From 24cc67de62eebbda3ce0c46bdd56582c00dccd03 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Thu, 20 Jan 2011 18:50:55 +1100 Subject: powerpc: Define CPU feature for Architected 2.06 HV mode This bit indicates that we are operating in hypervisor mode on a CPU compliant to architecture 2.06 or later (currently server only). We set it on POWER7 and have a boot-time CPU setup function that clears it if MSR:HV isn't set (booting under a hypervisor). Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/Makefile | 1 + arch/powerpc/kernel/cpu_setup_power7.S | 65 ++++++++++++++++++++++++++++++++++ arch/powerpc/kernel/cputable.c | 6 ++++ 3 files changed, 72 insertions(+) create mode 100644 arch/powerpc/kernel/cpu_setup_power7.S (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile index 3bb2a3e6a337..7c6eb4974f25 100644 --- a/arch/powerpc/kernel/Makefile +++ b/arch/powerpc/kernel/Makefile @@ -38,6 +38,7 @@ obj-$(CONFIG_PPC64) += setup_64.o sys_ppc32.o \ paca.o nvram_64.o firmware.o obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o obj-$(CONFIG_PPC_BOOK3S_64) += cpu_setup_ppc970.o cpu_setup_pa6t.o +obj-$(CONFIG_PPC_BOOK3S_64) += cpu_setup_power7.o obj64-$(CONFIG_RELOCATABLE) += reloc_64.o obj-$(CONFIG_PPC_BOOK3E_64) += exceptions-64e.o idle_book3e.o obj-$(CONFIG_PPC64) += vdso64/ diff --git a/arch/powerpc/kernel/cpu_setup_power7.S b/arch/powerpc/kernel/cpu_setup_power7.S new file mode 100644 index 000000000000..f2b317817c4e --- /dev/null +++ b/arch/powerpc/kernel/cpu_setup_power7.S @@ -0,0 +1,65 @@ +/* + * This file contains low level CPU setup functions. + * Copyright (C) 2003 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. + * + */ + +#include +#include +#include +#include +#include +#include + +/* Entry: r3 = crap, r4 = ptr to cputable entry + * + * Note that we can be called twice for pseudo-PVRs + */ +_GLOBAL(__setup_cpu_power7) + mflr r11 + bl __init_hvmode_206 + mtlr r11 + beqlr + bl __init_LPCR + mtlr r11 + blr + +_GLOBAL(__restore_cpu_power7) + mflr r11 + mfmsr r3 + rldicl. r0,r3,4,63 + beqlr + bl __init_LPCR + mtlr r11 + blr + +__init_hvmode_206: + /* Disable CPU_FTR_HVMODE_206 and exit if MSR:HV is not set */ + mfmsr r3 + rldicl. r0,r3,4,63 + bnelr + ld r5,CPU_SPEC_FEATURES(r4) + LOAD_REG_IMMEDIATE(r6,CPU_FTR_HVMODE_206) + xor r5,r5,r6 + std r5,CPU_SPEC_FEATURES(r4) + blr + +__init_LPCR: + /* Setup a sane LPCR: + * + * LPES = 0b11 (SRR0/1 used for 0x500) + * PECE = 0b111 + * + * Other bits untouched for now + */ + mfspr r3,SPRN_LPCR + ori r3,r3,(LPCR_LPES0|LPCR_LPES1) + ori r3,r3,(LPCR_PECE0|LPCR_PECE1|LPCR_PECE2) + mtspr SPRN_LPCR,r3 + isync + blr diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c index b9602ee06deb..b65b4908d3c7 100644 --- a/arch/powerpc/kernel/cputable.c +++ b/arch/powerpc/kernel/cputable.c @@ -423,6 +423,8 @@ static struct cpu_spec __initdata cpu_specs[] = { .dcache_bsize = 128, .oprofile_type = PPC_OPROFILE_POWER4, .oprofile_cpu_type = "ppc64/ibm-compat-v1", + .cpu_setup = __setup_cpu_power7, + .cpu_restore = __restore_cpu_power7, .platform = "power7", }, { /* Power7 */ @@ -439,6 +441,8 @@ static struct cpu_spec __initdata cpu_specs[] = { .pmc_type = PPC_PMC_IBM, .oprofile_cpu_type = "ppc64/power7", .oprofile_type = PPC_OPROFILE_POWER4, + .cpu_setup = __setup_cpu_power7, + .cpu_restore = __restore_cpu_power7, .platform = "power7", }, { /* Power7+ */ @@ -455,6 +459,8 @@ static struct cpu_spec __initdata cpu_specs[] = { .pmc_type = PPC_PMC_IBM, .oprofile_cpu_type = "ppc64/power7", .oprofile_type = PPC_OPROFILE_POWER4, + .cpu_setup = __setup_cpu_power7, + .cpu_restore = __restore_cpu_power7, .platform = "power7+", }, { /* Cell Broadband Engine */ -- cgit v1.2.3 From 2dd60d79e0202628a47af9812a84d502cc63628c Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Thu, 20 Jan 2011 17:50:21 +1100 Subject: powerpc: In HV mode, use HSPRG0 for PACA When running in Hypervisor mode (arch 2.06 or later), we store the PACA in HSPRG0 instead of SPRG1. The architecture specifies that SPRGs may be lost during a "nap" power management operation (though they aren't currently on POWER7) and this enables use of SPRG1 by KVM guests. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/entry_64.S | 4 ++-- arch/powerpc/kernel/exceptions-64s.S | 8 ++++---- arch/powerpc/kernel/head_64.S | 4 ++-- arch/powerpc/kernel/paca.c | 13 ++++++++++++- 4 files changed, 20 insertions(+), 9 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S index d82878c4daa6..dbf5bfafd7bc 100644 --- a/arch/powerpc/kernel/entry_64.S +++ b/arch/powerpc/kernel/entry_64.S @@ -838,7 +838,7 @@ _GLOBAL(enter_rtas) _STATIC(rtas_return_loc) /* relocation is off at this point */ - mfspr r4,SPRN_SPRG_PACA /* Get PACA */ + GET_PACA(r4) clrldi r4,r4,2 /* convert to realmode address */ bcl 20,31,$+4 @@ -869,7 +869,7 @@ _STATIC(rtas_restore_regs) REST_8GPRS(14, r1) /* Restore the non-volatiles */ REST_10GPRS(22, r1) /* ditto */ - mfspr r13,SPRN_SPRG_PACA + GET_PACA(r13) ld r4,_CCR(r1) mtcr r4 diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S index aeb739e18769..6784bf7090f6 100644 --- a/arch/powerpc/kernel/exceptions-64s.S +++ b/arch/powerpc/kernel/exceptions-64s.S @@ -53,7 +53,7 @@ data_access_pSeries: DO_KVM 0x300 mtspr SPRN_SPRG_SCRATCH0,r13 BEGIN_FTR_SECTION - mfspr r13,SPRN_SPRG_PACA + GET_PACA(r13) std r9,PACA_EXSLB+EX_R9(r13) std r10,PACA_EXSLB+EX_R10(r13) mfspr r10,SPRN_DAR @@ -82,7 +82,7 @@ data_access_slb_pSeries: HMT_MEDIUM DO_KVM 0x380 mtspr SPRN_SPRG_SCRATCH0,r13 - mfspr r13,SPRN_SPRG_PACA /* get paca address into r13 */ + GET_PACA(r13) std r3,PACA_EXSLB+EX_R3(r13) mfspr r3,SPRN_DAR std r9,PACA_EXSLB+EX_R9(r13) /* save r9 - r12 */ @@ -121,7 +121,7 @@ instruction_access_slb_pSeries: HMT_MEDIUM DO_KVM 0x480 mtspr SPRN_SPRG_SCRATCH0,r13 - mfspr r13,SPRN_SPRG_PACA /* get paca address into r13 */ + GET_PACA(r13) std r3,PACA_EXSLB+EX_R3(r13) mfspr r3,SPRN_SRR0 /* SRR0 is faulting address */ std r9,PACA_EXSLB+EX_R9(r13) /* save r9 - r12 */ @@ -165,7 +165,7 @@ BEGIN_FTR_SECTION beq- 1f END_FTR_SECTION_IFSET(CPU_FTR_REAL_LE) mr r9,r13 - mfspr r13,SPRN_SPRG_PACA + GET_PACA(r13) mfspr r11,SPRN_SRR0 ld r12,PACAKBASE(r13) ld r10,PACAKMSR(r13) diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S index 3a319f9c9d3e..39a40400f3f2 100644 --- a/arch/powerpc/kernel/head_64.S +++ b/arch/powerpc/kernel/head_64.S @@ -228,7 +228,7 @@ generic_secondary_common_init: mr r3,r24 /* not found, copy phys to r3 */ b .kexec_wait /* next kernel might do better */ -2: mtspr SPRN_SPRG_PACA,r13 /* Save vaddr of paca in an SPRG */ +2: SET_PACA(r13) #ifdef CONFIG_PPC_BOOK3E addi r12,r13,PACA_EXTLB /* and TLB exc frame in another */ mtspr SPRN_SPRG_TLB_EXFRAME,r12 @@ -534,7 +534,7 @@ _GLOBAL(pmac_secondary_start) ld r4,0(r4) /* Get base vaddr of paca array */ mulli r13,r24,PACA_SIZE /* Calculate vaddr of right paca */ add r13,r13,r4 /* for this processor. */ - mtspr SPRN_SPRG_PACA,r13 /* Save vaddr of paca in an SPRG*/ + SET_PACA(r13) /* Save vaddr of paca in an SPRG*/ /* Mark interrupts soft and hard disabled (they might be enabled * in the PACA when doing hotplug) diff --git a/arch/powerpc/kernel/paca.c b/arch/powerpc/kernel/paca.c index 10f0aadee95b..102244edecf0 100644 --- a/arch/powerpc/kernel/paca.c +++ b/arch/powerpc/kernel/paca.c @@ -156,11 +156,22 @@ void __init initialise_paca(struct paca_struct *new_paca, int cpu) /* Put the paca pointer into r13 and SPRG_PACA */ void setup_paca(struct paca_struct *new_paca) { + /* Setup r13 */ local_paca = new_paca; - mtspr(SPRN_SPRG_PACA, local_paca); + #ifdef CONFIG_PPC_BOOK3E + /* On Book3E, initialize the TLB miss exception frames */ mtspr(SPRN_SPRG_TLB_EXFRAME, local_paca->extlb); +#else + /* In HV mode, we setup both HPACA and PACA to avoid problems + * if we do a GET_PACA() before the feature fixups have been + * applied + */ + if (cpu_has_feature(CPU_FTR_HVMODE_206)) + mtspr(SPRN_SPRG_HPACA, local_paca); #endif + mtspr(SPRN_SPRG_PACA, local_paca); + } static int __initdata paca_size; -- cgit v1.2.3 From a5d4f3ad3a28cf046836b9bfae61d532b8f77036 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 5 Apr 2011 14:20:31 +1000 Subject: powerpc: Base support for exceptions using HSRR0/1 Pass the register type to the prolog, also provides alternate "HV" version of hardware interrupt (0x500) and adjust LPES accordingly We tag those interrupts by setting bit 0x2 in the trap number Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/cpu_setup_power7.S | 3 ++- arch/powerpc/kernel/exceptions-64s.S | 48 ++++++++++++++++++++++++++-------- 2 files changed, 39 insertions(+), 12 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/cpu_setup_power7.S b/arch/powerpc/kernel/cpu_setup_power7.S index f2b317817c4e..e801ef15d6d0 100644 --- a/arch/powerpc/kernel/cpu_setup_power7.S +++ b/arch/powerpc/kernel/cpu_setup_power7.S @@ -52,13 +52,14 @@ __init_hvmode_206: __init_LPCR: /* Setup a sane LPCR: * - * LPES = 0b11 (SRR0/1 used for 0x500) + * LPES = 0b01 (HSRR0/1 used for 0x500) * PECE = 0b111 * * Other bits untouched for now */ mfspr r3,SPRN_LPCR ori r3,r3,(LPCR_LPES0|LPCR_LPES1) + xori r3,r3, LPCR_LPES0 ori r3,r3,(LPCR_PECE0|LPCR_PECE1|LPCR_PECE2) mtspr SPRN_LPCR,r3 isync diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S index 6784bf7090f6..17f1d6670635 100644 --- a/arch/powerpc/kernel/exceptions-64s.S +++ b/arch/powerpc/kernel/exceptions-64s.S @@ -44,7 +44,7 @@ _machine_check_pSeries: HMT_MEDIUM DO_KVM 0x200 mtspr SPRN_SPRG_SCRATCH0,r13 /* save r13 */ - EXCEPTION_PROLOG_PSERIES(PACA_EXMC, machine_check_common) + EXCEPTION_PROLOG_PSERIES(PACA_EXMC, machine_check_common, EXC_STD) . = 0x300 .globl data_access_pSeries @@ -71,9 +71,9 @@ BEGIN_FTR_SECTION std r10,PACA_EXGEN+EX_R10(r13) std r11,PACA_EXGEN+EX_R9(r13) std r12,PACA_EXGEN+EX_R13(r13) - EXCEPTION_PROLOG_PSERIES_1(data_access_common) + EXCEPTION_PROLOG_PSERIES_1(data_access_common, EXC_STD) FTR_SECTION_ELSE - EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, data_access_common) + EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, data_access_common, EXC_STD) ALT_FTR_SECTION_END_IFCLR(CPU_FTR_SLB) . = 0x380 @@ -147,11 +147,24 @@ instruction_access_slb_pSeries: bctr #endif - MASKABLE_EXCEPTION_PSERIES(0x500, hardware_interrupt) + . = 0x500; + .globl hardware_interrupt_pSeries +hardware_interrupt_pSeries: + BEGIN_FTR_SECTION + MASKABLE_EXCEPTION_PSERIES(0x500, hardware_interrupt, EXC_STD) + FTR_SECTION_ELSE + MASKABLE_EXCEPTION_PSERIES(0x502, hardware_interrupt, EXC_HV) + ALT_FTR_SECTION_END_IFCLR(CPU_FTR_HVMODE_206) + STD_EXCEPTION_PSERIES(0x600, alignment) STD_EXCEPTION_PSERIES(0x700, program_check) STD_EXCEPTION_PSERIES(0x800, fp_unavailable) - MASKABLE_EXCEPTION_PSERIES(0x900, decrementer) + + . = 0x900; + .globl decrementer_pSeries +decrementer_pSeries: + MASKABLE_EXCEPTION_PSERIES(0x900, decrementer, EXC_STD) + STD_EXCEPTION_PSERIES(0xa00, trap_0a) STD_EXCEPTION_PSERIES(0xb00, trap_0b) @@ -207,15 +220,15 @@ vsx_unavailable_pSeries_1: b vsx_unavailable_pSeries #ifdef CONFIG_CBE_RAS - HSTD_EXCEPTION_PSERIES(0x1200, cbe_system_error) + HSTD_EXCEPTION_PSERIES(0x1202, cbe_system_error) #endif /* CONFIG_CBE_RAS */ STD_EXCEPTION_PSERIES(0x1300, instruction_breakpoint) #ifdef CONFIG_CBE_RAS - HSTD_EXCEPTION_PSERIES(0x1600, cbe_maintenance) + HSTD_EXCEPTION_PSERIES(0x1602, cbe_maintenance) #endif /* CONFIG_CBE_RAS */ STD_EXCEPTION_PSERIES(0x1700, altivec_assist) #ifdef CONFIG_CBE_RAS - HSTD_EXCEPTION_PSERIES(0x1800, cbe_thermal) + HSTD_EXCEPTION_PSERIES(0x1802, cbe_thermal) #endif /* CONFIG_CBE_RAS */ . = 0x3000 @@ -244,13 +257,26 @@ masked_interrupt: rfid b . +masked_Hinterrupt: + stb r10,PACAHARDIRQEN(r13) + mtcrf 0x80,r9 + ld r9,PACA_EXGEN+EX_R9(r13) + mfspr r10,SPRN_HSRR1 + rldicl r10,r10,48,1 /* clear MSR_EE */ + rotldi r10,r10,16 + mtspr SPRN_HSRR1,r10 + ld r10,PACA_EXGEN+EX_R10(r13) + mfspr r13,SPRN_SPRG_HSCRATCH0 + hrfid + b . + .align 7 do_stab_bolted_pSeries: std r11,PACA_EXSLB+EX_R11(r13) std r12,PACA_EXSLB+EX_R12(r13) mfspr r10,SPRN_SPRG_SCRATCH0 std r10,PACA_EXSLB+EX_R13(r13) - EXCEPTION_PROLOG_PSERIES_1(.do_stab_bolted) + EXCEPTION_PROLOG_PSERIES_1(.do_stab_bolted, EXC_STD) #ifdef CONFIG_PPC_PSERIES /* @@ -261,14 +287,14 @@ do_stab_bolted_pSeries: system_reset_fwnmi: HMT_MEDIUM mtspr SPRN_SPRG_SCRATCH0,r13 /* save r13 */ - EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, system_reset_common) + EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, system_reset_common, EXC_STD) .globl machine_check_fwnmi .align 7 machine_check_fwnmi: HMT_MEDIUM mtspr SPRN_SPRG_SCRATCH0,r13 /* save r13 */ - EXCEPTION_PROLOG_PSERIES(PACA_EXMC, machine_check_common) + EXCEPTION_PROLOG_PSERIES(PACA_EXMC, machine_check_common, EXC_STD) #endif /* CONFIG_PPC_PSERIES */ -- cgit v1.2.3 From b3e6b5dfcf0974069a8ddcce7dd071120d20d79c Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 5 Apr 2011 14:27:11 +1000 Subject: powerpc: More work to support HV exceptions Rework exception macros a bit to split offset from vector and add some basic support for HDEC, HDSI, HISI and a few more. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/exceptions-64s.S | 92 +++++++++++++++++++++++++----------- 1 file changed, 64 insertions(+), 28 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S index 17f1d6670635..805e20657868 100644 --- a/arch/powerpc/kernel/exceptions-64s.S +++ b/arch/powerpc/kernel/exceptions-64s.S @@ -37,7 +37,7 @@ .globl __start_interrupts __start_interrupts: - STD_EXCEPTION_PSERIES(0x100, system_reset) + STD_EXCEPTION_PSERIES(0x100, 0x100, system_reset) . = 0x200 _machine_check_pSeries: @@ -113,7 +113,7 @@ data_access_slb_pSeries: bctr #endif - STD_EXCEPTION_PSERIES(0x400, instruction_access) + STD_EXCEPTION_PSERIES(0x400, 0x400, instruction_access) . = 0x480 .globl instruction_access_slb_pSeries @@ -147,26 +147,29 @@ instruction_access_slb_pSeries: bctr #endif + /* We open code these as we can't have a ". = x" (even with + * x = "." within a feature section + */ . = 0x500; - .globl hardware_interrupt_pSeries + .globl hardware_interrupt_pSeries; + .globl hardware_interrupt_hv; hardware_interrupt_pSeries: +hardware_interrupt_hv: BEGIN_FTR_SECTION - MASKABLE_EXCEPTION_PSERIES(0x500, hardware_interrupt, EXC_STD) + _MASKABLE_EXCEPTION_PSERIES(0x500, hardware_interrupt, EXC_STD) FTR_SECTION_ELSE - MASKABLE_EXCEPTION_PSERIES(0x502, hardware_interrupt, EXC_HV) + _MASKABLE_EXCEPTION_PSERIES(0x502, hardware_interrupt, EXC_HV) ALT_FTR_SECTION_END_IFCLR(CPU_FTR_HVMODE_206) - STD_EXCEPTION_PSERIES(0x600, alignment) - STD_EXCEPTION_PSERIES(0x700, program_check) - STD_EXCEPTION_PSERIES(0x800, fp_unavailable) + STD_EXCEPTION_PSERIES(0x600, 0x600, alignment) + STD_EXCEPTION_PSERIES(0x700, 0x700, program_check) + STD_EXCEPTION_PSERIES(0x800, 0x800, fp_unavailable) - . = 0x900; - .globl decrementer_pSeries -decrementer_pSeries: - MASKABLE_EXCEPTION_PSERIES(0x900, decrementer, EXC_STD) + MASKABLE_EXCEPTION_PSERIES(0x900, 0x900, decrementer) + MASKABLE_EXCEPTION_HV(0x980, 0x980, decrementer) - STD_EXCEPTION_PSERIES(0xa00, trap_0a) - STD_EXCEPTION_PSERIES(0xb00, trap_0b) + STD_EXCEPTION_PSERIES(0xa00, 0xa00, trap_0a) + STD_EXCEPTION_PSERIES(0xb00, 0xb00, trap_0b) . = 0xc00 .globl system_call_pSeries @@ -196,8 +199,21 @@ END_FTR_SECTION_IFSET(CPU_FTR_REAL_LE) rfid /* return to userspace */ b . - STD_EXCEPTION_PSERIES(0xd00, single_step) - STD_EXCEPTION_PSERIES(0xe00, trap_0e) + STD_EXCEPTION_PSERIES(0xd00, 0xd00, single_step) + + /* At 0xe??? we have a bunch of hypervisor exceptions, we branch + * out of line to handle them + */ + . = 0xe00 + b h_data_storage_hv + . = 0xe20 + b h_instr_storage_hv + . = 0xe40 + b emulation_assist_hv + . = 0xe50 + b hmi_exception_hv + . = 0xe60 + b hmi_exception_hv /* We need to deal with the Altivec unavailable exception * here which is at 0xf20, thus in the middle of the @@ -206,39 +222,42 @@ END_FTR_SECTION_IFSET(CPU_FTR_REAL_LE) */ performance_monitor_pSeries_1: . = 0xf00 - DO_KVM 0xf00 b performance_monitor_pSeries altivec_unavailable_pSeries_1: . = 0xf20 - DO_KVM 0xf20 b altivec_unavailable_pSeries vsx_unavailable_pSeries_1: . = 0xf40 - DO_KVM 0xf40 b vsx_unavailable_pSeries #ifdef CONFIG_CBE_RAS - HSTD_EXCEPTION_PSERIES(0x1202, cbe_system_error) + STD_EXCEPTION_HV(0x1200, 0x1202, cbe_system_error) #endif /* CONFIG_CBE_RAS */ - STD_EXCEPTION_PSERIES(0x1300, instruction_breakpoint) + STD_EXCEPTION_PSERIES(0x1300, 0x1300, instruction_breakpoint) #ifdef CONFIG_CBE_RAS - HSTD_EXCEPTION_PSERIES(0x1602, cbe_maintenance) + STD_EXCEPTION_HV(0x1600, 0x1602, cbe_maintenance) #endif /* CONFIG_CBE_RAS */ - STD_EXCEPTION_PSERIES(0x1700, altivec_assist) + STD_EXCEPTION_PSERIES(0x1700, 0x1700, altivec_assist) #ifdef CONFIG_CBE_RAS - HSTD_EXCEPTION_PSERIES(0x1802, cbe_thermal) + STD_EXCEPTION_HV(0x1800, 0x1802, cbe_thermal) #endif /* CONFIG_CBE_RAS */ . = 0x3000 -/*** pSeries interrupt support ***/ +/*** Out of line interrupts support ***/ + + /* moved from 0xe00 */ + STD_EXCEPTION_HV(., 0xe00, h_data_storage) + STD_EXCEPTION_HV(., 0xe20, h_instr_storage) + STD_EXCEPTION_HV(., 0xe40, emulation_assist) + STD_EXCEPTION_HV(., 0xe60, hmi_exception) /* need to flush cache ? */ /* moved from 0xf00 */ - STD_EXCEPTION_PSERIES(., performance_monitor) - STD_EXCEPTION_PSERIES(., altivec_unavailable) - STD_EXCEPTION_PSERIES(., vsx_unavailable) + STD_EXCEPTION_PSERIES(., 0xf00, performance_monitor) + STD_EXCEPTION_PSERIES(., 0xf20, altivec_unavailable) + STD_EXCEPTION_PSERIES(., 0xf40, vsx_unavailable) /* * An interrupt came in while soft-disabled; clear EE in SRR1, @@ -368,6 +387,8 @@ machine_check_common: STD_EXCEPTION_COMMON(0xb00, trap_0b, .unknown_exception) STD_EXCEPTION_COMMON(0xd00, single_step, .single_step_exception) STD_EXCEPTION_COMMON(0xe00, trap_0e, .unknown_exception) + STD_EXCEPTION_COMMON(0xe40, emulation_assist, .program_check_exception) + STD_EXCEPTION_COMMON(0xe60, hmi_exception, .unknown_exception) STD_EXCEPTION_COMMON_IDLE(0xf00, performance_monitor, .performance_monitor_exception) STD_EXCEPTION_COMMON(0x1300, instruction_breakpoint, .instruction_breakpoint_exception) #ifdef CONFIG_ALTIVEC @@ -445,6 +466,19 @@ data_access_common: li r5,0x300 b .do_hash_page /* Try to handle as hpte fault */ + .align 7 + .globl h_data_storage_common +h_data_storage_common: + mfspr r10,SPRN_HDAR + std r10,PACA_EXGEN+EX_DAR(r13) + mfspr r10,SPRN_HDSISR + stw r10,PACA_EXGEN+EX_DSISR(r13) + EXCEPTION_PROLOG_COMMON(0xe00, PACA_EXGEN) + bl .save_nvgprs + addi r3,r1,STACK_FRAME_OVERHEAD + bl .unknown_exception + b .ret_from_except + .align 7 .globl instruction_access_common instruction_access_common: @@ -454,6 +488,8 @@ instruction_access_common: li r5,0x400 b .do_hash_page /* Try to handle as hpte fault */ + STD_EXCEPTION_COMMON(0xe20, h_instr_storage, .unknown_exception) + /* * Here is the common SLB miss user that is used when going to virtual * mode for SLB misses, that is currently not used -- cgit v1.2.3 From 673b189a2e3353061fa8c49515d1014dab6ad9b9 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Tue, 5 Apr 2011 13:59:58 +1000 Subject: powerpc: Always use SPRN_SPRG_HSCRATCH0 when running in HV mode This uses feature sections to arrange that we always use HSPRG1 as the scratch register in the interrupt entry code rather than SPRG2 when we're running in hypervisor mode on POWER7. This will ensure that we don't trash the guest's SPRG2 when we are running KVM guests. To simplify the code, we define GET_SCRATCH0() and SET_SCRATCH0() macros like the GET_PACA/SET_PACA macros. Signed-off-by: Paul Mackerras Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/exceptions-64s.S | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S index 805e20657868..e513c1d35b2a 100644 --- a/arch/powerpc/kernel/exceptions-64s.S +++ b/arch/powerpc/kernel/exceptions-64s.S @@ -43,7 +43,7 @@ __start_interrupts: _machine_check_pSeries: HMT_MEDIUM DO_KVM 0x200 - mtspr SPRN_SPRG_SCRATCH0,r13 /* save r13 */ + SET_SCRATCH0(r13) EXCEPTION_PROLOG_PSERIES(PACA_EXMC, machine_check_common, EXC_STD) . = 0x300 @@ -51,7 +51,7 @@ _machine_check_pSeries: data_access_pSeries: HMT_MEDIUM DO_KVM 0x300 - mtspr SPRN_SPRG_SCRATCH0,r13 + SET_SCRATCH0(r13) BEGIN_FTR_SECTION GET_PACA(r13) std r9,PACA_EXSLB+EX_R9(r13) @@ -67,7 +67,7 @@ BEGIN_FTR_SECTION std r11,PACA_EXGEN+EX_R11(r13) ld r11,PACA_EXSLB+EX_R9(r13) std r12,PACA_EXGEN+EX_R12(r13) - mfspr r12,SPRN_SPRG_SCRATCH0 + GET_SCRATCH0(r12) std r10,PACA_EXGEN+EX_R10(r13) std r11,PACA_EXGEN+EX_R9(r13) std r12,PACA_EXGEN+EX_R13(r13) @@ -81,7 +81,7 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_SLB) data_access_slb_pSeries: HMT_MEDIUM DO_KVM 0x380 - mtspr SPRN_SPRG_SCRATCH0,r13 + SET_SCRATCH0(r13) GET_PACA(r13) std r3,PACA_EXSLB+EX_R3(r13) mfspr r3,SPRN_DAR @@ -95,7 +95,7 @@ data_access_slb_pSeries: std r10,PACA_EXSLB+EX_R10(r13) std r11,PACA_EXSLB+EX_R11(r13) std r12,PACA_EXSLB+EX_R12(r13) - mfspr r10,SPRN_SPRG_SCRATCH0 + GET_SCRATCH0(r10) std r10,PACA_EXSLB+EX_R13(r13) mfspr r12,SPRN_SRR1 /* and SRR1 */ #ifndef CONFIG_RELOCATABLE @@ -120,7 +120,7 @@ data_access_slb_pSeries: instruction_access_slb_pSeries: HMT_MEDIUM DO_KVM 0x480 - mtspr SPRN_SPRG_SCRATCH0,r13 + SET_SCRATCH0(r13) GET_PACA(r13) std r3,PACA_EXSLB+EX_R3(r13) mfspr r3,SPRN_SRR0 /* SRR0 is faulting address */ @@ -134,7 +134,7 @@ instruction_access_slb_pSeries: std r10,PACA_EXSLB+EX_R10(r13) std r11,PACA_EXSLB+EX_R11(r13) std r12,PACA_EXSLB+EX_R12(r13) - mfspr r10,SPRN_SPRG_SCRATCH0 + GET_SCRATCH0(r10) std r10,PACA_EXSLB+EX_R13(r13) mfspr r12,SPRN_SRR1 /* and SRR1 */ #ifndef CONFIG_RELOCATABLE @@ -272,7 +272,7 @@ masked_interrupt: rotldi r10,r10,16 mtspr SPRN_SRR1,r10 ld r10,PACA_EXGEN+EX_R10(r13) - mfspr r13,SPRN_SPRG_SCRATCH0 + GET_SCRATCH0(r13) rfid b . @@ -285,7 +285,7 @@ masked_Hinterrupt: rotldi r10,r10,16 mtspr SPRN_HSRR1,r10 ld r10,PACA_EXGEN+EX_R10(r13) - mfspr r13,SPRN_SPRG_HSCRATCH0 + GET_SCRATCH0(r13) hrfid b . @@ -293,7 +293,7 @@ masked_Hinterrupt: do_stab_bolted_pSeries: std r11,PACA_EXSLB+EX_R11(r13) std r12,PACA_EXSLB+EX_R12(r13) - mfspr r10,SPRN_SPRG_SCRATCH0 + GET_SCRATCH0(r10) std r10,PACA_EXSLB+EX_R13(r13) EXCEPTION_PROLOG_PSERIES_1(.do_stab_bolted, EXC_STD) @@ -305,14 +305,14 @@ do_stab_bolted_pSeries: .align 7 system_reset_fwnmi: HMT_MEDIUM - mtspr SPRN_SPRG_SCRATCH0,r13 /* save r13 */ + SET_SCRATCH0(r13) /* save r13 */ EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, system_reset_common, EXC_STD) .globl machine_check_fwnmi .align 7 machine_check_fwnmi: HMT_MEDIUM - mtspr SPRN_SPRG_SCRATCH0,r13 /* save r13 */ + SET_SCRATCH0(r13) /* save r13 */ EXCEPTION_PROLOG_PSERIES(PACA_EXMC, machine_check_common, EXC_STD) #endif /* CONFIG_PPC_PSERIES */ @@ -327,7 +327,7 @@ slb_miss_user_pseries: std r10,PACA_EXGEN+EX_R10(r13) std r11,PACA_EXGEN+EX_R11(r13) std r12,PACA_EXGEN+EX_R12(r13) - mfspr r10,SPRG_SCRATCH0 + GET_SCRATCH0(r10) ld r11,PACA_EXSLB+EX_R9(r13) ld r12,PACA_EXSLB+EX_R3(r13) std r10,PACA_EXGEN+EX_R13(r13) -- cgit v1.2.3 From 895796a8ab548fe03b6fea410dcb1b86e1913708 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Mon, 24 Jan 2011 13:25:55 +1100 Subject: powerpc: Initialize LPCR:DPFD on power7 to a sane default This sets the default data stream prefetch size for operating systems that don't set their own value in DSCR. We use 4 which is "medium". Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/cpu_setup_power7.S | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/cpu_setup_power7.S b/arch/powerpc/kernel/cpu_setup_power7.S index e801ef15d6d0..2390f6f7c478 100644 --- a/arch/powerpc/kernel/cpu_setup_power7.S +++ b/arch/powerpc/kernel/cpu_setup_power7.S @@ -54,6 +54,7 @@ __init_LPCR: * * LPES = 0b01 (HSRR0/1 used for 0x500) * PECE = 0b111 + * DPFD = 4 * * Other bits untouched for now */ @@ -61,6 +62,12 @@ __init_LPCR: ori r3,r3,(LPCR_LPES0|LPCR_LPES1) xori r3,r3, LPCR_LPES0 ori r3,r3,(LPCR_PECE0|LPCR_PECE1|LPCR_PECE2) + li r5,7 + sldi r5,r5,LPCR_DPFD_SH + andc r3,r3,r5 + li r5,4 + sldi r5,r5,LPCR_DPFD_SH + or r3,r3,r5 mtspr SPRN_LPCR,r3 isync blr -- cgit v1.2.3 From b144871cb5f2c268e94258ae8f1ec810db2e1120 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 1 Mar 2011 15:46:09 +1100 Subject: powerpc: Initialize TLB and LPID register on HV mode Power7 In case entry from the bootloader isn't "clean" Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/cpu_setup_power7.S | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/cpu_setup_power7.S b/arch/powerpc/kernel/cpu_setup_power7.S index 2390f6f7c478..4f9a93fcfe07 100644 --- a/arch/powerpc/kernel/cpu_setup_power7.S +++ b/arch/powerpc/kernel/cpu_setup_power7.S @@ -25,7 +25,10 @@ _GLOBAL(__setup_cpu_power7) bl __init_hvmode_206 mtlr r11 beqlr + li r0,0 + mtspr SPRN_LPID,r0 bl __init_LPCR + bl __init_TLB mtlr r11 blr @@ -34,7 +37,10 @@ _GLOBAL(__restore_cpu_power7) mfmsr r3 rldicl. r0,r3,4,63 beqlr + li r0,0 + mtspr SPRN_LPID,r0 bl __init_LPCR + bl __init_TLB mtlr r11 blr @@ -71,3 +77,15 @@ __init_LPCR: mtspr SPRN_LPCR,r3 isync blr + +__init_TLB: + /* Clear the TLB */ + li r6,128 + mtctr r6 + li r7,0xc00 /* IS field = 0b11 */ + ptesync +2: tlbiel r7 + addi r7,r7,0x1000 + bdnz 2b + ptesync +1: blr -- cgit v1.2.3 From ad0693ee722b93b63a89c845e99513f242e43aa6 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 1 Feb 2011 12:13:09 +1100 Subject: powerpc: Call CPU ->restore callback earlier on secondary CPUs We do it before we loop on the PACA start flag. This way, we get a chance to set critical SPRs on all CPUs before Linux tries to start them up, which avoids problems when changing some bits such as LPCR bits that need to be identical on all threads of a core or similar things like that. Ideally, some of that should also be done before the MMU is enabled, but that's a separate issue which would require moving some of the SMP startup code earlier, let's not get there for now, it works with that change alone. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/head_64.S | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S index 39a40400f3f2..95944278380c 100644 --- a/arch/powerpc/kernel/head_64.S +++ b/arch/powerpc/kernel/head_64.S @@ -236,17 +236,6 @@ generic_secondary_common_init: /* From now on, r24 is expected to be logical cpuid */ mr r24,r5 -3: HMT_LOW - lbz r23,PACAPROCSTART(r13) /* Test if this processor should */ - /* start. */ - -#ifndef CONFIG_SMP - b 3b /* Never go on non-SMP */ -#else - cmpwi 0,r23,0 - beq 3b /* Loop until told to go */ - - sync /* order paca.run and cur_cpu_spec */ /* See if we need to call a cpu state restore handler */ LOAD_REG_ADDR(r23, cur_cpu_spec) @@ -258,6 +247,17 @@ generic_secondary_common_init: mtctr r23 bctrl +3: HMT_LOW + lbz r23,PACAPROCSTART(r13) /* Test if this processor should */ + /* start. */ +#ifndef CONFIG_SMP + b 3b /* Never go on non-SMP */ +#else + cmpwi 0,r23,0 + beq 3b /* Loop until told to go */ + + sync /* order paca.run and cur_cpu_spec */ + 4: /* Create a temp kernel stack for use before relocation is on. */ ld r1,PACAEMERGSP(r13) subi r1,r1,STACK_FRAME_OVERHEAD -- cgit v1.2.3 From 9d07bc841c9779b4d7902e417f4e509996ce805d Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 16 Mar 2011 14:54:35 +1100 Subject: powerpc: Properly handshake CPUs going out of boot spin loop We need to wait a bit for them to have done their CPU setup or we might end up with translation and EE on with different LPCR values between threads Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/head_64.S | 18 +++++++++++++----- arch/powerpc/kernel/prom.c | 27 ++++++++++----------------- arch/powerpc/kernel/setup_32.c | 1 + arch/powerpc/kernel/setup_64.c | 13 ++++++++++++- 4 files changed, 36 insertions(+), 23 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S index 95944278380c..0700e1135c91 100644 --- a/arch/powerpc/kernel/head_64.S +++ b/arch/powerpc/kernel/head_64.S @@ -242,23 +242,31 @@ generic_secondary_common_init: ld r23,0(r23) ld r23,CPU_SPEC_RESTORE(r23) cmpdi 0,r23,0 - beq 4f + beq 3f ld r23,0(r23) mtctr r23 bctrl -3: HMT_LOW +3: LOAD_REG_ADDR(r3, boot_cpu_count) /* Decrement boot_cpu_count */ + lwarx r4,0,r3 + subi r4,r4,1 + stwcx. r4,0,r3 + bne 3b + isync + +4: HMT_LOW lbz r23,PACAPROCSTART(r13) /* Test if this processor should */ /* start. */ #ifndef CONFIG_SMP - b 3b /* Never go on non-SMP */ + b 4b /* Never go on non-SMP */ #else cmpwi 0,r23,0 - beq 3b /* Loop until told to go */ + beq 4b /* Loop until told to go */ sync /* order paca.run and cur_cpu_spec */ + isync /* In case code patching happened */ -4: /* Create a temp kernel stack for use before relocation is on. */ + /* Create a temp kernel stack for use before relocation is on. */ ld r1,PACAEMERGSP(r13) subi r1,r1,STACK_FRAME_OVERHEAD diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c index e74fa12afc82..c391dc4c8bad 100644 --- a/arch/powerpc/kernel/prom.c +++ b/arch/powerpc/kernel/prom.c @@ -268,13 +268,12 @@ static int __init early_init_dt_scan_cpus(unsigned long node, const char *uname, int depth, void *data) { - static int logical_cpuid = 0; char *type = of_get_flat_dt_prop(node, "device_type", NULL); const u32 *prop; const u32 *intserv; int i, nthreads; unsigned long len; - int found = 0; + int found = -1; /* We are scanning "cpu" nodes only */ if (type == NULL || strcmp(type, "cpu") != 0) @@ -299,11 +298,8 @@ static int __init early_init_dt_scan_cpus(unsigned long node, * booted proc. */ if (initial_boot_params && initial_boot_params->version >= 2) { - if (intserv[i] == - initial_boot_params->boot_cpuid_phys) { - found = 1; - break; - } + if (intserv[i] == initial_boot_params->boot_cpuid_phys) + found = boot_cpu_count; } else { /* * Check if it's the boot-cpu, set it's hw index now, @@ -311,23 +307,20 @@ static int __init early_init_dt_scan_cpus(unsigned long node, * off secondary threads. */ if (of_get_flat_dt_prop(node, - "linux,boot-cpu", NULL) != NULL) { - found = 1; - break; - } + "linux,boot-cpu", NULL) != NULL) + found = boot_cpu_count; } - #ifdef CONFIG_SMP /* logical cpu id is always 0 on UP kernels */ - logical_cpuid++; + boot_cpu_count++; #endif } - if (found) { - DBG("boot cpu: logical %d physical %d\n", logical_cpuid, + if (found >= 0) { + DBG("boot cpu: logical %d physical %d\n", found, intserv[i]); - boot_cpuid = logical_cpuid; - set_hard_smp_processor_id(boot_cpuid, intserv[i]); + boot_cpuid = found; + set_hard_smp_processor_id(found, intserv[i]); /* * PAPR defines "logical" PVR values for cpus that diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c index 1d2fbc905303..620d792b52e4 100644 --- a/arch/powerpc/kernel/setup_32.c +++ b/arch/powerpc/kernel/setup_32.c @@ -48,6 +48,7 @@ extern void bootx_init(unsigned long r4, unsigned long phys); int boot_cpuid = -1; EXPORT_SYMBOL_GPL(boot_cpuid); +int __initdata boot_cpu_count; int boot_cpuid_phys; int smp_hw_index[NR_CPUS]; diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index 5a0401fcaebd..91a5cc5f0d02 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -72,6 +72,7 @@ #endif int boot_cpuid = 0; +int __initdata boot_cpu_count; u64 ppc64_pft_size; /* Pick defaults since we might want to patch instructions @@ -233,6 +234,7 @@ void early_setup_secondary(void) void smp_release_cpus(void) { unsigned long *ptr; + int i; DBG(" -> smp_release_cpus()\n"); @@ -245,7 +247,16 @@ void smp_release_cpus(void) ptr = (unsigned long *)((unsigned long)&__secondary_hold_spinloop - PHYSICAL_START); *ptr = __pa(generic_secondary_smp_init); - mb(); + + /* And wait a bit for them to catch up */ + for (i = 0; i < 100000; i++) { + mb(); + HMT_low(); + if (boot_cpu_count == 0) + break; + udelay(1); + } + DBG("boot_cpu_count = %d\n", boot_cpu_count); DBG(" <- smp_release_cpus()\n"); } -- cgit v1.2.3 From 948cf67c4726cca2fc57533dccadfb54d890689d Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Mon, 24 Jan 2011 18:42:41 +1100 Subject: powerpc: Add NAP mode support on Power7 in HV mode Wakeup comes from the system reset handler with a potential loss of the non-hypervisor CPU state. We save the non-volatile state on the stack and a pointer to it in the PACA, which the system reset handler uses to restore things Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/Makefile | 1 + arch/powerpc/kernel/exceptions-64s.S | 30 ++++++++++- arch/powerpc/kernel/idle_power7.S | 97 ++++++++++++++++++++++++++++++++++++ 3 files changed, 127 insertions(+), 1 deletion(-) create mode 100644 arch/powerpc/kernel/idle_power7.S (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile index 7c6eb4974f25..0fd6273bb8a9 100644 --- a/arch/powerpc/kernel/Makefile +++ b/arch/powerpc/kernel/Makefile @@ -44,6 +44,7 @@ obj-$(CONFIG_PPC_BOOK3E_64) += exceptions-64e.o idle_book3e.o obj-$(CONFIG_PPC64) += vdso64/ obj-$(CONFIG_ALTIVEC) += vecemu.o obj-$(CONFIG_PPC_970_NAP) += idle_power4.o +obj-$(CONFIG_PPC_P7_NAP) += idle_power7.o obj-$(CONFIG_PPC_OF) += of_platform.o prom_parse.o obj-$(CONFIG_PPC_CLOCK) += clock.o procfs-y := proc_powerpc.o diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S index e513c1d35b2a..ad06333631ac 100644 --- a/arch/powerpc/kernel/exceptions-64s.S +++ b/arch/powerpc/kernel/exceptions-64s.S @@ -37,7 +37,35 @@ .globl __start_interrupts __start_interrupts: - STD_EXCEPTION_PSERIES(0x100, 0x100, system_reset) + .globl system_reset_pSeries; +system_reset_pSeries: + HMT_MEDIUM; + DO_KVM 0x100; + SET_SCRATCH0(r13) +#ifdef CONFIG_PPC_P7_NAP +BEGIN_FTR_SECTION + /* Running native on arch 2.06 or later, check if we are + * waking up from nap. We only handle no state loss and + * supervisor state loss. We do -not- handle hypervisor + * state loss at this time. + */ + mfspr r13,SPRN_SRR1 + rlwinm r13,r13,47-31,30,31 + cmpwi cr0,r13,1 + bne 1f + b .power7_wakeup_noloss +1: cmpwi cr0,r13,2 + bne 1f + b .power7_wakeup_loss + /* Total loss of HV state is fatal, we could try to use the + * PIR to locate a PACA, then use an emergency stack etc... + * but for now, let's just stay stuck here + */ +1: cmpwi cr0,r13,3 + beq . +END_FTR_SECTION_IFSET(CPU_FTR_HVMODE_206) +#endif /* CONFIG_PPC_P7_NAP */ + EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, system_reset_common, EXC_STD) . = 0x200 _machine_check_pSeries: diff --git a/arch/powerpc/kernel/idle_power7.S b/arch/powerpc/kernel/idle_power7.S new file mode 100644 index 000000000000..f8f0bc7f1d4f --- /dev/null +++ b/arch/powerpc/kernel/idle_power7.S @@ -0,0 +1,97 @@ +/* + * This file contains the power_save function for 970-family CPUs. + * + * 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 +#include +#include +#include +#include +#include +#include +#include + +#undef DEBUG + + .text + +_GLOBAL(power7_idle) + /* Now check if user or arch enabled NAP mode */ + LOAD_REG_ADDRBASE(r3,powersave_nap) + lwz r4,ADDROFF(powersave_nap)(r3) + cmpwi 0,r4,0 + beqlr + + /* NAP is a state loss, we create a regs frame on the + * stack, fill it up with the state we care about and + * stick a pointer to it in PACAR1. We really only + * need to save PC, some CR bits and the NV GPRs, + * but for now an interrupt frame will do. + */ + mflr r0 + std r0,16(r1) + stdu r1,-INT_FRAME_SIZE(r1) + std r0,_LINK(r1) + std r0,_NIP(r1) + +#ifndef CONFIG_SMP + /* Make sure FPU, VSX etc... are flushed as we may lose + * state when going to nap mode + */ + bl .discard_lazy_cpu_state +#endif /* CONFIG_SMP */ + + /* Hard disable interrupts */ + mfmsr r9 + rldicl r9,r9,48,1 + rotldi r9,r9,16 + mtmsrd r9,1 /* hard-disable interrupts */ + li r0,0 + stb r0,PACASOFTIRQEN(r13) /* we'll hard-enable shortly */ + stb r0,PACAHARDIRQEN(r13) + + /* Continue saving state */ + SAVE_GPR(2, r1) + SAVE_NVGPRS(r1) + mfcr r3 + std r3,_CCR(r1) + std r9,_MSR(r1) + std r1,PACAR1(r13) + + /* Magic NAP mode enter sequence */ + std r0,0(r1) + ptesync + ld r0,0(r1) +1: cmp cr0,r0,r0 + bne 1b + PPC_NAP + b . + +_GLOBAL(power7_wakeup_loss) + GET_PACA(r13) + ld r1,PACAR1(r13) + REST_NVGPRS(r1) + REST_GPR(2, r1) + ld r3,_CCR(r1) + ld r4,_MSR(r1) + ld r5,_NIP(r1) + addi r1,r1,INT_FRAME_SIZE + mtcr r3 + mtspr SPRN_SRR1,r4 + mtspr SPRN_SRR0,r5 + rfid + +_GLOBAL(power7_wakeup_noloss) + GET_PACA(r13) + ld r1,PACAR1(r13) + ld r4,_MSR(r1) + ld r5,_NIP(r1) + addi r1,r1,INT_FRAME_SIZE + mtspr SPRN_SRR1,r4 + mtspr SPRN_SRR0,r5 + rfid -- cgit v1.2.3 From dd797738643cd3c2dd9cdff7e4c3a04d318ab23a Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 5 Apr 2011 14:34:58 +1000 Subject: powerpc: Perform an isync to synchronize CPUs coming out of secondary_hold We need to do that to guarantee they see any code change done by dynamic patching during boot. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/head_64.S | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S index 0700e1135c91..6d17c37f22a1 100644 --- a/arch/powerpc/kernel/head_64.S +++ b/arch/powerpc/kernel/head_64.S @@ -147,6 +147,8 @@ __secondary_hold: mtctr r4 mr r3,r24 li r4,0 + /* Make sure that patched code is visible */ + isync bctr #else BUG_OPCODE -- cgit v1.2.3 From af2771493a1bf79cd9a1ab4f30327c428b5bd67c Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 6 Apr 2011 10:51:17 +1000 Subject: powerpc: Improve prom_printf() Adds the ability to print decimal numbers and adds some more format string variants Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/prom_init.c | 26 +++++++++++++++++++++++++- 1 file changed, 25 insertions(+), 1 deletion(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c index 941ff4dbc567..7839bd7bfd15 100644 --- a/arch/powerpc/kernel/prom_init.c +++ b/arch/powerpc/kernel/prom_init.c @@ -335,6 +335,7 @@ static void __init prom_printf(const char *format, ...) const char *p, *q, *s; va_list args; unsigned long v; + long vs; struct prom_t *_prom = &RELOC(prom); va_start(args, format); @@ -368,12 +369,35 @@ static void __init prom_printf(const char *format, ...) v = va_arg(args, unsigned long); prom_print_hex(v); break; + case 'd': + ++q; + vs = va_arg(args, int); + if (vs < 0) { + prom_print(RELOC("-")); + vs = -vs; + } + prom_print_dec(vs); + break; case 'l': ++q; - if (*q == 'u') { /* '%lu' */ + if (*q == 0) + break; + else if (*q == 'x') { + ++q; + v = va_arg(args, unsigned long); + prom_print_hex(v); + } else if (*q == 'u') { /* '%lu' */ ++q; v = va_arg(args, unsigned long); prom_print_dec(v); + } else if (*q == 'd') { /* %ld */ + ++q; + vs = va_arg(args, long); + if (vs < 0) { + prom_print(RELOC("-")); + vs = -vs; + } + prom_print_dec(vs); } break; } -- cgit v1.2.3 From de300974761d92f71cb583730ac9e1d4eb1b7156 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Mon, 11 Apr 2011 21:46:19 +0000 Subject: powerpc/smp: smp_ops->kick_cpu() should be able to fail When we start a cpu we use smp_ops->kick_cpu(), which currently returns void, it should be able to fail. Convert it to return int, and update all uses. Convert all the current error cases to return -ENOENT, which is what would eventually be returned by __cpu_up() currently when it doesn't detect the cpu as coming up in time. Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/smp.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index cbdbb14be4b0..b6083f4f39b1 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -95,7 +95,7 @@ int smt_enabled_at_boot = 1; static void (*crash_ipi_function_ptr)(struct pt_regs *) = NULL; #ifdef CONFIG_PPC64 -void __devinit smp_generic_kick_cpu(int nr) +int __devinit smp_generic_kick_cpu(int nr) { BUG_ON(nr < 0 || nr >= NR_CPUS); @@ -106,6 +106,8 @@ void __devinit smp_generic_kick_cpu(int nr) */ paca[nr].cpu_start = 1; smp_mb(); + + return 0; } #endif @@ -434,7 +436,11 @@ int __cpuinit __cpu_up(unsigned int cpu) /* wake up cpus */ DBG("smp: kicking cpu %d\n", cpu); - smp_ops->kick_cpu(cpu); + rc = smp_ops->kick_cpu(cpu); + if (rc) { + pr_err("smp: failed starting cpu %d (rc %d)\n", cpu, rc); + return rc; + } /* * wait to see if the cpu made a callin (is actually up). -- cgit v1.2.3 From 76b4eda866c4936af8d696f040abea56bf688e16 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Thu, 14 Apr 2011 22:32:01 +0000 Subject: powerpc: Add A2 cpu support Add the cputable entry, regs and setup & restore entries for the PowerPC A2 core. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/Makefile | 1 + arch/powerpc/kernel/cpu_setup_a2.S | 114 +++++++++++++++++++++++++++++++++++++ arch/powerpc/kernel/cputable.c | 25 +++++++- 3 files changed, 138 insertions(+), 2 deletions(-) create mode 100644 arch/powerpc/kernel/cpu_setup_a2.S (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile index 0fd6273bb8a9..058bc8bac488 100644 --- a/arch/powerpc/kernel/Makefile +++ b/arch/powerpc/kernel/Makefile @@ -41,6 +41,7 @@ obj-$(CONFIG_PPC_BOOK3S_64) += cpu_setup_ppc970.o cpu_setup_pa6t.o obj-$(CONFIG_PPC_BOOK3S_64) += cpu_setup_power7.o obj64-$(CONFIG_RELOCATABLE) += reloc_64.o obj-$(CONFIG_PPC_BOOK3E_64) += exceptions-64e.o idle_book3e.o +obj-$(CONFIG_PPC_A2) += cpu_setup_a2.o obj-$(CONFIG_PPC64) += vdso64/ obj-$(CONFIG_ALTIVEC) += vecemu.o obj-$(CONFIG_PPC_970_NAP) += idle_power4.o diff --git a/arch/powerpc/kernel/cpu_setup_a2.S b/arch/powerpc/kernel/cpu_setup_a2.S new file mode 100644 index 000000000000..7f818feaa7a5 --- /dev/null +++ b/arch/powerpc/kernel/cpu_setup_a2.S @@ -0,0 +1,114 @@ +/* + * A2 specific assembly support code + * + * Copyright 2009 Ben Herrenschmidt, 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 +#include +#include +#include +#include +#include +#include + +/* + * Disable thdid and class fields in ERATs to bump PID to full 14 bits capacity. + * This also prevents external LPID accesses but that isn't a problem when not a + * guest. Under PV, this setting will be ignored and MMUCR will return the right + * number of PID bits we can use. + */ +#define MMUCR1_EXTEND_PID \ + (MMUCR1_ICTID | MMUCR1_ITTID | MMUCR1_DCTID | \ + MMUCR1_DTTID | MMUCR1_DCCD) + +/* + * Use extended PIDs if enabled. + * Don't clear the ERATs on context sync events and enable I & D LRU. + * Enable ERAT back invalidate when tlbwe overwrites an entry. + */ +#define INITIAL_MMUCR1 \ + (MMUCR1_EXTEND_PID | MMUCR1_CSINV_NEVER | MMUCR1_IRRE | \ + MMUCR1_DRRE | MMUCR1_TLBWE_BINV) + +_GLOBAL(__setup_cpu_a2) + /* Some of these are actually thread local and some are + * core local but doing it always won't hurt + */ + +#ifdef CONFIG_PPC_WSP_COPRO + /* Make sure ACOP starts out as zero */ + li r3,0 + mtspr SPRN_ACOP,r3 + + /* Enable icswx instruction */ + mfspr r3,SPRN_A2_CCR2 + ori r3,r3,A2_CCR2_ENABLE_ICSWX + mtspr SPRN_A2_CCR2,r3 + + /* Unmask all CTs in HACOP */ + li r3,-1 + mtspr SPRN_HACOP,r3 +#endif /* CONFIG_PPC_WSP_COPRO */ + + /* Enable doorbell */ + mfspr r3,SPRN_A2_CCR2 + oris r3,r3,A2_CCR2_ENABLE_PC@h + mtspr SPRN_A2_CCR2,r3 + isync + + /* Setup CCR0 to disable power saving for now as it's busted + * in the current implementations. Setup CCR1 to wake on + * interrupts normally (we write the default value but who + * knows what FW may have clobbered...) + */ + li r3,0 + mtspr SPRN_A2_CCR0, r3 + LOAD_REG_IMMEDIATE(r3,0x0f0f0f0f) + mtspr SPRN_A2_CCR1, r3 + + /* Initialise MMUCR1 */ + lis r3,INITIAL_MMUCR1@h + ori r3,r3,INITIAL_MMUCR1@l + mtspr SPRN_MMUCR1,r3 + + /* Set MMUCR2 to enable 4K, 64K, 1M, 16M and 1G pages */ + LOAD_REG_IMMEDIATE(r3, 0x000a7531) + mtspr SPRN_MMUCR2,r3 + + /* Set MMUCR3 to write all thids bit to the TLB */ + LOAD_REG_IMMEDIATE(r3, 0x0000000f) + mtspr SPRN_MMUCR3,r3 + + /* Don't do ERAT stuff if running guest mode */ + mfmsr r3 + andis. r0,r3,MSR_GS@h + bne 1f + + /* Now set the I-ERAT watermark to 15 */ + lis r4,(MMUCR0_TLBSEL_I|MMUCR0_ECL)@h + mtspr SPRN_MMUCR0, r4 + li r4,A2_IERAT_SIZE-1 + PPC_ERATWE(r4,r4,3) + + /* Now set the D-ERAT watermark to 31 */ + lis r4,(MMUCR0_TLBSEL_D|MMUCR0_ECL)@h + mtspr SPRN_MMUCR0, r4 + li r4,A2_DERAT_SIZE-1 + PPC_ERATWE(r4,r4,3) + + /* And invalidate the beast just in case. That won't get rid of + * a bolted entry though it will be in LRU and so will go away eventually + * but let's not bother for now + */ + PPC_ERATILX(0,0,0) +1: + blr + +_GLOBAL(__restore_cpu_a2) + b __setup_cpu_a2 diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c index b65b4908d3c7..3d7b65ad4962 100644 --- a/arch/powerpc/kernel/cputable.c +++ b/arch/powerpc/kernel/cputable.c @@ -62,10 +62,12 @@ extern void __setup_cpu_745x(unsigned long offset, struct cpu_spec* spec); extern void __setup_cpu_ppc970(unsigned long offset, struct cpu_spec* spec); extern void __setup_cpu_ppc970MP(unsigned long offset, struct cpu_spec* spec); extern void __setup_cpu_pa6t(unsigned long offset, struct cpu_spec* spec); +extern void __setup_cpu_a2(unsigned long offset, struct cpu_spec* spec); extern void __restore_cpu_pa6t(void); extern void __restore_cpu_ppc970(void); extern void __setup_cpu_power7(unsigned long offset, struct cpu_spec* spec); extern void __restore_cpu_power7(void); +extern void __restore_cpu_a2(void); #endif /* CONFIG_PPC64 */ #if defined(CONFIG_E500) extern void __setup_cpu_e5500(unsigned long offset, struct cpu_spec* spec); @@ -2011,7 +2013,26 @@ static struct cpu_spec __initdata cpu_specs[] = { #endif /* CONFIG_PPC32 */ #endif /* CONFIG_E500 */ -#ifdef CONFIG_PPC_BOOK3E_64 +#ifdef CONFIG_PPC_A2 + { /* Standard A2 (>= DD2) + FPU core */ + .pvr_mask = 0xffff0000, + .pvr_value = 0x00480000, + .cpu_name = "A2 (>= DD2)", + .cpu_features = CPU_FTRS_A2, + .cpu_user_features = COMMON_USER_PPC64, + .mmu_features = MMU_FTR_TYPE_3E | MMU_FTR_USE_TLBILX | + MMU_FTR_USE_TLBIVAX_BCAST | + MMU_FTR_LOCK_BCAST_INVAL | + MMU_FTR_USE_TLBRSRV | + MMU_FTR_USE_PAIRED_MAS, + .icache_bsize = 64, + .dcache_bsize = 64, + .num_pmcs = 0, + .cpu_setup = __setup_cpu_a2, + .cpu_restore = __restore_cpu_a2, + .machine_check = machine_check_generic, + .platform = "ppca2", + }, { /* This is a default entry to get going, to be replaced by * a real one at some stage */ @@ -2032,7 +2053,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .machine_check = machine_check_generic, .platform = "power6", }, -#endif +#endif /* CONFIG_PPC_A2 */ }; static struct cpu_spec the_cpu_spec; -- cgit v1.2.3 From ca1769f7a372898f5e3dbb8e4ff53f53f0626ef4 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Thu, 14 Apr 2011 22:32:04 +0000 Subject: powerpc: Index crit/dbg/mcheck stacks using cpu number on 64bit In exc_lvl_ctx_init() we index into the crit/dbg/mcheck stacks using the hard cpu id, but that assumes the hard cpu id is zero based and contiguous. That is not the case on A2. The root of the problem is that the 32bit code has no equivalent of the paca to allow it to do the hard->soft mapping in assembler. Until the 32bit code is updated to handle that, index the stacks using the soft cpu ids on 64bit and hard on 32 bit. Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/irq.c | 26 +++++++++++++++----------- 1 file changed, 15 insertions(+), 11 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index f621b7d2d869..ea09512a68c3 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -397,24 +397,28 @@ struct thread_info *mcheckirq_ctx[NR_CPUS] __read_mostly; void exc_lvl_ctx_init(void) { struct thread_info *tp; - int i, hw_cpu; + int i, cpu_nr; for_each_possible_cpu(i) { - hw_cpu = get_hard_smp_processor_id(i); - memset((void *)critirq_ctx[hw_cpu], 0, THREAD_SIZE); - tp = critirq_ctx[hw_cpu]; - tp->cpu = i; +#ifdef CONFIG_PPC64 + cpu_nr = i; +#else + cpu_nr = get_hard_smp_processor_id(i); +#endif + memset((void *)critirq_ctx[cpu_nr], 0, THREAD_SIZE); + tp = critirq_ctx[cpu_nr]; + tp->cpu = cpu_nr; tp->preempt_count = 0; #ifdef CONFIG_BOOKE - memset((void *)dbgirq_ctx[hw_cpu], 0, THREAD_SIZE); - tp = dbgirq_ctx[hw_cpu]; - tp->cpu = i; + memset((void *)dbgirq_ctx[cpu_nr], 0, THREAD_SIZE); + tp = dbgirq_ctx[cpu_nr]; + tp->cpu = cpu_nr; tp->preempt_count = 0; - memset((void *)mcheckirq_ctx[hw_cpu], 0, THREAD_SIZE); - tp = mcheckirq_ctx[hw_cpu]; - tp->cpu = i; + memset((void *)mcheckirq_ctx[cpu_nr], 0, THREAD_SIZE); + tp = mcheckirq_ctx[cpu_nr]; + tp->cpu = cpu_nr; tp->preempt_count = HARDIRQ_OFFSET; #endif } -- cgit v1.2.3 From 1a51dde139d5305b2592c716c50c005d6ab9624b Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Thu, 14 Apr 2011 22:32:04 +0000 Subject: powerpc/book3e: Use way 3 for linear mapping bolted entry An erratum on A2 can lead to the bolted entry we insert for the linear mapping being evicted, to avoid that write the bolted entry to way 3. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/exceptions-64e.S | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/exceptions-64e.S b/arch/powerpc/kernel/exceptions-64e.S index 9651acc3504a..8fe0fc233f02 100644 --- a/arch/powerpc/kernel/exceptions-64e.S +++ b/arch/powerpc/kernel/exceptions-64e.S @@ -864,8 +864,9 @@ have_hes: * that will have to be made dependent on whether we are running under * a hypervisor I suppose. */ - ori r3,r3,MAS0_HES | MAS0_WQ_ALLWAYS - mtspr SPRN_MAS0,r3 + ori r11,r3,MAS0_WQ_ALLWAYS + oris r11,r11,MAS0_ESEL(3)@h /* Use way 3: workaround A2 erratum 376 */ + mtspr SPRN_MAS0,r11 lis r3,(MAS1_VALID | MAS1_IPROT)@h ori r3,r3,BOOK3E_PAGESZ_1GB << MAS1_TSIZE_SHIFT mtspr SPRN_MAS1,r3 -- cgit v1.2.3 From f0aae3238fc1c28b543cbaaa0e7c5d57685f5f89 Mon Sep 17 00:00:00 2001 From: Jack Miller Date: Thu, 14 Apr 2011 22:32:05 +0000 Subject: powerpc/book3e: Flush IPROT protected TLB entries leftover by firmware When we set up the TLB for ourselves on Book3E, we need to flush out any old mappings established by the firmware or bootloader. At present we attempt this with a tlbilx to flush everything, but this will leave behind any entries with the IPROT bit set. There are several good reason firmware might establish mappings with IPROT, and in fact ePAPR compliant firmwares are required to establish their initial mapped area with IPROT. This patch, therefore adds more complex code to scan through the TLB upon entry and flush away any entries that are not our own. Signed-off-by: Jack Miller Signed-off-by: David Gibson Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/exceptions-64e.S | 45 +++++++++++++++++++++++++++++++++++- 1 file changed, 44 insertions(+), 1 deletion(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/exceptions-64e.S b/arch/powerpc/kernel/exceptions-64e.S index 8fe0fc233f02..23bd83b20be4 100644 --- a/arch/powerpc/kernel/exceptions-64e.S +++ b/arch/powerpc/kernel/exceptions-64e.S @@ -886,8 +886,51 @@ have_hes: bctr 1: /* We are now running at PAGE_OFFSET, clean the TLB of everything - * else (XXX we should scan for bolted crap from the firmware too) + * else (including IPROTed things left by firmware) + * r4 = TLBnCFG + * r3 = current address (more or less) */ + + li r5,0 + mtspr SPRN_MAS6,r5 + tlbsx 0,r3 + + rlwinm r9,r4,0,TLBnCFG_N_ENTRY + rlwinm r10,r4,8,0xff + addi r10,r10,-1 /* Get inner loop mask */ + + li r3,1 + + mfspr r5,SPRN_MAS1 + rlwinm r5,r5,0,(~(MAS1_VALID|MAS1_IPROT)) + + mfspr r6,SPRN_MAS2 + rldicr r6,r6,0,51 /* Extract EPN */ + + mfspr r7,SPRN_MAS0 + rlwinm r7,r7,0,0xffff0fff /* Clear HES and WQ */ + + rlwinm r8,r7,16,0xfff /* Extract ESEL */ + +2: add r4,r3,r8 + and r4,r4,r10 + + rlwimi r7,r4,16,MAS0_ESEL_MASK + + mtspr SPRN_MAS0,r7 + mtspr SPRN_MAS1,r5 + mtspr SPRN_MAS2,r6 + tlbwe + + addi r3,r3,1 + and. r4,r3,r10 + + bne 3f + addis r6,r6,(1<<30)@h +3: + cmpw r3,r9 + blt 2b + PPC_TLBILX(0,0,0) sync isync -- cgit v1.2.3 From efcac6589a277c10060e4be44b9455cf43838dc1 Mon Sep 17 00:00:00 2001 From: Alexey Kardashevskiy Date: Wed, 2 Mar 2011 15:18:48 +0000 Subject: powerpc: Per process DSCR + some fixes (try#4) The DSCR (aka Data Stream Control Register) is supported on some server PowerPC chips and allow some control over the prefetch of data streams. This patch allows the value to be specified per thread by emulating the corresponding mfspr and mtspr instructions. Children of such threads inherit the value. Other threads use a default value that can be specified in sysfs - /sys/devices/system/cpu/dscr_default. If a thread starts with non default value in the sysfs entry, all children threads inherit this non default value even if the sysfs value is changed later. Signed-off-by: Alexey Kardashevskiy Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/asm-offsets.c | 1 + arch/powerpc/kernel/entry_64.S | 15 +++++++++++++++ arch/powerpc/kernel/process.c | 16 ++++++++++++++++ arch/powerpc/kernel/sysfs.c | 38 ++++++++++++++++++++++++++++++++++++++ arch/powerpc/kernel/traps.c | 24 ++++++++++++++++++++++++ 5 files changed, 94 insertions(+) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c index 23e6a93145ab..6887661ac072 100644 --- a/arch/powerpc/kernel/asm-offsets.c +++ b/arch/powerpc/kernel/asm-offsets.c @@ -74,6 +74,7 @@ int main(void) DEFINE(AUDITCONTEXT, offsetof(struct task_struct, audit_context)); DEFINE(SIGSEGV, SIGSEGV); DEFINE(NMI_MASK, NMI_MASK); + DEFINE(THREAD_DSCR, offsetof(struct thread_struct, dscr)); #else DEFINE(THREAD_INFO, offsetof(struct task_struct, stack)); #endif /* CONFIG_PPC64 */ diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S index dbf5bfafd7bc..64693706ebfd 100644 --- a/arch/powerpc/kernel/entry_64.S +++ b/arch/powerpc/kernel/entry_64.S @@ -421,6 +421,12 @@ BEGIN_FTR_SECTION std r24,THREAD_VRSAVE(r3) END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) #endif /* CONFIG_ALTIVEC */ +#ifdef CONFIG_PPC64 +BEGIN_FTR_SECTION + mfspr r25,SPRN_DSCR + std r25,THREAD_DSCR(r3) +END_FTR_SECTION_IFSET(CPU_FTR_DSCR) +#endif and. r0,r0,r22 beq+ 1f andc r22,r22,r0 @@ -522,6 +528,15 @@ BEGIN_FTR_SECTION mtspr SPRN_VRSAVE,r0 /* if G4, restore VRSAVE reg */ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) #endif /* CONFIG_ALTIVEC */ +#ifdef CONFIG_PPC64 +BEGIN_FTR_SECTION + ld r0,THREAD_DSCR(r4) + cmpd r0,r25 + beq 1f + mtspr SPRN_DSCR,r0 +1: +END_FTR_SECTION_IFSET(CPU_FTR_DSCR) +#endif /* r3-r13 are destroyed -- Cort */ REST_8GPRS(14, r1) diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c index f74f355a9617..a01c2d93fd2f 100644 --- a/arch/powerpc/kernel/process.c +++ b/arch/powerpc/kernel/process.c @@ -702,6 +702,8 @@ void prepare_to_copy(struct task_struct *tsk) /* * Copy a thread.. */ +extern unsigned long dscr_default; /* defined in arch/powerpc/kernel/sysfs.c */ + int copy_thread(unsigned long clone_flags, unsigned long usp, unsigned long unused, struct task_struct *p, struct pt_regs *regs) @@ -769,6 +771,20 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, p->thread.ksp_vsid = sp_vsid; } #endif /* CONFIG_PPC_STD_MMU_64 */ +#ifdef CONFIG_PPC64 + if (cpu_has_feature(CPU_FTR_DSCR)) { + if (current->thread.dscr_inherit) { + p->thread.dscr_inherit = 1; + p->thread.dscr = current->thread.dscr; + } else if (0 != dscr_default) { + p->thread.dscr_inherit = 1; + p->thread.dscr = dscr_default; + } else { + p->thread.dscr_inherit = 0; + p->thread.dscr = 0; + } + } +#endif /* * The PPC64 ABI makes use of a TOC to contain function diff --git a/arch/powerpc/kernel/sysfs.c b/arch/powerpc/kernel/sysfs.c index c0d8c2006bf4..f0f2199e64e1 100644 --- a/arch/powerpc/kernel/sysfs.c +++ b/arch/powerpc/kernel/sysfs.c @@ -182,6 +182,41 @@ static SYSDEV_ATTR(mmcra, 0600, show_mmcra, store_mmcra); static SYSDEV_ATTR(spurr, 0600, show_spurr, NULL); static SYSDEV_ATTR(dscr, 0600, show_dscr, store_dscr); static SYSDEV_ATTR(purr, 0600, show_purr, store_purr); + +unsigned long dscr_default = 0; +EXPORT_SYMBOL(dscr_default); + +static ssize_t show_dscr_default(struct sysdev_class *class, + struct sysdev_class_attribute *attr, char *buf) +{ + return sprintf(buf, "%lx\n", dscr_default); +} + +static ssize_t __used store_dscr_default(struct sysdev_class *class, + struct sysdev_class_attribute *attr, const char *buf, + size_t count) +{ + unsigned long val; + int ret = 0; + + ret = sscanf(buf, "%lx", &val); + if (ret != 1) + return -EINVAL; + dscr_default = val; + + return count; +} + +static SYSDEV_CLASS_ATTR(dscr_default, 0600, + show_dscr_default, store_dscr_default); + +static void sysfs_create_dscr_default(void) +{ + int err = 0; + if (cpu_has_feature(CPU_FTR_DSCR)) + err = sysfs_create_file(&cpu_sysdev_class.kset.kobj, + &attr_dscr_default.attr); +} #endif /* CONFIG_PPC64 */ #ifdef HAS_PPC_PMC_PA6T @@ -617,6 +652,9 @@ static int __init topology_init(void) if (cpu_online(cpu)) register_cpu_online(cpu); } +#ifdef CONFIG_PPC64 + sysfs_create_dscr_default(); +#endif /* CONFIG_PPC64 */ return 0; } diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c index 5ddb801bc154..cb71cf29edea 100644 --- a/arch/powerpc/kernel/traps.c +++ b/arch/powerpc/kernel/traps.c @@ -909,6 +909,26 @@ static int emulate_instruction(struct pt_regs *regs) return emulate_isel(regs, instword); } +#ifdef CONFIG_PPC64 + /* Emulate the mfspr rD, DSCR. */ + if (((instword & PPC_INST_MFSPR_DSCR_MASK) == PPC_INST_MFSPR_DSCR) && + cpu_has_feature(CPU_FTR_DSCR)) { + PPC_WARN_EMULATED(mfdscr, regs); + rd = (instword >> 21) & 0x1f; + regs->gpr[rd] = mfspr(SPRN_DSCR); + return 0; + } + /* Emulate the mtspr DSCR, rD. */ + if (((instword & PPC_INST_MTSPR_DSCR_MASK) == PPC_INST_MTSPR_DSCR) && + cpu_has_feature(CPU_FTR_DSCR)) { + PPC_WARN_EMULATED(mtdscr, regs); + rd = (instword >> 21) & 0x1f; + mtspr(SPRN_DSCR, regs->gpr[rd]); + current->thread.dscr_inherit = 1; + return 0; + } +#endif + return -EINVAL; } @@ -1506,6 +1526,10 @@ struct ppc_emulated ppc_emulated = { #ifdef CONFIG_VSX WARN_EMULATED_SETUP(vsx), #endif +#ifdef CONFIG_PPC64 + WARN_EMULATED_SETUP(mfdscr), + WARN_EMULATED_SETUP(mtdscr), +#endif }; u32 ppc_warn_emulated; -- cgit v1.2.3 From 3cc30d0726d258ac336283bcde66a8ab58283b61 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Mon, 11 Apr 2011 21:25:01 +0000 Subject: powerpc/pci: Move IO workarounds to the common kernel dir Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/Makefile | 2 + arch/powerpc/kernel/io-workarounds.c | 184 +++++++++++++++++++++++++++++++++++ 2 files changed, 186 insertions(+) create mode 100644 arch/powerpc/kernel/io-workarounds.c (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile index 058bc8bac488..82e0bed0650d 100644 --- a/arch/powerpc/kernel/Makefile +++ b/arch/powerpc/kernel/Makefile @@ -106,6 +106,8 @@ obj-$(CONFIG_KEXEC) += machine_kexec.o crash.o \ obj-$(CONFIG_AUDIT) += audit.o obj64-$(CONFIG_AUDIT) += compat_audit.o +obj-$(CONFIG_PPC_IO_WORKAROUNDS) += io-workarounds.o + obj-$(CONFIG_DYNAMIC_FTRACE) += ftrace.o obj-$(CONFIG_FUNCTION_GRAPH_TRACER) += ftrace.o obj-$(CONFIG_PERF_EVENTS) += perf_callchain.o diff --git a/arch/powerpc/kernel/io-workarounds.c b/arch/powerpc/kernel/io-workarounds.c new file mode 100644 index 000000000000..7e5845798788 --- /dev/null +++ b/arch/powerpc/kernel/io-workarounds.c @@ -0,0 +1,184 @@ +/* + * Support PCI IO workaround + * + * Copyright (C) 2006 Benjamin Herrenschmidt + * IBM, Corp. + * (C) Copyright 2007-2008 TOSHIBA 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. + */ +#undef DEBUG + +#include + +#include +#include +#include +#include +#include + +#define IOWA_MAX_BUS 8 + +static struct iowa_bus iowa_busses[IOWA_MAX_BUS]; +static unsigned int iowa_bus_count; + +static struct iowa_bus *iowa_pci_find(unsigned long vaddr, unsigned long paddr) +{ + int i, j; + struct resource *res; + unsigned long vstart, vend; + + for (i = 0; i < iowa_bus_count; i++) { + struct iowa_bus *bus = &iowa_busses[i]; + struct pci_controller *phb = bus->phb; + + if (vaddr) { + vstart = (unsigned long)phb->io_base_virt; + vend = vstart + phb->pci_io_size - 1; + if ((vaddr >= vstart) && (vaddr <= vend)) + return bus; + } + + if (paddr) + for (j = 0; j < 3; j++) { + res = &phb->mem_resources[j]; + if (paddr >= res->start && paddr <= res->end) + return bus; + } + } + + return NULL; +} + +struct iowa_bus *iowa_mem_find_bus(const PCI_IO_ADDR addr) +{ + struct iowa_bus *bus; + int token; + + token = PCI_GET_ADDR_TOKEN(addr); + + if (token && token <= iowa_bus_count) + bus = &iowa_busses[token - 1]; + else { + unsigned long vaddr, paddr; + pte_t *ptep; + + vaddr = (unsigned long)PCI_FIX_ADDR(addr); + if (vaddr < PHB_IO_BASE || vaddr >= PHB_IO_END) + return NULL; + + ptep = find_linux_pte(init_mm.pgd, vaddr); + if (ptep == NULL) + paddr = 0; + else + paddr = pte_pfn(*ptep) << PAGE_SHIFT; + bus = iowa_pci_find(vaddr, paddr); + + if (bus == NULL) + return NULL; + } + + return bus; +} + +struct iowa_bus *iowa_pio_find_bus(unsigned long port) +{ + unsigned long vaddr = (unsigned long)pci_io_base + port; + return iowa_pci_find(vaddr, 0); +} + + +#define DEF_PCI_AC_RET(name, ret, at, al, space, aa) \ +static ret iowa_##name at \ +{ \ + struct iowa_bus *bus; \ + bus = iowa_##space##_find_bus(aa); \ + if (bus && bus->ops && bus->ops->name) \ + return bus->ops->name al; \ + return __do_##name al; \ +} + +#define DEF_PCI_AC_NORET(name, at, al, space, aa) \ +static void iowa_##name at \ +{ \ + struct iowa_bus *bus; \ + bus = iowa_##space##_find_bus(aa); \ + if (bus && bus->ops && bus->ops->name) { \ + bus->ops->name al; \ + return; \ + } \ + __do_##name al; \ +} + +#include + +#undef DEF_PCI_AC_RET +#undef DEF_PCI_AC_NORET + +static const struct ppc_pci_io __devinitconst iowa_pci_io = { + +#define DEF_PCI_AC_RET(name, ret, at, al, space, aa) .name = iowa_##name, +#define DEF_PCI_AC_NORET(name, at, al, space, aa) .name = iowa_##name, + +#include + +#undef DEF_PCI_AC_RET +#undef DEF_PCI_AC_NORET + +}; + +static void __iomem *iowa_ioremap(phys_addr_t addr, unsigned long size, + unsigned long flags, void *caller) +{ + struct iowa_bus *bus; + void __iomem *res = __ioremap_caller(addr, size, flags, caller); + int busno; + + bus = iowa_pci_find(0, (unsigned long)addr); + if (bus != NULL) { + busno = bus - iowa_busses; + PCI_SET_ADDR_TOKEN(res, busno + 1); + } + return res; +} + +/* Regist new bus to support workaround */ +void __devinit iowa_register_bus(struct pci_controller *phb, + struct ppc_pci_io *ops, + int (*initfunc)(struct iowa_bus *, void *), void *data) +{ + struct iowa_bus *bus; + struct device_node *np = phb->dn; + + if (iowa_bus_count >= IOWA_MAX_BUS) { + pr_err("IOWA:Too many pci bridges, " + "workarounds disabled for %s\n", np->full_name); + return; + } + + bus = &iowa_busses[iowa_bus_count]; + bus->phb = phb; + bus->ops = ops; + + if (initfunc) + if ((*initfunc)(bus, data)) + return; + + iowa_bus_count++; + + pr_debug("IOWA:[%d]Add bus, %s.\n", iowa_bus_count-1, np->full_name); +} + +/* enable IO workaround */ +void __devinit io_workaround_init(void) +{ + static int io_workaround_inited; + + if (io_workaround_inited) + return; + ppc_pci_io = iowa_pci_io; + ppc_md.ioremap = iowa_ioremap; + io_workaround_inited = 1; +} -- cgit v1.2.3 From d1109b7529f362c06c47140ae09dbd2b853ffddc Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Mon, 11 Apr 2011 21:25:02 +0000 Subject: powerpc/pci: Make IO workarounds init implicit when first bus is registered Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/io-workarounds.c | 27 +++++++++++++++------------ 1 file changed, 15 insertions(+), 12 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/io-workarounds.c b/arch/powerpc/kernel/io-workarounds.c index 7e5845798788..d36515eaa314 100644 --- a/arch/powerpc/kernel/io-workarounds.c +++ b/arch/powerpc/kernel/io-workarounds.c @@ -144,7 +144,19 @@ static void __iomem *iowa_ioremap(phys_addr_t addr, unsigned long size, return res; } -/* Regist new bus to support workaround */ +/* Enable IO workaround */ +static void __devinit io_workaround_init(void) +{ + static int io_workaround_inited; + + if (io_workaround_inited) + return; + ppc_pci_io = iowa_pci_io; + ppc_md.ioremap = iowa_ioremap; + io_workaround_inited = 1; +} + +/* Register new bus to support workaround */ void __devinit iowa_register_bus(struct pci_controller *phb, struct ppc_pci_io *ops, int (*initfunc)(struct iowa_bus *, void *), void *data) @@ -152,6 +164,8 @@ void __devinit iowa_register_bus(struct pci_controller *phb, struct iowa_bus *bus; struct device_node *np = phb->dn; + io_workaround_init(); + if (iowa_bus_count >= IOWA_MAX_BUS) { pr_err("IOWA:Too many pci bridges, " "workarounds disabled for %s\n", np->full_name); @@ -171,14 +185,3 @@ void __devinit iowa_register_bus(struct pci_controller *phb, pr_debug("IOWA:[%d]Add bus, %s.\n", iowa_bus_count-1, np->full_name); } -/* enable IO workaround */ -void __devinit io_workaround_init(void) -{ - static int io_workaround_inited; - - if (io_workaround_inited) - return; - ppc_pci_io = iowa_pci_io; - ppc_md.ioremap = iowa_ioremap; - io_workaround_inited = 1; -} -- cgit v1.2.3 From 69b123684b50040b0926eed1e02795dac8cb9587 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Mon, 11 Apr 2011 21:25:02 +0000 Subject: powerpc/pci: Properly initialize IO workaround "private" Even when no initfunc is provided. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/io-workarounds.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/io-workarounds.c b/arch/powerpc/kernel/io-workarounds.c index d36515eaa314..ffafaea3d261 100644 --- a/arch/powerpc/kernel/io-workarounds.c +++ b/arch/powerpc/kernel/io-workarounds.c @@ -175,6 +175,7 @@ void __devinit iowa_register_bus(struct pci_controller *phb, bus = &iowa_busses[iowa_bus_count]; bus->phb = phb; bus->ops = ops; + bus->private = data; if (initfunc) if ((*initfunc)(bus, data)) -- cgit v1.2.3 From 73706c3283d755d3725c6a48a18e677a15ced8be Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Sun, 10 Apr 2011 20:26:15 +0000 Subject: powerpc/irq: Dump chip data pointer in virq_mapping This can be useful for differentiating interrupts on the same host but with different chip data. Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/irq.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index ea09512a68c3..4f5d6e751a65 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -1086,10 +1086,11 @@ static int virq_debug_show(struct seq_file *m, void *private) struct irq_desc *desc; const char *p; static const char none[] = "none"; + void *data; int i; - seq_printf(m, "%-5s %-7s %-15s %s\n", "virq", "hwirq", - "chip name", "host name"); + seq_printf(m, "%-5s %-7s %-15s %-18s %s\n", "virq", "hwirq", + "chip name", "chip data", "host name"); for (i = 1; i < nr_irqs; i++) { desc = irq_to_desc(i); @@ -1111,6 +1112,9 @@ static int virq_debug_show(struct seq_file *m, void *private) p = none; seq_printf(m, "%-15s ", p); + data = irq_desc_get_chip_data(desc); + seq_printf(m, "0x%16p ", data); + if (irq_map[i].host && irq_map[i].host->of_node) p = irq_map[i].host->of_node->full_name; else -- cgit v1.2.3 From 9f0b079320ad1cc71ad7ea4e0ed0b64cd72bbd6d Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Thu, 7 Apr 2011 21:56:03 +0000 Subject: powerpc: Use MSR_64BIT in places Use the new MSR_64BIT in a few places. Some of these are already ifdef'ed for BOOKE vs BOOKS, but it's still clearer, MSR_SF does not immediately parse as "MSR bit for 64bit". Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/head_64.S | 2 +- arch/powerpc/kernel/signal_64.c | 4 ++-- arch/powerpc/kernel/traps.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S index 6d17c37f22a1..73d6e9afcdf1 100644 --- a/arch/powerpc/kernel/head_64.S +++ b/arch/powerpc/kernel/head_64.S @@ -655,7 +655,7 @@ _GLOBAL(enable_64b_mode) oris r11,r11,0x8000 /* CM bit set, we'll set ICM later */ mtmsr r11 #else /* CONFIG_PPC_BOOK3E */ - li r12,(MSR_SF | MSR_ISF)@highest + li r12,(MSR_64BIT | MSR_ISF)@highest sldi r12,r12,48 or r11,r11,r12 mtmsrd r11 diff --git a/arch/powerpc/kernel/signal_64.c b/arch/powerpc/kernel/signal_64.c index 27c4a4584f80..da989fff19cc 100644 --- a/arch/powerpc/kernel/signal_64.c +++ b/arch/powerpc/kernel/signal_64.c @@ -381,7 +381,7 @@ badframe: regs, uc, &uc->uc_mcontext); #endif if (show_unhandled_signals && printk_ratelimit()) - printk(regs->msr & MSR_SF ? fmt64 : fmt32, + printk(regs->msr & MSR_64BIT ? fmt64 : fmt32, current->comm, current->pid, "rt_sigreturn", (long)uc, regs->nip, regs->link); @@ -469,7 +469,7 @@ badframe: regs, frame, newsp); #endif if (show_unhandled_signals && printk_ratelimit()) - printk(regs->msr & MSR_SF ? fmt64 : fmt32, + printk(regs->msr & MSR_64BIT ? fmt64 : fmt32, current->comm, current->pid, "setup_rt_frame", (long)frame, regs->nip, regs->link); diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c index cb71cf29edea..4a6a109b6816 100644 --- a/arch/powerpc/kernel/traps.c +++ b/arch/powerpc/kernel/traps.c @@ -199,7 +199,7 @@ void _exception(int signr, struct pt_regs *regs, int code, unsigned long addr) } else if (show_unhandled_signals && unhandled_signal(current, signr) && printk_ratelimit()) { - printk(regs->msr & MSR_SF ? fmt64 : fmt32, + printk(regs->msr & MSR_64BIT ? fmt64 : fmt32, current->comm, current->pid, signr, addr, regs->nip, regs->link, code); } -- cgit v1.2.3 From eca590f402332ab873d13f2d8d00fa0b91cfff36 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Thu, 7 Apr 2011 01:54:07 +0000 Subject: powerpc/rtas: Only sleep in rtas_busy_delay if we have useful work to do RTAS returns extended error codes as a hint of how long the OS might want to wait before retrying a call. If we have nothing else useful to do we may as well call back straight away. This was found when testing the new dynamic dma window feature. Firmware split the zeroing of the TCE table into 32k chunks but returned 9901 (which is a suggested wait of 10ms). All up this took about 10 minutes to complete since msleep is jiffies based and will round 10ms up to 20ms. With the patch below we take 3 seconds to complete the same test. The hint firmware is returning in the RTAS call should definitely be decreased, but even if we slept 1ms each iteration this would take 32s. Signed-off-by: Anton Blanchard Acked-by: Nishanth Aravamudan Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/rtas.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c index 2097f2b3cba8..f48446635c89 100644 --- a/arch/powerpc/kernel/rtas.c +++ b/arch/powerpc/kernel/rtas.c @@ -494,7 +494,7 @@ unsigned int rtas_busy_delay(int status) might_sleep(); ms = rtas_busy_delay_time(status); - if (ms) + if (ms && need_resched()) msleep(ms); return ms; -- cgit v1.2.3 From 44ae3ab3358e962039c36ad4ae461ae9fb29596c Mon Sep 17 00:00:00 2001 From: Matt Evans Date: Wed, 6 Apr 2011 19:48:50 +0000 Subject: powerpc: Free up some CPU feature bits by moving out MMU-related features Some of the 64bit PPC CPU features are MMU-related, so this patch moves them to MMU_FTR_ bits. All cpu_has_feature()-style tests are moved to mmu_has_feature(), and seven feature bits are freed as a result. Signed-off-by: Matt Evans Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/cputable.c | 45 +++++++++++++++--------------------- arch/powerpc/kernel/entry_64.S | 8 +++---- arch/powerpc/kernel/exceptions-64s.S | 4 ++-- arch/powerpc/kernel/process.c | 4 ++-- arch/powerpc/kernel/prom.c | 17 ++++++++------ arch/powerpc/kernel/setup_64.c | 2 +- 6 files changed, 38 insertions(+), 42 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c index 3d7b65ad4962..34d2722b9451 100644 --- a/arch/powerpc/kernel/cputable.c +++ b/arch/powerpc/kernel/cputable.c @@ -201,7 +201,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER4 (gp)", .cpu_features = CPU_FTRS_POWER4, .cpu_user_features = COMMON_USER_POWER4, - .mmu_features = MMU_FTR_HPTE_TABLE, + .mmu_features = MMU_FTRS_POWER4, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 8, @@ -216,7 +216,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER4+ (gq)", .cpu_features = CPU_FTRS_POWER4, .cpu_user_features = COMMON_USER_POWER4, - .mmu_features = MMU_FTR_HPTE_TABLE, + .mmu_features = MMU_FTRS_POWER4, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 8, @@ -232,7 +232,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_features = CPU_FTRS_PPC970, .cpu_user_features = COMMON_USER_POWER4 | PPC_FEATURE_HAS_ALTIVEC_COMP, - .mmu_features = MMU_FTR_HPTE_TABLE, + .mmu_features = MMU_FTRS_PPC970, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 8, @@ -250,7 +250,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_features = CPU_FTRS_PPC970, .cpu_user_features = COMMON_USER_POWER4 | PPC_FEATURE_HAS_ALTIVEC_COMP, - .mmu_features = MMU_FTR_HPTE_TABLE, + .mmu_features = MMU_FTRS_PPC970, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 8, @@ -286,7 +286,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_features = CPU_FTRS_PPC970, .cpu_user_features = COMMON_USER_POWER4 | PPC_FEATURE_HAS_ALTIVEC_COMP, - .mmu_features = MMU_FTR_HPTE_TABLE, + .mmu_features = MMU_FTRS_PPC970, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 8, @@ -304,7 +304,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_features = CPU_FTRS_PPC970, .cpu_user_features = COMMON_USER_POWER4 | PPC_FEATURE_HAS_ALTIVEC_COMP, - .mmu_features = MMU_FTR_HPTE_TABLE, + .mmu_features = MMU_FTRS_PPC970, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 8, @@ -320,7 +320,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER5 (gr)", .cpu_features = CPU_FTRS_POWER5, .cpu_user_features = COMMON_USER_POWER5, - .mmu_features = MMU_FTR_HPTE_TABLE, + .mmu_features = MMU_FTRS_POWER5, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 6, @@ -340,7 +340,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER5+ (gs)", .cpu_features = CPU_FTRS_POWER5, .cpu_user_features = COMMON_USER_POWER5_PLUS, - .mmu_features = MMU_FTR_HPTE_TABLE, + .mmu_features = MMU_FTRS_POWER5, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 6, @@ -356,7 +356,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER5+ (gs)", .cpu_features = CPU_FTRS_POWER5, .cpu_user_features = COMMON_USER_POWER5_PLUS, - .mmu_features = MMU_FTR_HPTE_TABLE, + .mmu_features = MMU_FTRS_POWER5, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 6, @@ -373,7 +373,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER5+", .cpu_features = CPU_FTRS_POWER5, .cpu_user_features = COMMON_USER_POWER5_PLUS, - .mmu_features = MMU_FTR_HPTE_TABLE, + .mmu_features = MMU_FTRS_POWER5, .icache_bsize = 128, .dcache_bsize = 128, .oprofile_cpu_type = "ppc64/ibm-compat-v1", @@ -387,7 +387,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_features = CPU_FTRS_POWER6, .cpu_user_features = COMMON_USER_POWER6 | PPC_FEATURE_POWER6_EXT, - .mmu_features = MMU_FTR_HPTE_TABLE, + .mmu_features = MMU_FTRS_POWER6, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 6, @@ -406,7 +406,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER6 (architected)", .cpu_features = CPU_FTRS_POWER6, .cpu_user_features = COMMON_USER_POWER6, - .mmu_features = MMU_FTR_HPTE_TABLE, + .mmu_features = MMU_FTRS_POWER6, .icache_bsize = 128, .dcache_bsize = 128, .oprofile_cpu_type = "ppc64/ibm-compat-v1", @@ -419,8 +419,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER7 (architected)", .cpu_features = CPU_FTRS_POWER7, .cpu_user_features = COMMON_USER_POWER7, - .mmu_features = MMU_FTR_HPTE_TABLE | - MMU_FTR_TLBIE_206, + .mmu_features = MMU_FTRS_POWER7, .icache_bsize = 128, .dcache_bsize = 128, .oprofile_type = PPC_OPROFILE_POWER4, @@ -435,8 +434,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER7 (raw)", .cpu_features = CPU_FTRS_POWER7, .cpu_user_features = COMMON_USER_POWER7, - .mmu_features = MMU_FTR_HPTE_TABLE | - MMU_FTR_TLBIE_206, + .mmu_features = MMU_FTRS_POWER7, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 6, @@ -453,8 +451,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER7+ (raw)", .cpu_features = CPU_FTRS_POWER7, .cpu_user_features = COMMON_USER_POWER7, - .mmu_features = MMU_FTR_HPTE_TABLE | - MMU_FTR_TLBIE_206, + .mmu_features = MMU_FTRS_POWER7, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 6, @@ -473,7 +470,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_user_features = COMMON_USER_PPC64 | PPC_FEATURE_CELL | PPC_FEATURE_HAS_ALTIVEC_COMP | PPC_FEATURE_SMT, - .mmu_features = MMU_FTR_HPTE_TABLE, + .mmu_features = MMU_FTRS_CELL, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 4, @@ -488,7 +485,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "PA6T", .cpu_features = CPU_FTRS_PA6T, .cpu_user_features = COMMON_USER_PA6T, - .mmu_features = MMU_FTR_HPTE_TABLE, + .mmu_features = MMU_FTRS_PA6T, .icache_bsize = 64, .dcache_bsize = 64, .num_pmcs = 6, @@ -505,7 +502,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "POWER4 (compatible)", .cpu_features = CPU_FTRS_COMPATIBLE, .cpu_user_features = COMMON_USER_PPC64, - .mmu_features = MMU_FTR_HPTE_TABLE, + .mmu_features = MMU_FTRS_DEFAULT_HPTE_ARCH_V2, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 6, @@ -2020,11 +2017,7 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_name = "A2 (>= DD2)", .cpu_features = CPU_FTRS_A2, .cpu_user_features = COMMON_USER_PPC64, - .mmu_features = MMU_FTR_TYPE_3E | MMU_FTR_USE_TLBILX | - MMU_FTR_USE_TLBIVAX_BCAST | - MMU_FTR_LOCK_BCAST_INVAL | - MMU_FTR_USE_TLBRSRV | - MMU_FTR_USE_PAIRED_MAS, + .mmu_features = MMU_FTRS_A2, .icache_bsize = 64, .dcache_bsize = 64, .num_pmcs = 0, diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S index 64693706ebfd..d834425186ae 100644 --- a/arch/powerpc/kernel/entry_64.S +++ b/arch/powerpc/kernel/entry_64.S @@ -468,10 +468,10 @@ BEGIN_FTR_SECTION FTR_SECTION_ELSE_NESTED(95) clrrdi r6,r8,40 /* get its 1T ESID */ clrrdi r9,r1,40 /* get current sp 1T ESID */ - ALT_FTR_SECTION_END_NESTED_IFCLR(CPU_FTR_1T_SEGMENT, 95) + ALT_MMU_FTR_SECTION_END_NESTED_IFCLR(MMU_FTR_1T_SEGMENT, 95) FTR_SECTION_ELSE b 2f -ALT_FTR_SECTION_END_IFSET(CPU_FTR_SLB) +ALT_MMU_FTR_SECTION_END_IFSET(MMU_FTR_SLB) clrldi. r0,r6,2 /* is new ESID c00000000? */ cmpd cr1,r6,r9 /* or is new ESID the same as current ESID? */ cror eq,4*cr1+eq,eq @@ -485,7 +485,7 @@ BEGIN_FTR_SECTION li r9,MMU_SEGSIZE_1T /* insert B field */ oris r6,r6,(MMU_SEGSIZE_1T << SLBIE_SSIZE_SHIFT)@h rldimi r7,r9,SLB_VSID_SSIZE_SHIFT,0 -END_FTR_SECTION_IFSET(CPU_FTR_1T_SEGMENT) +END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) /* Update the last bolted SLB. No write barriers are needed * here, provided we only update the current CPU's SLB shadow @@ -497,7 +497,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_1T_SEGMENT) std r7,SLBSHADOW_STACKVSID(r9) /* Save VSID */ std r0,SLBSHADOW_STACKESID(r9) /* Save ESID */ - /* No need to check for CPU_FTR_NO_SLBIE_B here, since when + /* No need to check for MMU_FTR_NO_SLBIE_B here, since when * we have 1TB segments, the only CPUs known to have the errata * only support less than 1TB of system memory and we'll never * actually hit this code path. diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S index ad06333631ac..226cc8c62224 100644 --- a/arch/powerpc/kernel/exceptions-64s.S +++ b/arch/powerpc/kernel/exceptions-64s.S @@ -102,7 +102,7 @@ BEGIN_FTR_SECTION EXCEPTION_PROLOG_PSERIES_1(data_access_common, EXC_STD) FTR_SECTION_ELSE EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, data_access_common, EXC_STD) -ALT_FTR_SECTION_END_IFCLR(CPU_FTR_SLB) +ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_SLB) . = 0x380 .globl data_access_slb_pSeries @@ -840,7 +840,7 @@ _STATIC(do_hash_page) BEGIN_FTR_SECTION andis. r0,r4,0x0020 /* Is it a segment table fault? */ bne- do_ste_alloc /* If so handle it */ -END_FTR_SECTION_IFCLR(CPU_FTR_SLB) +END_MMU_FTR_SECTION_IFCLR(MMU_FTR_SLB) clrrdi r11,r1,THREAD_SHIFT lwz r0,TI_PREEMPT(r11) /* If we're in an "NMI" */ diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c index a01c2d93fd2f..095043d79946 100644 --- a/arch/powerpc/kernel/process.c +++ b/arch/powerpc/kernel/process.c @@ -757,11 +757,11 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, _ALIGN_UP(sizeof(struct thread_info), 16); #ifdef CONFIG_PPC_STD_MMU_64 - if (cpu_has_feature(CPU_FTR_SLB)) { + if (mmu_has_feature(MMU_FTR_SLB)) { unsigned long sp_vsid; unsigned long llp = mmu_psize_defs[mmu_linear_psize].sllp; - if (cpu_has_feature(CPU_FTR_1T_SEGMENT)) + if (mmu_has_feature(MMU_FTR_1T_SEGMENT)) sp_vsid = get_kernel_vsid(sp, MMU_SEGSIZE_1T) << SLB_VSID_SHIFT_1T; else diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c index c391dc4c8bad..5f5e6aed2b70 100644 --- a/arch/powerpc/kernel/prom.c +++ b/arch/powerpc/kernel/prom.c @@ -123,18 +123,19 @@ static void __init move_device_tree(void) */ static struct ibm_pa_feature { unsigned long cpu_features; /* CPU_FTR_xxx bit */ + unsigned long mmu_features; /* MMU_FTR_xxx bit */ unsigned int cpu_user_ftrs; /* PPC_FEATURE_xxx bit */ unsigned char pabyte; /* byte number in ibm,pa-features */ unsigned char pabit; /* bit number (big-endian) */ unsigned char invert; /* if 1, pa bit set => clear feature */ } ibm_pa_features[] __initdata = { - {0, PPC_FEATURE_HAS_MMU, 0, 0, 0}, - {0, PPC_FEATURE_HAS_FPU, 0, 1, 0}, - {CPU_FTR_SLB, 0, 0, 2, 0}, - {CPU_FTR_CTRL, 0, 0, 3, 0}, - {CPU_FTR_NOEXECUTE, 0, 0, 6, 0}, - {CPU_FTR_NODSISRALIGN, 0, 1, 1, 1}, - {CPU_FTR_CI_LARGE_PAGE, 0, 1, 2, 0}, + {0, 0, PPC_FEATURE_HAS_MMU, 0, 0, 0}, + {0, 0, PPC_FEATURE_HAS_FPU, 0, 1, 0}, + {0, MMU_FTR_SLB, 0, 0, 2, 0}, + {CPU_FTR_CTRL, 0, 0, 0, 3, 0}, + {CPU_FTR_NOEXECUTE, 0, 0, 0, 6, 0}, + {CPU_FTR_NODSISRALIGN, 0, 0, 1, 1, 1}, + {0, MMU_FTR_CI_LARGE_PAGE, 0, 1, 2, 0}, {CPU_FTR_REAL_LE, PPC_FEATURE_TRUE_LE, 5, 0, 0}, }; @@ -166,9 +167,11 @@ static void __init scan_features(unsigned long node, unsigned char *ftrs, if (bit ^ fp->invert) { cur_cpu_spec->cpu_features |= fp->cpu_features; cur_cpu_spec->cpu_user_features |= fp->cpu_user_ftrs; + cur_cpu_spec->mmu_features |= fp->mmu_features; } else { cur_cpu_spec->cpu_features &= ~fp->cpu_features; cur_cpu_spec->cpu_user_features &= ~fp->cpu_user_ftrs; + cur_cpu_spec->mmu_features &= ~fp->mmu_features; } } } diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index 91a5cc5f0d02..959c63cf62e4 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -436,7 +436,7 @@ void __init setup_system(void) static u64 slb0_limit(void) { - if (cpu_has_feature(CPU_FTR_1T_SEGMENT)) { + if (mmu_has_feature(MMU_FTR_1T_SEGMENT)) { return 1UL << SID_SHIFT_1T; } return 1UL << SID_SHIFT; -- cgit v1.2.3 From 476eb4912601a8c01e6702b9a029f476b4b131d2 Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Wed, 4 May 2011 15:02:15 +1000 Subject: powerpc/irq: Stop exporting irq_map First step in eliminating irq_map[] table entirely Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/irq.c | 26 ++++++++++++++++++++++++-- 1 file changed, 24 insertions(+), 2 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index 4f5d6e751a65..a81dd74414bf 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -481,20 +481,42 @@ void do_softirq(void) * IRQ controller and virtual interrupts */ +/* The main irq map itself is an array of NR_IRQ entries containing the + * associate host and irq number. An entry with a host of NULL is free. + * An entry can be allocated if it's free, the allocator always then sets + * hwirq first to the host's invalid irq number and then fills ops. + */ +struct irq_map_entry { + irq_hw_number_t hwirq; + struct irq_host *host; +}; + static LIST_HEAD(irq_hosts); static DEFINE_RAW_SPINLOCK(irq_big_lock); static unsigned int revmap_trees_allocated; static DEFINE_MUTEX(revmap_trees_mutex); -struct irq_map_entry irq_map[NR_IRQS]; +static struct irq_map_entry irq_map[NR_IRQS]; static unsigned int irq_virq_count = NR_IRQS; static struct irq_host *irq_default_host; +irq_hw_number_t irqd_to_hwirq(struct irq_data *d) +{ + return irq_map[d->irq].hwirq; +} +EXPORT_SYMBOL_GPL(irqd_to_hwirq); + irq_hw_number_t virq_to_hw(unsigned int virq) { return irq_map[virq].hwirq; } EXPORT_SYMBOL_GPL(virq_to_hw); +struct irq_host *virq_to_host(unsigned int virq) +{ + return irq_map[virq].host; +} +EXPORT_SYMBOL_GPL(virq_to_host); + static int default_irq_host_match(struct irq_host *h, struct device_node *np) { return h->of_node != NULL && h->of_node == np; @@ -1103,7 +1125,7 @@ static int virq_debug_show(struct seq_file *m, void *private) struct irq_chip *chip; seq_printf(m, "%5d ", i); - seq_printf(m, "0x%05lx ", virq_to_hw(i)); + seq_printf(m, "0x%05lx ", irq_map[i].hwirq); chip = irq_desc_get_chip(desc); if (chip && chip->name) -- cgit v1.2.3 From 1977b502120d44b9b4897703adfb2e2fab346880 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Sun, 1 May 2011 19:46:44 +0000 Subject: powerpc: Save register r9-r13 values accurately on interrupt with bad stack When we take an interrupt or exception from kernel mode and the stack pointer is obviously not a kernel address (i.e. the top bit is 0), we switch to an emergency stack, save register values and panic. However, on 64-bit server machines, we don't actually save the values of r9 - r13 at the time of the interrupt, but rather values corrupted by the exception entry code for r12-r13, and nothing at all for r9-r11. This fixes it by passing a pointer to the register save area in the paca through to the bad_stack code in r3. The register values are saved in one of the paca register save areas (depending on which exception this is). Using the pointer in r3, the bad_stack code now retrieves the saved values of r9 - r13 and stores them in the exception frame on the emergency stack. This also stores the normal exception frame marker ("regshere") in the exception frame. Signed-off-by: Paul Mackerras Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/exceptions-64s.S | 20 +++++++++++++++++--- 1 file changed, 17 insertions(+), 3 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S index 226cc8c62224..27ca8b7c6002 100644 --- a/arch/powerpc/kernel/exceptions-64s.S +++ b/arch/powerpc/kernel/exceptions-64s.S @@ -461,9 +461,20 @@ bad_stack: std r12,_XER(r1) SAVE_GPR(0,r1) SAVE_GPR(2,r1) - SAVE_4GPRS(3,r1) - SAVE_2GPRS(7,r1) - SAVE_10GPRS(12,r1) + ld r10,EX_R3(r3) + std r10,GPR3(r1) + SAVE_GPR(4,r1) + SAVE_4GPRS(5,r1) + ld r9,EX_R9(r3) + ld r10,EX_R10(r3) + SAVE_2GPRS(9,r1) + ld r9,EX_R11(r3) + ld r10,EX_R12(r3) + ld r11,EX_R13(r3) + std r9,GPR11(r1) + std r10,GPR12(r1) + std r11,GPR13(r1) + SAVE_8GPRS(14,r1) SAVE_10GPRS(22,r1) lhz r12,PACA_TRAP_SAVE(r13) std r12,_TRAP(r1) @@ -472,6 +483,9 @@ bad_stack: li r12,0 std r12,0(r11) ld r2,PACATOC(r13) + ld r11,exception_marker@toc(r2) + std r12,RESULT(r1) + std r11,STACK_FRAME_OVERHEAD-16(r1) 1: addi r3,r1,STACK_FRAME_OVERHEAD bl .kernel_bad_stack b 1b -- cgit v1.2.3 From 48404f2e95ef0ffd8134d89c8abcd1a15e15f1b0 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Sun, 1 May 2011 19:48:20 +0000 Subject: powerpc: Save Come-From Address Register (CFAR) in exception frame Recent 64-bit server processors (POWER6 and POWER7) have a "Come-From Address Register" (CFAR), that records the address of the most recent branch or rfid (return from interrupt) instruction for debugging purposes. This saves the value of the CFAR in the exception entry code and stores it in the exception frame. We also make xmon print the CFAR value in its register dump code. Rather than extend the pt_regs struct at this time, we steal the orig_gpr3 field, which is only used for system calls, and use it for the CFAR value for all exceptions/interrupts other than system calls. This means we don't save the CFAR on system calls, which is not a great problem since system calls tend not to happen unexpectedly, and also avoids adding the overhead of reading the CFAR to the system call entry path. Signed-off-by: Paul Mackerras Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/exceptions-64s.S | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S index 27ca8b7c6002..0ec3b42717d7 100644 --- a/arch/powerpc/kernel/exceptions-64s.S +++ b/arch/powerpc/kernel/exceptions-64s.S @@ -474,6 +474,10 @@ bad_stack: std r9,GPR11(r1) std r10,GPR12(r1) std r11,GPR13(r1) +BEGIN_FTR_SECTION + ld r10,EX_CFAR(r3) + std r10,ORIG_GPR3(r1) +END_FTR_SECTION_IFSET(CPU_FTR_CFAR) SAVE_8GPRS(14,r1) SAVE_10GPRS(22,r1) lhz r12,PACA_TRAP_SAVE(r13) -- cgit v1.2.3 From 104699c0ab473535793b5fea156adaf309afd29b Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Thu, 28 Apr 2011 05:07:23 +0000 Subject: powerpc: Convert old cpumask API into new one Adapt new API. Almost change is trivial. Most important change is the below line because we plan to change task->cpus_allowed implementation. - ctx->cpus_allowed = current->cpus_allowed; Signed-off-by: KOSAKI Motohiro Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/crash.c | 32 ++++++++++++++++---------------- arch/powerpc/kernel/setup-common.c | 4 ++-- arch/powerpc/kernel/smp.c | 4 ++-- arch/powerpc/kernel/traps.c | 2 +- 4 files changed, 21 insertions(+), 21 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/crash.c b/arch/powerpc/kernel/crash.c index 5b5e1f002a8e..ccc2198e6b23 100644 --- a/arch/powerpc/kernel/crash.c +++ b/arch/powerpc/kernel/crash.c @@ -64,9 +64,9 @@ void crash_ipi_callback(struct pt_regs *regs) return; hard_irq_disable(); - if (!cpu_isset(cpu, cpus_in_crash)) + if (!cpumask_test_cpu(cpu, &cpus_in_crash)) crash_save_cpu(regs, cpu); - cpu_set(cpu, cpus_in_crash); + cpumask_set_cpu(cpu, &cpus_in_crash); /* * Entered via soft-reset - could be the kdump @@ -77,8 +77,8 @@ void crash_ipi_callback(struct pt_regs *regs) * Tell the kexec CPU that entered via soft-reset and ready * to go down. */ - if (cpu_isset(cpu, cpus_in_sr)) { - cpu_clear(cpu, cpus_in_sr); + if (cpumask_test_cpu(cpu, &cpus_in_sr)) { + cpumask_clear_cpu(cpu, &cpus_in_sr); atomic_inc(&enter_on_soft_reset); } @@ -87,7 +87,7 @@ void crash_ipi_callback(struct pt_regs *regs) * This barrier is needed to make sure that all CPUs are stopped. * If not, soft-reset will be invoked to bring other CPUs. */ - while (!cpu_isset(crashing_cpu, cpus_in_crash)) + while (!cpumask_test_cpu(crashing_cpu, &cpus_in_crash)) cpu_relax(); if (ppc_md.kexec_cpu_down) @@ -109,7 +109,7 @@ static void crash_soft_reset_check(int cpu) { unsigned int ncpus = num_online_cpus() - 1;/* Excluding the panic cpu */ - cpu_clear(cpu, cpus_in_sr); + cpumask_clear_cpu(cpu, &cpus_in_sr); while (atomic_read(&enter_on_soft_reset) != ncpus) cpu_relax(); } @@ -132,7 +132,7 @@ static void crash_kexec_prepare_cpus(int cpu) */ printk(KERN_EMERG "Sending IPI to other cpus...\n"); msecs = 10000; - while ((cpus_weight(cpus_in_crash) < ncpus) && (--msecs > 0)) { + while ((cpumask_weight(&cpus_in_crash) < ncpus) && (--msecs > 0)) { cpu_relax(); mdelay(1); } @@ -144,20 +144,20 @@ static void crash_kexec_prepare_cpus(int cpu) * user to do soft reset such that we get all. * Soft-reset will be used until better mechanism is implemented. */ - if (cpus_weight(cpus_in_crash) < ncpus) { + if (cpumask_weight(&cpus_in_crash) < ncpus) { printk(KERN_EMERG "done waiting: %d cpu(s) not responding\n", - ncpus - cpus_weight(cpus_in_crash)); + ncpus - cpumask_weight(&cpus_in_crash)); printk(KERN_EMERG "Activate soft-reset to stop other cpu(s)\n"); - cpus_in_sr = CPU_MASK_NONE; + cpumask_clear(&cpus_in_sr); atomic_set(&enter_on_soft_reset, 0); - while (cpus_weight(cpus_in_crash) < ncpus) + while (cpumask_weight(&cpus_in_crash) < ncpus) cpu_relax(); } /* * Make sure all CPUs are entered via soft-reset if the kdump is * invoked using soft-reset. */ - if (cpu_isset(cpu, cpus_in_sr)) + if (cpumask_test_cpu(cpu, &cpus_in_sr)) crash_soft_reset_check(cpu); /* Leave the IPI callback set */ } @@ -210,7 +210,7 @@ void crash_kexec_secondary(struct pt_regs *regs) * exited using 'x'(exit and recover) or * kexec_should_crash() failed for all running tasks. */ - cpu_clear(cpu, cpus_in_sr); + cpumask_clear_cpu(cpu, &cpus_in_sr); local_irq_restore(flags); return; } @@ -224,7 +224,7 @@ void crash_kexec_secondary(struct pt_regs *regs) * then start kexec boot. */ crash_soft_reset_check(cpu); - cpu_set(crashing_cpu, cpus_in_crash); + cpumask_set_cpu(crashing_cpu, &cpus_in_crash); if (ppc_md.kexec_cpu_down) ppc_md.kexec_cpu_down(1, 0); machine_kexec(kexec_crash_image); @@ -253,7 +253,7 @@ static void crash_kexec_prepare_cpus(int cpu) void crash_kexec_secondary(struct pt_regs *regs) { - cpus_in_sr = CPU_MASK_NONE; + cpumask_clear(&cpus_in_sr); } #endif /* CONFIG_SMP */ @@ -345,7 +345,7 @@ void default_machine_crash_shutdown(struct pt_regs *regs) crashing_cpu = smp_processor_id(); crash_save_cpu(regs, crashing_cpu); crash_kexec_prepare_cpus(crashing_cpu); - cpu_set(crashing_cpu, cpus_in_crash); + cpumask_set_cpu(crashing_cpu, &cpus_in_crash); crash_kexec_wait_realmode(crashing_cpu); machine_kexec_mask_interrupts(); diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c index 21f30cb68077..1475df6e403f 100644 --- a/arch/powerpc/kernel/setup-common.c +++ b/arch/powerpc/kernel/setup-common.c @@ -381,7 +381,7 @@ static void __init cpu_init_thread_core_maps(int tpc) int i; threads_per_core = tpc; - threads_core_mask = CPU_MASK_NONE; + cpumask_clear(&threads_core_mask); /* This implementation only supports power of 2 number of threads * for simplicity and performance @@ -390,7 +390,7 @@ static void __init cpu_init_thread_core_maps(int tpc) BUG_ON(tpc != (1 << threads_shift)); for (i = 0; i < tpc; i++) - cpu_set(i, threads_core_mask); + cpumask_set_cpu(i, &threads_core_mask); printk(KERN_INFO "CPU maps initialized for %d thread%s per core\n", tpc, tpc > 1 ? "s" : ""); diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index b6083f4f39b1..87517ab6d365 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -513,7 +513,7 @@ int cpu_first_thread_of_core(int core) } EXPORT_SYMBOL_GPL(cpu_first_thread_of_core); -/* Must be called when no change can occur to cpu_present_map, +/* Must be called when no change can occur to cpu_present_mask, * i.e. during cpu online or offline. */ static struct device_node *cpu_to_l2cache(int cpu) @@ -614,7 +614,7 @@ void __init smp_cpus_done(unsigned int max_cpus) * se we pin us down to CPU 0 for a short while */ alloc_cpumask_var(&old_mask, GFP_NOWAIT); - cpumask_copy(old_mask, ¤t->cpus_allowed); + cpumask_copy(old_mask, tsk_cpus_allowed(current)); set_cpus_allowed_ptr(current, cpumask_of(boot_cpuid)); if (smp_ops && smp_ops->setup_cpu) diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c index 4a6a109b6816..06b9d457d0a7 100644 --- a/arch/powerpc/kernel/traps.c +++ b/arch/powerpc/kernel/traps.c @@ -221,7 +221,7 @@ void system_reset_exception(struct pt_regs *regs) } #ifdef CONFIG_KEXEC - cpu_set(smp_processor_id(), cpus_in_sr); + cpumask_set_cpu(smp_processor_id(), &cpus_in_sr); #endif die("System Reset", regs, SIGABRT); -- cgit v1.2.3 From 7707e4110e5692fe85e7e6c471c9bb2a9254d313 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Sun, 24 Apr 2011 15:04:31 +0000 Subject: powerpc/kexec: Fix build failure on 32-bit SMP Commit b987812b3fcaf70fdf0037589e5d2f5f2453e6ce left crash_kexec_wait_realmode() undefined for UP. Commit 7c7a81b53e581d727d069cc45df5510516faac31 defined it for UP but left it undefined for 32-bit SMP. Seems like people are getting confused by nested #ifdef's, so move the definitions of crash_kexec_wait_realmode() after the #ifdef CONFIG_SMP section. Compile-tested with 32-bit UP, 32-bit SMP and 64-bit SMP configurations. Signed-off-by: Ben Hutchings Tested-by: Paul Gortmaker Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/crash.c | 59 +++++++++++++++++++++++---------------------- 1 file changed, 30 insertions(+), 29 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/crash.c b/arch/powerpc/kernel/crash.c index ccc2198e6b23..21f2c781ded1 100644 --- a/arch/powerpc/kernel/crash.c +++ b/arch/powerpc/kernel/crash.c @@ -162,34 +162,6 @@ static void crash_kexec_prepare_cpus(int cpu) /* Leave the IPI callback set */ } -/* wait for all the CPUs to hit real mode but timeout if they don't come in */ -#ifdef CONFIG_PPC_STD_MMU_64 -static void crash_kexec_wait_realmode(int cpu) -{ - unsigned int msecs; - int i; - - msecs = 10000; - for (i=0; i < NR_CPUS && msecs > 0; i++) { - if (i == cpu) - continue; - - while (paca[i].kexec_state < KEXEC_STATE_REAL_MODE) { - barrier(); - if (!cpu_possible(i)) { - break; - } - if (!cpu_online(i)) { - break; - } - msecs--; - mdelay(1); - } - } - mb(); -} -#endif /* CONFIG_PPC_STD_MMU_64 */ - /* * This function will be called by secondary cpus or by kexec cpu * if soft-reset is activated to stop some CPUs. @@ -234,7 +206,6 @@ void crash_kexec_secondary(struct pt_regs *regs) } #else /* ! CONFIG_SMP */ -static inline void crash_kexec_wait_realmode(int cpu) {} static void crash_kexec_prepare_cpus(int cpu) { @@ -257,6 +228,36 @@ void crash_kexec_secondary(struct pt_regs *regs) } #endif /* CONFIG_SMP */ +/* wait for all the CPUs to hit real mode but timeout if they don't come in */ +#if defined(CONFIG_SMP) && defined(CONFIG_PPC_STD_MMU_64) +static void crash_kexec_wait_realmode(int cpu) +{ + unsigned int msecs; + int i; + + msecs = 10000; + for (i=0; i < NR_CPUS && msecs > 0; i++) { + if (i == cpu) + continue; + + while (paca[i].kexec_state < KEXEC_STATE_REAL_MODE) { + barrier(); + if (!cpu_possible(i)) { + break; + } + if (!cpu_online(i)) { + break; + } + msecs--; + mdelay(1); + } + } + mb(); +} +#else +static inline void crash_kexec_wait_realmode(int cpu) {} +#endif /* CONFIG_SMP && CONFIG_PPC_STD_MMU_64 */ + /* * Register a function to be called on shutdown. Only use this if you * can't reset your device in the second kernel. -- cgit v1.2.3 From 9ee820fa005254dfc816330f6654f14dcb2beee1 Mon Sep 17 00:00:00 2001 From: Brian King Date: Wed, 4 May 2011 16:01:20 +1000 Subject: powerpc/pseries: Add page coalescing support Adds support for page coalescing, which is a feature on IBM Power servers which allows for coalescing identical pages between logical partitions. Hint text pages as coalesce candidates, since they are the most likely pages to be able to be coalesced between partitions. This patch also exports some page coalescing statistics available from firmware via lparcfg. [BenH: Moved a couple of things around to fix compile problems] Signed-off-by: Brian King Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/lparcfg.c | 53 +++++++++++++++++++---------------------- arch/powerpc/kernel/prom_init.c | 4 +++- arch/powerpc/kernel/rtas.c | 2 ++ 3 files changed, 30 insertions(+), 29 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c index 301db65f05a1..84daabe2fcba 100644 --- a/arch/powerpc/kernel/lparcfg.c +++ b/arch/powerpc/kernel/lparcfg.c @@ -132,34 +132,6 @@ static int iseries_lparcfg_data(struct seq_file *m, void *v) /* * Methods used to fetch LPAR data when running on a pSeries platform. */ -/** - * h_get_mpp - * H_GET_MPP hcall returns info in 7 parms - */ -int h_get_mpp(struct hvcall_mpp_data *mpp_data) -{ - int rc; - unsigned long retbuf[PLPAR_HCALL9_BUFSIZE]; - - rc = plpar_hcall9(H_GET_MPP, retbuf); - - mpp_data->entitled_mem = retbuf[0]; - mpp_data->mapped_mem = retbuf[1]; - - mpp_data->group_num = (retbuf[2] >> 2 * 8) & 0xffff; - mpp_data->pool_num = retbuf[2] & 0xffff; - - mpp_data->mem_weight = (retbuf[3] >> 7 * 8) & 0xff; - mpp_data->unallocated_mem_weight = (retbuf[3] >> 6 * 8) & 0xff; - mpp_data->unallocated_entitlement = retbuf[3] & 0xffffffffffff; - - mpp_data->pool_size = retbuf[4]; - mpp_data->loan_request = retbuf[5]; - mpp_data->backing_mem = retbuf[6]; - - return rc; -} -EXPORT_SYMBOL(h_get_mpp); struct hvcall_ppp_data { u64 entitlement; @@ -345,6 +317,30 @@ static void parse_mpp_data(struct seq_file *m) seq_printf(m, "backing_memory=%ld bytes\n", mpp_data.backing_mem); } +/** + * parse_mpp_x_data + * Parse out data returned from h_get_mpp_x + */ +static void parse_mpp_x_data(struct seq_file *m) +{ + struct hvcall_mpp_x_data mpp_x_data; + + if (!firmware_has_feature(FW_FEATURE_XCMO)) + return; + if (h_get_mpp_x(&mpp_x_data)) + return; + + seq_printf(m, "coalesced_bytes=%ld\n", mpp_x_data.coalesced_bytes); + + if (mpp_x_data.pool_coalesced_bytes) + seq_printf(m, "pool_coalesced_bytes=%ld\n", + mpp_x_data.pool_coalesced_bytes); + if (mpp_x_data.pool_purr_cycles) + seq_printf(m, "coalesce_pool_purr=%ld\n", mpp_x_data.pool_purr_cycles); + if (mpp_x_data.pool_spurr_cycles) + seq_printf(m, "coalesce_pool_spurr=%ld\n", mpp_x_data.pool_spurr_cycles); +} + #define SPLPAR_CHARACTERISTICS_TOKEN 20 #define SPLPAR_MAXLENGTH 1026*(sizeof(char)) @@ -520,6 +516,7 @@ static int pseries_lparcfg_data(struct seq_file *m, void *v) parse_system_parameter_string(m); parse_ppp_data(m); parse_mpp_data(m); + parse_mpp_x_data(m); pseries_cmo_data(m); splpar_dispatch_data(m); diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c index 7839bd7bfd15..c016033ba78d 100644 --- a/arch/powerpc/kernel/prom_init.c +++ b/arch/powerpc/kernel/prom_init.c @@ -700,8 +700,10 @@ static void __init early_cmdline_parse(void) #endif /* CONFIG_PCI_MSI */ #ifdef CONFIG_PPC_SMLPAR #define OV5_CMO 0x80 /* Cooperative Memory Overcommitment */ +#define OV5_XCMO 0x40 /* Page Coalescing */ #else #define OV5_CMO 0x00 +#define OV5_XCMO 0x00 #endif #define OV5_TYPE1_AFFINITY 0x80 /* Type 1 NUMA affinity */ @@ -756,7 +758,7 @@ static unsigned char ibm_architecture_vec[] = { OV5_LPAR | OV5_SPLPAR | OV5_LARGE_PAGES | OV5_DRCONF_MEMORY | OV5_DONATE_DEDICATE_CPU | OV5_MSI, 0, - OV5_CMO, + OV5_CMO | OV5_XCMO, OV5_TYPE1_AFFINITY, 0, 0, diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c index f48446635c89..271ff6318eda 100644 --- a/arch/powerpc/kernel/rtas.c +++ b/arch/powerpc/kernel/rtas.c @@ -42,6 +42,7 @@ #include #include #include +#include struct rtas_t rtas = { .lock = __ARCH_SPIN_LOCK_UNLOCKED @@ -731,6 +732,7 @@ static int __rtas_suspend_last_cpu(struct rtas_suspend_me_data *data, int wake_w atomic_set(&data->error, rc); start_topology_update(); + pSeries_coalesce_init(); if (wake_when_done) { atomic_set(&data->done, 1); -- cgit v1.2.3 From 40bd587a88fcd425f489f3d9f0be7daa84014141 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 3 May 2011 14:07:01 +0000 Subject: powerpc: Rename slb0_limit() to safe_stack_limit() and add Book3E support slb0_limit() wasn't a very descriptive name. This changes it along with a comment explaining what it's used for, and provides a 64-bit BookE implementation. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/setup_64.c | 23 ++++++++++++++++++----- 1 file changed, 18 insertions(+), 5 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index 959c63cf62e4..c2ec0a12e14f 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -434,17 +434,30 @@ void __init setup_system(void) DBG(" <- setup_system()\n"); } -static u64 slb0_limit(void) +/* This returns the limit below which memory accesses to the linear + * mapping are guarnateed not to cause a TLB or SLB miss. This is + * used to allocate interrupt or emergency stacks for which our + * exception entry path doesn't deal with being interrupted. + */ +static u64 safe_stack_limit(void) { - if (mmu_has_feature(MMU_FTR_1T_SEGMENT)) { +#ifdef CONFIG_PPC_BOOK3E + /* Freescale BookE bolts the entire linear mapping */ + if (mmu_has_feature(MMU_FTR_TYPE_FSL_E)) + return linear_map_top; + /* Other BookE, we assume the first GB is bolted */ + return 1ul << 30; +#else + /* BookS, the first segment is bolted */ + if (mmu_has_feature(MMU_FTR_1T_SEGMENT)) return 1UL << SID_SHIFT_1T; - } return 1UL << SID_SHIFT; +#endif } static void __init irqstack_early_init(void) { - u64 limit = slb0_limit(); + u64 limit = safe_stack_limit(); unsigned int i; /* @@ -497,7 +510,7 @@ static void __init emergency_stack_init(void) * bringup, we need to get at them in real mode. This means they * must also be within the RMO region. */ - limit = min(slb0_limit(), ppc64_rma_size); + limit = min(safe_stack_limit(), ppc64_rma_size); for_each_possible_cpu(i) { unsigned long sp; -- cgit v1.2.3 From a1d0d98daf6ce580d017a43b09fe30a375cde3e8 Mon Sep 17 00:00:00 2001 From: David Gibson Date: Thu, 14 Apr 2011 22:32:06 +0000 Subject: powerpc: Add WSP platform Add a platform for the Wire Speed Processor, based on the PPC A2. This includes code for the ICS & OPB interrupt controllers, as well as a SCOM backend, and SCOM based cpu bringup. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: David Gibson Signed-off-by: Jack Miller Signed-off-by: Ian Munsie Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/exceptions-64e.S | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/exceptions-64e.S b/arch/powerpc/kernel/exceptions-64e.S index 23bd83b20be4..c98e9d260621 100644 --- a/arch/powerpc/kernel/exceptions-64e.S +++ b/arch/powerpc/kernel/exceptions-64e.S @@ -864,6 +864,20 @@ have_hes: * that will have to be made dependent on whether we are running under * a hypervisor I suppose. */ + + /* BEWARE, MAGIC + * This code is called as an ordinary function on the boot CPU. But to + * avoid duplication, this code is also used in SCOM bringup of + * secondary CPUs. We read the code between the initial_tlb_code_start + * and initial_tlb_code_end labels one instruction at a time and RAM it + * into the new core via SCOM. That doesn't process branches, so there + * must be none between those two labels. It also means if this code + * ever takes any parameters, the SCOM code must also be updated to + * provide them. + */ + .globl a2_tlbinit_code_start +a2_tlbinit_code_start: + ori r11,r3,MAS0_WQ_ALLWAYS oris r11,r11,MAS0_ESEL(3)@h /* Use way 3: workaround A2 erratum 376 */ mtspr SPRN_MAS0,r11 @@ -880,6 +894,9 @@ have_hes: /* Write the TLB entry */ tlbwe + .globl a2_tlbinit_after_linear_map +a2_tlbinit_after_linear_map: + /* Now we branch the new virtual address mapped by this entry */ LOAD_REG_IMMEDIATE(r3,1f) mtctr r3 @@ -931,10 +948,16 @@ have_hes: cmpw r3,r9 blt 2b + .globl a2_tlbinit_after_iprot_flush +a2_tlbinit_after_iprot_flush: + PPC_TLBILX(0,0,0) sync isync + .globl a2_tlbinit_code_end +a2_tlbinit_code_end: + /* We translate LR and return */ mflr r3 tovirt(r3,r3) -- cgit v1.2.3 From a0496d450ab8c17f6c4d86979b1f6ba486fe9365 Mon Sep 17 00:00:00 2001 From: Jack Miller Date: Thu, 14 Apr 2011 22:32:08 +0000 Subject: powerpc: Add early debug for WSP platforms Signed-off-by: Jack Miller Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/exceptions-64e.S | 17 ++++++++++++ arch/powerpc/kernel/udbg.c | 2 ++ arch/powerpc/kernel/udbg_16550.c | 51 ++++++++++++++++++++++++++++++++++++ 3 files changed, 70 insertions(+) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/exceptions-64e.S b/arch/powerpc/kernel/exceptions-64e.S index c98e9d260621..4d0abb4930a1 100644 --- a/arch/powerpc/kernel/exceptions-64e.S +++ b/arch/powerpc/kernel/exceptions-64e.S @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -951,6 +952,22 @@ a2_tlbinit_after_linear_map: .globl a2_tlbinit_after_iprot_flush a2_tlbinit_after_iprot_flush: +#ifdef CONFIG_PPC_EARLY_DEBUG_WSP + /* Now establish early debug mappings if applicable */ + /* Restore the MAS0 we used for linear mapping load */ + mtspr SPRN_MAS0,r11 + + lis r3,(MAS1_VALID | MAS1_IPROT)@h + ori r3,r3,(BOOK3E_PAGESZ_4K << MAS1_TSIZE_SHIFT) + mtspr SPRN_MAS1,r3 + LOAD_REG_IMMEDIATE(r3, WSP_UART_VIRT | MAS2_I | MAS2_G) + mtspr SPRN_MAS2,r3 + LOAD_REG_IMMEDIATE(r3, WSP_UART_PHYS | MAS3_SR | MAS3_SW) + mtspr SPRN_MAS7_MAS3,r3 + /* re-use the MAS8 value from the linear mapping */ + tlbwe +#endif /* CONFIG_PPC_EARLY_DEBUG_WSP */ + PPC_TLBILX(0,0,0) sync isync diff --git a/arch/powerpc/kernel/udbg.c b/arch/powerpc/kernel/udbg.c index e39cad83c884..23d65abbedce 100644 --- a/arch/powerpc/kernel/udbg.c +++ b/arch/powerpc/kernel/udbg.c @@ -62,6 +62,8 @@ void __init udbg_early_init(void) udbg_init_cpm(); #elif defined(CONFIG_PPC_EARLY_DEBUG_USBGECKO) udbg_init_usbgecko(); +#elif defined(CONFIG_PPC_EARLY_DEBUG_WSP) + udbg_init_wsp(); #endif #ifdef CONFIG_PPC_EARLY_DEBUG diff --git a/arch/powerpc/kernel/udbg_16550.c b/arch/powerpc/kernel/udbg_16550.c index baa33a7517bc..6837f839ab78 100644 --- a/arch/powerpc/kernel/udbg_16550.c +++ b/arch/powerpc/kernel/udbg_16550.c @@ -11,6 +11,7 @@ #include #include #include +#include extern u8 real_readb(volatile u8 __iomem *addr); extern void real_writeb(u8 data, volatile u8 __iomem *addr); @@ -298,3 +299,53 @@ void __init udbg_init_40x_realmode(void) udbg_getc_poll = NULL; } #endif /* CONFIG_PPC_EARLY_DEBUG_40x */ + +#ifdef CONFIG_PPC_EARLY_DEBUG_WSP +static void udbg_wsp_flush(void) +{ + if (udbg_comport) { + while ((readb(&udbg_comport->lsr) & LSR_THRE) == 0) + /* wait for idle */; + } +} + +static void udbg_wsp_putc(char c) +{ + if (udbg_comport) { + if (c == '\n') + udbg_wsp_putc('\r'); + udbg_wsp_flush(); + writeb(c, &udbg_comport->thr); eieio(); + } +} + +static int udbg_wsp_getc(void) +{ + if (udbg_comport) { + while ((readb(&udbg_comport->lsr) & LSR_DR) == 0) + ; /* wait for char */ + return readb(&udbg_comport->rbr); + } + return -1; +} + +static int udbg_wsp_getc_poll(void) +{ + if (udbg_comport) + if (readb(&udbg_comport->lsr) & LSR_DR) + return readb(&udbg_comport->rbr); + return -1; +} + +void __init udbg_init_wsp(void) +{ + udbg_comport = (struct NS16550 __iomem *)WSP_UART_VIRT; + + udbg_init_uart(udbg_comport, 57600, 50000000); + + udbg_putc = udbg_wsp_putc; + udbg_flush = udbg_wsp_flush; + udbg_getc = udbg_wsp_getc; + udbg_getc_poll = udbg_wsp_getc_poll; +} +#endif /* CONFIG_PPC_EARLY_DEBUG_WSP */ -- cgit v1.2.3 From 69e3cea8d5fd52677f2b6219542d0f8b53fe4c80 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Thu, 19 May 2011 13:07:12 +1000 Subject: powerpc/smp: Make start_secondary_resume available to all CPU variants This should fix SMP & Hotplug builds on FSL BookE and 476 Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/head_32.S | 9 --------- arch/powerpc/kernel/misc_32.S | 11 +++++++++++ 2 files changed, 11 insertions(+), 9 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/head_32.S b/arch/powerpc/kernel/head_32.S index c5c24beb8387..98c4b29a56f4 100644 --- a/arch/powerpc/kernel/head_32.S +++ b/arch/powerpc/kernel/head_32.S @@ -890,15 +890,6 @@ __secondary_start: mtspr SPRN_SRR1,r4 SYNC RFI - -_GLOBAL(start_secondary_resume) - /* Reset stack */ - rlwinm r1,r1,0,0,(31-THREAD_SHIFT) /* current_thread_info() */ - addi r1,r1,THREAD_SIZE-STACK_FRAME_OVERHEAD - li r3,0 - std r3,0(r1) /* Zero the stack frame pointer */ - bl start_secondary - b . #endif /* CONFIG_SMP */ #ifdef CONFIG_KVM_BOOK3S_HANDLER diff --git a/arch/powerpc/kernel/misc_32.S b/arch/powerpc/kernel/misc_32.S index 094bd9821ad4..402560e957bd 100644 --- a/arch/powerpc/kernel/misc_32.S +++ b/arch/powerpc/kernel/misc_32.S @@ -694,6 +694,17 @@ _GLOBAL(kernel_thread) addi r1,r1,16 blr +#ifdef CONFIG_SMP +_GLOBAL(start_secondary_resume) + /* Reset stack */ + rlwinm r1,r1,0,0,(31-THREAD_SHIFT) /* current_thread_info() */ + addi r1,r1,THREAD_SIZE-STACK_FRAME_OVERHEAD + li r3,0 + std r3,0(r1) /* Zero the stack frame pointer */ + bl start_secondary + b . +#endif /* CONFIG_SMP */ + /* * This routine is just here to keep GCC happy - sigh... */ -- cgit v1.2.3 From 35d215fbe4f4d3569f2adabd8d53510a26ecb9c5 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Sun, 24 Apr 2011 15:04:31 +0000 Subject: powerpc/kexec: Fix build failure on 32-bit SMP Commit b987812b3fcaf70fdf0037589e5d2f5f2453e6ce left crash_kexec_wait_realmode() undefined for UP. Commit 7c7a81b53e581d727d069cc45df5510516faac31 defined it for UP but left it undefined for 32-bit SMP. Seems like people are getting confused by nested #ifdef's, so move the definitions of crash_kexec_wait_realmode() after the #ifdef CONFIG_SMP section. Compile-tested with 32-bit UP, 32-bit SMP and 64-bit SMP configurations. Signed-off-by: Ben Hutchings Tested-by: Paul Gortmaker Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/crash.c | 59 +++++++++++++++++++++++---------------------- 1 file changed, 30 insertions(+), 29 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/crash.c b/arch/powerpc/kernel/crash.c index 5b5e1f002a8e..5eb0ee37f6cd 100644 --- a/arch/powerpc/kernel/crash.c +++ b/arch/powerpc/kernel/crash.c @@ -162,34 +162,6 @@ static void crash_kexec_prepare_cpus(int cpu) /* Leave the IPI callback set */ } -/* wait for all the CPUs to hit real mode but timeout if they don't come in */ -#ifdef CONFIG_PPC_STD_MMU_64 -static void crash_kexec_wait_realmode(int cpu) -{ - unsigned int msecs; - int i; - - msecs = 10000; - for (i=0; i < NR_CPUS && msecs > 0; i++) { - if (i == cpu) - continue; - - while (paca[i].kexec_state < KEXEC_STATE_REAL_MODE) { - barrier(); - if (!cpu_possible(i)) { - break; - } - if (!cpu_online(i)) { - break; - } - msecs--; - mdelay(1); - } - } - mb(); -} -#endif /* CONFIG_PPC_STD_MMU_64 */ - /* * This function will be called by secondary cpus or by kexec cpu * if soft-reset is activated to stop some CPUs. @@ -234,7 +206,6 @@ void crash_kexec_secondary(struct pt_regs *regs) } #else /* ! CONFIG_SMP */ -static inline void crash_kexec_wait_realmode(int cpu) {} static void crash_kexec_prepare_cpus(int cpu) { @@ -257,6 +228,36 @@ void crash_kexec_secondary(struct pt_regs *regs) } #endif /* CONFIG_SMP */ +/* wait for all the CPUs to hit real mode but timeout if they don't come in */ +#if defined(CONFIG_SMP) && defined(CONFIG_PPC_STD_MMU_64) +static void crash_kexec_wait_realmode(int cpu) +{ + unsigned int msecs; + int i; + + msecs = 10000; + for (i=0; i < NR_CPUS && msecs > 0; i++) { + if (i == cpu) + continue; + + while (paca[i].kexec_state < KEXEC_STATE_REAL_MODE) { + barrier(); + if (!cpu_possible(i)) { + break; + } + if (!cpu_online(i)) { + break; + } + msecs--; + mdelay(1); + } + } + mb(); +} +#else +static inline void crash_kexec_wait_realmode(int cpu) {} +#endif /* CONFIG_SMP && CONFIG_PPC_STD_MMU_64 */ + /* * Register a function to be called on shutdown. Only use this if you * can't reset your device in the second kernel. -- cgit v1.2.3 From c560bbceaf6b06e52f1ef20131b76a3fdc0a2c19 Mon Sep 17 00:00:00 2001 From: kerstin jonsson Date: Tue, 17 May 2011 23:57:11 +0000 Subject: powerpc/4xx: Fix regression in SMP on 476 commit c56e58537d504706954a06570b4034c04e5b7500 breaks SMP support in PPC_47x chip. secondary_ti must be set to current thread info before callin kick_cpu or else start_secondary_47x will jump into void when trying to return to c-code. In the current setup secondary_ti is initialized before the CPU idle task is started and only the boot core will start. I am not sure this is the correct solution, but it makes SMP possible in my chip. Note! The HOTPLUG support probably need some fixing to, There is no trampoline code available in head_44x.S - start_secondary_resume? Signed-off-by: Kerstin Jonsson Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/smp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index cbdbb14be4b0..f2dcab7aadc8 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -410,8 +410,6 @@ int __cpuinit __cpu_up(unsigned int cpu) { int rc, c; - secondary_ti = current_set[cpu]; - if (smp_ops == NULL || (smp_ops->cpu_bootable && !smp_ops->cpu_bootable(cpu))) return -EINVAL; @@ -421,6 +419,8 @@ int __cpuinit __cpu_up(unsigned int cpu) if (rc) return rc; + secondary_ti = current_set[cpu]; + /* Make sure callin-map entry is 0 (can be leftover a CPU * hotplug */ -- cgit v1.2.3 From 93395febdb51ed812ac1003852f537177a172aad Mon Sep 17 00:00:00 2001 From: Justin Mattock Date: Tue, 5 Apr 2011 06:58:22 +0000 Subject: powerpc: Remove unused config in the Makefile The patch below removes an unused config variable found by using a kernel cleanup script. Note: I did try to cross compile these but hit erros while doing so.. (gcc is not setup to cross compile) and am unsure if anymore needs to be done. Please have a look if/when anybody has free time. Signed-off-by: Justin P. Mattock Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/Makefile | 1 - 1 file changed, 1 deletion(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile index 82e0bed0650d..9aab36312572 100644 --- a/arch/powerpc/kernel/Makefile +++ b/arch/powerpc/kernel/Makefile @@ -78,7 +78,6 @@ obj-$(CONFIG_PPC_FSL_BOOK3E) += cpu_setup_fsl_booke.o dbell.o obj-$(CONFIG_PPC_BOOK3E_64) += dbell.o extra-y := head_$(CONFIG_WORD_SIZE).o -extra-$(CONFIG_PPC_BOOK3E_32) := head_new_booke.o extra-$(CONFIG_40x) := head_40x.o extra-$(CONFIG_44x) := head_44x.o extra-$(CONFIG_FSL_BOOKE) := head_fsl_booke.o -- cgit v1.2.3 From d988f0e3f84cb8a4f85ccdbca6f6fefcc37bedcb Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Sun, 8 May 2011 21:18:38 +0000 Subject: powerpc: Simplify 4k/64k copy_page logic To make it easier to add optimised versions of copy_page, remove the 4kB loop for 64kB pages and just do all the work in copy_page. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/ppc_ksyms.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/ppc_ksyms.c b/arch/powerpc/kernel/ppc_ksyms.c index ef3ef566235e..7d28f540200c 100644 --- a/arch/powerpc/kernel/ppc_ksyms.c +++ b/arch/powerpc/kernel/ppc_ksyms.c @@ -54,7 +54,6 @@ extern void single_step_exception(struct pt_regs *regs); extern int sys_sigreturn(struct pt_regs *regs); EXPORT_SYMBOL(clear_pages); -EXPORT_SYMBOL(copy_page); EXPORT_SYMBOL(ISA_DMA_THRESHOLD); EXPORT_SYMBOL(DMA_MODE_READ); EXPORT_SYMBOL(DMA_MODE_WRITE); @@ -88,9 +87,7 @@ EXPORT_SYMBOL(__copy_tofrom_user); EXPORT_SYMBOL(__clear_user); EXPORT_SYMBOL(__strncpy_from_user); EXPORT_SYMBOL(__strnlen_user); -#ifdef CONFIG_PPC64 -EXPORT_SYMBOL(copy_4K_page); -#endif +EXPORT_SYMBOL(copy_page); #if defined(CONFIG_PCI) && defined(CONFIG_PPC32) EXPORT_SYMBOL(isa_io_base); -- cgit v1.2.3 From ba00ce1d6e08ad06f19f2ac53fd5c60bbe3fbeeb Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Sun, 8 May 2011 21:20:19 +0000 Subject: powerpc: Remove static branch hint in giveup_altivec A static branch hint will override dynamic branch prediction on recent POWER CPUs. Since we are about to use more altivec in the kernel remove the static hint in giveup_altivec that assumes a userspace task is using altivec. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/vector.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/vector.S b/arch/powerpc/kernel/vector.S index 9de6f396cf85..4d5a3edff49e 100644 --- a/arch/powerpc/kernel/vector.S +++ b/arch/powerpc/kernel/vector.S @@ -102,7 +102,7 @@ _GLOBAL(giveup_altivec) MTMSRD(r5) /* enable use of VMX now */ isync PPC_LCMPI 0,r3,0 - beqlr- /* if no previous owner, done */ + beqlr /* if no previous owner, done */ addi r3,r3,THREAD /* want THREAD of task */ PPC_LL r5,PT_REGS(r3) PPC_LCMPI 0,r5,0 -- cgit v1.2.3 From f5f0307f42d39a51a925ca4841f76a2f2ea330ff Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Sun, 8 May 2011 21:36:44 +0000 Subject: powerpc: Improve scheduling of system call entry instructions After looking at our system call path, Mary Brown suggested that we should put all mfspr SRR* instructions before any mtspr SRR*. To test this I used a very simple null syscall (actually getppid) testcase at http://ozlabs.org/~anton/junkcode/null_syscall.c I tested with the following changes against the pseries_defconfig: CONFIG_VIRT_CPU_ACCOUNTING=n CONFIG_AUDIT=n to remove the overhead of virtual CPU accounting and syscall auditing. POWER6: baseline: mean = 757.2 cycles sd = 2.108 modified: mean = 759.1 cycles sd = 2.020 POWER7: baseline: mean = 411.4 cycles sd = 0.138 modified: mean = 404.1 cycles sd = 0.109 So we have 1.77% improvement on POWER7 which looks significant. The POWER6 suggest a 0.25% slowdown, but the results are within 1 standard deviation and may be in the noise. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/exceptions-64s.S | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S index 0ec3b42717d7..a85f4874cba7 100644 --- a/arch/powerpc/kernel/exceptions-64s.S +++ b/arch/powerpc/kernel/exceptions-64s.S @@ -211,11 +211,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_REAL_LE) mr r9,r13 GET_PACA(r13) mfspr r11,SPRN_SRR0 - ld r12,PACAKBASE(r13) - ld r10,PACAKMSR(r13) - LOAD_HANDLER(r12, system_call_entry) - mtspr SPRN_SRR0,r12 mfspr r12,SPRN_SRR1 + ld r10,PACAKBASE(r13) + LOAD_HANDLER(r10, system_call_entry) + mtspr SPRN_SRR0,r10 + ld r10,PACAKMSR(r13) mtspr SPRN_SRR1,r10 rfid b . /* prevent speculative execution */ -- cgit v1.2.3 From 3d2cea732d68aa270c360f55d8669820ebce188a Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:28:33 +0000 Subject: powerpc/kexec: Fix memory corruption from unallocated slaves Commit 1fc711f7ffb01089efc58042cfdbac8573d1b59a (powerpc/kexec: Fix race in kexec shutdown) moved the write to signal the cpu had exited the kernel from before the transition to real mode in kexec_smp_wait to kexec_wait. Unfornately it missed that kexec_wait is used both by cpus leaving the kernel and by secondary slave cpus that were not allocated a paca for what ever reason -- they could be beyond nr_cpus or not described in the current device tree for whatever reason (for example, kexec-load was not refreshed after a cpu hotplug operation). Cpus coming through that path they will write to paca[NR_CPUS] which is beyond the space allocated for the paca data and overwrite memory not allocated to pacas but very likely still real mode accessable). Move the write back to kexec_smp_wait, which is used only by cpus that found their paca, but after the transition to real mode. Signed-off-by: Milton Miller Cc: # (1fc711f was backported to 2.6.32) Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/misc_64.S | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/misc_64.S b/arch/powerpc/kernel/misc_64.S index 206a321a71d3..e89df59cdc5a 100644 --- a/arch/powerpc/kernel/misc_64.S +++ b/arch/powerpc/kernel/misc_64.S @@ -462,7 +462,8 @@ _GLOBAL(disable_kernel_fp) * wait for the flag to change, indicating this kernel is going away but * the slave code for the next one is at addresses 0 to 100. * - * This is used by all slaves. + * This is used by all slaves, even those that did not find a matching + * paca in the secondary startup code. * * Physical (hardware) cpu id should be in r3. */ @@ -471,10 +472,6 @@ _GLOBAL(kexec_wait) 1: mflr r5 addi r5,r5,kexec_flag-1b - li r4,KEXEC_STATE_REAL_MODE - stb r4,PACAKEXECSTATE(r13) - SYNC - 99: HMT_LOW #ifdef CONFIG_KEXEC /* use no memory without kexec */ lwz r4,0(r5) @@ -499,11 +496,17 @@ kexec_flag: * * get phys id from paca * switch to real mode + * mark the paca as no longer used * join other cpus in kexec_wait(phys_id) */ _GLOBAL(kexec_smp_wait) lhz r3,PACAHWCPUID(r13) bl real_mode + + li r4,KEXEC_STATE_REAL_MODE + stb r4,PACAKEXECSTATE(r13) + SYNC + b .kexec_wait /* -- cgit v1.2.3 From 768d18ad6d5e600d911f9499ca287d6986d8d81b Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:28:37 +0000 Subject: powerpc: Don't search for paca in freed memory Starting with 1426d5a3bd07589534286375998c0c8c6fdc5260 (powerpc: Dynamically allocate pacas) we free the memory for pacas beyond cpu_possible, but we failed to update the loop the secondary cpus use to find their paca. If the system has running cpu threads for which the kernel did not allocate a paca for they will search the memory that was freed. For instance this could happen when the device tree for a kdump kernel was not updated after a cpu hotplug, or the kernel is running with more cpus than the kernel was configured. Since c1854e00727f50f7ac99e98d26ece04c087ef785 (powerpc: Set nr_cpu_ids early and use it to free PACAs) we set nr_cpu_ids before telling the cpus to advance, so use that to limit the search. We can't reference nr_cpu_ids without CONFIG_SMP because it is defined as 1 instead of a memory location, but any extra threads should be sent to kexec_wait in that case anyways, so make that explicit and remove the search loop for UP. Note to stable: The fix also requires c1854e00727f50f7ac99e98d26ece04c087ef785 (powerpc: Set nr_cpu_ids early and use it to free PACAs) to function. Also 9d07bc841c9779b4d7902e417f4e509996ce805d (Properly handshake CPUs going out of boot spin loop) affects the second chunk, specifically the branch target was 3b before and is 4b after that patch, and there was a blank line before the #ifdef CONFIG_SMP that was removed Cc: # .34.x: c1854e0072 powerpc: Set nr_cpu_ids early Cc: # .34.x Signed-off-by: Milton Miller Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/head_64.S | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S index 73d6e9afcdf1..ba504099844a 100644 --- a/arch/powerpc/kernel/head_64.S +++ b/arch/powerpc/kernel/head_64.S @@ -218,13 +218,19 @@ generic_secondary_common_init: */ LOAD_REG_ADDR(r13, paca) /* Load paca pointer */ ld r13,0(r13) /* Get base vaddr of paca array */ +#ifndef CONFIG_SMP + addi r13,r13,PACA_SIZE /* know r13 if used accidentally */ + b .kexec_wait /* wait for next kernel if !SMP */ +#else + LOAD_REG_ADDR(r7, nr_cpu_ids) /* Load nr_cpu_ids address */ + lwz r7,0(r7) /* also the max paca allocated */ li r5,0 /* logical cpu id */ 1: lhz r6,PACAHWCPUID(r13) /* Load HW procid from paca */ cmpw r6,r24 /* Compare to our id */ beq 2f addi r13,r13,PACA_SIZE /* Loop to next PACA on miss */ addi r5,r5,1 - cmpwi r5,NR_CPUS + cmpw r5,r7 /* Check if more pacas exist */ blt 1b mr r3,r24 /* not found, copy phys to r3 */ @@ -259,9 +265,6 @@ generic_secondary_common_init: 4: HMT_LOW lbz r23,PACAPROCSTART(r13) /* Test if this processor should */ /* start. */ -#ifndef CONFIG_SMP - b 4b /* Never go on non-SMP */ -#else cmpwi 0,r23,0 beq 4b /* Loop until told to go */ @@ -273,7 +276,7 @@ generic_secondary_common_init: subi r1,r1,STACK_FRAME_OVERHEAD b __secondary_start -#endif +#endif /* SMP */ /* * Turn the MMU off. -- cgit v1.2.3 From bd9e5eefecb3d69018bb95796298019d309cbec8 Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:28:41 +0000 Subject: powerpc/kdump64: Don't reference freed memory as pacas Starting with 1426d5a3bd07589534286375998c0c8c6fdc5260 (powerpc: Dynamically allocate pacas) the space for pacas beyond cpu_possible is freed, but we failed to update the loop in crash.c. Since c1854e00727f50f7ac99e98d26ece04c087ef785 (powerpc: Set nr_cpu_ids early and use it to free PACAs) the number of pacas allocated is always nr_cpu_ids. Signed-off-by: Milton Miller Cc: # .34.x Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/crash.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/crash.c b/arch/powerpc/kernel/crash.c index 21f2c781ded1..4e6ee944495a 100644 --- a/arch/powerpc/kernel/crash.c +++ b/arch/powerpc/kernel/crash.c @@ -236,7 +236,7 @@ static void crash_kexec_wait_realmode(int cpu) int i; msecs = 10000; - for (i=0; i < NR_CPUS && msecs > 0; i++) { + for (i=0; i < nr_cpu_ids && msecs > 0; i++) { if (i == cpu) continue; -- cgit v1.2.3 From 8657ae28ddd34db0f52b0730a6a25992c0173264 Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:28:48 +0000 Subject: powerpc: Respect nr_cpu_ids when calling set_cpu_possible and set_cpu_present We should not set cpus above nr_cpu_ids to possible. While we will trigger a warning with CONFIG_CPUMASK_DEBUG, even then the mask initializers will set the bits beyond what the iterators check and cause nr_cpu_ids to increase. Respecting nr_cpu_ids during setup will allow us to use it in our initial paca allocation. It can be reduced from NR_CPUS by the existing early param nr_cpus=, which was added in 2b633e3fac5efada088b57d31e65401f22bcc18f (smp: Use nr_cpus= to set nr_cpu_ids early). We already call parse_early_parms between finding the command line and allocating the pacas. Signed-off-by: Milton Miller Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/setup-common.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c index 1475df6e403f..fce759ba31f3 100644 --- a/arch/powerpc/kernel/setup-common.c +++ b/arch/powerpc/kernel/setup-common.c @@ -404,7 +404,7 @@ static void __init cpu_init_thread_core_maps(int tpc) * cpu_present_mask * * Having the possible map set up early allows us to restrict allocations - * of things like irqstacks to num_possible_cpus() rather than NR_CPUS. + * of things like irqstacks to nr_cpu_ids rather than NR_CPUS. * * We do not initialize the online map here; cpus set their own bits in * cpu_online_mask as they come up. @@ -424,7 +424,7 @@ void __init smp_setup_cpu_maps(void) DBG("smp_setup_cpu_maps()\n"); - while ((dn = of_find_node_by_type(dn, "cpu")) && cpu < NR_CPUS) { + while ((dn = of_find_node_by_type(dn, "cpu")) && cpu < nr_cpu_ids) { const int *intserv; int j, len; @@ -443,7 +443,7 @@ void __init smp_setup_cpu_maps(void) intserv = &cpu; /* assume logical == phys */ } - for (j = 0; j < nthreads && cpu < NR_CPUS; j++) { + for (j = 0; j < nthreads && cpu < nr_cpu_ids; j++) { DBG(" thread %d -> cpu %d (hard id %d)\n", j, cpu, intserv[j]); set_cpu_present(cpu, true); @@ -483,12 +483,12 @@ void __init smp_setup_cpu_maps(void) if (cpu_has_feature(CPU_FTR_SMT)) maxcpus *= nthreads; - if (maxcpus > NR_CPUS) { + if (maxcpus > nr_cpu_ids) { printk(KERN_WARNING "Partition configured for %d cpus, " "operating system maximum is %d.\n", - maxcpus, NR_CPUS); - maxcpus = NR_CPUS; + maxcpus, nr_cpu_ids); + maxcpus = nr_cpu_ids; } else printk(KERN_INFO "Partition configured for %d cpus.\n", maxcpus); -- cgit v1.2.3 From 2cd947f1757fb937806535be13caf2ddd813d60b Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:28:52 +0000 Subject: powerpc: Use nr_cpu_ids in initial paca allocation Now that we never set a cpu above nr_cpu_ids possible we can limit our initial paca allocation to nr_cpu_ids. We can then clamp the number of cpus in platforms/iseries/setup.c. Signed-off-by: Milton Miller Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/paca.c | 17 ++++++----------- 1 file changed, 6 insertions(+), 11 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/paca.c b/arch/powerpc/kernel/paca.c index 102244edecf0..efeb88184182 100644 --- a/arch/powerpc/kernel/paca.c +++ b/arch/powerpc/kernel/paca.c @@ -7,7 +7,7 @@ * 2 of the License, or (at your option) any later version. */ -#include +#include #include #include @@ -178,7 +178,7 @@ static int __initdata paca_size; void __init allocate_pacas(void) { - int nr_cpus, cpu, limit; + int cpu, limit; /* * We can't take SLB misses on the paca, and we want to access them @@ -190,23 +190,18 @@ void __init allocate_pacas(void) if (firmware_has_feature(FW_FEATURE_ISERIES)) limit = min(limit, HvPagesToMap * HVPAGESIZE); - nr_cpus = NR_CPUS; - /* On iSeries we know we can never have more than 64 cpus */ - if (firmware_has_feature(FW_FEATURE_ISERIES)) - nr_cpus = min(64, nr_cpus); - - paca_size = PAGE_ALIGN(sizeof(struct paca_struct) * nr_cpus); + paca_size = PAGE_ALIGN(sizeof(struct paca_struct) * nr_cpu_ids); paca = __va(memblock_alloc_base(paca_size, PAGE_SIZE, limit)); memset(paca, 0, paca_size); printk(KERN_DEBUG "Allocated %u bytes for %d pacas at %p\n", - paca_size, nr_cpus, paca); + paca_size, nr_cpu_ids, paca); - allocate_lppacas(nr_cpus, limit); + allocate_lppacas(nr_cpu_ids, limit); /* Can't use for_each_*_cpu, as they aren't functional yet */ - for (cpu = 0; cpu < nr_cpus; cpu++) + for (cpu = 0; cpu < nr_cpu_ids; cpu++) initialise_paca(&paca[cpu], cpu); } -- cgit v1.2.3 From aa79bc2167104581cc1d77762394f2c01d3bf3f3 Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:28:55 +0000 Subject: powerpc: Call no-longer static setup_nr_cpu_ids instead of replicating it c1854e00727f50f7ac99e98d26ece04c087ef785 (powerpc: Set nr_cpu_ids early and use it to free PACAs) copied the formerly static setup_nr_cpu_ids from init/main.c but 34db18a054c600b6f81787165669dc572fe4de25 (smp: move smp setup functions to kernel/smp.c) moved it to kernel/smp.c with a declaration in include/linux/smp.h, so we can call it instead of replicating it. Signed-off-by: Milton Miller Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/setup-common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c index fce759ba31f3..ef33a084fcf4 100644 --- a/arch/powerpc/kernel/setup-common.c +++ b/arch/powerpc/kernel/setup-common.c @@ -510,7 +510,7 @@ void __init smp_setup_cpu_maps(void) cpu_init_thread_core_maps(nthreads); /* Now that possible cpus are set, set nr_cpu_ids for later use */ - nr_cpu_ids = find_last_bit(cpumask_bits(cpu_possible_mask),NR_CPUS) + 1; + setup_nr_cpu_ids(); free_unused_pacas(); } -- cgit v1.2.3 From e04763713286b1e00e1c2a33fe2741caf9470f2b Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:29:06 +0000 Subject: powerpc: Remove call sites of MSG_ALL_BUT_SELF The only user of MSG_ALL_BUT_SELF in the whole kernel tree is powerpc, and it only uses it to start the debugger. Both debuggers always call smp_send_debugger_break with MSG_ALL_BUT_SELF, and only mpic can do anything more optimal than a loop over all online cpus, but all message passing implementations have to code for this special delivery target. Convert smp_send_debugger_break to take void and loop calling the smp_ops message_pass function for each of the other cpus in the online cpumask. Use raw_smp_processor_id() because we are either entering the debugger or trying to start kdump and the additional warning it not useful were it to trigger. Signed-off-by: Milton Miller Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/kgdb.c | 2 +- arch/powerpc/kernel/smp.c | 19 +++++++++++++------ 2 files changed, 14 insertions(+), 7 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/kgdb.c b/arch/powerpc/kernel/kgdb.c index 42850ee00ada..bd9d35f59cf4 100644 --- a/arch/powerpc/kernel/kgdb.c +++ b/arch/powerpc/kernel/kgdb.c @@ -109,7 +109,7 @@ static int kgdb_call_nmi_hook(struct pt_regs *regs) #ifdef CONFIG_SMP void kgdb_roundup_cpus(unsigned long flags) { - smp_send_debugger_break(MSG_ALL_BUT_SELF); + smp_send_debugger_break(); } #endif diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index 87517ab6d365..b74411446922 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -218,11 +218,18 @@ void arch_send_call_function_ipi_mask(const struct cpumask *mask) smp_ops->message_pass(cpu, PPC_MSG_CALL_FUNCTION); } -#ifdef CONFIG_DEBUGGER -void smp_send_debugger_break(int cpu) +#if defined(CONFIG_DEBUGGER) || defined(CONFIG_KEXEC) +void smp_send_debugger_break(void) { - if (likely(smp_ops)) - smp_ops->message_pass(cpu, PPC_MSG_DEBUGGER_BREAK); + int cpu; + int me = raw_smp_processor_id(); + + if (unlikely(!smp_ops)) + return; + + for_each_online_cpu(cpu) + if (cpu != me) + smp_ops->message_pass(cpu, PPC_MSG_DEBUGGER_BREAK); } #endif @@ -230,9 +237,9 @@ void smp_send_debugger_break(int cpu) void crash_send_ipi(void (*crash_ipi_callback)(struct pt_regs *)) { crash_ipi_function_ptr = crash_ipi_callback; - if (crash_ipi_callback && smp_ops) { + if (crash_ipi_callback) { mb(); - smp_ops->message_pass(MSG_ALL_BUT_SELF, PPC_MSG_DEBUGGER_BREAK); + smp_send_debugger_break(); } } #endif -- cgit v1.2.3 From f1072939b6dd01d038d47db0bdc01b33e5f90f28 Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:29:10 +0000 Subject: powerpc: Remove checks for MSG_ALL and MSG_ALL_BUT_SELF Now that smp_ops->smp_message_pass is always called with an (online) cpu number for the target remove the checks for MSG_ALL and MSG_ALL_BUT_SELF. Signed-off-by: Milton Miller Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/dbell.c | 29 +++++------------------------ 1 file changed, 5 insertions(+), 24 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/dbell.c b/arch/powerpc/kernel/dbell.c index 3307a52d797f..e49b24c84133 100644 --- a/arch/powerpc/kernel/dbell.c +++ b/arch/powerpc/kernel/dbell.c @@ -34,32 +34,13 @@ void doorbell_setup_this_cpu(void) info->tag = mfspr(SPRN_PIR) & 0x3fff; } -void doorbell_message_pass(int target, int msg) +void doorbell_message_pass(int cpu, int msg) { struct doorbell_cpu_info *info; - int i; - - if (target < NR_CPUS) { - info = &per_cpu(doorbell_cpu_info, target); - set_bit(msg, &info->messages); - ppc_msgsnd(PPC_DBELL, 0, info->tag); - } - else if (target == MSG_ALL_BUT_SELF) { - for_each_online_cpu(i) { - if (i == smp_processor_id()) - continue; - info = &per_cpu(doorbell_cpu_info, i); - set_bit(msg, &info->messages); - ppc_msgsnd(PPC_DBELL, 0, info->tag); - } - } - else { /* target == MSG_ALL */ - for_each_online_cpu(i) { - info = &per_cpu(doorbell_cpu_info, i); - set_bit(msg, &info->messages); - } - ppc_msgsnd(PPC_DBELL, PPC_DBELL_MSG_BRDCAST, 0); - } + + info = &per_cpu(doorbell_cpu_info, cpu); + set_bit(msg, &info->messages); + ppc_msgsnd(PPC_DBELL, 0, info->tag); } void doorbell_exception(struct pt_regs *regs) -- cgit v1.2.3 From a56555e573d3740d588d912aada506d57759cf5d Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:29:24 +0000 Subject: powerpc: Remove alloc_maybe_bootmem for zalloc version Replace all remaining callers of alloc_maybe_bootmem with zalloc_maybe_bootmem. The callsite in pci_dn is followed with a memset to clear the memory, and not zeroing at the other callsites in the celleb fake pci code could lead to following uninitialized memory as pointers or even freeing said pointers on error paths. Signed-off-by: Milton Miller Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/pci_dn.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/pci_dn.c b/arch/powerpc/kernel/pci_dn.c index d225d99fe39d..6baabc13306a 100644 --- a/arch/powerpc/kernel/pci_dn.c +++ b/arch/powerpc/kernel/pci_dn.c @@ -43,10 +43,9 @@ void * __devinit update_dn_pci_info(struct device_node *dn, void *data) const u32 *regs; struct pci_dn *pdn; - pdn = alloc_maybe_bootmem(sizeof(*pdn), GFP_KERNEL); + pdn = zalloc_maybe_bootmem(sizeof(*pdn), GFP_KERNEL); if (pdn == NULL) return NULL; - memset(pdn, 0, sizeof(*pdn)); dn->data = pdn; pdn->node = dn; pdn->phb = phb; -- cgit v1.2.3 From 23d72bfd8f9f24aa9efafed3586a99f5669c23d7 Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:29:39 +0000 Subject: powerpc: Consolidate ipi message mux and demux Consolidate the mux and demux of ipi messages into smp.c and call a new smp_ops callback to actually trigger the ipi. The powerpc architecture code is optimised for having 4 distinct ipi triggers, which are mapped to 4 distinct messages (ipi many, ipi single, scheduler ipi, and enter debugger). However, several interrupt controllers only provide a single software triggered interrupt that can be delivered to each cpu. To resolve this limitation, each smp_ops implementation created a per-cpu variable that is manipulated with atomic bitops. Since these lines will be contended they are optimialy marked as shared_aligned and take a full cache line for each cpu. Distro kernels may have 2 or 3 of these in their config, each taking per-cpu space even though at most one will be in use. This consolidation removes smp_message_recv and replaces the single call actions cases with direct calls from the common message recognition loop. The complicated debugger ipi case with its muxed crash handling code is moved to debug_ipi_action which is now called from the demux code (instead of the multi-message action calling smp_message_recv). I put a call to reschedule_action to increase the likelyhood of correctly merging the anticipated scheduler_ipi() hook coming from the scheduler tree; that single required call can be inlined later. The actual message decode is a copy of the old pseries xics code with its memory barriers and cache line spacing, augmented with a per-cpu unsigned long based on the book-e doorbell code. The optional data is set via a callback from the implementation and is passed to the new cause-ipi hook along with the logical cpu number. While currently only the doorbell implemntation uses this data it should be almost zero cost to retrieve and pass it -- it adds a single register load for the argument from the same cache line to which we just completed a store and the register is dead on return from the call. I extended the data element from unsigned int to unsigned long in case some other code wanted to associate a pointer. The doorbell check_self is replaced by a call to smp_muxed_ipi_resend, conditioned on the CPU_DBELL feature. The ifdef guard could be relaxed to CONFIG_SMP but I left it with BOOKE for now. Also, the doorbell interrupt vector for book-e was not calling irq_enter and irq_exit, which throws off cpu accounting and causes code to not realize it is running in interrupt context. Add the missing calls. Signed-off-by: Milton Miller Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/dbell.c | 46 ++++------------------ arch/powerpc/kernel/irq.c | 4 +- arch/powerpc/kernel/smp.c | 94 ++++++++++++++++++++++++++++++--------------- 3 files changed, 73 insertions(+), 71 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/dbell.c b/arch/powerpc/kernel/dbell.c index e49b24c84133..2cc451aaaca7 100644 --- a/arch/powerpc/kernel/dbell.c +++ b/arch/powerpc/kernel/dbell.c @@ -13,65 +13,35 @@ #include #include #include -#include +#include #include #include #ifdef CONFIG_SMP -struct doorbell_cpu_info { - unsigned long messages; /* current messages bits */ - unsigned int tag; /* tag value */ -}; - -static DEFINE_PER_CPU(struct doorbell_cpu_info, doorbell_cpu_info); - void doorbell_setup_this_cpu(void) { - struct doorbell_cpu_info *info = &__get_cpu_var(doorbell_cpu_info); + unsigned long tag = mfspr(SPRN_PIR) & 0x3fff; - info->messages = 0; - info->tag = mfspr(SPRN_PIR) & 0x3fff; + smp_muxed_ipi_set_data(smp_processor_id(), tag); } -void doorbell_message_pass(int cpu, int msg) +void doorbell_cause_ipi(int cpu, unsigned long data) { - struct doorbell_cpu_info *info; - - info = &per_cpu(doorbell_cpu_info, cpu); - set_bit(msg, &info->messages); - ppc_msgsnd(PPC_DBELL, 0, info->tag); + ppc_msgsnd(PPC_DBELL, 0, data); } void doorbell_exception(struct pt_regs *regs) { struct pt_regs *old_regs = set_irq_regs(regs); - struct doorbell_cpu_info *info = &__get_cpu_var(doorbell_cpu_info); - int msg; - /* Warning: regs can be NULL when called from irq enable */ + irq_enter(); - if (!info->messages || (num_online_cpus() < 2)) - goto out; + smp_ipi_demux(); - for (msg = 0; msg < 4; msg++) - if (test_and_clear_bit(msg, &info->messages)) - smp_message_recv(msg); - -out: + irq_exit(); set_irq_regs(old_regs); } - -void doorbell_check_self(void) -{ - struct doorbell_cpu_info *info = &__get_cpu_var(doorbell_cpu_info); - - if (!info->messages) - return; - - ppc_msgsnd(PPC_DBELL, 0, info->tag); -} - #else /* CONFIG_SMP */ void doorbell_exception(struct pt_regs *regs) { diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index a81dd74414bf..826552cecebd 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -66,7 +66,6 @@ #include #include #include -#include #include #ifdef CONFIG_PPC64 @@ -160,7 +159,8 @@ notrace void arch_local_irq_restore(unsigned long en) #if defined(CONFIG_BOOKE) && defined(CONFIG_SMP) /* Check for pending doorbell interrupts and resend to ourself */ - doorbell_check_self(); + if (cpu_has_feature(CPU_FTR_DBELL)) + smp_muxed_ipi_resend(); #endif /* diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index b74411446922..fa8e8700064b 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -111,35 +111,6 @@ int __devinit smp_generic_kick_cpu(int nr) } #endif -void smp_message_recv(int msg) -{ - switch(msg) { - case PPC_MSG_CALL_FUNCTION: - generic_smp_call_function_interrupt(); - break; - case PPC_MSG_RESCHEDULE: - /* we notice need_resched on exit */ - break; - case PPC_MSG_CALL_FUNC_SINGLE: - generic_smp_call_function_single_interrupt(); - break; - case PPC_MSG_DEBUGGER_BREAK: - if (crash_ipi_function_ptr) { - crash_ipi_function_ptr(get_irq_regs()); - break; - } -#ifdef CONFIG_DEBUGGER - debugger_ipi(get_irq_regs()); - break; -#endif /* CONFIG_DEBUGGER */ - /* FALLTHROUGH */ - default: - printk("SMP %d: smp_message_recv(): unknown msg %d\n", - smp_processor_id(), msg); - break; - } -} - static irqreturn_t call_function_action(int irq, void *data) { generic_smp_call_function_interrupt(); @@ -158,9 +129,17 @@ static irqreturn_t call_function_single_action(int irq, void *data) return IRQ_HANDLED; } -static irqreturn_t debug_ipi_action(int irq, void *data) +irqreturn_t debug_ipi_action(int irq, void *data) { - smp_message_recv(PPC_MSG_DEBUGGER_BREAK); + if (crash_ipi_function_ptr) { + crash_ipi_function_ptr(get_irq_regs()); + return IRQ_HANDLED; + } + +#ifdef CONFIG_DEBUGGER + debugger_ipi(get_irq_regs()); +#endif /* CONFIG_DEBUGGER */ + return IRQ_HANDLED; } @@ -199,6 +178,59 @@ int smp_request_message_ipi(int virq, int msg) return err; } +struct cpu_messages { + unsigned long messages; /* current messages bits */ + unsigned long data; /* data for cause ipi */ +}; +static DEFINE_PER_CPU_SHARED_ALIGNED(struct cpu_messages, ipi_message); + +void smp_muxed_ipi_set_data(int cpu, unsigned long data) +{ + struct cpu_messages *info = &per_cpu(ipi_message, cpu); + + info->data = data; +} + +void smp_muxed_ipi_message_pass(int cpu, int msg) +{ + struct cpu_messages *info = &per_cpu(ipi_message, cpu); + unsigned long *tgt = &info->messages; + + set_bit(msg, tgt); + mb(); + smp_ops->cause_ipi(cpu, info->data); +} + +void smp_muxed_ipi_resend(void) +{ + struct cpu_messages *info = &__get_cpu_var(ipi_message); + unsigned long *tgt = &info->messages; + + if (*tgt) + smp_ops->cause_ipi(smp_processor_id(), info->data); +} + +irqreturn_t smp_ipi_demux(void) +{ + struct cpu_messages *info = &__get_cpu_var(ipi_message); + unsigned long *tgt = &info->messages; + + mb(); /* order any irq clear */ + while (*tgt) { + if (test_and_clear_bit(PPC_MSG_CALL_FUNCTION, tgt)) + generic_smp_call_function_interrupt(); + if (test_and_clear_bit(PPC_MSG_RESCHEDULE, tgt)) + reschedule_action(0, NULL); /* upcoming sched hook */ + if (test_and_clear_bit(PPC_MSG_CALL_FUNC_SINGLE, tgt)) + generic_smp_call_function_single_interrupt(); +#if defined(CONFIG_DEBUGGER) || defined(CONFIG_KEXEC) + if (test_and_clear_bit(PPC_MSG_DEBUGGER_BREAK, tgt)) + debug_ipi_action(0, NULL); +#endif + } + return IRQ_HANDLED; +} + void smp_send_reschedule(int cpu) { if (likely(smp_ops)) -- cgit v1.2.3 From 1ece355b6825b7c61d1dc39a5c6cf49dc746e193 Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:29:42 +0000 Subject: powerpc: Add kconfig for muxed smp ipi support Compile the new smp ipi mux and demux code only if a platform will make use of it. The new config is selected as required. The new cause_ipi smp op is only available conditionally to point out configs where the select is required; this makes setting the op an immediate fail instead of a deferred unresolved symbol at link. This also creates a new config for power surge powermac upgrade support that can be disabled in expert mode but is default on. I also removed the depends / default y on CONFIG_XICS since it is selected by PSERIES. Signed-off-by: Milton Miller Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/smp.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index fa8e8700064b..d76f7d7929be 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -178,6 +178,7 @@ int smp_request_message_ipi(int virq, int msg) return err; } +#ifdef CONFIG_PPC_SMP_MUXED_IPI struct cpu_messages { unsigned long messages; /* current messages bits */ unsigned long data; /* data for cause ipi */ @@ -230,6 +231,7 @@ irqreturn_t smp_ipi_demux(void) } return IRQ_HANDLED; } +#endif /* CONFIG_PPC_SMP_MUXED_IPI */ void smp_send_reschedule(int cpu) { -- cgit v1.2.3 From 714542721b4a53a3ebbdd5f0619ac0f66e7df610 Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:29:46 +0000 Subject: powerpc: Use bytes instead of bitops in smp ipi multiplexing Since there are only 4 messages, we can replace the atomic bit set (which uses atomic load reserve and store conditional sequence) with a byte stores to seperate bytes. We still have to perform a load reserve and store conditional sequence to avoid loosing messages on reception but we can do that with a single call to xchg. The do {} while and __BIG_ENDIAN specific mask testing was chosen by looking at the generated asm code. On gcc-4.4, the bit masking becomes a simple bit mask and test of the register returned from xchg without storing and loading the value to the stack like attempts with a union of bytes and an int (or worse, loading single bit constants from the constant pool into non-voliatle registers that had to be preseved on the stack). The do {} while avoids an unconditional branch to the end of the loop to test the entry / repeat condition of a while loop and instead optimises for the expected single iteration of the loop. We have a full mb() at the beginning to cover ordering between send, ipi, and receive so we can use xchg_local and forgo the further acquire and release barriers of xchg. Signed-off-by: Milton Miller Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/smp.c | 31 ++++++++++++++++++------------- 1 file changed, 18 insertions(+), 13 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index d76f7d7929be..a8909aa50642 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -180,7 +180,7 @@ int smp_request_message_ipi(int virq, int msg) #ifdef CONFIG_PPC_SMP_MUXED_IPI struct cpu_messages { - unsigned long messages; /* current messages bits */ + int messages; /* current messages */ unsigned long data; /* data for cause ipi */ }; static DEFINE_PER_CPU_SHARED_ALIGNED(struct cpu_messages, ipi_message); @@ -195,9 +195,9 @@ void smp_muxed_ipi_set_data(int cpu, unsigned long data) void smp_muxed_ipi_message_pass(int cpu, int msg) { struct cpu_messages *info = &per_cpu(ipi_message, cpu); - unsigned long *tgt = &info->messages; + char *message = (char *)&info->messages; - set_bit(msg, tgt); + message[msg] = 1; mb(); smp_ops->cause_ipi(cpu, info->data); } @@ -205,30 +205,35 @@ void smp_muxed_ipi_message_pass(int cpu, int msg) void smp_muxed_ipi_resend(void) { struct cpu_messages *info = &__get_cpu_var(ipi_message); - unsigned long *tgt = &info->messages; - if (*tgt) + if (info->messages) smp_ops->cause_ipi(smp_processor_id(), info->data); } irqreturn_t smp_ipi_demux(void) { struct cpu_messages *info = &__get_cpu_var(ipi_message); - unsigned long *tgt = &info->messages; + unsigned int all; mb(); /* order any irq clear */ - while (*tgt) { - if (test_and_clear_bit(PPC_MSG_CALL_FUNCTION, tgt)) + + do { + all = xchg_local(&info->messages, 0); + +#ifdef __BIG_ENDIAN + if (all & (1 << (24 - 8 * PPC_MSG_CALL_FUNCTION))) generic_smp_call_function_interrupt(); - if (test_and_clear_bit(PPC_MSG_RESCHEDULE, tgt)) + if (all & (1 << (24 - 8 * PPC_MSG_RESCHEDULE))) reschedule_action(0, NULL); /* upcoming sched hook */ - if (test_and_clear_bit(PPC_MSG_CALL_FUNC_SINGLE, tgt)) + if (all & (1 << (24 - 8 * PPC_MSG_CALL_FUNC_SINGLE))) generic_smp_call_function_single_interrupt(); -#if defined(CONFIG_DEBUGGER) || defined(CONFIG_KEXEC) - if (test_and_clear_bit(PPC_MSG_DEBUGGER_BREAK, tgt)) + if (all & (1 << (24 - 8 * PPC_MSG_DEBUGGER_BREAK))) debug_ipi_action(0, NULL); +#else +#error Unsupported ENDIAN #endif - } + } while (info->messages); + return IRQ_HANDLED; } #endif /* CONFIG_PPC_SMP_MUXED_IPI */ -- cgit v1.2.3 From 3af259d1555a93b3b6f6545af13e0eb99b0d5d32 Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:29:53 +0000 Subject: powerpc: Radix trees are available before init_IRQ Since the generic irq code uses a radix tree for sparse interrupts, the initcall ordering has been changed to initialize radix trees before irqs. We no longer need to defer creating revmap radix trees to the arch_initcall irq_late_init. Also, the kmem caches are allocated so we don't need to use zalloc_maybe_bootmem. Signed-off-by: Milton Miller Reviewed-by: Grant Likely Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/irq.c | 78 +++-------------------------------------------- 1 file changed, 4 insertions(+), 74 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index 826552cecebd..f42e869ee3cc 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -493,7 +493,6 @@ struct irq_map_entry { static LIST_HEAD(irq_hosts); static DEFINE_RAW_SPINLOCK(irq_big_lock); -static unsigned int revmap_trees_allocated; static DEFINE_MUTEX(revmap_trees_mutex); static struct irq_map_entry irq_map[NR_IRQS]; static unsigned int irq_virq_count = NR_IRQS; @@ -537,7 +536,7 @@ struct irq_host *irq_alloc_host(struct device_node *of_node, /* Allocate structure and revmap table if using linear mapping */ if (revmap_type == IRQ_HOST_MAP_LINEAR) size += revmap_arg * sizeof(unsigned int); - host = zalloc_maybe_bootmem(size, GFP_KERNEL); + host = kzalloc(size, GFP_KERNEL); if (host == NULL) return NULL; @@ -605,6 +604,9 @@ struct irq_host *irq_alloc_host(struct device_node *of_node, smp_wmb(); host->revmap_data.linear.revmap = rmap; break; + case IRQ_HOST_MAP_TREE: + INIT_RADIX_TREE(&host->revmap_data.tree, GFP_KERNEL); + break; default: break; } @@ -839,13 +841,6 @@ void irq_dispose_mapping(unsigned int virq) host->revmap_data.linear.revmap[hwirq] = NO_IRQ; break; case IRQ_HOST_MAP_TREE: - /* - * Check if radix tree allocated yet, if not then nothing to - * remove. - */ - smp_rmb(); - if (revmap_trees_allocated < 1) - break; mutex_lock(&revmap_trees_mutex); radix_tree_delete(&host->revmap_data.tree, hwirq); mutex_unlock(&revmap_trees_mutex); @@ -905,14 +900,6 @@ unsigned int irq_radix_revmap_lookup(struct irq_host *host, WARN_ON(host->revmap_type != IRQ_HOST_MAP_TREE); - /* - * Check if the radix tree exists and has bee initialized. - * If not, we fallback to slow mode - */ - if (revmap_trees_allocated < 2) - return irq_find_mapping(host, hwirq); - - /* Now try to resolve */ /* * No rcu_read_lock(ing) needed, the ptr returned can't go under us * as it's referencing an entry in the static irq_map table. @@ -935,18 +922,8 @@ unsigned int irq_radix_revmap_lookup(struct irq_host *host, void irq_radix_revmap_insert(struct irq_host *host, unsigned int virq, irq_hw_number_t hwirq) { - WARN_ON(host->revmap_type != IRQ_HOST_MAP_TREE); - /* - * Check if the radix tree exists yet. - * If not, then the irq will be inserted into the tree when it gets - * initialized. - */ - smp_rmb(); - if (revmap_trees_allocated < 1) - return; - if (virq != NO_IRQ) { mutex_lock(&revmap_trees_mutex); radix_tree_insert(&host->revmap_data.tree, hwirq, @@ -1054,53 +1031,6 @@ int arch_early_irq_init(void) return 0; } -/* We need to create the radix trees late */ -static int irq_late_init(void) -{ - struct irq_host *h; - unsigned int i; - - /* - * No mutual exclusion with respect to accessors of the tree is needed - * here as the synchronization is done via the state variable - * revmap_trees_allocated. - */ - list_for_each_entry(h, &irq_hosts, link) { - if (h->revmap_type == IRQ_HOST_MAP_TREE) - INIT_RADIX_TREE(&h->revmap_data.tree, GFP_KERNEL); - } - - /* - * Make sure the radix trees inits are visible before setting - * the flag - */ - smp_wmb(); - revmap_trees_allocated = 1; - - /* - * Insert the reverse mapping for those interrupts already present - * in irq_map[]. - */ - mutex_lock(&revmap_trees_mutex); - for (i = 0; i < irq_virq_count; i++) { - if (irq_map[i].host && - (irq_map[i].host->revmap_type == IRQ_HOST_MAP_TREE)) - radix_tree_insert(&irq_map[i].host->revmap_data.tree, - irq_map[i].hwirq, &irq_map[i]); - } - mutex_unlock(&revmap_trees_mutex); - - /* - * Make sure the radix trees insertions are visible before setting - * the flag - */ - smp_wmb(); - revmap_trees_allocated = 2; - - return 0; -} -arch_initcall(irq_late_init); - #ifdef CONFIG_VIRQ_DEBUG static int virq_debug_show(struct seq_file *m, void *private) { -- cgit v1.2.3 From 2d441681a4df7822e6ef6fcc0320bb14d2a06dbb Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:29:57 +0000 Subject: powerpc: Return early if irq_host lookup type is wrong If for some reason the code incrorectly calls the wrong function to manage the revmap, not only should we warn, we should take action. However, in the paths we expect to be taken every delivered interrupt change to WARN_ON_ONCE. Use the if (WARN_ON(x)) format to get the unlikely for free. Signed-off-by: Milton Miller Reviewed-by: Grant Likely Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/irq.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index f42e869ee3cc..4a5aa8ca97a5 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -814,8 +814,7 @@ void irq_dispose_mapping(unsigned int virq) return; host = irq_map[virq].host; - WARN_ON (host == NULL); - if (host == NULL) + if (WARN_ON(host == NULL)) return; /* Never unmap legacy interrupts */ @@ -898,7 +897,8 @@ unsigned int irq_radix_revmap_lookup(struct irq_host *host, struct irq_map_entry *ptr; unsigned int virq; - WARN_ON(host->revmap_type != IRQ_HOST_MAP_TREE); + if (WARN_ON_ONCE(host->revmap_type != IRQ_HOST_MAP_TREE)) + return irq_find_mapping(host, hwirq); /* * No rcu_read_lock(ing) needed, the ptr returned can't go under us @@ -922,7 +922,8 @@ unsigned int irq_radix_revmap_lookup(struct irq_host *host, void irq_radix_revmap_insert(struct irq_host *host, unsigned int virq, irq_hw_number_t hwirq) { - WARN_ON(host->revmap_type != IRQ_HOST_MAP_TREE); + if (WARN_ON(host->revmap_type != IRQ_HOST_MAP_TREE)) + return; if (virq != NO_IRQ) { mutex_lock(&revmap_trees_mutex); @@ -937,7 +938,8 @@ unsigned int irq_linear_revmap(struct irq_host *host, { unsigned int *revmap; - WARN_ON(host->revmap_type != IRQ_HOST_MAP_LINEAR); + if (WARN_ON_ONCE(host->revmap_type != IRQ_HOST_MAP_LINEAR)) + return irq_find_mapping(host, hwirq); /* Check revmap bounds */ if (unlikely(hwirq >= host->revmap_data.linear.size)) -- cgit v1.2.3 From da0519800260a3c791b4fe3317a0c7560027a372 Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:30:26 +0000 Subject: powerpc: Remove irq_host_ops->remap hook It was called from irq_create_mapping if that was called for a host and hwirq that was previously mapped, "to update the flags". But the only implementation was in beat_interrupt and all it did was repeat a hypervisor call without error checking that was performed with error checking at the beginning of the map hook. In addition, the comment on the beat remap hook says it will only called once for a given mapping, which would apply to map not remap. All flags should be known by the time the match hook is called, before we call the map hook. Removing this mostly unused hook will simpify the requirements of irq_domain concept. Signed-off-by: Milton Miller Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/irq.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index 4a5aa8ca97a5..0715a09a4101 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -732,8 +732,6 @@ unsigned int irq_create_mapping(struct irq_host *host, */ virq = irq_find_mapping(host, hwirq); if (virq != NO_IRQ) { - if (host->ops->remap) - host->ops->remap(host, virq, hwirq); pr_debug("irq: -> existing mapping on virq %d\n", virq); return virq; } -- cgit v1.2.3 From 3ee62d365b519c0c18c774049efcde84fe51c60c Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:30:36 +0000 Subject: powerpc: Add virq_is_host to reduce virq_to_host usage Some irq_host implementations are using virq_to_host to check if they are the irq_host for a virtual irq. To allow us to make space versus time tradeoffs, replace this usage with an assertive virq_is_host that confirms or denies the irq is associated with the given irq_host. Signed-off-by: Milton Miller Acked-by: Grant Likely Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/irq.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index 0715a09a4101..73cf29078fef 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -510,6 +510,12 @@ irq_hw_number_t virq_to_hw(unsigned int virq) } EXPORT_SYMBOL_GPL(virq_to_hw); +bool virq_is_host(unsigned int virq, struct irq_host *host) +{ + return irq_map[virq].host == host; +} +EXPORT_SYMBOL_GPL(virq_is_host); + struct irq_host *virq_to_host(unsigned int virq) { return irq_map[virq].host; -- cgit v1.2.3 From 1e8c23013ed0d535e531b3b9cc30200e884f3ff0 Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:30:40 +0000 Subject: powerpc: Remove virq_to_host The only references to the irq_map[].host field are internal to arch/powerpc/kernel/irq.c Signed-off-by: Milton Miller Acked-by: Grant Likely Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/irq.c | 6 ------ 1 file changed, 6 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index 73cf29078fef..4368b5ed5604 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -516,12 +516,6 @@ bool virq_is_host(unsigned int virq, struct irq_host *host) } EXPORT_SYMBOL_GPL(virq_is_host); -struct irq_host *virq_to_host(unsigned int virq) -{ - return irq_map[virq].host; -} -EXPORT_SYMBOL_GPL(virq_to_host); - static int default_irq_host_match(struct irq_host *h, struct device_node *np) { return h->of_node != NULL && h->of_node == np; -- cgit v1.2.3 From d36b4c4f3cc6caae6d4a12d9f995513e4c3acdd5 Mon Sep 17 00:00:00 2001 From: Kumar Gala Date: Wed, 6 Apr 2011 00:18:48 -0500 Subject: powerpc/fsl-booke64: Add support for Debug Level exception handler Signed-off-by: Kumar Gala --- arch/powerpc/kernel/exceptions-64e.S | 65 ++++++++++++++++++++++++++++++++++-- arch/powerpc/kernel/setup_64.c | 8 +++++ 2 files changed, 70 insertions(+), 3 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/exceptions-64e.S b/arch/powerpc/kernel/exceptions-64e.S index 4d0abb4930a1..cf27a8fa0d29 100644 --- a/arch/powerpc/kernel/exceptions-64e.S +++ b/arch/powerpc/kernel/exceptions-64e.S @@ -253,9 +253,6 @@ exception_marker: .balign 0x1000 .globl interrupt_base_book3e interrupt_base_book3e: /* fake trap */ - /* Note: If real debug exceptions are supported by the HW, the vector - * below will have to be patched up to point to an appropriate handler - */ EXCEPTION_STUB(0x000, machine_check) /* 0x0200 */ EXCEPTION_STUB(0x020, critical_input) /* 0x0580 */ EXCEPTION_STUB(0x040, debug_crit) /* 0x0d00 */ @@ -455,6 +452,68 @@ interrupt_end_book3e: kernel_dbg_exc: b . /* NYI */ +/* Debug exception as a debug interrupt*/ + START_EXCEPTION(debug_debug); + DBG_EXCEPTION_PROLOG(0xd00, PROLOG_ADDITION_2REGS) + + /* + * If there is a single step or branch-taken exception in an + * exception entry sequence, it was probably meant to apply to + * the code where the exception occurred (since exception entry + * doesn't turn off DE automatically). We simulate the effect + * of turning off DE on entry to an exception handler by turning + * off DE in the DSRR1 value and clearing the debug status. + */ + + mfspr r14,SPRN_DBSR /* check single-step/branch taken */ + andis. r15,r14,DBSR_IC@h + beq+ 1f + + LOAD_REG_IMMEDIATE(r14,interrupt_base_book3e) + LOAD_REG_IMMEDIATE(r15,interrupt_end_book3e) + cmpld cr0,r10,r14 + cmpld cr1,r10,r15 + blt+ cr0,1f + bge+ cr1,1f + + /* here it looks like we got an inappropriate debug exception. */ + lis r14,DBSR_IC@h /* clear the IC event */ + rlwinm r11,r11,0,~MSR_DE /* clear DE in the DSRR1 value */ + mtspr SPRN_DBSR,r14 + mtspr SPRN_DSRR1,r11 + lwz r10,PACA_EXDBG+EX_CR(r13) /* restore registers */ + ld r1,PACA_EXDBG+EX_R1(r13) + ld r14,PACA_EXDBG+EX_R14(r13) + ld r15,PACA_EXDBG+EX_R15(r13) + mtcr r10 + ld r10,PACA_EXDBG+EX_R10(r13) /* restore registers */ + ld r11,PACA_EXDBG+EX_R11(r13) + mfspr r13,SPRN_SPRG_DBG_SCRATCH + rfdi + + /* Normal debug exception */ + /* XXX We only handle coming from userspace for now since we can't + * quite save properly an interrupted kernel state yet + */ +1: andi. r14,r11,MSR_PR; /* check for userspace again */ + beq kernel_dbg_exc; /* if from kernel mode */ + + /* Now we mash up things to make it look like we are coming on a + * normal exception + */ + mfspr r15,SPRN_SPRG_DBG_SCRATCH + mtspr SPRN_SPRG_GEN_SCRATCH,r15 + mfspr r14,SPRN_DBSR + EXCEPTION_COMMON(0xd00, PACA_EXDBG, INTS_DISABLE_ALL) + std r14,_DSISR(r1) + addi r3,r1,STACK_FRAME_OVERHEAD + mr r4,r14 + ld r14,PACA_EXDBG+EX_R14(r13) + ld r15,PACA_EXDBG+EX_R15(r13) + bl .save_nvgprs + bl .DebugException + b .ret_from_except + /* Doorbell interrupt */ MASKABLE_EXCEPTION(0x2070, doorbell, .doorbell_exception, ACK_NONE) diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index c2ec0a12e14f..a88bf2713d41 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -62,6 +62,7 @@ #include #include #include +#include #include "setup.h" @@ -477,6 +478,9 @@ static void __init irqstack_early_init(void) #ifdef CONFIG_PPC_BOOK3E static void __init exc_lvl_early_init(void) { + extern unsigned int interrupt_base_book3e; + extern unsigned int exc_debug_debug_book3e; + unsigned int i; for_each_possible_cpu(i) { @@ -487,6 +491,10 @@ static void __init exc_lvl_early_init(void) mcheckirq_ctx[i] = (struct thread_info *) __va(memblock_alloc(THREAD_SIZE, THREAD_SIZE)); } + + if (cpu_has_feature(CPU_FTR_DEBUG_LVL_EXC)) + patch_branch(&interrupt_base_book3e + (0x040 / 4) + 1, + (unsigned long)&exc_debug_debug_book3e, 0); } #else #define exc_lvl_early_init() -- cgit v1.2.3 From 3a6e9bd7f60b29efc205485ceb11a768032c40d4 Mon Sep 17 00:00:00 2001 From: Scott Wood Date: Mon, 9 May 2011 16:26:00 -0500 Subject: powerpc/e5500: set non-base IVORs Without this, we attempt to use doorbells for IPIs, and end up branching to some bad address. Plus, even for the exceptions we don't implement, it's good to handle it and get a message out. Signed-off-by: Scott Wood Signed-off-by: Kumar Gala --- arch/powerpc/kernel/cpu_setup_fsl_booke.S | 3 ++ arch/powerpc/kernel/exceptions-64e.S | 47 +++++++++++++++++++++++++++++++ 2 files changed, 50 insertions(+) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/cpu_setup_fsl_booke.S b/arch/powerpc/kernel/cpu_setup_fsl_booke.S index 913611105c1f..8053db02b85e 100644 --- a/arch/powerpc/kernel/cpu_setup_fsl_booke.S +++ b/arch/powerpc/kernel/cpu_setup_fsl_booke.S @@ -88,6 +88,9 @@ _GLOBAL(__setup_cpu_e5500) bl __e500_dcache_setup #ifdef CONFIG_PPC_BOOK3E_64 bl .__setup_base_ivors + bl .setup_perfmon_ivor + bl .setup_doorbell_ivors + bl .setup_ehv_ivors #else bl __setup_e500mc_ivors #endif diff --git a/arch/powerpc/kernel/exceptions-64e.S b/arch/powerpc/kernel/exceptions-64e.S index cf27a8fa0d29..d24d4400cc79 100644 --- a/arch/powerpc/kernel/exceptions-64e.S +++ b/arch/powerpc/kernel/exceptions-64e.S @@ -269,8 +269,13 @@ interrupt_base_book3e: /* fake trap */ EXCEPTION_STUB(0x1a0, watchdog) /* 0x09f0 */ EXCEPTION_STUB(0x1c0, data_tlb_miss) EXCEPTION_STUB(0x1e0, instruction_tlb_miss) + EXCEPTION_STUB(0x260, perfmon) EXCEPTION_STUB(0x280, doorbell) EXCEPTION_STUB(0x2a0, doorbell_crit) + EXCEPTION_STUB(0x2c0, guest_doorbell) + EXCEPTION_STUB(0x2e0, guest_doorbell_crit) + EXCEPTION_STUB(0x300, hypercall) + EXCEPTION_STUB(0x320, ehpriv) .globl interrupt_end_book3e interrupt_end_book3e: @@ -514,6 +519,8 @@ kernel_dbg_exc: bl .DebugException b .ret_from_except + MASKABLE_EXCEPTION(0x260, perfmon, .performance_monitor_exception, ACK_NONE) + /* Doorbell interrupt */ MASKABLE_EXCEPTION(0x2070, doorbell, .doorbell_exception, ACK_NONE) @@ -528,6 +535,11 @@ kernel_dbg_exc: // b ret_from_crit_except b . + MASKABLE_EXCEPTION(0x2c0, guest_doorbell, .unknown_exception, ACK_NONE) + MASKABLE_EXCEPTION(0x2e0, guest_doorbell_crit, .unknown_exception, ACK_NONE) + MASKABLE_EXCEPTION(0x310, hypercall, .unknown_exception, ACK_NONE) + MASKABLE_EXCEPTION(0x320, ehpriv, .unknown_exception, ACK_NONE) + /* * An interrupt came in while soft-disabled; clear EE in SRR1, @@ -647,7 +659,12 @@ fast_exception_return: BAD_STACK_TRAMPOLINE(0x000) BAD_STACK_TRAMPOLINE(0x100) BAD_STACK_TRAMPOLINE(0x200) +BAD_STACK_TRAMPOLINE(0x260) +BAD_STACK_TRAMPOLINE(0x2c0) +BAD_STACK_TRAMPOLINE(0x2e0) BAD_STACK_TRAMPOLINE(0x300) +BAD_STACK_TRAMPOLINE(0x310) +BAD_STACK_TRAMPOLINE(0x320) BAD_STACK_TRAMPOLINE(0x400) BAD_STACK_TRAMPOLINE(0x500) BAD_STACK_TRAMPOLINE(0x600) @@ -1183,3 +1200,33 @@ _GLOBAL(__setup_base_ivors) sync blr + +_GLOBAL(setup_perfmon_ivor) + SET_IVOR(35, 0x260) /* Performance Monitor */ + blr + +_GLOBAL(setup_doorbell_ivors) + SET_IVOR(36, 0x280) /* Processor Doorbell */ + SET_IVOR(37, 0x2a0) /* Processor Doorbell Crit */ + + /* Check MMUCFG[LPIDSIZE] to determine if we have category E.HV */ + mfspr r10,SPRN_MMUCFG + rlwinm. r10,r10,0,MMUCFG_LPIDSIZE + beqlr + + SET_IVOR(38, 0x2c0) /* Guest Processor Doorbell */ + SET_IVOR(39, 0x2e0) /* Guest Processor Doorbell Crit/MC */ + blr + +_GLOBAL(setup_ehv_ivors) + /* + * We may be running as a guest and lack E.HV even on a chip + * that normally has it. + */ + mfspr r10,SPRN_MMUCFG + rlwinm. r10,r10,0,MMUCFG_LPIDSIZE + beqlr + + SET_IVOR(40, 0x300) /* Embedded Hypervisor System Call */ + SET_IVOR(41, 0x320) /* Embedded Hypervisor Privilege */ + blr -- cgit v1.2.3 From 41fb5e62604c5ddd00a784ffb7672dd8df5d76f2 Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Tue, 10 May 2011 19:30:44 +0000 Subject: powerpc: Make IRQ_NOREQUEST last to clear, first to set When creating an irq, don't allow a concurent driver request until we have caled map, which will likley call set_chip_and_handler to change the irq_chip and its operations. Similarly, when tearing down an IRQ, make sure no new uses come along while we change the irq back to the nop chip and then reset the descriptor to freed status. Signed-off-by: Milton Miller Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/irq.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index 4368b5ed5604..a24d37d4cf51 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -586,14 +586,14 @@ struct irq_host *irq_alloc_host(struct device_node *of_node, irq_map[i].host = host; smp_wmb(); - /* Clear norequest flags */ - irq_clear_status_flags(i, IRQ_NOREQUEST); - /* Legacy flags are left to default at this point, * one can then use irq_create_mapping() to * explicitly change them */ ops->map(host, i, i); + + /* Clear norequest flags */ + irq_clear_status_flags(i, IRQ_NOREQUEST); } break; case IRQ_HOST_MAP_LINEAR: @@ -664,8 +664,6 @@ static int irq_setup_virq(struct irq_host *host, unsigned int virq, goto error; } - irq_clear_status_flags(virq, IRQ_NOREQUEST); - /* map it */ smp_wmb(); irq_map[virq].hwirq = hwirq; @@ -676,6 +674,8 @@ static int irq_setup_virq(struct irq_host *host, unsigned int virq, goto errdesc; } + irq_clear_status_flags(virq, IRQ_NOREQUEST); + return 0; errdesc: @@ -819,6 +819,8 @@ void irq_dispose_mapping(unsigned int virq) if (host->revmap_type == IRQ_HOST_MAP_LEGACY) return; + irq_set_status_flags(virq, IRQ_NOREQUEST); + /* remove chip and handler */ irq_set_chip_and_handler(virq, NULL, NULL); @@ -848,8 +850,6 @@ void irq_dispose_mapping(unsigned int virq) smp_mb(); irq_map[virq].hwirq = host->inval_irq; - irq_set_status_flags(virq, IRQ_NOREQUEST); - irq_free_descs(virq, 1); /* Free it */ irq_free_virt(virq, 1); -- cgit v1.2.3 From 03bf469add176afd8a1a4c493d9f4e0e520db12b Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 11 May 2011 20:58:18 +0000 Subject: powerpc: Make early memory scan more resilient to out of order nodes We keep track of the size of the lowest block of memory and call setup_initial_memory_limit() only after we've parsed them all Signed-off-by: Benjamin Herrenschmidt Acked-by: Milton Miller --- arch/powerpc/kernel/prom.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c index 5f5e6aed2b70..5311a26dcf46 100644 --- a/arch/powerpc/kernel/prom.c +++ b/arch/powerpc/kernel/prom.c @@ -68,6 +68,7 @@ int __initdata iommu_force_on; unsigned long tce_alloc_start, tce_alloc_end; u64 ppc64_rma_size; #endif +static phys_addr_t first_memblock_size; static int __init early_parse_mem(char *p) { @@ -505,11 +506,14 @@ void __init early_init_dt_add_memory_arch(u64 base, u64 size) size = 0x80000000ul - base; } #endif - - /* First MEMBLOCK added, do some special initializations */ - if (memstart_addr == ~(phys_addr_t)0) - setup_initial_memory_limit(base, size); - memstart_addr = min((u64)memstart_addr, base); + /* Keep track of the beginning of memory -and- the size of + * the very first block in the device-tree as it represents + * the RMA on ppc64 server + */ + if (base < memstart_addr) { + memstart_addr = base; + first_memblock_size = size; + } /* Add the chunk to the MEMBLOCK list */ memblock_add(base, size); @@ -694,6 +698,7 @@ void __init early_init_devtree(void *params) of_scan_flat_dt(early_init_dt_scan_root, NULL); of_scan_flat_dt(early_init_dt_scan_memory_ppc, NULL); + setup_initial_memory_limit(memstart_addr, first_memblock_size); /* Save command line for /proc/cmdline and then parse parameters */ strlcpy(boot_command_line, cmd_line, COMMAND_LINE_SIZE); -- cgit v1.2.3 From 2c78027a62ea38585da1ff944afdc6146335cb7c Mon Sep 17 00:00:00 2001 From: Gabriel Paubert Date: Fri, 13 May 2011 01:03:13 +0000 Subject: powerpc: Fix for Pegasos keyboard and mouse [See http://lists.ozlabs.org/pipermail/linuxppc-dev/2010-October/086424.html and followups. Part of the commit message is directly copied from that.] Commit 540c6c392f01887dcc96bef0a41e63e6c1334f01 tries to find i8042 IRQs in the device-tree but doesn't fall back to the old hardcoded 1 and 12 in all failure cases. Specifically, the case where the device-tree contains nothing matching pnpPNP,303 or pnpPNP,f03 doesn't seem to be handled well. It sort of falls through to the old code, but leaves the IRQs set to 0. Signed-off-by: Gabriel Paubert Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/setup-common.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c index ef33a084fcf4..79fca2651b65 100644 --- a/arch/powerpc/kernel/setup-common.c +++ b/arch/powerpc/kernel/setup-common.c @@ -602,6 +602,10 @@ int check_legacy_ioport(unsigned long base_port) * name instead */ if (!np) np = of_find_node_by_name(NULL, "8042"); + if (np) { + of_i8042_kbd_irq = 1; + of_i8042_aux_irq = 12; + } break; case FDC_BASE: /* FDC1 */ np = of_find_node_by_type(NULL, "fdc"); -- cgit v1.2.3 From f38aa708776aefd9e3ba7ec1211c07efe9fa3227 Mon Sep 17 00:00:00 2001 From: Sebastian Siewior Date: Mon, 16 May 2011 08:58:13 +0000 Subject: powerpc: Remove last piece of GEMINI It seems that Adrian is getting old. He removed almost everything of GEMINI in commit c53653130 ("[POWERPC] Remove the broken Gemini support") except this piece. Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/head_32.S | 13 ------------- 1 file changed, 13 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/head_32.S b/arch/powerpc/kernel/head_32.S index 98c4b29a56f4..ba250d505e07 100644 --- a/arch/powerpc/kernel/head_32.S +++ b/arch/powerpc/kernel/head_32.S @@ -805,19 +805,6 @@ _ENTRY(copy_and_flush) blr #ifdef CONFIG_SMP -#ifdef CONFIG_GEMINI - .globl __secondary_start_gemini -__secondary_start_gemini: - mfspr r4,SPRN_HID0 - ori r4,r4,HID0_ICFI - li r3,0 - ori r3,r3,HID0_ICE - andc r4,r4,r3 - mtspr SPRN_HID0,r4 - sync - b __secondary_start -#endif /* CONFIG_GEMINI */ - .globl __secondary_start_mpc86xx __secondary_start_mpc86xx: mfspr r3, SPRN_PIR -- cgit v1.2.3 From 208b3a4c196e733b9cec006dc132cfc149b2810a Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Fri, 20 May 2011 17:50:18 +1000 Subject: powerpc: Fix hard CPU IDs detection commit 9d07bc841c9779b4d7902e417f4e509996ce805d "powerpc: Properly handshake CPUs going out of boot spin loop" Would cause a miscalculation of the hard CPU ID. It removes breaking out of the loop when finding a match with a processor, thus the "i" used as an index in the intserv array is always incorrect This broke interrupt on my PowerMac laptop. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/prom.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c index 5311a26dcf46..48aeb55faae9 100644 --- a/arch/powerpc/kernel/prom.c +++ b/arch/powerpc/kernel/prom.c @@ -278,6 +278,7 @@ static int __init early_init_dt_scan_cpus(unsigned long node, int i, nthreads; unsigned long len; int found = -1; + int found_thread = 0; /* We are scanning "cpu" nodes only */ if (type == NULL || strcmp(type, "cpu") != 0) @@ -301,9 +302,11 @@ static int __init early_init_dt_scan_cpus(unsigned long node, * version 2 of the kexec param format adds the phys cpuid of * booted proc. */ - if (initial_boot_params && initial_boot_params->version >= 2) { - if (intserv[i] == initial_boot_params->boot_cpuid_phys) + if (initial_boot_params->version >= 2) { + if (intserv[i] == initial_boot_params->boot_cpuid_phys) { found = boot_cpu_count; + found_thread = i; + } } else { /* * Check if it's the boot-cpu, set it's hw index now, @@ -322,9 +325,9 @@ static int __init early_init_dt_scan_cpus(unsigned long node, if (found >= 0) { DBG("boot cpu: logical %d physical %d\n", found, - intserv[i]); + intserv[found_thread]); boot_cpuid = found; - set_hard_smp_processor_id(found, intserv[i]); + set_hard_smp_processor_id(found, intserv[found_thread]); /* * PAPR defines "logical" PVR values for cpus that -- cgit v1.2.3