dev *viodev; - unsigned int *unit_address; - unsigned int *irq_p; - - /* we need the 'device_type' property, in order to match with drivers */ - if ((NULL == of_node->type)) { - printk(KERN_WARNING - "%s: node %s missing 'device_type'\n", __FUNCTION__, - of_node->name ? of_node->name : "<unknown>"); - return NULL; - } - - unit_address = (unsigned int *)get_property(of_node, "reg", NULL); - if (!unit_address) { - printk(KERN_WARNING "%s: node %s missing 'reg'\n", __FUNCTION__, - of_node->name ? of_node->name : "<unknown>"); - return NULL; - } - - /* allocate a vio_dev for this node */ - viodev = kmalloc(sizeof(struct vio_dev), GFP_KERNEL); - if (!viodev) { - return NULL; - } - memset(viodev, 0, sizeof(struct vio_dev)); - - viodev->dev.platform_data = of_node_get(of_node); - - viodev->irq = NO_IRQ; - irq_p = (unsigned int *)get_property(of_node, "interrupts", NULL); - if (irq_p) { - int virq = virt_irq_create_mapping(*irq_p); - if (virq == NO_IRQ) { - printk(KERN_ERR "Unable to allocate interrupt " - "number for %s\n", of_node->full_name); - } else - viodev->irq = irq_offset_up(virq); - } - - snprintf(viodev->dev.bus_id, BUS_ID_SIZE, "%x", *unit_address); - viodev->name = of_node->name; - viodev->type = of_node->type; - viodev->unit_address = *unit_address; - viodev->iommu_table = vio_build_iommu_table(viodev); - - /* register with generic device framework */ - if (vio_register_device(viodev) == NULL) { - /* XXX free TCE table */ - kfree(viodev); - return NULL; - } - device_create_file(&viodev->dev, &dev_attr_devspec); - - return viodev; -} -EXPORT_SYMBOL(vio_register_device_node); - -/** - * vio_get_attribute: - get attribute for virtual device - * @vdev: The vio device to get property. - * @which: The property/attribute to be extracted. - * @length: Pointer to length of returned data size (unused if NULL). - * - * Calls prom.c's get_property() to return the value of the - * attribute specified by the preprocessor constant @which -*/ -const void * vio_get_attribute(struct vio_dev *vdev, void* which, int* length) -{ - return get_property(vdev->dev.platform_data, (char*)which, length); -} -EXPORT_SYMBOL(vio_get_attribute); - -/* vio_find_name() - internal because only vio.c knows how we formatted the - * kobject name - * XXX once vio_bus_type.devices is actually used as a kset in - * drivers/base/bus.c, this function should be removed in favor of - * "device_find(kobj_name, &vio_bus_type)" - */ -static struct vio_dev *vio_find_name(const char *kobj_name) -{ - struct kobject *found; - - found = kset_find_obj(&devices_subsys.kset, kobj_name); - if (!found) - return NULL; - - return to_vio_dev(container_of(found, struct device, kobj)); -} - -/** - * vio_find_node - find an already-registered vio_dev - * @vnode: device_node of the virtual device we're looking for - */ -struct vio_dev *vio_find_node(struct device_node *vnode) -{ - uint32_t *unit_address; - char kobj_name[BUS_ID_SIZE]; - - /* construct the kobject name from the device node */ - unit_address = (uint32_t *)get_property(vnode, "reg", NULL); - if (!unit_address) - return NULL; - snprintf(kobj_name, BUS_ID_SIZE, "%x", *unit_address); - - return vio_find_name(kobj_name); -} -EXPORT_SYMBOL(vio_find_node); - -int vio_enable_interrupts(struct vio_dev *dev) -{ - int rc = h_vio_signal(dev->unit_address, VIO_IRQ_ENABLE); - if (rc != H_Success) - printk(KERN_ERR "vio: Error 0x%x enabling interrupts\n", rc); - return rc; -} -EXPORT_SYMBOL(vio_enable_interrupts); - -int vio_disable_interrupts(struct vio_dev *dev) -{ - int rc = h_vio_signal(dev->unit_address, VIO_IRQ_DISABLE); - if (rc != H_Success) - printk(KERN_ERR "vio: Error 0x%x disabling interrupts\n", rc); - return rc; -} -EXPORT_SYMBOL(vio_disable_interrupts); diff --git a/arch/ppc64/kernel/pci.c b/arch/ppc64/kernel/pci.c index ff4be1da69d5..b2fb6746f00b 100644 --- a/arch/ppc64/kernel/pci.c +++ b/arch/ppc64/kernel/pci.c @@ -31,8 +31,7 @@ #include <asm/irq.h> #include <asm/machdep.h> #include <asm/udbg.h> - -#include "pci.h" +#include <asm/ppc-pci.h> #ifdef DEBUG #define DBG(fmt...) udbg_printf(fmt) @@ -881,9 +880,9 @@ static void __devinit pci_process_ISA_OF_ranges(struct device_node *isa_node, } void __devinit pci_process_bridge_OF_ranges(struct pci_controller *hose, - struct device_node *dev) + struct device_node *dev, int prim) { - unsigned int *ranges; + unsigned int *ranges, pci_space; unsigned long size; int rlen = 0; int memno = 0; @@ -906,16 +905,39 @@ void __devinit pci_process_bridge_OF_ranges(struct pci_controller *hose, ranges = (unsigned int *) get_property(dev, "ranges", &rlen); while ((rlen -= np * sizeof(unsigned int)) >= 0) { res = NULL; - pci_addr = (unsigned long)ranges[1] << 32 | ranges[2]; + pci_space = ranges[0]; + pci_addr = ((unsigned long)ranges[1] << 32) | ranges[2]; cpu_phys_addr = ranges[3]; - if (na == 2) - cpu_phys_addr = cpu_phys_addr << 32 | ranges[4]; + if (na >= 2) + cpu_phys_addr = (cpu_phys_addr << 32) | ranges[4]; - size = (unsigned long)ranges[na+3] << 32 | ranges[na+4]; + size = ((unsigned long)ranges[na+3] << 32) | ranges[na+4]; + ranges += np; if (size == 0) continue; - switch ((ranges[0] >> 24) & 0x3) { + + /* Now consume following elements while they are contiguous */ + while (rlen >= np * sizeof(unsigned int)) { + unsigned long addr, phys; + + if (ranges[0] != pci_space) + break; + addr = ((unsigned long)ranges[1] << 32) | ranges[2]; + phys = ranges[3]; + if (na >= 2) + phys = (phys << 32) | ranges[4]; + if (addr != pci_addr + size || + phys != cpu_phys_addr + size) + break; + + size += ((unsigned long)ranges[na+3] << 32) + | ranges[na+4]; + ranges += np; + rlen -= np * sizeof(unsigned int); + } + + switch ((pci_space >> 24) & 0x3) { case 1: /* I/O space */ hose->io_base_phys = cpu_phys_addr; hose->pci_io_size = size; @@ -949,7 +971,6 @@ void __devinit pci_process_bridge_OF_ranges(struct pci_controller *hose, res->sibling = NULL; res->child = NULL; } - ranges += np; } } diff --git a/arch/ppc64/kernel/pci.h b/arch/ppc64/kernel/pci.h deleted file mode 100644 index 5eb2cc320566..000000000000 --- a/arch/ppc64/kernel/pci.h +++ /dev/null @@ -1,54 +0,0 @@ -/* - * c 2001 PPC 64 Team, IBM Corp - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#ifndef __PPC_KERNEL_PCI_H__ -#define __PPC_KERNEL_PCI_H__ - -#include <linux/pci.h> -#include <asm/pci-bridge.h> - -extern unsigned long isa_io_base; - -extern void pci_setup_pci_controller(struct pci_controller *hose); -extern void pci_setup_phb_io(struct pci_controller *hose, int primary); -extern void pci_setup_phb_io_dynamic(struct pci_controller *hose, int primary); - - -extern struct list_head hose_list; -extern int global_phb_number; - -extern unsigned long find_and_init_phbs(void); - -extern struct pci_dev *ppc64_isabridge_dev; /* may be NULL if no ISA bus */ - -/* PCI device_node operations */ -struct device_node; -typedef void *(*traverse_func)(struct device_node *me, void *data); -void *traverse_pci_devices(struct device_node *start, traverse_func pre, - void *data); - -void pci_devs_phb_init(void); -void pci_devs_phb_init_dynamic(struct pci_controller *phb); - -/* PCI address cache management routines */ -void pci_addr_cache_insert_device(struct pci_dev *dev); -void pci_addr_cache_remove_device(struct pci_dev *dev); - -/* From rtas_pci.h */ -void init_pci_config_tokens (void); -unsigned long get_phb_buid (struct device_node *); - -/* From pSeries_pci.h */ -extern void pSeries_final_fixup(void); -extern void pSeries_irq_bus_setup(struct pci_bus *bus); - -extern unsigned long pci_probe_only; -extern unsigned long pci_assign_all_buses; -extern int pci_read_irq_line(struct pci_dev *pci_dev); - -#endif /* __PPC_KERNEL_PCI_H__ */ diff --git a/arch/ppc64/kernel/pci_direct_iommu.c b/arch/ppc64/kernel/pci_direct_iommu.c index 54055c81017a..e1a32f802c0b 100644 --- a/arch/ppc64/kernel/pci_direct_iommu.c +++ b/arch/ppc64/kernel/pci_direct_iommu.c @@ -27,8 +27,7 @@ #include <asm/machdep.h> #include <asm/pmac_feature.h> #include <asm/abs_addr.h> - -#include "pci.h" +#include <asm/ppc-pci.h> static void *pci_direct_alloc_coherent(struct device *hwdev, size_t size, dma_addr_t *dma_handle, gfp_t flag) diff --git a/arch/ppc64/kernel/pci_dn.c b/arch/ppc64/kernel/pci_dn.c index a86389d07d57..493bbe43f5b4 100644 --- a/arch/ppc64/kernel/pci_dn.c +++ b/arch/ppc64/kernel/pci_dn.c @@ -30,8 +30,7 @@ #include <asm/prom.h> #include <asm/pci-bridge.h> #include <asm/pSeries_reconfig.h> - -#include "pci.h" +#include <asm/ppc-pci.h> /* * Traverse_func that inits the PCI fields of the device node. diff --git a/arch/ppc64/kernel/pci_iommu.c b/arch/ppc64/kernel/pci_iommu.c index d9e33b7d4203..bdf15dbbf4f0 100644 --- a/arch/ppc64/kernel/pci_iommu.c +++ b/arch/ppc64/kernel/pci_iommu.c @@ -1,8 +1,8 @@ /* * arch/ppc64/kernel/pci_iommu.c * Copyright (C) 2001 Mike Corrigan & Dave Engebretsen, IBM Corporation - * - * Rewrite, cleanup, new allocation schemes: + * + * Rewrite, cleanup, new allocation schemes: * Copyright (C) 2004 Olof Johansson, IBM Corporation * * Dynamic DMA mapping support, platform-independent parts. @@ -11,19 +11,18 @@ * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2 of the License, or * (at your option) any later version. - * + * * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. - * + * * You should have received a copy of the GNU General Public License * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#include <linux/config.h> #include <linux/init.h> #include <linux/types.h> #include <linux/slab.h> @@ -37,11 +36,7 @@ #include <asm/iommu.h> #include <asm/pci-bridge.h> #include <asm/machdep.h> -#include "pci.h" - -#ifdef CONFIG_PPC_ISERIES -#include <asm/iSeries/iSeries_pci.h> -#endif /* CONFIG_PPC_ISERIES */ +#include <asm/ppc-pci.h> /* * We can use ->sysdata directly and avoid the extra work in @@ -61,13 +56,7 @@ static inline struct iommu_table *devnode_table(struct device *dev) } else pdev = to_pci_dev(dev); -#ifdef CONFIG_PPC_ISERIES - return ISERIES_DEVNODE(pdev)->iommu_table; -#endif /* CONFIG_PPC_ISERIES */ - -#ifdef CONFIG_PPC_MULTIPLATFORM return PCI_DN(PCI_GET_DN(pdev))->iommu_table; -#endif /* CONFIG_PPC_MULTIPLATFORM */ } diff --git a/arch/ppc64/kernel/pmac.h b/arch/ppc64/kernel/pmac.h deleted file mode 100644 index 40e1c5030f74..000000000000 --- a/arch/ppc64/kernel/pmac.h +++ /dev/null @@ -1,31 +0,0 @@ -#ifndef __PMAC_H__ -#define __PMAC_H__ - -#include <linux/pci.h> -#include <linux/ide.h> - -/* - * Declaration for the various functions exported by the - * pmac_* files. Mostly for use by pmac_setup - */ - -extern void pmac_get_boot_time(struct rtc_time *tm); -extern void pmac_get_rtc_time(struct rtc_time *tm); -extern int pmac_set_rtc_time(struct rtc_time *tm); -extern void pmac_read_rtc_time(void); -extern void pmac_calibrate_decr(void); - -extern void pmac_pcibios_fixup(void); -extern void pmac_pci_init(void); -extern void pmac_setup_pci_dma(void); -extern void pmac_check_ht_link(void); - -extern void pmac_setup_smp(void); - -extern unsigned long pmac_ide_get_base(int index); -extern void pmac_ide_init_hwif_ports(hw_regs_t *hw, - unsigned long data_port, unsigned long ctrl_port, int *irq); - -extern void pmac_nvram_init(void); - -#endif /* __PMAC_H__ */ diff --git a/arch/ppc64/kernel/pmac_feature.c b/arch/ppc64/kernel/pmac_feature.c deleted file mode 100644 index eb4e6c3f694d..000000000000 --- a/arch/ppc64/kernel/pmac_feature.c +++ /dev/null @@ -1,767 +0,0 @@ -/* - * arch/ppc/platforms/pmac_feature.c - * - * Copyright (C) 1996-2001 Paul Mackerras (paulus@cs.anu.edu.au) - * Ben. Herrenschmidt (benh@kernel.crashing.org) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * TODO: - * - * - Replace mdelay with some schedule loop if possible - * - Shorten some obfuscated delays on some routines (like modem - * power) - * - Refcount some clocks (see darwin) - * - Split split split... - * - */ -#include <linux/config.h> -#include <linux/types.h> -#include <linux/init.h> -#include <linux/delay.h> -#include <linux/kernel.h> -#include <linux/sched.h> -#include <linux/spinlock.h> -#include <linux/adb.h> -#include <linux/pmu.h> -#include <linux/ioport.h> -#include <linux/pci.h> -#include <asm/sections.h> -#include <asm/errno.h> -#include <asm/keylargo.h> -#include <asm/uninorth.h> -#include <asm/io.h> -#include <asm/prom.h> -#include <asm/machdep.h> -#include <asm/pmac_feature.h> -#include <asm/dbdma.h> -#include <asm/pci-bridge.h> -#include <asm/pmac_low_i2c.h> - -#undef DEBUG_FEATURE - -#ifdef DEBUG_FEATURE -#define DBG(fmt...) printk(KERN_DEBUG fmt) -#else -#define DBG(fmt...) -#endif - -/* - * We use a single global lock to protect accesses. Each driver has - * to take care of its own locking - */ -static DEFINE_SPINLOCK(feature_lock __pmacdata); - -#define LOCK(flags) spin_lock_irqsave(&feature_lock, flags); -#define UNLOCK(flags) spin_unlock_irqrestore(&feature_lock, flags); - - -/* - * Instance of some macio stuffs - */ -struct macio_chip macio_chips[MAX_MACIO_CHIPS] __pmacdata; - -struct macio_chip* __pmac macio_find(struct device_node* child, int type) -{ - while(child) { - int i; - - for (i=0; i < MAX_MACIO_CHIPS && macio_chips[i].of_node; i++) - if (child == macio_chips[i].of_node && - (!type || macio_chips[i].type == type)) - return &macio_chips[i]; - child = child->parent; - } - return NULL; -} -EXPORT_SYMBOL_GPL(macio_find); - -static const char* macio_names[] __pmacdata = -{ - "Unknown", - "Grand Central", - "OHare", - "OHareII", - "Heathrow", - "Gatwick", - "Paddington", - "Keylargo", - "Pangea", - "Intrepid", - "K2" -}; - - - -/* - * Uninorth reg. access. Note that Uni-N regs are big endian - */ - -#define UN_REG(r) (uninorth_base + ((r) >> 2)) -#define UN_IN(r) (in_be32(UN_REG(r))) -#define UN_OUT(r,v) (out_be32(UN_REG(r), (v))) -#define UN_BIS(r,v) (UN_OUT((r), UN_IN(r) | (v))) -#define UN_BIC(r,v) (UN_OUT((r), UN_IN(r) & ~(v))) - -static struct device_node* uninorth_node __pmacdata; -static u32* uninorth_base __pmacdata; -static u32 uninorth_rev __pmacdata; -static void *u3_ht; - -extern struct device_node *k2_skiplist[2]; - -/* - * For each motherboard family, we have a table of functions pointers - * that handle the various features. - */ - -typedef long (*feature_call)(struct device_node* node, long param, long value); - -struct feature_table_entry { - unsigned int selector; - feature_call function; -}; - -struct pmac_mb_def -{ - const char* model_string; - const char* model_name; - int model_id; - struct feature_table_entry* features; - unsigned long board_flags; -}; -static struct pmac_mb_def pmac_mb __pmacdata; - -/* - * Here are the chip specific feature functions - */ - - -static long __pmac g5_read_gpio(struct device_node* node, long param, long value) -{ - struct macio_chip* macio = &macio_chips[0]; - - return MACIO_IN8(param); -} - - -static long __pmac g5_write_gpio(struct device_node* node, long param, long value) -{ - struct macio_chip* macio = &macio_chips[0]; - - MACIO_OUT8(param, (u8)(value & 0xff)); - return 0; -} - -static long __pmac g5_gmac_enable(struct device_node* node, long param, long value) -{ - struct macio_chip* macio = &macio_chips[0]; - unsigned long flags; - - if (node == NULL) - return -ENODEV; - - LOCK(flags); - if (value) { - MACIO_BIS(KEYLARGO_FCR1, K2_FCR1_GMAC_CLK_ENABLE); - mb(); - k2_skiplist[0] = NULL; - } else { - k2_skiplist[0] = node; - mb(); - MACIO_BIC(KEYLARGO_FCR1, K2_FCR1_GMAC_CLK_ENABLE); - } - - UNLOCK(flags); - mdelay(1); - - return 0; -} - -static long __pmac g5_fw_enable(struct device_node* node, long param, long value) -{ - struct macio_chip* macio = &macio_chips[0]; - unsigned long flags; - - if (node == NULL) - return -ENODEV; - - LOCK(flags); - if (value) { - MACIO_BIS(KEYLARGO_FCR1, K2_FCR1_FW_CLK_ENABLE); - mb(); - k2_skiplist[1] = NULL; - } else { - k2_skiplist[1] = node; - mb(); - MACIO_BIC(KEYLARGO_FCR1, K2_FCR1_FW_CLK_ENABLE); - } - - UNLOCK(flags); - mdelay(1); - - return 0; -} - -static long __pmac g5_mpic_enable(struct device_node* node, long param, long value) -{ - unsigned long flags; - - if (node->parent == NULL || strcmp(node->parent->name, "u3")) - return 0; - - LOCK(flags); - UN_BIS(U3_TOGGLE_REG, U3_MPIC_RESET | U3_MPIC_OUTPUT_ENABLE); - UNLOCK(flags); - - return 0; -} - -static long __pmac g5_eth_phy_reset(struct device_node* node, long param, long value) -{ - struct macio_chip* macio = &macio_chips[0]; - struct device_node *phy; - int need_reset; - - /* - * We must not reset the combo PHYs, only the BCM5221 found in - * the iMac G5. - */ - phy = of_get_next_child(node, NULL); - if (!phy) - return -ENODEV; - need_reset = device_is_compatible(phy, "B5221"); - of_node_put(phy); - if (!need_reset) - return 0; - - /* PHY reset is GPIO 29, not in device-tree unfortunately */ - MACIO_OUT8(K2_GPIO_EXTINT_0 + 29, - KEYLARGO_GPIO_OUTPUT_ENABLE | KEYLARGO_GPIO_OUTOUT_DATA); - /* Thankfully, this is now always called at a time when we can - * schedule by sungem. - */ - msleep(10); - MACIO_OUT8(K2_GPIO_EXTINT_0 + 29, 0); - - return 0; -} - -static long __pmac g5_i2s_enable(struct device_node *node, long param, long value) -{ - /* Very crude implementation for now */ - struct macio_chip* macio = &macio_chips[0]; - unsigned long flags; - - if (value == 0) - return 0; /* don't disable yet */ - - LOCK(flags); - MACIO_BIS(KEYLARGO_FCR3, KL3_CLK45_ENABLE | KL3_CLK49_ENABLE | - KL3_I2S0_CLK18_ENABLE); - udelay(10); - MACIO_BIS(KEYLARGO_FCR1, K2_FCR1_I2S0_CELL_ENABLE | - K2_FCR1_I2S0_CLK_ENABLE_BIT | K2_FCR1_I2S0_ENABLE); - udelay(10); - MACIO_BIC(KEYLARGO_FCR1, K2_FCR1_I2S0_RESET); - UNLOCK(flags); - udelay(10); - - return 0; -} - - -#ifdef CONFIG_SMP -static long __pmac g5_reset_cpu(struct device_node* node, long param, long value) -{ - unsigned int reset_io = 0; - unsigned long flags; - struct macio_chip* macio; - struct device_node* np; - - macio = &macio_chips[0]; - if (macio->type != macio_keylargo2) - return -ENODEV; - - np = find_path_device("/cpus"); - if (np == NULL) - return -ENODEV; - for (np = np->child; np != NULL; np = np->sibling) { - u32* num = (u32 *)get_property(np, "reg", NULL); - u32* rst = (u32 *)get_property(np, "soft-reset", NULL); - if (num == NULL || rst == NULL) - continue; - if (param == *num) { - reset_io = *rst; - break; - } - } - if (np == NULL || reset_io == 0) - return -ENODEV; - - LOCK(flags); - MACIO_OUT8(reset_io, KEYLARGO_GPIO_OUTPUT_ENABLE); - (void)MACIO_IN8(reset_io); - udelay(1); - MACIO_OUT8(reset_io, 0); - (void)MACIO_IN8(reset_io); - UNLOCK(flags); - - return 0; -} -#endif /* CONFIG_SMP */ - -/* - * This can be called from pmac_smp so isn't static - * - * This takes the second CPU off the bus on dual CPU machines - * running UP - */ -void __pmac g5_phy_disable_cpu1(void) -{ - UN_OUT(U3_API_PHY_CONFIG_1, 0); -} - -static long __pmac generic_get_mb_info(struct device_node* node, long param, long value) -{ - switch(param) { - case PMAC_MB_INFO_MODEL: - return pmac_mb.model_id; - case PMAC_MB_INFO_FLAGS: - return pmac_mb.board_flags; - case PMAC_MB_INFO_NAME: - /* hack hack hack... but should work */ - *((const char **)value) = pmac_mb.model_name; - return 0; - } - return -EINVAL; -} - - -/* - * Table definitions - */ - -/* Used on any machine - */ -static struct feature_table_entry any_features[] __pmacdata = { - { PMAC_FTR_GET_MB_INFO, generic_get_mb_info }, - { 0, NULL } -}; - -/* G5 features - */ -static struct feature_table_entry g5_features[] __pmacdata = { - { PMAC_FTR_GMAC_ENABLE, g5_gmac_enable }, - { PMAC_FTR_1394_ENABLE, g5_fw_enable }, - { PMAC_FTR_ENABLE_MPIC, g5_mpic_enable }, - { PMAC_FTR_READ_GPIO, g5_read_gpio }, - { PMAC_FTR_WRITE_GPIO, g5_write_gpio }, - { PMAC_FTR_GMAC_PHY_RESET, g5_eth_phy_reset }, - { PMAC_FTR_SOUND_CHIP_ENABLE, g5_i2s_enable }, -#ifdef CONFIG_SMP - { PMAC_FTR_RESET_CPU, g5_reset_cpu }, -#endif /* CONFIG_SMP */ - { 0, NULL } -}; - -static struct pmac_mb_def pmac_mb_defs[] __pmacdata = { - { "PowerMac7,2", "PowerMac G5", - PMAC_TYPE_POWERMAC_G5, g5_features, - 0, - }, - { "PowerMac7,3", "PowerMac G5", - PMAC_TYPE_POWERMAC_G5, g5_features, - 0, - }, - { "PowerMac8,1", "iMac G5", - PMAC_TYPE_IMAC_G5, g5_features, - 0, - }, - { "PowerMac9,1", "PowerMac G5", - PMAC_TYPE_POWERMAC_G5_U3L, g5_features, - 0, - }, - { "RackMac3,1", "XServe G5", - PMAC_TYPE_XSERVE_G5, g5_features, - 0, - }, -}; - -/* - * The toplevel feature_call callback - */ -long __pmac pmac_do_feature_call(unsigned int selector, ...) -{ - struct device_node* node; - long param, value; - int i; - feature_call func = NULL; - va_list args; - - if (pmac_mb.features) - for (i=0; pmac_mb.features[i].function; i++) - if (pmac_mb.features[i].selector == selector) { - func = pmac_mb.features[i].function; - break; - } - if (!func) - for (i=0; any_features[i].function; i++) - if (any_features[i].selector == selector) { - func = any_features[i].function; - break; - } - if (!func) - return -ENODEV; - - va_start(args, selector); - node = (struct device_node*)va_arg(args, void*); - param = va_arg(args, long); - value = va_arg(args, long); - va_end(args); - - return func(node, param, value); -} - -static int __init probe_motherboard(void) -{ - int i; - struct macio_chip* macio = &macio_chips[0]; - const char* model = NULL; - struct device_node *dt; - - /* Lookup known motherboard type in device-tree. First try an - * exact match on the "model" property, then try a "compatible" - * match is none is found. - */ - dt = find_devices("device-tree"); - if (dt != NULL) - model = (const char *) get_property(dt, "model", NULL); - for(i=0; model && i<(sizeof(pmac_mb_defs)/sizeof(struct pmac_mb_def)); i++) { - if (strcmp(model, pmac_mb_defs[i].model_string) == 0) { - pmac_mb = pmac_mb_defs[i]; - goto found; - } - } - for(i=0; i<(sizeof(pmac_mb_defs)/sizeof(struct pmac_mb_def)); i++) { - if (machine_is_compatible(pmac_mb_defs[i].model_string)) { - pmac_mb = pmac_mb_defs[i]; - goto found; - } - } - - /* Fallback to selection depending on mac-io chip type */ - switch(macio->type) { - case macio_keylargo2: - pmac_mb.model_id = PMAC_TYPE_UNKNOWN_K2; - pmac_mb.model_name = "Unknown K2-based"; - pmac_mb.features = g5_features; - - default: - return -ENODEV; - } -found: - /* Check for "mobile" machine */ - if (model && (strncmp(model, "PowerBook", 9) == 0 - || strncmp(model, "iBook", 5) == 0)) - pmac_mb.board_flags |= PMAC_MB_MOBILE; - - - printk(KERN_INFO "PowerMac motherboard: %s\n", pmac_mb.model_name); - return 0; -} - -/* Initialize the Core99 UniNorth host bridge and memory controller - */ -static void __init probe_uninorth(void) -{ - uninorth_node = of_find_node_by_name(NULL, "u3"); - if (uninorth_node && uninorth_node->n_addrs > 0) { - /* Small hack until I figure out if parsing in prom.c is correct. I should - * get rid of those pre-parsed junk anyway - */ - unsigned long address = uninorth_node->addrs[0].address; - uninorth_base = ioremap(address, 0x40000); - uninorth_rev = in_be32(UN_REG(UNI_N_VERSION)); - u3_ht = ioremap(address + U3_HT_CONFIG_BASE, 0x1000); - } else - uninorth_node = NULL; - - if (!uninorth_node) - return; - - printk(KERN_INFO "Found U3 memory controller & host bridge, revision: %d\n", - uninorth_rev); - printk(KERN_INFO "Mapped at 0x%08lx\n", (unsigned long)uninorth_base); - -} - -static void __init probe_one_macio(const char* name, const char* compat, int type) -{ - struct device_node* node; - int i; - volatile u32* base; - u32* revp; - - node = find_devices(name); - if (!node || !node->n_addrs) - return; - if (compat) - do { - if (device_is_compatible(node, compat)) - break; - node = node->next; - } while (node); - if (!node) - return; - for(i=0; i<MAX_MACIO_CHIPS; i++) { - if (!macio_chips[i].of_node) - break; - if (macio_chips[i].of_node == node) - return; - } - if (i >= MAX_MACIO_CHIPS) { - printk(KERN_ERR "pmac_feature: Please increase MAX_MACIO_CHIPS !\n"); - printk(KERN_ERR "pmac_feature: %s skipped\n", node->full_name); - return; - } - base = (volatile u32*)ioremap(node->addrs[0].address, node->addrs[0].size); - if (!base) { - printk(KERN_ERR "pmac_feature: Can't map mac-io chip !\n"); - return; - } - if (type == macio_keylargo) { - u32* did = (u32 *)get_property(node, "device-id", NULL); - if (*did == 0x00000025) - type = macio_pangea; - if (*did == 0x0000003e) - type = macio_intrepid; - } - macio_chips[i].of_node = node; - macio_chips[i].type = type; - macio_chips[i].base = base; - macio_chips[i].flags = MACIO_FLAG_SCCB_ON | MACIO_FLAG_SCCB_ON; - macio_chips[i].name = macio_names[type]; - revp = (u32 *)get_property(node, "revision-id", NULL); - if (revp) - macio_chips[i].rev = *revp; - printk(KERN_INFO "Found a %s mac-io controller, rev: %d, mapped at 0x%p\n", - macio_names[type], macio_chips[i].rev, macio_chips[i].base); -} - -static int __init -probe_macios(void) -{ - probe_one_macio("mac-io", "K2-Keylargo", macio_keylargo2); - - macio_chips[0].lbus.index = 0; - macio_chips[1].lbus.index = 1; - - return (macio_chips[0].of_node == NULL) ? -ENODEV : 0; -} - -static void __init -set_initial_features(void) -{ - struct device_node *np; - - if (macio_chips[0].type == macio_keylargo2) { -#ifndef CONFIG_SMP - /* On SMP machines running UP, we have the second CPU eating - * bus cycles. We need to take it off the bus. This is done - * from pmac_smp for SMP kernels running on one CPU - */ - np = of_find_node_by_type(NULL, "cpu"); - if (np != NULL) - np = of_find_node_by_type(np, "cpu"); - if (np != NULL) { - g5_phy_disable_cpu1(); - of_node_put(np); - } -#endif /* CONFIG_SMP */ - /* Enable GMAC for now for PCI probing. It will be disabled - * later on after PCI probe - */ - np = of_find_node_by_name(NULL, "ethernet"); - while(np) { - if (device_is_compatible(np, "K2-GMAC")) - g5_gmac_enable(np, 0, 1); - np = of_find_node_by_name(np, "ethernet"); - } - - /* Enable FW before PCI probe. Will be disabled later on - * Note: We should have a batter way to check that we are - * dealing with uninorth internal cell and not a PCI cell - * on the external PCI. The code below works though. - */ - np = of_find_node_by_name(NULL, "firewire"); - while(np) { - if (device_is_compatible(np, "pci106b,5811")) { - macio_chips[0].flags |= MACIO_FLAG_FW_SUPPORTED; - g5_fw_enable(np, 0, 1); - } - np = of_find_node_by_name(np, "firewire"); - } - } -} - -void __init -pmac_feature_init(void) -{ - /* Detect the UniNorth memory controller */ - probe_uninorth(); - - /* Probe mac-io controllers */ - if (probe_macios()) { - printk(KERN_WARNING "No mac-io chip found\n"); - return; - } - - /* Setup low-level i2c stuffs */ - pmac_init_low_i2c(); - - /* Probe machine type */ - if (probe_motherboard()) - printk(KERN_WARNING "Unknown PowerMac !\n"); - - /* Set some initial features (turn off some chips that will - * be later turned on) - */ - set_initial_features(); -} - -int __init pmac_feature_late_init(void) -{ -#if 0 - struct device_node* np; - - /* Request some resources late */ - if (uninorth_node) - request_OF_resource(uninorth_node, 0, NULL); - np = find_devices("hammerhead"); - if (np) - request_OF_resource(np, 0, NULL); - np = find_devices("interrupt-controller"); - if (np) - request_OF_resource(np, 0, NULL); -#endif - return 0; -} - -device_initcall(pmac_feature_late_init); - -#if 0 -static void dump_HT_speeds(char *name, u32 cfg, u32 frq) -{ - int freqs[16] = { 200,300,400,500,600,800,1000,0,0,0,0,0,0,0,0,0 }; - int bits[8] = { 8,16,0,32,2,4,0,0 }; - int freq = (frq >> 8) & 0xf; - - if (freqs[freq] == 0) - printk("%s: Unknown HT link frequency %x\n", name, freq); - else - printk("%s: %d MHz on main link, (%d in / %d out) bits width\n", - name, freqs[freq], - bits[(cfg >> 28) & 0x7], bits[(cfg >> 24) & 0x7]); -} -#endif - -void __init pmac_check_ht_link(void) -{ -#if 0 /* Disabled for now */ - u32 ufreq, freq, ucfg, cfg; - struct device_node *pcix_node; - struct pci_dn *pdn; - u8 px_bus, px_devfn; - struct pci_controller *px_hose; - - (void)in_be32(u3_ht + U3_HT_LINK_COMMAND); - ucfg = cfg = in_be32(u3_ht + U3_HT_LINK_CONFIG); - ufreq = freq = in_be32(u3_ht + U3_HT_LINK_FREQ); - dump_HT_speeds("U3 HyperTransport", cfg, freq); - - pcix_node = of_find_compatible_node(NULL, "pci", "pci-x"); - if (pcix_node == NULL) { - printk("No PCI-X bridge found\n"); - return; - } - pdn = pcix_node->data; - px_hose = pdn->phb; - px_bus = pdn->busno; - px_devfn = pdn->devfn; - - early_read_config_dword(px_hose, px_bus, px_devfn, 0xc4, &cfg); - early_read_config_dword(px_hose, px_bus, px_devfn, 0xcc, &freq); - dump_HT_speeds("PCI-X HT Uplink", cfg, freq); - early_read_config_dword(px_hose, px_bus, px_devfn, 0xc8, &cfg); - early_read_config_dword(px_hose, px_bus, px_devfn, 0xd0, &freq); - dump_HT_speeds("PCI-X HT Downlink", cfg, freq); -#endif -} - -/* - * Early video resume hook - */ - -static void (*pmac_early_vresume_proc)(void *data) __pmacdata; -static void *pmac_early_vresume_data __pmacdata; - -void pmac_set_early_video_resume(void (*proc)(void *data), void *data) -{ - if (_machine != _MACH_Pmac) - return; - preempt_disable(); - pmac_early_vresume_proc = proc; - pmac_early_vresume_data = data; - preempt_enable(); -} -EXPORT_SYMBOL(pmac_set_early_video_resume); - - -/* - * AGP related suspend/resume code - */ - -static struct pci_dev *pmac_agp_bridge __pmacdata; -static int (*pmac_agp_suspend)(struct pci_dev *bridge) __pmacdata; -static int (*pmac_agp_resume)(struct pci_dev *bridge) __pmacdata; - -void __pmac pmac_register_agp_pm(struct pci_dev *bridge, - int (*suspend)(struct pci_dev *bridge), - int (*resume)(struct pci_dev *bridge)) -{ - if (suspend || resume) { - pmac_agp_bridge = bridge; - pmac_agp_suspend = suspend; - pmac_agp_resume = resume; - return; - } - if (bridge != pmac_agp_bridge) - return; - pmac_agp_suspend = pmac_agp_resume = NULL; - return; -} -EXPORT_SYMBOL(pmac_register_agp_pm); - -void __pmac pmac_suspend_agp_for_card(struct pci_dev *dev) -{ - if (pmac_agp_bridge == NULL || pmac_agp_suspend == NULL) - return; - if (pmac_agp_bridge->bus != dev->bus) - return; - pmac_agp_suspend(pmac_agp_bridge); -} -EXPORT_SYMBOL(pmac_suspend_agp_for_card); - -void __pmac pmac_resume_agp_for_card(struct pci_dev *dev) -{ - if (pmac_agp_bridge == NULL || pmac_agp_resume == NULL) - return; - if (pmac_agp_bridge->bus != dev->bus) - return; - pmac_agp_resume(pmac_agp_bridge); -} -EXPORT_SYMBOL(pmac_resume_agp_for_card); diff --git a/arch/ppc64/kernel/pmac_low_i2c.c b/arch/ppc64/kernel/pmac_low_i2c.c deleted file mode 100644 index f3f39e8e337a..000000000000 --- a/arch/ppc64/kernel/pmac_low_i2c.c +++ /dev/null @@ -1,523 +0,0 @@ -/* - * arch/ppc/platforms/pmac_low_i2c.c - * - * Copyright (C) 2003 Ben. Herrenschmidt (benh@kernel.crashing.org) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * This file contains some low-level i2c access routines that - * need to be used by various bits of the PowerMac platform code - * at times where the real asynchronous & interrupt driven driver - * cannot be used. The API borrows some semantics from the darwin - * driver in order to ease the implementation of the platform - * properties parser - */ - -#undef DEBUG - -#include <linux/config.h> -#include <linux/types.h> -#include <linux/sched.h> -#include <linux/init.h> -#include <linux/module.h> -#include <linux/adb.h> -#include <linux/pmu.h> -#include <asm/keylargo.h> -#include <asm/uninorth.h> -#include <asm/io.h> -#include <asm/prom.h> -#include <asm/machdep.h> -#include <asm/pmac_low_i2c.h> - -#define MAX_LOW_I2C_HOST 4 - -#ifdef DEBUG -#define DBG(x...) do {\ - printk(KERN_DEBUG "KW:" x); \ - } while(0) -#else -#define DBG(x...) -#endif - -struct low_i2c_host; - -typedef int (*low_i2c_func_t)(struct low_i2c_host *host, u8 addr, u8 sub, u8 *data, int len); - -struct low_i2c_host -{ - struct device_node *np; /* OF device node */ - struct semaphore mutex; /* Access mutex for use by i2c-keywest */ - low_i2c_func_t func; /* Access function */ - unsigned int is_open : 1; /* Poor man's access control */ - int mode; /* Current mode */ - int channel; /* Current channel */ - int num_channels; /* Number of channels */ - void __iomem *base; /* For keywest-i2c, base address */ - int bsteps; /* And register stepping */ - int speed; /* And speed */ -}; - -static struct low_i2c_host low_i2c_hosts[MAX_LOW_I2C_HOST]; - -/* No locking is necessary on allocation, we are running way before - * anything can race with us - */ -static struct low_i2c_host *find_low_i2c_host(struct device_node *np) -{ - int i; - - for (i = 0; i < MAX_LOW_I2C_HOST; i++) - if (low_i2c_hosts[i].np == np) - return &low_i2c_hosts[i]; - return NULL; -} - -/* - * - * i2c-keywest implementation (UniNorth, U2, U3, Keylargo's) - * - */ - -/* - * Keywest i2c definitions borrowed from drivers/i2c/i2c-keywest.h, - * should be moved somewhere in include/asm-ppc/ - */ -/* Register indices */ -typedef enum { - reg_mode = 0, - reg_control, - reg_status, - reg_isr, - reg_ier, - reg_addr, - reg_subaddr, - reg_data -} reg_t; - - -/* Mode register */ -#define KW_I2C_MODE_100KHZ 0x00 -#define KW_I2C_MODE_50KHZ 0x01 -#define KW_I2C_MODE_25KHZ 0x02 -#define KW_I2C_MODE_DUMB 0x00 -#define KW_I2C_MODE_STANDARD 0x04 -#define KW_I2C_MODE_STANDARDSUB 0x08 -#define KW_I2C_MODE_COMBINED 0x0C -#define KW_I2C_MODE_MODE_MASK 0x0C -#define KW_I2C_MODE_CHAN_MASK 0xF0 - -/* Control register */ -#define KW_I2C_CTL_AAK 0x01 -#define KW_I2C_CTL_XADDR 0x02 -#define KW_I2C_CTL_STOP 0x04 -#define KW_I2C_CTL_START 0x08 - -/* Status register */ -#define KW_I2C_STAT_BUSY 0x01 -#define KW_I2C_STAT_LAST_AAK 0x02 -#define KW_I2C_STAT_LAST_RW 0x04 -#define KW_I2C_STAT_SDA 0x08 -#define KW_I2C_STAT_SCL 0x10 - -/* IER & ISR registers */ -#define KW_I2C_IRQ_DATA 0x01 -#define KW_I2C_IRQ_ADDR 0x02 -#define KW_I2C_IRQ_STOP 0x04 -#define KW_I2C_IRQ_START 0x08 -#define KW_I2C_IRQ_MASK 0x0F - -/* State machine states */ -enum { - state_idle, - state_addr, - state_read, - state_write, - state_stop, - state_dead -}; - -#define WRONG_STATE(name) do {\ - printk(KERN_DEBUG "KW: wrong state. Got %s, state: %s (isr: %02x)\n", \ - name, __kw_state_names[state], isr); \ - } while(0) - -static const char *__kw_state_names[] = { - "state_idle", - "state_addr", - "state_read", - "state_write", - "state_stop", - "state_dead" -}; - -static inline u8 __kw_read_reg(struct low_i2c_host *host, reg_t reg) -{ - return readb(host->base + (((unsigned int)reg) << host->bsteps)); -} - -static inline void __kw_write_reg(struct low_i2c_host *host, reg_t reg, u8 val) -{ - writeb(val, host->base + (((unsigned)reg) << host->bsteps)); - (void)__kw_read_reg(host, reg_subaddr); -} - -#define kw_write_reg(reg, val) __kw_write_reg(host, reg, val) -#define kw_read_reg(reg) __kw_read_reg(host, reg) - - -/* Don't schedule, the g5 fan controller is too - * timing sensitive - */ -static u8 kw_wait_interrupt(struct low_i2c_host* host) -{ - int i, j; - u8 isr; - - for (i = 0; i < 100000; i++) { - isr = kw_read_reg(reg_isr) & KW_I2C_IRQ_MASK; - if (isr != 0) - return isr; - - /* This code is used with the timebase frozen, we cannot rely - * on udelay ! For now, just use a bogus loop - */ - for (j = 1; j < 10000; j++) - mb(); - } - return isr; -} - -static int kw_handle_interrupt(struct low_i2c_host *host, int state, int rw, int *rc, u8 **data, int *len, u8 isr) -{ - u8 ack; - - DBG("kw_handle_interrupt(%s, isr: %x)\n", __kw_state_names[state], isr); - - if (isr == 0) { - if (state != state_stop) { - DBG("KW: Timeout !\n"); - *rc = -EIO; - goto stop; - } - if (state == state_stop) { - ack = kw_read_reg(reg_status); - if (!(ack & KW_I2C_STAT_BUSY)) { - state = state_idle; - kw_write_reg(reg_ier, 0x00); - } - } - return state; - } - - if (isr & KW_I2C_IRQ_ADDR) { - ack = kw_read_reg(reg_status); - if (state != state_addr) { - kw_write_reg(reg_isr, KW_I2C_IRQ_ADDR); - WRONG_STATE("KW_I2C_IRQ_ADDR"); - *rc = -EIO; - goto stop; - } - if ((ack & KW_I2C_STAT_LAST_AAK) == 0) { - *rc = -ENODEV; - DBG("KW: NAK on address\n"); - return state_stop; - } else { - if (rw) { - state = state_read; - if (*len > 1) - kw_write_reg(reg_control, KW_I2C_CTL_AAK); - } else { - state = state_write; - kw_write_reg(reg_data, **data); - (*data)++; (*len)--; - } - } - kw_write_reg(reg_isr, KW_I2C_IRQ_ADDR); - } - - if (isr & KW_I2C_IRQ_DATA) { - if (state == state_read) { - **data = kw_read_reg(reg_data); - (*data)++; (*len)--; - kw_write_reg(reg_isr, KW_I2C_IRQ_DATA); - if ((*len) == 0) - state = state_stop; - else if ((*len) == 1) - kw_write_reg(reg_control, 0); - } else if (state == state_write) { - ack = kw_read_reg(reg_status); - if ((ack & KW_I2C_STAT_LAST_AAK) == 0) { - DBG("KW: nack on data write\n"); - *rc = -EIO; - goto stop; - } else if (*len) { - kw_write_reg(reg_data, **data); - (*data)++; (*len)--; - } else { - kw_write_reg(reg_control, KW_I2C_CTL_STOP); - state = state_stop; - *rc = 0; - } - kw_write_reg(reg_isr, KW_I2C_IRQ_DATA); - } else { - kw_write_reg(reg_isr, KW_I2C_IRQ_DATA); - WRONG_STATE("KW_I2C_IRQ_DATA"); - if (state != state_stop) { - *rc = -EIO; - goto stop; - } - } - } - - if (isr & KW_I2C_IRQ_STOP) { - kw_write_reg(reg_isr, KW_I2C_IRQ_STOP); - if (state != state_stop) { - WRONG_STATE("KW_I2C_IRQ_STOP"); - *rc = -EIO; - } - return state_idle; - } - - if (isr & KW_I2C_IRQ_START) - kw_write_reg(reg_isr, KW_I2C_IRQ_START); - - return state; - - stop: - kw_write_reg(reg_control, KW_I2C_CTL_STOP); - return state_stop; -} - -static int keywest_low_i2c_func(struct low_i2c_host *host, u8 addr, u8 subaddr, u8 *data, int len) -{ - u8 mode_reg = host->speed; - int state = state_addr; - int rc = 0; - - /* Setup mode & subaddress if any */ - switch(host->mode) { - case pmac_low_i2c_mode_dumb: - printk(KERN_ERR "low_i2c: Dumb mode not supported !\n"); - return -EINVAL; - case pmac_low_i2c_mode_std: - mode_reg |= KW_I2C_MODE_STANDARD; - break; - case pmac_low_i2c_mode_stdsub: - mode_reg |= KW_I2C_MODE_STANDARDSUB; - break; - case pmac_low_i2c_mode_combined: - mode_reg |= KW_I2C_MODE_COMBINED; - break; - } - - /* Setup channel & clear pending irqs */ - kw_write_reg(reg_isr, kw_read_reg(reg_isr)); - kw_write_reg(reg_mode, mode_reg | (host->channel << 4)); - kw_write_reg(reg_status, 0); - - /* Set up address and r/w bit */ - kw_write_reg(reg_addr, addr); - - /* Set up the sub address */ - if ((mode_reg & KW_I2C_MODE_MODE_MASK) == KW_I2C_MODE_STANDARDSUB - || (mode_reg & KW_I2C_MODE_MODE_MASK) == KW_I2C_MODE_COMBINED) - kw_write_reg(reg_subaddr, subaddr); - - /* Start sending address & disable interrupt*/ - kw_write_reg(reg_ier, 0 /*KW_I2C_IRQ_MASK*/); - kw_write_reg(reg_control, KW_I2C_CTL_XADDR); - - /* State machine, to turn into an interrupt handler */ - while(state != state_idle) { - u8 isr = kw_wait_interrupt(host); - state = kw_handle_interrupt(host, state, addr & 1, &rc, &data, &len, isr); - } - - return rc; -} - -static void keywest_low_i2c_add(struct device_node *np) -{ - struct low_i2c_host *host = find_low_i2c_host(NULL); - u32 *psteps, *prate, steps, aoffset = 0; - struct device_node *parent; - - if (host == NULL) { - printk(KERN_ERR "low_i2c: Can't allocate host for %s\n", - np->full_name); - return; - } - memset(host, 0, sizeof(*host)); - - init_MUTEX(&host->mutex); - host->np = of_node_get(np); - psteps = (u32 *)get_property(np, "AAPL,address-step", NULL); - steps = psteps ? (*psteps) : 0x10; - for (host->bsteps = 0; (steps & 0x01) == 0; host->bsteps++) - steps >>= 1; - parent = of_get_parent(np); - host->num_channels = 1; - if (parent && parent->name[0] == 'u') { - host->num_channels = 2; - aoffset = 3; - } - /* Select interface rate */ - host->speed = KW_I2C_MODE_100KHZ; - prate = (u32 *)get_property(np, "AAPL,i2c-rate", NULL); - if (prate) switch(*prate) { - case 100: - host->speed = KW_I2C_MODE_100KHZ; - break; - case 50: - host->speed = KW_I2C_MODE_50KHZ; - break; - case 25: - host->speed = KW_I2C_MODE_25KHZ; - break; - } - - host->mode = pmac_low_i2c_mode_std; - host->base = ioremap(np->addrs[0].address + aoffset, - np->addrs[0].size); - host->func = keywest_low_i2c_func; -} - -/* - * - * PMU implementation - * - */ - - -#ifdef CONFIG_ADB_PMU - -static int pmu_low_i2c_func(struct low_i2c_host *host, u8 addr, u8 sub, u8 *data, int len) -{ - // TODO - return -ENODEV; -} - -static void pmu_low_i2c_add(struct device_node *np) -{ - struct low_i2c_host *host = find_low_i2c_host(NULL); - - if (host == NULL) { - printk(KERN_ERR "low_i2c: Can't allocate host for %s\n", - np->full_name); - return; - } - memset(host, 0, sizeof(*host)); - - init_MUTEX(&host->mutex); - host->np = of_node_get(np); - host->num_channels = 3; - host->mode = pmac_low_i2c_mode_std; - host->func = pmu_low_i2c_func; -} - -#endif /* CONFIG_ADB_PMU */ - -void __init pmac_init_low_i2c(void) -{ - struct device_node *np; - - /* Probe keywest-i2c busses */ - np = of_find_compatible_node(NULL, "i2c", "keywest-i2c"); - while(np) { - keywest_low_i2c_add(np); - np = of_find_compatible_node(np, "i2c", "keywest-i2c"); - } - -#ifdef CONFIG_ADB_PMU - /* Probe PMU busses */ - np = of_find_node_by_name(NULL, "via-pmu"); - if (np) - pmu_low_i2c_add(np); -#endif /* CONFIG_ADB_PMU */ - - /* TODO: Add CUDA support as well */ -} - -int pmac_low_i2c_lock(struct device_node *np) -{ - struct low_i2c_host *host = find_low_i2c_host(np); - - if (!host) - return -ENODEV; - down(&host->mutex); - return 0; -} -EXPORT_SYMBOL(pmac_low_i2c_lock); - -int pmac_low_i2c_unlock(struct device_node *np) -{ - struct low_i2c_host *host = find_low_i2c_host(np); - - if (!host) - return -ENODEV; - up(&host->mutex); - return 0; -} -EXPORT_SYMBOL(pmac_low_i2c_unlock); - - -int pmac_low_i2c_open(struct device_node *np, int channel) -{ - struct low_i2c_host *host = find_low_i2c_host(np); - - if (!host) - return -ENODEV; - - if (channel >= host->num_channels) - return -EINVAL; - - down(&host->mutex); - host->is_open = 1; - host->channel = channel; - - return 0; -} -EXPORT_SYMBOL(pmac_low_i2c_open); - -int pmac_low_i2c_close(struct device_node *np) -{ - struct low_i2c_host *host = find_low_i2c_host(np); - - if (!host) - return -ENODEV; - - host->is_open = 0; - up(&host->mutex); - - return 0; -} -EXPORT_SYMBOL(pmac_low_i2c_close); - -int pmac_low_i2c_setmode(struct device_node *np, int mode) -{ - struct low_i2c_host *host = find_low_i2c_host(np); - - if (!host) - return -ENODEV; - WARN_ON(!host->is_open); - host->mode = mode; - - return 0; -} -EXPORT_SYMBOL(pmac_low_i2c_setmode); - -int pmac_low_i2c_xfer(struct device_node *np, u8 addrdir, u8 subaddr, u8 *data, int len) -{ - struct low_i2c_host *host = find_low_i2c_host(np); - - if (!host) - return -ENODEV; - WARN_ON(!host->is_open); - - return host->func(host, addrdir, subaddr, data, len); -} -EXPORT_SYMBOL(pmac_low_i2c_xfer); - diff --git a/arch/ppc64/kernel/pmac_nvram.c b/arch/ppc64/kernel/pmac_nvram.c deleted file mode 100644 index e32a902236e3..000000000000 --- a/arch/ppc64/kernel/pmac_nvram.c +++ /dev/null @@ -1,495 +0,0 @@ -/* - * arch/ppc/platforms/pmac_nvram.c - * - * Copyright (C) 2002 Benjamin Herrenschmidt (benh@kernel.crashing.org) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * Todo: - add support for the OF persistent properties - */ -#include <linux/config.h> -#include <linux/module.h> -#include <linux/kernel.h> -#include <linux/stddef.h> -#include <linux/string.h> -#include <linux/init.h> -#include <linux/slab.h> -#include <linux/delay.h> -#include <linux/errno.h> -#include <linux/bootmem.h> -#include <linux/completion.h> -#include <linux/spinlock.h> -#include <asm/sections.h> -#include <asm/io.h> -#include <asm/system.h> -#include <asm/prom.h> -#include <asm/machdep.h> -#include <asm/nvram.h> - -#define DEBUG - -#ifdef DEBUG -#define DBG(x...) printk(x) -#else -#define DBG(x...) -#endif - -#define NVRAM_SIZE 0x2000 /* 8kB of non-volatile RAM */ - -#define CORE99_SIGNATURE 0x5a -#define CORE99_ADLER_START 0x14 - -/* On Core99, nvram is either a sharp, a micron or an AMD flash */ -#define SM_FLASH_STATUS_DONE 0x80 -#define SM_FLASH_STATUS_ERR 0x38 - -#define SM_FLASH_CMD_ERASE_CONFIRM 0xd0 -#define SM_FLASH_CMD_ERASE_SETUP 0x20 -#define SM_FLASH_CMD_RESET 0xff -#define SM_FLASH_CMD_WRITE_SETUP 0x40 -#define SM_FLASH_CMD_CLEAR_STATUS 0x50 -#define SM_FLASH_CMD_READ_STATUS 0x70 - -/* CHRP NVRAM header */ -struct chrp_header { - u8 signature; - u8 cksum; - u16 len; - char name[12]; - u8 data[0]; -}; - -struct core99_header { - struct chrp_header hdr; - u32 adler; - u32 generation; - u32 reserved[2]; -}; - -/* - * Read and write the non-volatile RAM on PowerMacs and CHRP machines. - */ -static volatile unsigned char *nvram_data; -static int core99_bank = 0; -// XXX Turn that into a sem -static DEFINE_SPINLOCK(nv_lock); - -extern int system_running; - -static int (*core99_write_bank)(int bank, u8* datas); -static int (*core99_erase_bank)(int bank); - -static char *nvram_image __pmacdata; - - -static ssize_t __pmac core99_nvram_read(char *buf, size_t count, loff_t *index) -{ - int i; - - if (nvram_image == NULL) - return -ENODEV; - if (*index > NVRAM_SIZE) - return 0; - - i = *index; - if (i + count > NVRAM_SIZE) - count = NVRAM_SIZE - i; - - memcpy(buf, &nvram_image[i], count); - *index = i + count; - return count; -} - -static ssize_t __pmac core99_nvram_write(char *buf, size_t count, loff_t *index) -{ - int i; - - if (nvram_image == NULL) - return -ENODEV; - if (*index > NVRAM_SIZE) - return 0; - - i = *index; - if (i + count > NVRAM_SIZE) - count = NVRAM_SIZE - i; - - memcpy(&nvram_image[i], buf, count); - *index = i + count; - return count; -} - -static ssize_t __pmac core99_nvram_size(void) -{ - if (nvram_image == NULL) - return -ENODEV; - return NVRAM_SIZE; -} - -static u8 __pmac chrp_checksum(struct chrp_header* hdr) -{ - u8 *ptr; - u16 sum = hdr->signature; - for (ptr = (u8 *)&hdr->len; ptr < hdr->data; ptr++) - sum += *ptr; - while (sum > 0xFF) - sum = (sum & 0xFF) + (sum>>8); - return sum; -} - -static u32 __pmac core99_calc_adler(u8 *buffer) -{ - int cnt; - u32 low, high; - - buffer += CORE99_ADLER_START; - low = 1; - high = 0; - for (cnt=0; cnt<(NVRAM_SIZE-CORE99_ADLER_START); cnt++) { - if ((cnt % 5000) == 0) { - high %= 65521UL; - high %= 65521UL; - } - low += buffer[cnt]; - high += low; - } - low %= 65521UL; - high %= 65521UL; - - return (high << 16) | low; -} - -static u32 __pmac core99_check(u8* datas) -{ - struct core99_header* hdr99 = (struct core99_header*)datas; - - if (hdr99->hdr.signature != CORE99_SIGNATURE) { - DBG("Invalid signature\n"); - return 0; - } - if (hdr99->hdr.cksum != chrp_checksum(&hdr99->hdr)) { - DBG("Invalid checksum\n"); - return 0; - } - if (hdr99->adler != core99_calc_adler(datas)) { - DBG("Invalid adler\n"); - return 0; - } - return hdr99->generation; -} - -static int __pmac sm_erase_bank(int bank) -{ - int stat, i; - unsigned long timeout; - - u8* base = (u8 *)nvram_data + core99_bank*NVRAM_SIZE; - - DBG("nvram: Sharp/Micron Erasing bank %d...\n", bank); - - out_8(base, SM_FLASH_CMD_ERASE_SETUP); - out_8(base, SM_FLASH_CMD_ERASE_CONFIRM); - timeout = 0; - do { - if (++timeout > 1000000) { - printk(KERN_ERR "nvram: Sharp/Miron flash erase timeout !\n"); - break; - } - out_8(base, SM_FLASH_CMD_READ_STATUS); - stat = in_8(base); - } while (!(stat & SM_FLASH_STATUS_DONE)); - - out_8(base, SM_FLASH_CMD_CLEAR_STATUS); - out_8(base, SM_FLASH_CMD_RESET); - - for (i=0; i<NVRAM_SIZE; i++) - if (base[i] != 0xff) { - printk(KERN_ERR "nvram: Sharp/Micron flash erase failed !\n"); - return -ENXIO; - } - return 0; -} - -static int __pmac sm_write_bank(int bank, u8* datas) -{ - int i, stat = 0; - unsigned long timeout; - - u8* base = (u8 *)nvram_data + core99_bank*NVRAM_SIZE; - - DBG("nvram: Sharp/Micron Writing bank %d...\n", bank); - - for (i=0; i<NVRAM_SIZE; i++) { - out_8(base+i, SM_FLASH_CMD_WRITE_SETUP); - udelay(1); - out_8(base+i, datas[i]); - timeout = 0; - do { - if (++timeout > 1000000) { - printk(KERN_ERR "nvram: Sharp/Micron flash write timeout !\n"); - break; - } - out_8(base, SM_FLASH_CMD_READ_STATUS); - stat = in_8(base); - } while (!(stat & SM_FLASH_STATUS_DONE)); - if (!(stat & SM_FLASH_STATUS_DONE)) - break; - } - out_8(base, SM_FLASH_CMD_CLEAR_STATUS); - out_8(base, SM_FLASH_CMD_RESET); - for (i=0; i<NVRAM_SIZE; i++) - if (base[i] != datas[i]) { - printk(KERN_ERR "nvram: Sharp/Micron flash write failed !\n"); - return -ENXIO; - } - return 0; -} - -static int __pmac amd_erase_bank(int bank) -{ - int i, stat = 0; - unsigned long timeout; - - u8* base = (u8 *)nvram_data + core99_bank*NVRAM_SIZE; - - DBG("nvram: AMD Erasing bank %d...\n", bank); - - /* Unlock 1 */ - out_8(base+0x555, 0xaa); - udelay(1); - /* Unlock 2 */ - out_8(base+0x2aa, 0x55); - udelay(1); - - /* Sector-Erase */ - out_8(base+0x555, 0x80); - udelay(1); - out_8(base+0x555, 0xaa); - udelay(1); - out_8(base+0x2aa, 0x55); - udelay(1); - out_8(base, 0x30); - udelay(1); - - timeout = 0; - do { - if (++timeout > 1000000) { - printk(KERN_ERR "nvram: AMD flash erase timeout !\n"); - break; - } - stat = in_8(base) ^ in_8(base); - } while (stat != 0); - - /* Reset */ - out_8(base, 0xf0); - udelay(1); - - for (i=0; i<NVRAM_SIZE; i++) - if (base[i] != 0xff) { - printk(KERN_ERR "nvram: AMD flash erase failed !\n"); - return -ENXIO; - } - return 0; -} - -static int __pmac amd_write_bank(int bank, u8* datas) -{ - int i, stat = 0; - unsigned long timeout; - - u8* base = (u8 *)nvram_data + core99_bank*NVRAM_SIZE; - - DBG("nvram: AMD Writing bank %d...\n", bank); - - for (i=0; i<NVRAM_SIZE; i++) { - /* Unlock 1 */ - out_8(base+0x555, 0xaa); - udelay(1); - /* Unlock 2 */ - out_8(base+0x2aa, 0x55); - udelay(1); - - /* Write single word */ - out_8(base+0x555, 0xa0); - udelay(1); - out_8(base+i, datas[i]); - - timeout = 0; - do { - if (++timeout > 1000000) { - printk(KERN_ERR "nvram: AMD flash write timeout !\n"); - break; - } - stat = in_8(base) ^ in_8(base); - } while (stat != 0); - if (stat != 0) - break; - } - - /* Reset */ - out_8(base, 0xf0); - udelay(1); - - for (i=0; i<NVRAM_SIZE; i++) - if (base[i] != datas[i]) { - printk(KERN_ERR "nvram: AMD flash write failed !\n"); - return -ENXIO; - } - return 0; -} - - -static int __pmac core99_nvram_sync(void) -{ - struct core99_header* hdr99; - unsigned long flags; - - spin_lock_irqsave(&nv_lock, flags); - if (!memcmp(nvram_image, (u8*)nvram_data + core99_bank*NVRAM_SIZE, - NVRAM_SIZE)) - goto bail; - - DBG("Updating nvram...\n"); - - hdr99 = (struct core99_header*)nvram_image; - hdr99->generation++; - hdr99->hdr.signature = CORE99_SIGNATURE; - hdr99->hdr.cksum = chrp_checksum(&hdr99->hdr); - hdr99->adler = core99_calc_adler(nvram_image); - core99_bank = core99_bank ? 0 : 1; - if (core99_erase_bank) - if (core99_erase_bank(core99_bank)) { - printk("nvram: Error erasing bank %d\n", core99_bank); - goto bail; - } - if (core99_write_bank) - if (core99_write_bank(core99_bank, nvram_image)) - printk("nvram: Error writing bank %d\n", core99_bank); - bail: - spin_unlock_irqrestore(&nv_lock, flags); - - return 0; -} - -int __init pmac_nvram_init(void) -{ - struct device_node *dp; - u32 gen_bank0, gen_bank1; - int i; - - dp = find_devices("nvram"); - if (dp == NULL) { - printk(KERN_ERR "Can't find NVRAM device\n"); - return -ENODEV; - } - if (!device_is_compatible(dp, "nvram,flash")) { - printk(KERN_ERR "Incompatible type of NVRAM\n"); - return -ENXIO; - } - - nvram_image = alloc_bootmem(NVRAM_SIZE); - if (nvram_image == NULL) { - printk(KERN_ERR "nvram: can't allocate ram image\n"); - return -ENOMEM; - } - nvram_data = ioremap(dp->addrs[0].address, NVRAM_SIZE*2); - - DBG("nvram: Checking bank 0...\n"); - - gen_bank0 = core99_check((u8 *)nvram_data); - gen_bank1 = core99_check((u8 *)nvram_data + NVRAM_SIZE); - core99_bank = (gen_bank0 < gen_bank1) ? 1 : 0; - - DBG("nvram: gen0=%d, gen1=%d\n", gen_bank0, gen_bank1); - DBG("nvram: Active bank is: %d\n", core99_bank); - - for (i=0; i<NVRAM_SIZE; i++) - nvram_image[i] = nvram_data[i + core99_bank*NVRAM_SIZE]; - - ppc_md.nvram_read = core99_nvram_read; - ppc_md.nvram_write = core99_nvram_write; - ppc_md.nvram_size = core99_nvram_size; - ppc_md.nvram_sync = core99_nvram_sync; - - /* - * Maybe we could be smarter here though making an exclusive list - * of known flash chips is a bit nasty as older OF didn't provide us - * with a useful "compatible" entry. A solution would be to really - * identify the chip using flash id commands and base ourselves on - * a list of known chips IDs - */ - if (device_is_compatible(dp, "amd-0137")) { - core99_erase_bank = amd_erase_bank; - core99_write_bank = amd_write_bank; - } else { - core99_erase_bank = sm_erase_bank; - core99_write_bank = sm_write_bank; - } - - return 0; -} - -int __pmac pmac_get_partition(int partition) -{ - struct nvram_partition *part; - const char *name; - int sig; - - switch(partition) { - case pmac_nvram_OF: - name = "common"; - sig = NVRAM_SIG_SYS; - break; - case pmac_nvram_XPRAM: - name = "APL,MacOS75"; - sig = NVRAM_SIG_OS; - break; - case pmac_nvram_NR: - default: - /* Oldworld stuff */ - return -ENODEV; - } - - part = nvram_find_partition(sig, name); - if (part == NULL) - return 0; - - return part->index; -} - -u8 __pmac pmac_xpram_read(int xpaddr) -{ - int offset = pmac_get_partition(pmac_nvram_XPRAM); - loff_t index; - u8 buf; - ssize_t count; - - if (offset < 0 || xpaddr < 0 || xpaddr > 0x100) - return 0xff; - index = offset + xpaddr; - - count = ppc_md.nvram_read(&buf, 1, &index); - if (count != 1) - return 0xff; - return buf; -} - -void __pmac pmac_xpram_write(int xpaddr, u8 data) -{ - int offset = pmac_get_partition(pmac_nvram_XPRAM); - loff_t index; - u8 buf; - - if (offset < 0 || xpaddr < 0 || xpaddr > 0x100) - return; - index = offset + xpaddr; - buf = data; - - ppc_md.nvram_write(&buf, 1, &index); -} - -EXPORT_SYMBOL(pmac_get_partition); -EXPORT_SYMBOL(pmac_xpram_read); -EXPORT_SYMBOL(pmac_xpram_write); diff --git a/arch/ppc64/kernel/pmac_pci.c b/arch/ppc64/kernel/pmac_pci.c deleted file mode 100644 index dc40a0cad0b4..000000000000 --- a/arch/ppc64/kernel/pmac_pci.c +++ /dev/null @@ -1,793 +0,0 @@ -/* - * Support for PCI bridges found on Power Macintoshes. - * At present the "bandit" and "chaos" bridges are supported. - * Fortunately you access configuration space in the same - * way with either bridge. - * - * Copyright (C) 2003 Benjamin Herrenschmuidt (benh@kernel.crashing.org) - * Copyright (C) 1997 Paul Mackerras (paulus@samba.org) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/kernel.h> -#include <linux/pci.h> -#include <linux/delay.h> -#include <linux/string.h> -#include <linux/init.h> -#include <linux/bootmem.h> - -#include <asm/sections.h> -#include <asm/io.h> -#include <asm/prom.h> -#include <asm/pci-bridge.h> -#include <asm/machdep.h> -#include <asm/pmac_feature.h> -#include <asm/iommu.h> - -#include "pci.h" -#include "pmac.h" - -#define DEBUG - -#ifdef DEBUG -#define DBG(x...) printk(x) -#else -#define DBG(x...) -#endif - -/* XXX Could be per-controller, but I don't think we risk anything by - * assuming we won't have both UniNorth and Bandit */ -static int has_uninorth; -static struct pci_controller *u3_agp; -struct device_node *k2_skiplist[2]; - -static int __init fixup_one_level_bus_range(struct device_node *node, int higher) -{ - for (; node != 0;node = node->sibling) { - int * bus_range; - unsigned int *class_code; - int len; - - /* For PCI<->PCI bridges or CardBus bridges, we go down */ - class_code = (unsigned int *) get_property(node, "class-code", NULL); - if (!class_code || ((*class_code >> 8) != PCI_CLASS_BRIDGE_PCI && - (*class_code >> 8) != PCI_CLASS_BRIDGE_CARDBUS)) - continue; - bus_range = (int *) get_property(node, "bus-range", &len); - if (bus_range != NULL && len > 2 * sizeof(int)) { - if (bus_range[1] > higher) - higher = bus_range[1]; - } - higher = fixup_one_level_bus_range(node->child, higher); - } - return higher; -} - -/* This routine fixes the "bus-range" property of all bridges in the - * system since they tend to have their "last" member wrong on macs - * - * Note that the bus numbers manipulated here are OF bus numbers, they - * are not Linux bus numbers. - */ -static void __init fixup_bus_range(struct device_node *bridge) -{ - int * bus_range; - int len; - - /* Lookup the "bus-range" property for the hose */ - bus_range = (int *) get_property(bridge, "bus-range", &len); - if (bus_range == NULL || len < 2 * sizeof(int)) { - printk(KERN_WARNING "Can't get bus-range for %s\n", - bridge->full_name); - return; - } - bus_range[1] = fixup_one_level_bus_range(bridge->child, bus_range[1]); -} - -/* - * Apple MacRISC (U3, UniNorth, Bandit, Chaos) PCI controllers. - * - * The "Bandit" version is present in all early PCI PowerMacs, - * and up to the first ones using Grackle. Some machines may - * have 2 bandit controllers (2 PCI busses). - * - * "Chaos" is used in some "Bandit"-type machines as a bridge - * for the separate display bus. It is accessed the same - * way as bandit, but cannot be probed for devices. It therefore - * has its own config access functions. - * - * The "UniNorth" version is present in all Core99 machines - * (iBook, G4, new IMacs, and all the recent Apple machines). - * It contains 3 controllers in one ASIC. - * - * The U3 is the bridge used on G5 machines. It contains on - * AGP bus which is dealt with the old UniNorth access routines - * and an HyperTransport bus which uses its own set of access - * functions. - */ - -#define MACRISC_CFA0(devfn, off) \ - ((1 << (unsigned long)PCI_SLOT(dev_fn)) \ - | (((unsigned long)PCI_FUNC(dev_fn)) << 8) \ - | (((unsigned long)(off)) & 0xFCUL)) - -#define MACRISC_CFA1(bus, devfn, off) \ - ((((unsigned long)(bus)) << 16) \ - |(((unsigned long)(devfn)) << 8) \ - |(((unsigned long)(off)) & 0xFCUL) \ - |1UL) - -static unsigned long __pmac macrisc_cfg_access(struct pci_controller* hose, - u8 bus, u8 dev_fn, u8 offset) -{ - unsigned int caddr; - - if (bus == hose->first_busno) { - if (dev_fn < (11 << 3)) - return 0; - caddr = MACRISC_CFA0(dev_fn, offset); - } else - caddr = MACRISC_CFA1(bus, dev_fn, offset); - - /* Uninorth will return garbage if we don't read back the value ! */ - do { - out_le32(hose->cfg_addr, caddr); - } while (in_le32(hose->cfg_addr) != caddr); - - offset &= has_uninorth ? 0x07 : 0x03; - return ((unsigned long)hose->cfg_data) + offset; -} - -static int __pmac macrisc_read_config(struct pci_bus *bus, unsigned int devfn, - int offset, int len, u32 *val) -{ - struct pci_controller *hose; - unsigned long addr; - - hose = pci_bus_to_host(bus); - if (hose == NULL) - return PCIBIOS_DEVICE_NOT_FOUND; - - addr = macrisc_cfg_access(hose, bus->number, devfn, offset); - if (!addr) - return PCIBIOS_DEVICE_NOT_FOUND; - /* - * Note: the caller has already checked that offset is - * suitably aligned and that len is 1, 2 or 4. - */ - switch (len) { - case 1: - *val = in_8((u8 *)addr); - break; - case 2: - *val = in_le16((u16 *)addr); - break; - default: - *val = in_le32((u32 *)addr); - break; - } - return PCIBIOS_SUCCESSFUL; -} - -static int __pmac macrisc_write_config(struct pci_bus *bus, unsigned int devfn, - int offset, int len, u32 val) -{ - struct pci_controller *hose; - unsigned long addr; - - hose = pci_bus_to_host(bus); - if (hose == NULL) - return PCIBIOS_DEVICE_NOT_FOUND; - - addr = macrisc_cfg_access(hose, bus->number, devfn, offset); - if (!addr) - return PCIBIOS_DEVICE_NOT_FOUND; - /* - * Note: the caller has already checked that offset is - * suitably aligned and that len is 1, 2 or 4. - */ - switch (len) { - case 1: - out_8((u8 *)addr, val); - (void) in_8((u8 *)addr); - break; - case 2: - out_le16((u16 *)addr, val); - (void) in_le16((u16 *)addr); - break; - default: - out_le32((u32 *)addr, val); - (void) in_le32((u32 *)addr); - break; - } - return PCIBIOS_SUCCESSFUL; -} - -static struct pci_ops macrisc_pci_ops = -{ - macrisc_read_config, - macrisc_write_config -}; - -/* - * These versions of U3 HyperTransport config space access ops do not - * implement self-view of the HT host yet - */ - -/* - * This function deals with some "special cases" devices. - * - * 0 -> No special case - * 1 -> Skip the device but act as if the access was successfull - * (return 0xff's on reads, eventually, cache config space - * accesses in a later version) - * -1 -> Hide the device (unsuccessful acess) - */ -static int u3_ht_skip_device(struct pci_controller *hose, - struct pci_bus *bus, unsigned int devfn) -{ - struct device_node *busdn, *dn; - int i; - - /* We only allow config cycles to devices that are in OF device-tree - * as we are apparently having some weird things going on with some - * revs of K2 on recent G5s - */ - if (bus->self) - busdn = pci_device_to_OF_node(bus->self); - else - busdn = hose->arch_data; - for (dn = busdn->child; dn; dn = dn->sibling) - if (dn->data && PCI_DN(dn)->devfn == devfn) - break; - if (dn == NULL) - return -1; - - /* - * When a device in K2 is powered down, we die on config - * cycle accesses. Fix that here. - */ - for (i=0; i<2; i++) - if (k2_skiplist[i] == dn) - return 1; - - return 0; -} - -#define U3_HT_CFA0(devfn, off) \ - ((((unsigned long)devfn) << 8) | offset) -#define U3_HT_CFA1(bus, devfn, off) \ - (U3_HT_CFA0(devfn, off) \ - + (((unsigned long)bus) << 16) \ - + 0x01000000UL) - -static unsigned long __pmac u3_ht_cfg_access(struct pci_controller* hose, - u8 bus, u8 devfn, u8 offset) -{ - if (bus == hose->first_busno) { - /* For now, we don't self probe U3 HT bridge */ - if (PCI_SLOT(devfn) == 0) - return 0; - return ((unsigned long)hose->cfg_data) + U3_HT_CFA0(devfn, offset); - } else - return ((unsigned long)hose->cfg_data) + U3_HT_CFA1(bus, devfn, offset); -} - -static int __pmac u3_ht_read_config(struct pci_bus *bus, unsigned int devfn, - int offset, int len, u32 *val) -{ - struct pci_controller *hose; - unsigned long addr; - - - hose = pci_bus_to_host(bus); - if (hose == NULL) - return PCIBIOS_DEVICE_NOT_FOUND; - - addr = u3_ht_cfg_access(hose, bus->number, devfn, offset); - if (!addr) - return PCIBIOS_DEVICE_NOT_FOUND; - - switch (u3_ht_skip_device(hose, bus, devfn)) { - case 0: - break; - case 1: - switch (len) { - case 1: - *val = 0xff; break; - case 2: - *val = 0xffff; break; - default: - *val = 0xfffffffful; break; - } - return PCIBIOS_SUCCESSFUL; - default: - return PCIBIOS_DEVICE_NOT_FOUND; - } - - /* - * Note: the caller has already checked that offset is - * suitably aligned and that len is 1, 2 or 4. - */ - switch (len) { - case 1: - *val = in_8((u8 *)addr); - break; - case 2: - *val = in_le16((u16 *)addr); - break; - default: - *val = in_le32((u32 *)addr); - break; - } - return PCIBIOS_SUCCESSFUL; -} - -static int __pmac u3_ht_write_config(struct pci_bus *bus, unsigned int devfn, - int offset, int len, u32 val) -{ - struct pci_controller *hose; - unsigned long addr; - - hose = pci_bus_to_host(bus); - if (hose == NULL) - return PCIBIOS_DEVICE_NOT_FOUND; - - addr = u3_ht_cfg_access(hose, bus->number, devfn, offset); - if (!addr) - return PCIBIOS_DEVICE_NOT_FOUND; - - switch (u3_ht_skip_device(hose, bus, devfn)) { - case 0: - break; - case 1: - return PCIBIOS_SUCCESSFUL; - default: - return PCIBIOS_DEVICE_NOT_FOUND; - } - - /* - * Note: the caller has already checked that offset is - * suitably aligned and that len is 1, 2 or 4. - */ - switch (len) { - case 1: - out_8((u8 *)addr, val); - (void) in_8((u8 *)addr); - break; - case 2: - out_le16((u16 *)addr, val); - (void) in_le16((u16 *)addr); - break; - default: - out_le32((u32 *)addr, val); - (void) in_le32((u32 *)addr); - break; - } - return PCIBIOS_SUCCESSFUL; -} - -static struct pci_ops u3_ht_pci_ops = -{ - u3_ht_read_config, - u3_ht_write_config -}; - -static void __init setup_u3_agp(struct pci_controller* hose) -{ - /* On G5, we move AGP up to high bus number so we don't need - * to reassign bus numbers for HT. If we ever have P2P bridges - * on AGP, we'll have to move pci_assign_all_busses to the - * pci_controller structure so we enable it for AGP and not for - * HT childs. - * We hard code the address because of the different size of - * the reg address cell, we shall fix that by killing struct - * reg_property and using some accessor functions instead - */ - hose->first_busno = 0xf0; - hose->last_busno = 0xff; - has_uninorth = 1; - hose->ops = ¯isc_pci_ops; - hose->cfg_addr = ioremap(0xf0000000 + 0x800000, 0x1000); - hose->cfg_data = ioremap(0xf0000000 + 0xc00000, 0x1000); - - u3_agp = hose; -} - -static void __init setup_u3_ht(struct pci_controller* hose) -{ - struct device_node *np = (struct device_node *)hose->arch_data; - int i, cur; - - hose->ops = &u3_ht_pci_ops; - - /* We hard code the address because of the different size of - * the reg address cell, we shall fix that by killing struct - * reg_property and using some accessor functions instead - */ - hose->cfg_data = (volatile unsigned char *)ioremap(0xf2000000, 0x02000000); - - /* - * /ht node doesn't expose a "ranges" property, so we "remove" regions that - * have been allocated to AGP. So far, this version of the code doesn't assign - * any of the 0xfxxxxxxx "fine" memory regions to /ht. - * We need to fix that sooner or later by either parsing all child "ranges" - * properties or figuring out the U3 address space decoding logic and - * then read it's configuration register (if any). - */ - hose->io_base_phys = 0xf4000000; - hose->io_base_virt = ioremap(hose->io_base_phys, 0x00400000); - isa_io_base = pci_io_base = (unsigned long) hose->io_base_virt; - hose->io_resource.name = np->full_name; - hose->io_resource.start = 0; - hose->io_resource.end = 0x003fffff; - hose->io_resource.flags = IORESOURCE_IO; - hose->pci_mem_offset = 0; - hose->first_busno = 0; - hose->last_busno = 0xef; - hose->mem_resources[0].name = np->full_name; - hose->mem_resources[0].start = 0x80000000; - hose->mem_resources[0].end = 0xefffffff; - hose->mem_resources[0].flags = IORESOURCE_MEM; - - if (u3_agp == NULL) { - DBG("U3 has no AGP, using full resource range\n"); - return; - } - - /* We "remove" the AGP resources from the resources allocated to HT, that - * is we create "holes". However, that code does assumptions that so far - * happen to be true (cross fingers...), typically that resources in the - * AGP node are properly ordered - */ - cur = 0; - for (i=0; i<3; i++) { - struct resource *res = &u3_agp->mem_resources[i]; - if (res->flags != IORESOURCE_MEM) - continue; - /* We don't care about "fine" resources */ - if (res->start >= 0xf0000000) - continue; - /* Check if it's just a matter of "shrinking" us in one direction */ - if (hose->mem_resources[cur].start == res->start) { - DBG("U3/HT: shrink start of %d, %08lx -> %08lx\n", - cur, hose->mem_resources[cur].start, res->end + 1); - hose->mem_resources[cur].start = res->end + 1; - continue; - } - if (hose->mem_resources[cur].end == res->end) { - DBG("U3/HT: shrink end of %d, %08lx -> %08lx\n", - cur, hose->mem_resources[cur].end, res->start - 1); - hose->mem_resources[cur].end = res->start - 1; - continue; - } - /* No, it's not the case, we need a hole */ - if (cur == 2) { - /* not enough resources for a hole, we drop part of the range */ - printk(KERN_WARNING "Running out of resources for /ht host !\n"); - hose->mem_resources[cur].end = res->start - 1; - continue; - } - cur++; - DBG("U3/HT: hole, %d end at %08lx, %d start at %08lx\n", - cur-1, res->start - 1, cur, res->end + 1); - hose->mem_resources[cur].name = np->full_name; - hose->mem_resources[cur].flags = IORESOURCE_MEM; - hose->mem_resources[cur].start = res->end + 1; - hose->mem_resources[cur].end = hose->mem_resources[cur-1].end; - hose->mem_resources[cur-1].end = res->start - 1; - } -} - -static void __init pmac_process_bridge_OF_ranges(struct pci_controller *hose, - struct device_node *dev, int primary) -{ - static unsigned int static_lc_ranges[2024]; - unsigned int *dt_ranges, *lc_ranges, *ranges, *prev; - unsigned int size; - int rlen = 0, orig_rlen; - int memno = 0; - struct resource *res; - int np, na = prom_n_addr_cells(dev); - - np = na + 5; - - /* First we try to merge ranges to fix a problem with some pmacs - * that can have more than 3 ranges, fortunately using contiguous - * addresses -- BenH - */ - dt_ranges = (unsigned int *) get_property(dev, "ranges", &rlen); - if (!dt_ranges) - return; - /* lc_ranges = alloc_bootmem(rlen);*/ - lc_ranges = static_lc_ranges; - if (!lc_ranges) - return; /* what can we do here ? */ - memcpy(lc_ranges, dt_ranges, rlen); - orig_rlen = rlen; - - /* Let's work on a copy of the "ranges" property instead of damaging - * the device-tree image in memory - */ - ranges = lc_ranges; - prev = NULL; - while ((rlen -= np * sizeof(unsigned int)) >= 0) { - if (prev) { - if (prev[0] == ranges[0] && prev[1] == ranges[1] && - (prev[2] + prev[na+4]) == ranges[2] && - (prev[na+2] + prev[na+4]) == ranges[na+2]) { - prev[na+4] += ranges[na+4]; - ranges[0] = 0; - ranges += np; - continue; - } - } - prev = ranges; - ranges += np; - } - - /* - * The ranges property is laid out as an array of elements, - * each of which comprises: - * cells 0 - 2: a PCI address - * cells 3 or 3+4: a CPU physical address - * (size depending on dev->n_addr_cells) - * cells 4+5 or 5+6: the size of the range - */ - ranges = lc_ranges; - rlen = orig_rlen; - while (ranges && (rlen -= np * sizeof(unsigned int)) >= 0) { - res = NULL; - size = ranges[na+4]; - switch (ranges[0] >> 24) { - case 1: /* I/O space */ - if (ranges[2] != 0) - break; - hose->io_base_phys = ranges[na+2]; - /* limit I/O space to 16MB */ - if (size > 0x01000000) - size = 0x01000000; - hose->io_base_virt = ioremap(ranges[na+2], size); - if (primary) - isa_io_base = (unsigned long) hose->io_base_virt; - res = &hose->io_resource; - res->flags = IORESOURCE_IO; - res->start = ranges[2]; - break; - case 2: /* memory space */ - memno = 0; - if (ranges[1] == 0 && ranges[2] == 0 - && ranges[na+4] <= (16 << 20)) { - /* 1st 16MB, i.e. ISA memory area */ -#if 0 - if (primary) - isa_mem_base = ranges[na+2]; -#endif - memno = 1; - } - while (memno < 3 && hose->mem_resources[memno].flags) - ++memno; - if (memno == 0) - hose->pci_mem_offset = ranges[na+2] - ranges[2]; - if (memno < 3) { - res = &hose->mem_resources[memno]; - res->flags = IORESOURCE_MEM; - res->start = ranges[na+2]; - } - break; - } - if (res != NULL) { - res->name = dev->full_name; - res->end = res->start + size - 1; - res->parent = NULL; - res->sibling = NULL; - res->child = NULL; - } - ranges += np; - } -} - -/* - * We assume that if we have a G3 powermac, we have one bridge called - * "pci" (a MPC106) and no bandit or chaos bridges, and contrariwise, - * if we have one or more bandit or chaos bridges, we don't have a MPC106. - */ -static int __init add_bridge(struct device_node *dev) -{ - int len; - struct pci_controller *hose; - char* disp_name; - int *bus_range; - int primary = 1; - struct property *of_prop; - - DBG("Adding PCI host bridge %s\n", dev->full_name); - - bus_range = (int *) get_property(dev, "bus-range", &len); - if (bus_range == NULL || len < 2 * sizeof(int)) { - printk(KERN_WARNING "Can't get bus-range for %s, assume bus 0\n", - dev->full_name); - } - - hose = alloc_bootmem(sizeof(struct pci_controller)); - if (hose == NULL) - return -ENOMEM; - pci_setup_pci_controller(hose); - - hose->arch_data = dev; - hose->first_busno = bus_range ? bus_range[0] : 0; - hose->last_busno = bus_range ? bus_range[1] : 0xff; - - of_prop = alloc_bootmem(sizeof(struct property) + - sizeof(hose->global_number)); - if (of_prop) { - memset(of_prop, 0, sizeof(struct property)); - of_prop->name = "linux,pci-domain"; - of_prop->length = sizeof(hose->global_number); - of_prop->value = (unsigned char *)&of_prop[1]; - memcpy(of_prop->value, &hose->global_number, sizeof(hose->global_number)); - prom_add_property(dev, of_prop); - } - - disp_name = NULL; - if (device_is_compatible(dev, "u3-agp")) { - setup_u3_agp(hose); - disp_name = "U3-AGP"; - primary = 0; - } else if (device_is_compatible(dev, "u3-ht")) { - setup_u3_ht(hose); - disp_name = "U3-HT"; - primary = 1; - } - printk(KERN_INFO "Found %s PCI host bridge. Firmware bus number: %d->%d\n", - disp_name, hose->first_busno, hose->last_busno); - - /* Interpret the "ranges" property */ - /* This also maps the I/O region and sets isa_io/mem_base */ - pmac_process_bridge_OF_ranges(hose, dev, primary); - - /* Fixup "bus-range" OF property */ - fixup_bus_range(dev); - - return 0; -} - -/* - * We use our own read_irq_line here because PCI_INTERRUPT_PIN is - * crap on some of Apple ASICs. We unconditionally use the Open Firmware - * interrupt number as this is always right. - */ -static int pmac_pci_read_irq_line(struct pci_dev *pci_dev) -{ - struct device_node *node; - - node = pci_device_to_OF_node(pci_dev); - if (node == NULL) - return -1; - if (node->n_intrs == 0) - return -1; - pci_dev->irq = node->intrs[0].line; - pci_write_config_byte(pci_dev, PCI_INTERRUPT_LINE, pci_dev->irq); - - return 0; -} - -void __init pmac_pcibios_fixup(void) -{ - struct pci_dev *dev = NULL; - - for_each_pci_dev(dev) - pmac_pci_read_irq_line(dev); -} - -static void __init pmac_fixup_phb_resources(void) -{ - struct pci_controller *hose, *tmp; - - list_for_each_entry_safe(hose, tmp, &hose_list, list_node) { - unsigned long offset = (unsigned long)hose->io_base_virt - pci_io_base; - hose->io_resource.start += offset; - hose->io_resource.end += offset; - printk(KERN_INFO "PCI Host %d, io start: %lx; io end: %lx\n", - hose->global_number, - hose->io_resource.start, hose->io_resource.end); - } -} - -void __init pmac_pci_init(void) -{ - struct device_node *np, *root; - struct device_node *ht = NULL; - - /* Probe root PCI hosts, that is on U3 the AGP host and the - * HyperTransport host. That one is actually "kept" around - * and actually added last as it's resource management relies - * on the AGP resources to have been setup first - */ - root = of_find_node_by_path("/"); - if (root == NULL) { - printk(KERN_CRIT "pmac_find_bridges: can't find root of device tree\n"); - return; - } - for (np = NULL; (np = of_get_next_child(root, np)) != NULL;) { - if (np->name == NULL) - continue; - if (strcmp(np->name, "pci") == 0) { - if (add_bridge(np) == 0) - of_node_get(np); - } - if (strcmp(np->name, "ht") == 0) { - of_node_get(np); - ht = np; - } - } - of_node_put(root); - - /* Now setup the HyperTransport host if we found any - */ - if (ht && add_bridge(ht) != 0) - of_node_put(ht); - - /* Fixup the IO resources on our host bridges as the common code - * does it only for childs of the host bridges - */ - pmac_fixup_phb_resources(); - - /* Setup the linkage between OF nodes and PHBs */ - pci_devs_phb_init(); - - /* Fixup the PCI<->OF mapping for U3 AGP due to bus renumbering. We - * assume there is no P2P bridge on the AGP bus, which should be a - * safe assumptions hopefully. - */ - if (u3_agp) { - struct device_node *np = u3_agp->arch_data; - PCI_DN(np)->busno = 0xf0; - for (np = np->child; np; np = np->sibling) - PCI_DN(np)->busno = 0xf0; - } - - pmac_check_ht_link(); - - /* Tell pci.c to not use the common resource allocation mecanism */ - pci_probe_only = 1; - - /* Allow all IO */ - io_page_mask = -1; -} - -/* - * Disable second function on K2-SATA, it's broken - * and disable IO BARs on first one - */ -static void fixup_k2_sata(struct pci_dev* dev) -{ - int i; - u16 cmd; - - if (PCI_FUNC(dev->devfn) > 0) { - pci_read_config_word(dev, PCI_COMMAND, &cmd); - cmd &= ~(PCI_COMMAND_IO | PCI_COMMAND_MEMORY); - pci_write_config_word(dev, PCI_COMMAND, cmd); - for (i = 0; i < 6; i++) { - dev->resource[i].start = dev->resource[i].end = 0; - dev->resource[i].flags = 0; - pci_write_config_dword(dev, PCI_BASE_ADDRESS_0 + 4 * i, 0); - } - } else { - pci_read_config_word(dev, PCI_COMMAND, &cmd); - cmd &= ~PCI_COMMAND_IO; - pci_write_config_word(dev, PCI_COMMAND, cmd); - for (i = 0; i < 5; i++) { - dev->resource[i].start = dev->resource[i].end = 0; - dev->resource[i].flags = 0; - pci_write_config_dword(dev, PCI_BASE_ADDRESS_0 + 4 * i, 0); - } - } -} -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_SERVERWORKS, 0x0240, fixup_k2_sata); - diff --git a/arch/ppc64/kernel/pmac_setup.c b/arch/ppc64/kernel/pmac_setup.c deleted file mode 100644 index fa8121d53b89..000000000000 --- a/arch/ppc64/kernel/pmac_setup.c +++ /dev/null @@ -1,525 +0,0 @@ -/* - * arch/ppc/platforms/setup.c - * - * PowerPC version - * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org) - * - * Adapted for Power Macintosh by Paul Mackerras - * Copyright (C) 1996 Paul Mackerras (paulus@cs.anu.edu.au) - * - * Derived from "arch/alpha/kernel/setup.c" - * Copyright (C) 1995 Linus Torvalds - * - * Maintained by Benjamin Herrenschmidt (benh@kernel.crashing.org) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - */ - -/* - * bootup setup stuff.. - */ - -#undef DEBUG - -#include <linux/config.h> -#include <linux/init.h> -#include <linux/errno.h> -#include <linux/sched.h> -#include <linux/kernel.h> -#include <linux/mm.h> -#include <linux/stddef.h> -#include <linux/unistd.h> -#include <linux/ptrace.h> -#include <linux/slab.h> -#include <linux/user.h> -#include <linux/a.out.h> -#include <linux/tty.h> -#include <linux/string.h> -#include <linux/delay.h> -#include <linux/ioport.h> -#include <linux/major.h> -#include <linux/initrd.h> -#include <linux/vt_kern.h> -#include <linux/console.h> -#include <linux/ide.h> -#include <linux/pci.h> -#include <linux/adb.h> -#include <linux/cuda.h> -#include <linux/pmu.h> -#include <linux/irq.h> -#include <linux/seq_file.h> -#include <linux/root_dev.h> -#include <linux/bitops.h> - -#include <asm/processor.h> -#include <asm/sections.h> -#include <asm/prom.h> -#include <asm/system.h> -#include <asm/io.h> -#include <asm/pci-bridge.h> -#include <asm/iommu.h> -#include <asm/machdep.h> -#include <asm/dma.h> -#include <asm/btext.h> -#include <asm/cputable.h> -#include <asm/pmac_feature.h> -#include <asm/time.h> -#include <asm/of_device.h> -#include <asm/lmb.h> -#include <asm/smu.h> -#include <asm/pmc.h> - -#include "pmac.h" -#include "mpic.h" - -#ifdef DEBUG -#define DBG(fmt...) udbg_printf(fmt) -#else -#define DBG(fmt...) -#endif - -static int current_root_goodness = -1; -#define DEFAULT_ROOT_DEVICE Root_SDA1 /* sda1 - slightly silly choice */ - -extern int powersave_nap; -int sccdbg; - -sys_ctrler_t sys_ctrler; -EXPORT_SYMBOL(sys_ctrler); - -#ifdef CONFIG_PMAC_SMU -unsigned long smu_cmdbuf_abs; -EXPORT_SYMBOL(smu_cmdbuf_abs); -#endif - -extern void udbg_init_scc(struct device_node *np); - -static void __pmac pmac_show_cpuinfo(struct seq_file *m) -{ - struct device_node *np; - char *pp; - int plen; - char* mbname; - int mbmodel = pmac_call_feature(PMAC_FTR_GET_MB_INFO, NULL, - PMAC_MB_INFO_MODEL, 0); - unsigned int mbflags = pmac_call_feature(PMAC_FTR_GET_MB_INFO, NULL, - PMAC_MB_INFO_FLAGS, 0); - - if (pmac_call_feature(PMAC_FTR_GET_MB_INFO, NULL, PMAC_MB_INFO_NAME, - (long)&mbname) != 0) - mbname = "Unknown"; - - /* find motherboard type */ - seq_printf(m, "machine\t\t: "); - np = of_find_node_by_path("/"); - if (np != NULL) { - pp = (char *) get_property(np, "model", NULL); - if (pp != NULL) - seq_printf(m, "%s\n", pp); - else - seq_printf(m, "PowerMac\n"); - pp = (char *) get_property(np, "compatible", &plen); - if (pp != NULL) { - seq_printf(m, "motherboard\t:"); - while (plen > 0) { - int l = strlen(pp) + 1; - seq_printf(m, " %s", pp); - plen -= l; - pp += l; - } - seq_printf(m, "\n"); - } - of_node_put(np); - } else - seq_printf(m, "PowerMac\n"); - - /* print parsed model */ - seq_printf(m, "detected as\t: %d (%s)\n", mbmodel, mbname); - seq_printf(m, "pmac flags\t: %08x\n", mbflags); - - /* Indicate newworld */ - seq_printf(m, "pmac-generation\t: NewWorld\n"); -} - - -static void __init pmac_setup_arch(void) -{ - /* init to some ~sane value until calibrate_delay() runs */ - loops_per_jiffy = 50000000; - - /* Probe motherboard chipset */ - pmac_feature_init(); -#if 0 - /* Lock-enable the SCC channel used for debug */ - if (sccdbg) { - np = of_find_node_by_name(NULL, "escc"); - if (np) - pmac_call_feature(PMAC_FTR_SCC_ENABLE, np, - PMAC_SCC_ASYNC | PMAC_SCC_FLAG_XMON, 1); - } -#endif - /* We can NAP */ - powersave_nap = 1; - -#ifdef CONFIG_ADB_PMU - /* Initialize the PMU if any */ - find_via_pmu(); -#endif -#ifdef CONFIG_PMAC_SMU - /* Initialize the SMU if any */ - smu_init(); -#endif - - /* Init NVRAM access */ - pmac_nvram_init(); - - /* Setup SMP callback */ -#ifdef CONFIG_SMP - pmac_setup_smp(); -#endif - - /* Lookup PCI hosts */ - pmac_pci_init(); - -#ifdef CONFIG_DUMMY_CONSOLE - conswitchp = &dummy_con; -#endif - - printk(KERN_INFO "Using native/NAP idle loop\n"); -} - -#ifdef CONFIG_SCSI -void note_scsi_host(struct device_node *node, void *host) -{ - /* Obsolete */ -} -#endif - - -static int initializing = 1; - -static int pmac_late_init(void) -{ - initializing = 0; - return 0; -} - -late_initcall(pmac_late_init); - -/* can't be __init - can be called whenever a disk is first accessed */ -void __pmac note_bootable_part(dev_t dev, int part, int goodness) -{ - extern dev_t boot_dev; - char *p; - - if (!initializing) - return; - if ((goodness <= current_root_goodness) && - ROOT_DEV != DEFAULT_ROOT_DEVICE) - return; - p = strstr(saved_command_line, "root="); - if (p != NULL && (p == saved_command_line || p[-1] == ' ')) - return; - - if (!boot_dev || dev == boot_dev) { - ROOT_DEV = dev + part; - boot_dev = 0; - current_root_goodness = goodness; - } -} - -static void __pmac pmac_restart(char *cmd) -{ - switch(sys_ctrler) { -#ifdef CONFIG_ADB_PMU - case SYS_CTRLER_PMU: - pmu_restart(); - break; -#endif - -#ifdef CONFIG_PMAC_SMU - case SYS_CTRLER_SMU: - smu_restart(); - break; -#endif - default: - ; - } -} - -static void __pmac pmac_power_off(void) -{ - switch(sys_ctrler) { -#ifdef CONFIG_ADB_PMU - case SYS_CTRLER_PMU: - pmu_shutdown(); - break; -#endif -#ifdef CONFIG_PMAC_SMU - case SYS_CTRLER_SMU: - smu_shutdown(); - break; -#endif - default: - ; - } -} - -static void __pmac pmac_halt(void) -{ - pmac_power_off(); -} - -#ifdef CONFIG_BOOTX_TEXT -static void btext_putc(unsigned char c) -{ - btext_drawchar(c); -} - -static void __init init_boot_display(void) -{ - char *name; - struct device_node *np = NULL; - int rc = -ENODEV; - - printk("trying to initialize btext ...\n"); - - name = (char *)get_property(of_chosen, "linux,stdout-path", NULL); - if (name != NULL) { - np = of_find_node_by_path(name); - if (np != NULL) { - if (strcmp(np->type, "display") != 0) { - printk("boot stdout isn't a display !\n"); - of_node_put(np); - np = NULL; - } - } - } - if (np) - rc = btext_initialize(np); - if (rc == 0) - return; - - for (np = NULL; (np = of_find_node_by_type(np, "display"));) { - if (get_property(np, "linux,opened", NULL)) { - printk("trying %s ...\n", np->full_name); - rc = btext_initialize(np); - printk("result: %d\n", rc); - } - if (rc == 0) - return; - } -} -#endif /* CONFIG_BOOTX_TEXT */ - -/* - * Early initialization. - */ -static void __init pmac_init_early(void) -{ - DBG(" -> pmac_init_early\n"); - - /* Initialize hash table, from now on, we can take hash faults - * and call ioremap - */ - hpte_init_native(); - - /* Init SCC */ - if (strstr(cmd_line, "sccdbg")) { - sccdbg = 1; - udbg_init_scc(NULL); - } -#ifdef CONFIG_BOOTX_TEXT - else { - init_boot_display(); - - udbg_putc = btext_putc; - } -#endif /* CONFIG_BOOTX_TEXT */ - - /* Setup interrupt mapping options */ - ppc64_interrupt_controller = IC_OPEN_PIC; - - iommu_init_early_u3(); - - DBG(" <- pmac_init_early\n"); -} - -static int pmac_u3_cascade(struct pt_regs *regs, void *data) -{ - return mpic_get_one_irq((struct mpic *)data, regs); -} - -static __init void pmac_init_IRQ(void) -{ - struct device_node *irqctrler = NULL; - struct device_node *irqctrler2 = NULL; - struct device_node *np = NULL; - struct mpic *mpic1, *mpic2; - - /* We first try to detect Apple's new Core99 chipset, since mac-io - * is quite different on those machines and contains an IBM MPIC2. - */ - while ((np = of_find_node_by_type(np, "open-pic")) != NULL) { - struct device_node *parent = of_get_parent(np); - if (parent && !strcmp(parent->name, "u3")) - irqctrler2 = of_node_get(np); - else - irqctrler = of_node_get(np); - of_node_put(parent); - } - if (irqctrler != NULL && irqctrler->n_addrs > 0) { - unsigned char senses[128]; - - printk(KERN_INFO "PowerMac using OpenPIC irq controller at 0x%08x\n", - (unsigned int)irqctrler->addrs[0].address); - - prom_get_irq_senses(senses, 0, 128); - mpic1 = mpic_alloc(irqctrler->addrs[0].address, - MPIC_PRIMARY | MPIC_WANTS_RESET, - 0, 0, 128, 256, senses, 128, " K2-MPIC "); - BUG_ON(mpic1 == NULL); - mpic_init(mpic1); - - if (irqctrler2 != NULL && irqctrler2->n_intrs > 0 && - irqctrler2->n_addrs > 0) { - printk(KERN_INFO "Slave OpenPIC at 0x%08x hooked on IRQ %d\n", - (u32)irqctrler2->addrs[0].address, - irqctrler2->intrs[0].line); - - pmac_call_feature(PMAC_FTR_ENABLE_MPIC, irqctrler2, 0, 0); - prom_get_irq_senses(senses, 128, 128 + 128); - - /* We don't need to set MPIC_BROKEN_U3 here since we don't have - * hypertransport interrupts routed to it - */ - mpic2 = mpic_alloc(irqctrler2->addrs[0].address, - MPIC_BIG_ENDIAN | MPIC_WANTS_RESET, - 0, 128, 128, 0, senses, 128, " U3-MPIC "); - BUG_ON(mpic2 == NULL); - mpic_init(mpic2); - mpic_setup_cascade(irqctrler2->intrs[0].line, - pmac_u3_cascade, mpic2); - } - } - of_node_put(irqctrler); - of_node_put(irqctrler2); -} - -static void __init pmac_progress(char *s, unsigned short hex) -{ - if (sccdbg) { - udbg_puts(s); - udbg_puts("\n"); - } -#ifdef CONFIG_BOOTX_TEXT - else if (boot_text_mapped) { - btext_drawstring(s); - btext_drawstring("\n"); - } -#endif /* CONFIG_BOOTX_TEXT */ -} - -/* - * pmac has no legacy IO, anything calling this function has to - * fail or bad things will happen - */ -static int pmac_check_legacy_ioport(unsigned int baseport) -{ - return -ENODEV; -} - -static int __init pmac_declare_of_platform_devices(void) -{ - struct device_node *np, *npp; - - npp = of_find_node_by_name(NULL, "u3"); - if (npp) { - for (np = NULL; (np = of_get_next_child(npp, np)) != NULL;) { - if (strncmp(np->name, "i2c", 3) == 0) { - of_platform_device_create(np, "u3-i2c", NULL); - of_node_put(np); - break; - } - } - of_node_put(npp); - } - npp = of_find_node_by_type(NULL, "smu"); - if (npp) { - of_platform_device_create(npp, "smu", NULL); - of_node_put(npp); - } - - return 0; -} - -device_initcall(pmac_declare_of_platform_devices); - -/* - * Called very early, MMU is off, device-tree isn't unflattened - */ -static int __init pmac_probe(int platform) -{ - if (platform != PLATFORM_POWERMAC) - return 0; - /* - * On U3, the DART (iommu) must be allocated now since it - * has an impact on htab_initialize (due to the large page it - * occupies having to be broken up so the DART itself is not - * part of the cacheable linar mapping - */ - alloc_u3_dart_table(); - -#ifdef CONFIG_PMAC_SMU - /* - * SMU based G5s need some memory below 2Gb, at least the current - * driver needs that. We have to allocate it now. We allocate 4k - * (1 small page) for now. - */ - smu_cmdbuf_abs = lmb_alloc_base(4096, 4096, 0x80000000UL); -#endif /* CONFIG_PMAC_SMU */ - - return 1; -} - -static int pmac_probe_mode(struct pci_bus *bus) -{ - struct device_node *node = bus->sysdata; - - /* We need to use normal PCI probing for the AGP bus, - since the device for the AGP bridge isn't in the tree. */ - if (bus->self == NULL && device_is_compatible(node, "u3-agp")) - return PCI_PROBE_NORMAL; - - return PCI_PROBE_DEVTREE; -} - -struct machdep_calls __initdata pmac_md = { -#ifdef CONFIG_HOTPLUG_CPU - .cpu_die = generic_mach_cpu_die, -#endif - .probe = pmac_probe, - .setup_arch = pmac_setup_arch, - .init_early = pmac_init_early, - .get_cpuinfo = pmac_show_cpuinfo, - .init_IRQ = pmac_init_IRQ, - .get_irq = mpic_get_irq, - .pcibios_fixup = pmac_pcibios_fixup, - .pci_probe_mode = pmac_probe_mode, - .restart = pmac_restart, - .power_off = pmac_power_off, - .halt = pmac_halt, - .get_boot_time = pmac_get_boot_time, - .set_rtc_time = pmac_set_rtc_time, - .get_rtc_time = pmac_get_rtc_time, - .calibrate_decr = pmac_calibrate_decr, - .feature_call = pmac_do_feature_call, - .progress = pmac_progress, - .check_legacy_ioport = pmac_check_legacy_ioport, - .idle_loop = native_idle, - .enable_pmcs = power4_enable_pmcs, -}; diff --git a/arch/ppc64/kernel/pmac_smp.c b/arch/ppc64/kernel/pmac_smp.c deleted file mode 100644 index a23de37227bf..000000000000 --- a/arch/ppc64/kernel/pmac_smp.c +++ /dev/null @@ -1,330 +0,0 @@ -/* - * SMP support for power macintosh. - * - * We support both the old "powersurge" SMP architecture - * and the current Core99 (G4 PowerMac) machines. - * - * Note that we don't support the very first rev. of - * Apple/DayStar 2 CPUs board, the one with the funky - * watchdog. Hopefully, none of these should be there except - * maybe internally to Apple. I should probably still add some - * code to detect this card though and disable SMP. --BenH. - * - * Support Macintosh G4 SMP by Troy Benjegerdes (hozer@drgw.net) - * and Ben Herrenschmidt <benh@kernel.crashing.org>. - * - * Support for DayStar quad CPU cards - * Copyright (C) XLR8, Inc. 1994-2000 - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#undef DEBUG - -#include <linux/config.h> -#include <linux/kernel.h> -#include <linux/sched.h> -#include <linux/smp.h> -#include <linux/smp_lock.h> -#include <linux/interrupt.h> -#include <linux/kernel_stat.h> -#include <linux/init.h> -#include <linux/spinlock.h> -#include <linux/errno.h> -#include <linux/irq.h> - -#include <asm/ptrace.h> -#include <asm/atomic.h> -#include <asm/irq.h> -#include <asm/page.h> -#include <asm/pgtable.h> -#include <asm/sections.h> -#include <asm/io.h> -#include <asm/prom.h> -#include <asm/smp.h> -#include <asm/machdep.h> -#include <asm/pmac_feature.h> -#include <asm/time.h> -#include <asm/cacheflush.h> -#include <asm/keylargo.h> -#include <asm/pmac_low_i2c.h> - -#include "mpic.h" - -#ifdef DEBUG -#define DBG(fmt...) udbg_printf(fmt) -#else -#define DBG(fmt...) -#endif - -extern void pmac_secondary_start_1(void); -extern void pmac_secondary_start_2(void); -extern void pmac_secondary_start_3(void); - -extern struct smp_ops_t *smp_ops; - -static void (*pmac_tb_freeze)(int freeze); -static struct device_node *pmac_tb_clock_chip_host; -static u8 pmac_tb_pulsar_addr; -static DEFINE_SPINLOCK(timebase_lock); -static unsigned long timebase; - -static void smp_core99_cypress_tb_freeze(int freeze) -{ - u8 data; - int rc; - - /* Strangely, the device-tree says address is 0xd2, but darwin - * accesses 0xd0 ... - */ - pmac_low_i2c_setmode(pmac_tb_clock_chip_host, pmac_low_i2c_mode_combined); - rc = pmac_low_i2c_xfer(pmac_tb_clock_chip_host, - 0xd0 | pmac_low_i2c_read, - 0x81, &data, 1); - if (rc != 0) - goto bail; - - data = (data & 0xf3) | (freeze ? 0x00 : 0x0c); - - pmac_low_i2c_setmode(pmac_tb_clock_chip_host, pmac_low_i2c_mode_stdsub); - rc = pmac_low_i2c_xfer(pmac_tb_clock_chip_host, - 0xd0 | pmac_low_i2c_write, - 0x81, &data, 1); - - bail: - if (rc != 0) { - printk("Cypress Timebase %s rc: %d\n", - freeze ? "freeze" : "unfreeze", rc); - panic("Timebase freeze failed !\n"); - } -} - -static void smp_core99_pulsar_tb_freeze(int freeze) -{ - u8 data; - int rc; - - pmac_low_i2c_setmode(pmac_tb_clock_chip_host, pmac_low_i2c_mode_combined); - rc = pmac_low_i2c_xfer(pmac_tb_clock_chip_host, - pmac_tb_pulsar_addr | pmac_low_i2c_read, - 0x2e, &data, 1); - if (rc != 0) - goto bail; - - data = (data & 0x88) | (freeze ? 0x11 : 0x22); - - pmac_low_i2c_setmode(pmac_tb_clock_chip_host, pmac_low_i2c_mode_stdsub); - rc = pmac_low_i2c_xfer(pmac_tb_clock_chip_host, - pmac_tb_pulsar_addr | pmac_low_i2c_write, - 0x2e, &data, 1); - bail: - if (rc != 0) { - printk(KERN_ERR "Pulsar Timebase %s rc: %d\n", - freeze ? "freeze" : "unfreeze", rc); - panic("Timebase freeze failed !\n"); - } -} - - -static void smp_core99_give_timebase(void) -{ - /* Open i2c bus for synchronous access */ - if (pmac_low_i2c_open(pmac_tb_clock_chip_host, 0)) - panic("Can't open i2c for TB sync !\n"); - - spin_lock(&timebase_lock); - (*pmac_tb_freeze)(1); - mb(); - timebase = get_tb(); - spin_unlock(&timebase_lock); - - while (timebase) - barrier(); - - spin_lock(&timebase_lock); - (*pmac_tb_freeze)(0); - spin_unlock(&timebase_lock); - - /* Close i2c bus */ - pmac_low_i2c_close(pmac_tb_clock_chip_host); -} - - -static void __devinit smp_core99_take_timebase(void) -{ - while (!timebase) - barrier(); - spin_lock(&timebase_lock); - set_tb(timebase >> 32, timebase & 0xffffffff); - timebase = 0; - spin_unlock(&timebase_lock); -} - - -static int __init smp_core99_probe(void) -{ - struct device_node *cpus; - struct device_node *cc; - int ncpus = 0; - - /* Maybe use systemconfiguration here ? */ - if (ppc_md.progress) ppc_md.progress("smp_core99_probe", 0x345); - - /* Count CPUs in the device-tree */ - for (cpus = NULL; (cpus = of_find_node_by_type(cpus, "cpu")) != NULL;) - ++ncpus; - - printk(KERN_INFO "PowerMac SMP probe found %d cpus\n", ncpus); - - /* Nothing more to do if less than 2 of them */ - if (ncpus <= 1) - return 1; - - /* HW sync only on these platforms */ - if (!machine_is_compatible("PowerMac7,2") && - !machine_is_compatible("PowerMac7,3") && - !machine_is_compatible("RackMac3,1")) - goto nohwsync; - - /* Look for the clock chip */ - for (cc = NULL; (cc = of_find_node_by_name(cc, "i2c-hwclock")) != NULL;) { - struct device_node *p = of_get_parent(cc); - u32 *reg; - int ok; - ok = p && device_is_compatible(p, "uni-n-i2c"); - if (!ok) - goto next; - reg = (u32 *)get_property(cc, "reg", NULL); - if (reg == NULL) - goto next; - switch (*reg) { - case 0xd2: - if (device_is_compatible(cc, "pulsar-legacy-slewing")) { - pmac_tb_freeze = smp_core99_pulsar_tb_freeze; - pmac_tb_pulsar_addr = 0xd2; - printk(KERN_INFO "Timebase clock is Pulsar chip\n"); - } else if (device_is_compatible(cc, "cy28508")) { - pmac_tb_freeze = smp_core99_cypress_tb_freeze; - printk(KERN_INFO "Timebase clock is Cypress chip\n"); - } - break; - case 0xd4: - pmac_tb_freeze = smp_core99_pulsar_tb_freeze; - pmac_tb_pulsar_addr = 0xd4; - printk(KERN_INFO "Timebase clock is Pulsar chip\n"); - break; - } - if (pmac_tb_freeze != NULL) { - pmac_tb_clock_chip_host = p; - smp_ops->give_timebase = smp_core99_give_timebase; - smp_ops->take_timebase = smp_core99_take_timebase; - of_node_put(cc); - of_node_put(p); - break; - } - next: - of_node_put(p); - } - - nohwsync: - mpic_request_ipis(); - - return ncpus; -} - -static void __init smp_core99_kick_cpu(int nr) -{ - int save_vector, j; - unsigned long new_vector; - unsigned long flags; - volatile unsigned int *vector - = ((volatile unsigned int *)(KERNELBASE+0x100)); - - if (nr < 1 || nr > 3) - return; - if (ppc_md.progress) ppc_md.progress("smp_core99_kick_cpu", 0x346); - - local_irq_save(flags); - local_irq_disable(); - - /* Save reset vector */ - save_vector = *vector; - - /* Setup fake reset vector that does - * b .pmac_secondary_start - KERNELBASE - */ - switch(nr) { - case 1: - new_vector = (unsigned long)pmac_secondary_start_1; - break; - case 2: - new_vector = (unsigned long)pmac_secondary_start_2; - break; - case 3: - default: - new_vector = (unsigned long)pmac_secondary_start_3; - break; - } - *vector = 0x48000002 + (new_vector - KERNELBASE); - - /* flush data cache and inval instruction cache */ - flush_icache_range((unsigned long) vector, (unsigned long) vector + 4); - - /* Put some life in our friend */ - pmac_call_feature(PMAC_FTR_RESET_CPU, NULL, nr, 0); - paca[nr].cpu_start = 1; - - /* FIXME: We wait a bit for the CPU to take the exception, I should - * instead wait for the entry code to set something for me. Well, - * ideally, all that crap will be done in prom.c and the CPU left - * in a RAM-based wait loop like CHRP. - */ - for (j = 1; j < 1000000; j++) - mb(); - - /* Restore our exception vector */ - *vector = save_vector; - flush_icache_range((unsigned long) vector, (unsigned long) vector + 4); - - local_irq_restore(flags); - if (ppc_md.progress) ppc_md.progress("smp_core99_kick_cpu done", 0x347); -} - -static void __init smp_core99_setup_cpu(int cpu_nr) -{ - /* Setup MPIC */ - mpic_setup_this_cpu(); - - if (cpu_nr == 0) { - extern void g5_phy_disable_cpu1(void); - - /* If we didn't start the second CPU, we must take - * it off the bus - */ - if (num_online_cpus() < 2) - g5_phy_disable_cpu1(); - if (ppc_md.progress) ppc_md.progress("smp_core99_setup_cpu 0 done", 0x349); - } -} - -struct smp_ops_t core99_smp_ops __pmacdata = { - .message_pass = smp_mpic_message_pass, - .probe = smp_core99_probe, - .kick_cpu = smp_core99_kick_cpu, - .setup_cpu = smp_core99_setup_cpu, - .give_timebase = smp_generic_give_timebase, - .take_timebase = smp_generic_take_timebase, -}; - -void __init pmac_setup_smp(void) -{ - smp_ops = &core99_smp_ops; -#ifdef CONFIG_HOTPLUG_CPU - smp_ops->cpu_enable = generic_cpu_enable; - smp_ops->cpu_disable = generic_cpu_disable; - smp_ops->cpu_die = generic_cpu_die; -#endif -} diff --git a/arch/ppc64/kernel/pmac_time.c b/arch/ppc64/kernel/pmac_time.c deleted file mode 100644 index 41bbb8c59697..000000000000 --- a/arch/ppc64/kernel/pmac_time.c +++ /dev/null @@ -1,195 +0,0 @@ -/* - * Support for periodic interrupts (100 per second) and for getting - * the current time from the RTC on Power Macintoshes. - * - * We use the decrementer register for our periodic interrupts. - * - * Paul Mackerras August 1996. - * Copyright (C) 1996 Paul Mackerras. - * Copyright (C) 2003-2005 Benjamin Herrenschmidt. - * - */ -#include <linux/config.h> -#include <linux/errno.h> -#include <linux/sched.h> -#include <linux/kernel.h> -#include <linux/param.h> -#include <linux/string.h> -#include <linux/mm.h> -#include <linux/init.h> -#include <linux/time.h> -#include <linux/adb.h> -#include <linux/pmu.h> -#include <linux/interrupt.h> - -#include <asm/sections.h> -#include <asm/prom.h> -#include <asm/system.h> -#include <asm/io.h> -#include <asm/pgtable.h> -#include <asm/machdep.h> -#include <asm/time.h> -#include <asm/nvram.h> -#include <asm/smu.h> - -#undef DEBUG - -#ifdef DEBUG -#define DBG(x...) printk(x) -#else -#define DBG(x...) -#endif - -/* Apparently the RTC stores seconds since 1 Jan 1904 */ -#define RTC_OFFSET 2082844800 - -/* - * Calibrate the decrementer frequency with the VIA timer 1. - */ -#define VIA_TIMER_FREQ_6 4700000 /* time 1 frequency * 6 */ - -extern struct timezone sys_tz; -extern void to_tm(int tim, struct rtc_time * tm); - -void __pmac pmac_get_rtc_time(struct rtc_time *tm) -{ - switch(sys_ctrler) { -#ifdef CONFIG_ADB_PMU - case SYS_CTRLER_PMU: { - /* TODO: Move that to a function in the PMU driver */ - struct adb_request req; - unsigned int now; - - if (pmu_request(&req, NULL, 1, PMU_READ_RTC) < 0) - return; - pmu_wait_complete(&req); - if (req.reply_len != 4) - printk(KERN_ERR "pmac_get_rtc_time: PMU returned a %d" - " bytes reply\n", req.reply_len); - now = (req.reply[0] << 24) + (req.reply[1] << 16) - + (req.reply[2] << 8) + req.reply[3]; - DBG("get: %u -> %u\n", (int)now, (int)(now - RTC_OFFSET)); - now -= RTC_OFFSET; - - to_tm(now, tm); - tm->tm_year -= 1900; - tm->tm_mon -= 1; - - DBG("-> tm_mday: %d, tm_mon: %d, tm_year: %d, %d:%02d:%02d\n", - tm->tm_mday, tm->tm_mon, tm->tm_year, - tm->tm_hour, tm->tm_min, tm->tm_sec); - break; - } -#endif /* CONFIG_ADB_PMU */ - -#ifdef CONFIG_PMAC_SMU - case SYS_CTRLER_SMU: - smu_get_rtc_time(tm, 1); - break; -#endif /* CONFIG_PMAC_SMU */ - default: - ; - } -} - -int __pmac pmac_set_rtc_time(struct rtc_time *tm) -{ - switch(sys_ctrler) { -#ifdef CONFIG_ADB_PMU - case SYS_CTRLER_PMU: { - /* TODO: Move that to a function in the PMU driver */ - struct adb_request req; - unsigned int nowtime; - - DBG("set: tm_mday: %d, tm_mon: %d, tm_year: %d," - " %d:%02d:%02d\n", - tm->tm_mday, tm->tm_mon, tm->tm_year, - tm->tm_hour, tm->tm_min, tm->tm_sec); - - nowtime = mktime(tm->tm_year + 1900, tm->tm_mon + 1, - tm->tm_mday, tm->tm_hour, tm->tm_min, - tm->tm_sec); - - DBG("-> %u -> %u\n", (int)nowtime, - (int)(nowtime + RTC_OFFSET)); - nowtime += RTC_OFFSET; - - if (pmu_request(&req, NULL, 5, PMU_SET_RTC, - nowtime >> 24, nowtime >> 16, - nowtime >> 8, nowtime) < 0) - return -ENXIO; - pmu_wait_complete(&req); - if (req.reply_len != 0) - printk(KERN_ERR "pmac_set_rtc_time: PMU returned a %d" - " bytes reply\n", req.reply_len); - return 0; - } -#endif /* CONFIG_ADB_PMU */ - -#ifdef CONFIG_PMAC_SMU - case SYS_CTRLER_SMU: - return smu_set_rtc_time(tm, 1); -#endif /* CONFIG_PMAC_SMU */ - default: - return -ENODEV; - } -} - -void __init pmac_get_boot_time(struct rtc_time *tm) -{ - pmac_get_rtc_time(tm); - -#ifdef disabled__CONFIG_NVRAM - s32 delta = 0; - int dst; - - delta = ((s32)pmac_xpram_read(PMAC_XPRAM_MACHINE_LOC + 0x9)) << 16; - delta |= ((s32)pmac_xpram_read(PMAC_XPRAM_MACHINE_LOC + 0xa)) << 8; - delta |= pmac_xpram_read(PMAC_XPRAM_MACHINE_LOC + 0xb); - if (delta & 0x00800000UL) - delta |= 0xFF000000UL; - dst = ((pmac_xpram_read(PMAC_XPRAM_MACHINE_LOC + 0x8) & 0x80) != 0); - printk("GMT Delta read from XPRAM: %d minutes, DST: %s\n", delta/60, - dst ? "on" : "off"); -#endif -} - -/* - * Query the OF and get the decr frequency. - * FIXME: merge this with generic_calibrate_decr - */ -void __init pmac_calibrate_decr(void) -{ - struct device_node *cpu; - unsigned int freq, *fp; - struct div_result divres; - - /* - * The cpu node should have a timebase-frequency property - * to tell us the rate at which the decrementer counts. - */ - cpu = find_type_devices("cpu"); - if (cpu == 0) - panic("can't find cpu node in time_init"); - fp = (unsigned int *) get_property(cpu, "timebase-frequency", NULL); - if (fp == 0) - panic("can't get cpu timebase frequency"); - freq = *fp; - printk("time_init: decrementer frequency = %u.%.6u MHz\n", - freq/1000000, freq%1000000); - tb_ticks_per_jiffy = freq / HZ; - tb_ticks_per_sec = tb_ticks_per_jiffy * HZ; - tb_ticks_per_usec = freq / 1000000; - tb_to_us = mulhwu_scale_factor(freq, 1000000); - div128_by_32( 1024*1024, 0, tb_ticks_per_sec, &divres ); - tb_to_xs = divres.result_low; - ppc_tb_freq = freq; - - fp = (unsigned int *)get_property(cpu, "clock-frequency", NULL); - if (fp == 0) - panic("can't get cpu processor frequency"); - ppc_proc_freq = *fp; - - setup_default_decr(); -} - diff --git a/arch/ppc64/kernel/pmc.c b/arch/ppc64/kernel/pmc.c deleted file mode 100644 index 63d9481c3ec2..000000000000 --- a/arch/ppc64/kernel/pmc.c +++ /dev/null @@ -1,88 +0,0 @@ -/* - * linux/arch/ppc64/kernel/pmc.c - * - * Copyright (C) 2004 David Gibson, IBM Corporation. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/config.h> -#include <linux/errno.h> -#include <linux/spinlock.h> -#include <linux/module.h> - -#include <asm/processor.h> -#include <asm/pmc.h> - -/* Ensure exceptions are disabled */ -static void dummy_perf(struct pt_regs *regs) -{ - unsigned int mmcr0 = mfspr(SPRN_MMCR0); - - mmcr0 &= ~(MMCR0_PMXE|MMCR0_PMAO); - mtspr(SPRN_MMCR0, mmcr0); -} - -static DEFINE_SPINLOCK(pmc_owner_lock); -static void *pmc_owner_caller; /* mostly for debugging */ -perf_irq_t perf_irq = dummy_perf; - -int reserve_pmc_hardware(perf_irq_t new_perf_irq) -{ - int err = 0; - - spin_lock(&pmc_owner_lock); - - if (pmc_owner_caller) { - printk(KERN_WARNING "reserve_pmc_hardware: " - "PMC hardware busy (reserved by caller %p)\n", - pmc_owner_caller); - err = -EBUSY; - goto out; - } - - pmc_owner_caller = __builtin_return_address(0); - perf_irq = new_perf_irq ? : dummy_perf; - - out: - spin_unlock(&pmc_owner_lock); - return err; -} -EXPORT_SYMBOL_GPL(reserve_pmc_hardware); - -void release_pmc_hardware(void) -{ - spin_lock(&pmc_owner_lock); - - WARN_ON(! pmc_owner_caller); - - pmc_owner_caller = NULL; - perf_irq = dummy_perf; - - spin_unlock(&pmc_owner_lock); -} -EXPORT_SYMBOL_GPL(release_pmc_hardware); - -void power4_enable_pmcs(void) -{ - unsigned long hid0; - - hid0 = mfspr(HID0); - hid0 |= 1UL << (63 - 20); - - /* POWER4 requires the following sequence */ - asm volatile( - "sync\n" - "mtspr %1, %0\n" - "mfspr %0, %1\n" - "mfspr %0, %1\n" - "mfspr %0, %1\n" - "mfspr %0, %1\n" - "mfspr %0, %1\n" - "mfspr %0, %1\n" - "isync" : "=&r" (hid0) : "i" (HID0), "0" (hid0): - "memory"); -} diff --git a/arch/ppc64/kernel/ppc_ksyms.c b/arch/ppc64/kernel/ppc_ksyms.c index 705742f4eec6..84006e26342c 100644 --- a/arch/ppc64/kernel/ppc_ksyms.c +++ b/arch/ppc64/kernel/ppc_ksyms.c @@ -19,7 +19,6 @@ #include <asm/hw_irq.h> #include <asm/abs_addr.h> #include <asm/cacheflush.h> -#include <asm/iSeries/HvCallSc.h> EXPORT_SYMBOL(strcpy); EXPORT_SYMBOL(strncpy); @@ -46,17 +45,6 @@ EXPORT_SYMBOL(__strnlen_user); EXPORT_SYMBOL(reloc_offset); -#ifdef CONFIG_PPC_ISERIES -EXPORT_SYMBOL(HvCall0); -EXPORT_SYMBOL(HvCall1); -EXPORT_SYMBOL(HvCall2); -EXPORT_SYMBOL(HvCall3); -EXPORT_SYMBOL(HvCall4); -EXPORT_SYMBOL(HvCall5); -EXPORT_SYMBOL(HvCall6); -EXPORT_SYMBOL(HvCall7); -#endif - EXPORT_SYMBOL(_insb); EXPORT_SYMBOL(_outsb); EXPORT_SYMBOL(_insw); @@ -77,14 +65,6 @@ EXPORT_SYMBOL(giveup_altivec); EXPORT_SYMBOL(__flush_icache_range); EXPORT_SYMBOL(flush_dcache_range); -#ifdef CONFIG_SMP -#ifdef CONFIG_PPC_ISERIES -EXPORT_SYMBOL(local_get_flags); -EXPORT_SYMBOL(local_irq_disable); -EXPORT_SYMBOL(local_irq_restore); -#endif -#endif - EXPORT_SYMBOL(memcpy); EXPORT_SYMBOL(memset); EXPORT_SYMBOL(memmove); diff --git a/arch/ppc64/kernel/process.c b/arch/ppc64/kernel/process.c deleted file mode 100644 index 887005358eb1..000000000000 --- a/arch/ppc64/kernel/process.c +++ /dev/null @@ -1,713 +0,0 @@ -/* - * linux/arch/ppc64/kernel/process.c - * - * Derived from "arch/i386/kernel/process.c" - * Copyright (C) 1995 Linus Torvalds - * - * Updated and modified by Cort Dougan (cort@cs.nmt.edu) and - * Paul Mackerras (paulus@cs.anu.edu.au) - * - * PowerPC version - * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/config.h> -#include <linux/module.h> -#include <linux/errno.h> -#include <linux/sched.h> -#include <linux/kernel.h> -#include <linux/mm.h> -#include <linux/smp.h> -#include <linux/smp_lock.h> -#include <linux/stddef.h> -#include <linux/unistd.h> -#include <linux/slab.h> -#include <linux/user.h> -#include <linux/elf.h> -#include <linux/init.h> -#include <linux/init_task.h> -#include <linux/prctl.h> -#include <linux/ptrace.h> -#include <linux/kallsyms.h> -#include <linux/interrupt.h> -#include <linux/utsname.h> -#include <linux/kprobes.h> - -#include <asm/pgtable.h> -#include <asm/uaccess.h> -#include <asm/system.h> -#include <asm/io.h> -#include <asm/processor.h> -#include <asm/mmu.h> -#include <asm/mmu_context.h> -#include <asm/prom.h> -#include <asm/ppcdebug.h> -#include <asm/machdep.h> -#include <asm/iSeries/HvCallHpt.h> -#include <asm/cputable.h> -#include <asm/firmware.h> -#include <asm/sections.h> -#include <asm/tlbflush.h> -#include <asm/time.h> -#include <asm/plpar_wrappers.h> - -#ifndef CONFIG_SMP -struct task_struct *last_task_used_math = NULL; -struct task_struct *last_task_used_altivec = NULL; -#endif - -/* - * Make sure the floating-point register state in the - * the thread_struct is up to date for task tsk. - */ -void flush_fp_to_thread(struct task_struct *tsk) -{ - if (tsk->thread.regs) { - /* - * We need to disable preemption here because if we didn't, - * another process could get scheduled after the regs->msr - * test but before we have finished saving the FP registers - * to the thread_struct. That process could take over the - * FPU, and then when we get scheduled again we would store - * bogus values for the remaining FP registers. - */ - preempt_disable(); - if (tsk->thread.regs->msr & MSR_FP) { -#ifdef CONFIG_SMP - /* - * This should only ever be called for current or - * for a stopped child process. Since we save away - * the FP register state on context switch on SMP, - * there is something wrong if a stopped child appears - * to still have its FP state in the CPU registers. - */ - BUG_ON(tsk != current); -#endif - giveup_fpu(current); - } - preempt_enable(); - } -} - -void enable_kernel_fp(void) -{ - WARN_ON(preemptible()); - -#ifdef CONFIG_SMP - if (current->thread.regs && (current->thread.regs->msr & MSR_FP)) - giveup_fpu(current); - else - giveup_fpu(NULL); /* just enables FP for kernel */ -#else - giveup_fpu(last_task_used_math); -#endif /* CONFIG_SMP */ -} -EXPORT_SYMBOL(enable_kernel_fp); - -int dump_task_fpu(struct task_struct *tsk, elf_fpregset_t *fpregs) -{ - if (!tsk->thread.regs) - return 0; - flush_fp_to_thread(current); - - memcpy(fpregs, &tsk->thread.fpr[0], sizeof(*fpregs)); - - return 1; -} - -#ifdef CONFIG_ALTIVEC - -void enable_kernel_altivec(void) -{ - WARN_ON(preemptible()); - -#ifdef CONFIG_SMP - if (current->thread.regs && (current->thread.regs->msr & MSR_VEC)) - giveup_altivec(current); - else - giveup_altivec(NULL); /* just enables FP for kernel */ -#else - giveup_altivec(last_task_used_altivec); -#endif /* CONFIG_SMP */ -} -EXPORT_SYMBOL(enable_kernel_altivec); - -/* - * Make sure the VMX/Altivec register state in the - * the thread_struct is up to date for task tsk. - */ -void flush_altivec_to_thread(struct task_struct *tsk) -{ - if (tsk->thread.regs) { - preempt_disable(); - if (tsk->thread.regs->msr & MSR_VEC) { -#ifdef CONFIG_SMP - BUG_ON(tsk != current); -#endif - giveup_altivec(current); - } - preempt_enable(); - } -} - -int dump_task_altivec(struct pt_regs *regs, elf_vrregset_t *vrregs) -{ - flush_altivec_to_thread(current); - memcpy(vrregs, ¤t->thread.vr[0], sizeof(*vrregs)); - return 1; -} - -#endif /* CONFIG_ALTIVEC */ - -static void set_dabr_spr(unsigned long val) -{ - mtspr(SPRN_DABR, val); -} - -int set_dabr(unsigned long dabr) -{ - int ret = 0; - - if (firmware_has_feature(FW_FEATURE_XDABR)) { - /* We want to catch accesses from kernel and userspace */ - unsigned long flags = H_DABRX_KERNEL|H_DABRX_USER; - ret = plpar_set_xdabr(dabr, flags); - } else if (firmware_has_feature(FW_FEATURE_DABR)) { - ret = plpar_set_dabr(dabr); - } else { - set_dabr_spr(dabr); - } - - return ret; -} - -DEFINE_PER_CPU(struct cpu_usage, cpu_usage_array); -static DEFINE_PER_CPU(unsigned long, current_dabr); - -struct task_struct *__switch_to(struct task_struct *prev, - struct task_struct *new) -{ - struct thread_struct *new_thread, *old_thread; - unsigned long flags; - struct task_struct *last; - -#ifdef CONFIG_SMP - /* avoid complexity of lazy save/restore of fpu - * by just saving it every time we switch out if - * this task used the fpu during the last quantum. - * - * If it tries to use the fpu again, it'll trap and - * reload its fp regs. So we don't have to do a restore - * every switch, just a save. - * -- Cort - */ - if (prev->thread.regs && (prev->thread.regs->msr & MSR_FP)) - giveup_fpu(prev); -#ifdef CONFIG_ALTIVEC - if (prev->thread.regs && (prev->thread.regs->msr & MSR_VEC)) - giveup_altivec(prev); -#endif /* CONFIG_ALTIVEC */ -#endif /* CONFIG_SMP */ - -#if defined(CONFIG_ALTIVEC) && !defined(CONFIG_SMP) - /* Avoid the trap. On smp this this never happens since - * we don't set last_task_used_altivec -- Cort - */ - if (new->thread.regs && last_task_used_altivec == new) - new->thread.regs->msr |= MSR_VEC; -#endif /* CONFIG_ALTIVEC */ - - if (unlikely(__get_cpu_var(current_dabr) != new->thread.dabr)) { - set_dabr(new->thread.dabr); - __get_cpu_var(current_dabr) = new->thread.dabr; - } - - flush_tlb_pending(); - - new_thread = &new->thread; - old_thread = ¤t->thread; - - /* Collect purr utilization data per process and per processor - * wise purr is nothing but processor time base - */ - if (firmware_has_feature(FW_FEATURE_SPLPAR)) { - struct cpu_usage *cu = &__get_cpu_var(cpu_usage_array); - long unsigned start_tb, current_tb; - start_tb = old_thread->start_tb; - cu->current_tb = current_tb = mfspr(SPRN_PURR); - old_thread->accum_tb += (current_tb - start_tb); - new_thread->start_tb = current_tb; - } - - local_irq_save(flags); - last = _switch(old_thread, new_thread); - - local_irq_restore(flags); - - return last; -} - -static int instructions_to_print = 16; - -static void show_instructions(struct pt_regs *regs) -{ - int i; - unsigned long pc = regs->nip - (instructions_to_print * 3 / 4 * - sizeof(int)); - - printk("Instruction dump:"); - - for (i = 0; i < instructions_to_print; i++) { - int instr; - - if (!(i % 8)) - printk("\n"); - - if (((REGION_ID(pc) != KERNEL_REGION_ID) && - (REGION_ID(pc) != VMALLOC_REGION_ID)) || - __get_user(instr, (unsigned int *)pc)) { - printk("XXXXXXXX "); - } else { - if (regs->nip == pc) - printk("<%08x> ", instr); - else - printk("%08x ", instr); - } - - pc += sizeof(int); - } - - printk("\n"); -} - -void show_regs(struct pt_regs * regs) -{ - int i; - unsigned long trap; - - printk("NIP: %016lX XER: %08X LR: %016lX CTR: %016lX\n", - regs->nip, (unsigned int)regs->xer, regs->link, regs->ctr); - printk("REGS: %p TRAP: %04lx %s (%s)\n", - regs, regs->trap, print_tainted(), system_utsname.release); - printk("MSR: %016lx EE: %01x PR: %01x FP: %01x ME: %01x " - "IR/DR: %01x%01x CR: %08X\n", - regs->msr, regs->msr&MSR_EE ? 1 : 0, regs->msr&MSR_PR ? 1 : 0, - regs->msr & MSR_FP ? 1 : 0,regs->msr&MSR_ME ? 1 : 0, - regs->msr&MSR_IR ? 1 : 0, - regs->msr&MSR_DR ? 1 : 0, - (unsigned int)regs->ccr); - trap = TRAP(regs); - printk("DAR: %016lx DSISR: %016lx\n", regs->dar, regs->dsisr); - printk("TASK: %p[%d] '%s' THREAD: %p", - current, current->pid, current->comm, current->thread_info); - -#ifdef CONFIG_SMP - printk(" CPU: %d", smp_processor_id()); -#endif /* CONFIG_SMP */ - - for (i = 0; i < 32; i++) { - if ((i % 4) == 0) { - printk("\n" KERN_INFO "GPR%02d: ", i); - } - - printk("%016lX ", regs->gpr[i]); - if (i == 13 && !FULL_REGS(regs)) - break; - } - printk("\n"); - /* - * Lookup NIP late so we have the best change of getting the - * above info out without failing - */ - printk("NIP [%016lx] ", regs->nip); - print_symbol("%s\n", regs->nip); - printk("LR [%016lx] ", regs->link); - print_symbol("%s\n", regs->link); - show_stack(current, (unsigned long *)regs->gpr[1]); - if (!user_mode(regs)) - show_instructions(regs); -} - -void exit_thread(void) -{ - kprobe_flush_task(current); - -#ifndef CONFIG_SMP - if (last_task_used_math == current) - last_task_used_math = NULL; -#ifdef CONFIG_ALTIVEC - if (last_task_used_altivec == current) - last_task_used_altivec = NULL; -#endif /* CONFIG_ALTIVEC */ -#endif /* CONFIG_SMP */ -} - -void flush_thread(void) -{ - struct thread_info *t = current_thread_info(); - - kprobe_flush_task(current); - if (t->flags & _TIF_ABI_PENDING) - t->flags ^= (_TIF_ABI_PENDING | _TIF_32BIT); - -#ifndef CONFIG_SMP - if (last_task_used_math == current) - last_task_used_math = NULL; -#ifdef CONFIG_ALTIVEC - if (last_task_used_altivec == current) - last_task_used_altivec = NULL; -#endif /* CONFIG_ALTIVEC */ -#endif /* CONFIG_SMP */ - - if (current->thread.dabr) { - current->thread.dabr = 0; - set_dabr(0); - } -} - -void -release_thread(struct task_struct *t) -{ -} - - -/* - * This gets called before we allocate a new thread and copy - * the current task into it. - */ -void prepare_to_copy(struct task_struct *tsk) -{ - flush_fp_to_thread(current); - flush_altivec_to_thread(current); -} - -/* - * Copy a thread.. - */ -int -copy_thread(int nr, unsigned long clone_flags, unsigned long usp, - unsigned long unused, struct task_struct *p, struct pt_regs *regs) -{ - struct pt_regs *childregs, *kregs; - extern void ret_from_fork(void); - unsigned long sp = (unsigned long)p->thread_info + THREAD_SIZE; - - /* Copy registers */ - sp -= sizeof(struct pt_regs); - childregs = (struct pt_regs *) sp; - *childregs = *regs; - if ((childregs->msr & MSR_PR) == 0) { - /* for kernel thread, set stackptr in new task */ - childregs->gpr[1] = sp + sizeof(struct pt_regs); - p->thread.regs = NULL; /* no user register state */ - clear_ti_thread_flag(p->thread_info, TIF_32BIT); - } else { - childregs->gpr[1] = usp; - p->thread.regs = childregs; - if (clone_flags & CLONE_SETTLS) { - if (test_thread_flag(TIF_32BIT)) - childregs->gpr[2] = childregs->gpr[6]; - else - childregs->gpr[13] = childregs->gpr[6]; - } - } - childregs->gpr[3] = 0; /* Result from fork() */ - sp -= STACK_FRAME_OVERHEAD; - - /* - * The way this works is that at some point in the future - * some task will call _switch to switch to the new task. - * That will pop off the stack frame created below and start - * the new task running at ret_from_fork. The new task will - * do some house keeping and then return from the fork or clone - * system call, using the stack frame created above. - */ - sp -= sizeof(struct pt_regs); - kregs = (struct pt_regs *) sp; - sp -= STACK_FRAME_OVERHEAD; - p->thread.ksp = sp; - if (cpu_has_feature(CPU_FTR_SLB)) { - unsigned long sp_vsid = get_kernel_vsid(sp); - - sp_vsid <<= SLB_VSID_SHIFT; - sp_vsid |= SLB_VSID_KERNEL; - if (cpu_has_feature(CPU_FTR_16M_PAGE)) - sp_vsid |= SLB_VSID_L; - - p->thread.ksp_vsid = sp_vsid; - } - - /* - * The PPC64 ABI makes use of a TOC to contain function - * pointers. The function (ret_from_except) is actually a pointer - * to the TOC entry. The first entry is a pointer to the actual - * function. - */ - kregs->nip = *((unsigned long *)ret_from_fork); - - return 0; -} - -/* - * Set up a thread for executing a new program - */ -void start_thread(struct pt_regs *regs, unsigned long fdptr, unsigned long sp) -{ - unsigned long entry, toc, load_addr = regs->gpr[2]; - - /* fdptr is a relocated pointer to the function descriptor for - * the elf _start routine. The first entry in the function - * descriptor is the entry address of _start and the second - * entry is the TOC value we need to use. - */ - set_fs(USER_DS); - __get_user(entry, (unsigned long __user *)fdptr); - __get_user(toc, (unsigned long __user *)fdptr+1); - - /* Check whether the e_entry function descriptor entries - * need to be relocated before we can use them. - */ - if (load_addr != 0) { - entry += load_addr; - toc += load_addr; - } - - /* - * If we exec out of a kernel thread then thread.regs will not be - * set. Do it now. - */ - if (!current->thread.regs) { - unsigned long childregs = (unsigned long)current->thread_info + - THREAD_SIZE; - childregs -= sizeof(struct pt_regs); - current->thread.regs = (struct pt_regs *)childregs; - } - - regs->nip = entry; - regs->gpr[1] = sp; - regs->gpr[2] = toc; - regs->msr = MSR_USER64; -#ifndef CONFIG_SMP - if (last_task_used_math == current) - last_task_used_math = 0; -#endif /* CONFIG_SMP */ - memset(current->thread.fpr, 0, sizeof(current->thread.fpr)); - current->thread.fpscr = 0; -#ifdef CONFIG_ALTIVEC -#ifndef CONFIG_SMP - if (last_task_used_altivec == current) - last_task_used_altivec = 0; -#endif /* CONFIG_SMP */ - memset(current->thread.vr, 0, sizeof(current->thread.vr)); - current->thread.vscr.u[0] = 0; - current->thread.vscr.u[1] = 0; - current->thread.vscr.u[2] = 0; - current->thread.vscr.u[3] = 0x00010000; /* Java mode disabled */ - current->thread.vrsave = 0; - current->thread.used_vr = 0; -#endif /* CONFIG_ALTIVEC */ -} -EXPORT_SYMBOL(start_thread); - -int set_fpexc_mode(struct task_struct *tsk, unsigned int val) -{ - struct pt_regs *regs = tsk->thread.regs; - - if (val > PR_FP_EXC_PRECISE) - return -EINVAL; - tsk->thread.fpexc_mode = __pack_fe01(val); - if (regs != NULL && (regs->msr & MSR_FP) != 0) - regs->msr = (regs->msr & ~(MSR_FE0|MSR_FE1)) - | tsk->thread.fpexc_mode; - return 0; -} - -int get_fpexc_mode(struct task_struct *tsk, unsigned long adr) -{ - unsigned int val; - - val = __unpack_fe01(tsk->thread.fpexc_mode); - return put_user(val, (unsigned int __user *) adr); -} - -int sys_clone(unsigned long clone_flags, unsigned long p2, unsigned long p3, - unsigned long p4, unsigned long p5, unsigned long p6, - struct pt_regs *regs) -{ - unsigned long parent_tidptr = 0; - unsigned long child_tidptr = 0; - - if (p2 == 0) - p2 = regs->gpr[1]; /* stack pointer for child */ - - if (clone_flags & (CLONE_PARENT_SETTID | CLONE_CHILD_SETTID | - CLONE_CHILD_CLEARTID)) { - parent_tidptr = p3; - child_tidptr = p5; - if (test_thread_flag(TIF_32BIT)) { - parent_tidptr &= 0xffffffff; - child_tidptr &= 0xffffffff; - } - } - - return do_fork(clone_flags, p2, regs, 0, - (int __user *)parent_tidptr, (int __user *)child_tidptr); -} - -int sys_fork(unsigned long p1, unsigned long p2, unsigned long p3, - unsigned long p4, unsigned long p5, unsigned long p6, - struct pt_regs *regs) -{ - return do_fork(SIGCHLD, regs->gpr[1], regs, 0, NULL, NULL); -} - -int sys_vfork(unsigned long p1, unsigned long p2, unsigned long p3, - unsigned long p4, unsigned long p5, unsigned long p6, - struct pt_regs *regs) -{ - return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs->gpr[1], regs, 0, - NULL, NULL); -} - -int sys_execve(unsigned long a0, unsigned long a1, unsigned long a2, - unsigned long a3, unsigned long a4, unsigned long a5, - struct pt_regs *regs) -{ - int error; - char * filename; - - filename = getname((char __user *) a0); - error = PTR_ERR(filename); - if (IS_ERR(filename)) - goto out; - flush_fp_to_thread(current); - flush_altivec_to_thread(current); - error = do_execve(filename, (char __user * __user *) a1, - (char __user * __user *) a2, regs); - - if (error == 0) { - task_lock(current); - current->ptrace &= ~PT_DTRACE; - task_unlock(current); - } - putname(filename); - -out: - return error; -} - -static int kstack_depth_to_print = 64; - -static int validate_sp(unsigned long sp, struct task_struct *p, - unsigned long nbytes) -{ - unsigned long stack_page = (unsigned long)p->thread_info; - - if (sp >= stack_page + sizeof(struct thread_struct) - && sp <= stack_page + THREAD_SIZE - nbytes) - return 1; - -#ifdef CONFIG_IRQSTACKS - stack_page = (unsigned long) hardirq_ctx[task_cpu(p)]; - if (sp >= stack_page + sizeof(struct thread_struct) - && sp <= stack_page + THREAD_SIZE - nbytes) - return 1; - - stack_page = (unsigned long) softirq_ctx[task_cpu(p)]; - if (sp >= stack_page + sizeof(struct thread_struct) - && sp <= stack_page + THREAD_SIZE - nbytes) - return 1; -#endif - - return 0; -} - -unsigned long get_wchan(struct task_struct *p) -{ - unsigned long ip, sp; - int count = 0; - - if (!p || p == current || p->state == TASK_RUNNING) - return 0; - - sp = p->thread.ksp; - if (!validate_sp(sp, p, 112)) - return 0; - - do { - sp = *(unsigned long *)sp; - if (!validate_sp(sp, p, 112)) - return 0; - if (count > 0) { - ip = *(unsigned long *)(sp + 16); - if (!in_sched_functions(ip)) - return ip; - } - } while (count++ < 16); - return 0; -} -EXPORT_SYMBOL(get_wchan); - -void show_stack(struct task_struct *p, unsigned long *_sp) -{ - unsigned long ip, newsp, lr; - int count = 0; - unsigned long sp = (unsigned long)_sp; - int firstframe = 1; - - if (sp == 0) { - if (p) { - sp = p->thread.ksp; - } else { - sp = __get_SP(); - p = current; - } - } - - lr = 0; - printk("Call Trace:\n"); - do { - if (!validate_sp(sp, p, 112)) - return; - - _sp = (unsigned long *) sp; - newsp = _sp[0]; - ip = _sp[2]; - if (!firstframe || ip != lr) { - printk("[%016lx] [%016lx] ", sp, ip); - print_symbol("%s", ip); - if (firstframe) - printk(" (unreliable)"); - printk("\n"); - } - firstframe = 0; - - /* - * See if this is an exception frame. - * We look for the "regshere" marker in the current frame. - */ - if (validate_sp(sp, p, sizeof(struct pt_regs) + 400) - && _sp[12] == 0x7265677368657265ul) { - struct pt_regs *regs = (struct pt_regs *) - (sp + STACK_FRAME_OVERHEAD); - printk("--- Exception: %lx", regs->trap); - print_symbol(" at %s\n", regs->nip); - lr = regs->link; - print_symbol(" LR = %s\n", lr); - firstframe = 1; - } - - sp = newsp; - } while (count++ < kstack_depth_to_print); -} - -void dump_stack(void) -{ - show_stack(current, (unsigned long *)__get_SP()); -} -EXPORT_SYMBOL(dump_stack); diff --git a/arch/ppc64/kernel/prom.c b/arch/ppc64/kernel/prom.c index 7035deb6de92..a0866f12647f 100644 --- a/arch/ppc64/kernel/prom.c +++ b/arch/ppc64/kernel/prom.c @@ -78,6 +78,7 @@ typedef int interpret_func(struct device_node *, unsigned long *, extern struct rtas_t rtas; extern struct lmb lmb; extern unsigned long klimit; +extern unsigned long memory_limit; static int __initdata dt_root_addr_cells; static int __initdata dt_root_size_cells; @@ -1063,7 +1064,7 @@ static int __init early_init_dt_scan_chosen(unsigned long node, { u32 *prop; u64 *prop64; - extern unsigned long memory_limit, tce_alloc_start, tce_alloc_end; + extern unsigned long tce_alloc_start, tce_alloc_end; DBG("search \"chosen\", depth: %d, uname: %s\n", depth, uname); @@ -1237,7 +1238,7 @@ void __init early_init_devtree(void *params) lmb_init(); scan_flat_dt(early_init_dt_scan_root, NULL); scan_flat_dt(early_init_dt_scan_memory, NULL); - lmb_enforce_memory_limit(); + lmb_enforce_memory_limit(memory_limit); lmb_analyze(); systemcfg->physicalMemorySize = lmb_phys_mem_size(); lmb_reserve(0, __pa(klimit)); diff --git a/arch/ppc64/kernel/ptrace.c b/arch/ppc64/kernel/ptrace.c deleted file mode 100644 index b1c044ca5756..000000000000 --- a/arch/ppc64/kernel/ptrace.c +++ /dev/null @@ -1,363 +0,0 @@ -/* - * linux/arch/ppc64/kernel/ptrace.c - * - * PowerPC version - * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org) - * - * Derived from "arch/m68k/kernel/ptrace.c" - * Copyright (C) 1994 by Hamish Macdonald - * Taken from linux/kernel/ptrace.c and modified for M680x0. - * linux/kernel/ptrace.c is by Ross Biro 1/23/92, edited by Linus Torvalds - * - * Modified by Cort Dougan (cort@hq.fsmlabs.com) - * and Paul Mackerras (paulus@linuxcare.com.au). - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file README.legal in the main directory of - * this archive for more details. - */ - -#include <linux/config.h> -#include <linux/kernel.h> -#include <linux/sched.h> -#include <linux/mm.h> -#include <linux/smp.h> -#include <linux/smp_lock.h> -#include <linux/errno.h> -#include <linux/ptrace.h> -#include <linux/user.h> -#include <linux/security.h> -#include <linux/audit.h> -#include <linux/seccomp.h> -#include <linux/signal.h> - -#include <asm/uaccess.h> -#include <asm/page.h> -#include <asm/pgtable.h> -#include <asm/system.h> -#include <asm/ptrace-common.h> - -/* - * does not yet catch signals sent when the child dies. - * in exit.c or in signal.c. - */ - -/* - * Called by kernel/ptrace.c when detaching.. - * - * Make sure single step bits etc are not set. - */ -void ptrace_disable(struct task_struct *child) -{ - /* make sure the single step bit is not set. */ - clear_single_step(child); -} - -int sys_ptrace(long request, long pid, long addr, long data) -{ - struct task_struct *child; - int ret = -EPERM; - - lock_kernel(); - if (request == PTRACE_TRACEME) { - /* are we already being traced? */ - if (current->ptrace & PT_PTRACED) - goto out; - ret = security_ptrace(current->parent, current); - if (ret) - goto out; - /* set the ptrace bit in the process flags. */ - current->ptrace |= PT_PTRACED; - ret = 0; - goto out; - } - ret = -ESRCH; - read_lock(&tasklist_lock); - child = find_task_by_pid(pid); - if (child) - get_task_struct(child); - read_unlock(&tasklist_lock); - if (!child) - goto out; - - ret = -EPERM; - if (pid == 1) /* you may not mess with init */ - goto out_tsk; - - if (request == PTRACE_ATTACH) { - ret = ptrace_attach(child); - goto out_tsk; - } - - ret = ptrace_check_attach(child, request == PTRACE_KILL); - if (ret < 0) - goto out_tsk; - - switch (request) { - /* when I and D space are separate, these will need to be fixed. */ - case PTRACE_PEEKTEXT: /* read word at location addr. */ - case PTRACE_PEEKDATA: { - unsigned long tmp; - int copied; - - copied = access_process_vm(child, addr, &tmp, sizeof(tmp), 0); - ret = -EIO; - if (copied != sizeof(tmp)) - break; - ret = put_user(tmp,(unsigned long __user *) data); - break; - } - - /* read the word at location addr in the USER area. */ - case PTRACE_PEEKUSR: { - unsigned long index; - unsigned long tmp; - - ret = -EIO; - /* convert to index and check */ - index = (unsigned long) addr >> 3; - if ((addr & 7) || (index > PT_FPSCR)) - break; - - if (index < PT_FPR0) { - tmp = get_reg(child, (int)index); - } else { - flush_fp_to_thread(child); - tmp = ((unsigned long *)child->thread.fpr)[index - PT_FPR0]; - } - ret = put_user(tmp,(unsigned long __user *) data); - break; - } - - /* If I and D space are separate, this will have to be fixed. */ - case PTRACE_POKETEXT: /* write the word at location addr. */ - case PTRACE_POKEDATA: - ret = 0; - if (access_process_vm(child, addr, &data, sizeof(data), 1) - == sizeof(data)) - break; - ret = -EIO; - break; - - /* write the word at location addr in the USER area */ - case PTRACE_POKEUSR: { - unsigned long index; - - ret = -EIO; - /* convert to index and check */ - index = (unsigned long) addr >> 3; - if ((addr & 7) || (index > PT_FPSCR)) - break; - - if (index == PT_ORIG_R3) - break; - if (index < PT_FPR0) { - ret = put_reg(child, index, data); - } else { - flush_fp_to_thread(child); - ((unsigned long *)child->thread.fpr)[index - PT_FPR0] = data; - ret = 0; - } - break; - } - - case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ - case PTRACE_CONT: { /* restart after signal. */ - ret = -EIO; - if (!valid_signal(data)) - break; - if (request == PTRACE_SYSCALL) - set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - else - clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - child->exit_code = data; - /* make sure the single step bit is not set. */ - clear_single_step(child); - wake_up_process(child); - ret = 0; - break; - } - - /* - * make the child exit. Best I can do is send it a sigkill. - * perhaps it should be put in the status that it wants to - * exit. - */ - case PTRACE_KILL: { - ret = 0; - if (child->exit_state == EXIT_ZOMBIE) /* already dead */ - break; - child->exit_code = SIGKILL; - /* make sure the single step bit is not set. */ - clear_single_step(child); - wake_up_process(child); - break; - } - - case PTRACE_SINGLESTEP: { /* set the trap flag. */ - ret = -EIO; - if (!valid_signal(data)) - break; - clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - set_single_step(child); - child->exit_code = data; - /* give it a chance to run. */ - wake_up_process(child); - ret = 0; - break; - } - - case PTRACE_GET_DEBUGREG: { - ret = -EINVAL; - /* We only support one DABR and no IABRS at the moment */ - if (addr > 0) - break; - ret = put_user(child->thread.dabr, - (unsigned long __user *)data); - break; - } - - case PTRACE_SET_DEBUGREG: - ret = ptrace_set_debugreg(child, addr, data); - break; - - case PTRACE_DETACH: - ret = ptrace_detach(child, data); - break; - - case PPC_PTRACE_GETREGS: { /* Get GPRs 0 - 31. */ - int i; - unsigned long *reg = &((unsigned long *)child->thread.regs)[0]; - unsigned long __user *tmp = (unsigned long __user *)addr; - - for (i = 0; i < 32; i++) { - ret = put_user(*reg, tmp); - if (ret) - break; - reg++; - tmp++; - } - break; - } - - case PPC_PTRACE_SETREGS: { /* Set GPRs 0 - 31. */ - int i; - unsigned long *reg = &((unsigned long *)child->thread.regs)[0]; - unsigned long __user *tmp = (unsigned long __user *)addr; - - for (i = 0; i < 32; i++) { - ret = get_user(*reg, tmp); - if (ret) - break; - reg++; - tmp++; - } - break; - } - - case PPC_PTRACE_GETFPREGS: { /* Get FPRs 0 - 31. */ - int i; - unsigned long *reg = &((unsigned long *)child->thread.fpr)[0]; - unsigned long __user *tmp = (unsigned long __user *)addr; - - flush_fp_to_thread(child); - - for (i = 0; i < 32; i++) { - ret = put_user(*reg, tmp); - if (ret) - break; - reg++; - tmp++; - } - break; - } - - case PPC_PTRACE_SETFPREGS: { /* Get FPRs 0 - 31. */ - int i; - unsigned long *reg = &((unsigned long *)child->thread.fpr)[0]; - unsigned long __user *tmp = (unsigned long __user *)addr; - - flush_fp_to_thread(child); - - for (i = 0; i < 32; i++) { - ret = get_user(*reg, tmp); - if (ret) - break; - reg++; - tmp++; - } - break; - } - -#ifdef CONFIG_ALTIVEC - case PTRACE_GETVRREGS: - /* Get the child altivec register state. */ - flush_altivec_to_thread(child); - ret = get_vrregs((unsigned long __user *)data, child); - break; - - case PTRACE_SETVRREGS: - /* Set the child altivec register state. */ - flush_altivec_to_thread(child); - ret = set_vrregs(child, (unsigned long __user *)data); - break; -#endif - - default: - ret = ptrace_request(child, request, addr, data); - break; - } -out_tsk: - put_task_struct(child); -out: - unlock_kernel(); - return ret; -} - -static void do_syscall_trace(void) -{ - /* the 0x80 provides a way for the tracing parent to distinguish - between a syscall stop and SIGTRAP delivery */ - ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD) - ? 0x80 : 0)); - - /* - * this isn't the same as continuing with a signal, but it will do - * for normal use. strace only continues with a signal if the - * stopping signal is not SIGTRAP. -brl - */ - if (current->exit_code) { - send_sig(current->exit_code, current, 1); - current->exit_code = 0; - } -} - -void do_syscall_trace_enter(struct pt_regs *regs) -{ - secure_computing(regs->gpr[0]); - - if (test_thread_flag(TIF_SYSCALL_TRACE) - && (current->ptrace & PT_PTRACED)) - do_syscall_trace(); - - if (unlikely(current->audit_context)) - audit_syscall_entry(current, - test_thread_flag(TIF_32BIT)?AUDIT_ARCH_PPC:AUDIT_ARCH_PPC64, - regs->gpr[0], - regs->gpr[3], regs->gpr[4], - regs->gpr[5], regs->gpr[6]); - -} - -void do_syscall_trace_leave(struct pt_regs *regs) -{ - if (unlikely(current->audit_context)) - audit_syscall_exit(current, - (regs->ccr&0x1000)?AUDITSC_FAILURE:AUDITSC_SUCCESS, - regs->result); - - if ((test_thread_flag(TIF_SYSCALL_TRACE) - || test_thread_flag(TIF_SINGLESTEP)) - && (current->ptrace & PT_PTRACED)) - do_syscall_trace(); -} diff --git a/arch/ppc64/kernel/ptrace32.c b/arch/ppc64/kernel/ptrace32.c deleted file mode 100644 index fb8c22d6084a..000000000000 --- a/arch/ppc64/kernel/ptrace32.c +++ /dev/null @@ -1,449 +0,0 @@ -/* - * linux/arch/ppc64/kernel/ptrace32.c - * - * PowerPC version - * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org) - * - * Derived from "arch/m68k/kernel/ptrace.c" - * Copyright (C) 1994 by Hamish Macdonald - * Taken from linux/kernel/ptrace.c and modified for M680x0. - * linux/kernel/ptrace.c is by Ross Biro 1/23/92, edited by Linus Torvalds - * - * Modified by Cort Dougan (cort@hq.fsmlabs.com) - * and Paul Mackerras (paulus@linuxcare.com.au). - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file README.legal in the main directory of - * this archive for more details. - */ - -#include <linux/config.h> -#include <linux/kernel.h> -#include <linux/sched.h> -#include <linux/mm.h> -#include <linux/smp.h> -#include <linux/smp_lock.h> -#include <linux/errno.h> -#include <linux/ptrace.h> -#include <linux/user.h> -#include <linux/security.h> -#include <linux/signal.h> - -#include <asm/uaccess.h> -#include <asm/page.h> -#include <asm/pgtable.h> -#include <asm/system.h> -#include <asm/ptrace-common.h> - -/* - * does not yet catch signals sent when the child dies. - * in exit.c or in signal.c. - */ - -int sys32_ptrace(long request, long pid, unsigned long addr, unsigned long data) -{ - struct task_struct *child; - int ret = -EPERM; - - lock_kernel(); - if (request == PTRACE_TRACEME) { - /* are we already being traced? */ - if (current->ptrace & PT_PTRACED) - goto out; - ret = security_ptrace(current->parent, current); - if (ret) - goto out; - /* set the ptrace bit in the process flags. */ - current->ptrace |= PT_PTRACED; - ret = 0; - goto out; - } - ret = -ESRCH; - read_lock(&tasklist_lock); - child = find_task_by_pid(pid); - if (child) - get_task_struct(child); - read_unlock(&tasklist_lock); - if (!child) - goto out; - - ret = -EPERM; - if (pid == 1) /* you may not mess with init */ - goto out_tsk; - - if (request == PTRACE_ATTACH) { - ret = ptrace_attach(child); - goto out_tsk; - } - - ret = ptrace_check_attach(child, request == PTRACE_KILL); - if (ret < 0) - goto out_tsk; - - switch (request) { - /* when I and D space are separate, these will need to be fixed. */ - case PTRACE_PEEKTEXT: /* read word at location addr. */ - case PTRACE_PEEKDATA: { - unsigned int tmp; - int copied; - - copied = access_process_vm(child, addr, &tmp, sizeof(tmp), 0); - ret = -EIO; - if (copied != sizeof(tmp)) - break; - ret = put_user(tmp, (u32 __user *)data); - break; - } - - /* - * Read 4 bytes of the other process' storage - * data is a pointer specifying where the user wants the - * 4 bytes copied into - * addr is a pointer in the user's storage that contains an 8 byte - * address in the other process of the 4 bytes that is to be read - * (this is run in a 32-bit process looking at a 64-bit process) - * when I and D space are separate, these will need to be fixed. - */ - case PPC_PTRACE_PEEKTEXT_3264: - case PPC_PTRACE_PEEKDATA_3264: { - u32 tmp; - int copied; - u32 __user * addrOthers; - - ret = -EIO; - - /* Get the addr in the other process that we want to read */ - if (get_user(addrOthers, (u32 __user * __user *)addr) != 0) - break; - - copied = access_process_vm(child, (u64)addrOthers, &tmp, - sizeof(tmp), 0); - if (copied != sizeof(tmp)) - break; - ret = put_user(tmp, (u32 __user *)data); - break; - } - - /* Read a register (specified by ADDR) out of the "user area" */ - case PTRACE_PEEKUSR: { - int index; - unsigned long tmp; - - ret = -EIO; - /* convert to index and check */ - index = (unsigned long) addr >> 2; - if ((addr & 3) || (index > PT_FPSCR32)) - break; - - if (index < PT_FPR0) { - tmp = get_reg(child, index); - } else { - flush_fp_to_thread(child); - /* - * the user space code considers the floating point - * to be an array of unsigned int (32 bits) - the - * index passed in is based on this assumption. - */ - tmp = ((unsigned int *)child->thread.fpr)[index - PT_FPR0]; - } - ret = put_user((unsigned int)tmp, (u32 __user *)data); - break; - } - - /* - * Read 4 bytes out of the other process' pt_regs area - * data is a pointer specifying where the user wants the - * 4 bytes copied into - * addr is the offset into the other process' pt_regs structure - * that is to be read - * (this is run in a 32-bit process looking at a 64-bit process) - */ - case PPC_PTRACE_PEEKUSR_3264: { - u32 index; - u32 reg32bits; - u64 tmp; - u32 numReg; - u32 part; - - ret = -EIO; - /* Determine which register the user wants */ - index = (u64)addr >> 2; - numReg = index / 2; - /* Determine which part of the register the user wants */ - if (index % 2) - part = 1; /* want the 2nd half of the register (right-most). */ - else - part = 0; /* want the 1st half of the register (left-most). */ - - /* Validate the input - check to see if address is on the wrong boundary or beyond the end of the user area */ - if ((addr & 3) || numReg > PT_FPSCR) - break; - - if (numReg >= PT_FPR0) { - flush_fp_to_thread(child); - tmp = ((unsigned long int *)child->thread.fpr)[numReg - PT_FPR0]; - } else { /* register within PT_REGS struct */ - tmp = get_reg(child, numReg); - } - reg32bits = ((u32*)&tmp)[part]; - ret = put_user(reg32bits, (u32 __user *)data); - break; - } - - /* If I and D space are separate, this will have to be fixed. */ - case PTRACE_POKETEXT: /* write the word at location addr. */ - case PTRACE_POKEDATA: { - unsigned int tmp; - tmp = data; - ret = 0; - if (access_process_vm(child, addr, &tmp, sizeof(tmp), 1) - == sizeof(tmp)) - break; - ret = -EIO; - break; - } - - /* - * Write 4 bytes into the other process' storage - * data is the 4 bytes that the user wants written - * addr is a pointer in the user's storage that contains an - * 8 byte address in the other process where the 4 bytes - * that is to be written - * (this is run in a 32-bit process looking at a 64-bit process) - * when I and D space are separate, these will need to be fixed. - */ - case PPC_PTRACE_POKETEXT_3264: - case PPC_PTRACE_POKEDATA_3264: { - u32 tmp = data; - u32 __user * addrOthers; - - /* Get the addr in the other process that we want to write into */ - ret = -EIO; - if (get_user(addrOthers, (u32 __user * __user *)addr) != 0) - break; - ret = 0; - if (access_process_vm(child, (u64)addrOthers, &tmp, - sizeof(tmp), 1) == sizeof(tmp)) - break; - ret = -EIO; - break; - } - - /* write the word at location addr in the USER area */ - case PTRACE_POKEUSR: { - unsigned long index; - - ret = -EIO; - /* convert to index and check */ - index = (unsigned long) addr >> 2; - if ((addr & 3) || (index > PT_FPSCR32)) - break; - - if (index == PT_ORIG_R3) - break; - if (index < PT_FPR0) { - ret = put_reg(child, index, data); - } else { - flush_fp_to_thread(child); - /* - * the user space code considers the floating point - * to be an array of unsigned int (32 bits) - the - * index passed in is based on this assumption. - */ - ((unsigned int *)child->thread.fpr)[index - PT_FPR0] = data; - ret = 0; - } - break; - } - - /* - * Write 4 bytes into the other process' pt_regs area - * data is the 4 bytes that the user wants written - * addr is the offset into the other process' pt_regs structure - * that is to be written into - * (this is run in a 32-bit process looking at a 64-bit process) - */ - case PPC_PTRACE_POKEUSR_3264: { - u32 index; - u32 numReg; - - ret = -EIO; - /* Determine which register the user wants */ - index = (u64)addr >> 2; - numReg = index / 2; - /* - * Validate the input - check to see if address is on the - * wrong boundary or beyond the end of the user area - */ - if ((addr & 3) || (numReg > PT_FPSCR)) - break; - /* Insure it is a register we let them change */ - if ((numReg == PT_ORIG_R3) - || ((numReg > PT_CCR) && (numReg < PT_FPR0))) - break; - if (numReg >= PT_FPR0) { - flush_fp_to_thread(child); - } - if (numReg == PT_MSR) - data = (data & MSR_DEBUGCHANGE) - | (child->thread.regs->msr & ~MSR_DEBUGCHANGE); - ((u32*)child->thread.regs)[index] = data; - ret = 0; - break; - } - - case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ - case PTRACE_CONT: { /* restart after signal. */ - ret = -EIO; - if (!valid_signal(data)) - break; - if (request == PTRACE_SYSCALL) - set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - else - clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - child->exit_code = data; - /* make sure the single step bit is not set. */ - clear_single_step(child); - wake_up_process(child); - ret = 0; - break; - } - - /* - * make the child exit. Best I can do is send it a sigkill. - * perhaps it should be put in the status that it wants to - * exit. - */ - case PTRACE_KILL: { - ret = 0; - if (child->exit_state == EXIT_ZOMBIE) /* already dead */ - break; - child->exit_code = SIGKILL; - /* make sure the single step bit is not set. */ - clear_single_step(child); - wake_up_process(child); - break; - } - - case PTRACE_SINGLESTEP: { /* set the trap flag. */ - ret = -EIO; - if (!valid_signal(data)) - break; - clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - set_single_step(child); - child->exit_code = data; - /* give it a chance to run. */ - wake_up_process(child); - ret = 0; - break; - } - - case PTRACE_GET_DEBUGREG: { - ret = -EINVAL; - /* We only support one DABR and no IABRS at the moment */ - if (addr > 0) - break; - ret = put_user(child->thread.dabr, (u32 __user *)data); - break; - } - - case PTRACE_SET_DEBUGREG: - ret = ptrace_set_debugreg(child, addr, data); - break; - - case PTRACE_DETACH: - ret = ptrace_detach(child, data); - break; - - case PPC_PTRACE_GETREGS: { /* Get GPRs 0 - 31. */ - int i; - unsigned long *reg = &((unsigned long *)child->thread.regs)[0]; - unsigned int __user *tmp = (unsigned int __user *)addr; - - for (i = 0; i < 32; i++) { - ret = put_user(*reg, tmp); - if (ret) - break; - reg++; - tmp++; - } - break; - } - - case PPC_PTRACE_SETREGS: { /* Set GPRs 0 - 31. */ - int i; - unsigned long *reg = &((unsigned long *)child->thread.regs)[0]; - unsigned int __user *tmp = (unsigned int __user *)addr; - - for (i = 0; i < 32; i++) { - ret = get_user(*reg, tmp); - if (ret) - break; - reg++; - tmp++; - } - break; - } - - case PPC_PTRACE_GETFPREGS: { /* Get FPRs 0 - 31. */ - int i; - unsigned long *reg = &((unsigned long *)child->thread.fpr)[0]; - unsigned int __user *tmp = (unsigned int __user *)addr; - - flush_fp_to_thread(child); - - for (i = 0; i < 32; i++) { - ret = put_user(*reg, tmp); - if (ret) - break; - reg++; - tmp++; - } - break; - } - - case PPC_PTRACE_SETFPREGS: { /* Get FPRs 0 - 31. */ - int i; - unsigned long *reg = &((unsigned long *)child->thread.fpr)[0]; - unsigned int __user *tmp = (unsigned int __user *)addr; - - flush_fp_to_thread(child); - - for (i = 0; i < 32; i++) { - ret = get_user(*reg, tmp); - if (ret) - break; - reg++; - tmp++; - } - break; - } - - case PTRACE_GETEVENTMSG: - ret = put_user(child->ptrace_message, (unsigned int __user *) data); - break; - -#ifdef CONFIG_ALTIVEC - case PTRACE_GETVRREGS: - /* Get the child altivec register state. */ - flush_altivec_to_thread(child); - ret = get_vrregs((unsigned long __user *)data, child); - break; - - case PTRACE_SETVRREGS: - /* Set the child altivec register state. */ - flush_altivec_to_thread(child); - ret = set_vrregs(child, (unsigned long __user *)data); - break; -#endif - - default: - ret = ptrace_request(child, request, addr, data); - break; - } -out_tsk: - put_task_struct(child); -out: - unlock_kernel(); - return ret; -} diff --git a/arch/ppc64/kernel/rtas-proc.c b/arch/ppc64/kernel/rtas-proc.c index 1f3ff860fdf0..5bdd5b079d96 100644 --- a/arch/ppc64/kernel/rtas-proc.c +++ b/arch/ppc64/kernel/rtas-proc.c @@ -23,6 +23,7 @@ #include <linux/init.h> #include <linux/seq_file.h> #include <linux/bitops.h> +#include <linux/rtc.h> #include <asm/uaccess.h> #include <asm/processor.h> diff --git a/arch/ppc64/kernel/rtas.c b/arch/ppc64/kernel/rtas.c index 5e8eb33b8e54..36adab591bd3 100644 --- a/arch/ppc64/kernel/rtas.c +++ b/arch/ppc64/kernel/rtas.c @@ -30,6 +30,7 @@ #include <asm/delay.h> #include <asm/uaccess.h> #include <asm/systemcfg.h> +#include <asm/ppcdebug.h> struct flash_block_list_header rtas_firmware_flash_list = {0, NULL}; diff --git a/arch/ppc64/kernel/rtas_pci.c b/arch/ppc64/kernel/rtas_pci.c index 4a9719b48abe..3ad15c90fbbd 100644 --- a/arch/ppc64/kernel/rtas_pci.c +++ b/arch/ppc64/kernel/rtas_pci.c @@ -38,9 +38,8 @@ #include <asm/pci-bridge.h> #include <asm/iommu.h> #include <asm/rtas.h> - -#include "mpic.h" -#include "pci.h" +#include <asm/mpic.h> +#include <asm/ppc-pci.h> /* RTAS tokens */ static int read_pci_config; @@ -401,7 +400,7 @@ unsigned long __init find_and_init_phbs(void) if (!phb) continue; - pci_process_bridge_OF_ranges(phb, node); + pci_process_bridge_OF_ranges(phb, node, 0); pci_setup_phb_io(phb, index == 0); #ifdef CONFIG_PPC_PSERIES if (ppc64_interrupt_controller == IC_OPEN_PIC && pSeries_mpic) { @@ -451,7 +450,7 @@ struct pci_controller * __devinit init_phb_dynamic(struct device_node *dn) if (!phb) return NULL; - pci_process_bridge_OF_ranges(phb, dn); + pci_process_bridge_OF_ranges(phb, dn, primary); pci_setup_phb_io_dynamic(phb, primary); of_node_put(root); diff --git a/arch/ppc64/kernel/rtc.c b/arch/ppc64/kernel/rtc.c index 6ff52bc61325..79e7ed2858dd 100644 --- a/arch/ppc64/kernel/rtc.c +++ b/arch/ppc64/kernel/rtc.c @@ -43,11 +43,8 @@ #include <asm/time.h> #include <asm/rtas.h> -#include <asm/iSeries/mf.h> #include <asm/machdep.h> -extern int piranha_simulator; - /* * We sponge a minor off of the misc major. No need slurping * up another valuable major dev number for this. If you add @@ -265,44 +262,10 @@ static int rtc_read_proc(char *page, char **start, off_t off, return len; } -#ifdef CONFIG_PPC_ISERIES -/* - * Get the RTC from the virtual service processor - * This requires flowing LpEvents to the primary partition - */ -void iSeries_get_rtc_time(struct rtc_time *rtc_tm) -{ - if (piranha_simulator) - return; - - mf_get_rtc(rtc_tm); - rtc_tm->tm_mon--; -} - -/* - * Set the RTC in the virtual service processor - * This requires flowing LpEvents to the primary partition - */ -int iSeries_set_rtc_time(struct rtc_time *tm) -{ - mf_set_rtc(tm); - return 0; -} - -void iSeries_get_boot_time(struct rtc_time *tm) -{ - if ( piranha_simulator ) - return; - - mf_get_boot_rtc(tm); - tm->tm_mon -= 1; -} -#endif - #ifdef CONFIG_PPC_RTAS #define MAX_RTC_WAIT 5000 /* 5 sec */ #define RTAS_CLOCK_BUSY (-2) -void rtas_get_boot_time(struct rtc_time *rtc_tm) +unsigned long rtas_get_boot_time(void) { int ret[8]; int error, wait_time; @@ -322,15 +285,10 @@ void rtas_get_boot_time(struct rtc_time *rtc_tm) if (error != 0 && printk_ratelimit()) { printk(KERN_WARNING "error: reading the clock failed (%d)\n", error); - return; + return 0; } - rtc_tm->tm_sec = ret[5]; - rtc_tm->tm_min = ret[4]; - rtc_tm->tm_hour = ret[3]; - rtc_tm->tm_mday = ret[2]; - rtc_tm->tm_mon = ret[1] - 1; - rtc_tm->tm_year = ret[0] - 1900; + return mktime(ret[0], ret[1], ret[2], ret[3], ret[4], ret[5]); } /* NOTE: get_rtc_time will get an error if executed in interrupt context diff --git a/arch/ppc64/kernel/setup.c b/arch/ppc64/kernel/setup.c index 5ac48bd64891..44ee6ebe9a60 100644 --- a/arch/ppc64/kernel/setup.c +++ b/arch/ppc64/kernel/setup.c @@ -58,6 +58,7 @@ #include <asm/mmu.h> #include <asm/lmb.h> #include <asm/iSeries/ItLpNaca.h> +#include <asm/firmware.h> #ifdef DEBUG #define DBG(fmt...) udbg_printf(fmt) @@ -153,7 +154,7 @@ struct screen_info screen_info = { .orig_video_points = 16 }; -#if defined(CONFIG_PPC_MULTIPLATFORM) && defined(CONFIG_SMP) +#ifdef CONFIG_SMP static int smt_enabled_cmdline; @@ -306,15 +307,13 @@ static void __init setup_cpu_maps(void) systemcfg->processorCount = num_present_cpus(); } -#endif /* defined(CONFIG_PPC_MULTIPLATFORM) && defined(CONFIG_SMP) */ - - -#ifdef CONFIG_PPC_MULTIPLATFORM +#endif /* CONFIG_SMP */ extern struct machdep_calls pSeries_md; extern struct machdep_calls pmac_md; extern struct machdep_calls maple_md; extern struct machdep_calls bpa_md; +extern struct machdep_calls iseries_md; /* Ultimately, stuff them in an elf section like initcalls... */ static struct machdep_calls __initdata *machines[] = { @@ -330,6 +329,9 @@ static struct machdep_calls __initdata *machines[] = { #ifdef CONFIG_PPC_BPA &bpa_md, #endif +#ifdef CONFIG_PPC_ISERIES + &iseries_md, +#endif NULL }; @@ -401,7 +403,8 @@ void __init early_setup(unsigned long dt_ptr) /* * Initialize stab / SLB management */ - stab_initialize(lpaca->stab_real); + if (!firmware_has_feature(FW_FEATURE_ISERIES)) + stab_initialize(lpaca->stab_real); /* * Initialize the MMU Hash table and create the linear mapping @@ -532,8 +535,6 @@ static void __init check_for_initrd(void) #endif /* CONFIG_BLK_DEV_INITRD */ } -#endif /* CONFIG_PPC_MULTIPLATFORM */ - /* * Do some initial setup of the system. The parameters are those which * were passed in from the bootloader. @@ -542,14 +543,6 @@ void __init setup_system(void) { DBG(" -> setup_system()\n"); -#ifdef CONFIG_PPC_ISERIES - /* pSeries systems are identified in prom.c via OF. */ - if (itLpNaca.xLparInstalled == 1) - systemcfg->platform = PLATFORM_ISERIES_LPAR; - - ppc_md.init_early(); -#else /* CONFIG_PPC_ISERIES */ - /* * Unflatten the device-tree passed by prom_init or kexec */ @@ -592,6 +585,10 @@ void __init setup_system(void) */ finish_device_tree(); +#ifdef CONFIG_BOOTX_TEXT + init_boot_display(); +#endif + /* * Initialize xmon */ @@ -607,9 +604,8 @@ void __init setup_system(void) strlcpy(saved_command_line, cmd_line, COMMAND_LINE_SIZE); parse_early_param(); -#endif /* !CONFIG_PPC_ISERIES */ -#if defined(CONFIG_SMP) && !defined(CONFIG_PPC_ISERIES) +#ifdef CONFIG_SMP /* * iSeries has already initialized the cpu maps at this point. */ @@ -619,7 +615,7 @@ void __init setup_system(void) * we can map physical -> logical CPU ids */ smp_release_cpus(); -#endif /* defined(CONFIG_SMP) && !defined(CONFIG_PPC_ISERIES) */ +#endif printk("Starting Linux PPC64 %s\n", system_utsname.version); @@ -711,8 +707,8 @@ static int show_cpuinfo(struct seq_file *m, void *v) if (cpu_id == NR_CPUS) { seq_printf(m, "timebase\t: %lu\n", ppc_tb_freq); - if (ppc_md.get_cpuinfo != NULL) - ppc_md.get_cpuinfo(m); + if (ppc_md.show_cpuinfo != NULL) + ppc_md.show_cpuinfo(m); return 0; } @@ -1047,6 +1043,10 @@ void __init setup_arch(char **cmdline_p) /* initialize the syscall map in systemcfg */ setup_syscall_map(); +#ifdef CONFIG_DUMMY_CONSOLE + conswitchp = &dummy_con; +#endif + ppc_md.setup_arch(); /* Use the default idle loop if the platform hasn't provided one. */ @@ -1091,15 +1091,6 @@ void ppc64_terminate_msg(unsigned int src, const char *msg) printk("[terminate]%04x %s\n", src, msg); } -/* This should only be called on processor 0 during calibrate decr */ -void __init setup_default_decr(void) -{ - struct paca_struct *lpaca = get_paca(); - - lpaca->default_decr = tb_ticks_per_jiffy; - lpaca->next_jiffy_update_tb = get_tb() + tb_ticks_per_jiffy; -} - #ifndef CONFIG_PPC_ISERIES /* * This function can be used by platforms to "find" legacy serial ports. diff --git a/arch/ppc64/kernel/signal32.c b/arch/ppc64/kernel/signal32.c deleted file mode 100644 index a8b7a5a56bb4..000000000000 --- a/arch/ppc64/kernel/signal32.c +++ /dev/null @@ -1,998 +0,0 @@ -/* - * signal32.c: Support 32bit signal syscalls. - * - * Copyright (C) 2001 IBM - * Copyright (C) 1997,1998 Jakub Jelinek (jj@sunsite.mff.cuni.cz) - * Copyright (C) 1997 David S. Miller (davem@caip.rutgers.edu) - * - * These routines maintain argument size conversion between 32bit and 64bit - * environment. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/config.h> -#include <linux/sched.h> -#include <linux/mm.h> -#include <linux/smp.h> -#include <linux/smp_lock.h> -#include <linux/kernel.h> -#include <linux/signal.h> -#include <linux/syscalls.h> -#include <linux/errno.h> -#include <linux/elf.h> -#include <linux/compat.h> -#include <linux/ptrace.h> -#include <asm/ppc32.h> -#include <asm/uaccess.h> -#include <asm/ppcdebug.h> -#include <asm/unistd.h> -#include <asm/cacheflush.h> -#include <asm/vdso.h> - -#define DEBUG_SIG 0 - -#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) - -#define GP_REGS_SIZE32 min(sizeof(elf_gregset_t32), sizeof(struct pt_regs32)) - -/* - * When we have signals to deliver, we set up on the - * user stack, going down from the original stack pointer: - * a sigregs32 struct - * a sigcontext32 struct - * a gap of __SIGNAL_FRAMESIZE32 bytes - * - * Each of these things must be a multiple of 16 bytes in size. - * - */ -struct sigregs32 { - struct mcontext32 mctx; /* all the register values */ - /* - * Programs using the rs6000/xcoff abi can save up to 19 gp - * regs and 18 fp regs below sp before decrementing it. - */ - int abigap[56]; -}; - -/* We use the mc_pad field for the signal return trampoline. */ -#define tramp mc_pad - -/* - * When we have rt signals to deliver, we set up on the - * user stack, going down from the original stack pointer: - * one rt_sigframe32 struct (siginfo + ucontext + ABI gap) - * a gap of __SIGNAL_FRAMESIZE32+16 bytes - * (the +16 is to get the siginfo and ucontext32 in the same - * positions as in older kernels). - * - * Each of these things must be a multiple of 16 bytes in size. - * - */ -struct rt_sigframe32 { - compat_siginfo_t info; - struct ucontext32 uc; - /* - * Programs using the rs6000/xcoff abi can save up to 19 gp - * regs and 18 fp regs below sp before decrementing it. - */ - int abigap[56]; -}; - - -/* - * Common utility functions used by signal and context support - * - */ - -/* - * Restore the user process's signal mask - * (implemented in signal.c) - */ -extern void restore_sigmask(sigset_t *set); - -/* - * Functions for flipping sigsets (thanks to brain dead generic - * implementation that makes things simple for little endian only - */ -static inline void compat_from_sigset(compat_sigset_t *compat, sigset_t *set) -{ - switch (_NSIG_WORDS) { - case 4: compat->sig[5] = set->sig[3] & 0xffffffffull ; - compat->sig[7] = set->sig[3] >> 32; - case 3: compat->sig[4] = set->sig[2] & 0xffffffffull ; - compat->sig[5] = set->sig[2] >> 32; - case 2: compat->sig[2] = set->sig[1] & 0xffffffffull ; - compat->sig[3] = set->sig[1] >> 32; - case 1: compat->sig[0] = set->sig[0] & 0xffffffffull ; - compat->sig[1] = set->sig[0] >> 32; - } -} - -static inline void sigset_from_compat(sigset_t *set, compat_sigset_t *compat) -{ - switch (_NSIG_WORDS) { - case 4: set->sig[3] = compat->sig[6] | (((long)compat->sig[7]) << 32); - case 3: set->sig[2] = compat->sig[4] | (((long)compat->sig[5]) << 32); - case 2: set->sig[1] = compat->sig[2] | (((long)compat->sig[3]) << 32); - case 1: set->sig[0] = compat->sig[0] | (((long)compat->sig[1]) << 32); - } -} - - -/* - * Save the current user registers on the user stack. - * We only save the altivec registers if the process has used - * altivec instructions at some point. - */ -static int save_user_regs(struct pt_regs *regs, struct mcontext32 __user *frame, int sigret) -{ - elf_greg_t64 *gregs = (elf_greg_t64 *)regs; - int i, err = 0; - - /* Make sure floating point registers are stored in regs */ - flush_fp_to_thread(current); - - /* save general and floating-point registers */ - for (i = 0; i <= PT_RESULT; i ++) - err |= __put_user((unsigned int)gregs[i], &frame->mc_gregs[i]); - err |= __copy_to_user(&frame->mc_fregs, current->thread.fpr, - ELF_NFPREG * sizeof(double)); - if (err) - return 1; - - current->thread.fpscr = 0; /* turn off all fp exceptions */ - -#ifdef CONFIG_ALTIVEC - /* save altivec registers */ - if (current->thread.used_vr) { - flush_altivec_to_thread(current); - if (__copy_to_user(&frame->mc_vregs, current->thread.vr, - ELF_NVRREG32 * sizeof(vector128))) - return 1; - /* set MSR_VEC in the saved MSR value to indicate that - frame->mc_vregs contains valid data */ - if (__put_user(regs->msr | MSR_VEC, &frame->mc_gregs[PT_MSR])) - return 1; - } - /* else assert((regs->msr & MSR_VEC) == 0) */ - - /* We always copy to/from vrsave, it's 0 if we don't have or don't - * use altivec. Since VSCR only contains 32 bits saved in the least - * significant bits of a vector, we "cheat" and stuff VRSAVE in the - * most significant bits of that same vector. --BenH - */ - if (__put_user(current->thread.vrsave, (u32 __user *)&frame->mc_vregs[32])) - return 1; -#endif /* CONFIG_ALTIVEC */ - - if (sigret) { - /* Set up the sigreturn trampoline: li r0,sigret; sc */ - if (__put_user(0x38000000UL + sigret, &frame->tramp[0]) - || __put_user(0x44000002UL, &frame->tramp[1])) - return 1; - flush_icache_range((unsigned long) &frame->tramp[0], - (unsigned long) &frame->tramp[2]); - } - - return 0; -} - -/* - * Restore the current user register values from the user stack, - * (except for MSR). - */ -static long restore_user_regs(struct pt_regs *regs, - struct mcontext32 __user *sr, int sig) -{ - elf_greg_t64 *gregs = (elf_greg_t64 *)regs; - int i; - long err = 0; - unsigned int save_r2 = 0; -#ifdef CONFIG_ALTIVEC - unsigned long msr; -#endif - - /* - * restore general registers but not including MSR or SOFTE. Also - * take care of keeping r2 (TLS) intact if not a signal - */ - if (!sig) - save_r2 = (unsigned int)regs->gpr[2]; - for (i = 0; i <= PT_RESULT; i++) { - if ((i == PT_MSR) || (i == PT_SOFTE)) - continue; - err |= __get_user(gregs[i], &sr->mc_gregs[i]); - } - if (!sig) - regs->gpr[2] = (unsigned long) save_r2; - if (err) - return 1; - - /* force the process to reload the FP registers from - current->thread when it next does FP instructions */ - regs->msr &= ~(MSR_FP | MSR_FE0 | MSR_FE1); - if (__copy_from_user(current->thread.fpr, &sr->mc_fregs, - sizeof(sr->mc_fregs))) - return 1; - -#ifdef CONFIG_ALTIVEC - /* force the process to reload the altivec registers from - current->thread when it next does altivec instructions */ - regs->msr &= ~MSR_VEC; - if (!__get_user(msr, &sr->mc_gregs[PT_MSR]) && (msr & MSR_VEC) != 0) { - /* restore altivec registers from the stack */ - if (__copy_from_user(current->thread.vr, &sr->mc_vregs, - sizeof(sr->mc_vregs))) - return 1; - } else if (current->thread.used_vr) - memset(current->thread.vr, 0, ELF_NVRREG32 * sizeof(vector128)); - - /* Always get VRSAVE back */ - if (__get_user(current->thread.vrsave, (u32 __user *)&sr->mc_vregs[32])) - return 1; -#endif /* CONFIG_ALTIVEC */ - -#ifndef CONFIG_SMP - preempt_disable(); - if (last_task_used_math == current) - last_task_used_math = NULL; - if (last_task_used_altivec == current) - last_task_used_altivec = NULL; - preempt_enable(); -#endif - return 0; -} - - -/* - * Start of nonRT signal support - * - * sigset_t is 32 bits for non-rt signals - * - * System Calls - * sigaction sys32_sigaction - * sigreturn sys32_sigreturn - * - * Note sigsuspend has no special 32 bit routine - uses the 64 bit routine - * - * Other routines - * setup_frame32 - */ - -/* - * Atomically swap in the new signal mask, and wait for a signal. - */ -long sys32_sigsuspend(old_sigset_t mask, int p2, int p3, int p4, int p6, int p7, - struct pt_regs *regs) -{ - sigset_t saveset; - - mask &= _BLOCKABLE; - spin_lock_irq(¤t->sighand->siglock); - saveset = current->blocked; - siginitset(¤t->blocked, mask); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - regs->result = -EINTR; - regs->gpr[3] = EINTR; - regs->ccr |= 0x10000000; - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - if (do_signal32(&saveset, regs)) - /* - * Returning 0 means we return to userspace via - * ret_from_except and thus restore all user - * registers from *regs. This is what we need - * to do when a signal has been delivered. - */ - return 0; - } -} - -long sys32_sigaction(int sig, struct old_sigaction32 __user *act, - struct old_sigaction32 __user *oact) -{ - struct k_sigaction new_ka, old_ka; - int ret; - - if (sig < 0) - sig = -sig; - - if (act) { - compat_old_sigset_t mask; - compat_uptr_t handler, restorer; - - if (get_user(handler, &act->sa_handler) || - __get_user(restorer, &act->sa_restorer) || - __get_user(new_ka.sa.sa_flags, &act->sa_flags) || - __get_user(mask, &act->sa_mask)) - return -EFAULT; - new_ka.sa.sa_handler = compat_ptr(handler); - new_ka.sa.sa_restorer = compat_ptr(restorer); - siginitset(&new_ka.sa.sa_mask, mask); - } - - ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL); - if (!ret && oact) { - if (put_user((long)old_ka.sa.sa_handler, &oact->sa_handler) || - __put_user((long)old_ka.sa.sa_restorer, &oact->sa_restorer) || - __put_user(old_ka.sa.sa_flags, &oact->sa_flags) || - __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask)) - return -EFAULT; - } - - return ret; -} - - - -/* - * Start of RT signal support - * - * sigset_t is 64 bits for rt signals - * - * System Calls - * sigaction sys32_rt_sigaction - * sigpending sys32_rt_sigpending - * sigprocmask sys32_rt_sigprocmask - * sigreturn sys32_rt_sigreturn - * sigqueueinfo sys32_rt_sigqueueinfo - * sigsuspend sys32_rt_sigsuspend - * - * Other routines - * setup_rt_frame32 - * copy_siginfo_to_user32 - * siginfo32to64 - */ - - -long sys32_rt_sigaction(int sig, const struct sigaction32 __user *act, - struct sigaction32 __user *oact, size_t sigsetsize) -{ - struct k_sigaction new_ka, old_ka; - int ret; - compat_sigset_t set32; - - /* XXX: Don't preclude handling different sized sigset_t's. */ - if (sigsetsize != sizeof(compat_sigset_t)) - return -EINVAL; - - if (act) { - compat_uptr_t handler; - - ret = get_user(handler, &act->sa_handler); - new_ka.sa.sa_handler = compat_ptr(handler); - ret |= __copy_from_user(&set32, &act->sa_mask, - sizeof(compat_sigset_t)); - sigset_from_compat(&new_ka.sa.sa_mask, &set32); - ret |= __get_user(new_ka.sa.sa_flags, &act->sa_flags); - if (ret) - return -EFAULT; - } - - ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL); - if (!ret && oact) { - compat_from_sigset(&set32, &old_ka.sa.sa_mask); - ret = put_user((long)old_ka.sa.sa_handler, &oact->sa_handler); - ret |= __copy_to_user(&oact->sa_mask, &set32, - sizeof(compat_sigset_t)); - ret |= __put_user(old_ka.sa.sa_flags, &oact->sa_flags); - } - return ret; -} - -/* - * Note: it is necessary to treat how as an unsigned int, with the - * corresponding cast to a signed int to insure that the proper - * conversion (sign extension) between the register representation - * of a signed int (msr in 32-bit mode) and the register representation - * of a signed int (msr in 64-bit mode) is performed. - */ -long sys32_rt_sigprocmask(u32 how, compat_sigset_t __user *set, - compat_sigset_t __user *oset, size_t sigsetsize) -{ - sigset_t s; - sigset_t __user *up; - compat_sigset_t s32; - int ret; - mm_segment_t old_fs = get_fs(); - - if (set) { - if (copy_from_user (&s32, set, sizeof(compat_sigset_t))) - return -EFAULT; - sigset_from_compat(&s, &s32); - } - - set_fs(KERNEL_DS); - /* This is valid because of the set_fs() */ - up = (sigset_t __user *) &s; - ret = sys_rt_sigprocmask((int)how, set ? up : NULL, oset ? up : NULL, - sigsetsize); - set_fs(old_fs); - if (ret) - return ret; - if (oset) { - compat_from_sigset(&s32, &s); - if (copy_to_user (oset, &s32, sizeof(compat_sigset_t))) - return -EFAULT; - } - return 0; -} - -long sys32_rt_sigpending(compat_sigset_t __user *set, compat_size_t sigsetsize) -{ - sigset_t s; - compat_sigset_t s32; - int ret; - mm_segment_t old_fs = get_fs(); - - set_fs(KERNEL_DS); - /* The __user pointer cast is valid because of the set_fs() */ - ret = sys_rt_sigpending((sigset_t __user *) &s, sigsetsize); - set_fs(old_fs); - if (!ret) { - compat_from_sigset(&s32, &s); - if (copy_to_user (set, &s32, sizeof(compat_sigset_t))) - return -EFAULT; - } - return ret; -} - - -int copy_siginfo_to_user32(struct compat_siginfo __user *d, siginfo_t *s) -{ - int err; - - if (!access_ok (VERIFY_WRITE, d, sizeof(*d))) - return -EFAULT; - - /* If you change siginfo_t structure, please be sure - * this code is fixed accordingly. - * It should never copy any pad contained in the structure - * to avoid security leaks, but must copy the generic - * 3 ints plus the relevant union member. - * This routine must convert siginfo from 64bit to 32bit as well - * at the same time. - */ - err = __put_user(s->si_signo, &d->si_signo); - err |= __put_user(s->si_errno, &d->si_errno); - err |= __put_user((short)s->si_code, &d->si_code); - if (s->si_code < 0) - err |= __copy_to_user(&d->_sifields._pad, &s->_sifields._pad, - SI_PAD_SIZE32); - else switch(s->si_code >> 16) { - case __SI_CHLD >> 16: - err |= __put_user(s->si_pid, &d->si_pid); - err |= __put_user(s->si_uid, &d->si_uid); - err |= __put_user(s->si_utime, &d->si_utime); - err |= __put_user(s->si_stime, &d->si_stime); - err |= __put_user(s->si_status, &d->si_status); - break; - case __SI_FAULT >> 16: - err |= __put_user((unsigned int)(unsigned long)s->si_addr, - &d->si_addr); - break; - case __SI_POLL >> 16: - err |= __put_user(s->si_band, &d->si_band); - err |= __put_user(s->si_fd, &d->si_fd); - break; - case __SI_TIMER >> 16: - err |= __put_user(s->si_tid, &d->si_tid); - err |= __put_user(s->si_overrun, &d->si_overrun); - err |= __put_user(s->si_int, &d->si_int); - break; - case __SI_RT >> 16: /* This is not generated by the kernel as of now. */ - case __SI_MESGQ >> 16: - err |= __put_user(s->si_int, &d->si_int); - /* fallthrough */ - case __SI_KILL >> 16: - default: - err |= __put_user(s->si_pid, &d->si_pid); - err |= __put_user(s->si_uid, &d->si_uid); - break; - } - return err; -} - -/* - * Note: it is necessary to treat pid and sig as unsigned ints, with the - * corresponding cast to a signed int to insure that the proper conversion - * (sign extension) between the register representation of a signed int - * (msr in 32-bit mode) and the register representation of a signed int - * (msr in 64-bit mode) is performed. - */ -long sys32_rt_sigqueueinfo(u32 pid, u32 sig, compat_siginfo_t __user *uinfo) -{ - siginfo_t info; - int ret; - mm_segment_t old_fs = get_fs(); - - if (copy_from_user (&info, uinfo, 3*sizeof(int)) || - copy_from_user (info._sifields._pad, uinfo->_sifields._pad, SI_PAD_SIZE32)) - return -EFAULT; - set_fs (KERNEL_DS); - /* The __user pointer cast is valid becasuse of the set_fs() */ - ret = sys_rt_sigqueueinfo((int)pid, (int)sig, (siginfo_t __user *) &info); - set_fs (old_fs); - return ret; -} - -int sys32_rt_sigsuspend(compat_sigset_t __user * unewset, size_t sigsetsize, int p3, - int p4, int p6, int p7, struct pt_regs *regs) -{ - sigset_t saveset, newset; - compat_sigset_t s32; - - /* XXX: Don't preclude handling different sized sigset_t's. */ - if (sigsetsize != sizeof(sigset_t)) - return -EINVAL; - - if (copy_from_user(&s32, unewset, sizeof(s32))) - return -EFAULT; - - /* - * Swap the 2 words of the 64-bit sigset_t (they are stored - * in the "wrong" endian in 32-bit user storage). - */ - sigset_from_compat(&newset, &s32); - - sigdelsetmask(&newset, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - saveset = current->blocked; - current->blocked = newset; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - regs->result = -EINTR; - regs->gpr[3] = EINTR; - regs->ccr |= 0x10000000; - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - if (do_signal32(&saveset, regs)) - /* - * Returning 0 means we return to userspace via - * ret_from_except and thus restore all user - * registers from *regs. This is what we need - * to do when a signal has been delivered. - */ - return 0; - } -} - -/* - * Start Alternate signal stack support - * - * System Calls - * sigaltatck sys32_sigaltstack - */ - -int sys32_sigaltstack(u32 __new, u32 __old, int r5, - int r6, int r7, int r8, struct pt_regs *regs) -{ - stack_32_t __user * newstack = (stack_32_t __user *)(long) __new; - stack_32_t __user * oldstack = (stack_32_t __user *)(long) __old; - stack_t uss, uoss; - int ret; - mm_segment_t old_fs; - unsigned long sp; - compat_uptr_t ss_sp; - - /* - * set sp to the user stack on entry to the system call - * the system call router sets R9 to the saved registers - */ - sp = regs->gpr[1]; - - /* Put new stack info in local 64 bit stack struct */ - if (newstack) { - if (get_user(ss_sp, &newstack->ss_sp) || - __get_user(uss.ss_flags, &newstack->ss_flags) || - __get_user(uss.ss_size, &newstack->ss_size)) - return -EFAULT; - uss.ss_sp = compat_ptr(ss_sp); - } - - old_fs = get_fs(); - set_fs(KERNEL_DS); - /* The __user pointer casts are valid because of the set_fs() */ - ret = do_sigaltstack( - newstack ? (stack_t __user *) &uss : NULL, - oldstack ? (stack_t __user *) &uoss : NULL, - sp); - set_fs(old_fs); - /* Copy the stack information to the user output buffer */ - if (!ret && oldstack && - (put_user((long)uoss.ss_sp, &oldstack->ss_sp) || - __put_user(uoss.ss_flags, &oldstack->ss_flags) || - __put_user(uoss.ss_size, &oldstack->ss_size))) - return -EFAULT; - return ret; -} - - -/* - * Set up a signal frame for a "real-time" signal handler - * (one which gets siginfo). - */ -static int handle_rt_signal32(unsigned long sig, struct k_sigaction *ka, - siginfo_t *info, sigset_t *oldset, - struct pt_regs * regs, unsigned long newsp) -{ - struct rt_sigframe32 __user *rt_sf; - struct mcontext32 __user *frame; - unsigned long origsp = newsp; - compat_sigset_t c_oldset; - - /* Set up Signal Frame */ - /* Put a Real Time Context onto stack */ - newsp -= sizeof(*rt_sf); - rt_sf = (struct rt_sigframe32 __user *)newsp; - - /* create a stack frame for the caller of the handler */ - newsp -= __SIGNAL_FRAMESIZE32 + 16; - - if (!access_ok(VERIFY_WRITE, (void __user *)newsp, origsp - newsp)) - goto badframe; - - compat_from_sigset(&c_oldset, oldset); - - /* Put the siginfo & fill in most of the ucontext */ - if (copy_siginfo_to_user32(&rt_sf->info, info) - || __put_user(0, &rt_sf->uc.uc_flags) - || __put_user(0, &rt_sf->uc.uc_link) - || __put_user(current->sas_ss_sp, &rt_sf->uc.uc_stack.ss_sp) - || __put_user(sas_ss_flags(regs->gpr[1]), - &rt_sf->uc.uc_stack.ss_flags) - || __put_user(current->sas_ss_size, &rt_sf->uc.uc_stack.ss_size) - || __put_user((u32)(u64)&rt_sf->uc.uc_mcontext, &rt_sf->uc.uc_regs) - || __copy_to_user(&rt_sf->uc.uc_sigmask, &c_oldset, sizeof(c_oldset))) - goto badframe; - - /* Save user registers on the stack */ - frame = &rt_sf->uc.uc_mcontext; - if (put_user(regs->gpr[1], (u32 __user *)newsp)) - goto badframe; - - if (vdso32_rt_sigtramp && current->thread.vdso_base) { - if (save_user_regs(regs, frame, 0)) - goto badframe; - regs->link = current->thread.vdso_base + vdso32_rt_sigtramp; - } else { - if (save_user_regs(regs, frame, __NR_rt_sigreturn)) - goto badframe; - regs->link = (unsigned long) frame->tramp; - } - regs->gpr[1] = (unsigned long) newsp; - regs->gpr[3] = sig; - regs->gpr[4] = (unsigned long) &rt_sf->info; - regs->gpr[5] = (unsigned long) &rt_sf->uc; - regs->gpr[6] = (unsigned long) rt_sf; - regs->nip = (unsigned long) ka->sa.sa_handler; - regs->trap = 0; - regs->result = 0; - - if (test_thread_flag(TIF_SINGLESTEP)) - ptrace_notify(SIGTRAP); - - return 1; - -badframe: -#if DEBUG_SIG - printk("badframe in handle_rt_signal, regs=%p frame=%p newsp=%lx\n", - regs, frame, newsp); -#endif - force_sigsegv(sig, current); - return 0; -} - -static long do_setcontext32(struct ucontext32 __user *ucp, struct pt_regs *regs, int sig) -{ - compat_sigset_t c_set; - sigset_t set; - u32 mcp; - - if (__copy_from_user(&c_set, &ucp->uc_sigmask, sizeof(c_set)) - || __get_user(mcp, &ucp->uc_regs)) - return -EFAULT; - sigset_from_compat(&set, &c_set); - restore_sigmask(&set); - if (restore_user_regs(regs, (struct mcontext32 __user *)(u64)mcp, sig)) - return -EFAULT; - - return 0; -} - -/* - * Handle {get,set,swap}_context operations for 32 bits processes - */ - -long sys32_swapcontext(struct ucontext32 __user *old_ctx, - struct ucontext32 __user *new_ctx, - int ctx_size, int r6, int r7, int r8, struct pt_regs *regs) -{ - unsigned char tmp; - compat_sigset_t c_set; - - /* Context size is for future use. Right now, we only make sure - * we are passed something we understand - */ - if (ctx_size < sizeof(struct ucontext32)) - return -EINVAL; - - if (old_ctx != NULL) { - compat_from_sigset(&c_set, ¤t->blocked); - if (!access_ok(VERIFY_WRITE, old_ctx, sizeof(*old_ctx)) - || save_user_regs(regs, &old_ctx->uc_mcontext, 0) - || __copy_to_user(&old_ctx->uc_sigmask, &c_set, sizeof(c_set)) - || __put_user((u32)(u64)&old_ctx->uc_mcontext, &old_ctx->uc_regs)) - return -EFAULT; - } - if (new_ctx == NULL) - return 0; - if (!access_ok(VERIFY_READ, new_ctx, sizeof(*new_ctx)) - || __get_user(tmp, (u8 __user *) new_ctx) - || __get_user(tmp, (u8 __user *) (new_ctx + 1) - 1)) - return -EFAULT; - - /* - * If we get a fault copying the context into the kernel's - * image of the user's registers, we can't just return -EFAULT - * because the user's registers will be corrupted. For instance - * the NIP value may have been updated but not some of the - * other registers. Given that we have done the access_ok - * and successfully read the first and last bytes of the region - * above, this should only happen in an out-of-memory situation - * or if another thread unmaps the region containing the context. - * We kill the task with a SIGSEGV in this situation. - */ - if (do_setcontext32(new_ctx, regs, 0)) - do_exit(SIGSEGV); - - return 0; -} - -long sys32_rt_sigreturn(int r3, int r4, int r5, int r6, int r7, int r8, - struct pt_regs *regs) -{ - struct rt_sigframe32 __user *rt_sf; - int ret; - - - /* Always make any pending restarted system calls return -EINTR */ - current_thread_info()->restart_block.fn = do_no_restart_syscall; - - rt_sf = (struct rt_sigframe32 __user *) - (regs->gpr[1] + __SIGNAL_FRAMESIZE32 + 16); - if (!access_ok(VERIFY_READ, rt_sf, sizeof(*rt_sf))) - goto bad; - if (do_setcontext32(&rt_sf->uc, regs, 1)) - goto bad; - - /* - * It's not clear whether or why it is desirable to save the - * sigaltstack setting on signal delivery and restore it on - * signal return. But other architectures do this and we have - * always done it up until now so it is probably better not to - * change it. -- paulus - * We use the sys32_ version that does the 32/64 bits conversion - * and takes userland pointer directly. What about error checking ? - * nobody does any... - */ - sys32_sigaltstack((u32)(u64)&rt_sf->uc.uc_stack, 0, 0, 0, 0, 0, regs); - - ret = regs->result; - - return ret; - - bad: - force_sig(SIGSEGV, current); - return 0; -} - - -/* - * OK, we're invoking a handler - */ -static int handle_signal32(unsigned long sig, struct k_sigaction *ka, - siginfo_t *info, sigset_t *oldset, - struct pt_regs * regs, unsigned long newsp) -{ - struct sigcontext32 __user *sc; - struct sigregs32 __user *frame; - unsigned long origsp = newsp; - - /* Set up Signal Frame */ - newsp -= sizeof(struct sigregs32); - frame = (struct sigregs32 __user *) newsp; - - /* Put a sigcontext on the stack */ - newsp -= sizeof(*sc); - sc = (struct sigcontext32 __user *) newsp; - - /* create a stack frame for the caller of the handler */ - newsp -= __SIGNAL_FRAMESIZE32; - - if (!access_ok(VERIFY_WRITE, (void __user *) newsp, origsp - newsp)) - goto badframe; - -#if _NSIG != 64 -#error "Please adjust handle_signal32()" -#endif - if (__put_user((u32)(u64)ka->sa.sa_handler, &sc->handler) - || __put_user(oldset->sig[0], &sc->oldmask) - || __put_user((oldset->sig[0] >> 32), &sc->_unused[3]) - || __put_user((u32)(u64)frame, &sc->regs) - || __put_user(sig, &sc->signal)) - goto badframe; - - if (vdso32_sigtramp && current->thread.vdso_base) { - if (save_user_regs(regs, &frame->mctx, 0)) - goto badframe; - regs->link = current->thread.vdso_base + vdso32_sigtramp; - } else { - if (save_user_regs(regs, &frame->mctx, __NR_sigreturn)) - goto badframe; - regs->link = (unsigned long) frame->mctx.tramp; - } - - if (put_user(regs->gpr[1], (u32 __user *)newsp)) - goto badframe; - regs->gpr[1] = (unsigned long) newsp; - regs->gpr[3] = sig; - regs->gpr[4] = (unsigned long) sc; - regs->nip = (unsigned long) ka->sa.sa_handler; - regs->trap = 0; - regs->result = 0; - - if (test_thread_flag(TIF_SINGLESTEP)) - ptrace_notify(SIGTRAP); - - return 1; - -badframe: -#if DEBUG_SIG - printk("badframe in handle_signal, regs=%p frame=%x newsp=%x\n", - regs, frame, *newspp); -#endif - force_sigsegv(sig, current); - return 0; -} - -/* - * Do a signal return; undo the signal stack. - */ -long sys32_sigreturn(int r3, int r4, int r5, int r6, int r7, int r8, - struct pt_regs *regs) -{ - struct sigcontext32 __user *sc; - struct sigcontext32 sigctx; - struct mcontext32 __user *sr; - sigset_t set; - int ret; - - /* Always make any pending restarted system calls return -EINTR */ - current_thread_info()->restart_block.fn = do_no_restart_syscall; - - sc = (struct sigcontext32 __user *)(regs->gpr[1] + __SIGNAL_FRAMESIZE32); - if (copy_from_user(&sigctx, sc, sizeof(sigctx))) - goto badframe; - - /* - * Note that PPC32 puts the upper 32 bits of the sigmask in the - * unused part of the signal stackframe - */ - set.sig[0] = sigctx.oldmask + ((long)(sigctx._unused[3]) << 32); - restore_sigmask(&set); - - sr = (struct mcontext32 __user *)(u64)sigctx.regs; - if (!access_ok(VERIFY_READ, sr, sizeof(*sr)) - || restore_user_regs(regs, sr, 1)) - goto badframe; - - ret = regs->result; - return ret; - -badframe: - force_sig(SIGSEGV, current); - return 0; -} - - - -/* - * Start of do_signal32 routine - * - * This routine gets control when a pending signal needs to be processed - * in the 32 bit target thread - - * - * It handles both rt and non-rt signals - */ - -/* - * Note that 'init' is a special process: it doesn't get signals it doesn't - * want to handle. Thus you cannot kill init even with a SIGKILL even by - * mistake. - */ - -int do_signal32(sigset_t *oldset, struct pt_regs *regs) -{ - siginfo_t info; - unsigned int frame, newsp; - int signr, ret; - struct k_sigaction ka; - - if (!oldset) - oldset = ¤t->blocked; - - newsp = frame = 0; - - signr = get_signal_to_deliver(&info, &ka, regs, NULL); - - if (TRAP(regs) == 0x0C00 /* System Call! */ - && regs->ccr & 0x10000000 /* error signalled */ - && ((ret = regs->gpr[3]) == ERESTARTSYS - || ret == ERESTARTNOHAND || ret == ERESTARTNOINTR - || ret == ERESTART_RESTARTBLOCK)) { - - if (signr > 0 - && (ret == ERESTARTNOHAND || ret == ERESTART_RESTARTBLOCK - || (ret == ERESTARTSYS - && !(ka.sa.sa_flags & SA_RESTART)))) { - /* make the system call return an EINTR error */ - regs->result = -EINTR; - regs->gpr[3] = EINTR; - /* note that the cr0.SO bit is already set */ - } else { - regs->nip -= 4; /* Back up & retry system call */ - regs->result = 0; - regs->trap = 0; - if (ret == ERESTART_RESTARTBLOCK) - regs->gpr[0] = __NR_restart_syscall; - else - regs->gpr[3] = regs->orig_gpr3; - } - } - - if (signr == 0) - return 0; /* no signals delivered */ - - if ((ka.sa.sa_flags & SA_ONSTACK) && current->sas_ss_size - && (!on_sig_stack(regs->gpr[1]))) - newsp = (current->sas_ss_sp + current->sas_ss_size); - else - newsp = regs->gpr[1]; - newsp &= ~0xfUL; - - /* - * Reenable the DABR before delivering the signal to - * user space. The DABR will have been cleared if it - * triggered inside the kernel. - */ - if (current->thread.dabr) - set_dabr(current->thread.dabr); - - /* Whee! Actually deliver the signal. */ - if (ka.sa.sa_flags & SA_SIGINFO) - ret = handle_rt_signal32(signr, &ka, &info, oldset, regs, newsp); - else - ret = handle_signal32(signr, &ka, &info, oldset, regs, newsp); - - if (ret) { - spin_lock_irq(¤t->sighand->siglock); - sigorsets(¤t->blocked, ¤t->blocked, - &ka.sa.sa_mask); - if (!(ka.sa.sa_flags & SA_NODEFER)) - sigaddset(¤t->blocked, signr); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - } - - return ret; -} diff --git a/arch/ppc64/kernel/smp.c b/arch/ppc64/kernel/smp.c index 793b562da653..017c12919832 100644 --- a/arch/ppc64/kernel/smp.c +++ b/arch/ppc64/kernel/smp.c @@ -45,8 +45,7 @@ #include <asm/cputable.h> #include <asm/system.h> #include <asm/abs_addr.h> - -#include "mpic.h" +#include <asm/mpic.h> #ifdef DEBUG #define DBG(fmt...) udbg_printf(fmt) @@ -70,28 +69,6 @@ void smp_call_function_interrupt(void); int smt_enabled_at_boot = 1; #ifdef CONFIG_MPIC -void smp_mpic_message_pass(int target, int msg) -{ - /* make sure we're sending something that translates to an IPI */ - if ( msg > 0x3 ){ - printk("SMP %d: smp_message_pass: unknown msg %d\n", - smp_processor_id(), msg); - return; - } - switch ( target ) - { - case MSG_ALL: - mpic_send_ipi(msg, 0xffffffff); - break; - case MSG_ALL_BUT_SELF: - mpic_send_ipi(msg, 0xffffffff & ~(1 << smp_processor_id())); - break; - default: - mpic_send_ipi(msg, 1 << target); - break; - } -} - int __init smp_mpic_probe(void) { int nr_cpus; @@ -128,21 +105,6 @@ void __devinit smp_generic_kick_cpu(int nr) #endif /* CONFIG_MPIC */ -static void __init smp_space_timers(unsigned int max_cpus) -{ - int i; - unsigned long offset = tb_ticks_per_jiffy / max_cpus; - unsigned long previous_tb = paca[boot_cpuid].next_jiffy_update_tb; - - for_each_cpu(i) { - if (i != boot_cpuid) { - paca[i].next_jiffy_update_tb = - previous_tb + offset; - previous_tb = paca[i].next_jiffy_update_tb; - } - } -} - void smp_message_recv(int msg, struct pt_regs *regs) { switch(msg) { diff --git a/arch/ppc64/kernel/sys_ppc32.c b/arch/ppc64/kernel/sys_ppc32.c deleted file mode 100644 index e93c13458910..000000000000 --- a/arch/ppc64/kernel/sys_ppc32.c +++ /dev/null @@ -1,1222 +0,0 @@ -/* - * sys_ppc32.c: Conversion between 32bit and 64bit native syscalls. - * - * Copyright (C) 2001 IBM - * Copyright (C) 1997,1998 Jakub Jelinek (jj@sunsite.mff.cuni.cz) - * Copyright (C) 1997 David S. Miller (davem@caip.rutgers.edu) - * - * These routines maintain argument size conversion between 32bit and 64bit - * environment. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/config.h> -#include <linux/kernel.h> -#include <linux/sched.h> -#include <linux/fs.h> -#include <linux/mm.h> -#include <linux/file.h> -#include <linux/signal.h> -#include <linux/resource.h> -#include <linux/times.h> -#include <linux/utsname.h> -#include <linux/timex.h> -#include <linux/smp.h> -#include <linux/smp_lock.h> -#include <linux/sem.h> -#include <linux/msg.h> -#include <linux/shm.h> -#include <linux/poll.h> -#include <linux/personality.h> -#include <linux/stat.h> -#include <linux/mman.h> -#include <linux/in.h> -#include <linux/syscalls.h> -#include <linux/unistd.h> -#include <linux/sysctl.h> -#include <linux/binfmts.h> -#include <linux/security.h> -#include <linux/compat.h> -#include <linux/ptrace.h> -#include <linux/elf.h> - -#include <asm/ptrace.h> -#include <asm/types.h> -#include <asm/ipc.h> -#include <asm/uaccess.h> -#include <asm/unistd.h> -#include <asm/semaphore.h> -#include <asm/time.h> -#include <asm/mmu_context.h> -#include <asm/systemcfg.h> - -#include "pci.h" - -/* readdir & getdents */ -#define NAME_OFFSET(de) ((int) ((de)->d_name - (char __user *) (de))) -#define ROUND_UP(x) (((x)+sizeof(u32)-1) & ~(sizeof(u32)-1)) - -struct old_linux_dirent32 { - u32 d_ino; - u32 d_offset; - unsigned short d_namlen; - char d_name[1]; -}; - -struct readdir_callback32 { - struct old_linux_dirent32 __user * dirent; - int count; -}; - -static int fillonedir(void * __buf, const char * name, int namlen, - off_t offset, ino_t ino, unsigned int d_type) -{ - struct readdir_callback32 * buf = (struct readdir_callback32 *) __buf; - struct old_linux_dirent32 __user * dirent; - - if (buf->count) - return -EINVAL; - buf->count++; - dirent = buf->dirent; - put_user(ino, &dirent->d_ino); - put_user(offset, &dirent->d_offset); - put_user(namlen, &dirent->d_namlen); - copy_to_user(dirent->d_name, name, namlen); - put_user(0, dirent->d_name + namlen); - return 0; -} - -asmlinkage int old32_readdir(unsigned int fd, struct old_linux_dirent32 __user *dirent, unsigned int count) -{ - int error = -EBADF; - struct file * file; - struct readdir_callback32 buf; - - file = fget(fd); - if (!file) - goto out; - - buf.count = 0; - buf.dirent = dirent; - - error = vfs_readdir(file, (filldir_t)fillonedir, &buf); - if (error < 0) - goto out_putf; - error = buf.count; - -out_putf: - fput(file); -out: - return error; -} - -struct linux_dirent32 { - u32 d_ino; - u32 d_off; - unsigned short d_reclen; - char d_name[1]; -}; - -struct getdents_callback32 { - struct linux_dirent32 __user * current_dir; - struct linux_dirent32 __user * previous; - int count; - int error; -}; - -static int filldir(void * __buf, const char * name, int namlen, off_t offset, - ino_t ino, unsigned int d_type) -{ - struct linux_dirent32 __user * dirent; - struct getdents_callback32 * buf = (struct getdents_callback32 *) __buf; - int reclen = ROUND_UP(NAME_OFFSET(dirent) + namlen + 2); - - buf->error = -EINVAL; /* only used if we fail.. */ - if (reclen > buf->count) - return -EINVAL; - dirent = buf->previous; - if (dirent) { - if (__put_user(offset, &dirent->d_off)) - goto efault; - } - dirent = buf->current_dir; - if (__put_user(ino, &dirent->d_ino)) - goto efault; - if (__put_user(reclen, &dirent->d_reclen)) - goto efault; - if (copy_to_user(dirent->d_name, name, namlen)) - goto efault; - if (__put_user(0, dirent->d_name + namlen)) - goto efault; - if (__put_user(d_type, (char __user *) dirent + reclen - 1)) - goto efault; - buf->previous = dirent; - dirent = (void __user *)dirent + reclen; - buf->current_dir = dirent; - buf->count -= reclen; - return 0; -efault: - buf->error = -EFAULT; - return -EFAULT; -} - -asmlinkage long sys32_getdents(unsigned int fd, struct linux_dirent32 __user *dirent, - unsigned int count) -{ - struct file * file; - struct linux_dirent32 __user * lastdirent; - struct getdents_callback32 buf; - int error; - - error = -EFAULT; - if (!access_ok(VERIFY_WRITE, dirent, count)) - goto out; - - error = -EBADF; - file = fget(fd); - if (!file) - goto out; - - buf.current_dir = dirent; - buf.previous = NULL; - buf.count = count; - buf.error = 0; - - error = vfs_readdir(file, (filldir_t)filldir, &buf); - if (error < 0) - goto out_putf; - error = buf.error; - lastdirent = buf.previous; - if (lastdirent) { - if (put_user(file->f_pos, &lastdirent->d_off)) - error = -EFAULT; - else - error = count - buf.count; - } - -out_putf: - fput(file); -out: - return error; -} - -asmlinkage long ppc32_select(u32 n, compat_ulong_t __user *inp, - compat_ulong_t __user *outp, compat_ulong_t __user *exp, - compat_uptr_t tvp_x) -{ - /* sign extend n */ - return compat_sys_select((int)n, inp, outp, exp, compat_ptr(tvp_x)); -} - -int cp_compat_stat(struct kstat *stat, struct compat_stat __user *statbuf) -{ - long err; - - if (stat->size > MAX_NON_LFS || !new_valid_dev(stat->dev) || - !new_valid_dev(stat->rdev)) - return -EOVERFLOW; - - err = access_ok(VERIFY_WRITE, statbuf, sizeof(*statbuf)) ? 0 : -EFAULT; - err |= __put_user(new_encode_dev(stat->dev), &statbuf->st_dev); - err |= __put_user(stat->ino, &statbuf->st_ino); - err |= __put_user(stat->mode, &statbuf->st_mode); - err |= __put_user(stat->nlink, &statbuf->st_nlink); - err |= __put_user(stat->uid, &statbuf->st_uid); - err |= __put_user(stat->gid, &statbuf->st_gid); - err |= __put_user(new_encode_dev(stat->rdev), &statbuf->st_rdev); - err |= __put_user(stat->size, &statbuf->st_size); - err |= __put_user(stat->atime.tv_sec, &statbuf->st_atime); - err |= __put_user(stat->atime.tv_nsec, &statbuf->st_atime_nsec); - err |= __put_user(stat->mtime.tv_sec, &statbuf->st_mtime); - err |= __put_user(stat->mtime.tv_nsec, &statbuf->st_mtime_nsec); - err |= __put_user(stat->ctime.tv_sec, &statbuf->st_ctime); - err |= __put_user(stat->ctime.tv_nsec, &statbuf->st_ctime_nsec); - err |= __put_user(stat->blksize, &statbuf->st_blksize); - err |= __put_user(stat->blocks, &statbuf->st_blocks); - err |= __put_user(0, &statbuf->__unused4[0]); - err |= __put_user(0, &statbuf->__unused4[1]); - - return err; -} - -/* Note: it is necessary to treat option as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_sysfs(u32 option, u32 arg1, u32 arg2) -{ - return sys_sysfs((int)option, arg1, arg2); -} - -/* Handle adjtimex compatibility. */ -struct timex32 { - u32 modes; - s32 offset, freq, maxerror, esterror; - s32 status, constant, precision, tolerance; - struct compat_timeval time; - s32 tick; - s32 ppsfreq, jitter, shift, stabil; - s32 jitcnt, calcnt, errcnt, stbcnt; - s32 :32; s32 :32; s32 :32; s32 :32; - s32 :32; s32 :32; s32 :32; s32 :32; - s32 :32; s32 :32; s32 :32; s32 :32; -}; - -extern int do_adjtimex(struct timex *); -extern void ppc_adjtimex(void); - -asmlinkage long sys32_adjtimex(struct timex32 __user *utp) -{ - struct timex txc; - int ret; - - memset(&txc, 0, sizeof(struct timex)); - - if(get_user(txc.modes, &utp->modes) || - __get_user(txc.offset, &utp->offset) || - __get_user(txc.freq, &utp->freq) || - __get_user(txc.maxerror, &utp->maxerror) || - __get_user(txc.esterror, &utp->esterror) || - __get_user(txc.status, &utp->status) || - __get_user(txc.constant, &utp->constant) || - __get_user(txc.precision, &utp->precision) || - __get_user(txc.tolerance, &utp->tolerance) || - __get_user(txc.time.tv_sec, &utp->time.tv_sec) || - __get_user(txc.time.tv_usec, &utp->time.tv_usec) || - __get_user(txc.tick, &utp->tick) || - __get_user(txc.ppsfreq, &utp->ppsfreq) || - __get_user(txc.jitter, &utp->jitter) || - __get_user(txc.shift, &utp->shift) || - __get_user(txc.stabil, &utp->stabil) || - __get_user(txc.jitcnt, &utp->jitcnt) || - __get_user(txc.calcnt, &utp->calcnt) || - __get_user(txc.errcnt, &utp->errcnt) || - __get_user(txc.stbcnt, &utp->stbcnt)) - return -EFAULT; - - ret = do_adjtimex(&txc); - - /* adjust the conversion of TB to time of day to track adjtimex */ - ppc_adjtimex(); - - if(put_user(txc.modes, &utp->modes) || - __put_user(txc.offset, &utp->offset) || - __put_user(txc.freq, &utp->freq) || - __put_user(txc.maxerror, &utp->maxerror) || - __put_user(txc.esterror, &utp->esterror) || - __put_user(txc.status, &utp->status) || - __put_user(txc.constant, &utp->constant) || - __put_user(txc.precision, &utp->precision) || - __put_user(txc.tolerance, &utp->tolerance) || - __put_user(txc.time.tv_sec, &utp->time.tv_sec) || - __put_user(txc.time.tv_usec, &utp->time.tv_usec) || - __put_user(txc.tick, &utp->tick) || - __put_user(txc.ppsfreq, &utp->ppsfreq) || - __put_user(txc.jitter, &utp->jitter) || - __put_user(txc.shift, &utp->shift) || - __put_user(txc.stabil, &utp->stabil) || - __put_user(txc.jitcnt, &utp->jitcnt) || - __put_user(txc.calcnt, &utp->calcnt) || - __put_user(txc.errcnt, &utp->errcnt) || - __put_user(txc.stbcnt, &utp->stbcnt)) - ret = -EFAULT; - - return ret; -} - -asmlinkage long sys32_pause(void) -{ - current->state = TASK_INTERRUPTIBLE; - schedule(); - - return -ERESTARTNOHAND; -} - -static inline long get_ts32(struct timespec *o, struct compat_timeval __user *i) -{ - long usec; - - if (!access_ok(VERIFY_READ, i, sizeof(*i))) - return -EFAULT; - if (__get_user(o->tv_sec, &i->tv_sec)) - return -EFAULT; - if (__get_user(usec, &i->tv_usec)) - return -EFAULT; - o->tv_nsec = usec * 1000; - return 0; -} - -static inline long put_tv32(struct compat_timeval __user *o, struct timeval *i) -{ - return (!access_ok(VERIFY_WRITE, o, sizeof(*o)) || - (__put_user(i->tv_sec, &o->tv_sec) | - __put_user(i->tv_usec, &o->tv_usec))); -} - -struct sysinfo32 { - s32 uptime; - u32 loads[3]; - u32 totalram; - u32 freeram; - u32 sharedram; - u32 bufferram; - u32 totalswap; - u32 freeswap; - unsigned short procs; - unsigned short pad; - u32 totalhigh; - u32 freehigh; - u32 mem_unit; - char _f[20-2*sizeof(int)-sizeof(int)]; -}; - -asmlinkage long sys32_sysinfo(struct sysinfo32 __user *info) -{ - struct sysinfo s; - int ret, err; - int bitcount=0; - mm_segment_t old_fs = get_fs (); - - /* The __user cast is valid due to set_fs() */ - set_fs (KERNEL_DS); - ret = sys_sysinfo((struct sysinfo __user *)&s); - set_fs (old_fs); - - /* Check to see if any memory value is too large for 32-bit and - * scale down if needed. - */ - if ((s.totalram >> 32) || (s.totalswap >> 32)) { - while (s.mem_unit < PAGE_SIZE) { - s.mem_unit <<= 1; - bitcount++; - } - s.totalram >>=bitcount; - s.freeram >>= bitcount; - s.sharedram >>= bitcount; - s.bufferram >>= bitcount; - s.totalswap >>= bitcount; - s.freeswap >>= bitcount; - s.totalhigh >>= bitcount; - s.freehigh >>= bitcount; - } - - err = put_user (s.uptime, &info->uptime); - err |= __put_user (s.loads[0], &info->loads[0]); - err |= __put_user (s.loads[1], &info->loads[1]); - err |= __put_user (s.loads[2], &info->loads[2]); - err |= __put_user (s.totalram, &info->totalram); - err |= __put_user (s.freeram, &info->freeram); - err |= __put_user (s.sharedram, &info->sharedram); - err |= __put_user (s.bufferram, &info->bufferram); - err |= __put_user (s.totalswap, &info->totalswap); - err |= __put_user (s.freeswap, &info->freeswap); - err |= __put_user (s.procs, &info->procs); - err |= __put_user (s.totalhigh, &info->totalhigh); - err |= __put_user (s.freehigh, &info->freehigh); - err |= __put_user (s.mem_unit, &info->mem_unit); - if (err) - return -EFAULT; - - return ret; -} - - - - -/* Translations due to time_t size differences. Which affects all - sorts of things, like timeval and itimerval. */ -extern struct timezone sys_tz; - -asmlinkage long sys32_gettimeofday(struct compat_timeval __user *tv, struct timezone __user *tz) -{ - if (tv) { - struct timeval ktv; - do_gettimeofday(&ktv); - if (put_tv32(tv, &ktv)) - return -EFAULT; - } - if (tz) { - if (copy_to_user(tz, &sys_tz, sizeof(sys_tz))) - return -EFAULT; - } - - return 0; -} - - - -asmlinkage long sys32_settimeofday(struct compat_timeval __user *tv, struct timezone __user *tz) -{ - struct timespec kts; - struct timezone ktz; - - if (tv) { - if (get_ts32(&kts, tv)) - return -EFAULT; - } - if (tz) { - if (copy_from_user(&ktz, tz, sizeof(ktz))) - return -EFAULT; - } - - return do_sys_settimeofday(tv ? &kts : NULL, tz ? &ktz : NULL); -} - -#ifdef CONFIG_SYSVIPC -long sys32_ipc(u32 call, u32 first, u32 second, u32 third, compat_uptr_t ptr, - u32 fifth) -{ - int version; - - version = call >> 16; /* hack for backward compatibility */ - call &= 0xffff; - - switch (call) { - - case SEMTIMEDOP: - if (fifth) - /* sign extend semid */ - return compat_sys_semtimedop((int)first, - compat_ptr(ptr), second, - compat_ptr(fifth)); - /* else fall through for normal semop() */ - case SEMOP: - /* struct sembuf is the same on 32 and 64bit :)) */ - /* sign extend semid */ - return sys_semtimedop((int)first, compat_ptr(ptr), second, - NULL); - case SEMGET: - /* sign extend key, nsems */ - return sys_semget((int)first, (int)second, third); - case SEMCTL: - /* sign extend semid, semnum */ - return compat_sys_semctl((int)first, (int)second, third, - compat_ptr(ptr)); - - case MSGSND: - /* sign extend msqid */ - return compat_sys_msgsnd((int)first, (int)second, third, - compat_ptr(ptr)); - case MSGRCV: - /* sign extend msqid, msgtyp */ - return compat_sys_msgrcv((int)first, second, (int)fifth, - third, version, compat_ptr(ptr)); - case MSGGET: - /* sign extend key */ - return sys_msgget((int)first, second); - case MSGCTL: - /* sign extend msqid */ - return compat_sys_msgctl((int)first, second, compat_ptr(ptr)); - - case SHMAT: - /* sign extend shmid */ - return compat_sys_shmat((int)first, second, third, version, - compat_ptr(ptr)); - case SHMDT: - return sys_shmdt(compat_ptr(ptr)); - case SHMGET: - /* sign extend key_t */ - return sys_shmget((int)first, second, third); - case SHMCTL: - /* sign extend shmid */ - return compat_sys_shmctl((int)first, second, compat_ptr(ptr)); - - default: - return -ENOSYS; - } - - return -ENOSYS; -} -#endif - -/* Note: it is necessary to treat out_fd and in_fd as unsigned ints, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_sendfile(u32 out_fd, u32 in_fd, compat_off_t __user * offset, u32 count) -{ - mm_segment_t old_fs = get_fs(); - int ret; - off_t of; - off_t __user *up; - - if (offset && get_user(of, offset)) - return -EFAULT; - - /* The __user pointer cast is valid because of the set_fs() */ - set_fs(KERNEL_DS); - up = offset ? (off_t __user *) &of : NULL; - ret = sys_sendfile((int)out_fd, (int)in_fd, up, count); - set_fs(old_fs); - - if (offset && put_user(of, offset)) - return -EFAULT; - - return ret; -} - -asmlinkage int sys32_sendfile64(int out_fd, int in_fd, compat_loff_t __user *offset, s32 count) -{ - mm_segment_t old_fs = get_fs(); - int ret; - loff_t lof; - loff_t __user *up; - - if (offset && get_user(lof, offset)) - return -EFAULT; - - /* The __user pointer cast is valid because of the set_fs() */ - set_fs(KERNEL_DS); - up = offset ? (loff_t __user *) &lof : NULL; - ret = sys_sendfile64(out_fd, in_fd, up, count); - set_fs(old_fs); - - if (offset && put_user(lof, offset)) - return -EFAULT; - - return ret; -} - -long sys32_execve(unsigned long a0, unsigned long a1, unsigned long a2, - unsigned long a3, unsigned long a4, unsigned long a5, - struct pt_regs *regs) -{ - int error; - char * filename; - - filename = getname((char __user *) a0); - error = PTR_ERR(filename); - if (IS_ERR(filename)) - goto out; - flush_fp_to_thread(current); - flush_altivec_to_thread(current); - - error = compat_do_execve(filename, compat_ptr(a1), compat_ptr(a2), regs); - - if (error == 0) { - task_lock(current); - current->ptrace &= ~PT_DTRACE; - task_unlock(current); - } - putname(filename); - -out: - return error; -} - -/* Set up a thread for executing a new program. */ -void start_thread32(struct pt_regs* regs, unsigned long nip, unsigned long sp) -{ - set_fs(USER_DS); - - /* - * If we exec out of a kernel thread then thread.regs will not be - * set. Do it now. - */ - if (!current->thread.regs) { - unsigned long childregs = (unsigned long)current->thread_info + - THREAD_SIZE; - childregs -= sizeof(struct pt_regs); - current->thread.regs = (struct pt_regs *)childregs; - } - - /* - * ELF_PLAT_INIT already clears all registers but it also sets r2. - * So just clear r2 here. - */ - regs->gpr[2] = 0; - - regs->nip = nip; - regs->gpr[1] = sp; - regs->msr = MSR_USER32; -#ifndef CONFIG_SMP - if (last_task_used_math == current) - last_task_used_math = 0; -#endif /* CONFIG_SMP */ - current->thread.fpscr = 0; - memset(current->thread.fpr, 0, sizeof(current->thread.fpr)); -#ifdef CONFIG_ALTIVEC -#ifndef CONFIG_SMP - if (last_task_used_altivec == current) - last_task_used_altivec = 0; -#endif /* CONFIG_SMP */ - memset(current->thread.vr, 0, sizeof(current->thread.vr)); - current->thread.vscr.u[0] = 0; - current->thread.vscr.u[1] = 0; - current->thread.vscr.u[2] = 0; - current->thread.vscr.u[3] = 0x00010000; /* Java mode disabled */ - current->thread.vrsave = 0; - current->thread.used_vr = 0; -#endif /* CONFIG_ALTIVEC */ -} - -/* Note: it is necessary to treat option as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_prctl(u32 option, u32 arg2, u32 arg3, u32 arg4, u32 arg5) -{ - return sys_prctl((int)option, - (unsigned long) arg2, - (unsigned long) arg3, - (unsigned long) arg4, - (unsigned long) arg5); -} - -/* Note: it is necessary to treat pid as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_sched_rr_get_interval(u32 pid, struct compat_timespec __user *interval) -{ - struct timespec t; - int ret; - mm_segment_t old_fs = get_fs (); - - /* The __user pointer cast is valid because of the set_fs() */ - set_fs (KERNEL_DS); - ret = sys_sched_rr_get_interval((int)pid, (struct timespec __user *) &t); - set_fs (old_fs); - if (put_compat_timespec(&t, interval)) - return -EFAULT; - return ret; -} - -asmlinkage int sys32_pciconfig_read(u32 bus, u32 dfn, u32 off, u32 len, u32 ubuf) -{ - return sys_pciconfig_read((unsigned long) bus, - (unsigned long) dfn, - (unsigned long) off, - (unsigned long) len, - compat_ptr(ubuf)); -} - -asmlinkage int sys32_pciconfig_write(u32 bus, u32 dfn, u32 off, u32 len, u32 ubuf) -{ - return sys_pciconfig_write((unsigned long) bus, - (unsigned long) dfn, - (unsigned long) off, - (unsigned long) len, - compat_ptr(ubuf)); -} - -asmlinkage int sys32_pciconfig_iobase(u32 which, u32 in_bus, u32 in_devfn) -{ - return sys_pciconfig_iobase(which, in_bus, in_devfn); -} - - -/* Note: it is necessary to treat mode as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_access(const char __user * filename, u32 mode) -{ - return sys_access(filename, (int)mode); -} - - -/* Note: it is necessary to treat mode as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_creat(const char __user * pathname, u32 mode) -{ - return sys_creat(pathname, (int)mode); -} - - -/* Note: it is necessary to treat pid and options as unsigned ints, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_waitpid(u32 pid, unsigned int __user * stat_addr, u32 options) -{ - return sys_waitpid((int)pid, stat_addr, (int)options); -} - - -/* Note: it is necessary to treat gidsetsize as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_getgroups(u32 gidsetsize, gid_t __user *grouplist) -{ - return sys_getgroups((int)gidsetsize, grouplist); -} - - -/* Note: it is necessary to treat pid as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_getpgid(u32 pid) -{ - return sys_getpgid((int)pid); -} - - - -/* Note: it is necessary to treat pid as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_getsid(u32 pid) -{ - return sys_getsid((int)pid); -} - - -/* Note: it is necessary to treat pid and sig as unsigned ints, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_kill(u32 pid, u32 sig) -{ - return sys_kill((int)pid, (int)sig); -} - - -/* Note: it is necessary to treat mode as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_mkdir(const char __user * pathname, u32 mode) -{ - return sys_mkdir(pathname, (int)mode); -} - -long sys32_nice(u32 increment) -{ - /* sign extend increment */ - return sys_nice((int)increment); -} - -off_t ppc32_lseek(unsigned int fd, u32 offset, unsigned int origin) -{ - /* sign extend n */ - return sys_lseek(fd, (int)offset, origin); -} - -/* Note: it is necessary to treat bufsiz as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_readlink(const char __user * path, char __user * buf, u32 bufsiz) -{ - return sys_readlink(path, buf, (int)bufsiz); -} - -/* Note: it is necessary to treat option as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_sched_get_priority_max(u32 policy) -{ - return sys_sched_get_priority_max((int)policy); -} - - -/* Note: it is necessary to treat policy as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_sched_get_priority_min(u32 policy) -{ - return sys_sched_get_priority_min((int)policy); -} - - -/* Note: it is necessary to treat pid as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_sched_getparam(u32 pid, struct sched_param __user *param) -{ - return sys_sched_getparam((int)pid, param); -} - - -/* Note: it is necessary to treat pid as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_sched_getscheduler(u32 pid) -{ - return sys_sched_getscheduler((int)pid); -} - - -/* Note: it is necessary to treat pid as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_sched_setparam(u32 pid, struct sched_param __user *param) -{ - return sys_sched_setparam((int)pid, param); -} - - -/* Note: it is necessary to treat pid and policy as unsigned ints, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_sched_setscheduler(u32 pid, u32 policy, struct sched_param __user *param) -{ - return sys_sched_setscheduler((int)pid, (int)policy, param); -} - - -/* Note: it is necessary to treat len as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_setdomainname(char __user *name, u32 len) -{ - return sys_setdomainname(name, (int)len); -} - - -/* Note: it is necessary to treat gidsetsize as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_setgroups(u32 gidsetsize, gid_t __user *grouplist) -{ - return sys_setgroups((int)gidsetsize, grouplist); -} - - -asmlinkage long sys32_sethostname(char __user *name, u32 len) -{ - /* sign extend len */ - return sys_sethostname(name, (int)len); -} - - -/* Note: it is necessary to treat pid and pgid as unsigned ints, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_setpgid(u32 pid, u32 pgid) -{ - return sys_setpgid((int)pid, (int)pgid); -} - -long sys32_getpriority(u32 which, u32 who) -{ - /* sign extend which and who */ - return sys_getpriority((int)which, (int)who); -} - -long sys32_setpriority(u32 which, u32 who, u32 niceval) -{ - /* sign extend which, who and niceval */ - return sys_setpriority((int)which, (int)who, (int)niceval); -} - -long sys32_ioprio_get(u32 which, u32 who) -{ - /* sign extend which and who */ - return sys_ioprio_get((int)which, (int)who); -} - -long sys32_ioprio_set(u32 which, u32 who, u32 ioprio) -{ - /* sign extend which, who and ioprio */ - return sys_ioprio_set((int)which, (int)who, (int)ioprio); -} - -/* Note: it is necessary to treat newmask as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_ssetmask(u32 newmask) -{ - return sys_ssetmask((int) newmask); -} - -asmlinkage long sys32_syslog(u32 type, char __user * buf, u32 len) -{ - /* sign extend len */ - return sys_syslog(type, buf, (int)len); -} - - -/* Note: it is necessary to treat mask as an unsigned int, - * with the corresponding cast to a signed int to insure that the - * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) - * and the register representation of a signed int (msr in 64-bit mode) is performed. - */ -asmlinkage long sys32_umask(u32 mask) -{ - return sys_umask((int)mask); -} - -#ifdef CONFIG_SYSCTL -struct __sysctl_args32 { - u32 name; - int nlen; - u32 oldval; - u32 oldlenp; - u32 newval; - u32 newlen; - u32 __unused[4]; -}; - -asmlinkage long sys32_sysctl(struct __sysctl_args32 __user *args) -{ - struct __sysctl_args32 tmp; - int error; - size_t oldlen; - size_t __user *oldlenp = NULL; - unsigned long addr = (((unsigned long)&args->__unused[0]) + 7) & ~7; - - if (copy_from_user(&tmp, args, sizeof(tmp))) - return -EFAULT; - - if (tmp.oldval && tmp.oldlenp) { - /* Duh, this is ugly and might not work if sysctl_args - is in read-only memory, but do_sysctl does indirectly - a lot of uaccess in both directions and we'd have to - basically copy the whole sysctl.c here, and - glibc's __sysctl uses rw memory for the structure - anyway. */ - oldlenp = (size_t __user *)addr; - if (get_user(oldlen, (compat_size_t __user *)compat_ptr(tmp.oldlenp)) || - put_user(oldlen, oldlenp)) - return -EFAULT; - } - - lock_kernel(); - error = do_sysctl(compat_ptr(tmp.name), tmp.nlen, - compat_ptr(tmp.oldval), oldlenp, - compat_ptr(tmp.newval), tmp.newlen); - unlock_kernel(); - if (oldlenp) { - if (!error) { - if (get_user(oldlen, oldlenp) || - put_user(oldlen, (compat_size_t __user *)compat_ptr(tmp.oldlenp))) - error = -EFAULT; - } - copy_to_user(args->__unused, tmp.__unused, sizeof(tmp.__unused)); - } - return error; -} -#endif - -asmlinkage int sys32_uname(struct old_utsname __user * name) -{ - int err = 0; - - down_read(&uts_sem); - if (copy_to_user(name, &system_utsname, sizeof(*name))) - err = -EFAULT; - up_read(&uts_sem); - if (!err && personality(current->personality) == PER_LINUX32) { - /* change "ppc64" to "ppc" */ - if (__put_user(0, name->machine + 3) - || __put_user(0, name->machine + 4)) - err = -EFAULT; - } - return err; -} - -asmlinkage int sys32_olduname(struct oldold_utsname __user * name) -{ - int error; - - if (!access_ok(VERIFY_WRITE,name,sizeof(struct oldold_utsname))) - return -EFAULT; - - down_read(&uts_sem); - error = __copy_to_user(&name->sysname,&system_utsname.sysname,__OLD_UTS_LEN); - error |= __put_user(0,name->sysname+__OLD_UTS_LEN); - error |= __copy_to_user(&name->nodename,&system_utsname.nodename,__OLD_UTS_LEN); - error |= __put_user(0,name->nodename+__OLD_UTS_LEN); - error |= __copy_to_user(&name->release,&system_utsname.release,__OLD_UTS_LEN); - error |= __put_user(0,name->release+__OLD_UTS_LEN); - error |= __copy_to_user(&name->version,&system_utsname.version,__OLD_UTS_LEN); - error |= __put_user(0,name->version+__OLD_UTS_LEN); - error |= __copy_to_user(&name->machine,&system_utsname.machine,__OLD_UTS_LEN); - error |= __put_user(0,name->machine+__OLD_UTS_LEN); - if (personality(current->personality) == PER_LINUX32) { - /* change "ppc64" to "ppc" */ - error |= __put_user(0, name->machine + 3); - error |= __put_user(0, name->machine + 4); - } - - up_read(&uts_sem); - - error = error ? -EFAULT : 0; - - return error; -} - -unsigned long sys32_mmap2(unsigned long addr, size_t len, - unsigned long prot, unsigned long flags, - unsigned long fd, unsigned long pgoff) -{ - /* This should remain 12 even if PAGE_SIZE changes */ - return sys_mmap(addr, len, prot, flags, fd, pgoff << 12); -} - -int get_compat_timeval(struct timeval *tv, struct compat_timeval __user *ctv) -{ - return (!access_ok(VERIFY_READ, ctv, sizeof(*ctv)) || - __get_user(tv->tv_sec, &ctv->tv_sec) || - __get_user(tv->tv_usec, &ctv->tv_usec)) ? -EFAULT : 0; -} - -asmlinkage long sys32_utimes(char __user *filename, struct compat_timeval __user *tvs) -{ - struct timeval ktvs[2], *ptr; - - ptr = NULL; - if (tvs) { - if (get_compat_timeval(&ktvs[0], &tvs[0]) || - get_compat_timeval(&ktvs[1], &tvs[1])) - return -EFAULT; - ptr = ktvs; - } - - return do_utimes(filename, ptr); -} - -long sys32_tgkill(u32 tgid, u32 pid, int sig) -{ - /* sign extend tgid, pid */ - return sys_tgkill((int)tgid, (int)pid, sig); -} - -/* - * long long munging: - * The 32 bit ABI passes long longs in an odd even register pair. - */ - -compat_ssize_t sys32_pread64(unsigned int fd, char __user *ubuf, compat_size_t count, - u32 reg6, u32 poshi, u32 poslo) -{ - return sys_pread64(fd, ubuf, count, ((loff_t)poshi << 32) | poslo); -} - -compat_ssize_t sys32_pwrite64(unsigned int fd, char __user *ubuf, compat_size_t count, - u32 reg6, u32 poshi, u32 poslo) -{ - return sys_pwrite64(fd, ubuf, count, ((loff_t)poshi << 32) | poslo); -} - -compat_ssize_t sys32_readahead(int fd, u32 r4, u32 offhi, u32 offlo, u32 count) -{ - return sys_readahead(fd, ((loff_t)offhi << 32) | offlo, count); -} - -asmlinkage int sys32_truncate64(const char __user * path, u32 reg4, - unsigned long high, unsigned long low) -{ - return sys_truncate(path, (high << 32) | low); -} - -asmlinkage int sys32_ftruncate64(unsigned int fd, u32 reg4, unsigned long high, - unsigned long low) -{ - return sys_ftruncate(fd, (high << 32) | low); -} - -long ppc32_lookup_dcookie(u32 cookie_high, u32 cookie_low, char __user *buf, - size_t len) -{ - return sys_lookup_dcookie((u64)cookie_high << 32 | cookie_low, - buf, len); -} - -long ppc32_fadvise64(int fd, u32 unused, u32 offset_high, u32 offset_low, - size_t len, int advice) -{ - return sys_fadvise64(fd, (u64)offset_high << 32 | offset_low, len, - advice); -} - -long ppc32_fadvise64_64(int fd, int advice, u32 offset_high, u32 offset_low, - u32 len_high, u32 len_low) -{ - return sys_fadvise64(fd, (u64)offset_high << 32 | offset_low, - (u64)len_high << 32 | len_low, advice); -} - -long ppc32_timer_create(clockid_t clock, - struct compat_sigevent __user *ev32, - timer_t __user *timer_id) -{ - sigevent_t event; - timer_t t; - long err; - mm_segment_t savefs; - - if (ev32 == NULL) - return sys_timer_create(clock, NULL, timer_id); - - if (get_compat_sigevent(&event, ev32)) - return -EFAULT; - - if (!access_ok(VERIFY_WRITE, timer_id, sizeof(timer_t))) - return -EFAULT; - - savefs = get_fs(); - set_fs(KERNEL_DS); - /* The __user pointer casts are valid due to the set_fs() */ - err = sys_timer_create(clock, - (sigevent_t __user *) &event, - (timer_t __user *) &t); - set_fs(savefs); - - if (err == 0) - err = __put_user(t, timer_id); - - return err; -} - -asmlinkage long sys32_add_key(const char __user *_type, - const char __user *_description, - const void __user *_payload, - u32 plen, - u32 ringid) -{ - return sys_add_key(_type, _description, _payload, plen, ringid); -} - -asmlinkage long sys32_request_key(const char __user *_type, - const char __user *_description, - const char __user *_callout_info, - u32 destringid) -{ - return sys_request_key(_type, _description, _callout_info, destringid); -} - diff --git a/arch/ppc64/kernel/syscalls.c b/arch/ppc64/kernel/syscalls.c deleted file mode 100644 index 05f16633bd2c..000000000000 --- a/arch/ppc64/kernel/syscalls.c +++ /dev/null @@ -1,263 +0,0 @@ -/* - * linux/arch/ppc64/kernel/sys_ppc.c - * - * PowerPC version - * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org) - * - * Derived from "arch/i386/kernel/sys_i386.c" - * Adapted from the i386 version by Gary Thomas - * Modified by Cort Dougan (cort@cs.nmt.edu) - * and Paul Mackerras (paulus@cs.anu.edu.au). - * - * This file contains various random system calls that - * have a non-standard calling sequence on the Linux/PPC - * platform. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - */ - -#include <linux/errno.h> -#include <linux/sched.h> -#include <linux/syscalls.h> -#include <linux/mm.h> -#include <linux/smp.h> -#include <linux/smp_lock.h> -#include <linux/sem.h> -#include <linux/msg.h> -#include <linux/shm.h> -#include <linux/stat.h> -#include <linux/mman.h> -#include <linux/sys.h> -#include <linux/ipc.h> -#include <linux/utsname.h> -#include <linux/file.h> -#include <linux/init.h> -#include <linux/personality.h> - -#include <asm/uaccess.h> -#include <asm/ipc.h> -#include <asm/semaphore.h> -#include <asm/time.h> -#include <asm/unistd.h> - -extern unsigned long wall_jiffies; - - -/* - * sys_ipc() is the de-multiplexer for the SysV IPC calls.. - * - * This is really horribly ugly. - */ -asmlinkage int -sys_ipc (uint call, int first, unsigned long second, long third, - void __user *ptr, long fifth) -{ - int version, ret; - - version = call >> 16; /* hack for backward compatibility */ - call &= 0xffff; - - ret = -ENOSYS; - switch (call) { - case SEMOP: - ret = sys_semtimedop(first, (struct sembuf __user *)ptr, - (unsigned)second, NULL); - break; - case SEMTIMEDOP: - ret = sys_semtimedop(first, (struct sembuf __user *)ptr, - (unsigned)second, - (const struct timespec __user *) fifth); - break; - case SEMGET: - ret = sys_semget (first, (int)second, third); - break; - case SEMCTL: { - union semun fourth; - - ret = -EINVAL; - if (!ptr) - break; - if ((ret = get_user(fourth.__pad, (void __user * __user *)ptr))) - break; - ret = sys_semctl(first, (int)second, third, fourth); - break; - } - case MSGSND: - ret = sys_msgsnd(first, (struct msgbuf __user *)ptr, - (size_t)second, third); - break; - case MSGRCV: - switch (version) { - case 0: { - struct ipc_kludge tmp; - - ret = -EINVAL; - if (!ptr) - break; - if ((ret = copy_from_user(&tmp, - (struct ipc_kludge __user *) ptr, - sizeof (tmp)) ? -EFAULT : 0)) - break; - ret = sys_msgrcv(first, tmp.msgp, (size_t) second, - tmp.msgtyp, third); - break; - } - default: - ret = sys_msgrcv (first, (struct msgbuf __user *) ptr, - (size_t)second, fifth, third); - break; - } - break; - case MSGGET: - ret = sys_msgget ((key_t)first, (int)second); - break; - case MSGCTL: - ret = sys_msgctl(first, (int)second, - (struct msqid_ds __user *)ptr); - break; - case SHMAT: - switch (version) { - default: { - ulong raddr; - ret = do_shmat(first, (char __user *) ptr, - (int)second, &raddr); - if (ret) - break; - ret = put_user (raddr, (ulong __user *) third); - break; - } - case 1: /* iBCS2 emulator entry point */ - ret = -EINVAL; - if (!segment_eq(get_fs(), get_ds())) - break; - ret = do_shmat(first, (char __user *)ptr, - (int)second, (ulong *)third); - break; - } - break; - case SHMDT: - ret = sys_shmdt ((char __user *)ptr); - break; - case SHMGET: - ret = sys_shmget (first, (size_t)second, third); - break; - case SHMCTL: - ret = sys_shmctl(first, (int)second, - (struct shmid_ds __user *)ptr); - break; - } - - return ret; -} - -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -asmlinkage int sys_pipe(int __user *fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - - return error; -} - -unsigned long sys_mmap(unsigned long addr, size_t len, - unsigned long prot, unsigned long flags, - unsigned long fd, off_t offset) -{ - struct file * file = NULL; - unsigned long ret = -EBADF; - - if (!(flags & MAP_ANONYMOUS)) { - if (!(file = fget(fd))) - goto out; - } - - flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); - down_write(¤t->mm->mmap_sem); - ret = do_mmap(file, addr, len, prot, flags, offset); - up_write(¤t->mm->mmap_sem); - if (file) - fput(file); - -out: - return ret; -} - -long ppc64_personality(unsigned long personality) -{ - long ret; - - if (personality(current->personality) == PER_LINUX32 - && personality == PER_LINUX) - personality = PER_LINUX32; - ret = sys_personality(personality); - if (ret == PER_LINUX32) - ret = PER_LINUX; - return ret; -} - -long ppc64_newuname(struct new_utsname __user * name) -{ - int err = 0; - - down_read(&uts_sem); - if (copy_to_user(name, &system_utsname, sizeof(*name))) - err = -EFAULT; - up_read(&uts_sem); - if (!err && personality(current->personality) == PER_LINUX32) { - /* change ppc64 to ppc */ - if (__put_user(0, name->machine + 3) - || __put_user(0, name->machine + 4)) - err = -EFAULT; - } - return err; -} - -asmlinkage time_t sys64_time(time_t __user * tloc) -{ - time_t secs; - time_t usecs; - - long tb_delta = tb_ticks_since(tb_last_stamp); - tb_delta += (jiffies - wall_jiffies) * tb_ticks_per_jiffy; - - secs = xtime.tv_sec; - usecs = (xtime.tv_nsec/1000) + tb_delta / tb_ticks_per_usec; - while (usecs >= USEC_PER_SEC) { - ++secs; - usecs -= USEC_PER_SEC; - } - - if (tloc) { - if (put_user(secs,tloc)) - secs = -EFAULT; - } - - return secs; -} - -void do_show_syscall(unsigned long r3, unsigned long r4, unsigned long r5, - unsigned long r6, unsigned long r7, unsigned long r8, - struct pt_regs *regs) -{ - printk("syscall %ld(%lx, %lx, %lx, %lx, %lx, %lx) regs=%p current=%p" - " cpu=%d\n", regs->gpr[0], r3, r4, r5, r6, r7, r8, regs, - current, smp_processor_id()); -} - -void do_show_syscall_exit(unsigned long r3) -{ - printk(" -> %lx, current=%p cpu=%d\n", r3, current, smp_processor_id()); -} diff --git a/arch/ppc64/kernel/time.c b/arch/ppc64/kernel/time.c deleted file mode 100644 index b56c6a324e17..000000000000 --- a/arch/ppc64/kernel/time.c +++ /dev/null @@ -1,881 +0,0 @@ -/* - * - * Common time routines among all ppc machines. - * - * Written by Cort Dougan (cort@cs.nmt.edu) to merge - * Paul Mackerras' version and mine for PReP and Pmac. - * MPC8xx/MBX changes by Dan Malek (dmalek@jlc.net). - * Converted for 64-bit by Mike Corrigan (mikejc@us.ibm.com) - * - * First round of bugfixes by Gabriel Paubert (paubert@iram.es) - * to make clock more stable (2.4.0-test5). The only thing - * that this code assumes is that the timebases have been synchronized - * by firmware on SMP and are never stopped (never do sleep - * on SMP then, nap and doze are OK). - * - * Speeded up do_gettimeofday by getting rid of references to - * xtime (which required locks for consistency). (mikejc@us.ibm.com) - * - * TODO (not necessarily in this file): - * - improve precision and reproducibility of timebase frequency - * measurement at boot time. (for iSeries, we calibrate the timebase - * against the Titan chip's clock.) - * - for astronomical applications: add a new function to get - * non ambiguous timestamps even around leap seconds. This needs - * a new timestamp format and a good name. - * - * 1997-09-10 Updated NTP code according to technical memorandum Jan '96 - * "A Kernel Model for Precision Timekeeping" by Dave Mills - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/config.h> -#include <linux/errno.h> -#include <linux/module.h> -#include <linux/sched.h> -#include <linux/kernel.h> -#include <linux/param.h> -#include <linux/string.h> -#include <linux/mm.h> -#include <linux/interrupt.h> -#include <linux/timex.h> -#include <linux/kernel_stat.h> -#include <linux/mc146818rtc.h> -#include <linux/time.h> -#include <linux/init.h> -#include <linux/profile.h> -#include <linux/cpu.h> -#include <linux/security.h> - -#include <asm/io.h> -#include <asm/processor.h> -#include <asm/nvram.h> -#include <asm/cache.h> -#include <asm/machdep.h> -#ifdef CONFIG_PPC_ISERIES -#include <asm/iSeries/ItLpQueue.h> -#include <asm/iSeries/HvCallXm.h> -#endif -#include <asm/uaccess.h> -#include <asm/time.h> -#include <asm/ppcdebug.h> -#include <asm/prom.h> -#include <asm/sections.h> -#include <asm/systemcfg.h> -#include <asm/firmware.h> - -u64 jiffies_64 __cacheline_aligned_in_smp = INITIAL_JIFFIES; - -EXPORT_SYMBOL(jiffies_64); - -/* keep track of when we need to update the rtc */ -time_t last_rtc_update; -extern int piranha_simulator; -#ifdef CONFIG_PPC_ISERIES -unsigned long iSeries_recal_titan = 0; -unsigned long iSeries_recal_tb = 0; -static unsigned long first_settimeofday = 1; -#endif - -#define XSEC_PER_SEC (1024*1024) - -unsigned long tb_ticks_per_jiffy; -unsigned long tb_ticks_per_usec = 100; /* sane default */ -EXPORT_SYMBOL(tb_ticks_per_usec); -unsigned long tb_ticks_per_sec; -unsigned long tb_to_xs; -unsigned tb_to_us; -unsigned long processor_freq; -DEFINE_SPINLOCK(rtc_lock); -EXPORT_SYMBOL_GPL(rtc_lock); - -unsigned long tb_to_ns_scale; -unsigned long tb_to_ns_shift; - -struct gettimeofday_struct do_gtod; - -extern unsigned long wall_jiffies; -extern int smp_tb_synchronized; - -extern struct timezone sys_tz; - -void ppc_adjtimex(void); - -static unsigned adjusting_time = 0; - -unsigned long ppc_proc_freq; -unsigned long ppc_tb_freq; - -static __inline__ void timer_check_rtc(void) -{ - /* - * update the rtc when needed, this should be performed on the - * right fraction of a second. Half or full second ? - * Full second works on mk48t59 clocks, others need testing. - * Note that this update is basically only used through - * the adjtimex system calls. Setting the HW clock in - * any other way is a /dev/rtc and userland business. - * This is still wrong by -0.5/+1.5 jiffies because of the - * timer interrupt resolution and possible delay, but here we - * hit a quantization limit which can only be solved by higher - * resolution timers and decoupling time management from timer - * interrupts. This is also wrong on the clocks - * which require being written at the half second boundary. - * We should have an rtc call that only sets the minutes and - * seconds like on Intel to avoid problems with non UTC clocks. - */ - if (ntp_synced() && - xtime.tv_sec - last_rtc_update >= 659 && - abs((xtime.tv_nsec/1000) - (1000000-1000000/HZ)) < 500000/HZ && - jiffies - wall_jiffies == 1) { - struct rtc_time tm; - to_tm(xtime.tv_sec+1, &tm); - tm.tm_year -= 1900; - tm.tm_mon -= 1; - if (ppc_md.set_rtc_time(&tm) == 0) - last_rtc_update = xtime.tv_sec+1; - else - /* Try again one minute later */ - last_rtc_update += 60; - } -} - -/* - * This version of gettimeofday has microsecond resolution. - */ -static inline void __do_gettimeofday(struct timeval *tv, unsigned long tb_val) -{ - unsigned long sec, usec, tb_ticks; - unsigned long xsec, tb_xsec; - struct gettimeofday_vars * temp_varp; - unsigned long temp_tb_to_xs, temp_stamp_xsec; - - /* - * These calculations are faster (gets rid of divides) - * if done in units of 1/2^20 rather than microseconds. - * The conversion to microseconds at the end is done - * without a divide (and in fact, without a multiply) - */ - temp_varp = do_gtod.varp; - tb_ticks = tb_val - temp_varp->tb_orig_stamp; - temp_tb_to_xs = temp_varp->tb_to_xs; - temp_stamp_xsec = temp_varp->stamp_xsec; - tb_xsec = mulhdu( tb_ticks, temp_tb_to_xs ); - xsec = temp_stamp_xsec + tb_xsec; - sec = xsec / XSEC_PER_SEC; - xsec -= sec * XSEC_PER_SEC; - usec = (xsec * USEC_PER_SEC)/XSEC_PER_SEC; - - tv->tv_sec = sec; - tv->tv_usec = usec; -} - -void do_gettimeofday(struct timeval *tv) -{ - __do_gettimeofday(tv, get_tb()); -} - -EXPORT_SYMBOL(do_gettimeofday); - -/* Synchronize xtime with do_gettimeofday */ - -static inline void timer_sync_xtime(unsigned long cur_tb) -{ - struct timeval my_tv; - - __do_gettimeofday(&my_tv, cur_tb); - - if (xtime.tv_sec <= my_tv.tv_sec) { - xtime.tv_sec = my_tv.tv_sec; - xtime.tv_nsec = my_tv.tv_usec * 1000; - } -} - -/* - * When the timebase - tb_orig_stamp gets too big, we do a manipulation - * between tb_orig_stamp and stamp_xsec. The goal here is to keep the - * difference tb - tb_orig_stamp small enough to always fit inside a - * 32 bits number. This is a requirement of our fast 32 bits userland - * implementation in the vdso. If we "miss" a call to this function - * (interrupt latency, CPU locked in a spinlock, ...) and we end up - * with a too big difference, then the vdso will fallback to calling - * the syscall - */ -static __inline__ void timer_recalc_offset(unsigned long cur_tb) -{ - struct gettimeofday_vars * temp_varp; - unsigned temp_idx; - unsigned long offset, new_stamp_xsec, new_tb_orig_stamp; - - if (((cur_tb - do_gtod.varp->tb_orig_stamp) & 0x80000000u) == 0) - return; - - temp_idx = (do_gtod.var_idx == 0); - temp_varp = &do_gtod.vars[temp_idx]; - - new_tb_orig_stamp = cur_tb; - offset = new_tb_orig_stamp - do_gtod.varp->tb_orig_stamp; - new_stamp_xsec = do_gtod.varp->stamp_xsec + mulhdu(offset, do_gtod.varp->tb_to_xs); - - temp_varp->tb_to_xs = do_gtod.varp->tb_to_xs; - temp_varp->tb_orig_stamp = new_tb_orig_stamp; - temp_varp->stamp_xsec = new_stamp_xsec; - smp_mb(); - do_gtod.varp = temp_varp; - do_gtod.var_idx = temp_idx; - - ++(systemcfg->tb_update_count); - smp_wmb(); - systemcfg->tb_orig_stamp = new_tb_orig_stamp; - systemcfg->stamp_xsec = new_stamp_xsec; - smp_wmb(); - ++(systemcfg->tb_update_count); -} - -#ifdef CONFIG_SMP -unsigned long profile_pc(struct pt_regs *regs) -{ - unsigned long pc = instruction_pointer(regs); - - if (in_lock_functions(pc)) - return regs->link; - - return pc; -} -EXPORT_SYMBOL(profile_pc); -#endif - -#ifdef CONFIG_PPC_ISERIES - -/* - * This function recalibrates the timebase based on the 49-bit time-of-day - * value in the Titan chip. The Titan is much more accurate than the value - * returned by the service processor for the timebase frequency. - */ - -static void iSeries_tb_recal(void) -{ - struct div_result divres; - unsigned long titan, tb; - tb = get_tb(); - titan = HvCallXm_loadTod(); - if ( iSeries_recal_titan ) { - unsigned long tb_ticks = tb - iSeries_recal_tb; - unsigned long titan_usec = (titan - iSeries_recal_titan) >> 12; - unsigned long new_tb_ticks_per_sec = (tb_ticks * USEC_PER_SEC)/titan_usec; - unsigned long new_tb_ticks_per_jiffy = (new_tb_ticks_per_sec+(HZ/2))/HZ; - long tick_diff = new_tb_ticks_per_jiffy - tb_ticks_per_jiffy; - char sign = '+'; - /* make sure tb_ticks_per_sec and tb_ticks_per_jiffy are consistent */ - new_tb_ticks_per_sec = new_tb_ticks_per_jiffy * HZ; - - if ( tick_diff < 0 ) { - tick_diff = -tick_diff; - sign = '-'; - } - if ( tick_diff ) { - if ( tick_diff < tb_ticks_per_jiffy/25 ) { - printk( "Titan recalibrate: new tb_ticks_per_jiffy = %lu (%c%ld)\n", - new_tb_ticks_per_jiffy, sign, tick_diff ); - tb_ticks_per_jiffy = new_tb_ticks_per_jiffy; - tb_ticks_per_sec = new_tb_ticks_per_sec; - div128_by_32( XSEC_PER_SEC, 0, tb_ticks_per_sec, &divres ); - do_gtod.tb_ticks_per_sec = tb_ticks_per_sec; - tb_to_xs = divres.result_low; - do_gtod.varp->tb_to_xs = tb_to_xs; - systemcfg->tb_ticks_per_sec = tb_ticks_per_sec; - systemcfg->tb_to_xs = tb_to_xs; - } - else { - printk( "Titan recalibrate: FAILED (difference > 4 percent)\n" - " new tb_ticks_per_jiffy = %lu\n" - " old tb_ticks_per_jiffy = %lu\n", - new_tb_ticks_per_jiffy, tb_ticks_per_jiffy ); - } - } - } - iSeries_recal_titan = titan; - iSeries_recal_tb = tb; -} -#endif - -/* - * For iSeries shared processors, we have to let the hypervisor - * set the hardware decrementer. We set a virtual decrementer - * in the lppaca and call the hypervisor if the virtual - * decrementer is less than the current value in the hardware - * decrementer. (almost always the new decrementer value will - * be greater than the current hardware decementer so the hypervisor - * call will not be needed) - */ - -unsigned long tb_last_stamp __cacheline_aligned_in_smp; - -/* - * timer_interrupt - gets called when the decrementer overflows, - * with interrupts disabled. - */ -int timer_interrupt(struct pt_regs * regs) -{ - int next_dec; - unsigned long cur_tb; - struct paca_struct *lpaca = get_paca(); - unsigned long cpu = smp_processor_id(); - - irq_enter(); - - profile_tick(CPU_PROFILING, regs); - - lpaca->lppaca.int_dword.fields.decr_int = 0; - - while (lpaca->next_jiffy_update_tb <= (cur_tb = get_tb())) { - /* - * We cannot disable the decrementer, so in the period - * between this cpu's being marked offline in cpu_online_map - * and calling stop-self, it is taking timer interrupts. - * Avoid calling into the scheduler rebalancing code if this - * is the case. - */ - if (!cpu_is_offline(cpu)) - update_process_times(user_mode(regs)); - /* - * No need to check whether cpu is offline here; boot_cpuid - * should have been fixed up by now. - */ - if (cpu == boot_cpuid) { - write_seqlock(&xtime_lock); - tb_last_stamp = lpaca->next_jiffy_update_tb; - timer_recalc_offset(lpaca->next_jiffy_update_tb); - do_timer(regs); - timer_sync_xtime(lpaca->next_jiffy_update_tb); - timer_check_rtc(); - write_sequnlock(&xtime_lock); - if ( adjusting_time && (time_adjust == 0) ) - ppc_adjtimex(); - } - lpaca->next_jiffy_update_tb += tb_ticks_per_jiffy; - } - - next_dec = lpaca->next_jiffy_update_tb - cur_tb; - if (next_dec > lpaca->default_decr) - next_dec = lpaca->default_decr; - set_dec(next_dec); - -#ifdef CONFIG_PPC_ISERIES - if (hvlpevent_is_pending()) - process_hvlpevents(regs); -#endif - - /* collect purr register values often, for accurate calculations */ - if (firmware_has_feature(FW_FEATURE_SPLPAR)) { - struct cpu_usage *cu = &__get_cpu_var(cpu_usage_array); - cu->current_tb = mfspr(SPRN_PURR); - } - - irq_exit(); - - return 1; -} - -/* - * Scheduler clock - returns current time in nanosec units. - * - * Note: mulhdu(a, b) (multiply high double unsigned) returns - * the high 64 bits of a * b, i.e. (a * b) >> 64, where a and b - * are 64-bit unsigned numbers. - */ -unsigned long long sched_clock(void) -{ - return mulhdu(get_tb(), tb_to_ns_scale) << tb_to_ns_shift; -} - -int do_settimeofday(struct timespec *tv) -{ - time_t wtm_sec, new_sec = tv->tv_sec; - long wtm_nsec, new_nsec = tv->tv_nsec; - unsigned long flags; - unsigned long delta_xsec; - long int tb_delta; - unsigned long new_xsec; - - if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC) - return -EINVAL; - - write_seqlock_irqsave(&xtime_lock, flags); - /* Updating the RTC is not the job of this code. If the time is - * stepped under NTP, the RTC will be update after STA_UNSYNC - * is cleared. Tool like clock/hwclock either copy the RTC - * to the system time, in which case there is no point in writing - * to the RTC again, or write to the RTC but then they don't call - * settimeofday to perform this operation. - */ -#ifdef CONFIG_PPC_ISERIES - if ( first_settimeofday ) { - iSeries_tb_recal(); - first_settimeofday = 0; - } -#endif - tb_delta = tb_ticks_since(tb_last_stamp); - tb_delta += (jiffies - wall_jiffies) * tb_ticks_per_jiffy; - - new_nsec -= tb_delta / tb_ticks_per_usec / 1000; - - wtm_sec = wall_to_monotonic.tv_sec + (xtime.tv_sec - new_sec); - wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - new_nsec); - - set_normalized_timespec(&xtime, new_sec, new_nsec); - set_normalized_timespec(&wall_to_monotonic, wtm_sec, wtm_nsec); - - /* In case of a large backwards jump in time with NTP, we want the - * clock to be updated as soon as the PLL is again in lock. - */ - last_rtc_update = new_sec - 658; - - ntp_clear(); - - delta_xsec = mulhdu( (tb_last_stamp-do_gtod.varp->tb_orig_stamp), - do_gtod.varp->tb_to_xs ); - - new_xsec = (new_nsec * XSEC_PER_SEC) / NSEC_PER_SEC; - new_xsec += new_sec * XSEC_PER_SEC; - if ( new_xsec > delta_xsec ) { - do_gtod.varp->stamp_xsec = new_xsec - delta_xsec; - systemcfg->stamp_xsec = new_xsec - delta_xsec; - } - else { - /* This is only for the case where the user is setting the time - * way back to a time such that the boot time would have been - * before 1970 ... eg. we booted ten days ago, and we are setting - * the time to Jan 5, 1970 */ - do_gtod.varp->stamp_xsec = new_xsec; - do_gtod.varp->tb_orig_stamp = tb_last_stamp; - systemcfg->stamp_xsec = new_xsec; - systemcfg->tb_orig_stamp = tb_last_stamp; - } - - systemcfg->tz_minuteswest = sys_tz.tz_minuteswest; - systemcfg->tz_dsttime = sys_tz.tz_dsttime; - - write_sequnlock_irqrestore(&xtime_lock, flags); - clock_was_set(); - return 0; -} - -EXPORT_SYMBOL(do_settimeofday); - -#if defined(CONFIG_PPC_PSERIES) || defined(CONFIG_PPC_MAPLE) || defined(CONFIG_PPC_BPA) -void __init generic_calibrate_decr(void) -{ - struct device_node *cpu; - struct div_result divres; - unsigned int *fp; - int node_found; - - /* - * The cpu node should have a timebase-frequency property - * to tell us the rate at which the decrementer counts. - */ - cpu = of_find_node_by_type(NULL, "cpu"); - - ppc_tb_freq = DEFAULT_TB_FREQ; /* hardcoded default */ - node_found = 0; - if (cpu != 0) { - fp = (unsigned int *)get_property(cpu, "timebase-frequency", - NULL); - if (fp != 0) { - node_found = 1; - ppc_tb_freq = *fp; - } - } - if (!node_found) - printk(KERN_ERR "WARNING: Estimating decrementer frequency " - "(not found)\n"); - - ppc_proc_freq = DEFAULT_PROC_FREQ; - node_found = 0; - if (cpu != 0) { - fp = (unsigned int *)get_property(cpu, "clock-frequency", - NULL); - if (fp != 0) { - node_found = 1; - ppc_proc_freq = *fp; - } - } - if (!node_found) - printk(KERN_ERR "WARNING: Estimating processor frequency " - "(not found)\n"); - - of_node_put(cpu); - - printk(KERN_INFO "time_init: decrementer frequency = %lu.%.6lu MHz\n", - ppc_tb_freq/1000000, ppc_tb_freq%1000000); - printk(KERN_INFO "time_init: processor frequency = %lu.%.6lu MHz\n", - ppc_proc_freq/1000000, ppc_proc_freq%1000000); - - tb_ticks_per_jiffy = ppc_tb_freq / HZ; - tb_ticks_per_sec = tb_ticks_per_jiffy * HZ; - tb_ticks_per_usec = ppc_tb_freq / 1000000; - tb_to_us = mulhwu_scale_factor(ppc_tb_freq, 1000000); - div128_by_32(1024*1024, 0, tb_ticks_per_sec, &divres); - tb_to_xs = divres.result_low; - - setup_default_decr(); -} -#endif - -void __init time_init(void) -{ - /* This function is only called on the boot processor */ - unsigned long flags; - struct rtc_time tm; - struct div_result res; - unsigned long scale, shift; - - ppc_md.calibrate_decr(); - - /* - * Compute scale factor for sched_clock. - * The calibrate_decr() function has set tb_ticks_per_sec, - * which is the timebase frequency. - * We compute 1e9 * 2^64 / tb_ticks_per_sec and interpret - * the 128-bit result as a 64.64 fixed-point number. - * We then shift that number right until it is less than 1.0, - * giving us the scale factor and shift count to use in - * sched_clock(). - */ - div128_by_32(1000000000, 0, tb_ticks_per_sec, &res); - scale = res.result_low; - for (shift = 0; res.result_high != 0; ++shift) { - scale = (scale >> 1) | (res.result_high << 63); - res.result_high >>= 1; - } - tb_to_ns_scale = scale; - tb_to_ns_shift = shift; - -#ifdef CONFIG_PPC_ISERIES - if (!piranha_simulator) -#endif - ppc_md.get_boot_time(&tm); - - write_seqlock_irqsave(&xtime_lock, flags); - xtime.tv_sec = mktime(tm.tm_year + 1900, tm.tm_mon + 1, tm.tm_mday, - tm.tm_hour, tm.tm_min, tm.tm_sec); - tb_last_stamp = get_tb(); - do_gtod.varp = &do_gtod.vars[0]; - do_gtod.var_idx = 0; - do_gtod.varp->tb_orig_stamp = tb_last_stamp; - get_paca()->next_jiffy_update_tb = tb_last_stamp + tb_ticks_per_jiffy; - do_gtod.varp->stamp_xsec = xtime.tv_sec * XSEC_PER_SEC; - do_gtod.tb_ticks_per_sec = tb_ticks_per_sec; - do_gtod.varp->tb_to_xs = tb_to_xs; - do_gtod.tb_to_us = tb_to_us; - systemcfg->tb_orig_stamp = tb_last_stamp; - systemcfg->tb_update_count = 0; - systemcfg->tb_ticks_per_sec = tb_ticks_per_sec; - systemcfg->stamp_xsec = xtime.tv_sec * XSEC_PER_SEC; - systemcfg->tb_to_xs = tb_to_xs; - - time_freq = 0; - - xtime.tv_nsec = 0; - last_rtc_update = xtime.tv_sec; - set_normalized_timespec(&wall_to_monotonic, - -xtime.tv_sec, -xtime.tv_nsec); - write_sequnlock_irqrestore(&xtime_lock, flags); - - /* Not exact, but the timer interrupt takes care of this */ - set_dec(tb_ticks_per_jiffy); -} - -/* - * After adjtimex is called, adjust the conversion of tb ticks - * to microseconds to keep do_gettimeofday synchronized - * with ntpd. - * - * Use the time_adjust, time_freq and time_offset computed by adjtimex to - * adjust the frequency. - */ - -/* #define DEBUG_PPC_ADJTIMEX 1 */ - -void ppc_adjtimex(void) -{ - unsigned long den, new_tb_ticks_per_sec, tb_ticks, old_xsec, new_tb_to_xs, new_xsec, new_stamp_xsec; - unsigned long tb_ticks_per_sec_delta; - long delta_freq, ltemp; - struct div_result divres; - unsigned long flags; - struct gettimeofday_vars * temp_varp; - unsigned temp_idx; - long singleshot_ppm = 0; - - /* Compute parts per million frequency adjustment to accomplish the time adjustment - implied by time_offset to be applied over the elapsed time indicated by time_constant. - Use SHIFT_USEC to get it into the same units as time_freq. */ - if ( time_offset < 0 ) { - ltemp = -time_offset; - ltemp <<= SHIFT_USEC - SHIFT_UPDATE; - ltemp >>= SHIFT_KG + time_constant; - ltemp = -ltemp; - } - else { - ltemp = time_offset; - ltemp <<= SHIFT_USEC - SHIFT_UPDATE; - ltemp >>= SHIFT_KG + time_constant; - } - - /* If there is a single shot time adjustment in progress */ - if ( time_adjust ) { -#ifdef DEBUG_PPC_ADJTIMEX - printk("ppc_adjtimex: "); - if ( adjusting_time == 0 ) - printk("starting "); - printk("single shot time_adjust = %ld\n", time_adjust); -#endif - - adjusting_time = 1; - - /* Compute parts per million frequency adjustment to match time_adjust */ - singleshot_ppm = tickadj * HZ; - /* - * The adjustment should be tickadj*HZ to match the code in - * linux/kernel/timer.c, but experiments show that this is too - * large. 3/4 of tickadj*HZ seems about right - */ - singleshot_ppm -= singleshot_ppm / 4; - /* Use SHIFT_USEC to get it into the same units as time_freq */ - singleshot_ppm <<= SHIFT_USEC; - if ( time_adjust < 0 ) - singleshot_ppm = -singleshot_ppm; - } - else { -#ifdef DEBUG_PPC_ADJTIMEX - if ( adjusting_time ) - printk("ppc_adjtimex: ending single shot time_adjust\n"); -#endif - adjusting_time = 0; - } - - /* Add up all of the frequency adjustments */ - delta_freq = time_freq + ltemp + singleshot_ppm; - - /* Compute a new value for tb_ticks_per_sec based on the frequency adjustment */ - den = 1000000 * (1 << (SHIFT_USEC - 8)); - if ( delta_freq < 0 ) { - tb_ticks_per_sec_delta = ( tb_ticks_per_sec * ( (-delta_freq) >> (SHIFT_USEC - 8))) / den; - new_tb_ticks_per_sec = tb_ticks_per_sec + tb_ticks_per_sec_delta; - } - else { - tb_ticks_per_sec_delta = ( tb_ticks_per_sec * ( delta_freq >> (SHIFT_USEC - 8))) / den; - new_tb_ticks_per_sec = tb_ticks_per_sec - tb_ticks_per_sec_delta; - } - -#ifdef DEBUG_PPC_ADJTIMEX - printk("ppc_adjtimex: ltemp = %ld, time_freq = %ld, singleshot_ppm = %ld\n", ltemp, time_freq, singleshot_ppm); - printk("ppc_adjtimex: tb_ticks_per_sec - base = %ld new = %ld\n", tb_ticks_per_sec, new_tb_ticks_per_sec); -#endif - - /* Compute a new value of tb_to_xs (used to convert tb to microseconds and a new value of - stamp_xsec which is the time (in 1/2^20 second units) corresponding to tb_orig_stamp. This - new value of stamp_xsec compensates for the change in frequency (implied by the new tb_to_xs) - which guarantees that the current time remains the same */ - write_seqlock_irqsave( &xtime_lock, flags ); - tb_ticks = get_tb() - do_gtod.varp->tb_orig_stamp; - div128_by_32( 1024*1024, 0, new_tb_ticks_per_sec, &divres ); - new_tb_to_xs = divres.result_low; - new_xsec = mulhdu( tb_ticks, new_tb_to_xs ); - - old_xsec = mulhdu( tb_ticks, do_gtod.varp->tb_to_xs ); - new_stamp_xsec = do_gtod.varp->stamp_xsec + old_xsec - new_xsec; - - /* There are two copies of tb_to_xs and stamp_xsec so that no lock is needed to access and use these - values in do_gettimeofday. We alternate the copies and as long as a reasonable time elapses between - changes, there will never be inconsistent values. ntpd has a minimum of one minute between updates */ - - temp_idx = (do_gtod.var_idx == 0); - temp_varp = &do_gtod.vars[temp_idx]; - - temp_varp->tb_to_xs = new_tb_to_xs; - temp_varp->stamp_xsec = new_stamp_xsec; - temp_varp->tb_orig_stamp = do_gtod.varp->tb_orig_stamp; - smp_mb(); - do_gtod.varp = temp_varp; - do_gtod.var_idx = temp_idx; - - /* - * tb_update_count is used to allow the problem state gettimeofday code - * to assure itself that it sees a consistent view of the tb_to_xs and - * stamp_xsec variables. It reads the tb_update_count, then reads - * tb_to_xs and stamp_xsec and then reads tb_update_count again. If - * the two values of tb_update_count match and are even then the - * tb_to_xs and stamp_xsec values are consistent. If not, then it - * loops back and reads them again until this criteria is met. - */ - ++(systemcfg->tb_update_count); - smp_wmb(); - systemcfg->tb_to_xs = new_tb_to_xs; - systemcfg->stamp_xsec = new_stamp_xsec; - smp_wmb(); - ++(systemcfg->tb_update_count); - - write_sequnlock_irqrestore( &xtime_lock, flags ); - -} - - -#define TICK_SIZE tick -#define FEBRUARY 2 -#define STARTOFTIME 1970 -#define SECDAY 86400L -#define SECYR (SECDAY * 365) -#define leapyear(year) ((year) % 4 == 0) -#define days_in_year(a) (leapyear(a) ? 366 : 365) -#define days_in_month(a) (month_days[(a) - 1]) - -static int month_days[12] = { - 31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31 -}; - -/* - * This only works for the Gregorian calendar - i.e. after 1752 (in the UK) - */ -void GregorianDay(struct rtc_time * tm) -{ - int leapsToDate; - int lastYear; - int day; - int MonthOffset[] = { 0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334 }; - - lastYear=tm->tm_year-1; - - /* - * Number of leap corrections to apply up to end of last year - */ - leapsToDate = lastYear/4 - lastYear/100 + lastYear/400; - - /* - * This year is a leap year if it is divisible by 4 except when it is - * divisible by 100 unless it is divisible by 400 - * - * e.g. 1904 was a leap year, 1900 was not, 1996 is, and 2000 will be - */ - if((tm->tm_year%4==0) && - ((tm->tm_year%100!=0) || (tm->tm_year%400==0)) && - (tm->tm_mon>2)) - { - /* - * We are past Feb. 29 in a leap year - */ - day=1; - } - else - { - day=0; - } - - day += lastYear*365 + leapsToDate + MonthOffset[tm->tm_mon-1] + - tm->tm_mday; - - tm->tm_wday=day%7; -} - -void to_tm(int tim, struct rtc_time * tm) -{ - register int i; - register long hms, day; - - day = tim / SECDAY; - hms = tim % SECDAY; - - /* Hours, minutes, seconds are easy */ - tm->tm_hour = hms / 3600; - tm->tm_min = (hms % 3600) / 60; - tm->tm_sec = (hms % 3600) % 60; - - /* Number of years in days */ - for (i = STARTOFTIME; day >= days_in_year(i); i++) - day -= days_in_year(i); - tm->tm_year = i; - - /* Number of months in days left */ - if (leapyear(tm->tm_year)) - days_in_month(FEBRUARY) = 29; - for (i = 1; day >= days_in_month(i); i++) - day -= days_in_month(i); - days_in_month(FEBRUARY) = 28; - tm->tm_mon = i; - - /* Days are what is left over (+1) from all that. */ - tm->tm_mday = day + 1; - - /* - * Determine the day of week - */ - GregorianDay(tm); -} - -/* Auxiliary function to compute scaling factors */ -/* Actually the choice of a timebase running at 1/4 the of the bus - * frequency giving resolution of a few tens of nanoseconds is quite nice. - * It makes this computation very precise (27-28 bits typically) which - * is optimistic considering the stability of most processor clock - * oscillators and the precision with which the timebase frequency - * is measured but does not harm. - */ -unsigned mulhwu_scale_factor(unsigned inscale, unsigned outscale) { - unsigned mlt=0, tmp, err; - /* No concern for performance, it's done once: use a stupid - * but safe and compact method to find the multiplier. - */ - - for (tmp = 1U<<31; tmp != 0; tmp >>= 1) { - if (mulhwu(inscale, mlt|tmp) < outscale) mlt|=tmp; - } - - /* We might still be off by 1 for the best approximation. - * A side effect of this is that if outscale is too large - * the returned value will be zero. - * Many corner cases have been checked and seem to work, - * some might have been forgotten in the test however. - */ - - err = inscale*(mlt+1); - if (err <= inscale/2) mlt++; - return mlt; - } - -/* - * Divide a 128-bit dividend by a 32-bit divisor, leaving a 128 bit - * result. - */ - -void div128_by_32( unsigned long dividend_high, unsigned long dividend_low, - unsigned divisor, struct div_result *dr ) -{ - unsigned long a,b,c,d, w,x,y,z, ra,rb,rc; - - a = dividend_high >> 32; - b = dividend_high & 0xffffffff; - c = dividend_low >> 32; - d = dividend_low & 0xffffffff; - - w = a/divisor; - ra = (a - (w * divisor)) << 32; - - x = (ra + b)/divisor; - rb = ((ra + b) - (x * divisor)) << 32; - - y = (rb + c)/divisor; - rc = ((rb + c) - (y * divisor)) << 32; - - z = (rc + d)/divisor; - - dr->result_high = (w << 32) + x; - dr->result_low = (y << 32) + z; - -} - diff --git a/arch/ppc64/kernel/traps.c b/arch/ppc64/kernel/traps.c deleted file mode 100644 index 7467ae508e6e..000000000000 --- a/arch/ppc64/kernel/traps.c +++ /dev/null @@ -1,568 +0,0 @@ -/* - * linux/arch/ppc64/kernel/traps.c - * - * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * Modified by Cort Dougan (cort@cs.nmt.edu) - * and Paul Mackerras (paulus@cs.anu.edu.au) - */ - -/* - * This file handles the architecture-dependent parts of hardware exceptions - */ - -#include <linux/config.h> -#include <linux/errno.h> -#include <linux/sched.h> -#include <linux/kernel.h> -#include <linux/mm.h> -#include <linux/stddef.h> -#include <linux/unistd.h> -#include <linux/slab.h> -#include <linux/user.h> -#include <linux/a.out.h> -#include <linux/interrupt.h> -#include <linux/init.h> -#include <linux/module.h> -#include <linux/delay.h> -#include <linux/kprobes.h> -#include <asm/kdebug.h> - -#include <asm/pgtable.h> -#include <asm/uaccess.h> -#include <asm/system.h> -#include <asm/io.h> -#include <asm/processor.h> -#include <asm/ppcdebug.h> -#include <asm/rtas.h> -#include <asm/systemcfg.h> -#include <asm/machdep.h> -#include <asm/pmc.h> - -#ifdef CONFIG_DEBUGGER -int (*__debugger)(struct pt_regs *regs); -int (*__debugger_ipi)(struct pt_regs *regs); -int (*__debugger_bpt)(struct pt_regs *regs); -int (*__debugger_sstep)(struct pt_regs *regs); -int (*__debugger_iabr_match)(struct pt_regs *regs); -int (*__debugger_dabr_match)(struct pt_regs *regs); -int (*__debugger_fault_handler)(struct pt_regs *regs); - -EXPORT_SYMBOL(__debugger); -EXPORT_SYMBOL(__debugger_ipi); -EXPORT_SYMBOL(__debugger_bpt); -EXPORT_SYMBOL(__debugger_sstep); -EXPORT_SYMBOL(__debugger_iabr_match); -EXPORT_SYMBOL(__debugger_dabr_match); -EXPORT_SYMBOL(__debugger_fault_handler); -#endif - -struct notifier_block *ppc64_die_chain; -static DEFINE_SPINLOCK(die_notifier_lock); - -int register_die_notifier(struct notifier_block *nb) -{ - int err = 0; - unsigned long flags; - - spin_lock_irqsave(&die_notifier_lock, flags); - err = notifier_chain_register(&ppc64_die_chain, nb); - spin_unlock_irqrestore(&die_notifier_lock, flags); - return err; -} - -/* - * Trap & Exception support - */ - -static DEFINE_SPINLOCK(die_lock); - -int die(const char *str, struct pt_regs *regs, long err) -{ - static int die_counter; - int nl = 0; - - if (debugger(regs)) - return 1; - - console_verbose(); - spin_lock_irq(&die_lock); - bust_spinlocks(1); - printk("Oops: %s, sig: %ld [#%d]\n", str, err, ++die_counter); -#ifdef CONFIG_PREEMPT - printk("PREEMPT "); - nl = 1; -#endif -#ifdef CONFIG_SMP - printk("SMP NR_CPUS=%d ", NR_CPUS); - nl = 1; -#endif -#ifdef CONFIG_DEBUG_PAGEALLOC - printk("DEBUG_PAGEALLOC "); - nl = 1; -#endif -#ifdef CONFIG_NUMA - printk("NUMA "); - nl = 1; -#endif - switch(systemcfg->platform) { - case PLATFORM_PSERIES: - printk("PSERIES "); - nl = 1; - break; - case PLATFORM_PSERIES_LPAR: - printk("PSERIES LPAR "); - nl = 1; - break; - case PLATFORM_ISERIES_LPAR: - printk("ISERIES LPAR "); - nl = 1; - break; - case PLATFORM_POWERMAC: - printk("POWERMAC "); - nl = 1; - break; - case PLATFORM_BPA: - printk("BPA "); - nl = 1; - break; - } - if (nl) - printk("\n"); - print_modules(); - show_regs(regs); - bust_spinlocks(0); - spin_unlock_irq(&die_lock); - - if (in_interrupt()) - panic("Fatal exception in interrupt"); - - if (panic_on_oops) { - printk(KERN_EMERG "Fatal exception: panic in 5 seconds\n"); - ssleep(5); - panic("Fatal exception"); - } - do_exit(SIGSEGV); - - return 0; -} - -void _exception(int signr, struct pt_regs *regs, int code, unsigned long addr) -{ - siginfo_t info; - - if (!user_mode(regs)) { - if (die("Exception in kernel mode", regs, signr)) - return; - } - - memset(&info, 0, sizeof(info)); - info.si_signo = signr; - info.si_code = code; - info.si_addr = (void __user *) addr; - force_sig_info(signr, &info, current); -} - -void system_reset_exception(struct pt_regs *regs) -{ - /* See if any machine dependent calls */ - if (ppc_md.system_reset_exception) - ppc_md.system_reset_exception(regs); - - die("System Reset", regs, 0); - - /* Must die if the interrupt is not recoverable */ - if (!(regs->msr & MSR_RI)) - panic("Unrecoverable System Reset"); - - /* What should we do here? We could issue a shutdown or hard reset. */ -} - -void machine_check_exception(struct pt_regs *regs) -{ - int recover = 0; - - /* See if any machine dependent calls */ - if (ppc_md.machine_check_exception) - recover = ppc_md.machine_check_exception(regs); - - if (recover) - return; - - if (debugger_fault_handler(regs)) - return; - die("Machine check", regs, 0); - - /* Must die if the interrupt is not recoverable */ - if (!(regs->msr & MSR_RI)) - panic("Unrecoverable Machine check"); -} - -void unknown_exception(struct pt_regs *regs) -{ - printk("Bad trap at PC: %lx, SR: %lx, vector=%lx\n", - regs->nip, regs->msr, regs->trap); - - _exception(SIGTRAP, regs, 0, 0); -} - -void instruction_breakpoint_exception(struct pt_regs *regs) -{ - if (notify_die(DIE_IABR_MATCH, "iabr_match", regs, 5, - 5, SIGTRAP) == NOTIFY_STOP) - return; - if (debugger_iabr_match(regs)) - return; - _exception(SIGTRAP, regs, TRAP_BRKPT, regs->nip); -} - -void __kprobes single_step_exception(struct pt_regs *regs) -{ - regs->msr &= ~MSR_SE; /* Turn off 'trace' bit */ - - if (notify_die(DIE_SSTEP, "single_step", regs, 5, - 5, SIGTRAP) == NOTIFY_STOP) - return; - if (debugger_sstep(regs)) - return; - - _exception(SIGTRAP, regs, TRAP_TRACE, regs->nip); -} - -/* - * After we have successfully emulated an instruction, we have to - * check if the instruction was being single-stepped, and if so, - * pretend we got a single-step exception. This was pointed out - * by Kumar Gala. -- paulus - */ -static inline void emulate_single_step(struct pt_regs *regs) -{ - if (regs->msr & MSR_SE) - single_step_exception(regs); -} - -static void parse_fpe(struct pt_regs *regs) -{ - int code = 0; - unsigned long fpscr; - - flush_fp_to_thread(current); - - fpscr = current->thread.fpscr; - - /* Invalid operation */ - if ((fpscr & FPSCR_VE) && (fpscr & FPSCR_VX)) - code = FPE_FLTINV; - - /* Overflow */ - else if ((fpscr & FPSCR_OE) && (fpscr & FPSCR_OX)) - code = FPE_FLTOVF; - - /* Underflow */ - else if ((fpscr & FPSCR_UE) && (fpscr & FPSCR_UX)) - code = FPE_FLTUND; - - /* Divide by zero */ - else if ((fpscr & FPSCR_ZE) && (fpscr & FPSCR_ZX)) - code = FPE_FLTDIV; - - /* Inexact result */ - else if ((fpscr & FPSCR_XE) && (fpscr & FPSCR_XX)) - code = FPE_FLTRES; - - _exception(SIGFPE, regs, code, regs->nip); -} - -/* - * Illegal instruction emulation support. Return non-zero if we can't - * emulate, or -EFAULT if the associated memory access caused an access - * fault. Return zero on success. - */ - -#define INST_MFSPR_PVR 0x7c1f42a6 -#define INST_MFSPR_PVR_MASK 0xfc1fffff - -#define INST_DCBA 0x7c0005ec -#define INST_DCBA_MASK 0x7c0007fe - -#define INST_MCRXR 0x7c000400 -#define INST_MCRXR_MASK 0x7c0007fe - -static int emulate_instruction(struct pt_regs *regs) -{ - unsigned int instword; - - if (!user_mode(regs)) - return -EINVAL; - - CHECK_FULL_REGS(regs); - - if (get_user(instword, (unsigned int __user *)(regs->nip))) - return -EFAULT; - - /* Emulate the mfspr rD, PVR. */ - if ((instword & INST_MFSPR_PVR_MASK) == INST_MFSPR_PVR) { - unsigned int rd; - - rd = (instword >> 21) & 0x1f; - regs->gpr[rd] = mfspr(SPRN_PVR); - return 0; - } - - /* Emulating the dcba insn is just a no-op. */ - if ((instword & INST_DCBA_MASK) == INST_DCBA) { - static int warned; - - if (!warned) { - printk(KERN_WARNING - "process %d (%s) uses obsolete 'dcba' insn\n", - current->pid, current->comm); - warned = 1; - } - return 0; - } - - /* Emulate the mcrxr insn. */ - if ((instword & INST_MCRXR_MASK) == INST_MCRXR) { - static int warned; - unsigned int shift; - - if (!warned) { - printk(KERN_WARNING - "process %d (%s) uses obsolete 'mcrxr' insn\n", - current->pid, current->comm); - warned = 1; - } - - shift = (instword >> 21) & 0x1c; - regs->ccr &= ~(0xf0000000 >> shift); - regs->ccr |= (regs->xer & 0xf0000000) >> shift; - regs->xer &= ~0xf0000000; - return 0; - } - - return -EINVAL; -} - -/* - * Look through the list of trap instructions that are used for BUG(), - * BUG_ON() and WARN_ON() and see if we hit one. At this point we know - * that the exception was caused by a trap instruction of some kind. - * Returns 1 if we should continue (i.e. it was a WARN_ON) or 0 - * otherwise. - */ -extern struct bug_entry __start___bug_table[], __stop___bug_table[]; - -#ifndef CONFIG_MODULES -#define module_find_bug(x) NULL -#endif - -struct bug_entry *find_bug(unsigned long bugaddr) -{ - struct bug_entry *bug; - - for (bug = __start___bug_table; bug < __stop___bug_table; ++bug) - if (bugaddr == bug->bug_addr) - return bug; - return module_find_bug(bugaddr); -} - -static int -check_bug_trap(struct pt_regs *regs) -{ - struct bug_entry *bug; - unsigned long addr; - - if (regs->msr & MSR_PR) - return 0; /* not in kernel */ - addr = regs->nip; /* address of trap instruction */ - if (addr < PAGE_OFFSET) - return 0; - bug = find_bug(regs->nip); - if (bug == NULL) - return 0; - if (bug->line & BUG_WARNING_TRAP) { - /* this is a WARN_ON rather than BUG/BUG_ON */ - printk(KERN_ERR "Badness in %s at %s:%d\n", - bug->function, bug->file, - (unsigned int)bug->line & ~BUG_WARNING_TRAP); - show_stack(current, (void *)regs->gpr[1]); - return 1; - } - printk(KERN_CRIT "kernel BUG in %s at %s:%d!\n", - bug->function, bug->file, (unsigned int)bug->line); - return 0; -} - -void __kprobes program_check_exception(struct pt_regs *regs) -{ - if (debugger_fault_handler(regs)) - return; - - if (regs->msr & 0x100000) { - /* IEEE FP exception */ - parse_fpe(regs); - } else if (regs->msr & 0x20000) { - /* trap exception */ - - if (notify_die(DIE_BPT, "breakpoint", regs, 5, - 5, SIGTRAP) == NOTIFY_STOP) - return; - if (debugger_bpt(regs)) - return; - - if (check_bug_trap(regs)) { - regs->nip += 4; - return; - } - _exception(SIGTRAP, regs, TRAP_BRKPT, regs->nip); - - } else { - /* Privileged or illegal instruction; try to emulate it. */ - switch (emulate_instruction(regs)) { - case 0: - regs->nip += 4; - emulate_single_step(regs); - break; - - case -EFAULT: - _exception(SIGSEGV, regs, SEGV_MAPERR, regs->nip); - break; - - default: - if (regs->msr & 0x40000) - /* priveleged */ - _exception(SIGILL, regs, ILL_PRVOPC, regs->nip); - else - /* illegal */ - _exception(SIGILL, regs, ILL_ILLOPC, regs->nip); - break; - } - } -} - -void kernel_fp_unavailable_exception(struct pt_regs *regs) -{ - printk(KERN_EMERG "Unrecoverable FP Unavailable Exception " - "%lx at %lx\n", regs->trap, regs->nip); - die("Unrecoverable FP Unavailable Exception", regs, SIGABRT); -} - -void altivec_unavailable_exception(struct pt_regs *regs) -{ - if (user_mode(regs)) { - /* A user program has executed an altivec instruction, - but this kernel doesn't support altivec. */ - _exception(SIGILL, regs, ILL_ILLOPC, regs->nip); - return; - } - printk(KERN_EMERG "Unrecoverable VMX/Altivec Unavailable Exception " - "%lx at %lx\n", regs->trap, regs->nip); - die("Unrecoverable VMX/Altivec Unavailable Exception", regs, SIGABRT); -} - -extern perf_irq_t perf_irq; - -void performance_monitor_exception(struct pt_regs *regs) -{ - perf_irq(regs); -} - -void alignment_exception(struct pt_regs *regs) -{ - int fixed; - - fixed = fix_alignment(regs); - - if (fixed == 1) { - regs->nip += 4; /* skip over emulated instruction */ - emulate_single_step(regs); - return; - } - - /* Operand address was bad */ - if (fixed == -EFAULT) { - if (user_mode(regs)) { - _exception(SIGSEGV, regs, SEGV_MAPERR, regs->dar); - } else { - /* Search exception table */ - bad_page_fault(regs, regs->dar, SIGSEGV); - } - - return; - } - - _exception(SIGBUS, regs, BUS_ADRALN, regs->nip); -} - -#ifdef CONFIG_ALTIVEC -void altivec_assist_exception(struct pt_regs *regs) -{ - int err; - siginfo_t info; - - if (!user_mode(regs)) { - printk(KERN_EMERG "VMX/Altivec assist exception in kernel mode" - " at %lx\n", regs->nip); - die("Kernel VMX/Altivec assist exception", regs, SIGILL); - } - - flush_altivec_to_thread(current); - - err = emulate_altivec(regs); - if (err == 0) { - regs->nip += 4; /* skip emulated instruction */ - emulate_single_step(regs); - return; - } - - if (err == -EFAULT) { - /* got an error reading the instruction */ - info.si_signo = SIGSEGV; - info.si_errno = 0; - info.si_code = SEGV_MAPERR; - info.si_addr = (void __user *) regs->nip; - force_sig_info(SIGSEGV, &info, current); - } else { - /* didn't recognize the instruction */ - /* XXX quick hack for now: set the non-Java bit in the VSCR */ - if (printk_ratelimit()) - printk(KERN_ERR "Unrecognized altivec instruction " - "in %s at %lx\n", current->comm, regs->nip); - current->thread.vscr.u[3] |= 0x10000; - } -} -#endif /* CONFIG_ALTIVEC */ - -/* - * We enter here if we get an unrecoverable exception, that is, one - * that happened at a point where the RI (recoverable interrupt) bit - * in the MSR is 0. This indicates that SRR0/1 are live, and that - * we therefore lost state by taking this exception. - */ -void unrecoverable_exception(struct pt_regs *regs) -{ - printk(KERN_EMERG "Unrecoverable exception %lx at %lx\n", - regs->trap, regs->nip); - die("Unrecoverable exception", regs, SIGABRT); -} - -/* - * We enter here if we discover during exception entry that we are - * running in supervisor mode with a userspace value in the stack pointer. - */ -void kernel_bad_stack(struct pt_regs *regs) -{ - printk(KERN_EMERG "Bad kernel stack pointer %lx at %lx\n", - regs->gpr[1], regs->nip); - die("Bad kernel stack pointer", regs, SIGABRT); -} - -void __init trap_init(void) -{ -} diff --git a/arch/ppc64/kernel/u3_iommu.c b/arch/ppc64/kernel/u3_iommu.c index 41ea09cb9ac7..fba871a1bda5 100644 --- a/arch/ppc64/kernel/u3_iommu.c +++ b/arch/ppc64/kernel/u3_iommu.c @@ -44,39 +44,11 @@ #include <asm/abs_addr.h> #include <asm/cacheflush.h> #include <asm/lmb.h> - -#include "pci.h" +#include <asm/dart.h> +#include <asm/ppc-pci.h> extern int iommu_force_on; -/* physical base of DART registers */ -#define DART_BASE 0xf8033000UL - -/* Offset from base to control register */ -#define DARTCNTL 0 -/* Offset from base to exception register */ -#define DARTEXCP 0x10 -/* Offset from base to TLB tag registers */ -#define DARTTAG 0x1000 - - -/* Control Register fields */ - -/* base address of table (pfn) */ -#define DARTCNTL_BASE_MASK 0xfffff -#define DARTCNTL_BASE_SHIFT 12 - -#define DARTCNTL_FLUSHTLB 0x400 -#define DARTCNTL_ENABLE 0x200 - -/* size of table in pages */ -#define DARTCNTL_SIZE_MASK 0x1ff -#define DARTCNTL_SIZE_SHIFT 0 - -/* DART table fields */ -#define DARTMAP_VALID 0x80000000 -#define DARTMAP_RPNMASK 0x00ffffff - /* Physical base address and size of the DART table */ unsigned long dart_tablebase; /* exported to htab_initialize */ static unsigned long dart_tablesize; @@ -152,18 +124,21 @@ static void dart_build(struct iommu_table *tbl, long index, DBG("dart: build at: %lx, %lx, addr: %x\n", index, npages, uaddr); + index <<= DART_PAGE_FACTOR; + npages <<= DART_PAGE_FACTOR; + dp = ((unsigned int*)tbl->it_base) + index; /* On U3, all memory is contigous, so we can move this * out of the loop. */ while (npages--) { - rpn = virt_to_abs(uaddr) >> PAGE_SHIFT; + rpn = virt_to_abs(uaddr) >> DART_PAGE_SHIFT; *(dp++) = DARTMAP_VALID | (rpn & DARTMAP_RPNMASK); rpn++; - uaddr += PAGE_SIZE; + uaddr += DART_PAGE_SIZE; } dart_dirty = 1; @@ -181,6 +156,9 @@ static void dart_free(struct iommu_table *tbl, long index, long npages) DBG("dart: free at: %lx, %lx\n", index, npages); + index <<= DART_PAGE_FACTOR; + npages <<= DART_PAGE_FACTOR; + dp = ((unsigned int *)tbl->it_base) + index; while (npages--) @@ -209,10 +187,10 @@ static int dart_init(struct device_node *dart_node) * that to work around what looks like a problem with the HT bridge * prefetching into invalid pages and corrupting data */ - tmp = lmb_alloc(PAGE_SIZE, PAGE_SIZE); + tmp = lmb_alloc(DART_PAGE_SIZE, DART_PAGE_SIZE); if (!tmp) panic("U3-DART: Cannot allocate spare page!"); - dart_emptyval = DARTMAP_VALID | ((tmp >> PAGE_SHIFT) & DARTMAP_RPNMASK); + dart_emptyval = DARTMAP_VALID | ((tmp >> DART_PAGE_SHIFT) & DARTMAP_RPNMASK); /* Map in DART registers. FIXME: Use device node to get base address */ dart = ioremap(DART_BASE, 0x7000); @@ -223,8 +201,8 @@ static int dart_init(struct device_node *dart_node) * table size and enable bit */ regword = DARTCNTL_ENABLE | - ((dart_tablebase >> PAGE_SHIFT) << DARTCNTL_BASE_SHIFT) | - (((dart_tablesize >> PAGE_SHIFT) & DARTCNTL_SIZE_MASK) + ((dart_tablebase >> DART_PAGE_SHIFT) << DARTCNTL_BASE_SHIFT) | + (((dart_tablesize >> DART_PAGE_SHIFT) & DARTCNTL_SIZE_MASK) << DARTCNTL_SIZE_SHIFT); dart_vbase = ioremap(virt_to_abs(dart_tablebase), dart_tablesize); diff --git a/arch/ppc64/kernel/vdso64/sigtramp.S b/arch/ppc64/kernel/vdso64/sigtramp.S index 8ae8f205e470..31b604ab56de 100644 --- a/arch/ppc64/kernel/vdso64/sigtramp.S +++ b/arch/ppc64/kernel/vdso64/sigtramp.S @@ -15,6 +15,7 @@ #include <asm/ppc_asm.h> #include <asm/unistd.h> #include <asm/vdso.h> +#include <asm/ptrace.h> /* XXX for __SIGNAL_FRAMESIZE */ .text diff --git a/arch/ppc64/kernel/vecemu.c b/arch/ppc64/kernel/vecemu.c deleted file mode 100644 index cb207629f21f..000000000000 --- a/arch/ppc64/kernel/vecemu.c +++ /dev/null @@ -1,346 +0,0 @@ -/* - * Routines to emulate some Altivec/VMX instructions, specifically - * those that can trap when given denormalized operands in Java mode. - */ -#include <linux/kernel.h> -#include <linux/errno.h> -#include <linux/sched.h> -#include <asm/ptrace.h> -#include <asm/processor.h> -#include <asm/uaccess.h> - -/* Functions in vector.S */ -extern void vaddfp(vector128 *dst, vector128 *a, vector128 *b); -extern void vsubfp(vector128 *dst, vector128 *a, vector128 *b); -extern void vmaddfp(vector128 *dst, vector128 *a, vector128 *b, vector128 *c); -extern void vnmsubfp(vector128 *dst, vector128 *a, vector128 *b, vector128 *c); -extern void vrefp(vector128 *dst, vector128 *src); -extern void vrsqrtefp(vector128 *dst, vector128 *src); -extern void vexptep(vector128 *dst, vector128 *src); - -static unsigned int exp2s[8] = { - 0x800000, - 0x8b95c2, - 0x9837f0, - 0xa5fed7, - 0xb504f3, - 0xc5672a, - 0xd744fd, - 0xeac0c7 -}; - -/* - * Computes an estimate of 2^x. The `s' argument is the 32-bit - * single-precision floating-point representation of x. - */ -static unsigned int eexp2(unsigned int s) -{ - int exp, pwr; - unsigned int mant, frac; - - /* extract exponent field from input */ - exp = ((s >> 23) & 0xff) - 127; - if (exp > 7) { - /* check for NaN input */ - if (exp == 128 && (s & 0x7fffff) != 0) - return s | 0x400000; /* return QNaN */ - /* 2^-big = 0, 2^+big = +Inf */ - return (s & 0x80000000)? 0: 0x7f800000; /* 0 or +Inf */ - } - if (exp < -23) - return 0x3f800000; /* 1.0 */ - - /* convert to fixed point integer in 9.23 representation */ - pwr = (s & 0x7fffff) | 0x800000; - if (exp > 0) - pwr <<= exp; - else - pwr >>= -exp; - if (s & 0x80000000) - pwr = -pwr; - - /* extract integer part, which becomes exponent part of result */ - exp = (pwr >> 23) + 126; - if (exp >= 254) - return 0x7f800000; - if (exp < -23) - return 0; - - /* table lookup on top 3 bits of fraction to get mantissa */ - mant = exp2s[(pwr >> 20) & 7]; - - /* linear interpolation using remaining 20 bits of fraction */ - asm("mulhwu %0,%1,%2" : "=r" (frac) - : "r" (pwr << 12), "r" (0x172b83ff)); - asm("mulhwu %0,%1,%2" : "=r" (frac) : "r" (frac), "r" (mant)); - mant += frac; - - if (exp >= 0) - return mant + (exp << 23); - - /* denormalized result */ - exp = -exp; - mant += 1 << (exp - 1); - return mant >> exp; -} - -/* - * Computes an estimate of log_2(x). The `s' argument is the 32-bit - * single-precision floating-point representation of x. - */ -static unsigned int elog2(unsigned int s) -{ - int exp, mant, lz, frac; - - exp = s & 0x7f800000; - mant = s & 0x7fffff; - if (exp == 0x7f800000) { /* Inf or NaN */ - if (mant != 0) - s |= 0x400000; /* turn NaN into QNaN */ - return s; - } - if ((exp | mant) == 0) /* +0 or -0 */ - return 0xff800000; /* return -Inf */ - - if (exp == 0) { - /* denormalized */ - asm("cntlzw %0,%1" : "=r" (lz) : "r" (mant)); - mant <<= lz - 8; - exp = (-118 - lz) << 23; - } else { - mant |= 0x800000; - exp -= 127 << 23; - } - - if (mant >= 0xb504f3) { /* 2^0.5 * 2^23 */ - exp |= 0x400000; /* 0.5 * 2^23 */ - asm("mulhwu %0,%1,%2" : "=r" (mant) - : "r" (mant), "r" (0xb504f334)); /* 2^-0.5 * 2^32 */ - } - if (mant >= 0x9837f0) { /* 2^0.25 * 2^23 */ - exp |= 0x200000; /* 0.25 * 2^23 */ - asm("mulhwu %0,%1,%2" : "=r" (mant) - : "r" (mant), "r" (0xd744fccb)); /* 2^-0.25 * 2^32 */ - } - if (mant >= 0x8b95c2) { /* 2^0.125 * 2^23 */ - exp |= 0x100000; /* 0.125 * 2^23 */ - asm("mulhwu %0,%1,%2" : "=r" (mant) - : "r" (mant), "r" (0xeac0c6e8)); /* 2^-0.125 * 2^32 */ - } - if (mant > 0x800000) { /* 1.0 * 2^23 */ - /* calculate (mant - 1) * 1.381097463 */ - /* 1.381097463 == 0.125 / (2^0.125 - 1) */ - asm("mulhwu %0,%1,%2" : "=r" (frac) - : "r" ((mant - 0x800000) << 1), "r" (0xb0c7cd3a)); - exp += frac; - } - s = exp & 0x80000000; - if (exp != 0) { - if (s) - exp = -exp; - asm("cntlzw %0,%1" : "=r" (lz) : "r" (exp)); - lz = 8 - lz; - if (lz > 0) - exp >>= lz; - else if (lz < 0) - exp <<= -lz; - s += ((lz + 126) << 23) + exp; - } - return s; -} - -#define VSCR_SAT 1 - -static int ctsxs(unsigned int x, int scale, unsigned int *vscrp) -{ - int exp, mant; - - exp = (x >> 23) & 0xff; - mant = x & 0x7fffff; - if (exp == 255 && mant != 0) - return 0; /* NaN -> 0 */ - exp = exp - 127 + scale; - if (exp < 0) - return 0; /* round towards zero */ - if (exp >= 31) { - /* saturate, unless the result would be -2^31 */ - if (x + (scale << 23) != 0xcf000000) - *vscrp |= VSCR_SAT; - return (x & 0x80000000)? 0x80000000: 0x7fffffff; - } - mant |= 0x800000; - mant = (mant << 7) >> (30 - exp); - return (x & 0x80000000)? -mant: mant; -} - -static unsigned int ctuxs(unsigned int x, int scale, unsigned int *vscrp) -{ - int exp; - unsigned int mant; - - exp = (x >> 23) & 0xff; - mant = x & 0x7fffff; - if (exp == 255 && mant != 0) - return 0; /* NaN -> 0 */ - exp = exp - 127 + scale; - if (exp < 0) - return 0; /* round towards zero */ - if (x & 0x80000000) { - /* negative => saturate to 0 */ - *vscrp |= VSCR_SAT; - return 0; - } - if (exp >= 32) { - /* saturate */ - *vscrp |= VSCR_SAT; - return 0xffffffff; - } - mant |= 0x800000; - mant = (mant << 8) >> (31 - exp); - return mant; -} - -/* Round to floating integer, towards 0 */ -static unsigned int rfiz(unsigned int x) -{ - int exp; - - exp = ((x >> 23) & 0xff) - 127; - if (exp == 128 && (x & 0x7fffff) != 0) - return x | 0x400000; /* NaN -> make it a QNaN */ - if (exp >= 23) - return x; /* it's an integer already (or Inf) */ - if (exp < 0) - return x & 0x80000000; /* |x| < 1.0 rounds to 0 */ - return x & ~(0x7fffff >> exp); -} - -/* Round to floating integer, towards +/- Inf */ -static unsigned int rfii(unsigned int x) -{ - int exp, mask; - - exp = ((x >> 23) & 0xff) - 127; - if (exp == 128 && (x & 0x7fffff) != 0) - return x | 0x400000; /* NaN -> make it a QNaN */ - if (exp >= 23) - return x; /* it's an integer already (or Inf) */ - if ((x & 0x7fffffff) == 0) - return x; /* +/-0 -> +/-0 */ - if (exp < 0) - /* 0 < |x| < 1.0 rounds to +/- 1.0 */ - return (x & 0x80000000) | 0x3f800000; - mask = 0x7fffff >> exp; - /* mantissa overflows into exponent - that's OK, - it can't overflow into the sign bit */ - return (x + mask) & ~mask; -} - -/* Round to floating integer, to nearest */ -static unsigned int rfin(unsigned int x) -{ - int exp, half; - - exp = ((x >> 23) & 0xff) - 127; - if (exp == 128 && (x & 0x7fffff) != 0) - return x | 0x400000; /* NaN -> make it a QNaN */ - if (exp >= 23) - return x; /* it's an integer already (or Inf) */ - if (exp < -1) - return x & 0x80000000; /* |x| < 0.5 -> +/-0 */ - if (exp == -1) - /* 0.5 <= |x| < 1.0 rounds to +/- 1.0 */ - return (x & 0x80000000) | 0x3f800000; - half = 0x400000 >> exp; - /* add 0.5 to the magnitude and chop off the fraction bits */ - return (x + half) & ~(0x7fffff >> exp); -} - -int -emulate_altivec(struct pt_regs *regs) -{ - unsigned int instr, i; - unsigned int va, vb, vc, vd; - vector128 *vrs; - - if (get_user(instr, (unsigned int __user *) regs->nip)) - return -EFAULT; - if ((instr >> 26) != 4) - return -EINVAL; /* not an altivec instruction */ - vd = (instr >> 21) & 0x1f; - va = (instr >> 16) & 0x1f; - vb = (instr >> 11) & 0x1f; - vc = (instr >> 6) & 0x1f; - - vrs = current->thread.vr; - switch (instr & 0x3f) { - case 10: - switch (vc) { - case 0: /* vaddfp */ - vaddfp(&vrs[vd], &vrs[va], &vrs[vb]); - break; - case 1: /* vsubfp */ - vsubfp(&vrs[vd], &vrs[va], &vrs[vb]); - break; - case 4: /* vrefp */ - vrefp(&vrs[vd], &vrs[vb]); - break; - case 5: /* vrsqrtefp */ - vrsqrtefp(&vrs[vd], &vrs[vb]); - break; - case 6: /* vexptefp */ - for (i = 0; i < 4; ++i) - vrs[vd].u[i] = eexp2(vrs[vb].u[i]); - break; - case 7: /* vlogefp */ - for (i = 0; i < 4; ++i) - vrs[vd].u[i] = elog2(vrs[vb].u[i]); - break; - case 8: /* vrfin */ - for (i = 0; i < 4; ++i) - vrs[vd].u[i] = rfin(vrs[vb].u[i]); - break; - case 9: /* vrfiz */ - for (i = 0; i < 4; ++i) - vrs[vd].u[i] = rfiz(vrs[vb].u[i]); - break; - case 10: /* vrfip */ - for (i = 0; i < 4; ++i) { - u32 x = vrs[vb].u[i]; - x = (x & 0x80000000)? rfiz(x): rfii(x); - vrs[vd].u[i] = x; - } - break; - case 11: /* vrfim */ - for (i = 0; i < 4; ++i) { - u32 x = vrs[vb].u[i]; - x = (x & 0x80000000)? rfii(x): rfiz(x); - vrs[vd].u[i] = x; - } - break; - case 14: /* vctuxs */ - for (i = 0; i < 4; ++i) - vrs[vd].u[i] = ctuxs(vrs[vb].u[i], va, - ¤t->thread.vscr.u[3]); - break; - case 15: /* vctsxs */ - for (i = 0; i < 4; ++i) - vrs[vd].u[i] = ctsxs(vrs[vb].u[i], va, - ¤t->thread.vscr.u[3]); - break; - default: - return -EINVAL; - } - break; - case 46: /* vmaddfp */ - vmaddfp(&vrs[vd], &vrs[va], &vrs[vb], &vrs[vc]); - break; - case 47: /* vnmsubfp */ - vnmsubfp(&vrs[vd], &vrs[va], &vrs[vb], &vrs[vc]); - break; - default: - return -EINVAL; - } - - return 0; -} diff --git a/arch/ppc64/kernel/vector.S b/arch/ppc64/kernel/vector.S deleted file mode 100644 index b79d33e4001e..000000000000 --- a/arch/ppc64/kernel/vector.S +++ /dev/null @@ -1,172 +0,0 @@ -#include <asm/ppc_asm.h> -#include <asm/processor.h> - -/* - * The routines below are in assembler so we can closely control the - * usage of floating-point registers. These routines must be called - * with preempt disabled. - */ - .section ".toc","aw" -fpzero: - .tc FD_0_0[TC],0 -fpone: - .tc FD_3ff00000_0[TC],0x3ff0000000000000 /* 1.0 */ -fphalf: - .tc FD_3fe00000_0[TC],0x3fe0000000000000 /* 0.5 */ - - .text -/* - * Internal routine to enable floating point and set FPSCR to 0. - * Don't call it from C; it doesn't use the normal calling convention. - */ -fpenable: - mfmsr r10 - ori r11,r10,MSR_FP - mtmsr r11 - isync - stfd fr31,-8(r1) - stfd fr0,-16(r1) - stfd fr1,-24(r1) - mffs fr31 - lfd fr1,fpzero@toc(r2) - mtfsf 0xff,fr1 - blr - -fpdisable: - mtlr r12 - mtfsf 0xff,fr31 - lfd fr1,-24(r1) - lfd fr0,-16(r1) - lfd fr31,-8(r1) - mtmsr r10 - isync - blr - -/* - * Vector add, floating point. - */ -_GLOBAL(vaddfp) - mflr r12 - bl fpenable - li r0,4 - mtctr r0 - li r6,0 -1: lfsx fr0,r4,r6 - lfsx fr1,r5,r6 - fadds fr0,fr0,fr1 - stfsx fr0,r3,r6 - addi r6,r6,4 - bdnz 1b - b fpdisable - -/* - * Vector subtract, floating point. - */ -_GLOBAL(vsubfp) - mflr r12 - bl fpenable - li r0,4 - mtctr r0 - li r6,0 -1: lfsx fr0,r4,r6 - lfsx fr1,r5,r6 - fsubs fr0,fr0,fr1 - stfsx fr0,r3,r6 - addi r6,r6,4 - bdnz 1b - b fpdisable - -/* - * Vector multiply and add, floating point. - */ -_GLOBAL(vmaddfp) - mflr r12 - bl fpenable - stfd fr2,-32(r1) - li r0,4 - mtctr r0 - li r7,0 -1: lfsx fr0,r4,r7 - lfsx fr1,r5,r7 - lfsx fr2,r6,r7 - fmadds fr0,fr0,fr2,fr1 - stfsx fr0,r3,r7 - addi r7,r7,4 - bdnz 1b - lfd fr2,-32(r1) - b fpdisable - -/* - * Vector negative multiply and subtract, floating point. - */ -_GLOBAL(vnmsubfp) - mflr r12 - bl fpenable - stfd fr2,-32(r1) - li r0,4 - mtctr r0 - li r7,0 -1: lfsx fr0,r4,r7 - lfsx fr1,r5,r7 - lfsx fr2,r6,r7 - fnmsubs fr0,fr0,fr2,fr1 - stfsx fr0,r3,r7 - addi r7,r7,4 - bdnz 1b - lfd fr2,-32(r1) - b fpdisable - -/* - * Vector reciprocal estimate. We just compute 1.0/x. - * r3 -> destination, r4 -> source. - */ -_GLOBAL(vrefp) - mflr r12 - bl fpenable - li r0,4 - lfd fr1,fpone@toc(r2) - mtctr r0 - li r6,0 -1: lfsx fr0,r4,r6 - fdivs fr0,fr1,fr0 - stfsx fr0,r3,r6 - addi r6,r6,4 - bdnz 1b - b fpdisable - -/* - * Vector reciprocal square-root estimate, floating point. - * We use the frsqrte instruction for the initial estimate followed - * by 2 iterations of Newton-Raphson to get sufficient accuracy. - * r3 -> destination, r4 -> source. - */ -_GLOBAL(vrsqrtefp) - mflr r12 - bl fpenable - stfd fr2,-32(r1) - stfd fr3,-40(r1) - stfd fr4,-48(r1) - stfd fr5,-56(r1) - li r0,4 - lfd fr4,fpone@toc(r2) - lfd fr5,fphalf@toc(r2) - mtctr r0 - li r6,0 -1: lfsx fr0,r4,r6 - frsqrte fr1,fr0 /* r = frsqrte(s) */ - fmuls fr3,fr1,fr0 /* r * s */ - fmuls fr2,fr1,fr5 /* r * 0.5 */ - fnmsubs fr3,fr1,fr3,fr4 /* 1 - s * r * r */ - fmadds fr1,fr2,fr3,fr1 /* r = r + 0.5 * r * (1 - s * r * r) */ - fmuls fr3,fr1,fr0 /* r * s */ - fmuls fr2,fr1,fr5 /* r * 0.5 */ - fnmsubs fr3,fr1,fr3,fr4 /* 1 - s * r * r */ - fmadds fr1,fr2,fr3,fr1 /* r = r + 0.5 * r * (1 - s * r * r) */ - stfsx fr1,r3,r6 - addi r6,r6,4 - bdnz 1b - lfd fr5,-56(r1) - lfd fr4,-48(r1) - lfd fr3,-40(r1) - lfd fr2,-32(r1) - b fpdisable diff --git a/arch/ppc64/kernel/viopath.c b/arch/ppc64/kernel/viopath.c deleted file mode 100644 index 2a6c4f01c45e..000000000000 --- a/arch/ppc64/kernel/viopath.c +++ /dev/null @@ -1,673 +0,0 @@ -/* -*- linux-c -*- - * arch/ppc64/kernel/viopath.c - * - * iSeries Virtual I/O Message Path code - * - * Authors: Dave Boutcher <boutcher@us.ibm.com> - * Ryan Arnold <ryanarn@us.ibm.com> - * Colin Devilbiss <devilbis@us.ibm.com> - * - * (C) Copyright 2000-2003 IBM Corporation - * - * This code is used by the iSeries virtual disk, cd, - * tape, and console to communicate with OS/400 in another - * partition. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of the - * License, or (at your option) anyu later version. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software Foundation, - * Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - * - */ -#include <linux/module.h> -#include <linux/kernel.h> -#include <linux/errno.h> -#include <linux/vmalloc.h> -#include <linux/string.h> -#include <linux/proc_fs.h> -#include <linux/dma-mapping.h> -#include <linux/wait.h> -#include <linux/seq_file.h> -#include <linux/smp_lock.h> -#include <linux/interrupt.h> - -#include <asm/system.h> -#include <asm/uaccess.h> -#include <asm/iSeries/HvTypes.h> -#include <asm/iSeries/ItExtVpdPanel.h> -#include <asm/iSeries/HvLpEvent.h> -#include <asm/iSeries/HvLpConfig.h> -#include <asm/iSeries/mf.h> -#include <asm/iSeries/vio.h> - -/* Status of the path to each other partition in the system. - * This is overkill, since we will only ever establish connections - * to our hosting partition and the primary partition on the system. - * But this allows for other support in the future. - */ -static struct viopathStatus { - int isOpen; /* Did we open the path? */ - int isActive; /* Do we have a mon msg outstanding */ - int users[VIO_MAX_SUBTYPES]; - HvLpInstanceId mSourceInst; - HvLpInstanceId mTargetInst; - int numberAllocated; -} viopathStatus[HVMAXARCHITECTEDLPS]; - -static DEFINE_SPINLOCK(statuslock); - -/* - * For each kind of event we allocate a buffer that is - * guaranteed not to cross a page boundary - */ -static unsigned char event_buffer[VIO_MAX_SUBTYPES * 256] __page_aligned; -static atomic_t event_buffer_available[VIO_MAX_SUBTYPES]; -static int event_buffer_initialised; - -static void handleMonitorEvent(struct HvLpEvent *event); - -/* - * We use this structure to handle asynchronous responses. The caller - * blocks on the semaphore and the handler posts the semaphore. However, - * if system_state is not SYSTEM_RUNNING, then wait_atomic is used ... - */ -struct alloc_parms { - struct semaphore sem; - int number; - atomic_t wait_atomic; - int used_wait_atomic; -}; - -/* Put a sequence number in each mon msg. The value is not - * important. Start at something other than 0 just for - * readability. wrapping this is ok. - */ -static u8 viomonseq = 22; - -/* Our hosting logical partition. We get this at startup - * time, and different modules access this variable directly. - */ -HvLpIndex viopath_hostLp = HvLpIndexInvalid; -EXPORT_SYMBOL(viopath_hostLp); -HvLpIndex viopath_ourLp = HvLpIndexInvalid; -EXPORT_SYMBOL(viopath_ourLp); - -/* For each kind of incoming event we set a pointer to a - * routine to call. - */ -static vio_event_handler_t *vio_handler[VIO_MAX_SUBTYPES]; - -#define VIOPATH_KERN_WARN KERN_WARNING "viopath: " -#define VIOPATH_KERN_INFO KERN_INFO "viopath: " - -static int proc_viopath_show(struct seq_file *m, void *v) -{ - char *buf; - u16 vlanMap; - dma_addr_t handle; - HvLpEvent_Rc hvrc; - DECLARE_MUTEX_LOCKED(Semaphore); - - buf = kmalloc(PAGE_SIZE, GFP_KERNEL); - if (!buf) - return 0; - memset(buf, 0, PAGE_SIZE); - - handle = dma_map_single(iSeries_vio_dev, buf, PAGE_SIZE, - DMA_FROM_DEVICE); - - hvrc = HvCallEvent_signalLpEventFast(viopath_hostLp, - HvLpEvent_Type_VirtualIo, - viomajorsubtype_config | vioconfigget, - HvLpEvent_AckInd_DoAck, HvLpEvent_AckType_ImmediateAck, - viopath_sourceinst(viopath_hostLp), - viopath_targetinst(viopath_hostLp), - (u64)(unsigned long)&Semaphore, VIOVERSION << 16, - ((u64)handle) << 32, PAGE_SIZE, 0, 0); - - if (hvrc != HvLpEvent_Rc_Good) - printk(VIOPATH_KERN_WARN "hv error on op %d\n", (int)hvrc); - - down(&Semaphore); - - vlanMap = HvLpConfig_getVirtualLanIndexMap(); - - buf[PAGE_SIZE-1] = '\0'; - seq_printf(m, "%s", buf); - seq_printf(m, "AVAILABLE_VETH=%x\n", vlanMap); - seq_printf(m, "SRLNBR=%c%c%c%c%c%c%c\n", - e2a(xItExtVpdPanel.mfgID[2]), - e2a(xItExtVpdPanel.mfgID[3]), - e2a(xItExtVpdPanel.systemSerial[1]), - e2a(xItExtVpdPanel.systemSerial[2]), - e2a(xItExtVpdPanel.systemSerial[3]), - e2a(xItExtVpdPanel.systemSerial[4]), - e2a(xItExtVpdPanel.systemSerial[5])); - - dma_unmap_single(iSeries_vio_dev, handle, PAGE_SIZE, DMA_FROM_DEVICE); - kfree(buf); - - return 0; -} - -static int proc_viopath_open(struct inode *inode, struct file *file) -{ - return single_open(file, proc_viopath_show, NULL); -} - -static struct file_operations proc_viopath_operations = { - .open = proc_viopath_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -static int __init vio_proc_init(void) -{ - struct proc_dir_entry *e; - - e = create_proc_entry("iSeries/config", 0, NULL); - if (e) - e->proc_fops = &proc_viopath_operations; - - return 0; -} -__initcall(vio_proc_init); - -/* See if a given LP is active. Allow for invalid lps to be passed in - * and just return invalid - */ -int viopath_isactive(HvLpIndex lp) -{ - if (lp == HvLpIndexInvalid) - return 0; - if (lp < HVMAXARCHITECTEDLPS) - return viopathStatus[lp].isActive; - else - return 0; -} -EXPORT_SYMBOL(viopath_isactive); - -/* - * We cache the source and target instance ids for each - * partition. - */ -HvLpInstanceId viopath_sourceinst(HvLpIndex lp) -{ - return viopathStatus[lp].mSourceInst; -} -EXPORT_SYMBOL(viopath_sourceinst); - -HvLpInstanceId viopath_targetinst(HvLpIndex lp) -{ - return viopathStatus[lp].mTargetInst; -} -EXPORT_SYMBOL(viopath_targetinst); - -/* - * Send a monitor message. This is a message with the acknowledge - * bit on that the other side will NOT explicitly acknowledge. When - * the other side goes down, the hypervisor will acknowledge any - * outstanding messages....so we will know when the other side dies. - */ -static void sendMonMsg(HvLpIndex remoteLp) -{ - HvLpEvent_Rc hvrc; - - viopathStatus[remoteLp].mSourceInst = - HvCallEvent_getSourceLpInstanceId(remoteLp, - HvLpEvent_Type_VirtualIo); - viopathStatus[remoteLp].mTargetInst = - HvCallEvent_getTargetLpInstanceId(remoteLp, - HvLpEvent_Type_VirtualIo); - - /* - * Deliberately ignore the return code here. if we call this - * more than once, we don't care. - */ - vio_setHandler(viomajorsubtype_monitor, handleMonitorEvent); - - hvrc = HvCallEvent_signalLpEventFast(remoteLp, HvLpEvent_Type_VirtualIo, - viomajorsubtype_monitor, HvLpEvent_AckInd_DoAck, - HvLpEvent_AckType_DeferredAck, - viopathStatus[remoteLp].mSourceInst, - viopathStatus[remoteLp].mTargetInst, - viomonseq++, 0, 0, 0, 0, 0); - - if (hvrc == HvLpEvent_Rc_Good) - viopathStatus[remoteLp].isActive = 1; - else { - printk(VIOPATH_KERN_WARN "could not connect to partition %d\n", - remoteLp); - viopathStatus[remoteLp].isActive = 0; - } -} - -static void handleMonitorEvent(struct HvLpEvent *event) -{ - HvLpIndex remoteLp; - int i; - - /* - * This handler is _also_ called as part of the loop - * at the end of this routine, so it must be able to - * ignore NULL events... - */ - if (!event) - return; - - /* - * First see if this is just a normal monitor message from the - * other partition - */ - if (event->xFlags.xFunction == HvLpEvent_Function_Int) { - remoteLp = event->xSourceLp; - if (!viopathStatus[remoteLp].isActive) - sendMonMsg(remoteLp); - return; - } - - /* - * This path is for an acknowledgement; the other partition - * died - */ - remoteLp = event->xTargetLp; - if ((event->xSourceInstanceId != viopathStatus[remoteLp].mSourceInst) || - (event->xTargetInstanceId != viopathStatus[remoteLp].mTargetInst)) { - printk(VIOPATH_KERN_WARN "ignoring ack....mismatched instances\n"); - return; - } - - printk(VIOPATH_KERN_WARN "partition %d ended\n", remoteLp); - - viopathStatus[remoteLp].isActive = 0; - - /* - * For each active handler, pass them a NULL - * message to indicate that the other partition - * died - */ - for (i = 0; i < VIO_MAX_SUBTYPES; i++) { - if (vio_handler[i] != NULL) - (*vio_handler[i])(NULL); - } -} - -int vio_setHandler(int subtype, vio_event_handler_t *beh) -{ - subtype = subtype >> VIOMAJOR_SUBTYPE_SHIFT; - if ((subtype < 0) || (subtype >= VIO_MAX_SUBTYPES)) - return -EINVAL; - if (vio_handler[subtype] != NULL) - return -EBUSY; - vio_handler[subtype] = beh; - return 0; -} -EXPORT_SYMBOL(vio_setHandler); - -int vio_clearHandler(int subtype) -{ - subtype = subtype >> VIOMAJOR_SUBTYPE_SHIFT; - if ((subtype < 0) || (subtype >= VIO_MAX_SUBTYPES)) - return -EINVAL; - if (vio_handler[subtype] == NULL) - return -EAGAIN; - vio_handler[subtype] = NULL; - return 0; -} -EXPORT_SYMBOL(vio_clearHandler); - -static void handleConfig(struct HvLpEvent *event) -{ - if (!event) - return; - if (event->xFlags.xFunction == HvLpEvent_Function_Int) { - printk(VIOPATH_KERN_WARN - "unexpected config request from partition %d", - event->xSourceLp); - - if ((event->xFlags.xFunction == HvLpEvent_Function_Int) && - (event->xFlags.xAckInd == HvLpEvent_AckInd_DoAck)) { - event->xRc = HvLpEvent_Rc_InvalidSubtype; - HvCallEvent_ackLpEvent(event); - } - return; - } - - up((struct semaphore *)event->xCorrelationToken); -} - -/* - * Initialization of the hosting partition - */ -void vio_set_hostlp(void) -{ - /* - * If this has already been set then we DON'T want to either change - * it or re-register the proc file system - */ - if (viopath_hostLp != HvLpIndexInvalid) - return; - - /* - * Figure out our hosting partition. This isn't allowed to change - * while we're active - */ - viopath_ourLp = HvLpConfig_getLpIndex(); - viopath_hostLp = HvLpConfig_getHostingLpIndex(viopath_ourLp); - - if (viopath_hostLp != HvLpIndexInvalid) - vio_setHandler(viomajorsubtype_config, handleConfig); -} -EXPORT_SYMBOL(vio_set_hostlp); - -static void vio_handleEvent(struct HvLpEvent *event, struct pt_regs *regs) -{ - HvLpIndex remoteLp; - int subtype = (event->xSubtype & VIOMAJOR_SUBTYPE_MASK) - >> VIOMAJOR_SUBTYPE_SHIFT; - - if (event->xFlags.xFunction == HvLpEvent_Function_Int) { - remoteLp = event->xSourceLp; - /* - * The isActive is checked because if the hosting partition - * went down and came back up it would not be active but it - * would have different source and target instances, in which - * case we'd want to reset them. This case really protects - * against an unauthorized active partition sending interrupts - * or acks to this linux partition. - */ - if (viopathStatus[remoteLp].isActive - && (event->xSourceInstanceId != - viopathStatus[remoteLp].mTargetInst)) { - printk(VIOPATH_KERN_WARN - "message from invalid partition. " - "int msg rcvd, source inst (%d) doesnt match (%d)\n", - viopathStatus[remoteLp].mTargetInst, - event->xSourceInstanceId); - return; - } - - if (viopathStatus[remoteLp].isActive - && (event->xTargetInstanceId != - viopathStatus[remoteLp].mSourceInst)) { - printk(VIOPATH_KERN_WARN - "message from invalid partition. " - "int msg rcvd, target inst (%d) doesnt match (%d)\n", - viopathStatus[remoteLp].mSourceInst, - event->xTargetInstanceId); - return; - } - } else { - remoteLp = event->xTargetLp; - if (event->xSourceInstanceId != - viopathStatus[remoteLp].mSourceInst) { - printk(VIOPATH_KERN_WARN - "message from invalid partition. " - "ack msg rcvd, source inst (%d) doesnt match (%d)\n", - viopathStatus[remoteLp].mSourceInst, - event->xSourceInstanceId); - return; - } - - if (event->xTargetInstanceId != - viopathStatus[remoteLp].mTargetInst) { - printk(VIOPATH_KERN_WARN - "message from invalid partition. " - "viopath: ack msg rcvd, target inst (%d) doesnt match (%d)\n", - viopathStatus[remoteLp].mTargetInst, - event->xTargetInstanceId); - return; - } - } - - if (vio_handler[subtype] == NULL) { - printk(VIOPATH_KERN_WARN - "unexpected virtual io event subtype %d from partition %d\n", - event->xSubtype, remoteLp); - /* No handler. Ack if necessary */ - if ((event->xFlags.xFunction == HvLpEvent_Function_Int) && - (event->xFlags.xAckInd == HvLpEvent_AckInd_DoAck)) { - event->xRc = HvLpEvent_Rc_InvalidSubtype; - HvCallEvent_ackLpEvent(event); - } - return; - } - - /* This innocuous little line is where all the real work happens */ - (*vio_handler[subtype])(event); -} - -static void viopath_donealloc(void *parm, int number) -{ - struct alloc_parms *parmsp = parm; - - parmsp->number = number; - if (parmsp->used_wait_atomic) - atomic_set(&parmsp->wait_atomic, 0); - else - up(&parmsp->sem); -} - -static int allocateEvents(HvLpIndex remoteLp, int numEvents) -{ - struct alloc_parms parms; - - if (system_state != SYSTEM_RUNNING) { - parms.used_wait_atomic = 1; - atomic_set(&parms.wait_atomic, 1); - } else { - parms.used_wait_atomic = 0; - init_MUTEX_LOCKED(&parms.sem); - } - mf_allocate_lp_events(remoteLp, HvLpEvent_Type_VirtualIo, 250, /* It would be nice to put a real number here! */ - numEvents, &viopath_donealloc, &parms); - if (system_state != SYSTEM_RUNNING) { - while (atomic_read(&parms.wait_atomic)) - mb(); - } else - down(&parms.sem); - return parms.number; -} - -int viopath_open(HvLpIndex remoteLp, int subtype, int numReq) -{ - int i; - unsigned long flags; - int tempNumAllocated; - - if ((remoteLp >= HVMAXARCHITECTEDLPS) || (remoteLp == HvLpIndexInvalid)) - return -EINVAL; - - subtype = subtype >> VIOMAJOR_SUBTYPE_SHIFT; - if ((subtype < 0) || (subtype >= VIO_MAX_SUBTYPES)) - return -EINVAL; - - spin_lock_irqsave(&statuslock, flags); - - if (!event_buffer_initialised) { - for (i = 0; i < VIO_MAX_SUBTYPES; i++) - atomic_set(&event_buffer_available[i], 1); - event_buffer_initialised = 1; - } - - viopathStatus[remoteLp].users[subtype]++; - - if (!viopathStatus[remoteLp].isOpen) { - viopathStatus[remoteLp].isOpen = 1; - HvCallEvent_openLpEventPath(remoteLp, HvLpEvent_Type_VirtualIo); - - /* - * Don't hold the spinlock during an operation that - * can sleep. - */ - spin_unlock_irqrestore(&statuslock, flags); - tempNumAllocated = allocateEvents(remoteLp, 1); - spin_lock_irqsave(&statuslock, flags); - - viopathStatus[remoteLp].numberAllocated += tempNumAllocated; - - if (viopathStatus[remoteLp].numberAllocated == 0) { - HvCallEvent_closeLpEventPath(remoteLp, - HvLpEvent_Type_VirtualIo); - - spin_unlock_irqrestore(&statuslock, flags); - return -ENOMEM; - } - - viopathStatus[remoteLp].mSourceInst = - HvCallEvent_getSourceLpInstanceId(remoteLp, - HvLpEvent_Type_VirtualIo); - viopathStatus[remoteLp].mTargetInst = - HvCallEvent_getTargetLpInstanceId(remoteLp, - HvLpEvent_Type_VirtualIo); - HvLpEvent_registerHandler(HvLpEvent_Type_VirtualIo, - &vio_handleEvent); - sendMonMsg(remoteLp); - printk(VIOPATH_KERN_INFO "opening connection to partition %d, " - "setting sinst %d, tinst %d\n", - remoteLp, viopathStatus[remoteLp].mSourceInst, - viopathStatus[remoteLp].mTargetInst); - } - - spin_unlock_irqrestore(&statuslock, flags); - tempNumAllocated = allocateEvents(remoteLp, numReq); - spin_lock_irqsave(&statuslock, flags); - viopathStatus[remoteLp].numberAllocated += tempNumAllocated; - spin_unlock_irqrestore(&statuslock, flags); - - return 0; -} -EXPORT_SYMBOL(viopath_open); - -int viopath_close(HvLpIndex remoteLp, int subtype, int numReq) -{ - unsigned long flags; - int i; - int numOpen; - struct alloc_parms parms; - - if ((remoteLp >= HVMAXARCHITECTEDLPS) || (remoteLp == HvLpIndexInvalid)) - return -EINVAL; - - subtype = subtype >> VIOMAJOR_SUBTYPE_SHIFT; - if ((subtype < 0) || (subtype >= VIO_MAX_SUBTYPES)) - return -EINVAL; - - spin_lock_irqsave(&statuslock, flags); - /* - * If the viopath_close somehow gets called before a - * viopath_open it could decrement to -1 which is a non - * recoverable state so we'll prevent this from - * happening. - */ - if (viopathStatus[remoteLp].users[subtype] > 0) - viopathStatus[remoteLp].users[subtype]--; - - spin_unlock_irqrestore(&statuslock, flags); - - parms.used_wait_atomic = 0; - init_MUTEX_LOCKED(&parms.sem); - mf_deallocate_lp_events(remoteLp, HvLpEvent_Type_VirtualIo, - numReq, &viopath_donealloc, &parms); - down(&parms.sem); - - spin_lock_irqsave(&statuslock, flags); - for (i = 0, numOpen = 0; i < VIO_MAX_SUBTYPES; i++) - numOpen += viopathStatus[remoteLp].users[i]; - - if ((viopathStatus[remoteLp].isOpen) && (numOpen == 0)) { - printk(VIOPATH_KERN_INFO "closing connection to partition %d", - remoteLp); - - HvCallEvent_closeLpEventPath(remoteLp, - HvLpEvent_Type_VirtualIo); - viopathStatus[remoteLp].isOpen = 0; - viopathStatus[remoteLp].isActive = 0; - - for (i = 0; i < VIO_MAX_SUBTYPES; i++) - atomic_set(&event_buffer_available[i], 0); - event_buffer_initialised = 0; - } - spin_unlock_irqrestore(&statuslock, flags); - return 0; -} -EXPORT_SYMBOL(viopath_close); - -void *vio_get_event_buffer(int subtype) -{ - subtype = subtype >> VIOMAJOR_SUBTYPE_SHIFT; - if ((subtype < 0) || (subtype >= VIO_MAX_SUBTYPES)) - return NULL; - - if (atomic_dec_if_positive(&event_buffer_available[subtype]) == 0) - return &event_buffer[subtype * 256]; - else - return NULL; -} -EXPORT_SYMBOL(vio_get_event_buffer); - -void vio_free_event_buffer(int subtype, void *buffer) -{ - subtype = subtype >> VIOMAJOR_SUBTYPE_SHIFT; - if ((subtype < 0) || (subtype >= VIO_MAX_SUBTYPES)) { - printk(VIOPATH_KERN_WARN - "unexpected subtype %d freeing event buffer\n", subtype); - return; - } - - if (atomic_read(&event_buffer_available[subtype]) != 0) { - printk(VIOPATH_KERN_WARN - "freeing unallocated event buffer, subtype %d\n", - subtype); - return; - } - - if (buffer != &event_buffer[subtype * 256]) { - printk(VIOPATH_KERN_WARN - "freeing invalid event buffer, subtype %d\n", subtype); - } - - atomic_set(&event_buffer_available[subtype], 1); -} -EXPORT_SYMBOL(vio_free_event_buffer); - -static const struct vio_error_entry vio_no_error = - { 0, 0, "Non-VIO Error" }; -static const struct vio_error_entry vio_unknown_error = - { 0, EIO, "Unknown Error" }; - -static const struct vio_error_entry vio_default_errors[] = { - {0x0001, EIO, "No Connection"}, - {0x0002, EIO, "No Receiver"}, - {0x0003, EIO, "No Buffer Available"}, - {0x0004, EBADRQC, "Invalid Message Type"}, - {0x0000, 0, NULL}, -}; - -const struct vio_error_entry *vio_lookup_rc( - const struct vio_error_entry *local_table, u16 rc) -{ - const struct vio_error_entry *cur; - - if (!rc) - return &vio_no_error; - if (local_table) - for (cur = local_table; cur->rc; ++cur) - if (cur->rc == rc) - return cur; - for (cur = vio_default_errors; cur->rc; ++cur) - if (cur->rc == rc) - return cur; - return &vio_unknown_error; -} -EXPORT_SYMBOL(vio_lookup_rc); diff --git a/arch/ppc64/kernel/vmlinux.lds.S b/arch/ppc64/kernel/vmlinux.lds.S index 0306510bc4ff..022f220e772f 100644 --- a/arch/ppc64/kernel/vmlinux.lds.S +++ b/arch/ppc64/kernel/vmlinux.lds.S @@ -1,3 +1,4 @@ +#include <asm/page.h> #include <asm-generic/vmlinux.lds.h> OUTPUT_ARCH(powerpc:common64) @@ -17,7 +18,7 @@ SECTIONS LOCK_TEXT KPROBES_TEXT *(.fixup) - . = ALIGN(4096); + . = ALIGN(PAGE_SIZE); _etext = .; } @@ -43,7 +44,7 @@ SECTIONS /* will be freed after init */ - . = ALIGN(4096); + . = ALIGN(PAGE_SIZE); __init_begin = .; .init.text : { @@ -83,7 +84,7 @@ SECTIONS SECURITY_INIT - . = ALIGN(4096); + . = ALIGN(PAGE_SIZE); .init.ramfs : { __initramfs_start = .; *(.init.ramfs) @@ -96,18 +97,22 @@ SECTIONS __per_cpu_end = .; } + . = ALIGN(PAGE_SIZE); . = ALIGN(16384); __init_end = .; /* freed after init ends here */ /* Read/write sections */ + . = ALIGN(PAGE_SIZE); . = ALIGN(16384); + _sdata = .; /* The initial task and kernel stack */ .data.init_task : { *(.data.init_task) } + . = ALIGN(PAGE_SIZE); .data.page_aligned : { *(.data.page_aligned) } @@ -129,18 +134,18 @@ SECTIONS __toc_start = .; *(.got) *(.toc) - . = ALIGN(4096); + . = ALIGN(PAGE_SIZE); _edata = .; } - . = ALIGN(4096); + . = ALIGN(PAGE_SIZE); .bss : { __bss_start = .; *(.bss) __bss_stop = .; } - . = ALIGN(4096); + . = ALIGN(PAGE_SIZE); _end = . ; } diff --git a/arch/ppc64/mm/Makefile b/arch/ppc64/mm/Makefile deleted file mode 100644 index 3695d00d347f..000000000000 --- a/arch/ppc64/mm/Makefile +++ /dev/null @@ -1,11 +0,0 @@ -# -# Makefile for the linux ppc-specific parts of the memory manager. -# - -EXTRA_CFLAGS += -mno-minimal-toc - -obj-y := fault.o init.o imalloc.o hash_utils.o hash_low.o tlb.o \ - slb_low.o slb.o stab.o mmap.o -obj-$(CONFIG_NEED_MULTIPLE_NODES) += numa.o -obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o -obj-$(CONFIG_PPC_MULTIPLATFORM) += hash_native.o diff --git a/arch/ppc64/mm/fault.c b/arch/ppc64/mm/fault.c deleted file mode 100644 index be3f25cf3e9f..000000000000 --- a/arch/ppc64/mm/fault.c +++ /dev/null @@ -1,333 +0,0 @@ -/* - * arch/ppc/mm/fault.c - * - * PowerPC version - * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org) - * - * Derived from "arch/i386/mm/fault.c" - * Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds - * - * Modified by Cort Dougan and Paul Mackerras. - * - * Modified for PPC64 by Dave Engebretsen (engebret@ibm.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/config.h> -#include <linux/signal.h> -#include <linux/sched.h> -#include <linux/kernel.h> -#include <linux/errno.h> -#include <linux/string.h> -#include <linux/types.h> -#include <linux/mman.h> -#include <linux/mm.h> -#include <linux/interrupt.h> -#include <linux/smp_lock.h> -#include <linux/module.h> -#include <linux/kprobes.h> - -#include <asm/page.h> -#include <asm/pgtable.h> -#include <asm/mmu.h> -#include <asm/mmu_context.h> -#include <asm/system.h> -#include <asm/uaccess.h> -#include <asm/kdebug.h> -#include <asm/siginfo.h> - -/* - * Check whether the instruction at regs->nip is a store using - * an update addressing form which will update r1. - */ -static int store_updates_sp(struct pt_regs *regs) -{ - unsigned int inst; - - if (get_user(inst, (unsigned int __user *)regs->nip)) - return 0; - /* check for 1 in the rA field */ - if (((inst >> 16) & 0x1f) != 1) - return 0; - /* check major opcode */ - switch (inst >> 26) { - case 37: /* stwu */ - case 39: /* stbu */ - case 45: /* sthu */ - case 53: /* stfsu */ - case 55: /* stfdu */ - return 1; - case 62: /* std or stdu */ - return (inst & 3) == 1; - case 31: - /* check minor opcode */ - switch ((inst >> 1) & 0x3ff) { - case 181: /* stdux */ - case 183: /* stwux */ - case 247: /* stbux */ - case 439: /* sthux */ - case 695: /* stfsux */ - case 759: /* stfdux */ - return 1; - } - } - return 0; -} - -static void do_dabr(struct pt_regs *regs, unsigned long error_code) -{ - siginfo_t info; - - if (notify_die(DIE_DABR_MATCH, "dabr_match", regs, error_code, - 11, SIGSEGV) == NOTIFY_STOP) - return; - - if (debugger_dabr_match(regs)) - return; - - /* Clear the DABR */ - set_dabr(0); - - /* Deliver the signal to userspace */ - info.si_signo = SIGTRAP; - info.si_errno = 0; - info.si_code = TRAP_HWBKPT; - info.si_addr = (void __user *)regs->nip; - force_sig_info(SIGTRAP, &info, current); -} - -/* - * The error_code parameter is - * - DSISR for a non-SLB data access fault, - * - SRR1 & 0x08000000 for a non-SLB instruction access fault - * - 0 any SLB fault. - * The return value is 0 if the fault was handled, or the signal - * number if this is a kernel fault that can't be handled here. - */ -int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address, - unsigned long error_code) -{ - struct vm_area_struct * vma; - struct mm_struct *mm = current->mm; - siginfo_t info; - unsigned long code = SEGV_MAPERR; - unsigned long is_write = error_code & DSISR_ISSTORE; - unsigned long trap = TRAP(regs); - unsigned long is_exec = trap == 0x400; - - BUG_ON((trap == 0x380) || (trap == 0x480)); - - if (notify_die(DIE_PAGE_FAULT, "page_fault", regs, error_code, - 11, SIGSEGV) == NOTIFY_STOP) - return 0; - - if (trap == 0x300) { - if (debugger_fault_handler(regs)) - return 0; - } - - /* On a kernel SLB miss we can only check for a valid exception entry */ - if (!user_mode(regs) && (address >= TASK_SIZE)) - return SIGSEGV; - - if (error_code & DSISR_DABRMATCH) { - do_dabr(regs, error_code); - return 0; - } - - if (in_atomic() || mm == NULL) { - if (!user_mode(regs)) - return SIGSEGV; - /* in_atomic() in user mode is really bad, - as is current->mm == NULL. */ - printk(KERN_EMERG "Page fault in user mode with" - "in_atomic() = %d mm = %p\n", in_atomic(), mm); - printk(KERN_EMERG "NIP = %lx MSR = %lx\n", - regs->nip, regs->msr); - die("Weird page fault", regs, SIGSEGV); - } - - /* When running in the kernel we expect faults to occur only to - * addresses in user space. All other faults represent errors in the - * kernel and should generate an OOPS. Unfortunatly, in the case of an - * erroneous fault occuring in a code path which already holds mmap_sem - * we will deadlock attempting to validate the fault against the - * address space. Luckily the kernel only validly references user - * space from well defined areas of code, which are listed in the - * exceptions table. - * - * As the vast majority of faults will be valid we will only perform - * the source reference check when there is a possibilty of a deadlock. - * Attempt to lock the address space, if we cannot we then validate the - * source. If this is invalid we can skip the address space check, - * thus avoiding the deadlock. - */ - if (!down_read_trylock(&mm->mmap_sem)) { - if (!user_mode(regs) && !search_exception_tables(regs->nip)) - goto bad_area_nosemaphore; - - down_read(&mm->mmap_sem); - } - - vma = find_vma(mm, address); - if (!vma) - goto bad_area; - - if (vma->vm_start <= address) { - goto good_area; - } - if (!(vma->vm_flags & VM_GROWSDOWN)) - goto bad_area; - - /* - * N.B. The POWER/Open ABI allows programs to access up to - * 288 bytes below the stack pointer. - * The kernel signal delivery code writes up to about 1.5kB - * below the stack pointer (r1) before decrementing it. - * The exec code can write slightly over 640kB to the stack - * before setting the user r1. Thus we allow the stack to - * expand to 1MB without further checks. - */ - if (address + 0x100000 < vma->vm_end) { - /* get user regs even if this fault is in kernel mode */ - struct pt_regs *uregs = current->thread.regs; - if (uregs == NULL) - goto bad_area; - - /* - * A user-mode access to an address a long way below - * the stack pointer is only valid if the instruction - * is one which would update the stack pointer to the - * address accessed if the instruction completed, - * i.e. either stwu rs,n(r1) or stwux rs,r1,rb - * (or the byte, halfword, float or double forms). - * - * If we don't check this then any write to the area - * between the last mapped region and the stack will - * expand the stack rather than segfaulting. - */ - if (address + 2048 < uregs->gpr[1] - && (!user_mode(regs) || !store_updates_sp(regs))) - goto bad_area; - } - - if (expand_stack(vma, address)) - goto bad_area; - -good_area: - code = SEGV_ACCERR; - - if (is_exec) { - /* protection fault */ - if (error_code & DSISR_PROTFAULT) - goto bad_area; - if (!(vma->vm_flags & VM_EXEC)) - goto bad_area; - /* a write */ - } else if (is_write) { - if (!(vma->vm_flags & VM_WRITE)) - goto bad_area; - /* a read */ - } else { - if (!(vma->vm_flags & VM_READ)) - goto bad_area; - } - - survive: - /* - * If for any reason at all we couldn't handle the fault, - * make sure we exit gracefully rather than endlessly redo - * the fault. - */ - switch (handle_mm_fault(mm, vma, address, is_write)) { - - case VM_FAULT_MINOR: - current->min_flt++; - break; - case VM_FAULT_MAJOR: - current->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto do_sigbus; - case VM_FAULT_OOM: - goto out_of_memory; - default: - BUG(); - } - - up_read(&mm->mmap_sem); - return 0; - -bad_area: - up_read(&mm->mmap_sem); - -bad_area_nosemaphore: - /* User mode accesses cause a SIGSEGV */ - if (user_mode(regs)) { - info.si_signo = SIGSEGV; - info.si_errno = 0; - info.si_code = code; - info.si_addr = (void __user *) address; - force_sig_info(SIGSEGV, &info, current); - return 0; - } - - if (trap == 0x400 && (error_code & DSISR_PROTFAULT) - && printk_ratelimit()) - printk(KERN_CRIT "kernel tried to execute NX-protected" - " page (%lx) - exploit attempt? (uid: %d)\n", - address, current->uid); - - return SIGSEGV; - -/* - * We ran out of memory, or some other thing happened to us that made - * us unable to handle the page fault gracefully. - */ -out_of_memory: - up_read(&mm->mmap_sem); - if (current->pid == 1) { - yield(); - down_read(&mm->mmap_sem); - goto survive; - } - printk("VM: killing process %s\n", current->comm); - if (user_mode(regs)) - do_exit(SIGKILL); - return SIGKILL; - -do_sigbus: - up_read(&mm->mmap_sem); - if (user_mode(regs)) { - info.si_signo = SIGBUS; - info.si_errno = 0; - info.si_code = BUS_ADRERR; - info.si_addr = (void __user *)address; - force_sig_info(SIGBUS, &info, current); - return 0; - } - return SIGBUS; -} - -/* - * bad_page_fault is called when we have a bad access from the kernel. - * It is called from do_page_fault above and from some of the procedures - * in traps.c. - */ -void bad_page_fault(struct pt_regs *regs, unsigned long address, int sig) -{ - const struct exception_table_entry *entry; - - /* Are we prepared to handle this fault? */ - if ((entry = search_exception_tables(regs->nip)) != NULL) { - regs->nip = entry->fixup; - return; - } - - /* kernel has accessed a bad area */ - die("Kernel access of bad area", regs, sig); -} diff --git a/arch/ppc64/mm/hash_low.S b/arch/ppc64/mm/hash_low.S deleted file mode 100644 index ee5a5d36bfa8..000000000000 --- a/arch/ppc64/mm/hash_low.S +++ /dev/null @@ -1,288 +0,0 @@ -/* - * ppc64 MMU hashtable management routines - * - * (c) Copyright IBM Corp. 2003 - * - * Maintained by: Benjamin Herrenschmidt - * <benh@kernel.crashing.org> - * - * This file is covered by the GNU Public Licence v2 as - * described in the kernel's COPYING file. - */ - -#include <asm/processor.h> -#include <asm/pgtable.h> -#include <asm/mmu.h> -#include <asm/page.h> -#include <asm/types.h> -#include <asm/ppc_asm.h> -#include <asm/asm-offsets.h> -#include <asm/cputable.h> - - .text - -/* - * Stackframe: - * - * +-> Back chain (SP + 256) - * | General register save area (SP + 112) - * | Parameter save area (SP + 48) - * | TOC save area (SP + 40) - * | link editor doubleword (SP + 32) - * | compiler doubleword (SP + 24) - * | LR save area (SP + 16) - * | CR save area (SP + 8) - * SP ---> +-- Back chain (SP + 0) - */ -#define STACKFRAMESIZE 256 - -/* Save parameters offsets */ -#define STK_PARM(i) (STACKFRAMESIZE + 48 + ((i)-3)*8) - -/* Save non-volatile offsets */ -#define STK_REG(i) (112 + ((i)-14)*8) - -/* - * _hash_page(unsigned long ea, unsigned long access, unsigned long vsid, - * pte_t *ptep, unsigned long trap, int local) - * - * Adds a page to the hash table. This is the non-LPAR version for now - */ - -_GLOBAL(__hash_page) - mflr r0 - std r0,16(r1) - stdu r1,-STACKFRAMESIZE(r1) - /* Save all params that we need after a function call */ - std r6,STK_PARM(r6)(r1) - std r8,STK_PARM(r8)(r1) - - /* Add _PAGE_PRESENT to access */ - ori r4,r4,_PAGE_PRESENT - - /* Save non-volatile registers. - * r31 will hold "old PTE" - * r30 is "new PTE" - * r29 is "va" - * r28 is a hash value - * r27 is hashtab mask (maybe dynamic patched instead ?) - */ - std r27,STK_REG(r27)(r1) - std r28,STK_REG(r28)(r1) - std r29,STK_REG(r29)(r1) - std r30,STK_REG(r30)(r1) - std r31,STK_REG(r31)(r1) - - /* Step 1: - * - * Check permissions, atomically mark the linux PTE busy - * and hashed. - */ -1: - ldarx r31,0,r6 - /* Check access rights (access & ~(pte_val(*ptep))) */ - andc. r0,r4,r31 - bne- htab_wrong_access - /* Check if PTE is busy */ - andi. r0,r31,_PAGE_BUSY - /* If so, just bail out and refault if needed. Someone else - * is changing this PTE anyway and might hash it. - */ - bne- bail_ok - /* Prepare new PTE value (turn access RW into DIRTY, then - * add BUSY,HASHPTE and ACCESSED) - */ - rlwinm r30,r4,32-9+7,31-7,31-7 /* _PAGE_RW -> _PAGE_DIRTY */ - or r30,r30,r31 - ori r30,r30,_PAGE_BUSY | _PAGE_ACCESSED | _PAGE_HASHPTE - /* Write the linux PTE atomically (setting busy) */ - stdcx. r30,0,r6 - bne- 1b - isync - - /* Step 2: - * - * Insert/Update the HPTE in the hash table. At this point, - * r4 (access) is re-useable, we use it for the new HPTE flags - */ - - /* Calc va and put it in r29 */ - rldicr r29,r5,28,63-28 - rldicl r3,r3,0,36 - or r29,r3,r29 - - /* Calculate hash value for primary slot and store it in r28 */ - rldicl r5,r5,0,25 /* vsid & 0x0000007fffffffff */ - rldicl r0,r3,64-12,48 /* (ea >> 12) & 0xffff */ - xor r28,r5,r0 - - /* Convert linux PTE bits into HW equivalents */ - andi. r3,r30,0x1fe /* Get basic set of flags */ - xori r3,r3,HW_NO_EXEC /* _PAGE_EXEC -> NOEXEC */ - rlwinm r0,r30,32-9+1,30,30 /* _PAGE_RW -> _PAGE_USER (r0) */ - rlwinm r4,r30,32-7+1,30,30 /* _PAGE_DIRTY -> _PAGE_USER (r4) */ - and r0,r0,r4 /* _PAGE_RW & _PAGE_DIRTY -> r0 bit 30 */ - andc r0,r30,r0 /* r0 = pte & ~r0 */ - rlwimi r3,r0,32-1,31,31 /* Insert result into PP lsb */ - - /* We eventually do the icache sync here (maybe inline that - * code rather than call a C function...) - */ -BEGIN_FTR_SECTION - mr r4,r30 - mr r5,r7 - bl .hash_page_do_lazy_icache -END_FTR_SECTION(CPU_FTR_NOEXECUTE|CPU_FTR_COHERENT_ICACHE, CPU_FTR_NOEXECUTE) - - /* At this point, r3 contains new PP bits, save them in - * place of "access" in the param area (sic) - */ - std r3,STK_PARM(r4)(r1) - - /* Get htab_hash_mask */ - ld r4,htab_hash_mask@got(2) - ld r27,0(r4) /* htab_hash_mask -> r27 */ - - /* Check if we may already be in the hashtable, in this case, we - * go to out-of-line code to try to modify the HPTE - */ - andi. r0,r31,_PAGE_HASHPTE - bne htab_modify_pte - -htab_insert_pte: - /* Clear hpte bits in new pte (we also clear BUSY btw) and - * add _PAGE_HASHPTE - */ - lis r0,_PAGE_HPTEFLAGS@h - ori r0,r0,_PAGE_HPTEFLAGS@l - andc r30,r30,r0 - ori r30,r30,_PAGE_HASHPTE - - /* page number in r5 */ - rldicl r5,r31,64-PTE_SHIFT,PTE_SHIFT - - /* Calculate primary group hash */ - and r0,r28,r27 - rldicr r3,r0,3,63-3 /* r0 = (hash & mask) << 3 */ - - /* Call ppc_md.hpte_insert */ - ld r7,STK_PARM(r4)(r1) /* Retreive new pp bits */ - mr r4,r29 /* Retreive va */ - li r6,0 /* no vflags */ -_GLOBAL(htab_call_hpte_insert1) - bl . /* Will be patched by htab_finish_init() */ - cmpdi 0,r3,0 - bge htab_pte_insert_ok /* Insertion successful */ - cmpdi 0,r3,-2 /* Critical failure */ - beq- htab_pte_insert_failure - - /* Now try secondary slot */ - - /* page number in r5 */ - rldicl r5,r31,64-PTE_SHIFT,PTE_SHIFT - - /* Calculate secondary group hash */ - andc r0,r27,r28 - rldicr r3,r0,3,63-3 /* r0 = (~hash & mask) << 3 */ - - /* Call ppc_md.hpte_insert */ - ld r7,STK_PARM(r4)(r1) /* Retreive new pp bits */ - mr r4,r29 /* Retreive va */ - li r6,HPTE_V_SECONDARY@l /* secondary slot */ -_GLOBAL(htab_call_hpte_insert2) - bl . /* Will be patched by htab_finish_init() */ - cmpdi 0,r3,0 - bge+ htab_pte_insert_ok /* Insertion successful */ - cmpdi 0,r3,-2 /* Critical failure */ - beq- htab_pte_insert_failure - - /* Both are full, we need to evict something */ - mftb r0 - /* Pick a random group based on TB */ - andi. r0,r0,1 - mr r5,r28 - bne 2f - not r5,r5 -2: and r0,r5,r27 - rldicr r3,r0,3,63-3 /* r0 = (hash & mask) << 3 */ - /* Call ppc_md.hpte_remove */ -_GLOBAL(htab_call_hpte_remove) - bl . /* Will be patched by htab_finish_init() */ - - /* Try all again */ - b htab_insert_pte - -bail_ok: - li r3,0 - b bail - -htab_pte_insert_ok: - /* Insert slot number & secondary bit in PTE */ - rldimi r30,r3,12,63-15 - - /* Write out the PTE with a normal write - * (maybe add eieio may be good still ?) - */ -htab_write_out_pte: - ld r6,STK_PARM(r6)(r1) - std r30,0(r6) - li r3, 0 -bail: - ld r27,STK_REG(r27)(r1) - ld r28,STK_REG(r28)(r1) - ld r29,STK_REG(r29)(r1) - ld r30,STK_REG(r30)(r1) - ld r31,STK_REG(r31)(r1) - addi r1,r1,STACKFRAMESIZE - ld r0,16(r1) - mtlr r0 - blr - -htab_modify_pte: - /* Keep PP bits in r4 and slot idx from the PTE around in r3 */ - mr r4,r3 - rlwinm r3,r31,32-12,29,31 - - /* Secondary group ? if yes, get a inverted hash value */ - mr r5,r28 - andi. r0,r31,_PAGE_SECONDARY - beq 1f - not r5,r5 -1: - /* Calculate proper slot value for ppc_md.hpte_updatepp */ - and r0,r5,r27 - rldicr r0,r0,3,63-3 /* r0 = (hash & mask) << 3 */ - add r3,r0,r3 /* add slot idx */ - - /* Call ppc_md.hpte_updatepp */ - mr r5,r29 /* va */ - li r6,0 /* large is 0 */ - ld r7,STK_PARM(r8)(r1) /* get "local" param */ -_GLOBAL(htab_call_hpte_updatepp) - bl . /* Will be patched by htab_finish_init() */ - - /* if we failed because typically the HPTE wasn't really here - * we try an insertion. - */ - cmpdi 0,r3,-1 - beq- htab_insert_pte - - /* Clear the BUSY bit and Write out the PTE */ - li r0,_PAGE_BUSY - andc r30,r30,r0 - b htab_write_out_pte - -htab_wrong_access: - /* Bail out clearing reservation */ - stdcx. r31,0,r6 - li r3,1 - b bail - -htab_pte_insert_failure: - /* Bail out restoring old PTE */ - ld r6,STK_PARM(r6)(r1) - std r31,0(r6) - li r3,-1 - b bail - - diff --git a/arch/ppc64/mm/hash_native.c b/arch/ppc64/mm/hash_native.c deleted file mode 100644 index bfd385b7713c..000000000000 --- a/arch/ppc64/mm/hash_native.c +++ /dev/null @@ -1,453 +0,0 @@ -/* - * native hashtable management. - * - * SMP scalability work: - * Copyright (C) 2001 Anton Blanchard <anton@au.ibm.com>, IBM - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#include <linux/spinlock.h> -#include <linux/bitops.h> -#include <linux/threads.h> -#include <linux/smp.h> - -#include <asm/abs_addr.h> -#include <asm/machdep.h> -#include <asm/mmu.h> -#include <asm/mmu_context.h> -#include <asm/pgtable.h> -#include <asm/tlbflush.h> -#include <asm/tlb.h> -#include <asm/cputable.h> - -#define HPTE_LOCK_BIT 3 - -static DEFINE_SPINLOCK(native_tlbie_lock); - -static inline void native_lock_hpte(hpte_t *hptep) -{ - unsigned long *word = &hptep->v; - - while (1) { - if (!test_and_set_bit(HPTE_LOCK_BIT, word)) - break; - while(test_bit(HPTE_LOCK_BIT, word)) - cpu_relax(); - } -} - -static inline void native_unlock_hpte(hpte_t *hptep) -{ - unsigned long *word = &hptep->v; - - asm volatile("lwsync":::"memory"); - clear_bit(HPTE_LOCK_BIT, word); -} - -long native_hpte_insert(unsigned long hpte_group, unsigned long va, - unsigned long prpn, unsigned long vflags, - unsigned long rflags) -{ - hpte_t *hptep = htab_address + hpte_group; - unsigned long hpte_v, hpte_r; - int i; - - for (i = 0; i < HPTES_PER_GROUP; i++) { - if (! (hptep->v & HPTE_V_VALID)) { - /* retry with lock held */ - native_lock_hpte(hptep); - if (! (hptep->v & HPTE_V_VALID)) - break; - native_unlock_hpte(hptep); - } - - hptep++; - } - - if (i == HPTES_PER_GROUP) - return -1; - - hpte_v = (va >> 23) << HPTE_V_AVPN_SHIFT | vflags | HPTE_V_VALID; - if (vflags & HPTE_V_LARGE) - va &= ~(1UL << HPTE_V_AVPN_SHIFT); - hpte_r = (prpn << HPTE_R_RPN_SHIFT) | rflags; - - hptep->r = hpte_r; - /* Guarantee the second dword is visible before the valid bit */ - __asm__ __volatile__ ("eieio" : : : "memory"); - /* - * Now set the first dword including the valid bit - * NOTE: this also unlocks the hpte - */ - hptep->v = hpte_v; - - __asm__ __volatile__ ("ptesync" : : : "memory"); - - return i | (!!(vflags & HPTE_V_SECONDARY) << 3); -} - -static long native_hpte_remove(unsigned long hpte_group) -{ - hpte_t *hptep; - int i; - int slot_offset; - unsigned long hpte_v; - - /* pick a random entry to start at */ - slot_offset = mftb() & 0x7; - - for (i = 0; i < HPTES_PER_GROUP; i++) { - hptep = htab_address + hpte_group + slot_offset; - hpte_v = hptep->v; - - if ((hpte_v & HPTE_V_VALID) && !(hpte_v & HPTE_V_BOLTED)) { - /* retry with lock held */ - native_lock_hpte(hptep); - hpte_v = hptep->v; - if ((hpte_v & HPTE_V_VALID) - && !(hpte_v & HPTE_V_BOLTED)) - break; - native_unlock_hpte(hptep); - } - - slot_offset++; - slot_offset &= 0x7; - } - - if (i == HPTES_PER_GROUP) - return -1; - - /* Invalidate the hpte. NOTE: this also unlocks it */ - hptep->v = 0; - - return i; -} - -static inline void set_pp_bit(unsigned long pp, hpte_t *addr) -{ - unsigned long old; - unsigned long *p = &addr->r; - - __asm__ __volatile__( - "1: ldarx %0,0,%3\n\ - rldimi %0,%2,0,61\n\ - stdcx. %0,0,%3\n\ - bne 1b" - : "=&r" (old), "=m" (*p) - : "r" (pp), "r" (p), "m" (*p) - : "cc"); -} - -/* - * Only works on small pages. Yes its ugly to have to check each slot in - * the group but we only use this during bootup. - */ -static long native_hpte_find(unsigned long vpn) -{ - hpte_t *hptep; - unsigned long hash; - unsigned long i, j; - long slot; - unsigned long hpte_v; - - hash = hpt_hash(vpn, 0); - - for (j = 0; j < 2; j++) { - slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; - for (i = 0; i < HPTES_PER_GROUP; i++) { - hptep = htab_address + slot; - hpte_v = hptep->v; - - if ((HPTE_V_AVPN_VAL(hpte_v) == (vpn >> 11)) - && (hpte_v & HPTE_V_VALID) - && ( !!(hpte_v & HPTE_V_SECONDARY) == j)) { - /* HPTE matches */ - if (j) - slot = -slot; - return slot; - } - ++slot; - } - hash = ~hash; - } - - return -1; -} - -static long native_hpte_updatepp(unsigned long slot, unsigned long newpp, - unsigned long va, int large, int local) -{ - hpte_t *hptep = htab_address + slot; - unsigned long hpte_v; - unsigned long avpn = va >> 23; - int ret = 0; - - if (large) - avpn &= ~1; - - native_lock_hpte(hptep); - - hpte_v = hptep->v; - - /* Even if we miss, we need to invalidate the TLB */ - if ((HPTE_V_AVPN_VAL(hpte_v) != avpn) - || !(hpte_v & HPTE_V_VALID)) { - native_unlock_hpte(hptep); - ret = -1; - } else { - set_pp_bit(newpp, hptep); - native_unlock_hpte(hptep); - } - - /* Ensure it is out of the tlb too */ - if (cpu_has_feature(CPU_FTR_TLBIEL) && !large && local) { - tlbiel(va); - } else { - int lock_tlbie = !cpu_has_feature(CPU_FTR_LOCKLESS_TLBIE); - - if (lock_tlbie) - spin_lock(&native_tlbie_lock); - tlbie(va, large); - if (lock_tlbie) - spin_unlock(&native_tlbie_lock); - } - - return ret; -} - -/* - * Update the page protection bits. Intended to be used to create - * guard pages for kernel data structures on pages which are bolted - * in the HPT. Assumes pages being operated on will not be stolen. - * Does not work on large pages. - * - * No need to lock here because we should be the only user. - */ -static void native_hpte_updateboltedpp(unsigned long newpp, unsigned long ea) -{ - unsigned long vsid, va, vpn, flags = 0; - long slot; - hpte_t *hptep; - int lock_tlbie = !cpu_has_feature(CPU_FTR_LOCKLESS_TLBIE); - - vsid = get_kernel_vsid(ea); - va = (vsid << 28) | (ea & 0x0fffffff); - vpn = va >> PAGE_SHIFT; - - slot = native_hpte_find(vpn); - if (slot == -1) - panic("could not find page to bolt\n"); - hptep = htab_address + slot; - - set_pp_bit(newpp, hptep); - - /* Ensure it is out of the tlb too */ - if (lock_tlbie) - spin_lock_irqsave(&native_tlbie_lock, flags); - tlbie(va, 0); - if (lock_tlbie) - spin_unlock_irqrestore(&native_tlbie_lock, flags); -} - -static void native_hpte_invalidate(unsigned long slot, unsigned long va, - int large, int local) -{ - hpte_t *hptep = htab_address + slot; - unsigned long hpte_v; - unsigned long avpn = va >> 23; - unsigned long flags; - int lock_tlbie = !cpu_has_feature(CPU_FTR_LOCKLESS_TLBIE); - - if (large) - avpn &= ~1; - - local_irq_save(flags); - native_lock_hpte(hptep); - - hpte_v = hptep->v; - - /* Even if we miss, we need to invalidate the TLB */ - if ((HPTE_V_AVPN_VAL(hpte_v) != avpn) - || !(hpte_v & HPTE_V_VALID)) { - native_unlock_hpte(hptep); - } else { - /* Invalidate the hpte. NOTE: this also unlocks it */ - hptep->v = 0; - } - - /* Invalidate the tlb */ - if (cpu_has_feature(CPU_FTR_TLBIEL) && !large && local) { - tlbiel(va); - } else { - if (lock_tlbie) - spin_lock(&native_tlbie_lock); - tlbie(va, large); - if (lock_tlbie) - spin_unlock(&native_tlbie_lock); - } - local_irq_restore(flags); -} - -/* - * clear all mappings on kexec. All cpus are in real mode (or they will - * be when they isi), and we are the only one left. We rely on our kernel - * mapping being 0xC0's and the hardware ignoring those two real bits. - * - * TODO: add batching support when enabled. remember, no dynamic memory here, - * athough there is the control page available... - */ -static void native_hpte_clear(void) -{ - unsigned long slot, slots, flags; - hpte_t *hptep = htab_address; - unsigned long hpte_v; - unsigned long pteg_count; - - pteg_count = htab_hash_mask + 1; - - local_irq_save(flags); - - /* we take the tlbie lock and hold it. Some hardware will - * deadlock if we try to tlbie from two processors at once. - */ - spin_lock(&native_tlbie_lock); - - slots = pteg_count * HPTES_PER_GROUP; - - for (slot = 0; slot < slots; slot++, hptep++) { - /* - * we could lock the pte here, but we are the only cpu - * running, right? and for crash dump, we probably - * don't want to wait for a maybe bad cpu. - */ - hpte_v = hptep->v; - - if (hpte_v & HPTE_V_VALID) { - hptep->v = 0; - tlbie(slot2va(hpte_v, slot), hpte_v & HPTE_V_LARGE); - } - } - - spin_unlock(&native_tlbie_lock); - local_irq_restore(flags); -} - -static void native_flush_hash_range(unsigned long context, - unsigned long number, int local) -{ - unsigned long vsid, vpn, va, hash, secondary, slot, flags, avpn; - int i, j; - hpte_t *hptep; - unsigned long hpte_v; - struct ppc64_tlb_batch *batch = &__get_cpu_var(ppc64_tlb_batch); - unsigned long large = batch->large; - - local_irq_save(flags); - - j = 0; - for (i = 0; i < number; i++) { - if (batch->addr[i] < KERNELBASE) - vsid = get_vsid(context, batch->addr[i]); - else - vsid = get_kernel_vsid(batch->addr[i]); - - va = (vsid << 28) | (batch->addr[i] & 0x0fffffff); - batch->vaddr[j] = va; - if (large) - vpn = va >> HPAGE_SHIFT; - else - vpn = va >> PAGE_SHIFT; - hash = hpt_hash(vpn, large); - secondary = (pte_val(batch->pte[i]) & _PAGE_SECONDARY) >> 15; - if (secondary) - hash = ~hash; - slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; - slot += (pte_val(batch->pte[i]) & _PAGE_GROUP_IX) >> 12; - - hptep = htab_address + slot; - - avpn = va >> 23; - if (large) - avpn &= ~0x1UL; - - native_lock_hpte(hptep); - - hpte_v = hptep->v; - - /* Even if we miss, we need to invalidate the TLB */ - if ((HPTE_V_AVPN_VAL(hpte_v) != avpn) - || !(hpte_v & HPTE_V_VALID)) { - native_unlock_hpte(hptep); - } else { - /* Invalidate the hpte. NOTE: this also unlocks it */ - hptep->v = 0; - } - - j++; - } - - if (cpu_has_feature(CPU_FTR_TLBIEL) && !large && local) { - asm volatile("ptesync":::"memory"); - - for (i = 0; i < j; i++) - __tlbiel(batch->vaddr[i]); - - asm volatile("ptesync":::"memory"); - } else { - int lock_tlbie = !cpu_has_feature(CPU_FTR_LOCKLESS_TLBIE); - - if (lock_tlbie) - spin_lock(&native_tlbie_lock); - - asm volatile("ptesync":::"memory"); - - for (i = 0; i < j; i++) - __tlbie(batch->vaddr[i], large); - - asm volatile("eieio; tlbsync; ptesync":::"memory"); - - if (lock_tlbie) - spin_unlock(&native_tlbie_lock); - } - - local_irq_restore(flags); -} - -#ifdef CONFIG_PPC_PSERIES -/* Disable TLB batching on nighthawk */ -static inline int tlb_batching_enabled(void) -{ - struct device_node *root = of_find_node_by_path("/"); - int enabled = 1; - - if (root) { - const char *model = get_property(root, "model", NULL); - if (model && !strcmp(model, "IBM,9076-N81")) - enabled = 0; - of_node_put(root); - } - - return enabled; -} -#else -static inline int tlb_batching_enabled(void) -{ - return 1; -} -#endif - -void hpte_init_native(void) -{ - ppc_md.hpte_invalidate = native_hpte_invalidate; - ppc_md.hpte_updatepp = native_hpte_updatepp; - ppc_md.hpte_updateboltedpp = native_hpte_updateboltedpp; - ppc_md.hpte_insert = native_hpte_insert; - ppc_md.hpte_remove = native_hpte_remove; - ppc_md.hpte_clear_all = native_hpte_clear; - if (tlb_batching_enabled()) - ppc_md.flush_hash_range = native_flush_hash_range; - htab_finish_init(); -} diff --git a/arch/ppc64/mm/hash_utils.c b/arch/ppc64/mm/hash_utils.c deleted file mode 100644 index 09475c8edf7c..000000000000 --- a/arch/ppc64/mm/hash_utils.c +++ /dev/null @@ -1,438 +0,0 @@ -/* - * PowerPC64 port by Mike Corrigan and Dave Engebretsen - * {mikejc|engebret}@us.ibm.com - * - * Copyright (c) 2000 Mike Corrigan <mikejc@us.ibm.com> - * - * SMP scalability work: - * Copyright (C) 2001 Anton Blanchard <anton@au.ibm.com>, IBM - * - * Module name: htab.c - * - * Description: - * PowerPC Hashed Page Table functions - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#undef DEBUG - -#include <linux/config.h> -#include <linux/spinlock.h> -#include <linux/errno.h> -#include <linux/sched.h> -#include <linux/proc_fs.h> -#include <linux/stat.h> -#include <linux/sysctl.h> -#include <linux/ctype.h> -#include <linux/cache.h> -#include <linux/init.h> -#include <linux/signal.h> - -#include <asm/ppcdebug.h> -#include <asm/processor.h> -#include <asm/pgtable.h> -#include <asm/mmu.h> -#include <asm/mmu_context.h> -#include <asm/page.h> -#include <asm/types.h> -#include <asm/system.h> -#include <asm/uaccess.h> -#include <asm/machdep.h> -#include <asm/lmb.h> -#include <asm/abs_addr.h> -#include <asm/tlbflush.h> -#include <asm/io.h> -#include <asm/eeh.h> -#include <asm/tlb.h> -#include <asm/cacheflush.h> -#include <asm/cputable.h> -#include <asm/abs_addr.h> -#include <asm/sections.h> - -#ifdef DEBUG -#define DBG(fmt...) udbg_printf(fmt) -#else -#define DBG(fmt...) -#endif - -/* - * Note: pte --> Linux PTE - * HPTE --> PowerPC Hashed Page Table Entry - * - * Execution context: - * htab_initialize is called with the MMU off (of course), but - * the kernel has been copied down to zero so it can directly - * reference global data. At this point it is very difficult - * to print debug info. - * - */ - -#ifdef CONFIG_U3_DART -extern unsigned long dart_tablebase; -#endif /* CONFIG_U3_DART */ - -hpte_t *htab_address; -unsigned long htab_hash_mask; - -extern unsigned long _SDR1; - -#define KB (1024) -#define MB (1024*KB) - -static inline void loop_forever(void) -{ - volatile unsigned long x = 1; - for(;x;x|=1) - ; -} - -#ifdef CONFIG_PPC_MULTIPLATFORM -static inline void create_pte_mapping(unsigned long start, unsigned long end, - unsigned long mode, int large) -{ - unsigned long addr; - unsigned int step; - unsigned long tmp_mode; - unsigned long vflags; - - if (large) { - step = 16*MB; - vflags = HPTE_V_BOLTED | HPTE_V_LARGE; - } else { - step = 4*KB; - vflags = HPTE_V_BOLTED; - } - - for (addr = start; addr < end; addr += step) { - unsigned long vpn, hash, hpteg; - unsigned long vsid = get_kernel_vsid(addr); - unsigned long va = (vsid << 28) | (addr & 0xfffffff); - int ret; - - if (large) - vpn = va >> HPAGE_SHIFT; - else - vpn = va >> PAGE_SHIFT; - - - tmp_mode = mode; - - /* Make non-kernel text non-executable */ - if (!in_kernel_text(addr)) - tmp_mode = mode | HW_NO_EXEC; - - hash = hpt_hash(vpn, large); - - hpteg = ((hash & htab_hash_mask) * HPTES_PER_GROUP); - -#ifdef CONFIG_PPC_PSERIES - if (systemcfg->platform & PLATFORM_LPAR) - ret = pSeries_lpar_hpte_insert(hpteg, va, - virt_to_abs(addr) >> PAGE_SHIFT, - vflags, tmp_mode); - else -#endif /* CONFIG_PPC_PSERIES */ - ret = native_hpte_insert(hpteg, va, - virt_to_abs(addr) >> PAGE_SHIFT, - vflags, tmp_mode); - - if (ret == -1) { - ppc64_terminate_msg(0x20, "create_pte_mapping"); - loop_forever(); - } - } -} - -void __init htab_initialize(void) -{ - unsigned long table, htab_size_bytes; - unsigned long pteg_count; - unsigned long mode_rw; - int i, use_largepages = 0; - unsigned long base = 0, size = 0; - extern unsigned long tce_alloc_start, tce_alloc_end; - - DBG(" -> htab_initialize()\n"); - - /* - * Calculate the required size of the htab. We want the number of - * PTEGs to equal one half the number of real pages. - */ - htab_size_bytes = 1UL << ppc64_pft_size; - pteg_count = htab_size_bytes >> 7; - - /* For debug, make the HTAB 1/8 as big as it normally would be. */ - ifppcdebug(PPCDBG_HTABSIZE) { - pteg_count >>= 3; - htab_size_bytes = pteg_count << 7; - } - - htab_hash_mask = pteg_count - 1; - - if (systemcfg->platform & PLATFORM_LPAR) { - /* Using a hypervisor which owns the htab */ - htab_address = NULL; - _SDR1 = 0; - } else { - /* Find storage for the HPT. Must be contiguous in - * the absolute address space. - */ - table = lmb_alloc(htab_size_bytes, htab_size_bytes); - - DBG("Hash table allocated at %lx, size: %lx\n", table, - htab_size_bytes); - - if ( !table ) { - ppc64_terminate_msg(0x20, "hpt space"); - loop_forever(); - } - htab_address = abs_to_virt(table); - - /* htab absolute addr + encoded htabsize */ - _SDR1 = table + __ilog2(pteg_count) - 11; - - /* Initialize the HPT with no entries */ - memset((void *)table, 0, htab_size_bytes); - } - - mode_rw = _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_COHERENT | PP_RWXX; - - /* On U3 based machines, we need to reserve the DART area and - * _NOT_ map it to avoid cache paradoxes as it's remapped non - * cacheable later on - */ - if (cpu_has_feature(CPU_FTR_16M_PAGE)) - use_largepages = 1; - - /* create bolted the linear mapping in the hash table */ - for (i=0; i < lmb.memory.cnt; i++) { - base = lmb.memory.region[i].base + KERNELBASE; - size = lmb.memory.region[i].size; - - DBG("creating mapping for region: %lx : %lx\n", base, size); - -#ifdef CONFIG_U3_DART - /* Do not map the DART space. Fortunately, it will be aligned - * in such a way that it will not cross two lmb regions and will - * fit within a single 16Mb page. - * The DART space is assumed to be a full 16Mb region even if we - * only use 2Mb of that space. We will use more of it later for - * AGP GART. We have to use a full 16Mb large page. - */ - DBG("DART base: %lx\n", dart_tablebase); - - if (dart_tablebase != 0 && dart_tablebase >= base - && dart_tablebase < (base + size)) { - if (base != dart_tablebase) - create_pte_mapping(base, dart_tablebase, mode_rw, - use_largepages); - if ((base + size) > (dart_tablebase + 16*MB)) - create_pte_mapping(dart_tablebase + 16*MB, base + size, - mode_rw, use_largepages); - continue; - } -#endif /* CONFIG_U3_DART */ - create_pte_mapping(base, base + size, mode_rw, use_largepages); - } - - /* - * If we have a memory_limit and we've allocated TCEs then we need to - * explicitly map the TCE area at the top of RAM. We also cope with the - * case that the TCEs start below memory_limit. - * tce_alloc_start/end are 16MB aligned so the mapping should work - * for either 4K or 16MB pages. - */ - if (tce_alloc_start) { - tce_alloc_start += KERNELBASE; - tce_alloc_end += KERNELBASE; - - if (base + size >= tce_alloc_start) - tce_alloc_start = base + size + 1; - - create_pte_mapping(tce_alloc_start, tce_alloc_end, - mode_rw, use_largepages); - } - - DBG(" <- htab_initialize()\n"); -} -#undef KB -#undef MB -#endif /* CONFIG_PPC_MULTIPLATFORM */ - -/* - * Called by asm hashtable.S for doing lazy icache flush - */ -unsigned int hash_page_do_lazy_icache(unsigned int pp, pte_t pte, int trap) -{ - struct page *page; - - if (!pfn_valid(pte_pfn(pte))) - return pp; - - page = pte_page(pte); - - /* page is dirty */ - if (!test_bit(PG_arch_1, &page->flags) && !PageReserved(page)) { - if (trap == 0x400) { - __flush_dcache_icache(page_address(page)); - set_bit(PG_arch_1, &page->flags); - } else - pp |= HW_NO_EXEC; - } - return pp; -} - -/* Result code is: - * 0 - handled - * 1 - normal page fault - * -1 - critical hash insertion error - */ -int hash_page(unsigned long ea, unsigned long access, unsigned long trap) -{ - void *pgdir; - unsigned long vsid; - struct mm_struct *mm; - pte_t *ptep; - int ret; - int user_region = 0; - int local = 0; - cpumask_t tmp; - - if ((ea & ~REGION_MASK) >= PGTABLE_RANGE) - return 1; - - switch (REGION_ID(ea)) { - case USER_REGION_ID: - user_region = 1; - mm = current->mm; - if (! mm) - return 1; - - vsid = get_vsid(mm->context.id, ea); - break; - case VMALLOC_REGION_ID: - mm = &init_mm; - vsid = get_kernel_vsid(ea); - break; -#if 0 - case KERNEL_REGION_ID: - /* - * Should never get here - entire 0xC0... region is bolted. - * Send the problem up to do_page_fault - */ -#endif - default: - /* Not a valid range - * Send the problem up to do_page_fault - */ - return 1; - break; - } - - pgdir = mm->pgd; - - if (pgdir == NULL) - return 1; - - tmp = cpumask_of_cpu(smp_processor_id()); - if (user_region && cpus_equal(mm->cpu_vm_mask, tmp)) - local = 1; - - /* Is this a huge page ? */ - if (unlikely(in_hugepage_area(mm->context, ea))) - ret = hash_huge_page(mm, access, ea, vsid, local); - else { - ptep = find_linux_pte(pgdir, ea); - if (ptep == NULL) - return 1; - ret = __hash_page(ea, access, vsid, ptep, trap, local); - } - - return ret; -} - -void flush_hash_page(unsigned long context, unsigned long ea, pte_t pte, - int local) -{ - unsigned long vsid, vpn, va, hash, secondary, slot; - unsigned long huge = pte_huge(pte); - - if (ea < KERNELBASE) - vsid = get_vsid(context, ea); - else - vsid = get_kernel_vsid(ea); - - va = (vsid << 28) | (ea & 0x0fffffff); - if (huge) - vpn = va >> HPAGE_SHIFT; - else - vpn = va >> PAGE_SHIFT; - hash = hpt_hash(vpn, huge); - secondary = (pte_val(pte) & _PAGE_SECONDARY) >> 15; - if (secondary) - hash = ~hash; - slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; - slot += (pte_val(pte) & _PAGE_GROUP_IX) >> 12; - - ppc_md.hpte_invalidate(slot, va, huge, local); -} - -void flush_hash_range(unsigned long context, unsigned long number, int local) -{ - if (ppc_md.flush_hash_range) { - ppc_md.flush_hash_range(context, number, local); - } else { - int i; - struct ppc64_tlb_batch *batch = &__get_cpu_var(ppc64_tlb_batch); - - for (i = 0; i < number; i++) - flush_hash_page(context, batch->addr[i], batch->pte[i], - local); - } -} - -static inline void make_bl(unsigned int *insn_addr, void *func) -{ - unsigned long funcp = *((unsigned long *)func); - int offset = funcp - (unsigned long)insn_addr; - - *insn_addr = (unsigned int)(0x48000001 | (offset & 0x03fffffc)); - flush_icache_range((unsigned long)insn_addr, 4+ - (unsigned long)insn_addr); -} - -/* - * low_hash_fault is called when we the low level hash code failed - * to instert a PTE due to an hypervisor error - */ -void low_hash_fault(struct pt_regs *regs, unsigned long address) -{ - if (user_mode(regs)) { - siginfo_t info; - - info.si_signo = SIGBUS; - info.si_errno = 0; - info.si_code = BUS_ADRERR; - info.si_addr = (void __user *)address; - force_sig_info(SIGBUS, &info, current); - return; - } - bad_page_fault(regs, address, SIGBUS); -} - -void __init htab_finish_init(void) -{ - extern unsigned int *htab_call_hpte_insert1; - extern unsigned int *htab_call_hpte_insert2; - extern unsigned int *htab_call_hpte_remove; - extern unsigned int *htab_call_hpte_updatepp; - - make_bl(htab_call_hpte_insert1, ppc_md.hpte_insert); - make_bl(htab_call_hpte_insert2, ppc_md.hpte_insert); - make_bl(htab_call_hpte_remove, ppc_md.hpte_remove); - make_bl(htab_call_hpte_updatepp, ppc_md.hpte_updatepp); -} diff --git a/arch/ppc64/mm/hugetlbpage.c b/arch/ppc64/mm/hugetlbpage.c deleted file mode 100644 index 0ea0994ed974..000000000000 --- a/arch/ppc64/mm/hugetlbpage.c +++ /dev/null @@ -1,745 +0,0 @@ -/* - * PPC64 (POWER4) Huge TLB Page Support for Kernel. - * - * Copyright (C) 2003 David Gibson, IBM Corporation. - * - * Based on the IA-32 version: - * Copyright (C) 2002, Rohit Seth <rohit.seth@intel.com> - */ - -#include <linux/init.h> -#include <linux/fs.h> -#include <linux/mm.h> -#include <linux/hugetlb.h> -#include <linux/pagemap.h> -#include <linux/smp_lock.h> -#include <linux/slab.h> -#include <linux/err.h> -#include <linux/sysctl.h> -#include <asm/mman.h> -#include <asm/pgalloc.h> -#include <asm/tlb.h> -#include <asm/tlbflush.h> -#include <asm/mmu_context.h> -#include <asm/machdep.h> -#include <asm/cputable.h> -#include <asm/tlb.h> - -#include <linux/sysctl.h> - -#define NUM_LOW_AREAS (0x100000000UL >> SID_SHIFT) -#define NUM_HIGH_AREAS (PGTABLE_RANGE >> HTLB_AREA_SHIFT) - -/* Modelled after find_linux_pte() */ -pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) -{ - pgd_t *pg; - pud_t *pu; - pmd_t *pm; - pte_t *pt; - - BUG_ON(! in_hugepage_area(mm->context, addr)); - - addr &= HPAGE_MASK; - - pg = pgd_offset(mm, addr); - if (!pgd_none(*pg)) { - pu = pud_offset(pg, addr); - if (!pud_none(*pu)) { - pm = pmd_offset(pu, addr); - pt = (pte_t *)pm; - BUG_ON(!pmd_none(*pm) - && !(pte_present(*pt) && pte_huge(*pt))); - return pt; - } - } - - return NULL; -} - -pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) -{ - pgd_t *pg; - pud_t *pu; - pmd_t *pm; - pte_t *pt; - - BUG_ON(! in_hugepage_area(mm->context, addr)); - - addr &= HPAGE_MASK; - - pg = pgd_offset(mm, addr); - pu = pud_alloc(mm, pg, addr); - - if (pu) { - pm = pmd_alloc(mm, pu, addr); - if (pm) { - pt = (pte_t *)pm; - BUG_ON(!pmd_none(*pm) - && !(pte_present(*pt) && pte_huge(*pt))); - return pt; - } - } - - return NULL; -} - -#define HUGEPTE_BATCH_SIZE (HPAGE_SIZE / PMD_SIZE) - -void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, - pte_t *ptep, pte_t pte) -{ - int i; - - if (pte_present(*ptep)) { - pte_clear(mm, addr, ptep); - flush_tlb_pending(); - } - - for (i = 0; i < HUGEPTE_BATCH_SIZE; i++) { - *ptep = __pte(pte_val(pte) & ~_PAGE_HPTEFLAGS); - ptep++; - } -} - -pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, - pte_t *ptep) -{ - unsigned long old = pte_update(ptep, ~0UL); - int i; - - if (old & _PAGE_HASHPTE) - hpte_update(mm, addr, old, 0); - - for (i = 1; i < HUGEPTE_BATCH_SIZE; i++) - ptep[i] = __pte(0); - - return __pte(old); -} - -/* - * This function checks for proper alignment of input addr and len parameters. - */ -int is_aligned_hugepage_range(unsigned long addr, unsigned long len) -{ - if (len & ~HPAGE_MASK) - return -EINVAL; - if (addr & ~HPAGE_MASK) - return -EINVAL; - if (! (within_hugepage_low_range(addr, len) - || within_hugepage_high_range(addr, len)) ) - return -EINVAL; - return 0; -} - -static void flush_low_segments(void *parm) -{ - u16 areas = (unsigned long) parm; - unsigned long i; - - asm volatile("isync" : : : "memory"); - - BUILD_BUG_ON((sizeof(areas)*8) != NUM_LOW_AREAS); - - for (i = 0; i < NUM_LOW_AREAS; i++) { - if (! (areas & (1U << i))) - continue; - asm volatile("slbie %0" - : : "r" ((i << SID_SHIFT) | SLBIE_C)); - } - - asm volatile("isync" : : : "memory"); -} - -static void flush_high_segments(void *parm) -{ - u16 areas = (unsigned long) parm; - unsigned long i, j; - - asm volatile("isync" : : : "memory"); - - BUILD_BUG_ON((sizeof(areas)*8) != NUM_HIGH_AREAS); - - for (i = 0; i < NUM_HIGH_AREAS; i++) { - if (! (areas & (1U << i))) - continue; - for (j = 0; j < (1UL << (HTLB_AREA_SHIFT-SID_SHIFT)); j++) - asm volatile("slbie %0" - :: "r" (((i << HTLB_AREA_SHIFT) - + (j << SID_SHIFT)) | SLBIE_C)); - } - - asm volatile("isync" : : : "memory"); -} - -static int prepare_low_area_for_htlb(struct mm_struct *mm, unsigned long area) -{ - unsigned long start = area << SID_SHIFT; - unsigned long end = (area+1) << SID_SHIFT; - struct vm_area_struct *vma; - - BUG_ON(area >= NUM_LOW_AREAS); - - /* Check no VMAs are in the region */ - vma = find_vma(mm, start); - if (vma && (vma->vm_start < end)) - return -EBUSY; - - return 0; -} - -static int prepare_high_area_for_htlb(struct mm_struct *mm, unsigned long area) -{ - unsigned long start = area << HTLB_AREA_SHIFT; - unsigned long end = (area+1) << HTLB_AREA_SHIFT; - struct vm_area_struct *vma; - - BUG_ON(area >= NUM_HIGH_AREAS); - - /* Check no VMAs are in the region */ - vma = find_vma(mm, start); - if (vma && (vma->vm_start < end)) - return -EBUSY; - - return 0; -} - -static int open_low_hpage_areas(struct mm_struct *mm, u16 newareas) -{ - unsigned long i; - - BUILD_BUG_ON((sizeof(newareas)*8) != NUM_LOW_AREAS); - BUILD_BUG_ON((sizeof(mm->context.low_htlb_areas)*8) != NUM_LOW_AREAS); - - newareas &= ~(mm->context.low_htlb_areas); - if (! newareas) - return 0; /* The segments we want are already open */ - - for (i = 0; i < NUM_LOW_AREAS; i++) - if ((1 << i) & newareas) - if (prepare_low_area_for_htlb(mm, i) != 0) - return -EBUSY; - - mm->context.low_htlb_areas |= newareas; - - /* update the paca copy of the context struct */ - get_paca()->context = mm->context; - - /* the context change must make it to memory before the flush, - * so that further SLB misses do the right thing. */ - mb(); - on_each_cpu(flush_low_segments, (void *)(unsigned long)newareas, 0, 1); - - return 0; -} - -static int open_high_hpage_areas(struct mm_struct *mm, u16 newareas) -{ - unsigned long i; - - BUILD_BUG_ON((sizeof(newareas)*8) != NUM_HIGH_AREAS); - BUILD_BUG_ON((sizeof(mm->context.high_htlb_areas)*8) - != NUM_HIGH_AREAS); - - newareas &= ~(mm->context.high_htlb_areas); - if (! newareas) - return 0; /* The areas we want are already open */ - - for (i = 0; i < NUM_HIGH_AREAS; i++) - if ((1 << i) & newareas) - if (prepare_high_area_for_htlb(mm, i) != 0) - return -EBUSY; - - mm->context.high_htlb_areas |= newareas; - - /* update the paca copy of the context struct */ - get_paca()->context = mm->context; - - /* the context change must make it to memory before the flush, - * so that further SLB misses do the right thing. */ - mb(); - on_each_cpu(flush_high_segments, (void *)(unsigned long)newareas, 0, 1); - - return 0; -} - -int prepare_hugepage_range(unsigned long addr, unsigned long len) -{ - int err; - - if ( (addr+len) < addr ) - return -EINVAL; - - if ((addr + len) < 0x100000000UL) - err = open_low_hpage_areas(current->mm, - LOW_ESID_MASK(addr, len)); - else - err = open_high_hpage_areas(current->mm, - HTLB_AREA_MASK(addr, len)); - if (err) { - printk(KERN_DEBUG "prepare_hugepage_range(%lx, %lx)" - " failed (lowmask: 0x%04hx, highmask: 0x%04hx)\n", - addr, len, - LOW_ESID_MASK(addr, len), HTLB_AREA_MASK(addr, len)); - return err; - } - - return 0; -} - -struct page * -follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) -{ - pte_t *ptep; - struct page *page; - - if (! in_hugepage_area(mm->context, address)) - return ERR_PTR(-EINVAL); - - ptep = huge_pte_offset(mm, address); - page = pte_page(*ptep); - if (page) - page += (address % HPAGE_SIZE) / PAGE_SIZE; - - return page; -} - -int pmd_huge(pmd_t pmd) -{ - return 0; -} - -struct page * -follow_huge_pmd(struct mm_struct *mm, unsigned long address, - pmd_t *pmd, int write) -{ - BUG(); - return NULL; -} - -/* Because we have an exclusive hugepage region which lies within the - * normal user address space, we have to take special measures to make - * non-huge mmap()s evade the hugepage reserved regions. */ -unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, - unsigned long len, unsigned long pgoff, - unsigned long flags) -{ - struct mm_struct *mm = current->mm; - struct vm_area_struct *vma; - unsigned long start_addr; - - if (len > TASK_SIZE) - return -ENOMEM; - - if (addr) { - addr = PAGE_ALIGN(addr); - vma = find_vma(mm, addr); - if (((TASK_SIZE - len) >= addr) - && (!vma || (addr+len) <= vma->vm_start) - && !is_hugepage_only_range(mm, addr,len)) - return addr; - } - if (len > mm->cached_hole_size) { - start_addr = addr = mm->free_area_cache; - } else { - start_addr = addr = TASK_UNMAPPED_BASE; - mm->cached_hole_size = 0; - } - -full_search: - vma = find_vma(mm, addr); - while (TASK_SIZE - len >= addr) { - BUG_ON(vma && (addr >= vma->vm_end)); - - if (touches_hugepage_low_range(mm, addr, len)) { - addr = ALIGN(addr+1, 1<<SID_SHIFT); - vma = find_vma(mm, addr); - continue; - } - if (touches_hugepage_high_range(mm, addr, len)) { - addr = ALIGN(addr+1, 1UL<<HTLB_AREA_SHIFT); - vma = find_vma(mm, addr); - continue; - } - if (!vma || addr + len <= vma->vm_start) { - /* - * Remember the place where we stopped the search: - */ - mm->free_area_cache = addr + len; - return addr; - } - if (addr + mm->cached_hole_size < vma->vm_start) - mm->cached_hole_size = vma->vm_start - addr; - addr = vma->vm_end; - vma = vma->vm_next; - } - - /* Make sure we didn't miss any holes */ - if (start_addr != TASK_UNMAPPED_BASE) { - start_addr = addr = TASK_UNMAPPED_BASE; - mm->cached_hole_size = 0; - goto full_search; - } - return -ENOMEM; -} - -/* - * This mmap-allocator allocates new areas top-down from below the - * stack's low limit (the base): - * - * Because we have an exclusive hugepage region which lies within the - * normal user address space, we have to take special measures to make - * non-huge mmap()s evade the hugepage reserved regions. - */ -unsigned long -arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, - const unsigned long len, const unsigned long pgoff, - const unsigned long flags) -{ - struct vm_area_struct *vma, *prev_vma; - struct mm_struct *mm = current->mm; - unsigned long base = mm->mmap_base, addr = addr0; - unsigned long largest_hole = mm->cached_hole_size; - int first_time = 1; - - /* requested length too big for entire address space */ - if (len > TASK_SIZE) - return -ENOMEM; - - /* dont allow allocations above current base */ - if (mm->free_area_cache > base) - mm->free_area_cache = base; - - /* requesting a specific address */ - if (addr) { - addr = PAGE_ALIGN(addr); - vma = find_vma(mm, addr); - if (TASK_SIZE - len >= addr && - (!vma || addr + len <= vma->vm_start) - && !is_hugepage_only_range(mm, addr,len)) - return addr; - } - - if (len <= largest_hole) { - largest_hole = 0; - mm->free_area_cache = base; - } -try_again: - /* make sure it can fit in the remaining address space */ - if (mm->free_area_cache < len) - goto fail; - - /* either no address requested or cant fit in requested address hole */ - addr = (mm->free_area_cache - len) & PAGE_MASK; - do { -hugepage_recheck: - if (touches_hugepage_low_range(mm, addr, len)) { - addr = (addr & ((~0) << SID_SHIFT)) - len; - goto hugepage_recheck; - } else if (touches_hugepage_high_range(mm, addr, len)) { - addr = (addr & ((~0UL) << HTLB_AREA_SHIFT)) - len; - goto hugepage_recheck; - } - - /* - * Lookup failure means no vma is above this address, - * i.e. return with success: - */ - if (!(vma = find_vma_prev(mm, addr, &prev_vma))) - return addr; - - /* - * new region fits between prev_vma->vm_end and - * vma->vm_start, use it: - */ - if (addr+len <= vma->vm_start && - (!prev_vma || (addr >= prev_vma->vm_end))) { - /* remember the address as a hint for next time */ - mm->cached_hole_size = largest_hole; - return (mm->free_area_cache = addr); - } else { - /* pull free_area_cache down to the first hole */ - if (mm->free_area_cache == vma->vm_end) { - mm->free_area_cache = vma->vm_start; - mm->cached_hole_size = largest_hole; - } - } - - /* remember the largest hole we saw so far */ - if (addr + largest_hole < vma->vm_start) - largest_hole = vma->vm_start - addr; - - /* try just below the current vma->vm_start */ - addr = vma->vm_start-len; - } while (len <= vma->vm_start); - -fail: - /* - * if hint left us with no space for the requested - * mapping then try again: - */ - if (first_time) { - mm->free_area_cache = base; - largest_hole = 0; - first_time = 0; - goto try_again; - } - /* - * A failed mmap() very likely causes application failure, - * so fall back to the bottom-up function here. This scenario - * can happen with large stack limits and large mmap() - * allocations. - */ - mm->free_area_cache = TASK_UNMAPPED_BASE; - mm->cached_hole_size = ~0UL; - addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags); - /* - * Restore the topdown base: - */ - mm->free_area_cache = base; - mm->cached_hole_size = ~0UL; - - return addr; -} - -static unsigned long htlb_get_low_area(unsigned long len, u16 segmask) -{ - unsigned long addr = 0; - struct vm_area_struct *vma; - - vma = find_vma(current->mm, addr); - while (addr + len <= 0x100000000UL) { - BUG_ON(vma && (addr >= vma->vm_end)); /* invariant */ - - if (! __within_hugepage_low_range(addr, len, segmask)) { - addr = ALIGN(addr+1, 1<<SID_SHIFT); - vma = find_vma(current->mm, addr); - continue; - } - - if (!vma || (addr + len) <= vma->vm_start) - return addr; - addr = ALIGN(vma->vm_end, HPAGE_SIZE); - /* Depending on segmask this might not be a confirmed - * hugepage region, so the ALIGN could have skipped - * some VMAs */ - vma = find_vma(current->mm, addr); - } - - return -ENOMEM; -} - -static unsigned long htlb_get_high_area(unsigned long len, u16 areamask) -{ - unsigned long addr = 0x100000000UL; - struct vm_area_struct *vma; - - vma = find_vma(current->mm, addr); - while (addr + len <= TASK_SIZE_USER64) { - BUG_ON(vma && (addr >= vma->vm_end)); /* invariant */ - - if (! __within_hugepage_high_range(addr, len, areamask)) { - addr = ALIGN(addr+1, 1UL<<HTLB_AREA_SHIFT); - vma = find_vma(current->mm, addr); - continue; - } - - if (!vma || (addr + len) <= vma->vm_start) - return addr; - addr = ALIGN(vma->vm_end, HPAGE_SIZE); - /* Depending on segmask this might not be a confirmed - * hugepage region, so the ALIGN could have skipped - * some VMAs */ - vma = find_vma(current->mm, addr); - } - - return -ENOMEM; -} - -unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr, - unsigned long len, unsigned long pgoff, - unsigned long flags) -{ - int lastshift; - u16 areamask, curareas; - - if (len & ~HPAGE_MASK) - return -EINVAL; - - if (!cpu_has_feature(CPU_FTR_16M_PAGE)) - return -EINVAL; - - if (test_thread_flag(TIF_32BIT)) { - curareas = current->mm->context.low_htlb_areas; - - /* First see if we can do the mapping in the existing - * low areas */ - addr = htlb_get_low_area(len, curareas); - if (addr != -ENOMEM) - return addr; - - lastshift = 0; - for (areamask = LOW_ESID_MASK(0x100000000UL-len, len); - ! lastshift; areamask >>=1) { - if (areamask & 1) - lastshift = 1; - - addr = htlb_get_low_area(len, curareas | areamask); - if ((addr != -ENOMEM) - && open_low_hpage_areas(current->mm, areamask) == 0) - return addr; - } - } else { - curareas = current->mm->context.high_htlb_areas; - - /* First see if we can do the mapping in the existing - * high areas */ - addr = htlb_get_high_area(len, curareas); - if (addr != -ENOMEM) - return addr; - - lastshift = 0; - for (areamask = HTLB_AREA_MASK(TASK_SIZE_USER64-len, len); - ! lastshift; areamask >>=1) { - if (areamask & 1) - lastshift = 1; - - addr = htlb_get_high_area(len, curareas | areamask); - if ((addr != -ENOMEM) - && open_high_hpage_areas(current->mm, areamask) == 0) - return addr; - } - } - printk(KERN_DEBUG "hugetlb_get_unmapped_area() unable to open" - " enough areas\n"); - return -ENOMEM; -} - -int hash_huge_page(struct mm_struct *mm, unsigned long access, - unsigned long ea, unsigned long vsid, int local) -{ - pte_t *ptep; - unsigned long va, vpn; - pte_t old_pte, new_pte; - unsigned long rflags, prpn; - long slot; - int err = 1; - - spin_lock(&mm->page_table_lock); - - ptep = huge_pte_offset(mm, ea); - - /* Search the Linux page table for a match with va */ - va = (vsid << 28) | (ea & 0x0fffffff); - vpn = va >> HPAGE_SHIFT; - - /* - * If no pte found or not present, send the problem up to - * do_page_fault - */ - if (unlikely(!ptep || pte_none(*ptep))) - goto out; - -/* BUG_ON(pte_bad(*ptep)); */ - - /* - * Check the user's access rights to the page. If access should be - * prevented then send the problem up to do_page_fault. - */ - if (unlikely(access & ~pte_val(*ptep))) - goto out; - /* - * At this point, we have a pte (old_pte) which can be used to build - * or update an HPTE. There are 2 cases: - * - * 1. There is a valid (present) pte with no associated HPTE (this is - * the most common case) - * 2. There is a valid (present) pte with an associated HPTE. The - * current values of the pp bits in the HPTE prevent access - * because we are doing software DIRTY bit management and the - * page is currently not DIRTY. - */ - - - old_pte = *ptep; - new_pte = old_pte; - - rflags = 0x2 | (! (pte_val(new_pte) & _PAGE_RW)); - /* _PAGE_EXEC -> HW_NO_EXEC since it's inverted */ - rflags |= ((pte_val(new_pte) & _PAGE_EXEC) ? 0 : HW_NO_EXEC); - - /* Check if pte already has an hpte (case 2) */ - if (unlikely(pte_val(old_pte) & _PAGE_HASHPTE)) { - /* There MIGHT be an HPTE for this pte */ - unsigned long hash, slot; - - hash = hpt_hash(vpn, 1); - if (pte_val(old_pte) & _PAGE_SECONDARY) - hash = ~hash; - slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; - slot += (pte_val(old_pte) & _PAGE_GROUP_IX) >> 12; - - if (ppc_md.hpte_updatepp(slot, rflags, va, 1, local) == -1) - pte_val(old_pte) &= ~_PAGE_HPTEFLAGS; - } - - if (likely(!(pte_val(old_pte) & _PAGE_HASHPTE))) { - unsigned long hash = hpt_hash(vpn, 1); - unsigned long hpte_group; - - prpn = pte_pfn(old_pte); - -repeat: - hpte_group = ((hash & htab_hash_mask) * - HPTES_PER_GROUP) & ~0x7UL; - - /* Update the linux pte with the HPTE slot */ - pte_val(new_pte) &= ~_PAGE_HPTEFLAGS; - pte_val(new_pte) |= _PAGE_HASHPTE; - - /* Add in WIMG bits */ - /* XXX We should store these in the pte */ - rflags |= _PAGE_COHERENT; - - slot = ppc_md.hpte_insert(hpte_group, va, prpn, - HPTE_V_LARGE, rflags); - - /* Primary is full, try the secondary */ - if (unlikely(slot == -1)) { - pte_val(new_pte) |= _PAGE_SECONDARY; - hpte_group = ((~hash & htab_hash_mask) * - HPTES_PER_GROUP) & ~0x7UL; - slot = ppc_md.hpte_insert(hpte_group, va, prpn, - HPTE_V_LARGE | - HPTE_V_SECONDARY, - rflags); - if (slot == -1) { - if (mftb() & 0x1) - hpte_group = ((hash & htab_hash_mask) * - HPTES_PER_GROUP)&~0x7UL; - - ppc_md.hpte_remove(hpte_group); - goto repeat; - } - } - - if (unlikely(slot == -2)) - panic("hash_huge_page: pte_insert failed\n"); - - pte_val(new_pte) |= (slot<<12) & _PAGE_GROUP_IX; - - /* - * No need to use ldarx/stdcx here because all who - * might be updating the pte will hold the - * page_table_lock - */ - *ptep = new_pte; - } - - err = 0; - - out: - spin_unlock(&mm->page_table_lock); - - return err; -} diff --git a/arch/ppc64/mm/imalloc.c b/arch/ppc64/mm/imalloc.c deleted file mode 100644 index c65b87b92756..000000000000 --- a/arch/ppc64/mm/imalloc.c +++ /dev/null @@ -1,317 +0,0 @@ -/* - * c 2001 PPC 64 Team, IBM Corp - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/slab.h> -#include <linux/vmalloc.h> - -#include <asm/uaccess.h> -#include <asm/pgalloc.h> -#include <asm/pgtable.h> -#include <asm/semaphore.h> -#include <asm/imalloc.h> -#include <asm/cacheflush.h> - -static DECLARE_MUTEX(imlist_sem); -struct vm_struct * imlist = NULL; - -static int get_free_im_addr(unsigned long size, unsigned long *im_addr) -{ - unsigned long addr; - struct vm_struct **p, *tmp; - - addr = ioremap_bot; - for (p = &imlist; (tmp = *p) ; p = &tmp->next) { - if (size + addr < (unsigned long) tmp->addr) - break; - if ((unsigned long)tmp->addr >= ioremap_bot) - addr = tmp->size + (unsigned long) tmp->addr; - if (addr >= IMALLOC_END-size) - return 1; - } - *im_addr = addr; - - return 0; -} - -/* Return whether the region described by v_addr and size is a subset - * of the region described by parent - */ -static inline int im_region_is_subset(unsigned long v_addr, unsigned long size, - struct vm_struct *parent) -{ - return (int) (v_addr >= (unsigned long) parent->addr && - v_addr < (unsigned long) parent->addr + parent->size && - size < parent->size); -} - -/* Return whether the region described by v_addr and size is a superset - * of the region described by child - */ -static int im_region_is_superset(unsigned long v_addr, unsigned long size, - struct vm_struct *child) -{ - struct vm_struct parent; - - parent.addr = (void *) v_addr; - parent.size = size; - - return im_region_is_subset((unsigned long) child->addr, child->size, - &parent); -} - -/* Return whether the region described by v_addr and size overlaps - * the region described by vm. Overlapping regions meet the - * following conditions: - * 1) The regions share some part of the address space - * 2) The regions aren't identical - * 3) Neither region is a subset of the other - */ -static int im_region_overlaps(unsigned long v_addr, unsigned long size, - struct vm_struct *vm) -{ - if (im_region_is_superset(v_addr, size, vm)) - return 0; - - return (v_addr + size > (unsigned long) vm->addr + vm->size && - v_addr < (unsigned long) vm->addr + vm->size) || - (v_addr < (unsigned long) vm->addr && - v_addr + size > (unsigned long) vm->addr); -} - -/* Determine imalloc status of region described by v_addr and size. - * Can return one of the following: - * IM_REGION_UNUSED - Entire region is unallocated in imalloc space. - * IM_REGION_SUBSET - Region is a subset of a region that is already - * allocated in imalloc space. - * vm will be assigned to a ptr to the parent region. - * IM_REGION_EXISTS - Exact region already allocated in imalloc space. - * vm will be assigned to a ptr to the existing imlist - * member. - * IM_REGION_OVERLAPS - Region overlaps an allocated region in imalloc space. - * IM_REGION_SUPERSET - Region is a superset of a region that is already - * allocated in imalloc space. - */ -static int im_region_status(unsigned long v_addr, unsigned long size, - struct vm_struct **vm) -{ - struct vm_struct *tmp; - - for (tmp = imlist; tmp; tmp = tmp->next) - if (v_addr < (unsigned long) tmp->addr + tmp->size) - break; - - if (tmp) { - if (im_region_overlaps(v_addr, size, tmp)) - return IM_REGION_OVERLAP; - - *vm = tmp; - if (im_region_is_subset(v_addr, size, tmp)) { - /* Return with tmp pointing to superset */ - return IM_REGION_SUBSET; - } - if (im_region_is_superset(v_addr, size, tmp)) { - /* Return with tmp pointing to first subset */ - return IM_REGION_SUPERSET; - } - else if (v_addr == (unsigned long) tmp->addr && - size == tmp->size) { - /* Return with tmp pointing to exact region */ - return IM_REGION_EXISTS; - } - } - - *vm = NULL; - return IM_REGION_UNUSED; -} - -static struct vm_struct * split_im_region(unsigned long v_addr, - unsigned long size, struct vm_struct *parent) -{ - struct vm_struct *vm1 = NULL; - struct vm_struct *vm2 = NULL; - struct vm_struct *new_vm = NULL; - - vm1 = (struct vm_struct *) kmalloc(sizeof(*vm1), GFP_KERNEL); - if (vm1 == NULL) { - printk(KERN_ERR "%s() out of memory\n", __FUNCTION__); - return NULL; - } - - if (v_addr == (unsigned long) parent->addr) { - /* Use existing parent vm_struct to represent child, allocate - * new one for the remainder of parent range - */ - vm1->size = parent->size - size; - vm1->addr = (void *) (v_addr + size); - vm1->next = parent->next; - - parent->size = size; - parent->next = vm1; - new_vm = parent; - } else if (v_addr + size == (unsigned long) parent->addr + - parent->size) { - /* Allocate new vm_struct to represent child, use existing - * parent one for remainder of parent range - */ - vm1->size = size; - vm1->addr = (void *) v_addr; - vm1->next = parent->next; - new_vm = vm1; - - parent->size -= size; - parent->next = vm1; - } else { - /* Allocate two new vm_structs for the new child and - * uppermost remainder, and use existing parent one for the - * lower remainder of parent range - */ - vm2 = (struct vm_struct *) kmalloc(sizeof(*vm2), GFP_KERNEL); - if (vm2 == NULL) { - printk(KERN_ERR "%s() out of memory\n", __FUNCTION__); - kfree(vm1); - return NULL; - } - - vm1->size = size; - vm1->addr = (void *) v_addr; - vm1->next = vm2; - new_vm = vm1; - - vm2->size = ((unsigned long) parent->addr + parent->size) - - (v_addr + size); - vm2->addr = (void *) v_addr + size; - vm2->next = parent->next; - - parent->size = v_addr - (unsigned long) parent->addr; - parent->next = vm1; - } - - return new_vm; -} - -static struct vm_struct * __add_new_im_area(unsigned long req_addr, - unsigned long size) -{ - struct vm_struct **p, *tmp, *area; - - for (p = &imlist; (tmp = *p) ; p = &tmp->next) { - if (req_addr + size <= (unsigned long)tmp->addr) - break; - } - - area = (struct vm_struct *) kmalloc(sizeof(*area), GFP_KERNEL); - if (!area) - return NULL; - area->flags = 0; - area->addr = (void *)req_addr; - area->size = size; - area->next = *p; - *p = area; - - return area; -} - -static struct vm_struct * __im_get_area(unsigned long req_addr, - unsigned long size, - int criteria) -{ - struct vm_struct *tmp; - int status; - - status = im_region_status(req_addr, size, &tmp); - if ((criteria & status) == 0) { - return NULL; - } - - switch (status) { - case IM_REGION_UNUSED: - tmp = __add_new_im_area(req_addr, size); - break; - case IM_REGION_SUBSET: - tmp = split_im_region(req_addr, size, tmp); - break; - case IM_REGION_EXISTS: - /* Return requested region */ - break; - case IM_REGION_SUPERSET: - /* Return first existing subset of requested region */ - break; - default: - printk(KERN_ERR "%s() unexpected imalloc region status\n", - __FUNCTION__); - tmp = NULL; - } - - return tmp; -} - -struct vm_struct * im_get_free_area(unsigned long size) -{ - struct vm_struct *area; - unsigned long addr; - - down(&imlist_sem); - if (get_free_im_addr(size, &addr)) { - printk(KERN_ERR "%s() cannot obtain addr for size 0x%lx\n", - __FUNCTION__, size); - area = NULL; - goto next_im_done; - } - - area = __im_get_area(addr, size, IM_REGION_UNUSED); - if (area == NULL) { - printk(KERN_ERR - "%s() cannot obtain area for addr 0x%lx size 0x%lx\n", - __FUNCTION__, addr, size); - } -next_im_done: - up(&imlist_sem); - return area; -} - -struct vm_struct * im_get_area(unsigned long v_addr, unsigned long size, - int criteria) -{ - struct vm_struct *area; - - down(&imlist_sem); - area = __im_get_area(v_addr, size, criteria); - up(&imlist_sem); - return area; -} - -void im_free(void * addr) -{ - struct vm_struct **p, *tmp; - - if (!addr) - return; - if ((unsigned long) addr & ~PAGE_MASK) { - printk(KERN_ERR "Trying to %s bad address (%p)\n", __FUNCTION__, addr); - return; - } - down(&imlist_sem); - for (p = &imlist ; (tmp = *p) ; p = &tmp->next) { - if (tmp->addr == addr) { - *p = tmp->next; - - /* XXX: do we need the lock? */ - spin_lock(&init_mm.page_table_lock); - unmap_vm_area(tmp); - spin_unlock(&init_mm.page_table_lock); - - kfree(tmp); - up(&imlist_sem); - return; - } - } - up(&imlist_sem); - printk(KERN_ERR "Trying to %s nonexistent area (%p)\n", __FUNCTION__, - addr); -} diff --git a/arch/ppc64/mm/init.c b/arch/ppc64/mm/init.c deleted file mode 100644 index be64b157afce..000000000000 --- a/arch/ppc64/mm/init.c +++ /dev/null @@ -1,869 +0,0 @@ -/* - * PowerPC version - * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org) - * - * Modifications by Paul Mackerras (PowerMac) (paulus@cs.anu.edu.au) - * and Cort Dougan (PReP) (cort@cs.nmt.edu) - * Copyright (C) 1996 Paul Mackerras - * Amiga/APUS changes by Jesper Skov (jskov@cygnus.co.uk). - * - * Derived from "arch/i386/mm/init.c" - * Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds - * - * Dave Engebretsen <engebret@us.ibm.com> - * Rework for PPC64 port. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - */ - -#include <linux/config.h> -#include <linux/signal.h> -#include <linux/sched.h> -#include <linux/kernel.h> -#include <linux/errno.h> -#include <linux/string.h> -#include <linux/types.h> -#include <linux/mman.h> -#include <linux/mm.h> -#include <linux/swap.h> -#include <linux/stddef.h> -#include <linux/vmalloc.h> -#include <linux/init.h> -#include <linux/delay.h> -#include <linux/bootmem.h> -#include <linux/highmem.h> -#include <linux/idr.h> -#include <linux/nodemask.h> -#include <linux/module.h> - -#include <asm/pgalloc.h> -#include <asm/page.h> -#include <asm/prom.h> -#include <asm/lmb.h> -#include <asm/rtas.h> -#include <asm/io.h> -#include <asm/mmu_context.h> -#include <asm/pgtable.h> -#include <asm/mmu.h> -#include <asm/uaccess.h> -#include <asm/smp.h> -#include <asm/machdep.h> -#include <asm/tlb.h> -#include <asm/eeh.h> -#include <asm/processor.h> -#include <asm/mmzone.h> -#include <asm/cputable.h> -#include <asm/ppcdebug.h> -#include <asm/sections.h> -#include <asm/system.h> -#include <asm/iommu.h> -#include <asm/abs_addr.h> -#include <asm/vdso.h> -#include <asm/imalloc.h> - -#if PGTABLE_RANGE > USER_VSID_RANGE -#warning Limited user VSID range means pagetable space is wasted -#endif - -#if (TASK_SIZE_USER64 < PGTABLE_RANGE) && (TASK_SIZE_USER64 < USER_VSID_RANGE) -#warning TASK_SIZE is smaller than it needs to be. -#endif - -int mem_init_done; -unsigned long ioremap_bot = IMALLOC_BASE; -static unsigned long phbs_io_bot = PHBS_IO_BASE; - -extern pgd_t swapper_pg_dir[]; -extern struct task_struct *current_set[NR_CPUS]; - -unsigned long klimit = (unsigned long)_end; - -unsigned long _SDR1=0; -unsigned long _ASR=0; - -/* max amount of RAM to use */ -unsigned long __max_memory; - -/* info on what we think the IO hole is */ -unsigned long io_hole_start; -unsigned long io_hole_size; - -void show_mem(void) -{ - unsigned long total = 0, reserved = 0; - unsigned long shared = 0, cached = 0; - struct page *page; - pg_data_t *pgdat; - unsigned long i; - - printk("Mem-info:\n"); - show_free_areas(); - printk("Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); - for_each_pgdat(pgdat) { - for (i = 0; i < pgdat->node_spanned_pages; i++) { - page = pgdat_page_nr(pgdat, i); - total++; - if (PageReserved(page)) - reserved++; - else if (PageSwapCache(page)) - cached++; - else if (page_count(page)) - shared += page_count(page) - 1; - } - } - printk("%ld pages of RAM\n", total); - printk("%ld reserved pages\n", reserved); - printk("%ld pages shared\n", shared); - printk("%ld pages swap cached\n", cached); -} - -#ifdef CONFIG_PPC_ISERIES - -void __iomem *ioremap(unsigned long addr, unsigned long size) -{ - return (void __iomem *)addr; -} - -extern void __iomem *__ioremap(unsigned long addr, unsigned long size, - unsigned long flags) -{ - return (void __iomem *)addr; -} - -void iounmap(volatile void __iomem *addr) -{ - return; -} - -#else - -/* - * map_io_page currently only called by __ioremap - * map_io_page adds an entry to the ioremap page table - * and adds an entry to the HPT, possibly bolting it - */ -static int map_io_page(unsigned long ea, unsigned long pa, int flags) -{ - pgd_t *pgdp; - pud_t *pudp; - pmd_t *pmdp; - pte_t *ptep; - unsigned long vsid; - - if (mem_init_done) { - spin_lock(&init_mm.page_table_lock); - pgdp = pgd_offset_k(ea); - pudp = pud_alloc(&init_mm, pgdp, ea); - if (!pudp) - return -ENOMEM; - pmdp = pmd_alloc(&init_mm, pudp, ea); - if (!pmdp) - return -ENOMEM; - ptep = pte_alloc_kernel(&init_mm, pmdp, ea); - if (!ptep) - return -ENOMEM; - set_pte_at(&init_mm, ea, ptep, pfn_pte(pa >> PAGE_SHIFT, - __pgprot(flags))); - spin_unlock(&init_mm.page_table_lock); - } else { - unsigned long va, vpn, hash, hpteg; - - /* - * If the mm subsystem is not fully up, we cannot create a - * linux page table entry for this mapping. Simply bolt an - * entry in the hardware page table. - */ - vsid = get_kernel_vsid(ea); - va = (vsid << 28) | (ea & 0xFFFFFFF); - vpn = va >> PAGE_SHIFT; - - hash = hpt_hash(vpn, 0); - - hpteg = ((hash & htab_hash_mask) * HPTES_PER_GROUP); - - /* Panic if a pte grpup is full */ - if (ppc_md.hpte_insert(hpteg, va, pa >> PAGE_SHIFT, - HPTE_V_BOLTED, - _PAGE_NO_CACHE|_PAGE_GUARDED|PP_RWXX) - == -1) { - panic("map_io_page: could not insert mapping"); - } - } - return 0; -} - - -static void __iomem * __ioremap_com(unsigned long addr, unsigned long pa, - unsigned long ea, unsigned long size, - unsigned long flags) -{ - unsigned long i; - - if ((flags & _PAGE_PRESENT) == 0) - flags |= pgprot_val(PAGE_KERNEL); - - for (i = 0; i < size; i += PAGE_SIZE) - if (map_io_page(ea+i, pa+i, flags)) - return NULL; - - return (void __iomem *) (ea + (addr & ~PAGE_MASK)); -} - - -void __iomem * -ioremap(unsigned long addr, unsigned long size) -{ - return __ioremap(addr, size, _PAGE_NO_CACHE | _PAGE_GUARDED); -} - -void __iomem * __ioremap(unsigned long addr, unsigned long size, - unsigned long flags) -{ - unsigned long pa, ea; - void __iomem *ret; - - /* - * Choose an address to map it to. - * Once the imalloc system is running, we use it. - * Before that, we map using addresses going - * up from ioremap_bot. imalloc will use - * the addresses from ioremap_bot through - * IMALLOC_END - * - */ - pa = addr & PAGE_MASK; - size = PAGE_ALIGN(addr + size) - pa; - - if (size == 0) - return NULL; - - if (mem_init_done) { - struct vm_struct *area; - area = im_get_free_area(size); - if (area == NULL) - return NULL; - ea = (unsigned long)(area->addr); - ret = __ioremap_com(addr, pa, ea, size, flags); - if (!ret) - im_free(area->addr); - } else { - ea = ioremap_bot; - ret = __ioremap_com(addr, pa, ea, size, flags); - if (ret) - ioremap_bot += size; - } - return ret; -} - -#define IS_PAGE_ALIGNED(_val) ((_val) == ((_val) & PAGE_MASK)) - -int __ioremap_explicit(unsigned long pa, unsigned long ea, - unsigned long size, unsigned long flags) -{ - struct vm_struct *area; - void __iomem *ret; - - /* For now, require page-aligned values for pa, ea, and size */ - if (!IS_PAGE_ALIGNED(pa) || !IS_PAGE_ALIGNED(ea) || - !IS_PAGE_ALIGNED(size)) { - printk(KERN_ERR "unaligned value in %s\n", __FUNCTION__); - return 1; - } - - if (!mem_init_done) { - /* Two things to consider in this case: - * 1) No records will be kept (imalloc, etc) that the region - * has been remapped - * 2) It won't be easy to iounmap() the region later (because - * of 1) - */ - ; - } else { - area = im_get_area(ea, size, - IM_REGION_UNUSED|IM_REGION_SUBSET|IM_REGION_EXISTS); - if (area == NULL) { - /* Expected when PHB-dlpar is in play */ - return 1; - } - if (ea != (unsigned long) area->addr) { - printk(KERN_ERR "unexpected addr return from " - "im_get_area\n"); - return 1; - } - } - - ret = __ioremap_com(pa, pa, ea, size, flags); - if (ret == NULL) { - printk(KERN_ERR "ioremap_explicit() allocation failure !\n"); - return 1; - } - if (ret != (void *) ea) { - printk(KERN_ERR "__ioremap_com() returned unexpected addr\n"); - return 1; - } - - return 0; -} - -/* - * Unmap an IO region and remove it from imalloc'd list. - * Access to IO memory should be serialized by driver. - * This code is modeled after vmalloc code - unmap_vm_area() - * - * XXX what about calls before mem_init_done (ie python_countermeasures()) - */ -void iounmap(volatile void __iomem *token) -{ - void *addr; - - if (!mem_init_done) - return; - - addr = (void *) ((unsigned long __force) token & PAGE_MASK); - - im_free(addr); -} - -static int iounmap_subset_regions(unsigned long addr, unsigned long size) -{ - struct vm_struct *area; - - /* Check whether subsets of this region exist */ - area = im_get_area(addr, size, IM_REGION_SUPERSET); - if (area == NULL) - return 1; - - while (area) { - iounmap((void __iomem *) area->addr); - area = im_get_area(addr, size, - IM_REGION_SUPERSET); - } - - return 0; -} - -int iounmap_explicit(volatile void __iomem *start, unsigned long size) -{ - struct vm_struct *area; - unsigned long addr; - int rc; - - addr = (unsigned long __force) start & PAGE_MASK; - - /* Verify that the region either exists or is a subset of an existing - * region. In the latter case, split the parent region to create - * the exact region - */ - area = im_get_area(addr, size, - IM_REGION_EXISTS | IM_REGION_SUBSET); - if (area == NULL) { - /* Determine whether subset regions exist. If so, unmap */ - rc = iounmap_subset_regions(addr, size); - if (rc) { - printk(KERN_ERR - "%s() cannot unmap nonexistent range 0x%lx\n", - __FUNCTION__, addr); - return 1; - } - } else { - iounmap((void __iomem *) area->addr); - } - /* - * FIXME! This can't be right: - iounmap(area->addr); - * Maybe it should be "iounmap(area);" - */ - return 0; -} - -#endif - -EXPORT_SYMBOL(ioremap); -EXPORT_SYMBOL(__ioremap); -EXPORT_SYMBOL(iounmap); - -void free_initmem(void) -{ - unsigned long addr; - - addr = (unsigned long)__init_begin; - for (; addr < (unsigned long)__init_end; addr += PAGE_SIZE) { - memset((void *)addr, 0xcc, PAGE_SIZE); - ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); - free_page(addr); - totalram_pages++; - } - printk ("Freeing unused kernel memory: %luk freed\n", - ((unsigned long)__init_end - (unsigned long)__init_begin) >> 10); -} - -#ifdef CONFIG_BLK_DEV_INITRD -void free_initrd_mem(unsigned long start, unsigned long end) -{ - if (start < end) - printk ("Freeing initrd memory: %ldk freed\n", (end - start) >> 10); - for (; start < end; start += PAGE_SIZE) { - ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); - free_page(start); - totalram_pages++; - } -} -#endif - -static DEFINE_SPINLOCK(mmu_context_lock); -static DEFINE_IDR(mmu_context_idr); - -int init_new_context(struct task_struct *tsk, struct mm_struct *mm) -{ - int index; - int err; - -again: - if (!idr_pre_get(&mmu_context_idr, GFP_KERNEL)) - return -ENOMEM; - - spin_lock(&mmu_context_lock); - err = idr_get_new_above(&mmu_context_idr, NULL, 1, &index); - spin_unlock(&mmu_context_lock); - - if (err == -EAGAIN) - goto again; - else if (err) - return err; - - if (index > MAX_CONTEXT) { - idr_remove(&mmu_context_idr, index); - return -ENOMEM; - } - - mm->context.id = index; - - return 0; -} - -void destroy_context(struct mm_struct *mm) -{ - spin_lock(&mmu_context_lock); - idr_remove(&mmu_context_idr, mm->context.id); - spin_unlock(&mmu_context_lock); - - mm->context.id = NO_CONTEXT; -} - -/* - * Do very early mm setup. - */ -void __init mm_init_ppc64(void) -{ -#ifndef CONFIG_PPC_ISERIES - unsigned long i; -#endif - - ppc64_boot_msg(0x100, "MM Init"); - - /* This is the story of the IO hole... please, keep seated, - * unfortunately, we are out of oxygen masks at the moment. - * So we need some rough way to tell where your big IO hole - * is. On pmac, it's between 2G and 4G, on POWER3, it's around - * that area as well, on POWER4 we don't have one, etc... - * We need that as a "hint" when sizing the TCE table on POWER3 - * So far, the simplest way that seem work well enough for us it - * to just assume that the first discontinuity in our physical - * RAM layout is the IO hole. That may not be correct in the future - * (and isn't on iSeries but then we don't care ;) - */ - -#ifndef CONFIG_PPC_ISERIES - for (i = 1; i < lmb.memory.cnt; i++) { - unsigned long base, prevbase, prevsize; - - prevbase = lmb.memory.region[i-1].base; - prevsize = lmb.memory.region[i-1].size; - base = lmb.memory.region[i].base; - if (base > (prevbase + prevsize)) { - io_hole_start = prevbase + prevsize; - io_hole_size = base - (prevbase + prevsize); - break; - } - } -#endif /* CONFIG_PPC_ISERIES */ - if (io_hole_start) - printk("IO Hole assumed to be %lx -> %lx\n", - io_hole_start, io_hole_start + io_hole_size - 1); - - ppc64_boot_msg(0x100, "MM Init Done"); -} - -/* - * This is called by /dev/mem to know if a given address has to - * be mapped non-cacheable or not - */ -int page_is_ram(unsigned long pfn) -{ - int i; - unsigned long paddr = (pfn << PAGE_SHIFT); - - for (i=0; i < lmb.memory.cnt; i++) { - unsigned long base; - - base = lmb.memory.region[i].base; - - if ((paddr >= base) && - (paddr < (base + lmb.memory.region[i].size))) { - return 1; - } - } - - return 0; -} -EXPORT_SYMBOL(page_is_ram); - -/* - * Initialize the bootmem system and give it all the memory we - * have available. - */ -#ifndef CONFIG_NEED_MULTIPLE_NODES -void __init do_init_bootmem(void) -{ - unsigned long i; - unsigned long start, bootmap_pages; - unsigned long total_pages = lmb_end_of_DRAM() >> PAGE_SHIFT; - int boot_mapsize; - - /* - * Find an area to use for the bootmem bitmap. Calculate the size of - * bitmap required as (Total Memory) / PAGE_SIZE / BITS_PER_BYTE. - * Add 1 additional page in case the address isn't page-aligned. - */ - bootmap_pages = bootmem_bootmap_pages(total_pages); - - start = lmb_alloc(bootmap_pages<<PAGE_SHIFT, PAGE_SIZE); - BUG_ON(!start); - - boot_mapsize = init_bootmem(start >> PAGE_SHIFT, total_pages); - - max_pfn = max_low_pfn; - - /* Add all physical memory to the bootmem map, mark each area - * present. - */ - for (i=0; i < lmb.memory.cnt; i++) - free_bootmem(lmb.memory.region[i].base, - lmb_size_bytes(&lmb.memory, i)); - - /* reserve the sections we're already using */ - for (i=0; i < lmb.reserved.cnt; i++) - reserve_bootmem(lmb.reserved.region[i].base, - lmb_size_bytes(&lmb.reserved, i)); - - for (i=0; i < lmb.memory.cnt; i++) - memory_present(0, lmb_start_pfn(&lmb.memory, i), - lmb_end_pfn(&lmb.memory, i)); -} - -/* - * paging_init() sets up the page tables - in fact we've already done this. - */ -void __init paging_init(void) -{ - unsigned long zones_size[MAX_NR_ZONES]; - unsigned long zholes_size[MAX_NR_ZONES]; - unsigned long total_ram = lmb_phys_mem_size(); - unsigned long top_of_ram = lmb_end_of_DRAM(); - - printk(KERN_INFO "Top of RAM: 0x%lx, Total RAM: 0x%lx\n", - top_of_ram, total_ram); - printk(KERN_INFO "Memory hole size: %ldMB\n", - (top_of_ram - total_ram) >> 20); - /* - * All pages are DMA-able so we put them all in the DMA zone. - */ - memset(zones_size, 0, sizeof(zones_size)); - memset(zholes_size, 0, sizeof(zholes_size)); - - zones_size[ZONE_DMA] = top_of_ram >> PAGE_SHIFT; - zholes_size[ZONE_DMA] = (top_of_ram - total_ram) >> PAGE_SHIFT; - - free_area_init_node(0, NODE_DATA(0), zones_size, - __pa(PAGE_OFFSET) >> PAGE_SHIFT, zholes_size); -} -#endif /* ! CONFIG_NEED_MULTIPLE_NODES */ - -static struct kcore_list kcore_vmem; - -static int __init setup_kcore(void) -{ - int i; - - for (i=0; i < lmb.memory.cnt; i++) { - unsigned long base, size; - struct kcore_list *kcore_mem; - - base = lmb.memory.region[i].base; - size = lmb.memory.region[i].size; - - /* GFP_ATOMIC to avoid might_sleep warnings during boot */ - kcore_mem = kmalloc(sizeof(struct kcore_list), GFP_ATOMIC); - if (!kcore_mem) - panic("mem_init: kmalloc failed\n"); - - kclist_add(kcore_mem, __va(base), size); - } - - kclist_add(&kcore_vmem, (void *)VMALLOC_START, VMALLOC_END-VMALLOC_START); - - return 0; -} -module_init(setup_kcore); - -void __init mem_init(void) -{ -#ifdef CONFIG_NEED_MULTIPLE_NODES - int nid; -#endif - pg_data_t *pgdat; - unsigned long i; - struct page *page; - unsigned long reservedpages = 0, codesize, initsize, datasize, bsssize; - - num_physpages = max_low_pfn; /* RAM is assumed contiguous */ - high_memory = (void *) __va(max_low_pfn * PAGE_SIZE); - -#ifdef CONFIG_NEED_MULTIPLE_NODES - for_each_online_node(nid) { - if (NODE_DATA(nid)->node_spanned_pages != 0) { - printk("freeing bootmem node %x\n", nid); - totalram_pages += - free_all_bootmem_node(NODE_DATA(nid)); - } - } -#else - max_mapnr = num_physpages; - totalram_pages += free_all_bootmem(); -#endif - - for_each_pgdat(pgdat) { - for (i = 0; i < pgdat->node_spanned_pages; i++) { - page = pgdat_page_nr(pgdat, i); - if (PageReserved(page)) - reservedpages++; - } - } - - codesize = (unsigned long)&_etext - (unsigned long)&_stext; - initsize = (unsigned long)&__init_end - (unsigned long)&__init_begin; - datasize = (unsigned long)&_edata - (unsigned long)&__init_end; - bsssize = (unsigned long)&__bss_stop - (unsigned long)&__bss_start; - - printk(KERN_INFO "Memory: %luk/%luk available (%luk kernel code, " - "%luk reserved, %luk data, %luk bss, %luk init)\n", - (unsigned long)nr_free_pages() << (PAGE_SHIFT-10), - num_physpages << (PAGE_SHIFT-10), - codesize >> 10, - reservedpages << (PAGE_SHIFT-10), - datasize >> 10, - bsssize >> 10, - initsize >> 10); - - mem_init_done = 1; - - /* Initialize the vDSO */ - vdso_init(); -} - -/* - * This is called when a page has been modified by the kernel. - * It just marks the page as not i-cache clean. We do the i-cache - * flush later when the page is given to a user process, if necessary. - */ -void flush_dcache_page(struct page *page) -{ - if (cpu_has_feature(CPU_FTR_COHERENT_ICACHE)) - return; - /* avoid an atomic op if possible */ - if (test_bit(PG_arch_1, &page->flags)) - clear_bit(PG_arch_1, &page->flags); -} -EXPORT_SYMBOL(flush_dcache_page); - -void clear_user_page(void *page, unsigned long vaddr, struct page *pg) -{ - clear_page(page); - - if (cpu_has_feature(CPU_FTR_COHERENT_ICACHE)) - return; - /* - * We shouldnt have to do this, but some versions of glibc - * require it (ld.so assumes zero filled pages are icache clean) - * - Anton - */ - - /* avoid an atomic op if possible */ - if (test_bit(PG_arch_1, &pg->flags)) - clear_bit(PG_arch_1, &pg->flags); -} -EXPORT_SYMBOL(clear_user_page); - -void copy_user_page(void *vto, void *vfrom, unsigned long vaddr, - struct page *pg) -{ - copy_page(vto, vfrom); - - /* - * We should be able to use the following optimisation, however - * there are two problems. - * Firstly a bug in some versions of binutils meant PLT sections - * were not marked executable. - * Secondly the first word in the GOT section is blrl, used - * to establish the GOT address. Until recently the GOT was - * not marked executable. - * - Anton - */ -#if 0 - if (!vma->vm_file && ((vma->vm_flags & VM_EXEC) == 0)) - return; -#endif - - if (cpu_has_feature(CPU_FTR_COHERENT_ICACHE)) - return; - - /* avoid an atomic op if possible */ - if (test_bit(PG_arch_1, &pg->flags)) - clear_bit(PG_arch_1, &pg->flags); -} - -void flush_icache_user_range(struct vm_area_struct *vma, struct page *page, - unsigned long addr, int len) -{ - unsigned long maddr; - - maddr = (unsigned long)page_address(page) + (addr & ~PAGE_MASK); - flush_icache_range(maddr, maddr + len); -} -EXPORT_SYMBOL(flush_icache_user_range); - -/* - * This is called at the end of handling a user page fault, when the - * fault has been handled by updating a PTE in the linux page tables. - * We use it to preload an HPTE into the hash table corresponding to - * the updated linux PTE. - * - * This must always be called with the mm->page_table_lock held - */ -void update_mmu_cache(struct vm_area_struct *vma, unsigned long ea, - pte_t pte) -{ - unsigned long vsid; - void *pgdir; - pte_t *ptep; - int local = 0; - cpumask_t tmp; - unsigned long flags; - - /* handle i-cache coherency */ - if (!cpu_has_feature(CPU_FTR_COHERENT_ICACHE) && - !cpu_has_feature(CPU_FTR_NOEXECUTE)) { - unsigned long pfn = pte_pfn(pte); - if (pfn_valid(pfn)) { - struct page *page = pfn_to_page(pfn); - if (!PageReserved(page) - && !test_bit(PG_arch_1, &page->flags)) { - __flush_dcache_icache(page_address(page)); - set_bit(PG_arch_1, &page->flags); - } - } - } - - /* We only want HPTEs for linux PTEs that have _PAGE_ACCESSED set */ - if (!pte_young(pte)) - return; - - pgdir = vma->vm_mm->pgd; - if (pgdir == NULL) - return; - - ptep = find_linux_pte(pgdir, ea); - if (!ptep) - return; - - vsid = get_vsid(vma->vm_mm->context.id, ea); - - local_irq_save(flags); - tmp = cpumask_of_cpu(smp_processor_id()); - if (cpus_equal(vma->vm_mm->cpu_vm_mask, tmp)) - local = 1; - - __hash_page(ea, 0, vsid, ptep, 0x300, local); - local_irq_restore(flags); -} - -void __iomem * reserve_phb_iospace(unsigned long size) -{ - void __iomem *virt_addr; - - if (phbs_io_bot >= IMALLOC_BASE) - panic("reserve_phb_iospace(): phb io space overflow\n"); - - virt_addr = (void __iomem *) phbs_io_bot; - phbs_io_bot += size; - - return virt_addr; -} - -static void zero_ctor(void *addr, kmem_cache_t *cache, unsigned long flags) -{ - memset(addr, 0, kmem_cache_size(cache)); -} - -static const int pgtable_cache_size[2] = { - PTE_TABLE_SIZE, PMD_TABLE_SIZE -}; -static const char *pgtable_cache_name[ARRAY_SIZE(pgtable_cache_size)] = { - "pgd_pte_cache", "pud_pmd_cache", -}; - -kmem_cache_t *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)]; - -void pgtable_cache_init(void) -{ - int i; - - BUILD_BUG_ON(PTE_TABLE_SIZE != pgtable_cache_size[PTE_CACHE_NUM]); - BUILD_BUG_ON(PMD_TABLE_SIZE != pgtable_cache_size[PMD_CACHE_NUM]); - BUILD_BUG_ON(PUD_TABLE_SIZE != pgtable_cache_size[PUD_CACHE_NUM]); - BUILD_BUG_ON(PGD_TABLE_SIZE != pgtable_cache_size[PGD_CACHE_NUM]); - - for (i = 0; i < ARRAY_SIZE(pgtable_cache_size); i++) { - int size = pgtable_cache_size[i]; - const char *name = pgtable_cache_name[i]; - - pgtable_cache[i] = kmem_cache_create(name, - size, size, - SLAB_HWCACHE_ALIGN - | SLAB_MUST_HWCACHE_ALIGN, - zero_ctor, - NULL); - if (! pgtable_cache[i]) - panic("pgtable_cache_init(): could not create %s!\n", - name); - } -} - -pgprot_t phys_mem_access_prot(struct file *file, unsigned long addr, - unsigned long size, pgprot_t vma_prot) -{ - if (ppc_md.phys_mem_access_prot) - return ppc_md.phys_mem_access_prot(file, addr, size, vma_prot); - - if (!page_is_ram(addr >> PAGE_SHIFT)) - vma_prot = __pgprot(pgprot_val(vma_prot) - | _PAGE_GUARDED | _PAGE_NO_CACHE); - return vma_prot; -} -EXPORT_SYMBOL(phys_mem_access_prot); diff --git a/arch/ppc64/mm/mmap.c b/arch/ppc64/mm/mmap.c deleted file mode 100644 index fe65f522aff3..000000000000 --- a/arch/ppc64/mm/mmap.c +++ /dev/null @@ -1,86 +0,0 @@ -/* - * linux/arch/ppc64/mm/mmap.c - * - * flexible mmap layout support - * - * Copyright 2003-2004 Red Hat Inc., Durham, North Carolina. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - * - * - * Started by Ingo Molnar <mingo@elte.hu> - */ - -#include <linux/personality.h> -#include <linux/mm.h> - -/* - * Top of mmap area (just below the process stack). - * - * Leave an at least ~128 MB hole. - */ -#define MIN_GAP (128*1024*1024) -#define MAX_GAP (TASK_SIZE/6*5) - -static inline unsigned long mmap_base(void) -{ - unsigned long gap = current->signal->rlim[RLIMIT_STACK].rlim_cur; - - if (gap < MIN_GAP) - gap = MIN_GAP; - else if (gap > MAX_GAP) - gap = MAX_GAP; - - return TASK_SIZE - (gap & PAGE_MASK); -} - -static inline int mmap_is_legacy(void) -{ - /* - * Force standard allocation for 64 bit programs. - */ - if (!test_thread_flag(TIF_32BIT)) - return 1; - - if (current->personality & ADDR_COMPAT_LAYOUT) - return 1; - - if (current->signal->rlim[RLIMIT_STACK].rlim_cur == RLIM_INFINITY) - return 1; - - return sysctl_legacy_va_layout; -} - -/* - * This function, called very early during the creation of a new - * process VM image, sets up which VM layout function to use: - */ -void arch_pick_mmap_layout(struct mm_struct *mm) -{ - /* - * Fall back to the standard layout if the personality - * bit is set, or if the expected stack growth is unlimited: - */ - if (mmap_is_legacy()) { - mm->mmap_base = TASK_UNMAPPED_BASE; - mm->get_unmapped_area = arch_get_unmapped_area; - mm->unmap_area = arch_unmap_area; - } else { - mm->mmap_base = mmap_base(); - mm->get_unmapped_area = arch_get_unmapped_area_topdown; - mm->unmap_area = arch_unmap_area_topdown; - } -} diff --git a/arch/ppc64/mm/numa.c b/arch/ppc64/mm/numa.c deleted file mode 100644 index cb864b8f2750..000000000000 --- a/arch/ppc64/mm/numa.c +++ /dev/null @@ -1,779 +0,0 @@ -/* - * pSeries NUMA support - * - * Copyright (C) 2002 Anton Blanchard <anton@au.ibm.com>, IBM - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#include <linux/threads.h> -#include <linux/bootmem.h> -#include <linux/init.h> -#include <linux/mm.h> -#include <linux/mmzone.h> -#include <linux/module.h> -#include <linux/nodemask.h> -#include <linux/cpu.h> -#include <linux/notifier.h> -#include <asm/lmb.h> -#include <asm/machdep.h> -#include <asm/abs_addr.h> - -static int numa_enabled = 1; - -static int numa_debug; -#define dbg(args...) if (numa_debug) { printk(KERN_INFO args); } - -#ifdef DEBUG_NUMA -#define ARRAY_INITIALISER -1 -#else -#define ARRAY_INITIALISER 0 -#endif - -int numa_cpu_lookup_table[NR_CPUS] = { [ 0 ... (NR_CPUS - 1)] = - ARRAY_INITIALISER}; -char *numa_memory_lookup_table; -cpumask_t numa_cpumask_lookup_table[MAX_NUMNODES]; -int nr_cpus_in_node[MAX_NUMNODES] = { [0 ... (MAX_NUMNODES -1)] = 0}; - -struct pglist_data *node_data[MAX_NUMNODES]; -bootmem_data_t __initdata plat_node_bdata[MAX_NUMNODES]; -static int min_common_depth; - -/* - * We need somewhere to store start/span for each node until we have - * allocated the real node_data structures. - */ -static struct { - unsigned long node_start_pfn; - unsigned long node_end_pfn; - unsigned long node_present_pages; -} init_node_data[MAX_NUMNODES] __initdata; - -EXPORT_SYMBOL(node_data); -EXPORT_SYMBOL(numa_cpu_lookup_table); -EXPORT_SYMBOL(numa_memory_lookup_table); -EXPORT_SYMBOL(numa_cpumask_lookup_table); -EXPORT_SYMBOL(nr_cpus_in_node); - -static inline void map_cpu_to_node(int cpu, int node) -{ - numa_cpu_lookup_table[cpu] = node; - if (!(cpu_isset(cpu, numa_cpumask_lookup_table[node]))) { - cpu_set(cpu, numa_cpumask_lookup_table[node]); - nr_cpus_in_node[node]++; - } -} - -#ifdef CONFIG_HOTPLUG_CPU -static void unmap_cpu_from_node(unsigned long cpu) -{ - int node = numa_cpu_lookup_table[cpu]; - - dbg("removing cpu %lu from node %d\n", cpu, node); - - if (cpu_isset(cpu, numa_cpumask_lookup_table[node])) { - cpu_clear(cpu, numa_cpumask_lookup_table[node]); - nr_cpus_in_node[node]--; - } else { - printk(KERN_ERR "WARNING: cpu %lu not found in node %d\n", - cpu, node); - } -} -#endif /* CONFIG_HOTPLUG_CPU */ - -static struct device_node * __devinit find_cpu_node(unsigned int cpu) -{ - unsigned int hw_cpuid = get_hard_smp_processor_id(cpu); - struct device_node *cpu_node = NULL; - unsigned int *interrupt_server, *reg; - int len; - - while ((cpu_node = of_find_node_by_type(cpu_node, "cpu")) != NULL) { - /* Try interrupt server first */ - interrupt_server = (unsigned int *)get_property(cpu_node, - "ibm,ppc-interrupt-server#s", &len); - - len = len / sizeof(u32); - - if (interrupt_server && (len > 0)) { - while (len--) { - if (interrupt_server[len] == hw_cpuid) - return cpu_node; - } - } else { - reg = (unsigned int *)get_property(cpu_node, - "reg", &len); - if (reg && (len > 0) && (reg[0] == hw_cpuid)) - return cpu_node; - } - } - - return NULL; -} - -/* must hold reference to node during call */ -static int *of_get_associativity(struct device_node *dev) -{ - return (unsigned int *)get_property(dev, "ibm,associativity", NULL); -} - -static int of_node_numa_domain(struct device_node *device) -{ - int numa_domain; - unsigned int *tmp; - - if (min_common_depth == -1) - return 0; - - tmp = of_get_associativity(device); - if (tmp && (tmp[0] >= min_common_depth)) { - numa_domain = tmp[min_common_depth]; - } else { - dbg("WARNING: no NUMA information for %s\n", - device->full_name); - numa_domain = 0; - } - return numa_domain; -} - -/* - * In theory, the "ibm,associativity" property may contain multiple - * associativity lists because a resource may be multiply connected - * into the machine. This resource then has different associativity - * characteristics relative to its multiple connections. We ignore - * this for now. We also assume that all cpu and memory sets have - * their distances represented at a common level. This won't be - * true for heirarchical NUMA. - * - * In any case the ibm,associativity-reference-points should give - * the correct depth for a normal NUMA system. - * - * - Dave Hansen <haveblue@us.ibm.com> - */ -static int __init find_min_common_depth(void) -{ - int depth; - unsigned int *ref_points; - struct device_node *rtas_root; - unsigned int len; - - rtas_root = of_find_node_by_path("/rtas"); - - if (!rtas_root) - return -1; - - /* - * this property is 2 32-bit integers, each representing a level of - * depth in the associativity nodes. The first is for an SMP - * configuration (should be all 0's) and the second is for a normal - * NUMA configuration. - */ - ref_points = (unsigned int *)get_property(rtas_root, - "ibm,associativity-reference-points", &len); - - if ((len >= 1) && ref_points) { - depth = ref_points[1]; - } else { - dbg("WARNING: could not find NUMA " - "associativity reference point\n"); - depth = -1; - } - of_node_put(rtas_root); - - return depth; -} - -static int __init get_mem_addr_cells(void) -{ - struct device_node *memory = NULL; - int rc; - - memory = of_find_node_by_type(memory, "memory"); - if (!memory) - return 0; /* it won't matter */ - - rc = prom_n_addr_cells(memory); - return rc; -} - -static int __init get_mem_size_cells(void) -{ - struct device_node *memory = NULL; - int rc; - - memory = of_find_node_by_type(memory, "memory"); - if (!memory) - return 0; /* it won't matter */ - rc = prom_n_size_cells(memory); - return rc; -} - -static unsigned long read_n_cells(int n, unsigned int **buf) -{ - unsigned long result = 0; - - while (n--) { - result = (result << 32) | **buf; - (*buf)++; - } - return result; -} - -/* - * Figure out to which domain a cpu belongs and stick it there. - * Return the id of the domain used. - */ -static int numa_setup_cpu(unsigned long lcpu) -{ - int numa_domain = 0; - struct device_node *cpu = find_cpu_node(lcpu); - - if (!cpu) { - WARN_ON(1); - goto out; - } - - numa_domain = of_node_numa_domain(cpu); - - if (numa_domain >= num_online_nodes()) { - /* - * POWER4 LPAR uses 0xffff as invalid node, - * dont warn in this case. - */ - if (numa_domain != 0xffff) - printk(KERN_ERR "WARNING: cpu %ld " - "maps to invalid NUMA node %d\n", - lcpu, numa_domain); - numa_domain = 0; - } -out: - node_set_online(numa_domain); - - map_cpu_to_node(lcpu, numa_domain); - - of_node_put(cpu); - - return numa_domain; -} - -static int cpu_numa_callback(struct notifier_block *nfb, - unsigned long action, - void *hcpu) -{ - unsigned long lcpu = (unsigned long)hcpu; - int ret = NOTIFY_DONE; - - switch (action) { - case CPU_UP_PREPARE: - if (min_common_depth == -1 || !numa_enabled) - map_cpu_to_node(lcpu, 0); - else - numa_setup_cpu(lcpu); - ret = NOTIFY_OK; - break; -#ifdef CONFIG_HOTPLUG_CPU - case CPU_DEAD: - case CPU_UP_CANCELED: - unmap_cpu_from_node(lcpu); - break; - ret = NOTIFY_OK; -#endif - } - return ret; -} - -/* - * Check and possibly modify a memory region to enforce the memory limit. - * - * Returns the size the region should have to enforce the memory limit. - * This will either be the original value of size, a truncated value, - * or zero. If the returned value of size is 0 the region should be - * discarded as it lies wholy above the memory limit. - */ -static unsigned long __init numa_enforce_memory_limit(unsigned long start, unsigned long size) -{ - /* - * We use lmb_end_of_DRAM() in here instead of memory_limit because - * we've already adjusted it for the limit and it takes care of - * having memory holes below the limit. - */ - extern unsigned long memory_limit; - - if (! memory_limit) - return size; - - if (start + size <= lmb_end_of_DRAM()) - return size; - - if (start >= lmb_end_of_DRAM()) - return 0; - - return lmb_end_of_DRAM() - start; -} - -static int __init parse_numa_properties(void) -{ - struct device_node *cpu = NULL; - struct device_node *memory = NULL; - int addr_cells, size_cells; - int max_domain = 0; - long entries = lmb_end_of_DRAM() >> MEMORY_INCREMENT_SHIFT; - unsigned long i; - - if (numa_enabled == 0) { - printk(KERN_WARNING "NUMA disabled by user\n"); - return -1; - } - - numa_memory_lookup_table = - (char *)abs_to_virt(lmb_alloc(entries * sizeof(char), 1)); - memset(numa_memory_lookup_table, 0, entries * sizeof(char)); - - for (i = 0; i < entries ; i++) - numa_memory_lookup_table[i] = ARRAY_INITIALISER; - - min_common_depth = find_min_common_depth(); - - dbg("NUMA associativity depth for CPU/Memory: %d\n", min_common_depth); - if (min_common_depth < 0) - return min_common_depth; - - max_domain = numa_setup_cpu(boot_cpuid); - - /* - * Even though we connect cpus to numa domains later in SMP init, - * we need to know the maximum node id now. This is because each - * node id must have NODE_DATA etc backing it. - * As a result of hotplug we could still have cpus appear later on - * with larger node ids. In that case we force the cpu into node 0. - */ - for_each_cpu(i) { - int numa_domain; - - cpu = find_cpu_node(i); - - if (cpu) { - numa_domain = of_node_numa_domain(cpu); - of_node_put(cpu); - - if (numa_domain < MAX_NUMNODES && - max_domain < numa_domain) - max_domain = numa_domain; - } - } - - addr_cells = get_mem_addr_cells(); - size_cells = get_mem_size_cells(); - memory = NULL; - while ((memory = of_find_node_by_type(memory, "memory")) != NULL) { - unsigned long start; - unsigned long size; - int numa_domain; - int ranges; - unsigned int *memcell_buf; - unsigned int len; - - memcell_buf = (unsigned int *)get_property(memory, "reg", &len); - if (!memcell_buf || len <= 0) - continue; - - ranges = memory->n_addrs; -new_range: - /* these are order-sensitive, and modify the buffer pointer */ - start = read_n_cells(addr_cells, &memcell_buf); - size = read_n_cells(size_cells, &memcell_buf); - - start = _ALIGN_DOWN(start, MEMORY_INCREMENT); - size = _ALIGN_UP(size, MEMORY_INCREMENT); - - numa_domain = of_node_numa_domain(memory); - - if (numa_domain >= MAX_NUMNODES) { - if (numa_domain != 0xffff) - printk(KERN_ERR "WARNING: memory at %lx maps " - "to invalid NUMA node %d\n", start, - numa_domain); - numa_domain = 0; - } - - if (max_domain < numa_domain) - max_domain = numa_domain; - - if (! (size = numa_enforce_memory_limit(start, size))) { - if (--ranges) - goto new_range; - else - continue; - } - - /* - * Initialize new node struct, or add to an existing one. - */ - if (init_node_data[numa_domain].node_end_pfn) { - if ((start / PAGE_SIZE) < - init_node_data[numa_domain].node_start_pfn) - init_node_data[numa_domain].node_start_pfn = - start / PAGE_SIZE; - if (((start / PAGE_SIZE) + (size / PAGE_SIZE)) > - init_node_data[numa_domain].node_end_pfn) - init_node_data[numa_domain].node_end_pfn = - (start / PAGE_SIZE) + - (size / PAGE_SIZE); - - init_node_data[numa_domain].node_present_pages += - size / PAGE_SIZE; - } else { - node_set_online(numa_domain); - - init_node_data[numa_domain].node_start_pfn = - start / PAGE_SIZE; - init_node_data[numa_domain].node_end_pfn = - init_node_data[numa_domain].node_start_pfn + - size / PAGE_SIZE; - init_node_data[numa_domain].node_present_pages = - size / PAGE_SIZE; - } - - for (i = start ; i < (start+size); i += MEMORY_INCREMENT) - numa_memory_lookup_table[i >> MEMORY_INCREMENT_SHIFT] = - numa_domain; - - if (--ranges) - goto new_range; - } - - for (i = 0; i <= max_domain; i++) - node_set_online(i); - - return 0; -} - -static void __init setup_nonnuma(void) -{ - unsigned long top_of_ram = lmb_end_of_DRAM(); - unsigned long total_ram = lmb_phys_mem_size(); - unsigned long i; - - printk(KERN_INFO "Top of RAM: 0x%lx, Total RAM: 0x%lx\n", - top_of_ram, total_ram); - printk(KERN_INFO "Memory hole size: %ldMB\n", - (top_of_ram - total_ram) >> 20); - - if (!numa_memory_lookup_table) { - long entries = top_of_ram >> MEMORY_INCREMENT_SHIFT; - numa_memory_lookup_table = - (char *)abs_to_virt(lmb_alloc(entries * sizeof(char), 1)); - memset(numa_memory_lookup_table, 0, entries * sizeof(char)); - for (i = 0; i < entries ; i++) - numa_memory_lookup_table[i] = ARRAY_INITIALISER; - } - - map_cpu_to_node(boot_cpuid, 0); - - node_set_online(0); - - init_node_data[0].node_start_pfn = 0; - init_node_data[0].node_end_pfn = lmb_end_of_DRAM() / PAGE_SIZE; - init_node_data[0].node_present_pages = total_ram / PAGE_SIZE; - - for (i = 0 ; i < top_of_ram; i += MEMORY_INCREMENT) - numa_memory_lookup_table[i >> MEMORY_INCREMENT_SHIFT] = 0; -} - -static void __init dump_numa_topology(void) -{ - unsigned int node; - unsigned int count; - - if (min_common_depth == -1 || !numa_enabled) - return; - - for_each_online_node(node) { - unsigned long i; - - printk(KERN_INFO "Node %d Memory:", node); - - count = 0; - - for (i = 0; i < lmb_end_of_DRAM(); i += MEMORY_INCREMENT) { - if (numa_memory_lookup_table[i >> MEMORY_INCREMENT_SHIFT] == node) { - if (count == 0) - printk(" 0x%lx", i); - ++count; - } else { - if (count > 0) - printk("-0x%lx", i); - count = 0; - } - } - - if (count > 0) - printk("-0x%lx", i); - printk("\n"); - } - return; -} - -/* - * Allocate some memory, satisfying the lmb or bootmem allocator where - * required. nid is the preferred node and end is the physical address of - * the highest address in the node. - * - * Returns the physical address of the memory. - */ -static unsigned long careful_allocation(int nid, unsigned long size, - unsigned long align, unsigned long end) -{ - unsigned long ret = lmb_alloc_base(size, align, end); - - /* retry over all memory */ - if (!ret) - ret = lmb_alloc_base(size, align, lmb_end_of_DRAM()); - - if (!ret) - panic("numa.c: cannot allocate %lu bytes on node %d", - size, nid); - - /* - * If the memory came from a previously allocated node, we must - * retry with the bootmem allocator. - */ - if (pa_to_nid(ret) < nid) { - nid = pa_to_nid(ret); - ret = (unsigned long)__alloc_bootmem_node(NODE_DATA(nid), - size, align, 0); - - if (!ret) - panic("numa.c: cannot allocate %lu bytes on node %d", - size, nid); - - ret = virt_to_abs(ret); - - dbg("alloc_bootmem %lx %lx\n", ret, size); - } - - return ret; -} - -void __init do_init_bootmem(void) -{ - int nid; - int addr_cells, size_cells; - struct device_node *memory = NULL; - static struct notifier_block ppc64_numa_nb = { - .notifier_call = cpu_numa_callback, - .priority = 1 /* Must run before sched domains notifier. */ - }; - - min_low_pfn = 0; - max_low_pfn = lmb_end_of_DRAM() >> PAGE_SHIFT; - max_pfn = max_low_pfn; - - if (parse_numa_properties()) - setup_nonnuma(); - else - dump_numa_topology(); - - register_cpu_notifier(&ppc64_numa_nb); - - for_each_online_node(nid) { - unsigned long start_paddr, end_paddr; - int i; - unsigned long bootmem_paddr; - unsigned long bootmap_pages; - - start_paddr = init_node_data[nid].node_start_pfn * PAGE_SIZE; - end_paddr = init_node_data[nid].node_end_pfn * PAGE_SIZE; - - /* Allocate the node structure node local if possible */ - NODE_DATA(nid) = (struct pglist_data *)careful_allocation(nid, - sizeof(struct pglist_data), - SMP_CACHE_BYTES, end_paddr); - NODE_DATA(nid) = abs_to_virt(NODE_DATA(nid)); - memset(NODE_DATA(nid), 0, sizeof(struct pglist_data)); - - dbg("node %d\n", nid); - dbg("NODE_DATA() = %p\n", NODE_DATA(nid)); - - NODE_DATA(nid)->bdata = &plat_node_bdata[nid]; - NODE_DATA(nid)->node_start_pfn = - init_node_data[nid].node_start_pfn; - NODE_DATA(nid)->node_spanned_pages = - end_paddr - start_paddr; - - if (NODE_DATA(nid)->node_spanned_pages == 0) - continue; - - dbg("start_paddr = %lx\n", start_paddr); - dbg("end_paddr = %lx\n", end_paddr); - - bootmap_pages = bootmem_bootmap_pages((end_paddr - start_paddr) >> PAGE_SHIFT); - - bootmem_paddr = careful_allocation(nid, - bootmap_pages << PAGE_SHIFT, - PAGE_SIZE, end_paddr); - memset(abs_to_virt(bootmem_paddr), 0, - bootmap_pages << PAGE_SHIFT); - dbg("bootmap_paddr = %lx\n", bootmem_paddr); - - init_bootmem_node(NODE_DATA(nid), bootmem_paddr >> PAGE_SHIFT, - start_paddr >> PAGE_SHIFT, - end_paddr >> PAGE_SHIFT); - - /* - * We need to do another scan of all memory sections to - * associate memory with the correct node. - */ - addr_cells = get_mem_addr_cells(); - size_cells = get_mem_size_cells(); - memory = NULL; - while ((memory = of_find_node_by_type(memory, "memory")) != NULL) { - unsigned long mem_start, mem_size; - int numa_domain, ranges; - unsigned int *memcell_buf; - unsigned int len; - - memcell_buf = (unsigned int *)get_property(memory, "reg", &len); - if (!memcell_buf || len <= 0) - continue; - - ranges = memory->n_addrs; /* ranges in cell */ -new_range: - mem_start = read_n_cells(addr_cells, &memcell_buf); - mem_size = read_n_cells(size_cells, &memcell_buf); - if (numa_enabled) { - numa_domain = of_node_numa_domain(memory); - if (numa_domain >= MAX_NUMNODES) - numa_domain = 0; - } else - numa_domain = 0; - - if (numa_domain != nid) - continue; - - mem_size = numa_enforce_memory_limit(mem_start, mem_size); - if (mem_size) { - dbg("free_bootmem %lx %lx\n", mem_start, mem_size); - free_bootmem_node(NODE_DATA(nid), mem_start, mem_size); - } - - if (--ranges) /* process all ranges in cell */ - goto new_range; - } - - /* - * Mark reserved regions on this node - */ - for (i = 0; i < lmb.reserved.cnt; i++) { - unsigned long physbase = lmb.reserved.region[i].base; - unsigned long size = lmb.reserved.region[i].size; - - if (pa_to_nid(physbase) != nid && - pa_to_nid(physbase+size-1) != nid) - continue; - - if (physbase < end_paddr && - (physbase+size) > start_paddr) { - /* overlaps */ - if (physbase < start_paddr) { - size -= start_paddr - physbase; - physbase = start_paddr; - } - - if (size > end_paddr - physbase) - size = end_paddr - physbase; - - dbg("reserve_bootmem %lx %lx\n", physbase, - size); - reserve_bootmem_node(NODE_DATA(nid), physbase, - size); - } - } - /* - * This loop may look famaliar, but we have to do it again - * after marking our reserved memory to mark memory present - * for sparsemem. - */ - addr_cells = get_mem_addr_cells(); - size_cells = get_mem_size_cells(); - memory = NULL; - while ((memory = of_find_node_by_type(memory, "memory")) != NULL) { - unsigned long mem_start, mem_size; - int numa_domain, ranges; - unsigned int *memcell_buf; - unsigned int len; - - memcell_buf = (unsigned int *)get_property(memory, "reg", &len); - if (!memcell_buf || len <= 0) - continue; - - ranges = memory->n_addrs; /* ranges in cell */ -new_range2: - mem_start = read_n_cells(addr_cells, &memcell_buf); - mem_size = read_n_cells(size_cells, &memcell_buf); - if (numa_enabled) { - numa_domain = of_node_numa_domain(memory); - if (numa_domain >= MAX_NUMNODES) - numa_domain = 0; - } else - numa_domain = 0; - - if (numa_domain != nid) - continue; - - mem_size = numa_enforce_memory_limit(mem_start, mem_size); - memory_present(numa_domain, mem_start >> PAGE_SHIFT, - (mem_start + mem_size) >> PAGE_SHIFT); - - if (--ranges) /* process all ranges in cell */ - goto new_range2; - } - - } -} - -void __init paging_init(void) -{ - unsigned long zones_size[MAX_NR_ZONES]; - unsigned long zholes_size[MAX_NR_ZONES]; - int nid; - - memset(zones_size, 0, sizeof(zones_size)); - memset(zholes_size, 0, sizeof(zholes_size)); - - for_each_online_node(nid) { - unsigned long start_pfn; - unsigned long end_pfn; - - start_pfn = init_node_data[nid].node_start_pfn; - end_pfn = init_node_data[nid].node_end_pfn; - - zones_size[ZONE_DMA] = end_pfn - start_pfn; - zholes_size[ZONE_DMA] = zones_size[ZONE_DMA] - - init_node_data[nid].node_present_pages; - - dbg("free_area_init node %d %lx %lx (hole: %lx)\n", nid, - zones_size[ZONE_DMA], start_pfn, zholes_size[ZONE_DMA]); - - free_area_init_node(nid, NODE_DATA(nid), zones_size, - start_pfn, zholes_size); - } -} - -static int __init early_numa(char *p) -{ - if (!p) - return 0; - - if (strstr(p, "off")) - numa_enabled = 0; - - if (strstr(p, "debug")) - numa_debug = 1; - - return 0; -} -early_param("numa", early_numa); diff --git a/arch/ppc64/mm/slb.c b/arch/ppc64/mm/slb.c deleted file mode 100644 index 0473953f6a37..000000000000 --- a/arch/ppc64/mm/slb.c +++ /dev/null @@ -1,158 +0,0 @@ -/* - * PowerPC64 SLB support. - * - * Copyright (C) 2004 David Gibson <dwg@au.ibm.com>, IBM - * Based on earlier code writteh by: - * Dave Engebretsen and Mike Corrigan {engebret|mikejc}@us.ibm.com - * Copyright (c) 2001 Dave Engebretsen - * Copyright (C) 2002 Anton Blanchard <anton@au.ibm.com>, IBM - * - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/config.h> -#include <asm/pgtable.h> -#include <asm/mmu.h> -#include <asm/mmu_context.h> -#include <asm/paca.h> -#include <asm/cputable.h> - -extern void slb_allocate(unsigned long ea); - -static inline unsigned long mk_esid_data(unsigned long ea, unsigned long slot) -{ - return (ea & ESID_MASK) | SLB_ESID_V | slot; -} - -static inline unsigned long mk_vsid_data(unsigned long ea, unsigned long flags) -{ - return (get_kernel_vsid(ea) << SLB_VSID_SHIFT) | flags; -} - -static inline void create_slbe(unsigned long ea, unsigned long flags, - unsigned long entry) -{ - asm volatile("slbmte %0,%1" : - : "r" (mk_vsid_data(ea, flags)), - "r" (mk_esid_data(ea, entry)) - : "memory" ); -} - -static void slb_flush_and_rebolt(void) -{ - /* If you change this make sure you change SLB_NUM_BOLTED - * appropriately too. */ - unsigned long ksp_flags = SLB_VSID_KERNEL; - unsigned long ksp_esid_data; - - WARN_ON(!irqs_disabled()); - - if (cpu_has_feature(CPU_FTR_16M_PAGE)) - ksp_flags |= SLB_VSID_L; - - ksp_esid_data = mk_esid_data(get_paca()->kstack, 2); - if ((ksp_esid_data & ESID_MASK) == KERNELBASE) - ksp_esid_data &= ~SLB_ESID_V; - - /* We need to do this all in asm, so we're sure we don't touch - * the stack between the slbia and rebolting it. */ - asm volatile("isync\n" - "slbia\n" - /* Slot 1 - first VMALLOC segment */ - "slbmte %0,%1\n" - /* Slot 2 - kernel stack */ - "slbmte %2,%3\n" - "isync" - :: "r"(mk_vsid_data(VMALLOCBASE, SLB_VSID_KERNEL)), - "r"(mk_esid_data(VMALLOCBASE, 1)), - "r"(mk_vsid_data(ksp_esid_data, ksp_flags)), - "r"(ksp_esid_data) - : "memory"); -} - -/* Flush all user entries from the segment table of the current processor. */ -void switch_slb(struct task_struct *tsk, struct mm_struct *mm) -{ - unsigned long offset = get_paca()->slb_cache_ptr; - unsigned long esid_data = 0; - unsigned long pc = KSTK_EIP(tsk); - unsigned long stack = KSTK_ESP(tsk); - unsigned long unmapped_base; - - if (offset <= SLB_CACHE_ENTRIES) { - int i; - asm volatile("isync" : : : "memory"); - for (i = 0; i < offset; i++) { - esid_data = ((unsigned long)get_paca()->slb_cache[i] - << SID_SHIFT) | SLBIE_C; - asm volatile("slbie %0" : : "r" (esid_data)); - } - asm volatile("isync" : : : "memory"); - } else { - slb_flush_and_rebolt(); - } - - /* Workaround POWER5 < DD2.1 issue */ - if (offset == 1 || offset > SLB_CACHE_ENTRIES) - asm volatile("slbie %0" : : "r" (esid_data)); - - get_paca()->slb_cache_ptr = 0; - get_paca()->context = mm->context; - - /* - * preload some userspace segments into the SLB. - */ - if (test_tsk_thread_flag(tsk, TIF_32BIT)) - unmapped_base = TASK_UNMAPPED_BASE_USER32; - else - unmapped_base = TASK_UNMAPPED_BASE_USER64; - - if (pc >= KERNELBASE) - return; - slb_allocate(pc); - - if (GET_ESID(pc) == GET_ESID(stack)) - return; - - if (stack >= KERNELBASE) - return; - slb_allocate(stack); - - if ((GET_ESID(pc) == GET_ESID(unmapped_base)) - || (GET_ESID(stack) == GET_ESID(unmapped_base))) - return; - - if (unmapped_base >= KERNELBASE) - return; - slb_allocate(unmapped_base); -} - -void slb_initialize(void) -{ - /* On iSeries the bolted entries have already been set up by - * the hypervisor from the lparMap data in head.S */ -#ifndef CONFIG_PPC_ISERIES - unsigned long flags = SLB_VSID_KERNEL; - - /* Invalidate the entire SLB (even slot 0) & all the ERATS */ - if (cpu_has_feature(CPU_FTR_16M_PAGE)) - flags |= SLB_VSID_L; - - asm volatile("isync":::"memory"); - asm volatile("slbmte %0,%0"::"r" (0) : "memory"); - asm volatile("isync; slbia; isync":::"memory"); - create_slbe(KERNELBASE, flags, 0); - create_slbe(VMALLOCBASE, SLB_VSID_KERNEL, 1); - /* We don't bolt the stack for the time being - we're in boot, - * so the stack is in the bolted segment. By the time it goes - * elsewhere, we'll call _switch() which will bolt in the new - * one. */ - asm volatile("isync":::"memory"); -#endif - - get_paca()->stab_rr = SLB_NUM_BOLTED; -} diff --git a/arch/ppc64/mm/slb_low.S b/arch/ppc64/mm/slb_low.S deleted file mode 100644 index a3a03da503bc..000000000000 --- a/arch/ppc64/mm/slb_low.S +++ /dev/null @@ -1,151 +0,0 @@ -/* - * arch/ppc64/mm/slb_low.S - * - * Low-level SLB routines - * - * Copyright (C) 2004 David Gibson <dwg@au.ibm.com>, IBM - * - * Based on earlier C version: - * Dave Engebretsen and Mike Corrigan {engebret|mikejc}@us.ibm.com - * Copyright (c) 2001 Dave Engebretsen - * Copyright (C) 2002 Anton Blanchard <anton@au.ibm.com>, IBM - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/config.h> -#include <asm/processor.h> -#include <asm/page.h> -#include <asm/mmu.h> -#include <asm/ppc_asm.h> -#include <asm/asm-offsets.h> -#include <asm/cputable.h> - -/* void slb_allocate(unsigned long ea); - * - * Create an SLB entry for the given EA (user or kernel). - * r3 = faulting address, r13 = PACA - * r9, r10, r11 are clobbered by this function - * No other registers are examined or changed. - */ -_GLOBAL(slb_allocate) - /* - * First find a slot, round robin. Previously we tried to find - * a free slot first but that took too long. Unfortunately we - * dont have any LRU information to help us choose a slot. - */ -#ifdef CONFIG_PPC_ISERIES - /* - * On iSeries, the "bolted" stack segment can be cast out on - * shared processor switch so we need to check for a miss on - * it and restore it to the right slot. - */ - ld r9,PACAKSAVE(r13) - clrrdi r9,r9,28 - clrrdi r11,r3,28 - li r10,SLB_NUM_BOLTED-1 /* Stack goes in last bolted slot */ - cmpld r9,r11 - beq 3f -#endif /* CONFIG_PPC_ISERIES */ - - ld r10,PACASTABRR(r13) - addi r10,r10,1 - /* use a cpu feature mask if we ever change our slb size */ - cmpldi r10,SLB_NUM_ENTRIES - - blt+ 4f - li r10,SLB_NUM_BOLTED - -4: - std r10,PACASTABRR(r13) -3: - /* r3 = faulting address, r10 = entry */ - - srdi r9,r3,60 /* get region */ - srdi r3,r3,28 /* get esid */ - cmpldi cr7,r9,0xc /* cmp KERNELBASE for later use */ - - rldimi r10,r3,28,0 /* r10= ESID<<28 | entry */ - oris r10,r10,SLB_ESID_V@h /* r10 |= SLB_ESID_V */ - - /* r3 = esid, r10 = esid_data, cr7 = <>KERNELBASE */ - - blt cr7,0f /* user or kernel? */ - - /* kernel address: proto-VSID = ESID */ - /* WARNING - MAGIC: we don't use the VSID 0xfffffffff, but - * this code will generate the protoVSID 0xfffffffff for the - * top segment. That's ok, the scramble below will translate - * it to VSID 0, which is reserved as a bad VSID - one which - * will never have any pages in it. */ - li r11,SLB_VSID_KERNEL -BEGIN_FTR_SECTION - bne cr7,9f - li r11,(SLB_VSID_KERNEL|SLB_VSID_L) -END_FTR_SECTION_IFSET(CPU_FTR_16M_PAGE) - b 9f - -0: /* user address: proto-VSID = context<<15 | ESID */ - srdi. r9,r3,USER_ESID_BITS - bne- 8f /* invalid ea bits set */ - -#ifdef CONFIG_HUGETLB_PAGE -BEGIN_FTR_SECTION - lhz r9,PACAHIGHHTLBAREAS(r13) - srdi r11,r3,(HTLB_AREA_SHIFT-SID_SHIFT) - srd r9,r9,r11 - lhz r11,PACALOWHTLBAREAS(r13) - srd r11,r11,r3 - or r9,r9,r11 -END_FTR_SECTION_IFSET(CPU_FTR_16M_PAGE) -#endif /* CONFIG_HUGETLB_PAGE */ - - li r11,SLB_VSID_USER - -#ifdef CONFIG_HUGETLB_PAGE -BEGIN_FTR_SECTION - rldimi r11,r9,8,55 /* shift masked bit into SLB_VSID_L */ -END_FTR_SECTION_IFSET(CPU_FTR_16M_PAGE) -#endif /* CONFIG_HUGETLB_PAGE */ - - ld r9,PACACONTEXTID(r13) - rldimi r3,r9,USER_ESID_BITS,0 - -9: /* r3 = protovsid, r11 = flags, r10 = esid_data, cr7 = <>KERNELBASE */ - ASM_VSID_SCRAMBLE(r3,r9) - - rldimi r11,r3,SLB_VSID_SHIFT,16 /* combine VSID and flags */ - - /* - * No need for an isync before or after this slbmte. The exception - * we enter with and the rfid we exit with are context synchronizing. - */ - slbmte r11,r10 - - bgelr cr7 /* we're done for kernel addresses */ - - /* Update the slb cache */ - lhz r3,PACASLBCACHEPTR(r13) /* offset = paca->slb_cache_ptr */ - cmpldi r3,SLB_CACHE_ENTRIES - bge 1f - - /* still room in the slb cache */ - sldi r11,r3,1 /* r11 = offset * sizeof(u16) */ - rldicl r10,r10,36,28 /* get low 16 bits of the ESID */ - add r11,r11,r13 /* r11 = (u16 *)paca + offset */ - sth r10,PACASLBCACHE(r11) /* paca->slb_cache[offset] = esid */ - addi r3,r3,1 /* offset++ */ - b 2f -1: /* offset >= SLB_CACHE_ENTRIES */ - li r3,SLB_CACHE_ENTRIES+1 -2: - sth r3,PACASLBCACHEPTR(r13) /* paca->slb_cache_ptr = offset */ - blr - -8: /* invalid EA */ - li r3,0 /* BAD_VSID */ - li r11,SLB_VSID_USER /* flags don't much matter */ - b 9b diff --git a/arch/ppc64/mm/stab.c b/arch/ppc64/mm/stab.c deleted file mode 100644 index 1b83f002bf27..000000000000 --- a/arch/ppc64/mm/stab.c +++ /dev/null @@ -1,279 +0,0 @@ -/* - * PowerPC64 Segment Translation Support. - * - * Dave Engebretsen and Mike Corrigan {engebret|mikejc}@us.ibm.com - * Copyright (c) 2001 Dave Engebretsen - * - * Copyright (C) 2002 Anton Blanchard <anton@au.ibm.com>, IBM - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/config.h> -#include <asm/pgtable.h> -#include <asm/mmu.h> -#include <asm/mmu_context.h> -#include <asm/paca.h> -#include <asm/cputable.h> -#include <asm/lmb.h> -#include <asm/abs_addr.h> - -struct stab_entry { - unsigned long esid_data; - unsigned long vsid_data; -}; - -/* Both the segment table and SLB code uses the following cache */ -#define NR_STAB_CACHE_ENTRIES 8 -DEFINE_PER_CPU(long, stab_cache_ptr); -DEFINE_PER_CPU(long, stab_cache[NR_STAB_CACHE_ENTRIES]); - -/* - * Create a segment table entry for the given esid/vsid pair. - */ -static int make_ste(unsigned long stab, unsigned long esid, unsigned long vsid) -{ - unsigned long esid_data, vsid_data; - unsigned long entry, group, old_esid, castout_entry, i; - unsigned int global_entry; - struct stab_entry *ste, *castout_ste; - unsigned long kernel_segment = (esid << SID_SHIFT) >= KERNELBASE; - - vsid_data = vsid << STE_VSID_SHIFT; - esid_data = esid << SID_SHIFT | STE_ESID_KP | STE_ESID_V; - if (! kernel_segment) - esid_data |= STE_ESID_KS; - - /* Search the primary group first. */ - global_entry = (esid & 0x1f) << 3; - ste = (struct stab_entry *)(stab | ((esid & 0x1f) << 7)); - - /* Find an empty entry, if one exists. */ - for (group = 0; group < 2; group++) { - for (entry = 0; entry < 8; entry++, ste++) { - if (!(ste->esid_data & STE_ESID_V)) { - ste->vsid_data = vsid_data; - asm volatile("eieio":::"memory"); - ste->esid_data = esid_data; - return (global_entry | entry); - } - } - /* Now search the secondary group. */ - global_entry = ((~esid) & 0x1f) << 3; - ste = (struct stab_entry *)(stab | (((~esid) & 0x1f) << 7)); - } - - /* - * Could not find empty entry, pick one with a round robin selection. - * Search all entries in the two groups. - */ - castout_entry = get_paca()->stab_rr; - for (i = 0; i < 16; i++) { - if (castout_entry < 8) { - global_entry = (esid & 0x1f) << 3; - ste = (struct stab_entry *)(stab | ((esid & 0x1f) << 7)); - castout_ste = ste + castout_entry; - } else { - global_entry = ((~esid) & 0x1f) << 3; - ste = (struct stab_entry *)(stab | (((~esid) & 0x1f) << 7)); - castout_ste = ste + (castout_entry - 8); - } - - /* Dont cast out the first kernel segment */ - if ((castout_ste->esid_data & ESID_MASK) != KERNELBASE) - break; - - castout_entry = (castout_entry + 1) & 0xf; - } - - get_paca()->stab_rr = (castout_entry + 1) & 0xf; - - /* Modify the old entry to the new value. */ - - /* Force previous translations to complete. DRENG */ - asm volatile("isync" : : : "memory"); - - old_esid = castout_ste->esid_data >> SID_SHIFT; - castout_ste->esid_data = 0; /* Invalidate old entry */ - - asm volatile("sync" : : : "memory"); /* Order update */ - - castout_ste->vsid_data = vsid_data; - asm volatile("eieio" : : : "memory"); /* Order update */ - castout_ste->esid_data = esid_data; - - asm volatile("slbie %0" : : "r" (old_esid << SID_SHIFT)); - /* Ensure completion of slbie */ - asm volatile("sync" : : : "memory"); - - return (global_entry | (castout_entry & 0x7)); -} - -/* - * Allocate a segment table entry for the given ea and mm - */ -static int __ste_allocate(unsigned long ea, struct mm_struct *mm) -{ - unsigned long vsid; - unsigned char stab_entry; - unsigned long offset; - - /* Kernel or user address? */ - if (ea >= KERNELBASE) { - vsid = get_kernel_vsid(ea); - } else { - if ((ea >= TASK_SIZE_USER64) || (! mm)) - return 1; - - vsid = get_vsid(mm->context.id, ea); - } - - stab_entry = make_ste(get_paca()->stab_addr, GET_ESID(ea), vsid); - - if (ea < KERNELBASE) { - offset = __get_cpu_var(stab_cache_ptr); - if (offset < NR_STAB_CACHE_ENTRIES) - __get_cpu_var(stab_cache[offset++]) = stab_entry; - else - offset = NR_STAB_CACHE_ENTRIES+1; - __get_cpu_var(stab_cache_ptr) = offset; - - /* Order update */ - asm volatile("sync":::"memory"); - } - - return 0; -} - -int ste_allocate(unsigned long ea) -{ - return __ste_allocate(ea, current->mm); -} - -/* - * Do the segment table work for a context switch: flush all user - * entries from the table, then preload some probably useful entries - * for the new task - */ -void switch_stab(struct task_struct *tsk, struct mm_struct *mm) -{ - struct stab_entry *stab = (struct stab_entry *) get_paca()->stab_addr; - struct stab_entry *ste; - unsigned long offset = __get_cpu_var(stab_cache_ptr); - unsigned long pc = KSTK_EIP(tsk); - unsigned long stack = KSTK_ESP(tsk); - unsigned long unmapped_base; - - /* Force previous translations to complete. DRENG */ - asm volatile("isync" : : : "memory"); - - if (offset <= NR_STAB_CACHE_ENTRIES) { - int i; - - for (i = 0; i < offset; i++) { - ste = stab + __get_cpu_var(stab_cache[i]); - ste->esid_data = 0; /* invalidate entry */ - } - } else { - unsigned long entry; - - /* Invalidate all entries. */ - ste = stab; - - /* Never flush the first entry. */ - ste += 1; - for (entry = 1; - entry < (PAGE_SIZE / sizeof(struct stab_entry)); - entry++, ste++) { - unsigned long ea; - ea = ste->esid_data & ESID_MASK; - if (ea < KERNELBASE) { - ste->esid_data = 0; - } - } - } - - asm volatile("sync; slbia; sync":::"memory"); - - __get_cpu_var(stab_cache_ptr) = 0; - - /* Now preload some entries for the new task */ - if (test_tsk_thread_flag(tsk, TIF_32BIT)) - unmapped_base = TASK_UNMAPPED_BASE_USER32; - else - unmapped_base = TASK_UNMAPPED_BASE_USER64; - - __ste_allocate(pc, mm); - - if (GET_ESID(pc) == GET_ESID(stack)) - return; - - __ste_allocate(stack, mm); - - if ((GET_ESID(pc) == GET_ESID(unmapped_base)) - || (GET_ESID(stack) == GET_ESID(unmapped_base))) - return; - - __ste_allocate(unmapped_base, mm); - - /* Order update */ - asm volatile("sync" : : : "memory"); -} - -extern void slb_initialize(void); - -/* - * Allocate segment tables for secondary CPUs. These must all go in - * the first (bolted) segment, so that do_stab_bolted won't get a - * recursive segment miss on the segment table itself. - */ -void stabs_alloc(void) -{ - int cpu; - - if (cpu_has_feature(CPU_FTR_SLB)) - return; - - for_each_cpu(cpu) { - unsigned long newstab; - - if (cpu == 0) - continue; /* stab for CPU 0 is statically allocated */ - - newstab = lmb_alloc_base(PAGE_SIZE, PAGE_SIZE, 1<<SID_SHIFT); - if (! newstab) - panic("Unable to allocate segment table for CPU %d.\n", - cpu); - - newstab += KERNELBASE; - - memset((void *)newstab, 0, PAGE_SIZE); - - paca[cpu].stab_addr = newstab; - paca[cpu].stab_real = virt_to_abs(newstab); - printk(KERN_DEBUG "Segment table for CPU %d at 0x%lx virtual, 0x%lx absolute\n", cpu, paca[cpu].stab_addr, paca[cpu].stab_real); - } -} - -/* - * Build an entry for the base kernel segment and put it into - * the segment table or SLB. All other segment table or SLB - * entries are faulted in. - */ -void stab_initialize(unsigned long stab) -{ - unsigned long vsid = get_kernel_vsid(KERNELBASE); - - if (cpu_has_feature(CPU_FTR_SLB)) { - slb_initialize(); - } else { - asm volatile("isync; slbia; isync":::"memory"); - make_ste(stab, GET_ESID(KERNELBASE), vsid); - - /* Order update */ - asm volatile("sync":::"memory"); - } -} diff --git a/arch/ppc64/mm/tlb.c b/arch/ppc64/mm/tlb.c deleted file mode 100644 index 21fbffb23a43..000000000000 --- a/arch/ppc64/mm/tlb.c +++ /dev/null @@ -1,197 +0,0 @@ -/* - * This file contains the routines for flushing entries from the - * TLB and MMU hash table. - * - * Derived from arch/ppc64/mm/init.c: - * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org) - * - * Modifications by Paul Mackerras (PowerMac) (paulus@cs.anu.edu.au) - * and Cort Dougan (PReP) (cort@cs.nmt.edu) - * Copyright (C) 1996 Paul Mackerras - * Amiga/APUS changes by Jesper Skov (jskov@cygnus.co.uk). - * - * Derived from "arch/i386/mm/init.c" - * Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds - * - * Dave Engebretsen <engebret@us.ibm.com> - * Rework for PPC64 port. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#include <linux/config.h> -#include <linux/kernel.h> -#include <linux/mm.h> -#include <linux/init.h> -#include <linux/percpu.h> -#include <linux/hardirq.h> -#include <asm/pgalloc.h> -#include <asm/tlbflush.h> -#include <asm/tlb.h> -#include <linux/highmem.h> - -DEFINE_PER_CPU(struct ppc64_tlb_batch, ppc64_tlb_batch); - -/* This is declared as we are using the more or less generic - * include/asm-ppc64/tlb.h file -- tgall - */ -DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); -DEFINE_PER_CPU(struct pte_freelist_batch *, pte_freelist_cur); -unsigned long pte_freelist_forced_free; - -struct pte_freelist_batch -{ - struct rcu_head rcu; - unsigned int index; - pgtable_free_t tables[0]; -}; - -DEFINE_PER_CPU(struct pte_freelist_batch *, pte_freelist_cur); -unsigned long pte_freelist_forced_free; - -#define PTE_FREELIST_SIZE \ - ((PAGE_SIZE - sizeof(struct pte_freelist_batch)) \ - / sizeof(pgtable_free_t)) - -#ifdef CONFIG_SMP -static void pte_free_smp_sync(void *arg) -{ - /* Do nothing, just ensure we sync with all CPUs */ -} -#endif - -/* This is only called when we are critically out of memory - * (and fail to get a page in pte_free_tlb). - */ -static void pgtable_free_now(pgtable_free_t pgf) -{ - pte_freelist_forced_free++; - - smp_call_function(pte_free_smp_sync, NULL, 0, 1); - - pgtable_free(pgf); -} - -static void pte_free_rcu_callback(struct rcu_head *head) -{ - struct pte_freelist_batch *batch = - container_of(head, struct pte_freelist_batch, rcu); - unsigned int i; - - for (i = 0; i < batch->index; i++) - pgtable_free(batch->tables[i]); - - free_page((unsigned long)batch); -} - -static void pte_free_submit(struct pte_freelist_batch *batch) -{ - INIT_RCU_HEAD(&batch->rcu); - call_rcu(&batch->rcu, pte_free_rcu_callback); -} - -void pgtable_free_tlb(struct mmu_gather *tlb, pgtable_free_t pgf) -{ - /* This is safe as we are holding page_table_lock */ - cpumask_t local_cpumask = cpumask_of_cpu(smp_processor_id()); - struct pte_freelist_batch **batchp = &__get_cpu_var(pte_freelist_cur); - - if (atomic_read(&tlb->mm->mm_users) < 2 || - cpus_equal(tlb->mm->cpu_vm_mask, local_cpumask)) { - pgtable_free(pgf); - return; - } - - if (*batchp == NULL) { - *batchp = (struct pte_freelist_batch *)__get_free_page(GFP_ATOMIC); - if (*batchp == NULL) { - pgtable_free_now(pgf); - return; - } - (*batchp)->index = 0; - } - (*batchp)->tables[(*batchp)->index++] = pgf; - if ((*batchp)->index == PTE_FREELIST_SIZE) { - pte_free_submit(*batchp); - *batchp = NULL; - } -} - -/* - * Update the MMU hash table to correspond with a change to - * a Linux PTE. If wrprot is true, it is permissible to - * change the existing HPTE to read-only rather than removing it - * (if we remove it we should clear the _PTE_HPTEFLAGS bits). - */ -void hpte_update(struct mm_struct *mm, unsigned long addr, - unsigned long pte, int wrprot) -{ - int i; - unsigned long context = 0; - struct ppc64_tlb_batch *batch = &__get_cpu_var(ppc64_tlb_batch); - - if (REGION_ID(addr) == USER_REGION_ID) - context = mm->context.id; - i = batch->index; - - /* - * This can happen when we are in the middle of a TLB batch and - * we encounter memory pressure (eg copy_page_range when it tries - * to allocate a new pte). If we have to reclaim memory and end - * up scanning and resetting referenced bits then our batch context - * will change mid stream. - */ - if (i != 0 && (context != batch->context || - batch->large != pte_huge(pte))) { - flush_tlb_pending(); - i = 0; - } - - if (i == 0) { - batch->context = context; - batch->mm = mm; - batch->large = pte_huge(pte); - } - batch->pte[i] = __pte(pte); - batch->addr[i] = addr; - batch->index = ++i; - if (i >= PPC64_TLB_BATCH_NR) - flush_tlb_pending(); -} - -void __flush_tlb_pending(struct ppc64_tlb_batch *batch) -{ - int i; - int cpu; - cpumask_t tmp; - int local = 0; - - BUG_ON(in_interrupt()); - - cpu = get_cpu(); - i = batch->index; - tmp = cpumask_of_cpu(cpu); - if (cpus_equal(batch->mm->cpu_vm_mask, tmp)) - local = 1; - - if (i == 1) - flush_hash_page(batch->context, batch->addr[0], batch->pte[0], - local); - else - flush_hash_range(batch->context, i, local); - batch->index = 0; - put_cpu(); -} - -void pte_free_finish(void) -{ - /* This is safe as we are holding page_table_lock */ - struct pte_freelist_batch **batchp = &__get_cpu_var(pte_freelist_cur); - - if (*batchp == NULL) - return; - pte_free_submit(*batchp); - *batchp = NULL; -} diff --git a/arch/ppc64/oprofile/Kconfig b/arch/ppc64/oprofile/Kconfig deleted file mode 100644 index 5ade19801b97..000000000000 --- a/arch/ppc64/oprofile/Kconfig +++ /dev/null @@ -1,23 +0,0 @@ - -menu "Profiling support" - depends on EXPERIMENTAL - -config PROFILING - bool "Profiling support (EXPERIMENTAL)" - help - Say Y here to enable the extended profiling support mechanisms used - by profilers such as OProfile. - - -config OPROFILE - tristate "OProfile system profiling (EXPERIMENTAL)" - depends on PROFILING - help - OProfile is a profiling system capable of profiling the - whole system, include the kernel, kernel modules, libraries, - and applications. - - If unsure, say N. - -endmenu - diff --git a/arch/ppc64/oprofile/Makefile b/arch/ppc64/oprofile/Makefile deleted file mode 100644 index 162dbf06c142..000000000000 --- a/arch/ppc64/oprofile/Makefile +++ /dev/null @@ -1,9 +0,0 @@ -obj-$(CONFIG_OPROFILE) += oprofile.o - -DRIVER_OBJS := $(addprefix ../../../drivers/oprofile/, \ - oprof.o cpu_buffer.o buffer_sync.o \ - event_buffer.o oprofile_files.o \ - oprofilefs.o oprofile_stats.o \ - timer_int.o ) - -oprofile-y := $(DRIVER_OBJS) common.o op_model_rs64.o op_model_power4.o diff --git a/arch/ppc64/oprofile/common.c b/arch/ppc64/oprofile/common.c deleted file mode 100644 index e5f572710aa0..000000000000 --- a/arch/ppc64/oprofile/common.c +++ /dev/null @@ -1,145 +0,0 @@ -/* - * Copyright (C) 2004 Anton Blanchard <anton@au.ibm.com>, IBM - * - * Based on alpha version. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/oprofile.h> -#include <linux/init.h> -#include <linux/smp.h> -#include <linux/errno.h> -#include <asm/ptrace.h> -#include <asm/system.h> -#include <asm/pmc.h> -#include <asm/cputable.h> -#include <asm/oprofile_impl.h> - -static struct op_ppc64_model *model; - -static struct op_counter_config ctr[OP_MAX_COUNTER]; -static struct op_system_config sys; - -static void op_handle_interrupt(struct pt_regs *regs) -{ - model->handle_interrupt(regs, ctr); -} - -static int op_ppc64_setup(void) -{ - int err; - - /* Grab the hardware */ - err = reserve_pmc_hardware(op_handle_interrupt); - if (err) - return err; - - /* Pre-compute the values to stuff in the hardware registers. */ - model->reg_setup(ctr, &sys, model->num_counters); - - /* Configure the registers on all cpus. */ - on_each_cpu(model->cpu_setup, NULL, 0, 1); - - return 0; -} - -static void op_ppc64_shutdown(void) -{ - release_pmc_hardware(); -} - -static void op_ppc64_cpu_start(void *dummy) -{ - model->start(ctr); -} - -static int op_ppc64_start(void) -{ - on_each_cpu(op_ppc64_cpu_start, NULL, 0, 1); - return 0; -} - -static inline void op_ppc64_cpu_stop(void *dummy) -{ - model->stop(); -} - -static void op_ppc64_stop(void) -{ - on_each_cpu(op_ppc64_cpu_stop, NULL, 0, 1); -} - -static int op_ppc64_create_files(struct super_block *sb, struct dentry *root) -{ - int i; - - /* - * There is one mmcr0, mmcr1 and mmcra for setting the events for - * all of the counters. - */ - oprofilefs_create_ulong(sb, root, "mmcr0", &sys.mmcr0); - oprofilefs_create_ulong(sb, root, "mmcr1", &sys.mmcr1); - oprofilefs_create_ulong(sb, root, "mmcra", &sys.mmcra); - - for (i = 0; i < model->num_counters; ++i) { - struct dentry *dir; - char buf[3]; - - snprintf(buf, sizeof buf, "%d", i); - dir = oprofilefs_mkdir(sb, root, buf); - - oprofilefs_create_ulong(sb, dir, "enabled", &ctr[i].enabled); - oprofilefs_create_ulong(sb, dir, "event", &ctr[i].event); - oprofilefs_create_ulong(sb, dir, "count", &ctr[i].count); - /* - * We dont support per counter user/kernel selection, but - * we leave the entries because userspace expects them - */ - oprofilefs_create_ulong(sb, dir, "kernel", &ctr[i].kernel); - oprofilefs_create_ulong(sb, dir, "user", &ctr[i].user); - oprofilefs_create_ulong(sb, dir, "unit_mask", &ctr[i].unit_mask); - } - - oprofilefs_create_ulong(sb, root, "enable_kernel", &sys.enable_kernel); - oprofilefs_create_ulong(sb, root, "enable_user", &sys.enable_user); - oprofilefs_create_ulong(sb, root, "backtrace_spinlocks", - &sys.backtrace_spinlocks); - - /* Default to tracing both kernel and user */ - sys.enable_kernel = 1; - sys.enable_user = 1; - - /* Turn on backtracing through spinlocks by default */ - sys.backtrace_spinlocks = 1; - - return 0; -} - -int __init oprofile_arch_init(struct oprofile_operations *ops) -{ - if (!cur_cpu_spec->oprofile_model || !cur_cpu_spec->oprofile_cpu_type) - return -ENODEV; - - model = cur_cpu_spec->oprofile_model; - model->num_counters = cur_cpu_spec->num_pmcs; - - ops->cpu_type = cur_cpu_spec->oprofile_cpu_type; - ops->create_files = op_ppc64_create_files; - ops->setup = op_ppc64_setup; - ops->shutdown = op_ppc64_shutdown; - ops->start = op_ppc64_start; - ops->stop = op_ppc64_stop; - - printk(KERN_INFO "oprofile: using %s performance monitoring.\n", - ops->cpu_type); - - return 0; -} - -void oprofile_arch_exit(void) -{ -} diff --git a/arch/ppc64/oprofile/op_model_power4.c b/arch/ppc64/oprofile/op_model_power4.c deleted file mode 100644 index 32b2bb5625fe..000000000000 --- a/arch/ppc64/oprofile/op_model_power4.c +++ /dev/null @@ -1,309 +0,0 @@ -/* - * Copyright (C) 2004 Anton Blanchard <anton@au.ibm.com>, IBM - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/oprofile.h> -#include <linux/init.h> -#include <linux/smp.h> -#include <asm/ptrace.h> -#include <asm/system.h> -#include <asm/processor.h> -#include <asm/cputable.h> -#include <asm/systemcfg.h> -#include <asm/rtas.h> -#include <asm/oprofile_impl.h> - -#define dbg(args...) - -static unsigned long reset_value[OP_MAX_COUNTER]; - -static int oprofile_running; -static int mmcra_has_sihv; - -/* mmcr values are set in power4_reg_setup, used in power4_cpu_setup */ -static u32 mmcr0_val; -static u64 mmcr1_val; -static u32 mmcra_val; - -/* - * Since we do not have an NMI, backtracing through spinlocks is - * only a best guess. In light of this, allow it to be disabled at - * runtime. - */ -static int backtrace_spinlocks; - -static void power4_reg_setup(struct op_counter_config *ctr, - struct op_system_config *sys, - int num_ctrs) -{ - int i; - - /* - * SIHV / SIPR bits are only implemented on POWER4+ (GQ) and above. - * However we disable it on all POWER4 until we verify it works - * (I was seeing some strange behaviour last time I tried). - * - * It has been verified to work on POWER5 so we enable it there. - */ - if (cpu_has_feature(CPU_FTR_MMCRA_SIHV)) - mmcra_has_sihv = 1; - - /* - * The performance counter event settings are given in the mmcr0, - * mmcr1 and mmcra values passed from the user in the - * op_system_config structure (sys variable). - */ - mmcr0_val = sys->mmcr0; - mmcr1_val = sys->mmcr1; - mmcra_val = sys->mmcra; - - backtrace_spinlocks = sys->backtrace_spinlocks; - - for (i = 0; i < cur_cpu_spec->num_pmcs; ++i) - reset_value[i] = 0x80000000UL - ctr[i].count; - - /* setup user and kernel profiling */ - if (sys->enable_kernel) - mmcr0_val &= ~MMCR0_KERNEL_DISABLE; - else - mmcr0_val |= MMCR0_KERNEL_DISABLE; - - if (sys->enable_user) - mmcr0_val &= ~MMCR0_PROBLEM_DISABLE; - else - mmcr0_val |= MMCR0_PROBLEM_DISABLE; -} - -extern void ppc64_enable_pmcs(void); - -static void power4_cpu_setup(void *unused) -{ - unsigned int mmcr0 = mmcr0_val; - unsigned long mmcra = mmcra_val; - - ppc64_enable_pmcs(); - - /* set the freeze bit */ - mmcr0 |= MMCR0_FC; - mtspr(SPRN_MMCR0, mmcr0); - - mmcr0 |= MMCR0_FCM1|MMCR0_PMXE|MMCR0_FCECE; - mmcr0 |= MMCR0_PMC1CE|MMCR0_PMCjCE; - mtspr(SPRN_MMCR0, mmcr0); - - mtspr(SPRN_MMCR1, mmcr1_val); - - mmcra |= MMCRA_SAMPLE_ENABLE; - mtspr(SPRN_MMCRA, mmcra); - - dbg("setup on cpu %d, mmcr0 %lx\n", smp_processor_id(), - mfspr(SPRN_MMCR0)); - dbg("setup on cpu %d, mmcr1 %lx\n", smp_processor_id(), - mfspr(SPRN_MMCR1)); - dbg("setup on cpu %d, mmcra %lx\n", smp_processor_id(), - mfspr(SPRN_MMCRA)); -} - -static void power4_start(struct op_counter_config *ctr) -{ - int i; - unsigned int mmcr0; - - /* set the PMM bit (see comment below) */ - mtmsrd(mfmsr() | MSR_PMM); - - for (i = 0; i < cur_cpu_spec->num_pmcs; ++i) { - if (ctr[i].enabled) { - ctr_write(i, reset_value[i]); - } else { - ctr_write(i, 0); - } - } - - mmcr0 = mfspr(SPRN_MMCR0); - - /* - * We must clear the PMAO bit on some (GQ) chips. Just do it - * all the time - */ - mmcr0 &= ~MMCR0_PMAO; - - /* - * now clear the freeze bit, counting will not start until we - * rfid from this excetion, because only at that point will - * the PMM bit be cleared - */ - mmcr0 &= ~MMCR0_FC; - mtspr(SPRN_MMCR0, mmcr0); - - oprofile_running = 1; - - dbg("start on cpu %d, mmcr0 %x\n", smp_processor_id(), mmcr0); -} - -static void power4_stop(void) -{ - unsigned int mmcr0; - - /* freeze counters */ - mmcr0 = mfspr(SPRN_MMCR0); - mmcr0 |= MMCR0_FC; - mtspr(SPRN_MMCR0, mmcr0); - - oprofile_running = 0; - - dbg("stop on cpu %d, mmcr0 %x\n", smp_processor_id(), mmcr0); - - mb(); -} - -/* Fake functions used by canonicalize_pc */ -static void __attribute_used__ hypervisor_bucket(void) -{ -} - -static void __attribute_used__ rtas_bucket(void) -{ -} - -static void __attribute_used__ kernel_unknown_bucket(void) -{ -} - -static unsigned long check_spinlock_pc(struct pt_regs *regs, - unsigned long profile_pc) -{ - unsigned long pc = instruction_pointer(regs); - - /* - * If both the SIAR (sampled instruction) and the perfmon exception - * occurred in a spinlock region then we account the sample to the - * calling function. This isnt 100% correct, we really need soft - * IRQ disable so we always get the perfmon exception at the - * point at which the SIAR is set. - */ - if (backtrace_spinlocks && in_lock_functions(pc) && - in_lock_functions(profile_pc)) - return regs->link; - else - return profile_pc; -} - -/* - * On GQ and newer the MMCRA stores the HV and PR bits at the time - * the SIAR was sampled. We use that to work out if the SIAR was sampled in - * the hypervisor, our exception vectors or RTAS. - */ -static unsigned long get_pc(struct pt_regs *regs) -{ - unsigned long pc = mfspr(SPRN_SIAR); - unsigned long mmcra; - - /* Cant do much about it */ - if (!mmcra_has_sihv) - return check_spinlock_pc(regs, pc); - - mmcra = mfspr(SPRN_MMCRA); - - /* Were we in the hypervisor? */ - if ((systemcfg->platform == PLATFORM_PSERIES_LPAR) && - (mmcra & MMCRA_SIHV)) - /* function descriptor madness */ - return *((unsigned long *)hypervisor_bucket); - - /* We were in userspace, nothing to do */ - if (mmcra & MMCRA_SIPR) - return pc; - -#ifdef CONFIG_PPC_RTAS - /* Were we in RTAS? */ - if (pc >= rtas.base && pc < (rtas.base + rtas.size)) - /* function descriptor madness */ - return *((unsigned long *)rtas_bucket); -#endif - - /* Were we in our exception vectors or SLB real mode miss handler? */ - if (pc < 0x1000000UL) - return (unsigned long)__va(pc); - - /* Not sure where we were */ - if (pc < KERNELBASE) - /* function descriptor madness */ - return *((unsigned long *)kernel_unknown_bucket); - - return check_spinlock_pc(regs, pc); -} - -static int get_kernel(unsigned long pc) -{ - int is_kernel; - - if (!mmcra_has_sihv) { - is_kernel = (pc >= KERNELBASE); - } else { - unsigned long mmcra = mfspr(SPRN_MMCRA); - is_kernel = ((mmcra & MMCRA_SIPR) == 0); - } - - return is_kernel; -} - -static void power4_handle_interrupt(struct pt_regs *regs, - struct op_counter_config *ctr) -{ - unsigned long pc; - int is_kernel; - int val; - int i; - unsigned int mmcr0; - - pc = get_pc(regs); - is_kernel = get_kernel(pc); - - /* set the PMM bit (see comment below) */ - mtmsrd(mfmsr() | MSR_PMM); - - for (i = 0; i < cur_cpu_spec->num_pmcs; ++i) { - val = ctr_read(i); - if (val < 0) { - if (oprofile_running && ctr[i].enabled) { - oprofile_add_pc(pc, is_kernel, i); - ctr_write(i, reset_value[i]); - } else { - ctr_write(i, 0); - } - } - } - - mmcr0 = mfspr(SPRN_MMCR0); - - /* reset the perfmon trigger */ - mmcr0 |= MMCR0_PMXE; - - /* - * We must clear the PMAO bit on some (GQ) chips. Just do it - * all the time - */ - mmcr0 &= ~MMCR0_PMAO; - - /* - * now clear the freeze bit, counting will not start until we - * rfid from this exception, because only at that point will - * the PMM bit be cleared - */ - mmcr0 &= ~MMCR0_FC; - mtspr(SPRN_MMCR0, mmcr0); -} - -struct op_ppc64_model op_model_power4 = { - .reg_setup = power4_reg_setup, - .cpu_setup = power4_cpu_setup, - .start = power4_start, - .stop = power4_stop, - .handle_interrupt = power4_handle_interrupt, -}; diff --git a/arch/ppc64/oprofile/op_model_rs64.c b/arch/ppc64/oprofile/op_model_rs64.c deleted file mode 100644 index 08c5b333f5c4..000000000000 --- a/arch/ppc64/oprofile/op_model_rs64.c +++ /dev/null @@ -1,218 +0,0 @@ -/* - * Copyright (C) 2004 Anton Blanchard <anton@au.ibm.com>, IBM - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/oprofile.h> -#include <linux/init.h> -#include <linux/smp.h> -#include <asm/ptrace.h> -#include <asm/system.h> -#include <asm/processor.h> -#include <asm/cputable.h> -#include <asm/oprofile_impl.h> - -#define dbg(args...) - -static void ctrl_write(unsigned int i, unsigned int val) -{ - unsigned int tmp = 0; - unsigned long shift = 0, mask = 0; - - dbg("ctrl_write %d %x\n", i, val); - - switch(i) { - case 0: - tmp = mfspr(SPRN_MMCR0); - shift = 6; - mask = 0x7F; - break; - case 1: - tmp = mfspr(SPRN_MMCR0); - shift = 0; - mask = 0x3F; - break; - case 2: - tmp = mfspr(SPRN_MMCR1); - shift = 31 - 4; - mask = 0x1F; - break; - case 3: - tmp = mfspr(SPRN_MMCR1); - shift = 31 - 9; - mask = 0x1F; - break; - case 4: - tmp = mfspr(SPRN_MMCR1); - shift = 31 - 14; - mask = 0x1F; - break; - case 5: - tmp = mfspr(SPRN_MMCR1); - shift = 31 - 19; - mask = 0x1F; - break; - case 6: - tmp = mfspr(SPRN_MMCR1); - shift = 31 - 24; - mask = 0x1F; - break; - case 7: - tmp = mfspr(SPRN_MMCR1); - shift = 31 - 28; - mask = 0xF; - break; - } - - tmp = tmp & ~(mask << shift); - tmp |= val << shift; - - switch(i) { - case 0: - case 1: - mtspr(SPRN_MMCR0, tmp); - break; - default: - mtspr(SPRN_MMCR1, tmp); - } - - dbg("ctrl_write mmcr0 %lx mmcr1 %lx\n", mfspr(SPRN_MMCR0), - mfspr(SPRN_MMCR1)); -} - -static unsigned long reset_value[OP_MAX_COUNTER]; - -static int num_counters; - -static void rs64_reg_setup(struct op_counter_config *ctr, - struct op_system_config *sys, - int num_ctrs) -{ - int i; - - num_counters = num_ctrs; - - for (i = 0; i < num_counters; ++i) - reset_value[i] = 0x80000000UL - ctr[i].count; - - /* XXX setup user and kernel profiling */ -} - -static void rs64_cpu_setup(void *unused) -{ - unsigned int mmcr0; - - /* reset MMCR0 and set the freeze bit */ - mmcr0 = MMCR0_FC; - mtspr(SPRN_MMCR0, mmcr0); - - /* reset MMCR1, MMCRA */ - mtspr(SPRN_MMCR1, 0); - - if (cpu_has_feature(CPU_FTR_MMCRA)) - mtspr(SPRN_MMCRA, 0); - - mmcr0 |= MMCR0_FCM1|MMCR0_PMXE|MMCR0_FCECE; - /* Only applies to POWER3, but should be safe on RS64 */ - mmcr0 |= MMCR0_PMC1CE|MMCR0_PMCjCE; - mtspr(SPRN_MMCR0, mmcr0); - - dbg("setup on cpu %d, mmcr0 %lx\n", smp_processor_id(), - mfspr(SPRN_MMCR0)); - dbg("setup on cpu %d, mmcr1 %lx\n", smp_processor_id(), - mfspr(SPRN_MMCR1)); -} - -static void rs64_start(struct op_counter_config *ctr) -{ - int i; - unsigned int mmcr0; - - /* set the PMM bit (see comment below) */ - mtmsrd(mfmsr() | MSR_PMM); - - for (i = 0; i < num_counters; ++i) { - if (ctr[i].enabled) { - ctr_write(i, reset_value[i]); - ctrl_write(i, ctr[i].event); - } else { - ctr_write(i, 0); - } - } - - mmcr0 = mfspr(SPRN_MMCR0); - - /* - * now clear the freeze bit, counting will not start until we - * rfid from this excetion, because only at that point will - * the PMM bit be cleared - */ - mmcr0 &= ~MMCR0_FC; - mtspr(SPRN_MMCR0, mmcr0); - - dbg("start on cpu %d, mmcr0 %x\n", smp_processor_id(), mmcr0); -} - -static void rs64_stop(void) -{ - unsigned int mmcr0; - - /* freeze counters */ - mmcr0 = mfspr(SPRN_MMCR0); - mmcr0 |= MMCR0_FC; - mtspr(SPRN_MMCR0, mmcr0); - - dbg("stop on cpu %d, mmcr0 %x\n", smp_processor_id(), mmcr0); - - mb(); -} - -static void rs64_handle_interrupt(struct pt_regs *regs, - struct op_counter_config *ctr) -{ - unsigned int mmcr0; - int val; - int i; - unsigned long pc = mfspr(SPRN_SIAR); - int is_kernel = (pc >= KERNELBASE); - - /* set the PMM bit (see comment below) */ - mtmsrd(mfmsr() | MSR_PMM); - - for (i = 0; i < num_counters; ++i) { - val = ctr_read(i); - if (val < 0) { - if (ctr[i].enabled) { - oprofile_add_pc(pc, is_kernel, i); - ctr_write(i, reset_value[i]); - } else { - ctr_write(i, 0); - } - } - } - - mmcr0 = mfspr(SPRN_MMCR0); - - /* reset the perfmon trigger */ - mmcr0 |= MMCR0_PMXE; - - /* - * now clear the freeze bit, counting will not start until we - * rfid from this exception, because only at that point will - * the PMM bit be cleared - */ - mmcr0 &= ~MMCR0_FC; - mtspr(SPRN_MMCR0, mmcr0); -} - -struct op_ppc64_model op_model_rs64 = { - .reg_setup = rs64_reg_setup, - .cpu_setup = rs64_cpu_setup, - .start = rs64_start, - .stop = rs64_stop, - .handle_interrupt = rs64_handle_interrupt, -}; |