From 63fe08177f92ce21929df8af6361491b98ad12cd Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 14 Apr 2009 12:01:53 +1000 Subject: md: tiny md.h cleanups - update inclusion guard and make sure it covers the whole file - remove superflous #ifdef CONFIG_BLOCK - make sure all required headers are included so that new users aren't required to include others before Signed-off-by: Christoph Hellwig Signed-off-by: NeilBrown --- drivers/md/md.h | 21 +++++++++++++-------- 1 file changed, 13 insertions(+), 8 deletions(-) diff --git a/drivers/md/md.h b/drivers/md/md.h index e9b7f54c24d6..8227ab909d44 100644 --- a/drivers/md/md.h +++ b/drivers/md/md.h @@ -12,10 +12,17 @@ Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ -#ifndef _MD_K_H -#define _MD_K_H - -#ifdef CONFIG_BLOCK +#ifndef _MD_MD_H +#define _MD_MD_H + +#include +#include +#include +#include +#include +#include +#include +#include #define MaxSector (~(sector_t)0) @@ -408,10 +415,6 @@ static inline void safe_put_page(struct page *p) if (p) put_page(p); } -#endif /* CONFIG_BLOCK */ -#endif - - extern int register_md_personality(struct mdk_personality *p); extern int unregister_md_personality(struct mdk_personality *p); extern mdk_thread_t * md_register_thread(void (*run) (mddev_t *mddev), @@ -434,3 +437,5 @@ extern void md_new_event(mddev_t *mddev); extern int md_allow_write(mddev_t *mddev); extern void md_wait_for_blocked_rdev(mdk_rdev_t *rdev, mddev_t *mddev); extern void md_set_array_sectors(mddev_t *mddev, sector_t array_sectors); + +#endif /* _MD_MD_H */ -- cgit v1.2.3 From 6d56e278444bc8323b1bcfcada126b8e4dbba0f4 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 14 Apr 2009 12:01:57 +1000 Subject: md: allow setting newly added device to 'in_sync' via sysfs. When adding devices to an active array via sysfs, there is currently no way to mark a device as 'in-sync' which is useful when incrementally assembling an array. So add that option. Signed-off-by: NeilBrown --- drivers/md/md.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index ed5727c089a9..298731b8e951 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -2086,6 +2086,7 @@ state_store(mdk_rdev_t *rdev, const char *buf, size_t len) * -writemostly - clears write_mostly * blocked - sets the Blocked flag * -blocked - clears the Blocked flag + * insync - sets Insync providing device isn't active */ int err = -EINVAL; if (cmd_match(buf, "faulty") && rdev->mddev->pers) { @@ -2117,6 +2118,9 @@ state_store(mdk_rdev_t *rdev, const char *buf, size_t len) set_bit(MD_RECOVERY_NEEDED, &rdev->mddev->recovery); md_wakeup_thread(rdev->mddev->thread); + err = 0; + } else if (cmd_match(buf, "insync") && rdev->raid_disk == -1) { + set_bit(In_sync, &rdev->flags); err = 0; } if (!err && rdev->sysfs_state) @@ -2190,7 +2194,7 @@ slot_store(mdk_rdev_t *rdev, const char *buf, size_t len) } else if (rdev->mddev->pers) { mdk_rdev_t *rdev2; /* Activating a spare .. or possibly reactivating - * if we every get bitmaps working here. + * if we ever get bitmaps working here. */ if (rdev->raid_disk != -1) -- cgit v1.2.3 From acb180b0e335ad88acfed6c8a33e39c05b95dc49 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 14 Apr 2009 16:28:34 +1000 Subject: md: improve usefulness and accuracy of sysfs file md/sync_completed. The sync_completed file reports how much of a resync (or recovery or reshape) has been completed. However due to the possibility of out-of-order completion of writes, it is not certain to be accurate. We have an internal value - mddev->curr_resync_completed - which is an accurate value (though it might not always be quite so uptodate). So: - make curr_resync_completed be uptodate a little more often, particularly when raid5 reshape updates status in the metadata - report curr_resync_completed in the sysfs file - allow poll/select to report all updates to md/sync_completed. This makes sync_completed completed usable by any external metadata handler that wants to record this status information in its metadata. Signed-off-by: NeilBrown --- drivers/md/bitmap.c | 1 + drivers/md/md.c | 34 ++++++++++++++++++++++------------ drivers/md/raid5.c | 4 ++++ 3 files changed, 27 insertions(+), 12 deletions(-) diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index f8a9f7ab2cb8..e4510c976b34 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -1479,6 +1479,7 @@ void bitmap_cond_end_sync(struct bitmap *bitmap, sector_t sector) s += blocks; } bitmap->last_end_sync = jiffies; + sysfs_notify(&bitmap->mddev->kobj, NULL, "sync_completed"); } static void bitmap_set_memory_bits(struct bitmap *bitmap, sector_t offset, int needed) diff --git a/drivers/md/md.c b/drivers/md/md.c index 298731b8e951..7af64f3846a6 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -2017,6 +2017,8 @@ repeat: clear_bit(MD_CHANGE_PENDING, &mddev->flags); spin_unlock_irq(&mddev->write_lock); wake_up(&mddev->sb_wait); + if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery)) + sysfs_notify(&mddev->kobj, NULL, "sync_completed"); } @@ -3486,12 +3488,15 @@ sync_completed_show(mddev_t *mddev, char *page) { unsigned long max_sectors, resync; + if (!test_bit(MD_RECOVERY_RUNNING, &mddev->recovery)) + return sprintf(page, "none\n"); + if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) max_sectors = mddev->resync_max_sectors; else max_sectors = mddev->dev_sectors; - resync = (mddev->curr_resync - atomic_read(&mddev->recovery_active)); + resync = mddev->curr_resync_completed; return sprintf(page, "%lu / %lu\n", resync, max_sectors); } @@ -6338,18 +6343,12 @@ void md_do_sync(mddev_t *mddev) sector_t sectors; skipped = 0; - if (j >= mddev->resync_max) { - sysfs_notify(&mddev->kobj, NULL, "sync_completed"); - wait_event(mddev->recovery_wait, - mddev->resync_max > j - || kthread_should_stop()); - } - if (kthread_should_stop()) - goto interrupted; - if (mddev->curr_resync > mddev->curr_resync_completed && - (mddev->curr_resync - mddev->curr_resync_completed) - > (max_sectors >> 4)) { + if ((mddev->curr_resync > mddev->curr_resync_completed && + (mddev->curr_resync - mddev->curr_resync_completed) + > (max_sectors >> 4)) || + j >= mddev->resync_max + ) { /* time to update curr_resync_completed */ blk_unplug(mddev->queue); wait_event(mddev->recovery_wait, @@ -6357,7 +6356,17 @@ void md_do_sync(mddev_t *mddev) mddev->curr_resync_completed = mddev->curr_resync; set_bit(MD_CHANGE_CLEAN, &mddev->flags); + sysfs_notify(&mddev->kobj, NULL, "sync_completed"); } + + if (j >= mddev->resync_max) + wait_event(mddev->recovery_wait, + mddev->resync_max > j + || kthread_should_stop()); + + if (kthread_should_stop()) + goto interrupted; + sectors = mddev->pers->sync_request(mddev, j, &skipped, currspeed < speed_min(mddev)); if (sectors == 0) { @@ -6465,6 +6474,7 @@ void md_do_sync(mddev_t *mddev) skip: mddev->curr_resync = 0; + mddev->curr_resync_completed = 0; mddev->resync_min = 0; mddev->resync_max = MaxSector; sysfs_notify(&mddev->kobj, NULL, "sync_completed"); diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 3bbc6d647044..76892ac72544 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -3845,6 +3845,7 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped wait_event(conf->wait_for_overlap, atomic_read(&conf->reshape_stripes)==0); mddev->reshape_position = conf->reshape_progress; + mddev->curr_resync_completed = mddev->curr_resync; conf->reshape_checkpoint = jiffies; set_bit(MD_CHANGE_DEVS, &mddev->flags); md_wakeup_thread(mddev->thread); @@ -3854,6 +3855,7 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped conf->reshape_safe = mddev->reshape_position; spin_unlock_irq(&conf->device_lock); wake_up(&conf->wait_for_overlap); + sysfs_notify(&mddev->kobj, NULL, "sync_completed"); } if (mddev->delta_disks < 0) { @@ -3943,6 +3945,7 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped wait_event(conf->wait_for_overlap, atomic_read(&conf->reshape_stripes) == 0); mddev->reshape_position = conf->reshape_progress; + mddev->curr_resync_completed = mddev->curr_resync; conf->reshape_checkpoint = jiffies; set_bit(MD_CHANGE_DEVS, &mddev->flags); md_wakeup_thread(mddev->thread); @@ -3953,6 +3956,7 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped conf->reshape_safe = mddev->reshape_position; spin_unlock_irq(&conf->device_lock); wake_up(&conf->wait_for_overlap); + sysfs_notify(&mddev->kobj, NULL, "sync_completed"); } return reshape_sectors; } -- cgit v1.2.3 From 063adc7502890a0b115da4af4e15cc2622b348de Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Thu, 16 Apr 2009 14:12:22 +0900 Subject: rtc: rtc-sh: clock framework support. This adds clock framework support to the rtc-sh driver. With this in place, platforms can default to leaving the clock disabled rather than placing it in the always enabled state. Signed-off-by: Paul Mundt --- drivers/rtc/Kconfig | 2 +- drivers/rtc/rtc-sh.c | 67 ++++++++++++++++++++++++++++++++++++++-------------- 2 files changed, 50 insertions(+), 19 deletions(-) diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig index ffe34a12f446..4e9851fc1746 100644 --- a/drivers/rtc/Kconfig +++ b/drivers/rtc/Kconfig @@ -573,7 +573,7 @@ config RTC_DRV_SA1100 config RTC_DRV_SH tristate "SuperH On-Chip RTC" - depends on RTC_CLASS && SUPERH + depends on RTC_CLASS && SUPERH && HAVE_CLK help Say Y here to enable support for the on-chip RTC found in most SuperH processors. diff --git a/drivers/rtc/rtc-sh.c b/drivers/rtc/rtc-sh.c index 9b1ff12bf947..d7310adb7152 100644 --- a/drivers/rtc/rtc-sh.c +++ b/drivers/rtc/rtc-sh.c @@ -1,7 +1,7 @@ /* * SuperH On-Chip RTC Support * - * Copyright (C) 2006, 2007, 2008 Paul Mundt + * Copyright (C) 2006 - 2009 Paul Mundt * Copyright (C) 2006 Jamie Lenehan * Copyright (C) 2008 Angelo Castello * @@ -25,10 +25,11 @@ #include #include #include +#include #include #define DRV_NAME "sh-rtc" -#define DRV_VERSION "0.2.1" +#define DRV_VERSION "0.2.2" #define RTC_REG(r) ((r) * rtc_reg_size) @@ -87,16 +88,17 @@ #define RCR2_START 0x01 /* Start bit */ struct sh_rtc { - void __iomem *regbase; - unsigned long regsize; - struct resource *res; - int alarm_irq; - int periodic_irq; - int carry_irq; - struct rtc_device *rtc_dev; - spinlock_t lock; - unsigned long capabilities; /* See asm-sh/rtc.h for cap bits */ - unsigned short periodic_freq; + void __iomem *regbase; + unsigned long regsize; + struct resource *res; + int alarm_irq; + int periodic_irq; + int carry_irq; + struct clk *clk; + struct rtc_device *rtc_dev; + spinlock_t lock; + unsigned long capabilities; /* See asm/rtc.h for cap bits */ + unsigned short periodic_freq; }; static int __sh_rtc_interrupt(struct sh_rtc *rtc) @@ -294,10 +296,10 @@ static inline void sh_rtc_setaie(struct device *dev, unsigned int enable) tmp = readb(rtc->regbase + RCR1); - if (!enable) - tmp &= ~RCR1_AIE; - else + if (enable) tmp |= RCR1_AIE; + else + tmp &= ~RCR1_AIE; writeb(tmp, rtc->regbase + RCR1); @@ -618,6 +620,7 @@ static int sh_rtc_irq_set_freq(struct device *dev, int freq) { if (!is_power_of_2(freq)) return -EINVAL; + return sh_rtc_ioctl(dev, RTC_IRQP_SET, freq); } @@ -637,7 +640,8 @@ static int __devinit sh_rtc_probe(struct platform_device *pdev) struct sh_rtc *rtc; struct resource *res; struct rtc_time r; - int ret; + char clk_name[6]; + int clk_id, ret; rtc = kzalloc(sizeof(struct sh_rtc), GFP_KERNEL); if (unlikely(!rtc)) @@ -652,6 +656,7 @@ static int __devinit sh_rtc_probe(struct platform_device *pdev) dev_err(&pdev->dev, "No IRQ resource\n"); goto err_badres; } + rtc->periodic_irq = ret; rtc->carry_irq = platform_get_irq(pdev, 1); rtc->alarm_irq = platform_get_irq(pdev, 2); @@ -663,7 +668,7 @@ static int __devinit sh_rtc_probe(struct platform_device *pdev) goto err_badres; } - rtc->regsize = res->end - res->start + 1; + rtc->regsize = resource_size(res); rtc->res = request_mem_region(res->start, rtc->regsize, pdev->name); if (unlikely(!rtc->res)) { @@ -677,6 +682,26 @@ static int __devinit sh_rtc_probe(struct platform_device *pdev) goto err_badmap; } + clk_id = pdev->id; + /* With a single device, the clock id is still "rtc0" */ + if (clk_id < 0) + clk_id = 0; + + snprintf(clk_name, sizeof(clk_name), "rtc%d", clk_id); + + rtc->clk = clk_get(&pdev->dev, clk_name); + if (IS_ERR(rtc->clk)) { + /* + * No error handling for rtc->clk intentionally, not all + * platforms will have a unique clock for the RTC, and + * the clk API can handle the struct clk pointer being + * NULL. + */ + rtc->clk = NULL; + } + + clk_enable(rtc->clk); + rtc->rtc_dev = rtc_device_register("sh", &pdev->dev, &sh_rtc_ops, THIS_MODULE); if (IS_ERR(rtc->rtc_dev)) { @@ -759,6 +784,8 @@ static int __devinit sh_rtc_probe(struct platform_device *pdev) return 0; err_unmap: + clk_disable(rtc->clk); + clk_put(rtc->clk); iounmap(rtc->regbase); err_badmap: release_resource(rtc->res); @@ -780,6 +807,7 @@ static int __devexit sh_rtc_remove(struct platform_device *pdev) sh_rtc_setcie(&pdev->dev, 0); free_irq(rtc->periodic_irq, rtc); + if (rtc->carry_irq > 0) { free_irq(rtc->carry_irq, rtc); free_irq(rtc->alarm_irq, rtc); @@ -789,6 +817,9 @@ static int __devexit sh_rtc_remove(struct platform_device *pdev) iounmap(rtc->regbase); + clk_disable(rtc->clk); + clk_put(rtc->clk); + platform_set_drvdata(pdev, NULL); kfree(rtc); @@ -802,11 +833,11 @@ static void sh_rtc_set_irq_wake(struct device *dev, int enabled) struct sh_rtc *rtc = platform_get_drvdata(pdev); set_irq_wake(rtc->periodic_irq, enabled); + if (rtc->carry_irq > 0) { set_irq_wake(rtc->carry_irq, enabled); set_irq_wake(rtc->alarm_irq, enabled); } - } static int sh_rtc_suspend(struct device *dev) -- cgit v1.2.3 From 6bf853aeb1d840222aebe532ad5afc861f4d0fbb Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Thu, 16 Apr 2009 14:15:38 +0900 Subject: sh: sh7722: Don't default enable the RTC clock. rtc-sh takes care of this now, so no need to have this always enabled. Signed-off-by: Paul Mundt --- arch/sh/kernel/cpu/sh4a/setup-sh7722.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7722.c b/arch/sh/kernel/cpu/sh4a/setup-sh7722.c index 0e5d204bc792..406747f07dc0 100644 --- a/arch/sh/kernel/cpu/sh4a/setup-sh7722.c +++ b/arch/sh/kernel/cpu/sh4a/setup-sh7722.c @@ -256,7 +256,6 @@ static int __init sh7722_devices_setup(void) { clk_always_enable("uram0"); /* URAM */ clk_always_enable("xymem0"); /* XYMEM */ - clk_always_enable("rtc0"); /* RTC */ clk_always_enable("veu0"); /* VEU */ clk_always_enable("vpu0"); /* VPU */ clk_always_enable("jpu0"); /* JPU */ -- cgit v1.2.3 From b8c193f88ebd8705b3e916532539031cd9fc0b4c Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Thu, 16 Apr 2009 14:16:33 +0900 Subject: sh: sh7723: Don't default enable the RTC clock. rtc-sh takes care of this now, so no need to have this always enabled. Signed-off-by: Paul Mundt --- arch/sh/kernel/cpu/sh4a/setup-sh7723.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7723.c b/arch/sh/kernel/cpu/sh4a/setup-sh7723.c index 5338dacbcfba..a800466b938c 100644 --- a/arch/sh/kernel/cpu/sh4a/setup-sh7723.c +++ b/arch/sh/kernel/cpu/sh4a/setup-sh7723.c @@ -267,7 +267,6 @@ static struct platform_device *sh7723_devices[] __initdata = { static int __init sh7723_devices_setup(void) { clk_always_enable("meram0"); /* MERAM */ - clk_always_enable("rtc0"); /* RTC */ clk_always_enable("veu1"); /* VEU2H1 */ clk_always_enable("veu0"); /* VEU2H0 */ clk_always_enable("vpu0"); /* VPU */ -- cgit v1.2.3 From c4cb768f027706b3a0190309416b13f07114fe56 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Tue, 14 Apr 2009 18:08:37 -0300 Subject: [IA64] export smp_send_reschedule KVM will use smp_send_reschedule to force a cpu out of guest mode. Signed-off-by: Marcelo Tosatti Signed-off-by: Tony Luck --- arch/ia64/kernel/smp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/ia64/kernel/smp.c b/arch/ia64/kernel/smp.c index 2ea4199d9c57..c2d982385dce 100644 --- a/arch/ia64/kernel/smp.c +++ b/arch/ia64/kernel/smp.c @@ -225,6 +225,7 @@ smp_send_reschedule (int cpu) { platform_send_ipi(cpu, IA64_IPI_RESCHEDULE, IA64_IPI_DM_INT, 0); } +EXPORT_SYMBOL_GPL(smp_send_reschedule); /* * Called with preemption disabled. -- cgit v1.2.3 From edb91dc01a216e84b78721b71a06db1e0db141b7 Mon Sep 17 00:00:00 2001 From: Dimitri Sivanich Date: Wed, 15 Apr 2009 10:56:25 -0500 Subject: [IA64] smp_flush_tlb_mm() should only send IPI's to cpus in cpu_vm_mask Having flush_tlb_mm->smp_flush_tlb_mm() send an IPI to every cpu on the system is occasionally triggering spin_lock contention in generic_smp_call_function_interrupt(). Follow x86 arch's lead and only sends IPIs to the cpus in mm->cpu_vm_mask. Experiments with this change have shown significant improvement in this contention issue. Signed-off-by: Dimitri Sivanich Signed-off-by: Tony Luck --- arch/ia64/kernel/smp.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) diff --git a/arch/ia64/kernel/smp.c b/arch/ia64/kernel/smp.c index c2d982385dce..5230eaafd83f 100644 --- a/arch/ia64/kernel/smp.c +++ b/arch/ia64/kernel/smp.c @@ -301,15 +301,12 @@ smp_flush_tlb_mm (struct mm_struct *mm) return; } + smp_call_function_mask(mm->cpu_vm_mask, + (void (*)(void *))local_finish_flush_tlb_mm, mm, 1); + local_irq_disable(); + local_finish_flush_tlb_mm(mm); + local_irq_enable(); preempt_enable(); - /* - * We could optimize this further by using mm->cpu_vm_mask to track which CPUs - * have been running in the address space. It's not clear that this is worth the - * trouble though: to avoid races, we have to raise the IPI on the target CPU - * anyhow, and once a CPU is interrupted, the cost of local_flush_tlb_all() is - * rather trivial. - */ - on_each_cpu((void (*)(void *))local_finish_flush_tlb_mm, mm, 1); } void arch_send_call_function_single_ipi(int cpu) -- cgit v1.2.3 From c03f6a19699fce4389888a45db8245969311d868 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 17 Apr 2009 11:06:30 +1000 Subject: md: update sync_completed and reshape_position even more often. There are circumstances when a user-space process might need to "oversee" a resync/reshape process. For example when doing an in-place reshape of a raid5, it is prudent to take a backup of each section before reshaping it as this is the only way to provide safety against an unplanned shutdown (i.e. crash/power failure). The sync_max sysfs value can be used to stop the resync from advancing beyond a particular point. So user-space can: suspend IO to the first section and back it up set 'sync_max' to the end of the section wait for 'sync_completed' to reach that point resume IO on the first section and move on to the next section. However this process requires the kernel and user-space to run in lock-step which could introduce unnecessary delays. It would be better if a 'double buffered' approach could be used with userspace and kernel space working on different sections with the 'next' section always ready when the 'current' section is finished. One problem with implementing this is that sync_completed is only guaranteed to be updated when the sync process reaches sync_max. (it is updated on a time basis at other times, but it is hard to rely on that). This defeats some of the double buffering. With this patch, sync_completed (and reshape_position) get updated as the current position approaches sync_max, so there is room for userspace to advance sync_max early without losing updates. To be precise, sync_completed is updated when the current sync position reaches half way between the current value of sync_completed and the value of sync_max. This will usually be a good time for user space to update sync_max. If sync_max does not get updated, the updates to sync_completed (together with associated metadata updates) will occur at an exponentially increasing frequency which will get unreasonably fast (one update every page) immediately before the process hits sync_max and stops. So the update rate will be unreasonably fast only for an insignificant period of time. Signed-off-by: NeilBrown --- drivers/md/md.c | 3 ++- drivers/md/raid5.c | 3 ++- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index 7af64f3846a6..612343fdde94 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -6347,7 +6347,8 @@ void md_do_sync(mddev_t *mddev) if ((mddev->curr_resync > mddev->curr_resync_completed && (mddev->curr_resync - mddev->curr_resync_completed) > (max_sectors >> 4)) || - j >= mddev->resync_max + (j - mddev->curr_resync_completed)*2 + >= mddev->resync_max - mddev->curr_resync_completed ) { /* time to update curr_resync_completed */ blk_unplug(mddev->queue); diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 76892ac72544..4616bc3a6e71 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -3940,7 +3940,8 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped * then we need to write out the superblock. */ sector_nr += reshape_sectors; - if (sector_nr >= mddev->resync_max) { + if ((sector_nr - mddev->curr_resync_completed) * 2 + >= mddev->resync_max - mddev->curr_resync_completed) { /* Cannot proceed until we've updated the superblock... */ wait_event(conf->wait_for_overlap, atomic_read(&conf->reshape_stripes) == 0); -- cgit v1.2.3 From 6a7c7eaf71b636f197d73b381a2ab729ebdcfb2e Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 19 Apr 2009 20:08:42 +0200 Subject: PM/Suspend: Introduce two new platform callbacks to avoid breakage Commit 900af0d973856d6feb6fc088c2d0d3fde57707d3 (PM: Change suspend code ordering) changed the ordering of suspend code in such a way that the platform .prepare() callback is now executed after the device drivers' late suspend callbacks have run. Unfortunately, this turns out to break ARM platforms that need to talk via I2C to power control devices during the .prepare() callback. For this reason introduce two new platform suspend callbacks, .prepare_late() and .wake(), that will be called just prior to disabling non-boot CPUs and right after bringing them back on line, respectively, and use them instead of .prepare() and .finish() for ACPI suspend. Make the PM core execute the .prepare() and .finish() platform suspend callbacks where they were executed previously (that is, right after calling the regular suspend methods provided by device drivers and right before executing their regular resume methods, respectively). It is not necessary to make analogous changes to the hibernation code and data structures at the moment, because they are only used by ACPI platforms. Signed-off-by: Rafael J. Wysocki Reported-by: Russell King Acked-by: Len Brown --- drivers/acpi/sleep.c | 8 ++++---- include/linux/suspend.h | 36 ++++++++++++++++++++++++++---------- kernel/power/main.c | 24 +++++++++++++++++------- 3 files changed, 47 insertions(+), 21 deletions(-) diff --git a/drivers/acpi/sleep.c b/drivers/acpi/sleep.c index 779e4e500df4..d060e6fd7fd5 100644 --- a/drivers/acpi/sleep.c +++ b/drivers/acpi/sleep.c @@ -300,9 +300,9 @@ static int acpi_suspend_state_valid(suspend_state_t pm_state) static struct platform_suspend_ops acpi_suspend_ops = { .valid = acpi_suspend_state_valid, .begin = acpi_suspend_begin, - .prepare = acpi_pm_prepare, + .prepare_late = acpi_pm_prepare, .enter = acpi_suspend_enter, - .finish = acpi_pm_finish, + .wake = acpi_pm_finish, .end = acpi_pm_end, }; @@ -328,9 +328,9 @@ static int acpi_suspend_begin_old(suspend_state_t pm_state) static struct platform_suspend_ops acpi_suspend_ops_old = { .valid = acpi_suspend_state_valid, .begin = acpi_suspend_begin_old, - .prepare = acpi_pm_disable_gpes, + .prepare_late = acpi_pm_disable_gpes, .enter = acpi_suspend_enter, - .finish = acpi_pm_finish, + .wake = acpi_pm_finish, .end = acpi_pm_end, .recover = acpi_pm_finish, }; diff --git a/include/linux/suspend.h b/include/linux/suspend.h index 3e3a4364cbff..795032edfc46 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -58,10 +58,17 @@ typedef int __bitwise suspend_state_t; * by @begin(). * @prepare() is called right after devices have been suspended (ie. the * appropriate .suspend() method has been executed for each device) and - * before the nonboot CPUs are disabled (it is executed with IRQs enabled). - * This callback is optional. It returns 0 on success or a negative - * error code otherwise, in which case the system cannot enter the desired - * sleep state (@enter() and @finish() will not be called in that case). + * before device drivers' late suspend callbacks are executed. It returns + * 0 on success or a negative error code otherwise, in which case the + * system cannot enter the desired sleep state (@prepare_late(), @enter(), + * @wake(), and @finish() will not be called in that case). + * + * @prepare_late: Finish preparing the platform for entering the system sleep + * state indicated by @begin(). + * @prepare_late is called before disabling nonboot CPUs and after + * device drivers' late suspend callbacks have been executed. It returns + * 0 on success or a negative error code otherwise, in which case the + * system cannot enter the desired sleep state (@enter() and @wake()). * * @enter: Enter the system sleep state indicated by @begin() or represented by * the argument if @begin() is not implemented. @@ -69,19 +76,26 @@ typedef int __bitwise suspend_state_t; * error code otherwise, in which case the system cannot enter the desired * sleep state. * - * @finish: Called when the system has just left a sleep state, right after - * the nonboot CPUs have been enabled and before devices are resumed (it is - * executed with IRQs enabled). + * @wake: Called when the system has just left a sleep state, right after + * the nonboot CPUs have been enabled and before device drivers' early + * resume callbacks are executed. + * This callback is optional, but should be implemented by the platforms + * that implement @prepare_late(). If implemented, it is always called + * after @enter(), even if @enter() fails. + * + * @finish: Finish wake-up of the platform. + * @finish is called right prior to calling device drivers' regular suspend + * callbacks. * This callback is optional, but should be implemented by the platforms * that implement @prepare(). If implemented, it is always called after - * @enter() (even if @enter() fails). + * @enter() and @wake(), if implemented, even if any of them fails. * * @end: Called by the PM core right after resuming devices, to indicate to * the platform that the system has returned to the working state or * the transition to the sleep state has been aborted. * This callback is optional, but should be implemented by the platforms - * that implement @begin(), but platforms implementing @begin() should - * also provide a @end() which cleans up transitions aborted before + * that implement @begin(). Accordingly, platforms implementing @begin() + * should also provide a @end() which cleans up transitions aborted before * @enter(). * * @recover: Recover the platform from a suspend failure. @@ -93,7 +107,9 @@ struct platform_suspend_ops { int (*valid)(suspend_state_t state); int (*begin)(suspend_state_t state); int (*prepare)(void); + int (*prepare_late)(void); int (*enter)(suspend_state_t state); + void (*wake)(void); void (*finish)(void); void (*end)(void); void (*recover)(void); diff --git a/kernel/power/main.c b/kernel/power/main.c index f172f41858bb..f99ed6a75eac 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -291,20 +291,26 @@ static int suspend_enter(suspend_state_t state) device_pm_lock(); + if (suspend_ops->prepare) { + error = suspend_ops->prepare(); + if (error) + goto Done; + } + error = device_power_down(PMSG_SUSPEND); if (error) { printk(KERN_ERR "PM: Some devices failed to power down\n"); - goto Done; + goto Platfrom_finish; } - if (suspend_ops->prepare) { - error = suspend_ops->prepare(); + if (suspend_ops->prepare_late) { + error = suspend_ops->prepare_late(); if (error) goto Power_up_devices; } if (suspend_test(TEST_PLATFORM)) - goto Platfrom_finish; + goto Platform_wake; error = disable_nonboot_cpus(); if (error || suspend_test(TEST_CPUS)) @@ -326,13 +332,17 @@ static int suspend_enter(suspend_state_t state) Enable_cpus: enable_nonboot_cpus(); - Platfrom_finish: - if (suspend_ops->finish) - suspend_ops->finish(); + Platform_wake: + if (suspend_ops->wake) + suspend_ops->wake(); Power_up_devices: device_power_up(PMSG_RESUME); + Platfrom_finish: + if (suspend_ops->finish) + suspend_ops->finish(); + Done: device_pm_unlock(); -- cgit v1.2.3 From d9c6f546469f33f9aa48ae5991c33da8cd535b37 Mon Sep 17 00:00:00 2001 From: Roel Kluin Date: Thu, 16 Apr 2009 22:57:46 +0200 Subject: drm: count reaches -1 With a postfix decrement in the test count will reach -1 rather than 0, subsequent tests fail. Signed-off-by: Roel Kluin Signed-off-by: Dave Airlie --- drivers/gpu/drm/via/via_dma.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/via/via_dma.c b/drivers/gpu/drm/via/via_dma.c index 7a339dba6a69..bfb92d283260 100644 --- a/drivers/gpu/drm/via/via_dma.c +++ b/drivers/gpu/drm/via/via_dma.c @@ -481,11 +481,13 @@ static int via_wait_idle(drm_via_private_t * dev_priv) { int count = 10000000; - while (!(VIA_READ(VIA_REG_STATUS) & VIA_VR_QUEUE_BUSY) && count--); + while (!(VIA_READ(VIA_REG_STATUS) & VIA_VR_QUEUE_BUSY) && --count) + ; - while (count-- && (VIA_READ(VIA_REG_STATUS) & + while (count && (VIA_READ(VIA_REG_STATUS) & (VIA_CMD_RGTR_BUSY | VIA_2D_ENG_BUSY | - VIA_3D_ENG_BUSY))) ; + VIA_3D_ENG_BUSY))) + --count; return count; } @@ -705,7 +707,7 @@ static int via_cmdbuf_size(struct drm_device *dev, void *data, struct drm_file * switch (d_siz->func) { case VIA_CMDBUF_SPACE: while (((tmp_size = via_cmdbuf_space(dev_priv)) < d_siz->size) - && count--) { + && --count) { if (!d_siz->wait) { break; } @@ -717,7 +719,7 @@ static int via_cmdbuf_size(struct drm_device *dev, void *data, struct drm_file * break; case VIA_CMDBUF_LAG: while (((tmp_size = via_cmdbuf_lag(dev_priv)) > d_siz->size) - && count--) { + && --count) { if (!d_siz->wait) { break; } -- cgit v1.2.3 From 77d26dc9b9805f322f5a1f6e559b18ad66205bd9 Mon Sep 17 00:00:00 2001 From: Ma Ling Date: Thu, 16 Apr 2009 17:51:25 +0800 Subject: drm: clean dirty memory after device release In current code we register/unregister connector object by drm_sysfs_connector_add/remove function. However under some cases, we need to dynamically register or unregister device multiple times, so we have to go through register -> unregister ->register routine. Because after device_unregister function our memory is dirty, we need to do clean operation in order to re-register the device, otherwise the system will crash. The patch intends to clean device after device release. Signed-off-by: Ma Ling Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_sysfs.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c index bc0c6849360c..022876ae34f0 100644 --- a/drivers/gpu/drm/drm_sysfs.c +++ b/drivers/gpu/drm/drm_sysfs.c @@ -132,6 +132,7 @@ void drm_sysfs_destroy(void) */ static void drm_sysfs_device_release(struct device *dev) { + memset(dev, 0, sizeof(struct device)); return; } -- cgit v1.2.3 From 6b0084266c1d4917ad9259759a1e7bd623cb3888 Mon Sep 17 00:00:00 2001 From: Jonas Bonn Date: Thu, 16 Apr 2009 09:00:02 +0200 Subject: drm: set/clear is_master when master changed The variable is_master is being used to track the drm_file that is currently master, so its value needs to be updated accordingly when the master is changed. Signed-off-by: Jonas Bonn Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_stub.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_stub.c b/drivers/gpu/drm/drm_stub.c index d009661781bc..1b3e0ff57b45 100644 --- a/drivers/gpu/drm/drm_stub.c +++ b/drivers/gpu/drm/drm_stub.c @@ -159,6 +159,9 @@ void drm_master_put(struct drm_master **master) int drm_setmaster_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv) { + if (file_priv->is_master) + return 0; + if (file_priv->minor->master && file_priv->minor->master != file_priv->master) return -EINVAL; @@ -169,6 +172,7 @@ int drm_setmaster_ioctl(struct drm_device *dev, void *data, file_priv->minor->master != file_priv->master) { mutex_lock(&dev->struct_mutex); file_priv->minor->master = drm_master_get(file_priv->master); + file_priv->is_master = 1; mutex_unlock(&dev->struct_mutex); } @@ -178,10 +182,12 @@ int drm_setmaster_ioctl(struct drm_device *dev, void *data, int drm_dropmaster_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv) { - if (!file_priv->master) + if (!file_priv->is_master) return -EINVAL; + mutex_lock(&dev->struct_mutex); drm_master_put(&file_priv->minor->master); + file_priv->is_master = 0; mutex_unlock(&dev->struct_mutex); return 0; } -- cgit v1.2.3 From 07f1c7a7f6736d9ec2eba57d209c5f48888d841e Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Mon, 20 Apr 2009 09:32:50 +1000 Subject: drm: check for minor master before allowing drop master. When fast user switching a lot eventually we get to the point, where we were checking for the wrong thing in this function. Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_stub.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/drm_stub.c b/drivers/gpu/drm/drm_stub.c index 1b3e0ff57b45..ef878615c49f 100644 --- a/drivers/gpu/drm/drm_stub.c +++ b/drivers/gpu/drm/drm_stub.c @@ -185,6 +185,9 @@ int drm_dropmaster_ioctl(struct drm_device *dev, void *data, if (!file_priv->is_master) return -EINVAL; + if (!file_priv->minor->master) + return -EINVAL; + mutex_lock(&dev->struct_mutex); drm_master_put(&file_priv->minor->master); file_priv->is_master = 0; -- cgit v1.2.3 From 59de2bebabc5027f93df999d59cc65df591c3e6e Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Mon, 20 Apr 2009 10:08:35 +1000 Subject: agp: zero pages before sending to userspace AGP pages might be mapped into userspace finally, so the pages should be set to zero before userspace can use it. Otherwise there is potential information leakage. Signed-off-by: Shaohua Li Signed-off-by: Dave Airlie --- drivers/char/agp/generic.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/char/agp/generic.c b/drivers/char/agp/generic.c index 10d6cbd7c05e..2224b762b7fb 100644 --- a/drivers/char/agp/generic.c +++ b/drivers/char/agp/generic.c @@ -1226,7 +1226,7 @@ int agp_generic_alloc_pages(struct agp_bridge_data *bridge, struct agp_memory *m int i, ret = -ENOMEM; for (i = 0; i < num_pages; i++) { - page = alloc_page(GFP_KERNEL | GFP_DMA32); + page = alloc_page(GFP_KERNEL | GFP_DMA32 | __GFP_ZERO); /* agp_free_memory() needs gart address */ if (page == NULL) goto out; @@ -1257,7 +1257,7 @@ void *agp_generic_alloc_page(struct agp_bridge_data *bridge) { struct page * page; - page = alloc_page(GFP_KERNEL | GFP_DMA32); + page = alloc_page(GFP_KERNEL | GFP_DMA32 | __GFP_ZERO); if (page == NULL) return NULL; -- cgit v1.2.3 From 1f59390339f263c0fe7908fbe54466dbf3a64b58 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 20 Apr 2009 11:50:24 +1000 Subject: md: support bitmaps on RAID10 arrays larger then 2 terabytes .. and other arrays with components larger than 2 terabytes. We use a "long" rather than a "sector_t" in part of the bitmap size calculations, which is sad. Reported-by: "Mario 'BitKoenig' Holbe" Signed-off-by: NeilBrown --- drivers/md/bitmap.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index e4510c976b34..1fb91edc7de2 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -1590,7 +1590,7 @@ void bitmap_destroy(mddev_t *mddev) int bitmap_create(mddev_t *mddev) { struct bitmap *bitmap; - unsigned long blocks = mddev->resync_max_sectors; + sector_t blocks = mddev->resync_max_sectors; unsigned long chunks; unsigned long pages; struct file *file = mddev->bitmap_file; @@ -1632,8 +1632,8 @@ int bitmap_create(mddev_t *mddev) bitmap->chunkshift = ffz(~bitmap->chunksize); /* now that chunksize and chunkshift are set, we can use these macros */ - chunks = (blocks + CHUNK_BLOCK_RATIO(bitmap) - 1) / - CHUNK_BLOCK_RATIO(bitmap); + chunks = (blocks + CHUNK_BLOCK_RATIO(bitmap) - 1) >> + CHUNK_BLOCK_SHIFT(bitmap); pages = (chunks + PAGE_COUNTER_RATIO - 1) / PAGE_COUNTER_RATIO; BUG_ON(!pages); -- cgit v1.2.3 From 52fcd11c0900b0cbc584eeda12a6e27dd6c9d046 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Mon, 20 Apr 2009 08:58:45 +0100 Subject: GFS2: Clear dirty bit at end of inode glock sync The dirty bit can get set during the inode glock sync. Its too complicated to change that at the moment, so this is the quick fix - to clear the bit again at the end of the function. Signed-off-by: Steven Whitehouse --- fs/gfs2/glops.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index bf23a62aa925..70f87f43afa2 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -156,6 +156,12 @@ static void inode_go_sync(struct gfs2_glock *gl) error = filemap_fdatawait(metamapping); mapping_set_error(metamapping, error); gfs2_ail_empty_gl(gl); + /* + * Writeback of the data mapping may cause the dirty flag to be set + * so we have to clear it again here. + */ + smp_mb__before_clear_bit(); + clear_bit(GLF_DIRTY, &gl->gl_flags); } /** -- cgit v1.2.3 From e56985da455b9dc0591b8cb2006cc94b6f4fb0f4 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Mon, 20 Apr 2009 09:45:54 +0100 Subject: GFS2: Fix page_mkwrite() return code This allows for the possibility of returning VM_FAULT_OOM as well as VM_FAULT_SIGBUS. This ensures that the correct action is taken. Signed-off-by: Steven Whitehouse --- fs/gfs2/ops_file.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/gfs2/ops_file.c b/fs/gfs2/ops_file.c index 101caf3ee861..5d82e91887e3 100644 --- a/fs/gfs2/ops_file.c +++ b/fs/gfs2/ops_file.c @@ -413,7 +413,9 @@ out_unlock: gfs2_glock_dq(&gh); out: gfs2_holder_uninit(&gh); - if (ret) + if (ret == -ENOMEM) + ret = VM_FAULT_OOM; + else if (ret) ret = VM_FAULT_SIGBUS; return ret; } -- cgit v1.2.3 From 8a577ffc75d9194fe8cdb7479236f2081c26ca1f Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Sat, 18 Apr 2009 15:05:45 -0700 Subject: driver: dont update dev_name via device_add path notice one system /proc/iomem some entries missed the name for pci_devices it turns that dev->dev.kobj name is changed after device_add. for pci code: via acpi_pci_root_driver.ops.add (aka acpi_pci_root_add) ==> pci_acpi_scan_root is used to scan pci bus/device, and at the same time we read the resource for pci_dev in the pci_read_bases, we have res->name = pci_name(pci_dev); pci_name is calling dev_name. later via acpi_pci_root_driver.ops.start (aka acpi_pci_root_start) ==> pci_bus_add_device to add all pci_dev in kobj tree. pci_bus_add_device will call device_add. actually in device_add /* first, register with generic layer. */ error = kobject_add(&dev->kobj, dev->kobj.parent, "%s", dev_name(dev)); if (error) goto Error; will get one new name for that kobj, old name is freed. [Impact: fix corrupted names in /proc/iomem ] Signed-off-by: Yinghai Lu Signed-off-by: Linus Torvalds --- drivers/base/core.c | 3 ++- lib/kobject.c | 3 +++ 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/base/core.c b/drivers/base/core.c index d230ff4b3eec..4aa527b8a913 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -891,7 +891,8 @@ int device_add(struct device *dev) set_dev_node(dev, dev_to_node(parent)); /* first, register with generic layer. */ - error = kobject_add(&dev->kobj, dev->kobj.parent, "%s", dev_name(dev)); + /* we require the name to be set before, and pass NULL */ + error = kobject_add(&dev->kobj, dev->kobj.parent, NULL); if (error) goto Error; diff --git a/lib/kobject.c b/lib/kobject.c index a6dec32f2ddd..bacf6fe4f7a0 100644 --- a/lib/kobject.c +++ b/lib/kobject.c @@ -218,6 +218,9 @@ int kobject_set_name_vargs(struct kobject *kobj, const char *fmt, const char *old_name = kobj->name; char *s; + if (kobj->name && !fmt) + return 0; + kobj->name = kvasprintf(GFP_KERNEL, fmt, vargs); if (!kobj->name) return -ENOMEM; -- cgit v1.2.3 From f297bfee71df4f1cc5bacd4dea330e1241ba61ae Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 20 Apr 2009 12:46:24 +0100 Subject: FRV: Don't attempt to #include as it doesn't exist Stop the FRV arch from attempting to #include as it doesn't exist. Reported-by: Robert P. J. Day Signed-off-by: David Howells Signed-off-by: Linus Torvalds --- arch/frv/kernel/setup.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/frv/kernel/setup.c b/arch/frv/kernel/setup.c index 0669e1382383..55e4fab7c0bc 100644 --- a/arch/frv/kernel/setup.c +++ b/arch/frv/kernel/setup.c @@ -46,7 +46,6 @@ #include #ifdef CONFIG_BLK_DEV_INITRD -#include #include #endif -- cgit v1.2.3 From a5432f5ad43822802ba6f8ba8662ca6361712044 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 20 Apr 2009 15:46:45 +0100 Subject: FS-Cache: Add MAINTAINERS record for FS-Cache and CacheFiles Add MAINTAINERS record for FS-Cache and CacheFiles. Signed-off-by: David Howells Signed-off-by: Linus Torvalds --- MAINTAINERS | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 1e067a675e53..6baace38b24d 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1287,6 +1287,14 @@ S: Maintained F: Documentation/video4linux/bttv/ F: drivers/media/video/bt8xx/bttv* +CACHEFILES: FS-CACHE BACKEND FOR CACHING ON MOUNTED FILESYSTEMS +P: David Howells +M: dhowells@redhat.com +L: linux-cachefs@redhat.com +S: Supported +F: Documentation/filesystems/caching/cachefiles.txt +F: fs/cachefiles/ + CAFE CMOS INTEGRATED CAMERA CONTROLLER DRIVER P: Jonathan Corbet M: corbet@lwn.net @@ -2325,6 +2333,15 @@ F: Documentation/power/freezing-of-tasks.txt F: include/linux/freezer.h F: kernel/freezer.c +FS-CACHE: LOCAL CACHING FOR NETWORK FILESYSTEMS +P: David Howells +M: dhowells@redhat.com +L: linux-cachefs@redhat.com +S: Supported +F: Documentation/filesystems/caching/ +F: fs/fscache/ +F: include/linux/fscache*.h + FTRACE P: Steven Rostedt M: rostedt@goodmis.org -- cgit v1.2.3 From 05f0ecbda5c215279f8e0f852e1606f441a11236 Mon Sep 17 00:00:00 2001 From: Isaku Yamahata Date: Sat, 18 Apr 2009 12:15:23 +0900 Subject: [IA64] fix allmodconfig compilation breakage. This patch fixes the following compilation error caused by recursive inclusion of kernel.h which defines BUILD_BUG_ON(). In this case, the case it catches will be caught by the case CONFIG_PARAVIRT=n, so removing it would not hurt compile time check very much. So fix the breakage by removing it. CC arch/ia64/kernel/asm-offsets.s In file included from include/linux/bitops.h:17, from include/linux/kernel.h:15, from include/linux/sched.h:52, from arch/ia64/kernel/asm-offsets.c:9: arch/ia64/include/asm/bitops.h: In function 'set_bit': arch/ia64/include/asm/bitops.h:47: error: implicit declaration of function 'BUILD_BUG_ON' Signed-off-by: Isaku Yamahata Signed-off-by: Tony Luck --- arch/ia64/include/asm/paravirt_privop.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/arch/ia64/include/asm/paravirt_privop.h b/arch/ia64/include/asm/paravirt_privop.h index 3d2951130b5f..8f6cb11c9fae 100644 --- a/arch/ia64/include/asm/paravirt_privop.h +++ b/arch/ia64/include/asm/paravirt_privop.h @@ -445,7 +445,6 @@ paravirt_set_rr0_to_rr4(unsigned long val0, unsigned long val1, register unsigned long ia64_intri_res asm ("r8"); \ register unsigned long __reg asm ("r8") = (reg); \ \ - BUILD_BUG_ON(!__builtin_constant_p(reg)); \ asm volatile (paravirt_alt_bundle(__PARAVIRT_BR, \ PARAVIRT_TYPE(GETREG) \ + (reg)) \ @@ -464,7 +463,6 @@ paravirt_set_rr0_to_rr4(unsigned long val0, unsigned long val1, register unsigned long ia64_clobber1 asm ("r8"); \ register unsigned long ia64_clobber2 asm ("r9"); \ \ - BUILD_BUG_ON(!__builtin_constant_p(reg)); \ asm volatile (paravirt_alt_bundle(__PARAVIRT_BR, \ PARAVIRT_TYPE(SETREG) \ + (reg)) \ -- cgit v1.2.3 From ffbd517d5a8c8e93ddd11046434fb029f3df73aa Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Mon, 20 Apr 2009 15:50:09 -0400 Subject: Btrfs: use WRITE_SYNC for synchronous writes Part of reducing fsync/O_SYNC/O_DIRECT latencies is using WRITE_SYNC for writes we plan on waiting on in the near future. This patch mirrors recent changes in other filesystems and the generic code to use WRITE_SYNC when WB_SYNC_ALL is passed and to use WRITE_SYNC for other latency critical writes. Btrfs uses async worker threads for checksumming before the write is done, and then again to actually submit the bios. The bio submission code just runs a per-device list of bios that need to be sent down the pipe. This list is split into low priority and high priority lists so the WRITE_SYNC IO happens first. Signed-off-by: Chris Mason --- fs/btrfs/disk-io.c | 4 +- fs/btrfs/extent_io.c | 44 +++++++++++------ fs/btrfs/ordered-data.c | 2 +- fs/btrfs/volumes.c | 124 +++++++++++++++++++++++++++++++++++++----------- fs/btrfs/volumes.h | 13 ++++- 5 files changed, 141 insertions(+), 46 deletions(-) diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 92caa8035f36..fec18b43c2c3 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -2095,10 +2095,10 @@ static int write_dev_supers(struct btrfs_device *device, device->barriers = 0; get_bh(bh); lock_buffer(bh); - ret = submit_bh(WRITE, bh); + ret = submit_bh(WRITE_SYNC, bh); } } else { - ret = submit_bh(WRITE, bh); + ret = submit_bh(WRITE_SYNC, bh); } if (!ret && wait) { diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index eb2bee8b7fbf..483b6727aaaf 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -50,7 +50,10 @@ struct extent_page_data { /* tells writepage not to lock the state bits for this range * it still does the unlocking */ - int extent_locked; + unsigned int extent_locked:1; + + /* tells the submit_bio code to use a WRITE_SYNC */ + unsigned int sync_io:1; }; int __init extent_io_init(void) @@ -2136,8 +2139,14 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, u64 delalloc_end; int page_started; int compressed; + int write_flags; unsigned long nr_written = 0; + if (wbc->sync_mode == WB_SYNC_ALL) + write_flags = WRITE_SYNC_PLUG; + else + write_flags = WRITE; + WARN_ON(!PageLocked(page)); pg_offset = i_size & (PAGE_CACHE_SIZE - 1); if (page->index > end_index || @@ -2314,9 +2323,9 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, (unsigned long long)end); } - ret = submit_extent_page(WRITE, tree, page, sector, - iosize, pg_offset, bdev, - &epd->bio, max_nr, + ret = submit_extent_page(write_flags, tree, page, + sector, iosize, pg_offset, + bdev, &epd->bio, max_nr, end_bio_extent_writepage, 0, 0, 0); if (ret) @@ -2460,15 +2469,23 @@ retry: return ret; } -static noinline void flush_write_bio(void *data) +static void flush_epd_write_bio(struct extent_page_data *epd) { - struct extent_page_data *epd = data; if (epd->bio) { - submit_one_bio(WRITE, epd->bio, 0, 0); + if (epd->sync_io) + submit_one_bio(WRITE_SYNC, epd->bio, 0, 0); + else + submit_one_bio(WRITE, epd->bio, 0, 0); epd->bio = NULL; } } +static noinline void flush_write_bio(void *data) +{ + struct extent_page_data *epd = data; + flush_epd_write_bio(epd); +} + int extent_write_full_page(struct extent_io_tree *tree, struct page *page, get_extent_t *get_extent, struct writeback_control *wbc) @@ -2480,6 +2497,7 @@ int extent_write_full_page(struct extent_io_tree *tree, struct page *page, .tree = tree, .get_extent = get_extent, .extent_locked = 0, + .sync_io = wbc->sync_mode == WB_SYNC_ALL, }; struct writeback_control wbc_writepages = { .bdi = wbc->bdi, @@ -2490,13 +2508,11 @@ int extent_write_full_page(struct extent_io_tree *tree, struct page *page, .range_end = (loff_t)-1, }; - ret = __extent_writepage(page, wbc, &epd); extent_write_cache_pages(tree, mapping, &wbc_writepages, __extent_writepage, &epd, flush_write_bio); - if (epd.bio) - submit_one_bio(WRITE, epd.bio, 0, 0); + flush_epd_write_bio(&epd); return ret; } @@ -2515,6 +2531,7 @@ int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode, .tree = tree, .get_extent = get_extent, .extent_locked = 1, + .sync_io = mode == WB_SYNC_ALL, }; struct writeback_control wbc_writepages = { .bdi = inode->i_mapping->backing_dev_info, @@ -2540,8 +2557,7 @@ int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode, start += PAGE_CACHE_SIZE; } - if (epd.bio) - submit_one_bio(WRITE, epd.bio, 0, 0); + flush_epd_write_bio(&epd); return ret; } @@ -2556,13 +2572,13 @@ int extent_writepages(struct extent_io_tree *tree, .tree = tree, .get_extent = get_extent, .extent_locked = 0, + .sync_io = wbc->sync_mode == WB_SYNC_ALL, }; ret = extent_write_cache_pages(tree, mapping, wbc, __extent_writepage, &epd, flush_write_bio); - if (epd.bio) - submit_one_bio(WRITE, epd.bio, 0, 0); + flush_epd_write_bio(&epd); return ret; } diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index 53c87b197d70..d6f0806c682f 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -489,7 +489,7 @@ again: /* start IO across the range first to instantiate any delalloc * extents */ - btrfs_fdatawrite_range(inode->i_mapping, start, orig_end, WB_SYNC_NONE); + btrfs_fdatawrite_range(inode->i_mapping, start, orig_end, WB_SYNC_ALL); /* The compression code will leave pages locked but return from * writepage without setting the page writeback. Starting again diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index e0913e469728..e53835b88594 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -125,6 +125,20 @@ static noinline struct btrfs_fs_devices *find_fsid(u8 *fsid) return NULL; } +static void requeue_list(struct btrfs_pending_bios *pending_bios, + struct bio *head, struct bio *tail) +{ + + struct bio *old_head; + + old_head = pending_bios->head; + pending_bios->head = head; + if (pending_bios->tail) + tail->bi_next = old_head; + else + pending_bios->tail = tail; +} + /* * we try to collect pending bios for a device so we don't get a large * number of procs sending bios down to the same device. This greatly @@ -141,10 +155,12 @@ static noinline int run_scheduled_bios(struct btrfs_device *device) struct bio *pending; struct backing_dev_info *bdi; struct btrfs_fs_info *fs_info; + struct btrfs_pending_bios *pending_bios; struct bio *tail; struct bio *cur; int again = 0; - unsigned long num_run = 0; + unsigned long num_run; + unsigned long num_sync_run; unsigned long limit; unsigned long last_waited = 0; @@ -153,20 +169,30 @@ static noinline int run_scheduled_bios(struct btrfs_device *device) limit = btrfs_async_submit_limit(fs_info); limit = limit * 2 / 3; + /* we want to make sure that every time we switch from the sync + * list to the normal list, we unplug + */ + num_sync_run = 0; + loop: spin_lock(&device->io_lock); + num_run = 0; loop_lock: + /* take all the bios off the list at once and process them * later on (without the lock held). But, remember the * tail and other pointers so the bios can be properly reinserted * into the list if we hit congestion */ - pending = device->pending_bios; - tail = device->pending_bio_tail; + if (device->pending_sync_bios.head) + pending_bios = &device->pending_sync_bios; + else + pending_bios = &device->pending_bios; + + pending = pending_bios->head; + tail = pending_bios->tail; WARN_ON(pending && !tail); - device->pending_bios = NULL; - device->pending_bio_tail = NULL; /* * if pending was null this time around, no bios need processing @@ -176,16 +202,41 @@ loop_lock: * device->running_pending is used to synchronize with the * schedule_bio code. */ - if (pending) { - again = 1; - device->running_pending = 1; - } else { + if (device->pending_sync_bios.head == NULL && + device->pending_bios.head == NULL) { again = 0; device->running_pending = 0; + } else { + again = 1; + device->running_pending = 1; } + + pending_bios->head = NULL; + pending_bios->tail = NULL; + spin_unlock(&device->io_lock); + /* + * if we're doing the regular priority list, make sure we unplug + * for any high prio bios we've sent down + */ + if (pending_bios == &device->pending_bios && num_sync_run > 0) { + num_sync_run = 0; + blk_run_backing_dev(bdi, NULL); + } + while (pending) { + + rmb(); + if (pending_bios != &device->pending_sync_bios && + device->pending_sync_bios.head && + num_run > 16) { + cond_resched(); + spin_lock(&device->io_lock); + requeue_list(pending_bios, pending, tail); + goto loop_lock; + } + cur = pending; pending = pending->bi_next; cur->bi_next = NULL; @@ -196,10 +247,18 @@ loop_lock: wake_up(&fs_info->async_submit_wait); BUG_ON(atomic_read(&cur->bi_cnt) == 0); - bio_get(cur); submit_bio(cur->bi_rw, cur); - bio_put(cur); num_run++; + if (bio_sync(cur)) + num_sync_run++; + + if (need_resched()) { + if (num_sync_run) { + blk_run_backing_dev(bdi, NULL); + num_sync_run = 0; + } + cond_resched(); + } /* * we made progress, there is more work to do and the bdi @@ -208,7 +267,6 @@ loop_lock: */ if (pending && bdi_write_congested(bdi) && num_run > 16 && fs_info->fs_devices->open_devices > 1) { - struct bio *old_head; struct io_context *ioc; ioc = current->io_context; @@ -233,17 +291,17 @@ loop_lock: * against it before looping */ last_waited = ioc->last_waited; + if (need_resched()) { + if (num_sync_run) { + blk_run_backing_dev(bdi, NULL); + num_sync_run = 0; + } + cond_resched(); + } continue; } spin_lock(&device->io_lock); - - old_head = device->pending_bios; - device->pending_bios = pending; - if (device->pending_bio_tail) - tail->bi_next = old_head; - else - device->pending_bio_tail = tail; - + requeue_list(pending_bios, pending, tail); device->running_pending = 1; spin_unlock(&device->io_lock); @@ -251,11 +309,18 @@ loop_lock: goto done; } } + + if (num_sync_run) { + num_sync_run = 0; + blk_run_backing_dev(bdi, NULL); + } + + cond_resched(); if (again) goto loop; spin_lock(&device->io_lock); - if (device->pending_bios) + if (device->pending_bios.head || device->pending_sync_bios.head) goto loop_lock; spin_unlock(&device->io_lock); @@ -2497,7 +2562,7 @@ again: max_errors = 1; } } - if (multi_ret && rw == WRITE && + if (multi_ret && (rw & (1 << BIO_RW)) && stripes_allocated < stripes_required) { stripes_allocated = map->num_stripes; free_extent_map(em); @@ -2762,6 +2827,7 @@ static noinline int schedule_bio(struct btrfs_root *root, int rw, struct bio *bio) { int should_queue = 1; + struct btrfs_pending_bios *pending_bios; /* don't bother with additional async steps for reads, right now */ if (!(rw & (1 << BIO_RW))) { @@ -2783,13 +2849,17 @@ static noinline int schedule_bio(struct btrfs_root *root, bio->bi_rw |= rw; spin_lock(&device->io_lock); + if (bio_sync(bio)) + pending_bios = &device->pending_sync_bios; + else + pending_bios = &device->pending_bios; - if (device->pending_bio_tail) - device->pending_bio_tail->bi_next = bio; + if (pending_bios->tail) + pending_bios->tail->bi_next = bio; - device->pending_bio_tail = bio; - if (!device->pending_bios) - device->pending_bios = bio; + pending_bios->tail = bio; + if (!pending_bios->head) + pending_bios->head = bio; if (device->running_pending) should_queue = 0; diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index 2185de72ff7d..5836327ba5dd 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -23,13 +23,22 @@ #include "async-thread.h" struct buffer_head; +struct btrfs_pending_bios { + struct bio *head; + struct bio *tail; +}; + struct btrfs_device { struct list_head dev_list; struct list_head dev_alloc_list; struct btrfs_fs_devices *fs_devices; struct btrfs_root *dev_root; - struct bio *pending_bios; - struct bio *pending_bio_tail; + + /* regular prio bios */ + struct btrfs_pending_bios pending_bios; + /* WRITE_SYNC bios */ + struct btrfs_pending_bios pending_sync_bios; + int running_pending; u64 generation; -- cgit v1.2.3 From d313d7a31a752c88f7288692bd98e66d0789779b Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Mon, 20 Apr 2009 15:50:09 -0400 Subject: Btrfs: add a priority queue to the async thread helpers Btrfs is using WRITE_SYNC_PLUG to send down synchronous IOs with a higher priority. But, the checksumming helper threads prevent it from being fully effective. There are two problems. First, a big queue of pending checksumming will delay the synchronous IO behind other lower priority writes. Second, the checksumming uses an ordered async work queue. The ordering makes sure that IOs are sent to the block layer in the same order they are sent to the checksumming threads. Usually this gives us less seeky IO. But, when we start mixing IO priorities, the lower priority IO can delay the higher priority IO. This patch solves both problems by adding a high priority list to the async helper threads, and a new btrfs_set_work_high_prio(), which is used to make put a new async work item onto the higher priority list. The ordering is still done on high priority IO, but all of the high priority bios are ordered separately from the low priority bios. This ordering is purely an IO optimization, it is not involved in data or metadata integrity. Signed-off-by: Chris Mason --- fs/btrfs/async-thread.c | 60 ++++++++++++++++++++++++++++++++++++++----------- fs/btrfs/async-thread.h | 2 ++ fs/btrfs/disk-io.c | 5 +++++ fs/btrfs/extent_io.c | 2 +- fs/btrfs/file.c | 2 +- 5 files changed, 56 insertions(+), 15 deletions(-) diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c index 51bfdfc8fcda..502c3d61de62 100644 --- a/fs/btrfs/async-thread.c +++ b/fs/btrfs/async-thread.c @@ -25,6 +25,7 @@ #define WORK_QUEUED_BIT 0 #define WORK_DONE_BIT 1 #define WORK_ORDER_DONE_BIT 2 +#define WORK_HIGH_PRIO_BIT 3 /* * container for the kthread task pointer and the list of pending work @@ -36,6 +37,7 @@ struct btrfs_worker_thread { /* list of struct btrfs_work that are waiting for service */ struct list_head pending; + struct list_head prio_pending; /* list of worker threads from struct btrfs_workers */ struct list_head worker_list; @@ -103,10 +105,16 @@ static noinline int run_ordered_completions(struct btrfs_workers *workers, spin_lock_irqsave(&workers->lock, flags); - while (!list_empty(&workers->order_list)) { - work = list_entry(workers->order_list.next, - struct btrfs_work, order_list); - + while (1) { + if (!list_empty(&workers->prio_order_list)) { + work = list_entry(workers->prio_order_list.next, + struct btrfs_work, order_list); + } else if (!list_empty(&workers->order_list)) { + work = list_entry(workers->order_list.next, + struct btrfs_work, order_list); + } else { + break; + } if (!test_bit(WORK_DONE_BIT, &work->flags)) break; @@ -143,8 +151,14 @@ static int worker_loop(void *arg) do { spin_lock_irq(&worker->lock); again_locked: - while (!list_empty(&worker->pending)) { - cur = worker->pending.next; + while (1) { + if (!list_empty(&worker->prio_pending)) + cur = worker->prio_pending.next; + else if (!list_empty(&worker->pending)) + cur = worker->pending.next; + else + break; + work = list_entry(cur, struct btrfs_work, list); list_del(&work->list); clear_bit(WORK_QUEUED_BIT, &work->flags); @@ -163,7 +177,6 @@ again_locked: spin_lock_irq(&worker->lock); check_idle_worker(worker); - } if (freezing(current)) { worker->working = 0; @@ -178,7 +191,8 @@ again_locked: * jump_in? */ smp_mb(); - if (!list_empty(&worker->pending)) + if (!list_empty(&worker->pending) || + !list_empty(&worker->prio_pending)) continue; /* @@ -191,7 +205,8 @@ again_locked: */ schedule_timeout(1); smp_mb(); - if (!list_empty(&worker->pending)) + if (!list_empty(&worker->pending) || + !list_empty(&worker->prio_pending)) continue; if (kthread_should_stop()) @@ -200,7 +215,8 @@ again_locked: /* still no more work?, sleep for real */ spin_lock_irq(&worker->lock); set_current_state(TASK_INTERRUPTIBLE); - if (!list_empty(&worker->pending)) + if (!list_empty(&worker->pending) || + !list_empty(&worker->prio_pending)) goto again_locked; /* @@ -248,6 +264,7 @@ void btrfs_init_workers(struct btrfs_workers *workers, char *name, int max) INIT_LIST_HEAD(&workers->worker_list); INIT_LIST_HEAD(&workers->idle_list); INIT_LIST_HEAD(&workers->order_list); + INIT_LIST_HEAD(&workers->prio_order_list); spin_lock_init(&workers->lock); workers->max_workers = max; workers->idle_thresh = 32; @@ -273,6 +290,7 @@ int btrfs_start_workers(struct btrfs_workers *workers, int num_workers) } INIT_LIST_HEAD(&worker->pending); + INIT_LIST_HEAD(&worker->prio_pending); INIT_LIST_HEAD(&worker->worker_list); spin_lock_init(&worker->lock); atomic_set(&worker->num_pending, 0); @@ -396,7 +414,10 @@ int btrfs_requeue_work(struct btrfs_work *work) goto out; spin_lock_irqsave(&worker->lock, flags); - list_add_tail(&work->list, &worker->pending); + if (test_bit(WORK_HIGH_PRIO_BIT, &work->flags)) + list_add_tail(&work->list, &worker->prio_pending); + else + list_add_tail(&work->list, &worker->pending); atomic_inc(&worker->num_pending); /* by definition we're busy, take ourselves off the idle @@ -422,6 +443,11 @@ out: return 0; } +void btrfs_set_work_high_prio(struct btrfs_work *work) +{ + set_bit(WORK_HIGH_PRIO_BIT, &work->flags); +} + /* * places a struct btrfs_work into the pending queue of one of the kthreads */ @@ -438,7 +464,12 @@ int btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work) worker = find_worker(workers); if (workers->ordered) { spin_lock_irqsave(&workers->lock, flags); - list_add_tail(&work->order_list, &workers->order_list); + if (test_bit(WORK_HIGH_PRIO_BIT, &work->flags)) { + list_add_tail(&work->order_list, + &workers->prio_order_list); + } else { + list_add_tail(&work->order_list, &workers->order_list); + } spin_unlock_irqrestore(&workers->lock, flags); } else { INIT_LIST_HEAD(&work->order_list); @@ -446,7 +477,10 @@ int btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work) spin_lock_irqsave(&worker->lock, flags); - list_add_tail(&work->list, &worker->pending); + if (test_bit(WORK_HIGH_PRIO_BIT, &work->flags)) + list_add_tail(&work->list, &worker->prio_pending); + else + list_add_tail(&work->list, &worker->pending); atomic_inc(&worker->num_pending); check_busy_worker(worker); diff --git a/fs/btrfs/async-thread.h b/fs/btrfs/async-thread.h index 31be4ed8b63e..1b511c109db6 100644 --- a/fs/btrfs/async-thread.h +++ b/fs/btrfs/async-thread.h @@ -85,6 +85,7 @@ struct btrfs_workers { * of work items waiting for completion */ struct list_head order_list; + struct list_head prio_order_list; /* lock for finding the next worker thread to queue on */ spinlock_t lock; @@ -98,4 +99,5 @@ int btrfs_start_workers(struct btrfs_workers *workers, int num_workers); int btrfs_stop_workers(struct btrfs_workers *workers); void btrfs_init_workers(struct btrfs_workers *workers, char *name, int max); int btrfs_requeue_work(struct btrfs_work *work); +void btrfs_set_work_high_prio(struct btrfs_work *work); #endif diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index fec18b43c2c3..a6b83744b05d 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -579,6 +579,10 @@ int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode, async->bio_flags = bio_flags; atomic_inc(&fs_info->nr_async_submits); + + if (rw & (1 << BIO_RW_SYNCIO)) + btrfs_set_work_high_prio(&async->work); + btrfs_queue_worker(&fs_info->workers, &async->work); #if 0 int limit = btrfs_async_submit_limit(fs_info); @@ -656,6 +660,7 @@ static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio, return btrfs_map_bio(BTRFS_I(inode)->root, rw, bio, mirror_num, 0); } + /* * kthread helpers are used to submit writes so that checksumming * can happen in parallel across all CPUs diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 483b6727aaaf..5d66cb27e422 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2501,7 +2501,7 @@ int extent_write_full_page(struct extent_io_tree *tree, struct page *page, }; struct writeback_control wbc_writepages = { .bdi = wbc->bdi, - .sync_mode = WB_SYNC_NONE, + .sync_mode = wbc->sync_mode, .older_than_this = NULL, .nr_to_write = 64, .range_start = page_offset(page) + PAGE_CACHE_SIZE, diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 9c9fb46ccd08..e21c0060ee73 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1131,7 +1131,7 @@ static ssize_t btrfs_file_write(struct file *file, const char __user *buf, if (will_write) { btrfs_fdatawrite_range(inode->i_mapping, pos, pos + write_bytes - 1, - WB_SYNC_NONE); + WB_SYNC_ALL); } else { balance_dirty_pages_ratelimited_nr(inode->i_mapping, num_pages); -- cgit v1.2.3 From 11c8349b4eb68f2b04cd8ece577377e6c0e5dd4b Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Mon, 20 Apr 2009 15:50:09 -0400 Subject: Btrfs: fix oops on page->mapping->host during writepage The extent_io writepage call updates the writepage index in the inode as it makes progress. But, it was doing the update after unlocking the page, which isn't legal because page->mapping can't be trusted once the page is unlocked. This lead to an oops, especially common with compression turned on. The fix here is to update the writeback index before unlocking the page. Signed-off-by: Chris Mason --- fs/btrfs/extent_io.c | 40 ++++++++++++++++++++++++++++++++-------- 1 file changed, 32 insertions(+), 8 deletions(-) diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 5d66cb27e422..05a1c42e25bf 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2104,6 +2104,16 @@ int extent_read_full_page(struct extent_io_tree *tree, struct page *page, return ret; } +static noinline void update_nr_written(struct page *page, + struct writeback_control *wbc, + unsigned long nr_written) +{ + wbc->nr_to_write -= nr_written; + if (wbc->range_cyclic || (wbc->nr_to_write > 0 && + wbc->range_start == 0 && wbc->range_end == LLONG_MAX)) + page->mapping->writeback_index = page->index + nr_written; +} + /* * the writepage semantics are similar to regular writepage. extent * records are inserted to lock ranges in the tree, and as dirty areas @@ -2173,6 +2183,12 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, delalloc_end = 0; page_started = 0; if (!epd->extent_locked) { + /* + * make sure the wbc mapping index is at least updated + * to this page. + */ + update_nr_written(page, wbc, 0); + while (delalloc_end < page_end) { nr_delalloc = find_lock_delalloc_range(inode, tree, page, @@ -2194,7 +2210,13 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, */ if (page_started) { ret = 0; - goto update_nr_written; + /* + * we've unlocked the page, so we can't update + * the mapping's writeback index, just update + * nr_to_write. + */ + wbc->nr_to_write -= nr_written; + goto done_unlocked; } } lock_extent(tree, start, page_end, GFP_NOFS); @@ -2207,13 +2229,18 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, if (ret == -EAGAIN) { unlock_extent(tree, start, page_end, GFP_NOFS); redirty_page_for_writepage(wbc, page); + update_nr_written(page, wbc, nr_written); unlock_page(page); ret = 0; - goto update_nr_written; + goto done_unlocked; } } - nr_written++; + /* + * we don't want to touch the inode after unlocking the page, + * so we update the mapping writeback index now + */ + update_nr_written(page, wbc, nr_written + 1); end = page_end; if (test_range_bit(tree, start, page_end, EXTENT_DELALLOC, 0)) @@ -2345,11 +2372,8 @@ done: unlock_extent(tree, unlock_start, page_end, GFP_NOFS); unlock_page(page); -update_nr_written: - wbc->nr_to_write -= nr_written; - if (wbc->range_cyclic || (wbc->nr_to_write > 0 && - wbc->range_start == 0 && wbc->range_end == LLONG_MAX)) - page->mapping->writeback_index = page->index + nr_written; +done_unlocked: + return 0; } -- cgit v1.2.3 From 8c594ea81d7abbbffdda447b127f8ba8d76f319d Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Mon, 20 Apr 2009 15:50:10 -0400 Subject: Btrfs: use the right node in reada_for_balance reada_for_balance was using the wrong index into the path node array, so it wasn't reading the right blocks. We never directly used the results of the read done by this function because the btree search is started over at the end. This fixes reada_for_balance to reada in the correct node and to avoid searching past the last slot in the node. It also makes sure to hold the parent lock while we are finding the nodes to read. Signed-off-by: Chris Mason --- fs/btrfs/ctree.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index e5b2533b691a..a99f1c2a710d 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -1325,12 +1325,12 @@ static noinline int reada_for_balance(struct btrfs_root *root, int ret = 0; int blocksize; - parent = path->nodes[level - 1]; + parent = path->nodes[level + 1]; if (!parent) return 0; nritems = btrfs_header_nritems(parent); - slot = path->slots[level]; + slot = path->slots[level + 1]; blocksize = btrfs_level_size(root, level); if (slot > 0) { @@ -1341,7 +1341,7 @@ static noinline int reada_for_balance(struct btrfs_root *root, block1 = 0; free_extent_buffer(eb); } - if (slot < nritems) { + if (slot + 1 < nritems) { block2 = btrfs_node_blockptr(parent, slot + 1); gen = btrfs_node_ptr_generation(parent, slot + 1); eb = btrfs_find_tree_block(root, block2, blocksize); @@ -1351,7 +1351,11 @@ static noinline int reada_for_balance(struct btrfs_root *root, } if (block1 || block2) { ret = -EAGAIN; + + /* release the whole path */ btrfs_release_path(root, path); + + /* read the blocks */ if (block1) readahead_tree_block(root, block1, blocksize, 0); if (block2) @@ -1361,7 +1365,7 @@ static noinline int reada_for_balance(struct btrfs_root *root, eb = read_tree_block(root, block1, blocksize, 0); free_extent_buffer(eb); } - if (block1) { + if (block2) { eb = read_tree_block(root, block2, blocksize, 0); free_extent_buffer(eb); } @@ -1481,12 +1485,15 @@ read_block_for_search(struct btrfs_trans_handle *trans, * of the btree by dropping locks before * we read. */ - btrfs_release_path(NULL, p); + btrfs_unlock_up_safe(p, level + 1); + btrfs_set_path_blocking(p); + if (tmp) free_extent_buffer(tmp); if (p->reada) reada_for_search(root, p, level, slot, key->objectid); + btrfs_release_path(NULL, p); tmp = read_tree_block(root, blocknr, blocksize, gen); if (tmp) free_extent_buffer(tmp); -- cgit v1.2.3 From 8c31813f31cd4403b46802866949a95a6e8fa584 Mon Sep 17 00:00:00 2001 From: Toshinobu Sugioka Date: Tue, 21 Apr 2009 07:34:53 +0900 Subject: sh: Fix mmap2 for handling differing PAGE_SIZEs. mmap2 uses a fixed page shift of 12, regardless of the PAGE_SIZE setting. Fix up the mmap2 code to add some sanity checks on the mapping, and to update pgoff accordingly. Error handling bits based on 4280e3126f641898f0ed1a931645373d3489e2a6 ("frv: fix mmap2 error handling"). Signed-off-by: Toshinobu Sugioka Signed-off-by: Paul Mundt --- arch/sh/kernel/sys_sh.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/arch/sh/kernel/sys_sh.c b/arch/sh/kernel/sys_sh.c index 58dfc02c7af1..e3a7e36639ef 100644 --- a/arch/sh/kernel/sys_sh.c +++ b/arch/sh/kernel/sys_sh.c @@ -63,6 +63,15 @@ asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff) { + /* + * The shift for mmap2 is constant, regardless of PAGE_SIZE + * setting. + */ + if (pgoff & ((1 << (PAGE_SHIFT - 12)) - 1)) + return -EINVAL; + + pgoff >>= PAGE_SHIFT - 12; + return do_mmap2(addr, len, prot, flags, fd, pgoff); } -- cgit v1.2.3 From cf2706a340ae98616d4e2a54900393e0e0b6b72c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 7 Apr 2009 09:03:30 -0400 Subject: Fix AUTOFS_DEV_IOCTL_REQUESTER_CMD Missing conversion from kernel to userland dev_t; this sucker breaks as soon as we get sufficiently many autofs mounts for new_encode_dev(s_dev) != s_dev. Note: this is the minimal fix. Signed-off-by: Al Viro --- fs/autofs4/dev-ioctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c index 9e5ae8a4f5c8..463f798f71b3 100644 --- a/fs/autofs4/dev-ioctl.c +++ b/fs/autofs4/dev-ioctl.c @@ -488,7 +488,7 @@ static int autofs_dev_ioctl_requester(struct file *fp, } path = param->path; - devid = sbi->sb->s_dev; + devid = new_encode_dev(sbi->sb->s_dev); param->requester.uid = param->requester.gid = -1; -- cgit v1.2.3 From e5d67f0715bc60f6c19abdd86d007d7bb29c9951 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 7 Apr 2009 12:15:39 -0400 Subject: Touch all affected namespaces on propagation of mount We shouldn't just touch the namespace of current process Caught-by: Trond Myklebust Signed-off-by: Al Viro --- fs/namespace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/namespace.c b/fs/namespace.c index d9138f81ec10..41196209a906 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1377,7 +1377,7 @@ static int attach_recursive_mnt(struct vfsmount *source_mnt, if (parent_path) { detach_mnt(source_mnt, parent_path); attach_mnt(source_mnt, path); - touch_mnt_namespace(current->nsproxy->mnt_ns); + touch_mnt_namespace(parent_path->mnt->mnt_ns); } else { mnt_set_mountpoint(dest_mnt, dest_dentry, source_mnt); commit_tree(source_mnt); -- cgit v1.2.3 From 1644ccc8a99ae73859c39372f96afdbf03c9f80d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 18 Apr 2009 02:32:31 -0400 Subject: Safer nfsd_cross_mnt() AFAICS, we have a subtle bug there: if we have crossed mountpoint *and* it got mount --move'd away, we'll be holding only one reference to fs containing dentry - exp->ex_path.mnt. IOW, we ought to dput() before exp_put(). Signed-off-by: Al Viro --- fs/nfsd/vfs.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index ab93fcfef254..46e6bd2d4f07 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -116,10 +116,15 @@ nfsd_cross_mnt(struct svc_rqst *rqstp, struct dentry **dpp, } if ((exp->ex_flags & NFSEXP_CROSSMOUNT) || EX_NOHIDE(exp2)) { /* successfully crossed mount point */ - exp_put(exp); - *expp = exp2; + /* + * This is subtle: dentry is *not* under mnt at this point. + * The only reason we are safe is that original mnt is pinned + * down by exp, so we should dput before putting exp. + */ dput(dentry); *dpp = mounts; + exp_put(exp); + *expp = exp2; } else { exp_put(exp2); dput(mounts); -- cgit v1.2.3 From 24b6f16ecf37f918a1934d590e9e71c100d6388f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 18 Apr 2009 03:25:41 -0400 Subject: No need for crossing to mountpoint in audit_tag_tree() is_under() will DTRT anyway. And yes, is_subdir() behaviour is intentional. Signed-off-by: Al Viro --- fs/dcache.c | 1 - kernel/audit_tree.c | 3 --- 2 files changed, 4 deletions(-) diff --git a/fs/dcache.c b/fs/dcache.c index 761d30be2683..1fcffebfb44f 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -2149,7 +2149,6 @@ int is_subdir(struct dentry *new_dentry, struct dentry *old_dentry) int result; unsigned long seq; - /* FIXME: This is old behavior, needed? Please check callers. */ if (new_dentry == old_dentry) return 1; diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c index 917ab9525568..6e7351739a82 100644 --- a/kernel/audit_tree.c +++ b/kernel/audit_tree.c @@ -734,9 +734,6 @@ int audit_tag_tree(char *old, char *new) dentry = dget(path.dentry); path_put(&path); - if (dentry == tagged->mnt_root && dentry == mnt->mnt_root) - follow_up(&mnt, &dentry); - list_add_tail(&list, &tagged->mnt_list); mutex_lock(&audit_filter_mutex); -- cgit v1.2.3 From 117aff744a20a2a04ccdb36cd5978316e1af0c3a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 18 Apr 2009 11:19:26 -0400 Subject: Fix autofs_expire() mnt should remain the same for all iterations through the list; as it is, if we have a busy mount, mnt follows into it and isn't restored for the next iteration. Signed-off-by: Al Viro --- fs/autofs/dirhash.c | 34 ++++++++++++++++++---------------- 1 file changed, 18 insertions(+), 16 deletions(-) diff --git a/fs/autofs/dirhash.c b/fs/autofs/dirhash.c index bf8c8af98004..4eb4d8dfb2f1 100644 --- a/fs/autofs/dirhash.c +++ b/fs/autofs/dirhash.c @@ -39,10 +39,12 @@ struct autofs_dir_ent *autofs_expire(struct super_block *sb, { struct autofs_dirhash *dh = &sbi->dirhash; struct autofs_dir_ent *ent; - struct dentry *dentry; unsigned long timeout = sbi->exp_timeout; while (1) { + struct path path; + int umount_ok; + if ( list_empty(&dh->expiry_head) || sbi->catatonic ) return NULL; /* No entries */ /* We keep the list sorted by last_usage and want old stuff */ @@ -57,17 +59,17 @@ struct autofs_dir_ent *autofs_expire(struct super_block *sb, return ent; /* Symlinks are always expirable */ /* Get the dentry for the autofs subdirectory */ - dentry = ent->dentry; + path.dentry = ent->dentry; - if ( !dentry ) { + if (!path.dentry) { /* Should only happen in catatonic mode */ printk("autofs: dentry == NULL but inode range is directory, entry %s\n", ent->name); autofs_delete_usage(ent); continue; } - if ( !dentry->d_inode ) { - dput(dentry); + if (!path.dentry->d_inode) { + dput(path.dentry); printk("autofs: negative dentry on expiry queue: %s\n", ent->name); autofs_delete_usage(ent); @@ -76,29 +78,29 @@ struct autofs_dir_ent *autofs_expire(struct super_block *sb, /* Make sure entry is mounted and unused; note that dentry will point to the mounted-on-top root. */ - if (!S_ISDIR(dentry->d_inode->i_mode)||!d_mountpoint(dentry)) { + if (!S_ISDIR(path.dentry->d_inode->i_mode) || + !d_mountpoint(path.dentry)) { DPRINTK(("autofs: not expirable (not a mounted directory): %s\n", ent->name)); continue; } - mntget(mnt); - dget(dentry); - if (!follow_down(&mnt, &dentry)) { - dput(dentry); - mntput(mnt); + path.mnt = mnt; + path_get(&path); + if (!follow_down(&path.mnt, &path.dentry)) { + path_put(&path); DPRINTK(("autofs: not expirable (not a mounted directory): %s\n", ent->name)); continue; } - while (d_mountpoint(dentry) && follow_down(&mnt, &dentry)) + while (d_mountpoint(path.dentry) && + follow_down(&path.mnt, &path.dentry)) ; - dput(dentry); + umount_ok = may_umount(path.mnt); + path_put(&path); - if ( may_umount(mnt) ) { - mntput(mnt); + if (umount_ok) { DPRINTK(("autofs: signaling expire on %s\n", ent->name)); return ent; /* Expirable! */ } DPRINTK(("autofs: didn't expire due to may_umount: %s\n", ent->name)); - mntput(mnt); } return NULL; /* No expirable entries */ } -- cgit v1.2.3 From 1ba0c7dbbbc24230394100c5f0d0df38cb400cff Mon Sep 17 00:00:00 2001 From: Alexander Beregalov Date: Mon, 20 Apr 2009 12:23:02 +0400 Subject: fs/compat_ioctl: fix build when !BLOCK In file included from fs/compat_ioctl.c:61: include/linux/loop.h:59: error: field 'lo_bio_list' has incomplete type Signed-off-by: Alexander Beregalov Signed-off-by: Al Viro --- fs/compat_ioctl.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 3e87ce443ea2..b83f6bcfa51a 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -58,7 +58,6 @@ #include #include #include -#include #include #include @@ -68,6 +67,7 @@ #include #ifdef CONFIG_BLOCK +#include #include #include #include @@ -2660,6 +2660,8 @@ HANDLE_IOCTL(SONET_GETFRAMING, do_atm_ioctl) HANDLE_IOCTL(SONET_GETFRSENSE, do_atm_ioctl) /* block stuff */ #ifdef CONFIG_BLOCK +/* loop */ +IGNORE_IOCTL(LOOP_CLR_FD) /* Raw devices */ HANDLE_IOCTL(RAW_SETBIND, raw_ioctl) HANDLE_IOCTL(RAW_GETBIND, raw_ioctl) @@ -2728,9 +2730,6 @@ HANDLE_IOCTL(LPSETTIMEOUT, lp_timeout_trans) IGNORE_IOCTL(VFAT_IOCTL_READDIR_BOTH32) IGNORE_IOCTL(VFAT_IOCTL_READDIR_SHORT32) -/* loop */ -IGNORE_IOCTL(LOOP_CLR_FD) - #ifdef CONFIG_SPARC /* Sparc framebuffers, handled in sbusfb_compat_ioctl() */ IGNORE_IOCTL(FBIOGTYPE) -- cgit v1.2.3 From 2f9092e1020246168b1309b35e085ecd7ff9ff72 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Mon, 20 Apr 2009 23:18:37 +0100 Subject: Fix i_mutex vs. readdir handling in nfsd Commit 14f7dd63 ("Copy XFS readdir hack into nfsd code") introduced a bug to generic code which had been extant for a long time in the XFS version -- it started to call through into lookup_one_len() and hence into the file systems' ->lookup() methods without i_mutex held on the directory. This patch fixes it by locking the directory's i_mutex again before calling the filldir functions. The original deadlocks which commit 14f7dd63 was designed to avoid are still avoided, because they were due to fs-internal locking, not i_mutex. While we're at it, fix the return type of nfsd_buffered_readdir() which should be a __be32 not an int -- it's an NFS errno, not a Linux errno. And return nfserrno(-ENOMEM) when allocation fails, not just -ENOMEM. Sparse would have caught that, if it wasn't so busy bitching about __cold__. Commit 05f4f678 ("nfsd4: don't do lookup within readdir in recovery code") introduced a similar problem with calling lookup_one_len() without i_mutex, which this patch also addresses. To fix that, it was necessary to fix the called functions so that they expect i_mutex to be held; that part was done by J. Bruce Fields. Signed-off-by: David Woodhouse Umm-I-can-live-with-that-by: Al Viro Reported-by: J. R. Okajima Tested-by: J. Bruce Fields LKML-Reference: <8036.1237474444@jrobl> Cc: stable@kernel.org Signed-off-by: Al Viro --- fs/namei.c | 2 ++ fs/nfsd/nfs4recover.c | 46 +++++++++------------------------------------- fs/nfsd/vfs.c | 25 +++++++++++++++++++------ 3 files changed, 30 insertions(+), 43 deletions(-) diff --git a/fs/namei.c b/fs/namei.c index b8433ebfae05..78f253cd2d4f 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1248,6 +1248,8 @@ struct dentry *lookup_one_len(const char *name, struct dentry *base, int len) int err; struct qstr this; + WARN_ON_ONCE(!mutex_is_locked(&base->d_inode->i_mutex)); + err = __lookup_one_len(name, &this, base, len); if (err) return ERR_PTR(err); diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 3444c0052a87..5275097a7565 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -229,21 +229,23 @@ nfsd4_list_rec_dir(struct dentry *dir, recdir_func *f) goto out; status = vfs_readdir(filp, nfsd4_build_namelist, &names); fput(filp); + mutex_lock(&dir->d_inode->i_mutex); while (!list_empty(&names)) { entry = list_entry(names.next, struct name_list, list); dentry = lookup_one_len(entry->name, dir, HEXDIR_LEN-1); if (IS_ERR(dentry)) { status = PTR_ERR(dentry); - goto out; + break; } status = f(dir, dentry); dput(dentry); if (status) - goto out; + break; list_del(&entry->list); kfree(entry); } + mutex_unlock(&dir->d_inode->i_mutex); out: while (!list_empty(&names)) { entry = list_entry(names.next, struct name_list, list); @@ -254,36 +256,6 @@ out: return status; } -static int -nfsd4_remove_clid_file(struct dentry *dir, struct dentry *dentry) -{ - int status; - - if (!S_ISREG(dir->d_inode->i_mode)) { - printk("nfsd4: non-file found in client recovery directory\n"); - return -EINVAL; - } - mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_PARENT); - status = vfs_unlink(dir->d_inode, dentry); - mutex_unlock(&dir->d_inode->i_mutex); - return status; -} - -static int -nfsd4_clear_clid_dir(struct dentry *dir, struct dentry *dentry) -{ - int status; - - /* For now this directory should already be empty, but we empty it of - * any regular files anyway, just in case the directory was created by - * a kernel from the future.... */ - nfsd4_list_rec_dir(dentry, nfsd4_remove_clid_file); - mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_PARENT); - status = vfs_rmdir(dir->d_inode, dentry); - mutex_unlock(&dir->d_inode->i_mutex); - return status; -} - static int nfsd4_unlink_clid_dir(char *name, int namlen) { @@ -294,18 +266,18 @@ nfsd4_unlink_clid_dir(char *name, int namlen) mutex_lock(&rec_dir.dentry->d_inode->i_mutex); dentry = lookup_one_len(name, rec_dir.dentry, namlen); - mutex_unlock(&rec_dir.dentry->d_inode->i_mutex); if (IS_ERR(dentry)) { status = PTR_ERR(dentry); - return status; + goto out_unlock; } status = -ENOENT; if (!dentry->d_inode) goto out; - - status = nfsd4_clear_clid_dir(rec_dir.dentry, dentry); + status = vfs_rmdir(rec_dir.dentry->d_inode, dentry); out: dput(dentry); +out_unlock: + mutex_unlock(&rec_dir.dentry->d_inode->i_mutex); return status; } @@ -348,7 +320,7 @@ purge_old(struct dentry *parent, struct dentry *child) if (nfs4_has_reclaimed_state(child->d_name.name, false)) return 0; - status = nfsd4_clear_clid_dir(parent, child); + status = vfs_rmdir(parent->d_inode, child); if (status) printk("failed to remove client recovery directory %s\n", child->d_name.name); diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 46e6bd2d4f07..6c68ffd6b4bb 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1890,8 +1890,8 @@ static int nfsd_buffered_filldir(void *__buf, const char *name, int namlen, return 0; } -static int nfsd_buffered_readdir(struct file *file, filldir_t func, - struct readdir_cd *cdp, loff_t *offsetp) +static __be32 nfsd_buffered_readdir(struct file *file, filldir_t func, + struct readdir_cd *cdp, loff_t *offsetp) { struct readdir_data buf; struct buffered_dirent *de; @@ -1901,11 +1901,12 @@ static int nfsd_buffered_readdir(struct file *file, filldir_t func, buf.dirent = (void *)__get_free_page(GFP_KERNEL); if (!buf.dirent) - return -ENOMEM; + return nfserrno(-ENOMEM); offset = *offsetp; while (1) { + struct inode *dir_inode = file->f_path.dentry->d_inode; unsigned int reclen; cdp->err = nfserr_eof; /* will be cleared on successful read */ @@ -1924,26 +1925,38 @@ static int nfsd_buffered_readdir(struct file *file, filldir_t func, if (!size) break; + /* + * Various filldir functions may end up calling back into + * lookup_one_len() and the file system's ->lookup() method. + * These expect i_mutex to be held, as it would within readdir. + */ + host_err = mutex_lock_killable(&dir_inode->i_mutex); + if (host_err) + break; + de = (struct buffered_dirent *)buf.dirent; while (size > 0) { offset = de->offset; if (func(cdp, de->name, de->namlen, de->offset, de->ino, de->d_type)) - goto done; + break; if (cdp->err != nfs_ok) - goto done; + break; reclen = ALIGN(sizeof(*de) + de->namlen, sizeof(u64)); size -= reclen; de = (struct buffered_dirent *)((char *)de + reclen); } + mutex_unlock(&dir_inode->i_mutex); + if (size > 0) /* We bailed out early */ + break; + offset = vfs_llseek(file, 0, SEEK_CUR); } - done: free_page((unsigned long)(buf.dirent)); if (host_err) -- cgit v1.2.3 From 66672fefaa91802fec51c3fe0cc55bc9baea5a2d Mon Sep 17 00:00:00 2001 From: Adrian McMenamin Date: Mon, 20 Apr 2009 18:38:28 -0700 Subject: Documentation/filesystems: remove out of date reference to BKL being held Documentation/filesystems/vfs.txt incorrectly states that the kernel is locked during the call to statfs (Documentation/filesystems/Locking correctly says it is not). This patch removes the offending sentence. remove reference to BKL being held in statfs Signed-off-by: Adrian McMenamin Signed-off-by: Randy Dunlap Cc: Alexander Viro Signed-off-by: Al Viro --- Documentation/filesystems/vfs.txt | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt index deeeed0faa8f..f49eecf2e573 100644 --- a/Documentation/filesystems/vfs.txt +++ b/Documentation/filesystems/vfs.txt @@ -277,8 +277,7 @@ or bottom half). unfreeze_fs: called when VFS is unlocking a filesystem and making it writable again. - statfs: called when the VFS needs to get filesystem statistics. This - is called with the kernel lock held + statfs: called when the VFS needs to get filesystem statistics. remount_fs: called when the filesystem is remounted. This is called with the kernel lock held -- cgit v1.2.3 From 3eac8778a237d83a1e553eba0c6f4fd4b39eeec0 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 7 Apr 2009 11:12:46 -0400 Subject: autofs4: use memchr() in invalid_string() Signed-off-by: Al Viro --- fs/autofs4/dev-ioctl.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c index 463f798f71b3..84168c0dcc2d 100644 --- a/fs/autofs4/dev-ioctl.c +++ b/fs/autofs4/dev-ioctl.c @@ -54,11 +54,10 @@ static int check_name(const char *name) * Check a string doesn't overrun the chunk of * memory we copied from user land. */ -static int invalid_str(char *str, void *end) +static int invalid_str(char *str, size_t size) { - while ((void *) str <= end) - if (!*str++) - return 0; + if (memchr(str, 0, size)) + return 0; return -EINVAL; } @@ -138,8 +137,7 @@ static int validate_dev_ioctl(int cmd, struct autofs_dev_ioctl *param) } if (param->size > sizeof(*param)) { - err = invalid_str(param->path, - (void *) ((size_t) param + param->size)); + err = invalid_str(param->path, param->size - sizeof(*param)); if (err) { AUTOFS_WARN( "path string terminator missing for cmd(0x%08x)", -- cgit v1.2.3 From 3939fcde24473dc09ce16e922c88df9b3bee45d9 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Wed, 8 Apr 2009 15:06:12 +0800 Subject: xattr: use memdup_user() Remove open-coded memdup_user() Signed-off-by: Li Zefan Signed-off-by: Al Viro --- fs/xattr.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) diff --git a/fs/xattr.c b/fs/xattr.c index 197c4fcac032..d51b8f9db921 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -237,13 +237,9 @@ setxattr(struct dentry *d, const char __user *name, const void __user *value, if (size) { if (size > XATTR_SIZE_MAX) return -E2BIG; - kvalue = kmalloc(size, GFP_KERNEL); - if (!kvalue) - return -ENOMEM; - if (copy_from_user(kvalue, value, size)) { - kfree(kvalue); - return -EFAULT; - } + kvalue = memdup_user(value, size); + if (IS_ERR(kvalue)) + return PTR_ERR(kvalue); } error = vfs_setxattr(d, kname, kvalue, size, flags); -- cgit v1.2.3 From dae7b665cf6d6e6e733f1c9c16cf55547dd37e33 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Wed, 8 Apr 2009 15:06:54 +0800 Subject: btrfs: use memdup_user() Remove open-coded memdup_user(). Note this changes some GFP_NOFS to GFP_KERNEL, since copy_from_user() may cause pagefault, it's pointless to pass GFP_NOFS to kmalloc(). Signed-off-by: Li Zefan Signed-off-by: Al Viro --- fs/btrfs/ioctl.c | 49 ++++++++++++------------------------------------- fs/btrfs/super.c | 13 ++++--------- 2 files changed, 16 insertions(+), 46 deletions(-) diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 7594bec1be10..9f135e878507 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -461,15 +461,9 @@ static int btrfs_ioctl_resize(struct btrfs_root *root, void __user *arg) if (!capable(CAP_SYS_ADMIN)) return -EPERM; - vol_args = kmalloc(sizeof(*vol_args), GFP_NOFS); - - if (!vol_args) - return -ENOMEM; - - if (copy_from_user(vol_args, arg, sizeof(*vol_args))) { - ret = -EFAULT; - goto out; - } + vol_args = memdup_user(arg, sizeof(*vol_args)); + if (IS_ERR(vol_args)) + return PTR_ERR(vol_args); vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; namelen = strlen(vol_args->name); @@ -545,7 +539,6 @@ static int btrfs_ioctl_resize(struct btrfs_root *root, void __user *arg) out_unlock: mutex_unlock(&root->fs_info->volume_mutex); -out: kfree(vol_args); return ret; } @@ -565,15 +558,9 @@ static noinline int btrfs_ioctl_snap_create(struct file *file, if (root->fs_info->sb->s_flags & MS_RDONLY) return -EROFS; - vol_args = kmalloc(sizeof(*vol_args), GFP_NOFS); - - if (!vol_args) - return -ENOMEM; - - if (copy_from_user(vol_args, arg, sizeof(*vol_args))) { - ret = -EFAULT; - goto out; - } + vol_args = memdup_user(arg, sizeof(*vol_args)); + if (IS_ERR(vol_args)) + return PTR_ERR(vol_args); vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; namelen = strlen(vol_args->name); @@ -675,19 +662,13 @@ static long btrfs_ioctl_add_dev(struct btrfs_root *root, void __user *arg) if (!capable(CAP_SYS_ADMIN)) return -EPERM; - vol_args = kmalloc(sizeof(*vol_args), GFP_NOFS); + vol_args = memdup_user(arg, sizeof(*vol_args)); + if (IS_ERR(vol_args)) + return PTR_ERR(vol_args); - if (!vol_args) - return -ENOMEM; - - if (copy_from_user(vol_args, arg, sizeof(*vol_args))) { - ret = -EFAULT; - goto out; - } vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; ret = btrfs_init_new_device(root, vol_args->name); -out: kfree(vol_args); return ret; } @@ -703,19 +684,13 @@ static long btrfs_ioctl_rm_dev(struct btrfs_root *root, void __user *arg) if (root->fs_info->sb->s_flags & MS_RDONLY) return -EROFS; - vol_args = kmalloc(sizeof(*vol_args), GFP_NOFS); + vol_args = memdup_user(arg, sizeof(*vol_args)); + if (IS_ERR(vol_args)) + return PTR_ERR(vol_args); - if (!vol_args) - return -ENOMEM; - - if (copy_from_user(vol_args, arg, sizeof(*vol_args))) { - ret = -EFAULT; - goto out; - } vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; ret = btrfs_rm_device(root, vol_args->name); -out: kfree(vol_args); return ret; } diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 9744af9d71e9..a7acfe639a44 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -635,14 +635,9 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd, if (!capable(CAP_SYS_ADMIN)) return -EPERM; - vol = kmalloc(sizeof(*vol), GFP_KERNEL); - if (!vol) - return -ENOMEM; - - if (copy_from_user(vol, (void __user *)arg, sizeof(*vol))) { - ret = -EFAULT; - goto out; - } + vol = memdup_user((void __user *)arg, sizeof(*vol)); + if (IS_ERR(vol)) + return PTR_ERR(vol); switch (cmd) { case BTRFS_IOC_SCAN_DEV: @@ -650,7 +645,7 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd, &btrfs_fs_type, &fs_devices); break; } -out: + kfree(vol); return ret; } -- cgit v1.2.3 From 1c8542c7bb239ef02fe21477acd9cdac04c1b640 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Wed, 8 Apr 2009 15:07:30 +0800 Subject: sysfs: use memdup_user() Remove open-coded memdup_user(). Signed-off-by: Li Zefan Signed-off-by: Al Viro --- fs/sysfs/bin.c | 13 +++---------- 1 file changed, 3 insertions(+), 10 deletions(-) diff --git a/fs/sysfs/bin.c b/fs/sysfs/bin.c index 93e0c0281d45..9345806c8853 100644 --- a/fs/sysfs/bin.c +++ b/fs/sysfs/bin.c @@ -157,14 +157,9 @@ static ssize_t write(struct file *file, const char __user *userbuf, count = size - offs; } - temp = kmalloc(count, GFP_KERNEL); - if (!temp) - return -ENOMEM; - - if (copy_from_user(temp, userbuf, count)) { - count = -EFAULT; - goto out_free; - } + temp = memdup_user(userbuf, count); + if (IS_ERR(temp)) + return PTR_ERR(temp); mutex_lock(&bb->mutex); @@ -176,8 +171,6 @@ static ssize_t write(struct file *file, const char __user *userbuf, if (count > 0) *off = offs + count; -out_free: - kfree(temp); return count; } -- cgit v1.2.3 From 0e639bdeef26faf287db77a15530f3f295a4ae04 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Wed, 8 Apr 2009 15:08:04 +0800 Subject: xfs: use memdup_user() Remove open-coded memdup_user() Signed-off-by: Li Zefan Signed-off-by: Al Viro --- fs/xfs/linux-2.6/xfs_ioctl.c | 23 +++++++---------------- fs/xfs/linux-2.6/xfs_ioctl32.c | 12 ++++-------- 2 files changed, 11 insertions(+), 24 deletions(-) diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index d0b499418a7d..34eaab608e6e 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -489,17 +489,12 @@ xfs_attrmulti_attr_set( if (len > XATTR_SIZE_MAX) return EINVAL; - kbuf = kmalloc(len, GFP_KERNEL); - if (!kbuf) - return ENOMEM; - - if (copy_from_user(kbuf, ubuf, len)) - goto out_kfree; + kbuf = memdup_user(ubuf, len); + if (IS_ERR(kbuf)) + return PTR_ERR(kbuf); error = xfs_attr_set(XFS_I(inode), name, kbuf, len, flags); - out_kfree: - kfree(kbuf); return error; } @@ -540,20 +535,16 @@ xfs_attrmulti_by_handle( if (!size || size > 16 * PAGE_SIZE) goto out_dput; - error = ENOMEM; - ops = kmalloc(size, GFP_KERNEL); - if (!ops) + ops = memdup_user(am_hreq.ops, size); + if (IS_ERR(ops)) { + error = PTR_ERR(ops); goto out_dput; - - error = EFAULT; - if (copy_from_user(ops, am_hreq.ops, size)) - goto out_kfree_ops; + } attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL); if (!attr_name) goto out_kfree_ops; - error = 0; for (i = 0; i < am_hreq.opcount; i++) { ops[i].am_error = strncpy_from_user(attr_name, diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.c b/fs/xfs/linux-2.6/xfs_ioctl32.c index c70c4e3db790..0882d166239a 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl32.c +++ b/fs/xfs/linux-2.6/xfs_ioctl32.c @@ -427,20 +427,16 @@ xfs_compat_attrmulti_by_handle( if (!size || size > 16 * PAGE_SIZE) goto out_dput; - error = ENOMEM; - ops = kmalloc(size, GFP_KERNEL); - if (!ops) + ops = memdup_user(compat_ptr(am_hreq.ops), size); + if (IS_ERR(ops)) { + error = PTR_ERR(ops); goto out_dput; - - error = EFAULT; - if (copy_from_user(ops, compat_ptr(am_hreq.ops), size)) - goto out_kfree_ops; + } attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL); if (!attr_name) goto out_kfree_ops; - error = 0; for (i = 0; i < am_hreq.opcount; i++) { ops[i].am_error = strncpy_from_user(attr_name, -- cgit v1.2.3 From a9482ebcdedbc5872ed34a266e6a45c35116f264 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Wed, 8 Apr 2009 15:08:53 +0800 Subject: ncpfs: use memdup_user() Remove open-coded memdup_user() Signed-off-by: Li Zefan Signed-off-by: Al Viro --- fs/ncpfs/ioctl.c | 21 +++++++-------------- 1 file changed, 7 insertions(+), 14 deletions(-) diff --git a/fs/ncpfs/ioctl.c b/fs/ncpfs/ioctl.c index f54360f50a9c..fa038df63ac8 100644 --- a/fs/ncpfs/ioctl.c +++ b/fs/ncpfs/ioctl.c @@ -660,13 +660,10 @@ outrel: if (user.object_name_len > NCP_OBJECT_NAME_MAX_LEN) return -ENOMEM; if (user.object_name_len) { - newname = kmalloc(user.object_name_len, GFP_USER); - if (!newname) - return -ENOMEM; - if (copy_from_user(newname, user.object_name, user.object_name_len)) { - kfree(newname); - return -EFAULT; - } + newname = memdup_user(user.object_name, + user.object_name_len); + if (IS_ERR(newname)) + return PTR_ERR(newname); } else { newname = NULL; } @@ -760,13 +757,9 @@ outrel: if (user.len > NCP_PRIVATE_DATA_MAX_LEN) return -ENOMEM; if (user.len) { - new = kmalloc(user.len, GFP_USER); - if (!new) - return -ENOMEM; - if (copy_from_user(new, user.data, user.len)) { - kfree(new); - return -EFAULT; - } + new = memdup_user(user.data, user.len); + if (IS_ERR(new)) + return PTR_ERR(new); } else { new = NULL; } -- cgit v1.2.3 From fd56d242b3b80b6f2ca174272b20029aae61df75 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Wed, 8 Apr 2009 15:09:29 +0800 Subject: ecryptfs: use memdup_user() Remove open-coded memdup_user(). Signed-off-by: Li Zefan Signed-off-by: Al Viro --- fs/ecryptfs/miscdev.c | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-) diff --git a/fs/ecryptfs/miscdev.c b/fs/ecryptfs/miscdev.c index a67fea655f49..dda3c58eefc0 100644 --- a/fs/ecryptfs/miscdev.c +++ b/fs/ecryptfs/miscdev.c @@ -418,18 +418,13 @@ ecryptfs_miscdev_write(struct file *file, const char __user *buf, if (count == 0) goto out; - data = kmalloc(count, GFP_KERNEL); - if (!data) { - printk(KERN_ERR "%s: Out of memory whilst attempting to " - "kmalloc([%zd], GFP_KERNEL)\n", __func__, count); + + data = memdup_user(buf, count); + if (IS_ERR(data)) { + printk(KERN_ERR "%s: memdup_user returned error [%ld]\n", + __func__, PTR_ERR(data)); goto out; } - rc = copy_from_user(data, buf, count); - if (rc) { - printk(KERN_ERR "%s: copy_from_user returned error [%d]\n", - __func__, rc); - goto out_free; - } sz = count; i = 0; switch (data[i++]) { -- cgit v1.2.3 From 0112fc2229847feb6c4eb011e6833d8f1742a375 Mon Sep 17 00:00:00 2001 From: Oleg Drokin Date: Wed, 8 Apr 2009 20:05:42 +0400 Subject: Separate out common fstatat code into vfs_fstatat This is a version incorporating Christoph's suggestion. Separate out common *fstatat functionality into a single function instead of duplicating it all over the code. Signed-off-by: Oleg Drokin Signed-off-by: Al Viro --- arch/arm/kernel/sys_oabi-compat.c | 19 ++++--------- arch/s390/kernel/compat_linux.c | 18 ++++--------- arch/sparc/kernel/sys_sparc32.c | 19 ++++--------- arch/x86/ia32/sys_ia32.c | 19 ++++--------- fs/compat.c | 19 ++++--------- fs/stat.c | 56 +++++++++++++++++++-------------------- include/linux/fs.h | 1 + 7 files changed, 54 insertions(+), 97 deletions(-) diff --git a/arch/arm/kernel/sys_oabi-compat.c b/arch/arm/kernel/sys_oabi-compat.c index e04173c7e621..d59a0cd537f0 100644 --- a/arch/arm/kernel/sys_oabi-compat.c +++ b/arch/arm/kernel/sys_oabi-compat.c @@ -177,21 +177,12 @@ asmlinkage long sys_oabi_fstatat64(int dfd, int flag) { struct kstat stat; - int error = -EINVAL; + int error; - if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) - goto out; - - if (flag & AT_SYMLINK_NOFOLLOW) - error = vfs_lstat_fd(dfd, filename, &stat); - else - error = vfs_stat_fd(dfd, filename, &stat); - - if (!error) - error = cp_oldabi_stat64(&stat, statbuf); - -out: - return error; + error = vfs_fstatat(dfd, filename, &stat, flag); + if (error) + return error; + return cp_oldabi_stat64(&stat, statbuf); } struct oabi_flock64 { diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c index 6cc87d8c8682..002c70d3cb75 100644 --- a/arch/s390/kernel/compat_linux.c +++ b/arch/s390/kernel/compat_linux.c @@ -702,20 +702,12 @@ asmlinkage long sys32_fstatat64(unsigned int dfd, char __user *filename, struct stat64_emu31 __user* statbuf, int flag) { struct kstat stat; - int error = -EINVAL; - - if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) - goto out; - - if (flag & AT_SYMLINK_NOFOLLOW) - error = vfs_lstat_fd(dfd, filename, &stat); - else - error = vfs_stat_fd(dfd, filename, &stat); + int error; - if (!error) - error = cp_stat64(statbuf, &stat); -out: - return error; + error = vfs_fstatat(dfd, filename, &stat, flag); + if (error) + return error; + return cp_stat64(statbuf, &stat); } /* diff --git a/arch/sparc/kernel/sys_sparc32.c b/arch/sparc/kernel/sys_sparc32.c index e800503879e4..f5000a460c05 100644 --- a/arch/sparc/kernel/sys_sparc32.c +++ b/arch/sparc/kernel/sys_sparc32.c @@ -206,21 +206,12 @@ asmlinkage long compat_sys_fstatat64(unsigned int dfd, char __user *filename, struct compat_stat64 __user * statbuf, int flag) { struct kstat stat; - int error = -EINVAL; - - if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) - goto out; - - if (flag & AT_SYMLINK_NOFOLLOW) - error = vfs_lstat_fd(dfd, filename, &stat); - else - error = vfs_stat_fd(dfd, filename, &stat); - - if (!error) - error = cp_compat_stat64(&stat, statbuf); + int error; -out: - return error; + error = vfs_fstatat(dfd, filename, &stat, flag); + if (error) + return error; + return cp_compat_stat64(&stat, statbuf); } asmlinkage long compat_sys_sysfs(int option, u32 arg1, u32 arg2) diff --git a/arch/x86/ia32/sys_ia32.c b/arch/x86/ia32/sys_ia32.c index efac92fd1efb..085a8c35f149 100644 --- a/arch/x86/ia32/sys_ia32.c +++ b/arch/x86/ia32/sys_ia32.c @@ -129,21 +129,12 @@ asmlinkage long sys32_fstatat(unsigned int dfd, char __user *filename, struct stat64 __user *statbuf, int flag) { struct kstat stat; - int error = -EINVAL; + int error; - if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) - goto out; - - if (flag & AT_SYMLINK_NOFOLLOW) - error = vfs_lstat_fd(dfd, filename, &stat); - else - error = vfs_stat_fd(dfd, filename, &stat); - - if (!error) - error = cp_stat64(statbuf, &stat); - -out: - return error; + error = vfs_fstatat(dfd, filename, &stat, flag); + if (error) + return error; + return cp_stat64(statbuf, &stat); } /* diff --git a/fs/compat.c b/fs/compat.c index 3f84d5f15889..dda72e267092 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -204,21 +204,12 @@ asmlinkage long compat_sys_newfstatat(unsigned int dfd, char __user *filename, struct compat_stat __user *statbuf, int flag) { struct kstat stat; - int error = -EINVAL; - - if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) - goto out; - - if (flag & AT_SYMLINK_NOFOLLOW) - error = vfs_lstat_fd(dfd, filename, &stat); - else - error = vfs_stat_fd(dfd, filename, &stat); - - if (!error) - error = cp_compat_stat(&stat, statbuf); + int error; -out: - return error; + error = vfs_fstatat(dfd, filename, &stat, flag); + if (error) + return error; + return cp_compat_stat(&stat, statbuf); } #endif diff --git a/fs/stat.c b/fs/stat.c index 2db740a0cfb5..54711662b855 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -109,6 +109,24 @@ int vfs_fstat(unsigned int fd, struct kstat *stat) EXPORT_SYMBOL(vfs_fstat); +int vfs_fstatat(int dfd, char __user *filename, struct kstat *stat, int flag) +{ + int error = -EINVAL; + + if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) + goto out; + + if (flag & AT_SYMLINK_NOFOLLOW) + error = vfs_lstat_fd(dfd, filename, stat); + else + error = vfs_stat_fd(dfd, filename, stat); +out: + return error; +} + +EXPORT_SYMBOL(vfs_fstatat); + + #ifdef __ARCH_WANT_OLD_STAT /* @@ -264,21 +282,12 @@ SYSCALL_DEFINE4(newfstatat, int, dfd, char __user *, filename, struct stat __user *, statbuf, int, flag) { struct kstat stat; - int error = -EINVAL; - - if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) - goto out; - - if (flag & AT_SYMLINK_NOFOLLOW) - error = vfs_lstat_fd(dfd, filename, &stat); - else - error = vfs_stat_fd(dfd, filename, &stat); - - if (!error) - error = cp_new_stat(&stat, statbuf); + int error; -out: - return error; + error = vfs_fstatat(dfd, filename, &stat, flag); + if (error) + return error; + return cp_new_stat(&stat, statbuf); } #endif @@ -404,21 +413,12 @@ SYSCALL_DEFINE4(fstatat64, int, dfd, char __user *, filename, struct stat64 __user *, statbuf, int, flag) { struct kstat stat; - int error = -EINVAL; - - if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) - goto out; - - if (flag & AT_SYMLINK_NOFOLLOW) - error = vfs_lstat_fd(dfd, filename, &stat); - else - error = vfs_stat_fd(dfd, filename, &stat); - - if (!error) - error = cp_new_stat64(&stat, statbuf); + int error; -out: - return error; + error = vfs_fstatat(dfd, filename, &stat, flag); + if (error) + return error; + return cp_new_stat64(&stat, statbuf); } #endif /* __ARCH_WANT_STAT64 */ diff --git a/include/linux/fs.h b/include/linux/fs.h index e766be0d4329..257f4d37ad23 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2302,6 +2302,7 @@ extern int vfs_lstat(char __user *, struct kstat *); extern int vfs_stat_fd(int dfd, char __user *, struct kstat *); extern int vfs_lstat_fd(int dfd, char __user *, struct kstat *); extern int vfs_fstat(unsigned int, struct kstat *); +extern int vfs_fstatat(int , char __user *, struct kstat *, int); extern int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd, unsigned long arg); -- cgit v1.2.3 From 2eae7a1874ca5be3232765d89e0250a449f1bc90 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 8 Apr 2009 16:34:03 -0400 Subject: kill vfs_stat_fd / vfs_lstat_fd There's really no reason to keep vfs_stat_fd and vfs_lstat_fd with Oleg's vfs_fstatat. Use vfs_fstatat for the few cases having the directory fd, and switch all others to vfs_stat / vfs_lstat. Reviewed-by: Christoph Hellwig Signed-off-by: Al Viro --- fs/compat.c | 18 +++++---- fs/stat.c | 105 +++++++++++++++++++++-------------------------------- include/linux/fs.h | 2 - 3 files changed, 52 insertions(+), 73 deletions(-) diff --git a/fs/compat.c b/fs/compat.c index dda72e267092..379a399bf5c3 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -181,22 +181,24 @@ asmlinkage long compat_sys_newstat(char __user * filename, struct compat_stat __user *statbuf) { struct kstat stat; - int error = vfs_stat_fd(AT_FDCWD, filename, &stat); + int error; - if (!error) - error = cp_compat_stat(&stat, statbuf); - return error; + error = vfs_stat(filename, &stat); + if (error) + return error; + return cp_compat_stat(&stat, statbuf); } asmlinkage long compat_sys_newlstat(char __user * filename, struct compat_stat __user *statbuf) { struct kstat stat; - int error = vfs_lstat_fd(AT_FDCWD, filename, &stat); + int error; - if (!error) - error = cp_compat_stat(&stat, statbuf); - return error; + error = vfs_lstat(filename, &stat); + if (error) + return error; + return cp_compat_stat(&stat, statbuf); } #ifndef __ARCH_WANT_STAT64 diff --git a/fs/stat.c b/fs/stat.c index 54711662b855..075694e31d8b 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -55,46 +55,6 @@ int vfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) EXPORT_SYMBOL(vfs_getattr); -int vfs_stat_fd(int dfd, char __user *name, struct kstat *stat) -{ - struct path path; - int error; - - error = user_path_at(dfd, name, LOOKUP_FOLLOW, &path); - if (!error) { - error = vfs_getattr(path.mnt, path.dentry, stat); - path_put(&path); - } - return error; -} - -int vfs_stat(char __user *name, struct kstat *stat) -{ - return vfs_stat_fd(AT_FDCWD, name, stat); -} - -EXPORT_SYMBOL(vfs_stat); - -int vfs_lstat_fd(int dfd, char __user *name, struct kstat *stat) -{ - struct path path; - int error; - - error = user_path_at(dfd, name, 0, &path); - if (!error) { - error = vfs_getattr(path.mnt, path.dentry, stat); - path_put(&path); - } - return error; -} - -int vfs_lstat(char __user *name, struct kstat *stat) -{ - return vfs_lstat_fd(AT_FDCWD, name, stat); -} - -EXPORT_SYMBOL(vfs_lstat); - int vfs_fstat(unsigned int fd, struct kstat *stat) { struct file *f = fget(fd); @@ -106,26 +66,43 @@ int vfs_fstat(unsigned int fd, struct kstat *stat) } return error; } - EXPORT_SYMBOL(vfs_fstat); int vfs_fstatat(int dfd, char __user *filename, struct kstat *stat, int flag) { + struct path path; int error = -EINVAL; + int lookup_flags = 0; if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) goto out; - if (flag & AT_SYMLINK_NOFOLLOW) - error = vfs_lstat_fd(dfd, filename, stat); - else - error = vfs_stat_fd(dfd, filename, stat); + if (!(flag & AT_SYMLINK_NOFOLLOW)) + lookup_flags |= LOOKUP_FOLLOW; + + error = user_path_at(dfd, filename, lookup_flags, &path); + if (error) + goto out; + + error = vfs_getattr(path.mnt, path.dentry, stat); + path_put(&path); out: return error; } - EXPORT_SYMBOL(vfs_fstatat); +int vfs_stat(char __user *name, struct kstat *stat) +{ + return vfs_fstatat(AT_FDCWD, name, stat, 0); +} +EXPORT_SYMBOL(vfs_stat); + +int vfs_lstat(char __user *name, struct kstat *stat) +{ + return vfs_fstatat(AT_FDCWD, name, stat, AT_SYMLINK_NOFOLLOW); +} +EXPORT_SYMBOL(vfs_lstat); + #ifdef __ARCH_WANT_OLD_STAT @@ -173,23 +150,25 @@ static int cp_old_stat(struct kstat *stat, struct __old_kernel_stat __user * sta SYSCALL_DEFINE2(stat, char __user *, filename, struct __old_kernel_stat __user *, statbuf) { struct kstat stat; - int error = vfs_stat_fd(AT_FDCWD, filename, &stat); + int error; - if (!error) - error = cp_old_stat(&stat, statbuf); + error = vfs_stat(filename, &stat); + if (error) + return error; - return error; + return cp_old_stat(&stat, statbuf); } SYSCALL_DEFINE2(lstat, char __user *, filename, struct __old_kernel_stat __user *, statbuf) { struct kstat stat; - int error = vfs_lstat_fd(AT_FDCWD, filename, &stat); + int error; - if (!error) - error = cp_old_stat(&stat, statbuf); + error = vfs_lstat(filename, &stat); + if (error) + return error; - return error; + return cp_old_stat(&stat, statbuf); } SYSCALL_DEFINE2(fstat, unsigned int, fd, struct __old_kernel_stat __user *, statbuf) @@ -258,23 +237,23 @@ static int cp_new_stat(struct kstat *stat, struct stat __user *statbuf) SYSCALL_DEFINE2(newstat, char __user *, filename, struct stat __user *, statbuf) { struct kstat stat; - int error = vfs_stat_fd(AT_FDCWD, filename, &stat); - - if (!error) - error = cp_new_stat(&stat, statbuf); + int error = vfs_stat(filename, &stat); - return error; + if (error) + return error; + return cp_new_stat(&stat, statbuf); } SYSCALL_DEFINE2(newlstat, char __user *, filename, struct stat __user *, statbuf) { struct kstat stat; - int error = vfs_lstat_fd(AT_FDCWD, filename, &stat); + int error; - if (!error) - error = cp_new_stat(&stat, statbuf); + error = vfs_lstat(filename, &stat); + if (error) + return error; - return error; + return cp_new_stat(&stat, statbuf); } #if !defined(__ARCH_WANT_STAT64) || defined(__ARCH_WANT_SYS_NEWFSTATAT) diff --git a/include/linux/fs.h b/include/linux/fs.h index 257f4d37ad23..8f42b35a7565 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2299,8 +2299,6 @@ extern int vfs_readdir(struct file *, filldir_t, void *); extern int vfs_stat(char __user *, struct kstat *); extern int vfs_lstat(char __user *, struct kstat *); -extern int vfs_stat_fd(int dfd, char __user *, struct kstat *); -extern int vfs_lstat_fd(int dfd, char __user *, struct kstat *); extern int vfs_fstat(unsigned int, struct kstat *); extern int vfs_fstatat(int , char __user *, struct kstat *, int); -- cgit v1.2.3 From 38e23c95f92a84fb8505a9f572b8a209c9c372c1 Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Thu, 9 Apr 2009 20:17:52 +0900 Subject: fs: Mark get_filesystem_list() as __init function. "int get_filesystem_list(char * buf)" is called by only "static void __init get_fs_names(char *page)". We can mark get_filesystem_list() as "__init". Signed-off-by: Tetsuo Handa Signed-off-by: Al Viro --- fs/filesystems.c | 2 +- include/linux/fs.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/filesystems.c b/fs/filesystems.c index 1aa70260e6d1..a24c58e181db 100644 --- a/fs/filesystems.c +++ b/fs/filesystems.c @@ -199,7 +199,7 @@ SYSCALL_DEFINE3(sysfs, int, option, unsigned long, arg1, unsigned long, arg2) return retval; } -int get_filesystem_list(char * buf) +int __init get_filesystem_list(char *buf) { int len = 0; struct file_system_type * tmp; diff --git a/include/linux/fs.h b/include/linux/fs.h index 8f42b35a7565..5bed436f4353 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2448,7 +2448,7 @@ struct ctl_table; int proc_nr_files(struct ctl_table *table, int write, struct file *filp, void __user *buffer, size_t *lenp, loff_t *ppos); -int get_filesystem_list(char * buf); +int __init get_filesystem_list(char *buf); #endif /* __KERNEL__ */ #endif /* _LINUX_FS_H */ -- cgit v1.2.3 From be9208dff23af904655807672dd8235abf6ac039 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Apr 2009 23:29:41 -0400 Subject: reiserfs: fix j_last_flush_trans_id type Conversion in commit 600ed41675d8c384519d8f0b3c76afed39ef2f4b had missed that one, but converted format from %lu to %u. As the result, /proc/..../journal got buggered on 64bit boxen. Signed-off-by: Al Viro --- include/linux/reiserfs_fs_sb.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h index 5621d87c4479..6b361d23a499 100644 --- a/include/linux/reiserfs_fs_sb.h +++ b/include/linux/reiserfs_fs_sb.h @@ -193,7 +193,7 @@ struct reiserfs_journal { atomic_t j_wcount; /* count of writers for current commit */ unsigned long j_bcount; /* batch count. allows turning X transactions into 1 */ unsigned long j_first_unflushed_offset; /* first unflushed transactions offset */ - unsigned long j_last_flush_trans_id; /* last fully flushed journal timestamp */ + unsigned j_last_flush_trans_id; /* last fully flushed journal timestamp */ struct buffer_head *j_header_bh; time_t j_trans_start_time; /* time this transaction started */ -- cgit v1.2.3 From 8340437210390676f687633a80e3748c40885dc8 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 20 Apr 2009 14:58:35 -0400 Subject: NFS: Fix the XDR iovec calculation in nfs3_xdr_setaclargs Commit ae46141ff08f1965b17c531b571953c39ce8b9e2 (NFSv3: Fix posix ACL code) introduces a bug in the calculation of the XDR header iovec. In the case where we are inlining the acls, we need to adjust the length of the iovec req->rq_svec, in addition to adjusting the total buffer length. Tested-by: Leonardo Chiquitto Tested-by: Suresh Jayaraman Signed-off-by: Trond Myklebust Cc: stable@kernel.org Signed-off-by: Linus Torvalds --- fs/nfs/nfs3xdr.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index e6a1932c7110..35869a4921f1 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -713,7 +713,8 @@ nfs3_xdr_setaclargs(struct rpc_rqst *req, __be32 *p, if (args->npages != 0) xdr_encode_pages(buf, args->pages, 0, args->len); else - req->rq_slen += args->len; + req->rq_slen = xdr_adjust_iovec(req->rq_svec, + p + XDR_QUADLEN(args->len)); err = nfsacl_encode(buf, base, args->inode, (args->mask & NFS_ACL) ? -- cgit v1.2.3 From 8b9cf76d0fa6cd98fe42dd2f86460d6ede55fed8 Mon Sep 17 00:00:00 2001 From: Thomas Bogendoerfer Date: Tue, 21 Apr 2009 13:44:13 +0200 Subject: Fix SYSCALL_ALIAS for older MIPS assembler Older MIPS assembler don't support .set for defining aliases. Using = works for old and new assembers. Signed-off-by: Thomas Bogendoerfer Acked-by: Ralf Baechle Signed-off-by: Linus Torvalds --- include/linux/syscalls.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index dabe4ad89141..40617c1d8976 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -148,7 +148,7 @@ struct old_linux_dirent; asm ("\t.globl " #alias "\n\t.set " #alias ", " #name "\n" \ "\t.globl ." #alias "\n\t.set ." #alias ", ." #name) #else -#ifdef CONFIG_ALPHA +#if defined(CONFIG_ALPHA) || defined(CONFIG_MIPS) #define SYSCALL_ALIAS(alias, name) \ asm ( #alias " = " #name "\n\t.globl " #alias) #else -- cgit v1.2.3 From 546888da82082555a56528730a83f0afd12f33bf Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Tue, 21 Apr 2009 11:53:38 -0400 Subject: Btrfs: fix btrfs fallocate oops and deadlock Btrfs fallocate was incorrectly starting a transaction with a lock held on the extent_io tree for the file, which could deadlock. Strictly speaking it was using join_transaction which would be safe, but it is better to move the transaction outside of the lock. When preallocated extents are overwritten, btrfs_mark_buffer_dirty was being called on an unlocked buffer. This was triggering an assertion and oops because the lock is supposed to be held. The bug was calling btrfs_mark_buffer_dirty on a leaf after btrfs_del_item had been run. btrfs_del_item takes care of dirtying things, so the solution is a to skip the btrfs_mark_buffer_dirty call in this case. Signed-off-by: Chris Mason --- fs/btrfs/file.c | 4 +++- fs/btrfs/inode.c | 36 ++++++++++++++++++++++++++++-------- 2 files changed, 31 insertions(+), 9 deletions(-) diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index e21c0060ee73..482f8db2cfd0 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -830,7 +830,7 @@ again: ret = btrfs_del_items(trans, root, path, del_slot, del_nr); BUG_ON(ret); - goto done; + goto release; } else if (split == start) { if (locked_end < extent_end) { ret = try_lock_extent(&BTRFS_I(inode)->io_tree, @@ -926,6 +926,8 @@ again: } done: btrfs_mark_buffer_dirty(leaf); + +release: btrfs_release_path(root, path); if (split_end && split == start) { split = end; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index a0d1dd492a58..65219f6a16a1 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4970,10 +4970,10 @@ out_fail: return err; } -static int prealloc_file_range(struct inode *inode, u64 start, u64 end, +static int prealloc_file_range(struct btrfs_trans_handle *trans, + struct inode *inode, u64 start, u64 end, u64 alloc_hint, int mode) { - struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_key ins; u64 alloc_size; @@ -4981,10 +4981,6 @@ static int prealloc_file_range(struct inode *inode, u64 start, u64 end, u64 num_bytes = end - start; int ret = 0; - trans = btrfs_join_transaction(root, 1); - BUG_ON(!trans); - btrfs_set_trans_block_group(trans, inode); - while (num_bytes > 0) { alloc_size = min(num_bytes, root->fs_info->max_extent); ret = btrfs_reserve_extent(trans, root, alloc_size, @@ -5015,7 +5011,6 @@ out: BUG_ON(ret); } - btrfs_end_transaction(trans, root); return ret; } @@ -5029,11 +5024,18 @@ static long btrfs_fallocate(struct inode *inode, int mode, u64 alloc_hint = 0; u64 mask = BTRFS_I(inode)->root->sectorsize - 1; struct extent_map *em; + struct btrfs_trans_handle *trans; int ret; alloc_start = offset & ~mask; alloc_end = (offset + len + mask) & ~mask; + /* + * wait for ordered IO before we have any locks. We'll loop again + * below with the locks held. + */ + btrfs_wait_ordered_range(inode, alloc_start, alloc_end - alloc_start); + mutex_lock(&inode->i_mutex); if (alloc_start > inode->i_size) { ret = btrfs_cont_expand(inode, alloc_start); @@ -5043,6 +5045,16 @@ static long btrfs_fallocate(struct inode *inode, int mode, while (1) { struct btrfs_ordered_extent *ordered; + + trans = btrfs_start_transaction(BTRFS_I(inode)->root, 1); + if (!trans) { + ret = -EIO; + goto out; + } + + /* the extent lock is ordered inside the running + * transaction + */ lock_extent(&BTRFS_I(inode)->io_tree, alloc_start, alloc_end - 1, GFP_NOFS); ordered = btrfs_lookup_first_ordered_extent(inode, @@ -5053,6 +5065,12 @@ static long btrfs_fallocate(struct inode *inode, int mode, btrfs_put_ordered_extent(ordered); unlock_extent(&BTRFS_I(inode)->io_tree, alloc_start, alloc_end - 1, GFP_NOFS); + btrfs_end_transaction(trans, BTRFS_I(inode)->root); + + /* + * we can't wait on the range with the transaction + * running or with the extent lock held + */ btrfs_wait_ordered_range(inode, alloc_start, alloc_end - alloc_start); } else { @@ -5070,7 +5088,7 @@ static long btrfs_fallocate(struct inode *inode, int mode, last_byte = min(extent_map_end(em), alloc_end); last_byte = (last_byte + mask) & ~mask; if (em->block_start == EXTENT_MAP_HOLE) { - ret = prealloc_file_range(inode, cur_offset, + ret = prealloc_file_range(trans, inode, cur_offset, last_byte, alloc_hint, mode); if (ret < 0) { free_extent_map(em); @@ -5089,6 +5107,8 @@ static long btrfs_fallocate(struct inode *inode, int mode, } unlock_extent(&BTRFS_I(inode)->io_tree, alloc_start, alloc_end - 1, GFP_NOFS); + + btrfs_end_transaction(trans, BTRFS_I(inode)->root); out: mutex_unlock(&inode->i_mutex); return ret; -- cgit v1.2.3 From 7400516ab40d8fe55031dc8d614e2b365bd95f1c Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Tue, 21 Apr 2009 21:47:22 +0200 Subject: go7007: Convert to the new i2c device binding model Move the go7007 driver away from the legacy i2c binding model, which is going away really soon now. The I2C addresses of the audio and video chips in s2250-board didn't look quite right, apparently they were left-aligned values when Linux wants right-aligned values, so I fixed them too. Signed-off-by: Jean Delvare Cc: Greg Kroah-Hartman --- drivers/staging/go7007/go7007-driver.c | 11 ++++- drivers/staging/go7007/go7007-i2c.c | 83 --------------------------------- drivers/staging/go7007/go7007-priv.h | 1 + drivers/staging/go7007/go7007-usb.c | 14 ++++-- drivers/staging/go7007/s2250-board.c | 71 ++++++++++++---------------- drivers/staging/go7007/wis-i2c.h | 9 ---- drivers/staging/go7007/wis-ov7640.c | 56 +++++++--------------- drivers/staging/go7007/wis-saa7113.c | 60 ++++++++---------------- drivers/staging/go7007/wis-saa7115.c | 60 ++++++++---------------- drivers/staging/go7007/wis-sony-tuner.c | 60 +++++++----------------- drivers/staging/go7007/wis-tw2804.c | 57 +++++++--------------- drivers/staging/go7007/wis-tw9903.c | 60 ++++++++---------------- drivers/staging/go7007/wis-uda1342.c | 52 ++++++--------------- 13 files changed, 176 insertions(+), 418 deletions(-) diff --git a/drivers/staging/go7007/go7007-driver.c b/drivers/staging/go7007/go7007-driver.c index f47c0ce2849a..77b1e769ac92 100644 --- a/drivers/staging/go7007/go7007-driver.c +++ b/drivers/staging/go7007/go7007-driver.c @@ -191,8 +191,10 @@ int go7007_reset_encoder(struct go7007 *go) /* * Attempt to instantiate an I2C client by ID, probably loading a module. */ -static int init_i2c_module(struct i2c_adapter *adapter, int id, int addr) +static int init_i2c_module(struct i2c_adapter *adapter, const char *type, + int id, int addr) { + struct i2c_board_info info; char *modname; switch (id) { @@ -226,7 +228,11 @@ static int init_i2c_module(struct i2c_adapter *adapter, int id, int addr) } if (modname != NULL) request_module(modname); - if (wis_i2c_probe_device(adapter, id, addr) == 1) + + memset(&info, 0, sizeof(struct i2c_board_info)); + info.addr = addr; + strlcpy(info.type, type, I2C_NAME_SIZE); + if (!i2c_new_device(adapter, &info)) return 0; if (modname != NULL) printk(KERN_INFO @@ -266,6 +272,7 @@ int go7007_register_encoder(struct go7007 *go) if (go->i2c_adapter_online) { for (i = 0; i < go->board_info->num_i2c_devs; ++i) init_i2c_module(&go->i2c_adapter, + go->board_info->i2c_devs[i].type, go->board_info->i2c_devs[i].id, go->board_info->i2c_devs[i].addr); if (go->board_id == GO7007_BOARDID_ADLINK_MPG24) diff --git a/drivers/staging/go7007/go7007-i2c.c b/drivers/staging/go7007/go7007-i2c.c index cd55b76eabc7..c82867fdd28d 100644 --- a/drivers/staging/go7007/go7007-i2c.c +++ b/drivers/staging/go7007/go7007-i2c.c @@ -31,87 +31,6 @@ #include "go7007-priv.h" #include "wis-i2c.h" -/************** Registration interface for I2C client drivers **************/ - -/* Since there's no way to auto-probe the I2C devices connected to the I2C - * bus on the go7007, we have this silly little registration system that - * client drivers can use to register their I2C driver ID and their - * detect_client function (the one that's normally passed to i2c_probe). - * - * When a new go7007 device is connected, we can look up in a board info - * table by the USB or PCI vendor/product/revision ID to determine - * which I2C client module to load. The client driver module will register - * itself here, and then we can call the registered detect_client function - * to force-load a new client at the address listed in the board info table. - * - * Really the I2C subsystem should have a way to force-load I2C client - * drivers when we have a priori knowledge of what's on the bus, especially - * since the existing I2C auto-probe mechanism is so hokey, but we'll use - * our own mechanism for the time being. */ - -struct wis_i2c_client_driver { - unsigned int id; - found_proc found_proc; - struct list_head list; -}; - -static LIST_HEAD(i2c_client_drivers); -static DECLARE_MUTEX(i2c_client_driver_list_lock); - -/* Client drivers register here by their I2C driver ID */ -int wis_i2c_add_driver(unsigned int id, found_proc found_proc) -{ - struct wis_i2c_client_driver *driver; - - driver = kmalloc(sizeof(struct wis_i2c_client_driver), GFP_KERNEL); - if (driver == NULL) - return -ENOMEM; - driver->id = id; - driver->found_proc = found_proc; - - down(&i2c_client_driver_list_lock); - list_add_tail(&driver->list, &i2c_client_drivers); - up(&i2c_client_driver_list_lock); - - return 0; -} -EXPORT_SYMBOL(wis_i2c_add_driver); - -void wis_i2c_del_driver(found_proc found_proc) -{ - struct wis_i2c_client_driver *driver, *next; - - down(&i2c_client_driver_list_lock); - list_for_each_entry_safe(driver, next, &i2c_client_drivers, list) - if (driver->found_proc == found_proc) { - list_del(&driver->list); - kfree(driver); - } - up(&i2c_client_driver_list_lock); -} -EXPORT_SYMBOL(wis_i2c_del_driver); - -/* The main go7007 driver calls this to instantiate a client by driver - * ID and bus address, which are both stored in the board info table */ -int wis_i2c_probe_device(struct i2c_adapter *adapter, - unsigned int id, int addr) -{ - struct wis_i2c_client_driver *driver; - int found = 0; - - if (addr < 0 || addr > 0x7f) - return -1; - down(&i2c_client_driver_list_lock); - list_for_each_entry(driver, &i2c_client_drivers, list) - if (driver->id == id) { - if (driver->found_proc(adapter, addr, 0) == 0) - found = 1; - break; - } - up(&i2c_client_driver_list_lock); - return found; -} - /********************* Driver for on-board I2C adapter *********************/ /* #define GO7007_I2C_DEBUG */ @@ -287,9 +206,7 @@ static struct i2c_algorithm go7007_algo = { static struct i2c_adapter go7007_adap_templ = { .owner = THIS_MODULE, - .class = I2C_CLASS_TV_ANALOG, .name = "WIS GO7007SB", - .id = I2C_ALGO_GO7007, .algo = &go7007_algo, }; diff --git a/drivers/staging/go7007/go7007-priv.h b/drivers/staging/go7007/go7007-priv.h index 372f1f1c09b2..178d18119faa 100644 --- a/drivers/staging/go7007/go7007-priv.h +++ b/drivers/staging/go7007/go7007-priv.h @@ -87,6 +87,7 @@ struct go7007_board_info { int audio_main_div; int num_i2c_devs; struct { + const char *type; int id; int addr; } i2c_devs[4]; diff --git a/drivers/staging/go7007/go7007-usb.c b/drivers/staging/go7007/go7007-usb.c index 83eec920c7d3..aa4a9e0b9954 100644 --- a/drivers/staging/go7007/go7007-usb.c +++ b/drivers/staging/go7007/go7007-usb.c @@ -91,6 +91,7 @@ static struct go7007_usb_board board_matrix_ii = { .num_i2c_devs = 1, .i2c_devs = { { + .type = "wis_saa7115", .id = I2C_DRIVERID_WIS_SAA7115, .addr = 0x20, }, @@ -127,6 +128,7 @@ static struct go7007_usb_board board_matrix_reload = { .num_i2c_devs = 1, .i2c_devs = { { + .type = "wis_saa7113", .id = I2C_DRIVERID_WIS_SAA7113, .addr = 0x25, }, @@ -164,6 +166,7 @@ static struct go7007_usb_board board_star_trek = { .num_i2c_devs = 1, .i2c_devs = { { + .type = "wis_saa7115", .id = I2C_DRIVERID_WIS_SAA7115, .addr = 0x20, }, @@ -209,14 +212,17 @@ static struct go7007_usb_board board_px_tv402u = { .num_i2c_devs = 3, .i2c_devs = { { + .type = "wis_saa7115", .id = I2C_DRIVERID_WIS_SAA7115, .addr = 0x20, }, { + .type = "wis_uda1342", .id = I2C_DRIVERID_WIS_UDA1342, .addr = 0x1a, }, { + .type = "wis_sony_tuner", .id = I2C_DRIVERID_WIS_SONY_TUNER, .addr = 0x60, }, @@ -264,6 +270,7 @@ static struct go7007_usb_board board_xmen = { .num_i2c_devs = 1, .i2c_devs = { { + .type = "wis_ov7640", .id = I2C_DRIVERID_WIS_OV7640, .addr = 0x21, }, @@ -296,6 +303,7 @@ static struct go7007_usb_board board_matrix_revolution = { .num_i2c_devs = 1, .i2c_devs = { { + .type = "wis_tw9903", .id = I2C_DRIVERID_WIS_TW9903, .addr = 0x44, }, @@ -385,6 +393,7 @@ static struct go7007_usb_board board_adlink_mpg24 = { .num_i2c_devs = 1, .i2c_devs = { { + .type = "wis_twTW2804", .id = I2C_DRIVERID_WIS_TW2804, .addr = 0x00, /* yes, really */ }, @@ -415,8 +424,9 @@ static struct go7007_usb_board board_sensoray_2250 = { .num_i2c_devs = 1, .i2c_devs = { { + .type = "s2250_board", .id = I2C_DRIVERID_S2250, - .addr = 0x34, + .addr = 0x43, }, }, .num_inputs = 2, @@ -943,9 +953,7 @@ static struct i2c_algorithm go7007_usb_algo = { static struct i2c_adapter go7007_usb_adap_templ = { .owner = THIS_MODULE, - .class = I2C_CLASS_TV_ANALOG, .name = "WIS GO7007SB EZ-USB", - .id = I2C_ALGO_GO7007_USB, .algo = &go7007_usb_algo, }; diff --git a/drivers/staging/go7007/s2250-board.c b/drivers/staging/go7007/s2250-board.c index d333ea2cd774..1706fbf06847 100644 --- a/drivers/staging/go7007/s2250-board.c +++ b/drivers/staging/go7007/s2250-board.c @@ -28,7 +28,6 @@ extern int s2250loader_init(void); extern void s2250loader_cleanup(void); #define TLV320_ADDRESS 0x34 -#define S2250_VIDDEC 0x86 #define VPX322_ADDR_ANALOGCONTROL1 0x02 #define VPX322_ADDR_BRIGHTNESS0 0x0127 #define VPX322_ADDR_BRIGHTNESS1 0x0131 @@ -123,6 +122,7 @@ struct s2250 { int hue; int reg12b_val; int audio_input; + struct i2c_client *audio; }; /* from go7007-usb.c which is Copyright (C) 2005-2006 Micronas USA Inc.*/ @@ -452,16 +452,15 @@ static int s2250_command(struct i2c_client *client, { struct v4l2_audio *audio = arg; - client->addr = TLV320_ADDRESS; switch (audio->index) { case 0: - write_reg(client, 0x08, 0x02); /* Line In */ + write_reg(dec->audio, 0x08, 0x02); /* Line In */ break; case 1: - write_reg(client, 0x08, 0x04); /* Mic */ + write_reg(dec->audio, 0x08, 0x04); /* Mic */ break; case 2: - write_reg(client, 0x08, 0x05); /* Mic Boost */ + write_reg(dec->audio, 0x08, 0x05); /* Mic Boost */ break; default: return -EINVAL; @@ -477,31 +476,23 @@ static int s2250_command(struct i2c_client *client, return 0; } -static struct i2c_driver s2250_driver; - -static struct i2c_client s2250_client_templ = { - .name = "Sensoray 2250", - .driver = &s2250_driver, -}; - -static int s2250_detect(struct i2c_adapter *adapter, int addr, int kind) +static int s2250_probe(struct i2c_client *client, + const struct i2c_device_id *id) { - struct i2c_client *client; + struct i2c_client *audio; + struct i2c_adapter *adapter = client->adapter; struct s2250 *dec; u8 *data; struct go7007 *go = i2c_get_adapdata(adapter); struct go7007_usb *usb = go->hpi_context; - client = kmalloc(sizeof(struct i2c_client), GFP_KERNEL); - if (client == NULL) + audio = i2c_new_dummy(adapter, TLV320_ADDRESS >> 1); + if (audio == NULL) return -ENOMEM; - memcpy(client, &s2250_client_templ, - sizeof(s2250_client_templ)); - client->adapter = adapter; dec = kmalloc(sizeof(struct s2250), GFP_KERNEL); if (dec == NULL) { - kfree(client); + i2c_unregister_device(audio); return -ENOMEM; } @@ -510,7 +501,7 @@ static int s2250_detect(struct i2c_adapter *adapter, int addr, int kind) dec->contrast = 50; dec->saturation = 50; dec->hue = 0; - client->addr = TLV320_ADDRESS; + dec->audio = audio; i2c_set_clientdata(client, dec); printk(KERN_DEBUG @@ -518,28 +509,25 @@ static int s2250_detect(struct i2c_adapter *adapter, int addr, int kind) adapter->name); /* initialize the audio */ - client->addr = TLV320_ADDRESS; - if (write_regs(client, aud_regs) < 0) { + if (write_regs(audio, aud_regs) < 0) { printk(KERN_ERR "s2250: error initializing audio\n"); - kfree(client); + i2c_unregister_device(audio); kfree(dec); return 0; } - client->addr = S2250_VIDDEC; - i2c_set_clientdata(client, dec); if (write_regs(client, vid_regs) < 0) { printk(KERN_ERR "s2250: error initializing decoder\n"); - kfree(client); + i2c_unregister_device(audio); kfree(dec); return 0; } if (write_regs_fp(client, vid_regs_fp) < 0) { printk(KERN_ERR "s2250: error initializing decoder\n"); - kfree(client); + i2c_unregister_device(audio); kfree(dec); return 0; } @@ -575,32 +563,33 @@ static int s2250_detect(struct i2c_adapter *adapter, int addr, int kind) up(&usb->i2c_lock); } - i2c_attach_client(client); printk("s2250: initialized successfully\n"); return 0; } -static int s2250_detach(struct i2c_client *client) +static int s2250_remove(struct i2c_client *client) { struct s2250 *dec = i2c_get_clientdata(client); - int r; - - r = i2c_detach_client(client); - if (r < 0) - return r; - kfree(client); + i2c_set_clientdata(client, NULL); + i2c_unregister_device(dec->audio); kfree(dec); return 0; } +static struct i2c_device_id s2250_id[] = { + { "s2250_board", 0 }, + { } +}; + static struct i2c_driver s2250_driver = { .driver = { .name = "Sensoray 2250 board driver", }, - .id = I2C_DRIVERID_S2250, - .detach_client = s2250_detach, + .probe = s2250_probe, + .remove = s2250_remove, .command = s2250_command, + .id_table = s2250_id, }; static int __init s2250_init(void) @@ -613,13 +602,13 @@ static int __init s2250_init(void) r = i2c_add_driver(&s2250_driver); if (r < 0) - return r; - return wis_i2c_add_driver(s2250_driver.id, s2250_detect); + s2250loader_cleanup(); + + return r; } static void __exit s2250_cleanup(void) { - wis_i2c_del_driver(s2250_detect); i2c_del_driver(&s2250_driver); s2250loader_cleanup(); diff --git a/drivers/staging/go7007/wis-i2c.h b/drivers/staging/go7007/wis-i2c.h index 431f41dd3966..3c2b9be455df 100644 --- a/drivers/staging/go7007/wis-i2c.h +++ b/drivers/staging/go7007/wis-i2c.h @@ -24,21 +24,12 @@ #define I2C_DRIVERID_WIS_OV7640 0xf0f5 #define I2C_DRIVERID_WIS_TW2804 0xf0f6 #define I2C_DRIVERID_S2250 0xf0f7 -#define I2C_ALGO_GO7007 0xf00000 -#define I2C_ALGO_GO7007_USB 0xf10000 /* Flag to indicate that the client needs to be accessed with SCCB semantics */ /* We re-use the I2C_M_TEN value so the flag passes through the masks in the * core I2C code. Major kludge, but the I2C layer ain't exactly flexible. */ #define I2C_CLIENT_SCCB 0x10 -typedef int (*found_proc) (struct i2c_adapter *, int, int); -int wis_i2c_add_driver(unsigned int id, found_proc found_proc); -void wis_i2c_del_driver(found_proc found_proc); - -int wis_i2c_probe_device(struct i2c_adapter *adapter, - unsigned int id, int addr); - /* Definitions for new video decoder commands */ struct video_decoder_resolution { diff --git a/drivers/staging/go7007/wis-ov7640.c b/drivers/staging/go7007/wis-ov7640.c index 2f9efca04606..04d6d3a498a3 100644 --- a/drivers/staging/go7007/wis-ov7640.c +++ b/drivers/staging/go7007/wis-ov7640.c @@ -50,76 +50,54 @@ static int write_regs(struct i2c_client *client, u8 *regs) return 0; } -static struct i2c_driver wis_ov7640_driver; - -static struct i2c_client wis_ov7640_client_templ = { - .name = "OV7640 (WIS)", - .driver = &wis_ov7640_driver, -}; - -static int wis_ov7640_detect(struct i2c_adapter *adapter, int addr, int kind) +static int wis_ov7640_probe(struct i2c_client *client, + const struct i2c_device_id *id) { - struct i2c_client *client; + struct i2c_adapter *adapter = client->adapter; if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA)) - return 0; - - client = kmalloc(sizeof(struct i2c_client), GFP_KERNEL); - if (client == NULL) - return -ENOMEM; - memcpy(client, &wis_ov7640_client_templ, - sizeof(wis_ov7640_client_templ)); - client->adapter = adapter; - client->addr = addr; + return -ENODEV; + client->flags = I2C_CLIENT_SCCB; printk(KERN_DEBUG "wis-ov7640: initializing OV7640 at address %d on %s\n", - addr, adapter->name); + client->addr, adapter->name); if (write_regs(client, initial_registers) < 0) { printk(KERN_ERR "wis-ov7640: error initializing OV7640\n"); - kfree(client); - return 0; + return -ENODEV; } - i2c_attach_client(client); return 0; } -static int wis_ov7640_detach(struct i2c_client *client) +static int wis_ov7640_remove(struct i2c_client *client) { - int r; - - r = i2c_detach_client(client); - if (r < 0) - return r; - - kfree(client); return 0; } +static struct i2c_device_id wis_ov7640_id[] = { + { "wis_ov7640", 0 }, + { } +}; + static struct i2c_driver wis_ov7640_driver = { .driver = { .name = "WIS OV7640 I2C driver", }, - .id = I2C_DRIVERID_WIS_OV7640, - .detach_client = wis_ov7640_detach, + .probe = wis_ov7640_probe, + .remove = wis_ov7640_remove, + .id_table = wis_ov7640_id, }; static int __init wis_ov7640_init(void) { - int r; - - r = i2c_add_driver(&wis_ov7640_driver); - if (r < 0) - return r; - return wis_i2c_add_driver(wis_ov7640_driver.id, wis_ov7640_detect); + return i2c_add_driver(&wis_ov7640_driver); } static void __exit wis_ov7640_cleanup(void) { - wis_i2c_del_driver(wis_ov7640_detect); i2c_del_driver(&wis_ov7640_driver); } diff --git a/drivers/staging/go7007/wis-saa7113.c b/drivers/staging/go7007/wis-saa7113.c index 11689723945e..9ab893bd204e 100644 --- a/drivers/staging/go7007/wis-saa7113.c +++ b/drivers/staging/go7007/wis-saa7113.c @@ -261,34 +261,19 @@ static int wis_saa7113_command(struct i2c_client *client, return 0; } -static struct i2c_driver wis_saa7113_driver; - -static struct i2c_client wis_saa7113_client_templ = { - .name = "SAA7113 (WIS)", - .driver = &wis_saa7113_driver, -}; - -static int wis_saa7113_detect(struct i2c_adapter *adapter, int addr, int kind) +static int wis_saa7113_probe(struct i2c_client *client, + const struct i2c_device_id *id) { - struct i2c_client *client; + struct i2c_adapter *adapter = client->adapter; struct wis_saa7113 *dec; if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA)) - return 0; - - client = kmalloc(sizeof(struct i2c_client), GFP_KERNEL); - if (client == NULL) - return -ENOMEM; - memcpy(client, &wis_saa7113_client_templ, - sizeof(wis_saa7113_client_templ)); - client->adapter = adapter; - client->addr = addr; + return -ENODEV; dec = kmalloc(sizeof(struct wis_saa7113), GFP_KERNEL); - if (dec == NULL) { - kfree(client); + if (dec == NULL) return -ENOMEM; - } + dec->norm = V4L2_STD_NTSC; dec->brightness = 128; dec->contrast = 71; @@ -298,56 +283,49 @@ static int wis_saa7113_detect(struct i2c_adapter *adapter, int addr, int kind) printk(KERN_DEBUG "wis-saa7113: initializing SAA7113 at address %d on %s\n", - addr, adapter->name); + client->addr, adapter->name); if (write_regs(client, initial_registers) < 0) { printk(KERN_ERR "wis-saa7113: error initializing SAA7113\n"); - kfree(client); kfree(dec); - return 0; + return -ENODEV; } - i2c_attach_client(client); return 0; } -static int wis_saa7113_detach(struct i2c_client *client) +static int wis_saa7113_remove(struct i2c_client *client) { struct wis_saa7113 *dec = i2c_get_clientdata(client); - int r; - - r = i2c_detach_client(client); - if (r < 0) - return r; - kfree(client); + i2c_set_clientdata(client, NULL); kfree(dec); return 0; } +static struct i2c_device_id wis_saa7113_id[] = { + { "wis_saa7113", 0 }, + { } +}; + static struct i2c_driver wis_saa7113_driver = { .driver = { .name = "WIS SAA7113 I2C driver", }, - .id = I2C_DRIVERID_WIS_SAA7113, - .detach_client = wis_saa7113_detach, + .probe = wis_saa7113_probe, + .remove = wis_saa7113_remove, .command = wis_saa7113_command, + .id_table = wis_saa7113_id, }; static int __init wis_saa7113_init(void) { - int r; - - r = i2c_add_driver(&wis_saa7113_driver); - if (r < 0) - return r; - return wis_i2c_add_driver(wis_saa7113_driver.id, wis_saa7113_detect); + return i2c_add_driver(&wis_saa7113_driver); } static void __exit wis_saa7113_cleanup(void) { - wis_i2c_del_driver(wis_saa7113_detect); i2c_del_driver(&wis_saa7113_driver); } diff --git a/drivers/staging/go7007/wis-saa7115.c b/drivers/staging/go7007/wis-saa7115.c index 59417a7174d7..8687ad2de761 100644 --- a/drivers/staging/go7007/wis-saa7115.c +++ b/drivers/staging/go7007/wis-saa7115.c @@ -394,34 +394,19 @@ static int wis_saa7115_command(struct i2c_client *client, return 0; } -static struct i2c_driver wis_saa7115_driver; - -static struct i2c_client wis_saa7115_client_templ = { - .name = "SAA7115 (WIS)", - .driver = &wis_saa7115_driver, -}; - -static int wis_saa7115_detect(struct i2c_adapter *adapter, int addr, int kind) +static int wis_saa7115_probe(struct i2c_client *client, + const struct i2c_device_id *id) { - struct i2c_client *client; + struct i2c_adapter *adapter = client->adapter; struct wis_saa7115 *dec; if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA)) - return 0; - - client = kmalloc(sizeof(struct i2c_client), GFP_KERNEL); - if (client == NULL) - return -ENOMEM; - memcpy(client, &wis_saa7115_client_templ, - sizeof(wis_saa7115_client_templ)); - client->adapter = adapter; - client->addr = addr; + return -ENODEV; dec = kmalloc(sizeof(struct wis_saa7115), GFP_KERNEL); - if (dec == NULL) { - kfree(client); + if (dec == NULL) return -ENOMEM; - } + dec->norm = V4L2_STD_NTSC; dec->brightness = 128; dec->contrast = 64; @@ -431,56 +416,49 @@ static int wis_saa7115_detect(struct i2c_adapter *adapter, int addr, int kind) printk(KERN_DEBUG "wis-saa7115: initializing SAA7115 at address %d on %s\n", - addr, adapter->name); + client->addr, adapter->name); if (write_regs(client, initial_registers) < 0) { printk(KERN_ERR "wis-saa7115: error initializing SAA7115\n"); - kfree(client); kfree(dec); - return 0; + return -ENODEV; } - i2c_attach_client(client); return 0; } -static int wis_saa7115_detach(struct i2c_client *client) +static int wis_saa7115_remove(struct i2c_client *client) { struct wis_saa7115 *dec = i2c_get_clientdata(client); - int r; - - r = i2c_detach_client(client); - if (r < 0) - return r; - kfree(client); + i2c_set_clientdata(client, NULL); kfree(dec); return 0; } +static struct i2c_device_id wis_saa7115_id[] = { + { "wis_saa7115", 0 }, + { } +}; + static struct i2c_driver wis_saa7115_driver = { .driver = { .name = "WIS SAA7115 I2C driver", }, - .id = I2C_DRIVERID_WIS_SAA7115, - .detach_client = wis_saa7115_detach, + .probe = wis_saa7115_probe, + .remove = wis_saa7115_remove, .command = wis_saa7115_command, + .id_table = wis_saa7115_id, }; static int __init wis_saa7115_init(void) { - int r; - - r = i2c_add_driver(&wis_saa7115_driver); - if (r < 0) - return r; - return wis_i2c_add_driver(wis_saa7115_driver.id, wis_saa7115_detect); + return i2c_add_driver(&wis_saa7115_driver); } static void __exit wis_saa7115_cleanup(void) { - wis_i2c_del_driver(wis_saa7115_detect); i2c_del_driver(&wis_saa7115_driver); } diff --git a/drivers/staging/go7007/wis-sony-tuner.c b/drivers/staging/go7007/wis-sony-tuner.c index 0a7eeef7c008..c965c601ac90 100644 --- a/drivers/staging/go7007/wis-sony-tuner.c +++ b/drivers/staging/go7007/wis-sony-tuner.c @@ -653,35 +653,19 @@ static int tuner_command(struct i2c_client *client, unsigned int cmd, void *arg) return 0; } -static struct i2c_driver wis_sony_tuner_driver; - -static struct i2c_client wis_sony_tuner_client_templ = { - .name = "Sony TV Tuner (WIS)", - .driver = &wis_sony_tuner_driver, -}; - -static int wis_sony_tuner_detect(struct i2c_adapter *adapter, - int addr, int kind) +static int wis_sony_tuner_probe(struct i2c_client *client, + const struct i2c_device_id *id) { - struct i2c_client *client; + struct i2c_adapter *adapter = client->adapter; struct wis_sony_tuner *t; if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_I2C_BLOCK)) - return 0; - - client = kmalloc(sizeof(struct i2c_client), GFP_KERNEL); - if (client == NULL) - return -ENOMEM; - memcpy(client, &wis_sony_tuner_client_templ, - sizeof(wis_sony_tuner_client_templ)); - client->adapter = adapter; - client->addr = addr; + return -ENODEV; t = kmalloc(sizeof(struct wis_sony_tuner), GFP_KERNEL); - if (t == NULL) { - kfree(client); + if (t == NULL) return -ENOMEM; - } + t->type = -1; t->freq = 0; t->mpxmode = 0; @@ -690,50 +674,42 @@ static int wis_sony_tuner_detect(struct i2c_adapter *adapter, printk(KERN_DEBUG "wis-sony-tuner: initializing tuner at address %d on %s\n", - addr, adapter->name); - - i2c_attach_client(client); + client->addr, adapter->name); return 0; } -static int wis_sony_tuner_detach(struct i2c_client *client) +static int wis_sony_tuner_remove(struct i2c_client *client) { struct wis_sony_tuner *t = i2c_get_clientdata(client); - int r; - - r = i2c_detach_client(client); - if (r < 0) - return r; + i2c_set_clientdata(client, NULL); kfree(t); - kfree(client); return 0; } +static struct i2c_device_id wis_sony_tuner_id[] = { + { "wis_sony_tuner", 0 }, + { } +}; + static struct i2c_driver wis_sony_tuner_driver = { .driver = { .name = "WIS Sony TV Tuner I2C driver", }, - .id = I2C_DRIVERID_WIS_SONY_TUNER, - .detach_client = wis_sony_tuner_detach, + .probe = wis_sony_tuner_probe, + .remove = wis_sony_tuner_remove, .command = tuner_command, + .id_table = wis_sony_tuner_id, }; static int __init wis_sony_tuner_init(void) { - int r; - - r = i2c_add_driver(&wis_sony_tuner_driver); - if (r < 0) - return r; - return wis_i2c_add_driver(wis_sony_tuner_driver.id, - wis_sony_tuner_detect); + return i2c_add_driver(&wis_sony_tuner_driver); } static void __exit wis_sony_tuner_cleanup(void) { - wis_i2c_del_driver(wis_sony_tuner_detect); i2c_del_driver(&wis_sony_tuner_driver); } diff --git a/drivers/staging/go7007/wis-tw2804.c b/drivers/staging/go7007/wis-tw2804.c index 57b8f2b1caa3..e15794a2a0ae 100644 --- a/drivers/staging/go7007/wis-tw2804.c +++ b/drivers/staging/go7007/wis-tw2804.c @@ -291,34 +291,19 @@ static int wis_tw2804_command(struct i2c_client *client, return 0; } -static struct i2c_driver wis_tw2804_driver; - -static struct i2c_client wis_tw2804_client_templ = { - .name = "TW2804 (WIS)", - .driver = &wis_tw2804_driver, -}; - -static int wis_tw2804_detect(struct i2c_adapter *adapter, int addr, int kind) +static int wis_tw2804_probe(struct i2c_client *client, + const struct i2c_device_id *id) { - struct i2c_client *client; + struct i2c_adapter *adapter = client->adapter; struct wis_tw2804 *dec; if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA)) - return 0; - - client = kmalloc(sizeof(struct i2c_client), GFP_KERNEL); - if (client == NULL) - return -ENOMEM; - memcpy(client, &wis_tw2804_client_templ, - sizeof(wis_tw2804_client_templ)); - client->adapter = adapter; - client->addr = addr; + return -ENODEV; dec = kmalloc(sizeof(struct wis_tw2804), GFP_KERNEL); - if (dec == NULL) { - kfree(client); + if (dec == NULL) return -ENOMEM; - } + dec->channel = -1; dec->norm = V4L2_STD_NTSC; dec->brightness = 128; @@ -328,48 +313,42 @@ static int wis_tw2804_detect(struct i2c_adapter *adapter, int addr, int kind) i2c_set_clientdata(client, dec); printk(KERN_DEBUG "wis-tw2804: creating TW2804 at address %d on %s\n", - addr, adapter->name); + client->addr, adapter->name); - i2c_attach_client(client); return 0; } -static int wis_tw2804_detach(struct i2c_client *client) +static int wis_tw2804_remove(struct i2c_client *client) { struct wis_tw2804 *dec = i2c_get_clientdata(client); - int r; - - r = i2c_detach_client(client); - if (r < 0) - return r; - kfree(client); + i2c_set_clientdata(client, NULL); kfree(dec); return 0; } +static struct i2c_device_id wis_tw2804_id[] = { + { "wis_tw2804", 0 }, + { } +}; + static struct i2c_driver wis_tw2804_driver = { .driver = { .name = "WIS TW2804 I2C driver", }, - .id = I2C_DRIVERID_WIS_TW2804, - .detach_client = wis_tw2804_detach, + .probe = wis_tw2804_probe, + .remove = wis_tw2804_remove, .command = wis_tw2804_command, + .id_table = wis_tw2804_id, }; static int __init wis_tw2804_init(void) { - int r; - - r = i2c_add_driver(&wis_tw2804_driver); - if (r < 0) - return r; - return wis_i2c_add_driver(wis_tw2804_driver.id, wis_tw2804_detect); + return i2c_add_driver(&wis_tw2804_driver); } static void __exit wis_tw2804_cleanup(void) { - wis_i2c_del_driver(wis_tw2804_detect); i2c_del_driver(&wis_tw2804_driver); } diff --git a/drivers/staging/go7007/wis-tw9903.c b/drivers/staging/go7007/wis-tw9903.c index 40627b282cb4..6c3427bb6f4c 100644 --- a/drivers/staging/go7007/wis-tw9903.c +++ b/drivers/staging/go7007/wis-tw9903.c @@ -267,34 +267,19 @@ static int wis_tw9903_command(struct i2c_client *client, return 0; } -static struct i2c_driver wis_tw9903_driver; - -static struct i2c_client wis_tw9903_client_templ = { - .name = "TW9903 (WIS)", - .driver = &wis_tw9903_driver, -}; - -static int wis_tw9903_detect(struct i2c_adapter *adapter, int addr, int kind) +static int wis_tw9903_probe(struct i2c_client *client, + const struct i2c_device_id *id) { - struct i2c_client *client; + struct i2c_adapter *adapter = client->adapter; struct wis_tw9903 *dec; if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA)) - return 0; - - client = kmalloc(sizeof(struct i2c_client), GFP_KERNEL); - if (client == NULL) - return -ENOMEM; - memcpy(client, &wis_tw9903_client_templ, - sizeof(wis_tw9903_client_templ)); - client->adapter = adapter; - client->addr = addr; + return -ENODEV; dec = kmalloc(sizeof(struct wis_tw9903), GFP_KERNEL); - if (dec == NULL) { - kfree(client); + if (dec == NULL) return -ENOMEM; - } + dec->norm = V4L2_STD_NTSC; dec->brightness = 0; dec->contrast = 0x60; @@ -303,55 +288,48 @@ static int wis_tw9903_detect(struct i2c_adapter *adapter, int addr, int kind) printk(KERN_DEBUG "wis-tw9903: initializing TW9903 at address %d on %s\n", - addr, adapter->name); + client->addr, adapter->name); if (write_regs(client, initial_registers) < 0) { printk(KERN_ERR "wis-tw9903: error initializing TW9903\n"); - kfree(client); kfree(dec); - return 0; + return -ENODEV; } - i2c_attach_client(client); return 0; } -static int wis_tw9903_detach(struct i2c_client *client) +static int wis_tw9903_remove(struct i2c_client *client) { struct wis_tw9903 *dec = i2c_get_clientdata(client); - int r; - - r = i2c_detach_client(client); - if (r < 0) - return r; - kfree(client); + i2c_set_clientdata(client, NULL); kfree(dec); return 0; } +static struct i2c_device_id wis_tw9903_id[] = { + { "wis_tw9903", 0 }, + { } +}; + static struct i2c_driver wis_tw9903_driver = { .driver = { .name = "WIS TW9903 I2C driver", }, - .id = I2C_DRIVERID_WIS_TW9903, - .detach_client = wis_tw9903_detach, + .probe = wis_tw9903_probe, + .remove = wis_tw9903_remove, .command = wis_tw9903_command, + .id_table = wis_tw9903_id, }; static int __init wis_tw9903_init(void) { - int r; - - r = i2c_add_driver(&wis_tw9903_driver); - if (r < 0) - return r; - return wis_i2c_add_driver(wis_tw9903_driver.id, wis_tw9903_detect); + return i2c_add_driver(&wis_tw9903_driver); } static void __exit wis_tw9903_cleanup(void) { - wis_i2c_del_driver(wis_tw9903_detect); i2c_del_driver(&wis_tw9903_driver); } diff --git a/drivers/staging/go7007/wis-uda1342.c b/drivers/staging/go7007/wis-uda1342.c index 555645c0cc1a..739c7ae8913f 100644 --- a/drivers/staging/go7007/wis-uda1342.c +++ b/drivers/staging/go7007/wis-uda1342.c @@ -59,73 +59,51 @@ static int wis_uda1342_command(struct i2c_client *client, return 0; } -static struct i2c_driver wis_uda1342_driver; - -static struct i2c_client wis_uda1342_client_templ = { - .name = "UDA1342 (WIS)", - .driver = &wis_uda1342_driver, -}; - -static int wis_uda1342_detect(struct i2c_adapter *adapter, int addr, int kind) +static int wis_uda1342_probe(struct i2c_client *client, + const struct i2c_device_id *id) { - struct i2c_client *client; + struct i2c_adapter *adapter = client->adapter; if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_WORD_DATA)) - return 0; - - client = kmalloc(sizeof(struct i2c_client), GFP_KERNEL); - if (client == NULL) - return -ENOMEM; - memcpy(client, &wis_uda1342_client_templ, - sizeof(wis_uda1342_client_templ)); - client->adapter = adapter; - client->addr = addr; + return -ENODEV; printk(KERN_DEBUG "wis-uda1342: initializing UDA1342 at address %d on %s\n", - addr, adapter->name); + client->addr, adapter->name); write_reg(client, 0x00, 0x8000); /* reset registers */ write_reg(client, 0x00, 0x1241); /* select input 1 */ - i2c_attach_client(client); return 0; } -static int wis_uda1342_detach(struct i2c_client *client) +static int wis_uda1342_remove(struct i2c_client *client) { - int r; - - r = i2c_detach_client(client); - if (r < 0) - return r; - - kfree(client); return 0; } +static struct i2c_device_id wis_uda1342_id[] = { + { "wis_uda1342", 0 }, + { } +}; + static struct i2c_driver wis_uda1342_driver = { .driver = { .name = "WIS UDA1342 I2C driver", }, - .id = I2C_DRIVERID_WIS_UDA1342, - .detach_client = wis_uda1342_detach, + .probe = wis_uda1342_probe, + .remove = wis_uda1342_remove, .command = wis_uda1342_command, + .id_table = wis_uda1342_id, }; static int __init wis_uda1342_init(void) { - int r; - - r = i2c_add_driver(&wis_uda1342_driver); - if (r < 0) - return r; - return wis_i2c_add_driver(wis_uda1342_driver.id, wis_uda1342_detect); + return i2c_add_driver(&wis_uda1342_driver); } static void __exit wis_uda1342_cleanup(void) { - wis_i2c_del_driver(wis_uda1342_detect); i2c_del_driver(&wis_uda1342_driver); } -- cgit v1.2.3 From 44aa417910ec9cda6da42be914105e789273b507 Mon Sep 17 00:00:00 2001 From: Vlada Peric Date: Tue, 21 Apr 2009 12:23:59 -0700 Subject: asiliantfb: add missing return statement Commit 032220ba (asiliantfb: fix cmap memory leaks) changed the function init_asiliant from void to int, resulting in the following compile warning: drivers/video/asiliantfb.c: In function `init_asiliant': drivers/video/asiliantfb.c:536: warning: control reaches end of non-void function Fix the warning by returning 0. Signed-off-by: Vlada Peric Cc: Andres Salomon Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/asiliantfb.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/video/asiliantfb.c b/drivers/video/asiliantfb.c index 1a1f946d8fef..9fe90ce928fb 100644 --- a/drivers/video/asiliantfb.c +++ b/drivers/video/asiliantfb.c @@ -533,6 +533,7 @@ static int __devinit init_asiliant(struct fb_info *p, unsigned long addr) writeb(0xff, mmio_base + 0x78c); chips_hw_init(p); + return 0; } static int __devinit -- cgit v1.2.3 From ff14ed5db6e7e5e5dc23712d3c877891d4d9a1a8 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Tue, 21 Apr 2009 12:23:59 -0700 Subject: pxafb: lcsr1 is unused without CONFIG_FB_PXA_OVERLAY Fixes the warning: drivers/video/pxafb.c: In function 'pxafb_handle_irq': drivers/video/pxafb.c:1442: warning: unused variable 'lcsr1' [akpm@linux-foundation.org: save an ifdef] Signed-off-by: Denis V. Lunev Cc: Eric Miao Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/pxafb.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/drivers/video/pxafb.c b/drivers/video/pxafb.c index 84f63205c46d..0889d50c3288 100644 --- a/drivers/video/pxafb.c +++ b/drivers/video/pxafb.c @@ -1439,7 +1439,7 @@ static void pxafb_disable_controller(struct pxafb_info *fbi) static irqreturn_t pxafb_handle_irq(int irq, void *dev_id) { struct pxafb_info *fbi = dev_id; - unsigned int lccr0, lcsr, lcsr1; + unsigned int lccr0, lcsr; lcsr = lcd_readl(fbi, LCSR); if (lcsr & LCSR_LDD) { @@ -1455,14 +1455,16 @@ static irqreturn_t pxafb_handle_irq(int irq, void *dev_id) lcd_writel(fbi, LCSR, lcsr); #ifdef CONFIG_FB_PXA_OVERLAY - lcsr1 = lcd_readl(fbi, LCSR1); - if (lcsr1 & LCSR1_BS(1)) - complete(&fbi->overlay[0].branch_done); + { + unsigned int lcsr1 = lcd_readl(fbi, LCSR1); + if (lcsr1 & LCSR1_BS(1)) + complete(&fbi->overlay[0].branch_done); - if (lcsr1 & LCSR1_BS(2)) - complete(&fbi->overlay[1].branch_done); + if (lcsr1 & LCSR1_BS(2)) + complete(&fbi->overlay[1].branch_done); - lcd_writel(fbi, LCSR1, lcsr1); + lcd_writel(fbi, LCSR1, lcsr1); + } #endif return IRQ_HANDLED; } -- cgit v1.2.3 From 8e19608e8b5c001e4a66ce482edc474f05fb7355 Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Tue, 21 Apr 2009 12:24:00 -0700 Subject: clocksource: pass clocksource to read() callback Pass clocksource pointer to the read() callback for clocksources. This allows us to share the callback between multiple instances. [hugh@veritas.com: fix powerpc build of clocksource pass clocksource mods] [akpm@linux-foundation.org: cleanup] Signed-off-by: Magnus Damm Acked-by: John Stultz Cc: Thomas Gleixner Signed-off-by: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/mach-at91/at91rm9200_time.c | 2 +- arch/arm/mach-at91/at91sam926x_time.c | 2 +- arch/arm/mach-davinci/time.c | 2 +- arch/arm/mach-imx/time.c | 2 +- arch/arm/mach-ixp4xx/common.c | 2 +- arch/arm/mach-msm/timer.c | 4 ++-- arch/arm/mach-netx/time.c | 2 +- arch/arm/mach-ns9xxx/time-ns9360.c | 2 +- arch/arm/mach-omap1/time.c | 2 +- arch/arm/mach-omap2/timer-gp.c | 2 +- arch/arm/mach-pxa/time.c | 2 +- arch/arm/mach-realview/core.c | 2 +- arch/arm/mach-versatile/core.c | 2 +- arch/arm/plat-mxc/time.c | 2 +- arch/arm/plat-omap/common.c | 4 ++-- arch/arm/plat-orion/time.c | 2 +- arch/avr32/kernel/time.c | 2 +- arch/blackfin/kernel/time-ts.c | 12 ++++++------ arch/ia64/kernel/cyclone.c | 2 +- arch/ia64/kernel/time.c | 4 ++-- arch/ia64/sn/kernel/sn2/timer.c | 2 +- arch/m68knommu/platform/68328/timers.c | 2 +- arch/m68knommu/platform/coldfire/dma_timer.c | 2 +- arch/m68knommu/platform/coldfire/pit.c | 2 +- arch/m68knommu/platform/coldfire/timers.c | 2 +- arch/mips/kernel/cevt-txx9.c | 2 +- arch/mips/kernel/csrc-bcm1480.c | 2 +- arch/mips/kernel/csrc-ioasic.c | 6 +++--- arch/mips/kernel/csrc-r4k.c | 2 +- arch/mips/kernel/csrc-sb1250.c | 2 +- arch/mips/kernel/i8253.c | 2 +- arch/mips/nxp/pnx8550/common/time.c | 2 +- arch/mips/sgi-ip27/ip27-timer.c | 2 +- arch/powerpc/kernel/time.c | 8 ++++---- arch/s390/kernel/time.c | 2 +- arch/sh/kernel/time_32.c | 2 +- arch/sh/kernel/timers/timer-tmu.c | 2 +- arch/sparc/kernel/time_64.c | 7 ++++++- arch/um/kernel/time.c | 2 +- arch/x86/kernel/hpet.c | 6 +++--- arch/x86/kernel/i8253.c | 2 +- arch/x86/kernel/kvmclock.c | 7 ++++++- arch/x86/kernel/tsc.c | 2 +- arch/x86/kernel/vmiclock_32.c | 2 +- arch/x86/lguest/boot.c | 2 +- arch/x86/xen/time.c | 7 ++++++- drivers/char/hpet.c | 2 +- drivers/clocksource/acpi_pm.c | 12 ++++++------ drivers/clocksource/cyclone.c | 2 +- drivers/clocksource/scx200_hrt.c | 2 +- drivers/clocksource/tcb_clksrc.c | 2 +- include/linux/clocksource.h | 6 +++--- kernel/time/clocksource.c | 8 ++++---- kernel/time/jiffies.c | 2 +- 54 files changed, 94 insertions(+), 79 deletions(-) diff --git a/arch/arm/mach-at91/at91rm9200_time.c b/arch/arm/mach-at91/at91rm9200_time.c index 1ff1bda0a894..309f3511aa20 100644 --- a/arch/arm/mach-at91/at91rm9200_time.c +++ b/arch/arm/mach-at91/at91rm9200_time.c @@ -85,7 +85,7 @@ static struct irqaction at91rm9200_timer_irq = { .handler = at91rm9200_timer_interrupt }; -static cycle_t read_clk32k(void) +static cycle_t read_clk32k(struct clocksource *cs) { return read_CRTR(); } diff --git a/arch/arm/mach-at91/at91sam926x_time.c b/arch/arm/mach-at91/at91sam926x_time.c index b63e1d5f1bad..4bd56aee4370 100644 --- a/arch/arm/mach-at91/at91sam926x_time.c +++ b/arch/arm/mach-at91/at91sam926x_time.c @@ -31,7 +31,7 @@ static u32 pit_cnt; /* access only w/system irq blocked */ * Clocksource: just a monotonic counter of MCK/16 cycles. * We don't care whether or not PIT irqs are enabled. */ -static cycle_t read_pit_clk(void) +static cycle_t read_pit_clk(struct clocksource *cs) { unsigned long flags; u32 elapsed; diff --git a/arch/arm/mach-davinci/time.c b/arch/arm/mach-davinci/time.c index f8bcd29d17a6..6c227d4ba998 100644 --- a/arch/arm/mach-davinci/time.c +++ b/arch/arm/mach-davinci/time.c @@ -238,7 +238,7 @@ static void __init timer_init(void) /* * clocksource */ -static cycle_t read_cycles(void) +static cycle_t read_cycles(struct clocksource *cs) { struct timer_s *t = &timers[TID_CLOCKSOURCE]; diff --git a/arch/arm/mach-imx/time.c b/arch/arm/mach-imx/time.c index aff0ebcfa847..5aef18b599e5 100644 --- a/arch/arm/mach-imx/time.c +++ b/arch/arm/mach-imx/time.c @@ -73,7 +73,7 @@ static void __init imx_timer_hardware_init(void) IMX_TCTL(TIMER_BASE) = TCTL_FRR | TCTL_CLK_PCLK1 | TCTL_TEN; } -cycle_t imx_get_cycles(void) +cycle_t imx_get_cycles(struct clocksource *cs) { return IMX_TCN(TIMER_BASE); } diff --git a/arch/arm/mach-ixp4xx/common.c b/arch/arm/mach-ixp4xx/common.c index f4656d2ac8a8..1e93dfee7543 100644 --- a/arch/arm/mach-ixp4xx/common.c +++ b/arch/arm/mach-ixp4xx/common.c @@ -401,7 +401,7 @@ void __init ixp4xx_sys_init(void) /* * clocksource */ -cycle_t ixp4xx_get_cycles(void) +cycle_t ixp4xx_get_cycles(struct clocksource *cs) { return *IXP4XX_OSTS; } diff --git a/arch/arm/mach-msm/timer.c b/arch/arm/mach-msm/timer.c index 444d9c0f5ca6..4855b8ca5101 100644 --- a/arch/arm/mach-msm/timer.c +++ b/arch/arm/mach-msm/timer.c @@ -57,12 +57,12 @@ static irqreturn_t msm_timer_interrupt(int irq, void *dev_id) return IRQ_HANDLED; } -static cycle_t msm_gpt_read(void) +static cycle_t msm_gpt_read(struct clocksource *cs) { return readl(MSM_GPT_BASE + TIMER_COUNT_VAL); } -static cycle_t msm_dgt_read(void) +static cycle_t msm_dgt_read(struct clocksource *cs) { return readl(MSM_DGT_BASE + TIMER_COUNT_VAL) >> MSM_DGT_SHIFT; } diff --git a/arch/arm/mach-netx/time.c b/arch/arm/mach-netx/time.c index f201fddb594f..82801dbf0579 100644 --- a/arch/arm/mach-netx/time.c +++ b/arch/arm/mach-netx/time.c @@ -104,7 +104,7 @@ static struct irqaction netx_timer_irq = { .handler = netx_timer_interrupt, }; -cycle_t netx_get_cycles(void) +cycle_t netx_get_cycles(struct clocksource *cs) { return readl(NETX_GPIO_COUNTER_CURRENT(TIMER_CLOCKSOURCE)); } diff --git a/arch/arm/mach-ns9xxx/time-ns9360.c b/arch/arm/mach-ns9xxx/time-ns9360.c index 41df69721769..77281260358a 100644 --- a/arch/arm/mach-ns9xxx/time-ns9360.c +++ b/arch/arm/mach-ns9xxx/time-ns9360.c @@ -25,7 +25,7 @@ #define TIMER_CLOCKEVENT 1 static u32 latch; -static cycle_t ns9360_clocksource_read(void) +static cycle_t ns9360_clocksource_read(struct clocksource *cs) { return __raw_readl(SYS_TR(TIMER_CLOCKSOURCE)); } diff --git a/arch/arm/mach-omap1/time.c b/arch/arm/mach-omap1/time.c index 495a32c287b4..4d56408d3cff 100644 --- a/arch/arm/mach-omap1/time.c +++ b/arch/arm/mach-omap1/time.c @@ -198,7 +198,7 @@ static struct irqaction omap_mpu_timer2_irq = { .handler = omap_mpu_timer2_interrupt, }; -static cycle_t mpu_read(void) +static cycle_t mpu_read(struct clocksource *cs) { return ~omap_mpu_timer_read(1); } diff --git a/arch/arm/mach-omap2/timer-gp.c b/arch/arm/mach-omap2/timer-gp.c index 9fc13a2cc3f4..1cb2c0909c2b 100644 --- a/arch/arm/mach-omap2/timer-gp.c +++ b/arch/arm/mach-omap2/timer-gp.c @@ -138,7 +138,7 @@ static inline void __init omap2_gp_clocksource_init(void) {} * clocksource */ static struct omap_dm_timer *gpt_clocksource; -static cycle_t clocksource_read_cycles(void) +static cycle_t clocksource_read_cycles(struct clocksource *cs) { return (cycle_t)omap_dm_timer_read_counter(gpt_clocksource); } diff --git a/arch/arm/mach-pxa/time.c b/arch/arm/mach-pxa/time.c index 8eb3830fbb0b..750c448db672 100644 --- a/arch/arm/mach-pxa/time.c +++ b/arch/arm/mach-pxa/time.c @@ -125,7 +125,7 @@ static struct clock_event_device ckevt_pxa_osmr0 = { .set_mode = pxa_osmr0_set_mode, }; -static cycle_t pxa_read_oscr(void) +static cycle_t pxa_read_oscr(struct clocksource *cs) { return OSCR; } diff --git a/arch/arm/mach-realview/core.c b/arch/arm/mach-realview/core.c index 9ab947c14f26..942e1a7eb9b2 100644 --- a/arch/arm/mach-realview/core.c +++ b/arch/arm/mach-realview/core.c @@ -715,7 +715,7 @@ static struct irqaction realview_timer_irq = { .handler = realview_timer_interrupt, }; -static cycle_t realview_get_cycles(void) +static cycle_t realview_get_cycles(struct clocksource *cs) { return ~readl(timer3_va_base + TIMER_VALUE); } diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c index 565776680d8c..1f929c391af7 100644 --- a/arch/arm/mach-versatile/core.c +++ b/arch/arm/mach-versatile/core.c @@ -948,7 +948,7 @@ static struct irqaction versatile_timer_irq = { .handler = versatile_timer_interrupt, }; -static cycle_t versatile_get_cycles(void) +static cycle_t versatile_get_cycles(struct clocksource *cs) { return ~readl(TIMER3_VA_BASE + TIMER_VALUE); } diff --git a/arch/arm/plat-mxc/time.c b/arch/arm/plat-mxc/time.c index ef1b3cd85bd3..dab3357196fb 100644 --- a/arch/arm/plat-mxc/time.c +++ b/arch/arm/plat-mxc/time.c @@ -36,7 +36,7 @@ static enum clock_event_mode clockevent_mode = CLOCK_EVT_MODE_UNUSED; /* clock source */ -static cycle_t mxc_get_cycles(void) +static cycle_t mxc_get_cycles(struct clocksource *cs) { return __raw_readl(TIMER_BASE + MXC_TCN); } diff --git a/arch/arm/plat-omap/common.c b/arch/arm/plat-omap/common.c index d1797147732f..433021f3d7cc 100644 --- a/arch/arm/plat-omap/common.c +++ b/arch/arm/plat-omap/common.c @@ -185,7 +185,7 @@ console_initcall(omap_add_serial_console); #include -static cycle_t omap_32k_read(void) +static cycle_t omap_32k_read(struct clocksource *cs) { return omap_readl(TIMER_32K_SYNCHRONIZED); } @@ -207,7 +207,7 @@ unsigned long long sched_clock(void) { unsigned long long ret; - ret = (unsigned long long)omap_32k_read(); + ret = (unsigned long long)omap_32k_read(&clocksource_32k); ret = (ret * clocksource_32k.mult_orig) >> clocksource_32k.shift; return ret; } diff --git a/arch/arm/plat-orion/time.c b/arch/arm/plat-orion/time.c index 6fa2923e6dca..2faf9dba4ef7 100644 --- a/arch/arm/plat-orion/time.c +++ b/arch/arm/plat-orion/time.c @@ -41,7 +41,7 @@ static u32 ticks_per_jiffy; /* * Clocksource handling. */ -static cycle_t orion_clksrc_read(void) +static cycle_t orion_clksrc_read(struct clocksource *cs) { return 0xffffffff - readl(TIMER0_VAL); } diff --git a/arch/avr32/kernel/time.c b/arch/avr32/kernel/time.c index 0ff46bf873b0..f27aa3b259fa 100644 --- a/arch/avr32/kernel/time.c +++ b/arch/avr32/kernel/time.c @@ -18,7 +18,7 @@ #include -static cycle_t read_cycle_count(void) +static cycle_t read_cycle_count(struct clocksource *cs) { return (cycle_t)sysreg_read(COUNT); } diff --git a/arch/blackfin/kernel/time-ts.c b/arch/blackfin/kernel/time-ts.c index 0ed2badfd746..27646121280a 100644 --- a/arch/blackfin/kernel/time-ts.c +++ b/arch/blackfin/kernel/time-ts.c @@ -58,16 +58,11 @@ static inline unsigned long long cycles_2_ns(cycle_t cyc) return (cyc * cyc2ns_scale) >> CYC2NS_SCALE_FACTOR; } -static cycle_t read_cycles(void) +static cycle_t read_cycles(struct clocksource *cs) { return __bfin_cycles_off + (get_cycles() << __bfin_cycles_mod); } -unsigned long long sched_clock(void) -{ - return cycles_2_ns(read_cycles()); -} - static struct clocksource clocksource_bfin = { .name = "bfin_cycles", .rating = 350, @@ -77,6 +72,11 @@ static struct clocksource clocksource_bfin = { .flags = CLOCK_SOURCE_IS_CONTINUOUS, }; +unsigned long long sched_clock(void) +{ + return cycles_2_ns(read_cycles(&clocksource_bfin)); +} + static int __init bfin_clocksource_init(void) { set_cyc2ns_scale(get_cclk() / 1000); diff --git a/arch/ia64/kernel/cyclone.c b/arch/ia64/kernel/cyclone.c index 790ef0d87e12..71e35864d2e2 100644 --- a/arch/ia64/kernel/cyclone.c +++ b/arch/ia64/kernel/cyclone.c @@ -21,7 +21,7 @@ void __init cyclone_setup(void) static void __iomem *cyclone_mc; -static cycle_t read_cyclone(void) +static cycle_t read_cyclone(struct clocksource *cs) { return (cycle_t)readq((void __iomem *)cyclone_mc); } diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c index 641c8b61c4f1..604c1a35db33 100644 --- a/arch/ia64/kernel/time.c +++ b/arch/ia64/kernel/time.c @@ -33,7 +33,7 @@ #include "fsyscall_gtod_data.h" -static cycle_t itc_get_cycles(void); +static cycle_t itc_get_cycles(struct clocksource *cs); struct fsyscall_gtod_data_t fsyscall_gtod_data = { .lock = SEQLOCK_UNLOCKED, @@ -383,7 +383,7 @@ ia64_init_itm (void) } } -static cycle_t itc_get_cycles(void) +static cycle_t itc_get_cycles(struct clocksource *cs) { u64 lcycle, now, ret; diff --git a/arch/ia64/sn/kernel/sn2/timer.c b/arch/ia64/sn/kernel/sn2/timer.c index cf67fc562054..21d6f09e3447 100644 --- a/arch/ia64/sn/kernel/sn2/timer.c +++ b/arch/ia64/sn/kernel/sn2/timer.c @@ -23,7 +23,7 @@ extern unsigned long sn_rtc_cycles_per_second; -static cycle_t read_sn2(void) +static cycle_t read_sn2(struct clocksource *cs) { return (cycle_t)readq(RTC_COUNTER_ADDR); } diff --git a/arch/m68knommu/platform/68328/timers.c b/arch/m68knommu/platform/68328/timers.c index 6bafefa546e5..309f725995bf 100644 --- a/arch/m68knommu/platform/68328/timers.c +++ b/arch/m68knommu/platform/68328/timers.c @@ -75,7 +75,7 @@ static struct irqaction m68328_timer_irq = { /***************************************************************************/ -static cycle_t m68328_read_clk(void) +static cycle_t m68328_read_clk(struct clocksource *cs) { unsigned long flags; u32 cycles; diff --git a/arch/m68knommu/platform/coldfire/dma_timer.c b/arch/m68knommu/platform/coldfire/dma_timer.c index 772578b1084f..a5f562823d7a 100644 --- a/arch/m68knommu/platform/coldfire/dma_timer.c +++ b/arch/m68knommu/platform/coldfire/dma_timer.c @@ -34,7 +34,7 @@ #define DMA_DTMR_CLK_DIV_16 (2 << 1) #define DMA_DTMR_ENABLE (1 << 0) -static cycle_t cf_dt_get_cycles(void) +static cycle_t cf_dt_get_cycles(struct clocksource *cs) { return __raw_readl(DTCN0); } diff --git a/arch/m68knommu/platform/coldfire/pit.c b/arch/m68knommu/platform/coldfire/pit.c index 2a12e7fa9748..61b96211f8ff 100644 --- a/arch/m68knommu/platform/coldfire/pit.c +++ b/arch/m68knommu/platform/coldfire/pit.c @@ -125,7 +125,7 @@ static struct irqaction pit_irq = { /***************************************************************************/ -static cycle_t pit_read_clk(void) +static cycle_t pit_read_clk(struct clocksource *cs) { unsigned long flags; u32 cycles; diff --git a/arch/m68knommu/platform/coldfire/timers.c b/arch/m68knommu/platform/coldfire/timers.c index 454f25493491..1ba8a3731653 100644 --- a/arch/m68knommu/platform/coldfire/timers.c +++ b/arch/m68knommu/platform/coldfire/timers.c @@ -78,7 +78,7 @@ static struct irqaction mcftmr_timer_irq = { /***************************************************************************/ -static cycle_t mcftmr_read_clk(void) +static cycle_t mcftmr_read_clk(struct clocksource *cs) { unsigned long flags; u32 cycles; diff --git a/arch/mips/kernel/cevt-txx9.c b/arch/mips/kernel/cevt-txx9.c index eccf7d6096bd..2e911e3da8d3 100644 --- a/arch/mips/kernel/cevt-txx9.c +++ b/arch/mips/kernel/cevt-txx9.c @@ -22,7 +22,7 @@ static struct txx9_tmr_reg __iomem *txx9_cs_tmrptr; -static cycle_t txx9_cs_read(void) +static cycle_t txx9_cs_read(struct clocksource *cs) { return __raw_readl(&txx9_cs_tmrptr->trr); } diff --git a/arch/mips/kernel/csrc-bcm1480.c b/arch/mips/kernel/csrc-bcm1480.c index 868745e7184b..51489f8a825e 100644 --- a/arch/mips/kernel/csrc-bcm1480.c +++ b/arch/mips/kernel/csrc-bcm1480.c @@ -28,7 +28,7 @@ #include -static cycle_t bcm1480_hpt_read(void) +static cycle_t bcm1480_hpt_read(struct clocksource *cs) { return (cycle_t) __raw_readq(IOADDR(A_SCD_ZBBUS_CYCLE_COUNT)); } diff --git a/arch/mips/kernel/csrc-ioasic.c b/arch/mips/kernel/csrc-ioasic.c index 1d5f63cf8997..b551f48d3a07 100644 --- a/arch/mips/kernel/csrc-ioasic.c +++ b/arch/mips/kernel/csrc-ioasic.c @@ -25,7 +25,7 @@ #include #include -static cycle_t dec_ioasic_hpt_read(void) +static cycle_t dec_ioasic_hpt_read(struct clocksource *cs) { return ioasic_read(IO_REG_FCTR); } @@ -47,13 +47,13 @@ void __init dec_ioasic_clocksource_init(void) while (!ds1287_timer_state()) ; - start = dec_ioasic_hpt_read(); + start = dec_ioasic_hpt_read(&clocksource_dec); while (i--) while (!ds1287_timer_state()) ; - end = dec_ioasic_hpt_read(); + end = dec_ioasic_hpt_read(&clocksource_dec); freq = (end - start) * 10; printk(KERN_INFO "I/O ASIC clock frequency %dHz\n", freq); diff --git a/arch/mips/kernel/csrc-r4k.c b/arch/mips/kernel/csrc-r4k.c index f1a2893931ed..e95a3cd48eea 100644 --- a/arch/mips/kernel/csrc-r4k.c +++ b/arch/mips/kernel/csrc-r4k.c @@ -10,7 +10,7 @@ #include -static cycle_t c0_hpt_read(void) +static cycle_t c0_hpt_read(struct clocksource *cs) { return read_c0_count(); } diff --git a/arch/mips/kernel/csrc-sb1250.c b/arch/mips/kernel/csrc-sb1250.c index 92212bbb8e45..d14d3d1907fa 100644 --- a/arch/mips/kernel/csrc-sb1250.c +++ b/arch/mips/kernel/csrc-sb1250.c @@ -33,7 +33,7 @@ * The HPT is free running from SB1250_HPT_VALUE down to 0 then starts over * again. */ -static cycle_t sb1250_hpt_read(void) +static cycle_t sb1250_hpt_read(struct clocksource *cs) { unsigned int count; diff --git a/arch/mips/kernel/i8253.c b/arch/mips/kernel/i8253.c index 689719e34f08..ed20e7fe65e3 100644 --- a/arch/mips/kernel/i8253.c +++ b/arch/mips/kernel/i8253.c @@ -128,7 +128,7 @@ void __init setup_pit_timer(void) * to just read by itself. So use jiffies to emulate a free * running counter: */ -static cycle_t pit_read(void) +static cycle_t pit_read(struct clocksource *cs) { unsigned long flags; int count; diff --git a/arch/mips/nxp/pnx8550/common/time.c b/arch/mips/nxp/pnx8550/common/time.c index cf293b279098..8df43e9e4d90 100644 --- a/arch/mips/nxp/pnx8550/common/time.c +++ b/arch/mips/nxp/pnx8550/common/time.c @@ -35,7 +35,7 @@ static unsigned long cpj; -static cycle_t hpt_read(void) +static cycle_t hpt_read(struct clocksource *cs) { return read_c0_count2(); } diff --git a/arch/mips/sgi-ip27/ip27-timer.c b/arch/mips/sgi-ip27/ip27-timer.c index f024057a35f8..f10a7cd64f7e 100644 --- a/arch/mips/sgi-ip27/ip27-timer.c +++ b/arch/mips/sgi-ip27/ip27-timer.c @@ -159,7 +159,7 @@ static void __init hub_rt_clock_event_global_init(void) setup_irq(irq, &hub_rt_irqaction); } -static cycle_t hub_rt_read(void) +static cycle_t hub_rt_read(struct clocksource *cs) { return REMOTE_HUB_L(cputonasid(0), PI_RT_COUNT); } diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index 926ea864e34f..48571ac56fb7 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -77,7 +77,7 @@ #include #include -static cycle_t rtc_read(void); +static cycle_t rtc_read(struct clocksource *); static struct clocksource clocksource_rtc = { .name = "rtc", .rating = 400, @@ -88,7 +88,7 @@ static struct clocksource clocksource_rtc = { .read = rtc_read, }; -static cycle_t timebase_read(void); +static cycle_t timebase_read(struct clocksource *); static struct clocksource clocksource_timebase = { .name = "timebase", .rating = 400, @@ -766,12 +766,12 @@ unsigned long read_persistent_clock(void) } /* clocksource code */ -static cycle_t rtc_read(void) +static cycle_t rtc_read(struct clocksource *cs) { return (cycle_t)get_rtc(); } -static cycle_t timebase_read(void) +static cycle_t timebase_read(struct clocksource *cs) { return (cycle_t)get_tb(); } diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index 6ded50dfa75a..ef596d020573 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -201,7 +201,7 @@ unsigned long read_persistent_clock(void) return ts.tv_sec; } -static cycle_t read_tod_clock(void) +static cycle_t read_tod_clock(struct clocksource *cs) { return get_clock(); } diff --git a/arch/sh/kernel/time_32.c b/arch/sh/kernel/time_32.c index c34e1e0f9b02..1700d2465f6c 100644 --- a/arch/sh/kernel/time_32.c +++ b/arch/sh/kernel/time_32.c @@ -208,7 +208,7 @@ unsigned long long sched_clock(void) if (!clocksource_sh.rating) return (unsigned long long)jiffies * (NSEC_PER_SEC / HZ); - cycles = clocksource_sh.read(); + cycles = clocksource_sh.read(&clocksource_sh); return cyc2ns(&clocksource_sh, cycles); } #endif diff --git a/arch/sh/kernel/timers/timer-tmu.c b/arch/sh/kernel/timers/timer-tmu.c index c5d3396f5960..fe8d8930ccb6 100644 --- a/arch/sh/kernel/timers/timer-tmu.c +++ b/arch/sh/kernel/timers/timer-tmu.c @@ -81,7 +81,7 @@ static int tmu_timer_stop(void) */ static int tmus_are_scaled; -static cycle_t tmu_timer_read(void) +static cycle_t tmu_timer_read(struct clocksource *cs) { return ((cycle_t)(~_tmu_read(TMU1)))<get_tick(); +} + void __init time_init(void) { unsigned long freq = sparc64_init_timers(); @@ -827,7 +832,7 @@ void __init time_init(void) clocksource_tick.mult = clocksource_hz2mult(freq, clocksource_tick.shift); - clocksource_tick.read = tick_ops->get_tick; + clocksource_tick.read = clocksource_tick_read; printk("clocksource: mult[%x] shift[%d]\n", clocksource_tick.mult, clocksource_tick.shift); diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c index b13a87a3ec95..c8b9c469fcd7 100644 --- a/arch/um/kernel/time.c +++ b/arch/um/kernel/time.c @@ -65,7 +65,7 @@ static irqreturn_t um_timer(int irq, void *dev) return IRQ_HANDLED; } -static cycle_t itimer_read(void) +static cycle_t itimer_read(struct clocksource *cs) { return os_nsecs() / 1000; } diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 648b3a2a3a44..3f0019e0a229 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -722,7 +722,7 @@ static int hpet_cpuhp_notify(struct notifier_block *n, /* * Clock source related code */ -static cycle_t read_hpet(void) +static cycle_t read_hpet(struct clocksource *cs) { return (cycle_t)hpet_readl(HPET_COUNTER); } @@ -756,7 +756,7 @@ static int hpet_clocksource_register(void) hpet_restart_counter(); /* Verify whether hpet counter works */ - t1 = read_hpet(); + t1 = hpet_readl(HPET_COUNTER); rdtscll(start); /* @@ -770,7 +770,7 @@ static int hpet_clocksource_register(void) rdtscll(now); } while ((now - start) < 200000UL); - if (t1 == read_hpet()) { + if (t1 == hpet_readl(HPET_COUNTER)) { printk(KERN_WARNING "HPET counter not counting. HPET disabled\n"); return -ENODEV; diff --git a/arch/x86/kernel/i8253.c b/arch/x86/kernel/i8253.c index 3475440baa54..c2e0bb0890d4 100644 --- a/arch/x86/kernel/i8253.c +++ b/arch/x86/kernel/i8253.c @@ -129,7 +129,7 @@ void __init setup_pit_timer(void) * to just read by itself. So use jiffies to emulate a free * running counter: */ -static cycle_t pit_read(void) +static cycle_t pit_read(struct clocksource *cs) { static int old_count; static u32 old_jifs; diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c index 137f2e8132df..223af43f1526 100644 --- a/arch/x86/kernel/kvmclock.c +++ b/arch/x86/kernel/kvmclock.c @@ -77,6 +77,11 @@ static cycle_t kvm_clock_read(void) return ret; } +static cycle_t kvm_clock_get_cycles(struct clocksource *cs) +{ + return kvm_clock_read(); +} + /* * If we don't do that, there is the possibility that the guest * will calibrate under heavy load - thus, getting a lower lpj - @@ -107,7 +112,7 @@ static void kvm_get_preset_lpj(void) static struct clocksource kvm_clock = { .name = "kvm-clock", - .read = kvm_clock_read, + .read = kvm_clock_get_cycles, .rating = 400, .mask = CLOCKSOURCE_MASK(64), .mult = 1 << KVM_SCALE, diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index 7a567ebe6361..d57de05dc430 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -699,7 +699,7 @@ static struct clocksource clocksource_tsc; * code, which is necessary to support wrapping clocksources like pm * timer. */ -static cycle_t read_tsc(void) +static cycle_t read_tsc(struct clocksource *cs) { cycle_t ret = (cycle_t)get_cycles(); diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c index d303369a7bad..2b3eb82efeeb 100644 --- a/arch/x86/kernel/vmiclock_32.c +++ b/arch/x86/kernel/vmiclock_32.c @@ -283,7 +283,7 @@ void __devinit vmi_time_ap_init(void) /** vmi clocksource */ static struct clocksource clocksource_vmi; -static cycle_t read_real_cycles(void) +static cycle_t read_real_cycles(struct clocksource *cs) { cycle_t ret = (cycle_t)vmi_timer_ops.get_cycle_counter(VMI_CYCLES_REAL); return max(ret, clocksource_vmi.cycle_last); diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c index a2085368a3dc..ca7ec44bafc3 100644 --- a/arch/x86/lguest/boot.c +++ b/arch/x86/lguest/boot.c @@ -663,7 +663,7 @@ static unsigned long lguest_tsc_khz(void) /* If we can't use the TSC, the kernel falls back to our lower-priority * "lguest_clock", where we read the time value given to us by the Host. */ -static cycle_t lguest_clock_read(void) +static cycle_t lguest_clock_read(struct clocksource *cs) { unsigned long sec, nsec; diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c index 14f240623497..0a5aa44299a5 100644 --- a/arch/x86/xen/time.c +++ b/arch/x86/xen/time.c @@ -213,6 +213,11 @@ cycle_t xen_clocksource_read(void) return ret; } +static cycle_t xen_clocksource_get_cycles(struct clocksource *cs) +{ + return xen_clocksource_read(); +} + static void xen_read_wallclock(struct timespec *ts) { struct shared_info *s = HYPERVISOR_shared_info; @@ -241,7 +246,7 @@ int xen_set_wallclock(unsigned long now) static struct clocksource xen_clocksource __read_mostly = { .name = "xen", .rating = 400, - .read = xen_clocksource_read, + .read = xen_clocksource_get_cycles, .mask = ~0, .mult = 1< value1) diff --git a/drivers/clocksource/cyclone.c b/drivers/clocksource/cyclone.c index 8615059a8729..64e528e8bfa6 100644 --- a/drivers/clocksource/cyclone.c +++ b/drivers/clocksource/cyclone.c @@ -19,7 +19,7 @@ int use_cyclone = 0; static void __iomem *cyclone_ptr; -static cycle_t read_cyclone(void) +static cycle_t read_cyclone(struct clocksource *cs) { return (cycle_t)readl(cyclone_ptr); } diff --git a/drivers/clocksource/scx200_hrt.c b/drivers/clocksource/scx200_hrt.c index b92da677aa5d..27f4d9637b62 100644 --- a/drivers/clocksource/scx200_hrt.c +++ b/drivers/clocksource/scx200_hrt.c @@ -43,7 +43,7 @@ MODULE_PARM_DESC(ppm, "+-adjust to actual XO freq (ppm)"); /* The base timer frequency, * 27 if selected */ #define HRT_FREQ 1000000 -static cycle_t read_hrt(void) +static cycle_t read_hrt(struct clocksource *cs) { /* Read the timer value */ return (cycle_t) inl(scx200_cb_base + SCx200_TIMER_OFFSET); diff --git a/drivers/clocksource/tcb_clksrc.c b/drivers/clocksource/tcb_clksrc.c index 254f1064d973..01b886e68822 100644 --- a/drivers/clocksource/tcb_clksrc.c +++ b/drivers/clocksource/tcb_clksrc.c @@ -39,7 +39,7 @@ static void __iomem *tcaddr; -static cycle_t tc_get_cycles(void) +static cycle_t tc_get_cycles(struct clocksource *cs) { unsigned long flags; u32 lower, upper; diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h index 573819ef4cc0..0d96cde9ee5d 100644 --- a/include/linux/clocksource.h +++ b/include/linux/clocksource.h @@ -143,7 +143,7 @@ extern u64 timecounter_cyc2time(struct timecounter *tc, * 400-499: Perfect * The ideal clocksource. A must-use where * available. - * @read: returns a cycle value + * @read: returns a cycle value, passes clocksource as argument * @mask: bitmask for two's complement * subtraction of non 64 bit counters * @mult: cycle to nanosecond multiplier (adjusted by NTP) @@ -162,7 +162,7 @@ struct clocksource { char *name; struct list_head list; int rating; - cycle_t (*read)(void); + cycle_t (*read)(struct clocksource *cs); cycle_t mask; u32 mult; u32 mult_orig; @@ -271,7 +271,7 @@ static inline u32 clocksource_hz2mult(u32 hz, u32 shift_constant) */ static inline cycle_t clocksource_read(struct clocksource *cs) { - return cs->read(); + return cs->read(cs); } /** diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index c46c931a7fe7..ecfd7b5187e0 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -181,12 +181,12 @@ static void clocksource_watchdog(unsigned long data) resumed = test_and_clear_bit(0, &watchdog_resumed); - wdnow = watchdog->read(); + wdnow = watchdog->read(watchdog); wd_nsec = cyc2ns(watchdog, (wdnow - watchdog_last) & watchdog->mask); watchdog_last = wdnow; list_for_each_entry_safe(cs, tmp, &watchdog_list, wd_list) { - csnow = cs->read(); + csnow = cs->read(cs); if (unlikely(resumed)) { cs->wd_last = csnow; @@ -247,7 +247,7 @@ static void clocksource_check_watchdog(struct clocksource *cs) list_add(&cs->wd_list, &watchdog_list); if (!started && watchdog) { - watchdog_last = watchdog->read(); + watchdog_last = watchdog->read(watchdog); watchdog_timer.expires = jiffies + WATCHDOG_INTERVAL; add_timer_on(&watchdog_timer, cpumask_first(cpu_online_mask)); @@ -268,7 +268,7 @@ static void clocksource_check_watchdog(struct clocksource *cs) cse->flags &= ~CLOCK_SOURCE_WATCHDOG; /* Start if list is not empty */ if (!list_empty(&watchdog_list)) { - watchdog_last = watchdog->read(); + watchdog_last = watchdog->read(watchdog); watchdog_timer.expires = jiffies + WATCHDOG_INTERVAL; add_timer_on(&watchdog_timer, diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c index 06f197560f3b..c3f6c30816e3 100644 --- a/kernel/time/jiffies.c +++ b/kernel/time/jiffies.c @@ -50,7 +50,7 @@ */ #define JIFFIES_SHIFT 8 -static cycle_t jiffies_read(void) +static cycle_t jiffies_read(struct clocksource *cs) { return (cycle_t) jiffies; } -- cgit v1.2.3 From 4614e6adafa2c5e6c3a9c245af2807fa7bc5117a Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Tue, 21 Apr 2009 12:24:02 -0700 Subject: clocksource: add enable() and disable() callbacks Add enable() and disable() callbacks for clocksources. This allows us to put unused clocksources in power save mode. The functions clocksource_enable() and clocksource_disable() wrap the callbacks and are inserted in the timekeeping code to enable before use and disable after switching to a new clocksource. Signed-off-by: Magnus Damm Acked-by: John Stultz Cc: Thomas Gleixner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/clocksource.h | 31 +++++++++++++++++++++++++++++++ kernel/time/timekeeping.c | 12 +++++++++--- 2 files changed, 40 insertions(+), 3 deletions(-) diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h index 0d96cde9ee5d..5a40d14daa9f 100644 --- a/include/linux/clocksource.h +++ b/include/linux/clocksource.h @@ -144,6 +144,8 @@ extern u64 timecounter_cyc2time(struct timecounter *tc, * The ideal clocksource. A must-use where * available. * @read: returns a cycle value, passes clocksource as argument + * @enable: optional function to enable the clocksource + * @disable: optional function to disable the clocksource * @mask: bitmask for two's complement * subtraction of non 64 bit counters * @mult: cycle to nanosecond multiplier (adjusted by NTP) @@ -163,6 +165,8 @@ struct clocksource { struct list_head list; int rating; cycle_t (*read)(struct clocksource *cs); + int (*enable)(struct clocksource *cs); + void (*disable)(struct clocksource *cs); cycle_t mask; u32 mult; u32 mult_orig; @@ -274,6 +278,33 @@ static inline cycle_t clocksource_read(struct clocksource *cs) return cs->read(cs); } +/** + * clocksource_enable: - enable clocksource + * @cs: pointer to clocksource + * + * Enables the specified clocksource. The clocksource callback + * function should start up the hardware and setup mult and field + * members of struct clocksource to reflect hardware capabilities. + */ +static inline int clocksource_enable(struct clocksource *cs) +{ + return cs->enable ? cs->enable(cs) : 0; +} + +/** + * clocksource_disable: - disable clocksource + * @cs: pointer to clocksource + * + * Disables the specified clocksource. The clocksource callback + * function should power down the now unused hardware block to + * save power. + */ +static inline void clocksource_disable(struct clocksource *cs) +{ + if (cs->disable) + cs->disable(cs); +} + /** * cyc2ns - converts clocksource cycles to nanoseconds * @cs: Pointer to clocksource diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 900f1b6598d1..687dff49f6e7 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -182,7 +182,7 @@ EXPORT_SYMBOL(do_settimeofday); */ static void change_clocksource(void) { - struct clocksource *new; + struct clocksource *new, *old; new = clocksource_get_next(); @@ -191,11 +191,16 @@ static void change_clocksource(void) clocksource_forward_now(); - new->raw_time = clock->raw_time; + if (clocksource_enable(new)) + return; + new->raw_time = clock->raw_time; + old = clock; clock = new; + clocksource_disable(old); + clock->cycle_last = 0; - clock->cycle_last = clocksource_read(new); + clock->cycle_last = clocksource_read(clock); clock->error = 0; clock->xtime_nsec = 0; clocksource_calculate_interval(clock, NTP_INTERVAL_LENGTH); @@ -292,6 +297,7 @@ void __init timekeeping_init(void) ntp_init(); clock = clocksource_get_next(); + clocksource_enable(clock); clocksource_calculate_interval(clock, NTP_INTERVAL_LENGTH); clock->cycle_last = clocksource_read(clock); -- cgit v1.2.3 From 8b32b5d0dca2f5ab632e8bedcd57fe4c109c13fe Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Tue, 21 Apr 2009 12:24:02 -0700 Subject: ipmi: fix platform return check The wrong return value is being tested when allocating a platform device in the IPMI SI code. Check the right value. Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/ipmi/ipmi_si_intf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index e58ea4cd55ce..2438fdf889b4 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -2863,7 +2863,7 @@ static int try_smi_init(struct smi_info *new_smi) */ new_smi->pdev = platform_device_alloc("ipmi_si", new_smi->intf_num); - if (rv) { + if (!new_smi->pdev) { printk(KERN_ERR "ipmi_si_intf:" " Unable to allocate platform device\n"); -- cgit v1.2.3 From 40112ae7504745799e75ef418057f0d2cb745050 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Tue, 21 Apr 2009 12:24:03 -0700 Subject: ipmi: test for event buffer before using The IPMI driver would attempt to use the event buffer even if that didn't exist on the BMC. This patch modified the IPMI driver to check for the event buffer's existence before trying to use it. Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/ipmi/ipmi_si_intf.c | 148 +++++++++++++++++++++++++++++++-------- include/linux/ipmi_msgdefs.h | 6 ++ 2 files changed, 125 insertions(+), 29 deletions(-) diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 2438fdf889b4..259644646b82 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -82,12 +82,6 @@ #define SI_SHORT_TIMEOUT_USEC 250 /* .25ms when the SM request a short timeout */ -/* Bit for BMC global enables. */ -#define IPMI_BMC_RCV_MSG_INTR 0x01 -#define IPMI_BMC_EVT_MSG_INTR 0x02 -#define IPMI_BMC_EVT_MSG_BUFF 0x04 -#define IPMI_BMC_SYS_LOG 0x08 - enum si_intf_state { SI_NORMAL, SI_GETTING_FLAGS, @@ -220,6 +214,9 @@ struct smi_info { OEM2_DATA_AVAIL) unsigned char msg_flags; + /* Does the BMC have an event buffer? */ + char has_event_buffer; + /* * If set to true, this will request events the next time the * state machine is idle. @@ -968,7 +965,8 @@ static void request_events(void *send_info) { struct smi_info *smi_info = send_info; - if (atomic_read(&smi_info->stop_operation)) + if (atomic_read(&smi_info->stop_operation) || + !smi_info->has_event_buffer) return; atomic_set(&smi_info->req_events, 1); @@ -2407,26 +2405,9 @@ static struct of_platform_driver ipmi_of_platform_driver = { }; #endif /* CONFIG_PPC_OF */ - -static int try_get_dev_id(struct smi_info *smi_info) +static int wait_for_msg_done(struct smi_info *smi_info) { - unsigned char msg[2]; - unsigned char *resp; - unsigned long resp_len; enum si_sm_result smi_result; - int rv = 0; - - resp = kmalloc(IPMI_MAX_MSG_LENGTH, GFP_KERNEL); - if (!resp) - return -ENOMEM; - - /* - * Do a Get Device ID command, since it comes back with some - * useful info. - */ - msg[0] = IPMI_NETFN_APP_REQUEST << 2; - msg[1] = IPMI_GET_DEVICE_ID_CMD; - smi_info->handlers->start_transaction(smi_info->si_sm, msg, 2); smi_result = smi_info->handlers->event(smi_info->si_sm, 0); for (;;) { @@ -2441,16 +2422,39 @@ static int try_get_dev_id(struct smi_info *smi_info) } else break; } - if (smi_result == SI_SM_HOSED) { + if (smi_result == SI_SM_HOSED) /* * We couldn't get the state machine to run, so whatever's at * the port is probably not an IPMI SMI interface. */ - rv = -ENODEV; + return -ENODEV; + + return 0; +} + +static int try_get_dev_id(struct smi_info *smi_info) +{ + unsigned char msg[2]; + unsigned char *resp; + unsigned long resp_len; + int rv = 0; + + resp = kmalloc(IPMI_MAX_MSG_LENGTH, GFP_KERNEL); + if (!resp) + return -ENOMEM; + + /* + * Do a Get Device ID command, since it comes back with some + * useful info. + */ + msg[0] = IPMI_NETFN_APP_REQUEST << 2; + msg[1] = IPMI_GET_DEVICE_ID_CMD; + smi_info->handlers->start_transaction(smi_info->si_sm, msg, 2); + + rv = wait_for_msg_done(smi_info); + if (rv) goto out; - } - /* Otherwise, we got some data. */ resp_len = smi_info->handlers->get_result(smi_info->si_sm, resp, IPMI_MAX_MSG_LENGTH); @@ -2462,6 +2466,88 @@ static int try_get_dev_id(struct smi_info *smi_info) return rv; } +static int try_enable_event_buffer(struct smi_info *smi_info) +{ + unsigned char msg[3]; + unsigned char *resp; + unsigned long resp_len; + int rv = 0; + + resp = kmalloc(IPMI_MAX_MSG_LENGTH, GFP_KERNEL); + if (!resp) + return -ENOMEM; + + msg[0] = IPMI_NETFN_APP_REQUEST << 2; + msg[1] = IPMI_GET_BMC_GLOBAL_ENABLES_CMD; + smi_info->handlers->start_transaction(smi_info->si_sm, msg, 2); + + rv = wait_for_msg_done(smi_info); + if (rv) { + printk(KERN_WARNING + "ipmi_si: Error getting response from get global," + " enables command, the event buffer is not" + " enabled.\n"); + goto out; + } + + resp_len = smi_info->handlers->get_result(smi_info->si_sm, + resp, IPMI_MAX_MSG_LENGTH); + + if (resp_len < 4 || + resp[0] != (IPMI_NETFN_APP_REQUEST | 1) << 2 || + resp[1] != IPMI_GET_BMC_GLOBAL_ENABLES_CMD || + resp[2] != 0) { + printk(KERN_WARNING + "ipmi_si: Invalid return from get global" + " enables command, cannot enable the event" + " buffer.\n"); + rv = -EINVAL; + goto out; + } + + if (resp[3] & IPMI_BMC_EVT_MSG_BUFF) + /* buffer is already enabled, nothing to do. */ + goto out; + + msg[0] = IPMI_NETFN_APP_REQUEST << 2; + msg[1] = IPMI_SET_BMC_GLOBAL_ENABLES_CMD; + msg[2] = resp[3] | IPMI_BMC_EVT_MSG_BUFF; + smi_info->handlers->start_transaction(smi_info->si_sm, msg, 3); + + rv = wait_for_msg_done(smi_info); + if (rv) { + printk(KERN_WARNING + "ipmi_si: Error getting response from set global," + " enables command, the event buffer is not" + " enabled.\n"); + goto out; + } + + resp_len = smi_info->handlers->get_result(smi_info->si_sm, + resp, IPMI_MAX_MSG_LENGTH); + + if (resp_len < 3 || + resp[0] != (IPMI_NETFN_APP_REQUEST | 1) << 2 || + resp[1] != IPMI_SET_BMC_GLOBAL_ENABLES_CMD) { + printk(KERN_WARNING + "ipmi_si: Invalid return from get global," + "enables command, not enable the event" + " buffer.\n"); + rv = -EINVAL; + goto out; + } + + if (resp[2] != 0) + /* + * An error when setting the event buffer bit means + * that the event buffer is not supported. + */ + rv = -ENOENT; + out: + kfree(resp); + return rv; +} + static int type_file_read_proc(char *page, char **start, off_t off, int count, int *eof, void *data) { @@ -2847,6 +2933,10 @@ static int try_smi_init(struct smi_info *new_smi) new_smi->intf_num = smi_num; smi_num++; + rv = try_enable_event_buffer(new_smi); + if (rv == 0) + new_smi->has_event_buffer = 1; + /* * Start clearing the flags before we enable interrupts or the * timer to avoid racing with the timer. diff --git a/include/linux/ipmi_msgdefs.h b/include/linux/ipmi_msgdefs.h index b56a158d587a..a079f586e907 100644 --- a/include/linux/ipmi_msgdefs.h +++ b/include/linux/ipmi_msgdefs.h @@ -58,6 +58,12 @@ #define IPMI_READ_EVENT_MSG_BUFFER_CMD 0x35 #define IPMI_GET_CHANNEL_INFO_CMD 0x42 +/* Bit for BMC global enables. */ +#define IPMI_BMC_RCV_MSG_INTR 0x01 +#define IPMI_BMC_EVT_MSG_INTR 0x02 +#define IPMI_BMC_EVT_MSG_BUFF 0x04 +#define IPMI_BMC_SYS_LOG 0x08 + #define IPMI_NETFN_STORAGE_REQUEST 0x0a #define IPMI_NETFN_STORAGE_RESPONSE 0x0b #define IPMI_ADD_SEL_ENTRY_CMD 0x44 -- cgit v1.2.3 From 25176ed670121e1e0aae5c8161713c332b786538 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Tue, 21 Apr 2009 12:24:04 -0700 Subject: ipmi: fix statistics counting issues Bela Lubkin noticed that the statistics for send IPMB and LAN commands in the IPMI driver could be incremented even if an error occurred. Move the increments to the proper place to avoid this. Also add some statistics for retransmissions that failed, and some little helper functions to neaten up the code a little. Signed-off-by: Corey Minyard Cc: Bela Lubkin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/ipmi/ipmi_msghandler.c | 73 ++++++++++++++++++++++++++----------- 1 file changed, 51 insertions(+), 22 deletions(-) diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index e93fc8d22fb2..83c7477ba801 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -285,6 +285,11 @@ enum ipmi_stat_indexes { /* Events that were received with the proper format. */ IPMI_STAT_events, + /* Retransmissions on IPMB that failed. */ + IPMI_STAT_dropped_rexmit_ipmb_commands, + + /* Retransmissions on LAN that failed. */ + IPMI_STAT_dropped_rexmit_lan_commands, /* This *must* remain last, add new values above this. */ IPMI_NUM_STATS @@ -445,6 +450,20 @@ static DEFINE_MUTEX(smi_watchers_mutex); #define ipmi_get_stat(intf, stat) \ ((unsigned int) atomic_read(&(intf)->stats[IPMI_STAT_ ## stat])) +static int is_lan_addr(struct ipmi_addr *addr) +{ + return addr->addr_type == IPMI_LAN_ADDR_TYPE; +} + +static int is_ipmb_addr(struct ipmi_addr *addr) +{ + return addr->addr_type == IPMI_IPMB_ADDR_TYPE; +} + +static int is_ipmb_bcast_addr(struct ipmi_addr *addr) +{ + return addr->addr_type == IPMI_IPMB_BROADCAST_ADDR_TYPE; +} static void free_recv_msg_list(struct list_head *q) { @@ -601,8 +620,7 @@ ipmi_addr_equal(struct ipmi_addr *addr1, struct ipmi_addr *addr2) return (smi_addr1->lun == smi_addr2->lun); } - if ((addr1->addr_type == IPMI_IPMB_ADDR_TYPE) - || (addr1->addr_type == IPMI_IPMB_BROADCAST_ADDR_TYPE)) { + if (is_ipmb_addr(addr1) || is_ipmb_bcast_addr(addr1)) { struct ipmi_ipmb_addr *ipmb_addr1 = (struct ipmi_ipmb_addr *) addr1; struct ipmi_ipmb_addr *ipmb_addr2 @@ -612,7 +630,7 @@ ipmi_addr_equal(struct ipmi_addr *addr1, struct ipmi_addr *addr2) && (ipmb_addr1->lun == ipmb_addr2->lun)); } - if (addr1->addr_type == IPMI_LAN_ADDR_TYPE) { + if (is_lan_addr(addr1)) { struct ipmi_lan_addr *lan_addr1 = (struct ipmi_lan_addr *) addr1; struct ipmi_lan_addr *lan_addr2 @@ -644,14 +662,13 @@ int ipmi_validate_addr(struct ipmi_addr *addr, int len) || (addr->channel < 0)) return -EINVAL; - if ((addr->addr_type == IPMI_IPMB_ADDR_TYPE) - || (addr->addr_type == IPMI_IPMB_BROADCAST_ADDR_TYPE)) { + if (is_ipmb_addr(addr) || is_ipmb_bcast_addr(addr)) { if (len < sizeof(struct ipmi_ipmb_addr)) return -EINVAL; return 0; } - if (addr->addr_type == IPMI_LAN_ADDR_TYPE) { + if (is_lan_addr(addr)) { if (len < sizeof(struct ipmi_lan_addr)) return -EINVAL; return 0; @@ -1503,8 +1520,7 @@ static int i_ipmi_request(ipmi_user_t user, memcpy(&(smi_msg->data[2]), msg->data, msg->data_len); smi_msg->data_size = msg->data_len + 2; ipmi_inc_stat(intf, sent_local_commands); - } else if ((addr->addr_type == IPMI_IPMB_ADDR_TYPE) - || (addr->addr_type == IPMI_IPMB_BROADCAST_ADDR_TYPE)) { + } else if (is_ipmb_addr(addr) || is_ipmb_bcast_addr(addr)) { struct ipmi_ipmb_addr *ipmb_addr; unsigned char ipmb_seq; long seqid; @@ -1583,8 +1599,6 @@ static int i_ipmi_request(ipmi_user_t user, spin_lock_irqsave(&(intf->seq_lock), flags); - ipmi_inc_stat(intf, sent_ipmb_commands); - /* * Create a sequence number with a 1 second * timeout and 4 retries. @@ -1606,6 +1620,8 @@ static int i_ipmi_request(ipmi_user_t user, goto out_err; } + ipmi_inc_stat(intf, sent_ipmb_commands); + /* * Store the sequence number in the message, * so that when the send message response @@ -1635,7 +1651,7 @@ static int i_ipmi_request(ipmi_user_t user, */ spin_unlock_irqrestore(&(intf->seq_lock), flags); } - } else if (addr->addr_type == IPMI_LAN_ADDR_TYPE) { + } else if (is_lan_addr(addr)) { struct ipmi_lan_addr *lan_addr; unsigned char ipmb_seq; long seqid; @@ -1696,8 +1712,6 @@ static int i_ipmi_request(ipmi_user_t user, spin_lock_irqsave(&(intf->seq_lock), flags); - ipmi_inc_stat(intf, sent_lan_commands); - /* * Create a sequence number with a 1 second * timeout and 4 retries. @@ -1719,6 +1733,8 @@ static int i_ipmi_request(ipmi_user_t user, goto out_err; } + ipmi_inc_stat(intf, sent_lan_commands); + /* * Store the sequence number in the message, * so that when the send message response @@ -1937,6 +1953,10 @@ static int stat_file_read_proc(char *page, char **start, off_t off, ipmi_get_stat(intf, invalid_events)); out += sprintf(out, "events: %u\n", ipmi_get_stat(intf, events)); + out += sprintf(out, "failed rexmit LAN msgs: %u\n", + ipmi_get_stat(intf, dropped_rexmit_lan_commands)); + out += sprintf(out, "failed rexmit IPMB msgs: %u\n", + ipmi_get_stat(intf, dropped_rexmit_ipmb_commands)); return (out - ((char *) page)); } @@ -3730,7 +3750,7 @@ static void check_msg_timeout(ipmi_smi_t intf, struct seq_table *ent, list_add_tail(&msg->link, timeouts); if (ent->broadcast) ipmi_inc_stat(intf, timed_out_ipmb_broadcasts); - else if (ent->recv_msg->addr.addr_type == IPMI_LAN_ADDR_TYPE) + else if (is_lan_addr(&ent->recv_msg->addr)) ipmi_inc_stat(intf, timed_out_lan_commands); else ipmi_inc_stat(intf, timed_out_ipmb_commands); @@ -3744,15 +3764,17 @@ static void check_msg_timeout(ipmi_smi_t intf, struct seq_table *ent, */ ent->timeout = MAX_MSG_TIMEOUT; ent->retries_left--; - if (ent->recv_msg->addr.addr_type == IPMI_LAN_ADDR_TYPE) - ipmi_inc_stat(intf, retransmitted_lan_commands); - else - ipmi_inc_stat(intf, retransmitted_ipmb_commands); - smi_msg = smi_from_recv_msg(intf, ent->recv_msg, slot, ent->seqid); - if (!smi_msg) + if (!smi_msg) { + if (is_lan_addr(&ent->recv_msg->addr)) + ipmi_inc_stat(intf, + dropped_rexmit_lan_commands); + else + ipmi_inc_stat(intf, + dropped_rexmit_ipmb_commands); return; + } spin_unlock_irqrestore(&intf->seq_lock, *flags); @@ -3764,10 +3786,17 @@ static void check_msg_timeout(ipmi_smi_t intf, struct seq_table *ent, * resent. */ handlers = intf->handlers; - if (handlers) + if (handlers) { + if (is_lan_addr(&ent->recv_msg->addr)) + ipmi_inc_stat(intf, + retransmitted_lan_commands); + else + ipmi_inc_stat(intf, + retransmitted_ipmb_commands); + intf->handlers->sender(intf->send_info, smi_msg, 0); - else + } else ipmi_free_smi_msg(smi_msg); spin_lock_irqsave(&intf->seq_lock, *flags); -- cgit v1.2.3 From 4dec302ff71ebf48f5784a2d2fc5e3745e6d4d52 Mon Sep 17 00:00:00 2001 From: dann frazier Date: Tue, 21 Apr 2009 12:24:05 -0700 Subject: ipmi: add oem message handling Enable userspace to receive messages that a BMC transmits using an OEM medium. This is used by the HP iLO2. Based on code originally written by Patrick Schoeller. Signed-off-by: dann frazier Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/ipmi/ipmi_msghandler.c | 138 ++++++++++++++++++++++++++++++++++-- include/linux/ipmi.h | 2 + include/linux/ipmi_msgdefs.h | 2 + 3 files changed, 137 insertions(+), 5 deletions(-) diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index 83c7477ba801..aa83a0865ec1 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -3284,6 +3284,114 @@ static int handle_lan_get_msg_cmd(ipmi_smi_t intf, return rv; } +/* + * This routine will handle "Get Message" command responses with + * channels that use an OEM Medium. The message format belongs to + * the OEM. See IPMI 2.0 specification, Chapter 6 and + * Chapter 22, sections 22.6 and 22.24 for more details. + */ +static int handle_oem_get_msg_cmd(ipmi_smi_t intf, + struct ipmi_smi_msg *msg) +{ + struct cmd_rcvr *rcvr; + int rv = 0; + unsigned char netfn; + unsigned char cmd; + unsigned char chan; + ipmi_user_t user = NULL; + struct ipmi_system_interface_addr *smi_addr; + struct ipmi_recv_msg *recv_msg; + + /* + * We expect the OEM SW to perform error checking + * so we just do some basic sanity checks + */ + if (msg->rsp_size < 4) { + /* Message not big enough, just ignore it. */ + ipmi_inc_stat(intf, invalid_commands); + return 0; + } + + if (msg->rsp[2] != 0) { + /* An error getting the response, just ignore it. */ + return 0; + } + + /* + * This is an OEM Message so the OEM needs to know how + * handle the message. We do no interpretation. + */ + netfn = msg->rsp[0] >> 2; + cmd = msg->rsp[1]; + chan = msg->rsp[3] & 0xf; + + rcu_read_lock(); + rcvr = find_cmd_rcvr(intf, netfn, cmd, chan); + if (rcvr) { + user = rcvr->user; + kref_get(&user->refcount); + } else + user = NULL; + rcu_read_unlock(); + + if (user == NULL) { + /* We didn't find a user, just give up. */ + ipmi_inc_stat(intf, unhandled_commands); + + /* + * Don't do anything with these messages, just allow + * them to be freed. + */ + + rv = 0; + } else { + /* Deliver the message to the user. */ + ipmi_inc_stat(intf, handled_commands); + + recv_msg = ipmi_alloc_recv_msg(); + if (!recv_msg) { + /* + * We couldn't allocate memory for the + * message, so requeue it for handling + * later. + */ + rv = 1; + kref_put(&user->refcount, free_user); + } else { + /* + * OEM Messages are expected to be delivered via + * the system interface to SMS software. We might + * need to visit this again depending on OEM + * requirements + */ + smi_addr = ((struct ipmi_system_interface_addr *) + &(recv_msg->addr)); + smi_addr->addr_type = IPMI_SYSTEM_INTERFACE_ADDR_TYPE; + smi_addr->channel = IPMI_BMC_CHANNEL; + smi_addr->lun = msg->rsp[0] & 3; + + recv_msg->user = user; + recv_msg->user_msg_data = NULL; + recv_msg->recv_type = IPMI_OEM_RECV_TYPE; + recv_msg->msg.netfn = msg->rsp[0] >> 2; + recv_msg->msg.cmd = msg->rsp[1]; + recv_msg->msg.data = recv_msg->msg_data; + + /* + * The message starts at byte 4 which follows the + * the Channel Byte in the "GET MESSAGE" command + */ + recv_msg->msg.data_len = msg->rsp_size - 4; + memcpy(recv_msg->msg_data, + &(msg->rsp[4]), + msg->rsp_size - 4); + deliver_response(recv_msg); + } + } + + return rv; +} + static void copy_event_into_recv_msg(struct ipmi_recv_msg *recv_msg, struct ipmi_smi_msg *msg) { @@ -3539,6 +3647,17 @@ static int handle_new_recv_msg(ipmi_smi_t intf, goto out; } + /* + ** We need to make sure the channels have been initialized. + ** The channel_handler routine will set the "curr_channel" + ** equal to or greater than IPMI_MAX_CHANNELS when all the + ** channels for this interface have been initialized. + */ + if (intf->curr_channel < IPMI_MAX_CHANNELS) { + requeue = 1; /* Just put the message back for now */ + goto out; + } + switch (intf->channels[chan].medium) { case IPMI_CHANNEL_MEDIUM_IPMB: if (msg->rsp[4] & 0x04) { @@ -3574,11 +3693,20 @@ static int handle_new_recv_msg(ipmi_smi_t intf, break; default: - /* - * We don't handle the channel type, so just - * free the message. - */ - requeue = 0; + /* Check for OEM Channels. Clients had better + register for these commands. */ + if ((intf->channels[chan].medium + >= IPMI_CHANNEL_MEDIUM_OEM_MIN) + && (intf->channels[chan].medium + <= IPMI_CHANNEL_MEDIUM_OEM_MAX)) { + requeue = handle_oem_get_msg_cmd(intf, msg); + } else { + /* + * We don't handle the channel type, so just + * free the message. + */ + requeue = 0; + } } } else if ((msg->rsp[0] == ((IPMI_NETFN_APP_REQUEST|1) << 2)) diff --git a/include/linux/ipmi.h b/include/linux/ipmi.h index 7ebdb4fb4e54..65aae34759de 100644 --- a/include/linux/ipmi.h +++ b/include/linux/ipmi.h @@ -198,6 +198,8 @@ struct kernel_ipmi_msg { response. When you send a response message, this will be returned. */ +#define IPMI_OEM_RECV_TYPE 5 /* The response for OEM Channels */ + /* Note that async events and received commands do not have a completion code as the first byte of the incoming data, unlike a response. */ diff --git a/include/linux/ipmi_msgdefs.h b/include/linux/ipmi_msgdefs.h index a079f586e907..df97e6e31e87 100644 --- a/include/linux/ipmi_msgdefs.h +++ b/include/linux/ipmi_msgdefs.h @@ -115,5 +115,7 @@ #define IPMI_CHANNEL_MEDIUM_USB1 10 #define IPMI_CHANNEL_MEDIUM_USB2 11 #define IPMI_CHANNEL_MEDIUM_SYSINTF 12 +#define IPMI_CHANNEL_MEDIUM_OEM_MIN 0x60 +#define IPMI_CHANNEL_MEDIUM_OEM_MAX 0x7f #endif /* __LINUX_IPMI_MSGDEFS_H */ -- cgit v1.2.3 From c12ddba09394c60e1120e6997794fa6ed52da884 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Tue, 21 Apr 2009 12:24:05 -0700 Subject: hugetlbfs: return negative error code for bad mount option This fixes the following BUG: # mount -o size=MM -t hugetlbfs none /huge hugetlbfs: Bad value 'MM' for mount option 'size=MM' ------------[ cut here ]------------ kernel BUG at fs/super.c:996! Due to BUG_ON(!mnt->mnt_sb); in vfs_kern_mount(). Also, remove unused #include Cc: William Irwin Cc: Signed-off-by: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hugetlbfs/inode.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 23a3c76711e0..153d9681192b 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -26,7 +26,6 @@ #include #include #include -#include #include #include #include @@ -842,7 +841,7 @@ hugetlbfs_parse_options(char *options, struct hugetlbfs_config *pconfig) bad_val: printk(KERN_ERR "hugetlbfs: Bad value '%s' for mount option '%s'\n", args[0].from, p); - return 1; + return -EINVAL; } static int -- cgit v1.2.3 From e638c1394010859a015a3b533ee452d768e62cea Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Tue, 21 Apr 2009 12:24:41 -0700 Subject: memcg: use rcu_dereference to access mm->owner mm->owner should be accessed with rcu_dereference(). Reported-by: KOSAKI Motohiro Signed-off-by: KAMEZAWA Hiroyuki Acked-by: Balbir Singh Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 18146c980b68..a9e3b76aa884 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -75,7 +75,7 @@ int mm_match_cgroup(const struct mm_struct *mm, const struct mem_cgroup *cgroup) { struct mem_cgroup *mem; rcu_read_lock(); - mem = mem_cgroup_from_task((mm)->owner); + mem = mem_cgroup_from_task(rcu_dereference((mm)->owner)); rcu_read_unlock(); return cgroup == mem; } -- cgit v1.2.3 From aa0b8f3687f06ac0e5a2b24547fdf431e923c475 Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Tue, 21 Apr 2009 12:24:42 -0700 Subject: drivers/input/serio/hp_sdc.c: fix crash when removing hp_sdc module On parisc machines, which don't have HIL, removing the hp_sdc module panics the kernel. Fix this by returning early in hp_sdc_exit() if no HP SDC controller was found. Add functionality to probe for the hp_sdc_mlc kernel module (which takes care of the upper layer HIL functionality on parisc) after two seconds. This is needed to get all the other HIL drivers (keyboard / mouse/ ..) drivers automatically loaded by udev later as well. Signed-off-by: Helge Deller Cc: Geert Uytterhoeven Cc: Frans Pop Cc: Kyle McMartin Cc: Grant Grundler Acked-by: Dmitry Torokhov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/input/serio/hp_sdc.c | 24 +++++++++++++++++++++++- 1 file changed, 23 insertions(+), 1 deletion(-) diff --git a/drivers/input/serio/hp_sdc.c b/drivers/input/serio/hp_sdc.c index bfe49243f38b..1c9410d1822c 100644 --- a/drivers/input/serio/hp_sdc.c +++ b/drivers/input/serio/hp_sdc.c @@ -819,6 +819,7 @@ static const struct parisc_device_id hp_sdc_tbl[] = { MODULE_DEVICE_TABLE(parisc, hp_sdc_tbl); static int __init hp_sdc_init_hppa(struct parisc_device *d); +static struct delayed_work moduleloader_work; static struct parisc_driver hp_sdc_driver = { .name = "hp_sdc", @@ -930,8 +931,15 @@ static int __init hp_sdc_init(void) #if defined(__hppa__) +static void request_module_delayed(struct work_struct *work) +{ + request_module("hp_sdc_mlc"); +} + static int __init hp_sdc_init_hppa(struct parisc_device *d) { + int ret; + if (!d) return 1; if (hp_sdc.dev != NULL) @@ -944,13 +952,26 @@ static int __init hp_sdc_init_hppa(struct parisc_device *d) hp_sdc.data_io = d->hpa.start + 0x800; hp_sdc.status_io = d->hpa.start + 0x801; - return hp_sdc_init(); + INIT_DELAYED_WORK(&moduleloader_work, request_module_delayed); + + ret = hp_sdc_init(); + /* after sucessfull initialization give SDC some time to settle + * and then load the hp_sdc_mlc upper layer driver */ + if (!ret) + schedule_delayed_work(&moduleloader_work, + msecs_to_jiffies(2000)); + + return ret; } #endif /* __hppa__ */ static void hp_sdc_exit(void) { + /* do nothing if we don't have a SDC */ + if (!hp_sdc.dev) + return; + write_lock_irq(&hp_sdc.lock); /* Turn off all maskable "sub-function" irq's. */ @@ -969,6 +990,7 @@ static void hp_sdc_exit(void) tasklet_kill(&hp_sdc.task); #if defined(__hppa__) + cancel_delayed_work_sync(&moduleloader_work); if (unregister_parisc_driver(&hp_sdc_driver)) printk(KERN_WARNING PREFIX "Error unregistering HP SDC"); #endif -- cgit v1.2.3 From 148da331200a0df8195e10eb8a38fd77bd7003af Mon Sep 17 00:00:00 2001 From: Daniel Ribeiro Date: Tue, 21 Apr 2009 12:24:43 -0700 Subject: pxa2xx_spi: restore DRCMR on resume If DMA is enabled, any spi_sync call after suspend/resume would block forever, because DRCMR is lost on suspend. This patch restores DRCMR to the same values set by probe. Signed-off-by: Daniel Ribeiro Signed-off-by: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/spi/pxa2xx_spi.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/spi/pxa2xx_spi.c b/drivers/spi/pxa2xx_spi.c index c76feea5fe25..1a00b415bcc4 100644 --- a/drivers/spi/pxa2xx_spi.c +++ b/drivers/spi/pxa2xx_spi.c @@ -1700,6 +1700,13 @@ static int pxa2xx_spi_resume(struct platform_device *pdev) struct ssp_device *ssp = drv_data->ssp; int status = 0; + if (drv_data->rx_channel != -1) + DRCMR(drv_data->ssp->drcmr_rx) = + DRCMR_MAPVLD | drv_data->rx_channel; + if (drv_data->tx_channel != -1) + DRCMR(drv_data->ssp->drcmr_tx) = + DRCMR_MAPVLD | drv_data->tx_channel; + /* Enable the SSP clock */ clk_enable(ssp->clk); -- cgit v1.2.3 From 3285821bc5850f5bb099e803a55b0fd884a153b8 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Tue, 21 Apr 2009 12:24:44 -0700 Subject: MAINTAINERS: remove include/asm-*/suspend* file patterns There are no more arches with suspend support using these directories. Signed-off-by: Joe Perches Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 2 -- 1 file changed, 2 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 6baace38b24d..8106118e0675 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2562,7 +2562,6 @@ F: kernel/power/ F: include/linux/suspend.h F: include/linux/freezer.h F: include/linux/pm.h -F: include/asm-*/suspend*.h F: arch/*/include/asm/suspend*.h HID CORE LAYER @@ -5404,7 +5403,6 @@ F: kernel/power/ F: include/linux/suspend.h F: include/linux/freezer.h F: include/linux/pm.h -F: include/asm-*/suspend.h SVGA HANDLING P: Martin Mares -- cgit v1.2.3 From dc8c7f8919aeb2d89247bdf302ad68b569a2b1e2 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Tue, 21 Apr 2009 12:24:45 -0700 Subject: MAINTAINERS: update KMEMTRACE pattern after file rename Signed-off-by: Joe Perches Acked-by: Pekka Enberg Acked-by: Eduard - Gabriel Munteanu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 8106118e0675..72485b64dabe 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3339,7 +3339,7 @@ P: Eduard - Gabriel Munteanu M: eduard.munteanu@linux360.ro L: linux-kernel@vger.kernel.org S: Maintained -F: Documentation/vm/kmemtrace.txt +F: Documentation/trace/kmemtrace.txt F: include/trace/kmemtrace.h F: kernel/trace/kmemtrace.c -- cgit v1.2.3 From 306c68aaa7c62010428196d309fda30b6bf57710 Mon Sep 17 00:00:00 2001 From: Roel Kluin Date: Tue, 21 Apr 2009 12:24:46 -0700 Subject: spi: pxa2xx: limit reaches -1 On line 944 the return value of flush() is considered as a boolean, but limit reaches -1 upon timeout which evaluates to true. On 540, 594, 720 the same occurs for wait_ssp_rx_stall() On 536 the same occurs for wait_dma_channel_stop() Signed-off-by: Roel Kluin Acked-by: Eric Miao Cc: David Brownell Cc: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/spi/pxa2xx_spi.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/spi/pxa2xx_spi.c b/drivers/spi/pxa2xx_spi.c index 1a00b415bcc4..885194a07418 100644 --- a/drivers/spi/pxa2xx_spi.c +++ b/drivers/spi/pxa2xx_spi.c @@ -213,7 +213,7 @@ static int flush(struct driver_data *drv_data) while (read_SSSR(reg) & SSSR_RNE) { read_SSDR(reg); } - } while ((read_SSSR(reg) & SSSR_BSY) && limit--); + } while ((read_SSSR(reg) & SSSR_BSY) && --limit); write_SSSR(SSSR_ROR, reg); return limit; @@ -484,7 +484,7 @@ static int wait_ssp_rx_stall(void const __iomem *ioaddr) { unsigned long limit = loops_per_jiffy << 1; - while ((read_SSSR(ioaddr) & SSSR_BSY) && limit--) + while ((read_SSSR(ioaddr) & SSSR_BSY) && --limit) cpu_relax(); return limit; @@ -494,7 +494,7 @@ static int wait_dma_channel_stop(int channel) { unsigned long limit = loops_per_jiffy << 1; - while (!(DCSR(channel) & DCSR_STOPSTATE) && limit--) + while (!(DCSR(channel) & DCSR_STOPSTATE) && --limit) cpu_relax(); return limit; -- cgit v1.2.3 From 0f249221813150b92c9d85b6bd232e7ee695ec6e Mon Sep 17 00:00:00 2001 From: Matt Mackall Date: Tue, 21 Apr 2009 12:24:47 -0700 Subject: MAINTAINERS: add Matt Mackall to embedded maintainers Impact: make more work for myself Signed-off-by: Matt Mackall Cc: David Woodhouse Acked-by: Paul Gortmaker Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 2 ++ 1 file changed, 2 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 72485b64dabe..8aa117bad169 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2065,6 +2065,8 @@ F: drivers/infiniband/hw/ehca/ EMBEDDED LINUX P: Paul Gortmaker M: paul.gortmaker@windriver.com +P: Matt Mackall +M: mpm@selenic.com P: David Woodhouse M: dwmw2@infradead.org L: linux-embedded@vger.kernel.org -- cgit v1.2.3 From 14fadca793e39742f442df53391cdd1437b9262f Mon Sep 17 00:00:00 2001 From: "Robert P. J. Day" Date: Tue, 21 Apr 2009 12:24:47 -0700 Subject: MAINTAINERS: add a more searchable string for the H8300 architecture. Add a parenthesized string of "H8300" for more convenient searchability in the MAINTAINERS file. Signed-off-by: Robert P. J. Day Cc: Yoshinori Sato Cc: Joe Perches Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 8aa117bad169..cb9571042f21 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5638,7 +5638,7 @@ L: uclinux-dev@uclinux.org (subscribers-only) S: Maintained F: arch/m68knommu/ -UCLINUX FOR RENESAS H8/300 +UCLINUX FOR RENESAS H8/300 (H8300) P: Yoshinori Sato M: ysato@users.sourceforge.jp W: http://uclinux-h8.sourceforge.jp/ -- cgit v1.2.3 From 6e538aaf50ae782a890cbc02c27950448d8193e1 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Tue, 21 Apr 2009 12:24:49 -0700 Subject: spi: documentation: emphasise spi_master.setup() semantics This is a doc-only patch which I hope will reduce the number of spi_master controller driver patches starting out with a common implementation bug. (As in: almost every spi_master driver I see starts out with its version of this bug. Sigh.) It just re-emphasizes that the setup() method may be called for one device while a transfer is active on another ... which means that most driver implementations shouldn't touch any registers. Signed-off-by: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/spi/spi-summary | 6 ++++++ include/linux/spi/spi.h | 7 ++++++- 2 files changed, 12 insertions(+), 1 deletion(-) diff --git a/Documentation/spi/spi-summary b/Documentation/spi/spi-summary index 0f5122eb282b..4a02d2508bc8 100644 --- a/Documentation/spi/spi-summary +++ b/Documentation/spi/spi-summary @@ -511,10 +511,16 @@ SPI MASTER METHODS This sets up the device clock rate, SPI mode, and word sizes. Drivers may change the defaults provided by board_info, and then call spi_setup(spi) to invoke this routine. It may sleep. + Unless each SPI slave has its own configuration registers, don't change them right away ... otherwise drivers could corrupt I/O that's in progress for other SPI devices. + ** BUG ALERT: for some reason the first version of + ** many spi_master drivers seems to get this wrong. + ** When you code setup(), ASSUME that the controller + ** is actively processing transfers for another device. + master->transfer(struct spi_device *spi, struct spi_message *message) This must not sleep. Its responsibility is arrange that the transfer happens and its complete() callback is issued. The two diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h index 2cc43fa380cb..a0faa18f7b1b 100644 --- a/include/linux/spi/spi.h +++ b/include/linux/spi/spi.h @@ -245,7 +245,12 @@ struct spi_master { */ u16 dma_alignment; - /* setup mode and clock, etc (spi driver may call many times) */ + /* Setup mode and clock, etc (spi driver may call many times). + * + * IMPORTANT: this may be called when transfers to another + * device are active. DO NOT UPDATE SHARED REGISTERS in ways + * which could break those transfers. + */ int (*setup)(struct spi_device *spi); /* bidirectional bulk transfers -- cgit v1.2.3 From 6d029b645175ae71fdeedea84b246ecb1362d003 Mon Sep 17 00:00:00 2001 From: Krzysztof Halasa Date: Tue, 21 Apr 2009 12:24:49 -0700 Subject: rtc-cmos: fix printk output With no IRQ available/defined, RTC-CMOS driver prints something like: rtc0: alarms up to one no, y3k, 114 bytes nvram ^^^^ I guess the following is a bit easier to understand: rtc0: no alarms, y3k, 114 bytes nvram Signed-off-by: Krzysztof Halasa Cc: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-cmos.c | 20 +++++++++----------- 1 file changed, 9 insertions(+), 11 deletions(-) diff --git a/drivers/rtc/rtc-cmos.c b/drivers/rtc/rtc-cmos.c index b6d35f50e404..23e10b6263d6 100644 --- a/drivers/rtc/rtc-cmos.c +++ b/drivers/rtc/rtc-cmos.c @@ -797,17 +797,15 @@ cmos_do_probe(struct device *dev, struct resource *ports, int rtc_irq) goto cleanup2; } - pr_info("%s: alarms up to one %s%s, %zd bytes nvram%s\n", - dev_name(&cmos_rtc.rtc->dev), - is_valid_irq(rtc_irq) - ? (cmos_rtc.mon_alrm - ? "year" - : (cmos_rtc.day_alrm - ? "month" : "day")) - : "no", - cmos_rtc.century ? ", y3k" : "", - nvram.size, - is_hpet_enabled() ? ", hpet irqs" : ""); + pr_info("%s: %s%s, %zd bytes nvram%s\n", + dev_name(&cmos_rtc.rtc->dev), + !is_valid_irq(rtc_irq) ? "no alarms" : + cmos_rtc.mon_alrm ? "alarms up to one year" : + cmos_rtc.day_alrm ? "alarms up to one month" : + "alarms up to one day", + cmos_rtc.century ? ", y3k" : "", + nvram.size, + is_hpet_enabled() ? ", hpet irqs" : ""); return 0; -- cgit v1.2.3 From 77e38a554aae2c3cdbf852117bc09bac6f95dae1 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 21 Apr 2009 12:24:51 -0700 Subject: frv: insert PCI root bus resources for the MB93090 devel motherboard Insert PCI root bus resources for the FRV-based MB93090 development kit motherboard. This is required because the CPU's window onto the PCI bus address space is considerably smaller than the CPU's full address space and non-PCI devices lie outside of the PCI window that we might want to access. Without this patch, the PCI root bus uses the platform-level bus resources, and these are then confined to the PCI window, thus making platform_device_add() reject devices outside of this window. Signed-off-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/frv/mb93090-mb00/pci-vdk.c | 63 ++++++++++++++++++++++++++++++++--------- 1 file changed, 50 insertions(+), 13 deletions(-) diff --git a/arch/frv/mb93090-mb00/pci-vdk.c b/arch/frv/mb93090-mb00/pci-vdk.c index 0f41c3a72da5..c0dcec65c6b7 100644 --- a/arch/frv/mb93090-mb00/pci-vdk.c +++ b/arch/frv/mb93090-mb00/pci-vdk.c @@ -30,6 +30,29 @@ int __nongpreldata pcibios_last_bus = -1; struct pci_bus *__nongpreldata pci_root_bus; struct pci_ops *__nongpreldata pci_root_ops; +/* + * The accessible PCI window does not cover the entire CPU address space, but + * there are devices we want to access outside of that window, so we need to + * insert specific PCI bus resources instead of using the platform-level bus + * resources directly for the PCI root bus. + * + * These are configured and inserted by pcibios_init() and are attached to the + * root bus by pcibios_fixup_bus(). + */ +static struct resource pci_ioport_resource = { + .name = "PCI IO", + .start = 0, + .end = IO_SPACE_LIMIT, + .flags = IORESOURCE_IO, +}; + +static struct resource pci_iomem_resource = { + .name = "PCI mem", + .start = 0, + .end = -1, + .flags = IORESOURCE_MEM, +}; + /* * Functions for accessing PCI configuration space */ @@ -304,6 +327,12 @@ void __init pcibios_fixup_bus(struct pci_bus *bus) #if 0 printk("### PCIBIOS_FIXUP_BUS(%d)\n",bus->number); #endif + + if (bus->number == 0) { + bus->resource[0] = &pci_ioport_resource; + bus->resource[1] = &pci_iomem_resource; + } + pci_read_bridge_bases(bus); if (bus->number == 0) { @@ -350,28 +379,36 @@ int __init pcibios_init(void) /* enable PCI arbitration */ __reg_MB86943_pci_arbiter = MB86943_PCIARB_EN; - ioport_resource.start = (__reg_MB86943_sl_pci_io_base << 9) & 0xfffffc00; - ioport_resource.end = (__reg_MB86943_sl_pci_io_range << 9) | 0x3ff; - ioport_resource.end += ioport_resource.start; + pci_ioport_resource.start = (__reg_MB86943_sl_pci_io_base << 9) & 0xfffffc00; + pci_ioport_resource.end = (__reg_MB86943_sl_pci_io_range << 9) | 0x3ff; + pci_ioport_resource.end += pci_ioport_resource.start; printk("PCI IO window: %08llx-%08llx\n", - (unsigned long long) ioport_resource.start, - (unsigned long long) ioport_resource.end); + (unsigned long long) pci_ioport_resource.start, + (unsigned long long) pci_ioport_resource.end); - iomem_resource.start = (__reg_MB86943_sl_pci_mem_base << 9) & 0xfffffc00; + pci_iomem_resource.start = (__reg_MB86943_sl_pci_mem_base << 9) & 0xfffffc00; + pci_iomem_resource.end = (__reg_MB86943_sl_pci_mem_range << 9) | 0x3ff; + pci_iomem_resource.end += pci_iomem_resource.start; - /* Reserve somewhere to write to flush posted writes. */ - iomem_resource.start += 0x400; - - iomem_resource.end = (__reg_MB86943_sl_pci_mem_range << 9) | 0x3ff; - iomem_resource.end += iomem_resource.start; + /* Reserve somewhere to write to flush posted writes. This is used by + * __flush_PCI_writes() from asm/io.h to force the write FIFO in the + * CPU-PCI bridge to flush as this doesn't happen automatically when a + * read is performed on the MB93090 development kit motherboard. + */ + pci_iomem_resource.start += 0x400; printk("PCI MEM window: %08llx-%08llx\n", - (unsigned long long) iomem_resource.start, - (unsigned long long) iomem_resource.end); + (unsigned long long) pci_iomem_resource.start, + (unsigned long long) pci_iomem_resource.end); printk("PCI DMA memory: %08lx-%08lx\n", dma_coherent_mem_start, dma_coherent_mem_end); + if (insert_resource(&iomem_resource, &pci_iomem_resource) < 0) + panic("Unable to insert PCI IOMEM resource\n"); + if (insert_resource(&ioport_resource, &pci_ioport_resource) < 0) + panic("Unable to insert PCI IOPORT resource\n"); + if (!pci_probe) return -ENXIO; -- cgit v1.2.3 From 2d7197f412e75587037449960bc4c7ea06155a7f Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Tue, 21 Apr 2009 12:24:52 -0700 Subject: uml: kill a kconfig warning Got this warning from Kconfig: boolean symbol INPUT tested for 'm'? test forced to 'n' because INPUT is tristate, not bool. Signed-off-by: WANG Cong Cc: Sam Ravnborg Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/Kconfig.rest | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/um/Kconfig.rest b/arch/um/Kconfig.rest index 7b5cea75a6c4..0ccad0ff6d6e 100644 --- a/arch/um/Kconfig.rest +++ b/arch/um/Kconfig.rest @@ -36,7 +36,7 @@ source "drivers/leds/Kconfig" #This is just to shut up some Kconfig warnings, so no prompt. config INPUT - bool + tristate default n source "arch/um/Kconfig.debug" -- cgit v1.2.3 From e873cff0fada2b0532ca5104d5b5f5b02aa442b2 Mon Sep 17 00:00:00 2001 From: Robin Holt Date: Tue, 21 Apr 2009 12:24:53 -0700 Subject: sgi-xp/sgi-gru: allow modules to load on non-uv systems For an upcoming distro release, we need to have the xp kernel module loadable even when not on UV equipment. The xpc module will not load. This will allow one set of modules dependent upon xp to work on either UV or non-UV equipment. Signed-off-by: Robin Holt Signed-off-by: Jack Steiner Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/misc/sgi-gru/grufile.c | 2 +- drivers/misc/sgi-xp/xp_main.c | 12 ++++++------ 2 files changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/misc/sgi-gru/grufile.c b/drivers/misc/sgi-gru/grufile.c index 3e6e42d2f01b..bbefe77c67a9 100644 --- a/drivers/misc/sgi-gru/grufile.c +++ b/drivers/misc/sgi-gru/grufile.c @@ -375,7 +375,7 @@ static int __init gru_init(void) void *gru_start_vaddr; if (!is_uv_system()) - return -ENODEV; + return 0; #if defined CONFIG_IA64 gru_start_paddr = 0xd000000000UL; /* ZZZZZZZZZZZZZZZZZZZ fixme */ diff --git a/drivers/misc/sgi-xp/xp_main.c b/drivers/misc/sgi-xp/xp_main.c index 16f8dcab2da4..7896849b16dc 100644 --- a/drivers/misc/sgi-xp/xp_main.c +++ b/drivers/misc/sgi-xp/xp_main.c @@ -248,19 +248,19 @@ xp_init(void) enum xp_retval ret; int ch_number; + /* initialize the connection registration mutex */ + for (ch_number = 0; ch_number < XPC_MAX_NCHANNELS; ch_number++) + mutex_init(&xpc_registrations[ch_number].mutex); + if (is_shub()) ret = xp_init_sn2(); else if (is_uv()) ret = xp_init_uv(); else - ret = xpUnsupported; + ret = 0; if (ret != xpSuccess) - return -ENODEV; - - /* initialize the connection registration mutex */ - for (ch_number = 0; ch_number < XPC_MAX_NCHANNELS; ch_number++) - mutex_init(&xpc_registrations[ch_number].mutex); + return ret; return 0; } -- cgit v1.2.3 From b298cecb3deddf76d60022473a57f1cb776cbdcd Mon Sep 17 00:00:00 2001 From: Eric Paris Date: Tue, 21 Apr 2009 12:24:54 -0700 Subject: scsi: mpt: suppress debugobjects warning Addresses http://bugzilla.kernel.org/show_bug.cgi?id=13133 ODEBUG: object is on stack, but not annotated ------------[ cut here ]------------ WARNING: at lib/debugobjects.c:253 __debug_object_init+0x1f3/0x276() Hardware name: VMware Virtual Platform Modules linked in: mptspi(+) mptscsih mptbase scsi_transport_spi ext3 jbd mbcache Pid: 540, comm: insmod Not tainted 2.6.28-mm1 #2 Call Trace: [] warn_slowpath+0x74/0x8a [] ? start_critical_timing+0x96/0xb7 [] ? _spin_unlock_irqrestore+0x2f/0x3c [] ? trace_hardirqs_off_caller+0x18/0xaf [] ? trace_hardirqs_off+0xb/0xd [] ? _spin_unlock_irqrestore+0x2f/0x3c [] ? release_console_sem+0x1a5/0x1ad [] __debug_object_init+0x1f3/0x276 [] debug_object_init+0x13/0x17 [] init_timer+0x10/0x1a [] mpt_config+0x1c1/0x2b7 [mptbase] [] ? kmalloc+0x8/0xa [mptbase] [] ? kmalloc+0x8/0xa [mptbase] [] mpt_do_ioc_recovery+0x950/0x1212 [mptbase] [] ? __lock_acquire+0xa69/0xacc [] ? _spin_unlock_irqrestore+0x36/0x3c [] ? _spin_unlock_irq+0x22/0x26 [] ? string+0x2b/0x76 [] ? vsnprintf+0x338/0x7b3 [] ? __lock_acquire+0xa69/0xacc [] ? _spin_unlock_irqrestore+0x2f/0x3c [] ? __lock_acquire+0xa69/0xacc [] ? debug_check_no_locks_freed+0xeb/0x105 [] ? _spin_unlock_irqrestore+0x36/0x3c [] ? debug_check_no_locks_freed+0x2a/0x105 [] ? lock_release_holdtime+0x43/0x48 [] ? up_read+0x16/0x29 [] ? pci_get_slot+0x66/0x72 [] mpt_attach+0x881/0x9b1 [mptbase] [] mptspi_probe+0x11/0x354 [mptspi] Noticing that every caller of mpt_config has its CONFIGPARMS struct declared on the stack and thus the &pCfg->timer is always on the stack I changed init_timer() to init_timer_on_stack() and it seems to have shut up..... Cc: "Moore, Eric Dean" Cc: James Bottomley Cc: Thomas Gleixner Acked-by: "Desai, Kashyap" Cc: [2.6.29.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/message/fusion/mptbase.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/message/fusion/mptbase.c b/drivers/message/fusion/mptbase.c index d0d126c69354..5d496a99e034 100644 --- a/drivers/message/fusion/mptbase.c +++ b/drivers/message/fusion/mptbase.c @@ -5934,7 +5934,7 @@ mpt_config(MPT_ADAPTER *ioc, CONFIGPARMS *pCfg) /* Initalize the timer */ - init_timer(&pCfg->timer); + init_timer_on_stack(&pCfg->timer); pCfg->timer.data = (unsigned long) ioc; pCfg->timer.function = mpt_timer_expired; pCfg->wait_done = 0; -- cgit v1.2.3 From 55e5750b3e979bac853c0809ad0ef75b7cebd18c Mon Sep 17 00:00:00 2001 From: Dave Jiang Date: Tue, 21 Apr 2009 12:24:56 -0700 Subject: edac: ppc mpc85xx fix mc err detect Error found by Jeff Haran. The error detect register is 0s when no errors are detected. The check code is incorrect, so reverse check sense. Reported-by: Jeff Haran Signed-off-by: Dave Jiang Signed-off-by: Doug Thompson Cc: Benjamin Herrenschmidt Acked-by: Kumar Gala Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/edac/mpc85xx_edac.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/edac/mpc85xx_edac.c b/drivers/edac/mpc85xx_edac.c index 4637a4a757df..7c8c2d72916f 100644 --- a/drivers/edac/mpc85xx_edac.c +++ b/drivers/edac/mpc85xx_edac.c @@ -674,7 +674,7 @@ static void mpc85xx_mc_check(struct mem_ctl_info *mci) int row_index; err_detect = in_be32(pdata->mc_vbase + MPC85XX_MC_ERR_DETECT); - if (err_detect) + if (!err_detect) return; mpc85xx_mc_printk(mci, KERN_ERR, "Err Detect Register: %#8.8x\n", -- cgit v1.2.3 From 2e2e425989080cc534fc0fca154cae515f971cf5 Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Tue, 21 Apr 2009 12:24:57 -0700 Subject: vmscan,memcg: reintroduce sc->may_swap Commit a6dc60f8975ad96d162915e07703a4439c80dcf0 ("vmscan: rename sc.may_swap to may_unmap") removed the may_swap flag, but memcg had used it as a flag for "we need to use swap?", as the name indicate. And in the current implementation, memcg cannot reclaim mapped file caches when mem+swap hits the limit. re-introduce may_swap flag and handle it at get_scan_ratio(). This patch doesn't influence any scan_control users other than memcg. Signed-off-by: KOSAKI Motohiro Signed-off-by: Daisuke Nishimura Acked-by: Johannes Weiner Cc: Balbir Singh Cc: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmscan.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/mm/vmscan.c b/mm/vmscan.c index 99155b7b8123..eac9577941f9 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -63,6 +63,9 @@ struct scan_control { /* Can mapped pages be reclaimed? */ int may_unmap; + /* Can pages be swapped as part of reclaim? */ + int may_swap; + /* This context's SWAP_CLUSTER_MAX. If freeing memory for * suspend, we effectively ignore SWAP_CLUSTER_MAX. * In this context, it doesn't matter that we scan the @@ -1380,7 +1383,7 @@ static void get_scan_ratio(struct zone *zone, struct scan_control *sc, struct zone_reclaim_stat *reclaim_stat = get_reclaim_stat(zone, sc); /* If we have no swap space, do not bother scanning anon pages. */ - if (nr_swap_pages <= 0) { + if (!sc->may_swap || (nr_swap_pages <= 0)) { percent[0] = 0; percent[1] = 100; return; @@ -1697,6 +1700,7 @@ unsigned long try_to_free_pages(struct zonelist *zonelist, int order, .may_writepage = !laptop_mode, .swap_cluster_max = SWAP_CLUSTER_MAX, .may_unmap = 1, + .may_swap = 1, .swappiness = vm_swappiness, .order = order, .mem_cgroup = NULL, @@ -1717,6 +1721,7 @@ unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem_cont, struct scan_control sc = { .may_writepage = !laptop_mode, .may_unmap = 1, + .may_swap = !noswap, .swap_cluster_max = SWAP_CLUSTER_MAX, .swappiness = swappiness, .order = 0, @@ -1726,9 +1731,6 @@ unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem_cont, }; struct zonelist *zonelist; - if (noswap) - sc.may_unmap = 0; - sc.gfp_mask = (gfp_mask & GFP_RECLAIM_MASK) | (GFP_HIGHUSER_MOVABLE & ~GFP_RECLAIM_MASK); zonelist = NODE_DATA(numa_node_id())->node_zonelists; @@ -1767,6 +1769,7 @@ static unsigned long balance_pgdat(pg_data_t *pgdat, int order) struct scan_control sc = { .gfp_mask = GFP_KERNEL, .may_unmap = 1, + .may_swap = 1, .swap_cluster_max = SWAP_CLUSTER_MAX, .swappiness = vm_swappiness, .order = order, @@ -2298,6 +2301,7 @@ static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) struct scan_control sc = { .may_writepage = !!(zone_reclaim_mode & RECLAIM_WRITE), .may_unmap = !!(zone_reclaim_mode & RECLAIM_SWAP), + .may_swap = 1, .swap_cluster_max = max_t(unsigned long, nr_pages, SWAP_CLUSTER_MAX), .gfp_mask = gfp_mask, -- cgit v1.2.3 From daba0280fdebce0761088cf541d6e6dd7c232850 Mon Sep 17 00:00:00 2001 From: Roel Kluin Date: Tue, 21 Apr 2009 12:24:58 -0700 Subject: bfin_5xx: misplaced parentheses `!' has a higher precedence than `&', parentheses are misplaced. Signed-off-by: Roel Kluin Cc: Alan Cox Acked-by: Sonic Zhang Cc: Bryan Wu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/serial/bfin_5xx.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/serial/bfin_5xx.c b/drivers/serial/bfin_5xx.c index 18ba812a4f84..d86123e03391 100644 --- a/drivers/serial/bfin_5xx.c +++ b/drivers/serial/bfin_5xx.c @@ -166,7 +166,7 @@ static void bfin_serial_start_tx(struct uart_port *port) struct tty_struct *tty = uart->port.info->port.tty; #ifdef CONFIG_SERIAL_BFIN_HARD_CTSRTS - if (uart->scts && (!bfin_serial_get_mctrl(&uart->port)&TIOCM_CTS)) { + if (uart->scts && !(bfin_serial_get_mctrl(&uart->port) & TIOCM_CTS)) { uart->scts = 0; uart_handle_cts_change(&uart->port, uart->scts); } @@ -368,7 +368,7 @@ static irqreturn_t bfin_serial_tx_int(int irq, void *dev_id) struct bfin_serial_port *uart = dev_id; #ifdef CONFIG_SERIAL_BFIN_HARD_CTSRTS - if (uart->scts && (!bfin_serial_get_mctrl(&uart->port)&TIOCM_CTS)) { + if (uart->scts && !(bfin_serial_get_mctrl(&uart->port) & TIOCM_CTS)) { uart->scts = 0; uart_handle_cts_change(&uart->port, uart->scts); } @@ -504,7 +504,7 @@ static irqreturn_t bfin_serial_dma_tx_int(int irq, void *dev_id) struct circ_buf *xmit = &uart->port.info->xmit; #ifdef CONFIG_SERIAL_BFIN_HARD_CTSRTS - if (uart->scts && (!bfin_serial_get_mctrl(&uart->port)&TIOCM_CTS)) { + if (uart->scts && !(bfin_serial_get_mctrl(&uart->port)&TIOCM_CTS)) { uart->scts = 0; uart_handle_cts_change(&uart->port, uart->scts); } -- cgit v1.2.3 From 9b8de7479d0dbab1ed98b5b015d44232c9d3d08e Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 21 Apr 2009 23:00:24 +0100 Subject: FRV: Fix the section attribute on UP DECLARE_PER_CPU() In non-SMP mode, the variable section attribute specified by DECLARE_PER_CPU() does not agree with that specified by DEFINE_PER_CPU(). This means that architectures that have a small data section references relative to a base register may throw up linkage errors due to too great a displacement between where the base register points and the per-CPU variable. On FRV, the .h declaration says that the variable is in the .sdata section, but the .c definition says it's actually in the .data section. The linker throws up the following errors: kernel/built-in.o: In function `release_task': kernel/exit.c:78: relocation truncated to fit: R_FRV_GPREL12 against symbol `per_cpu__process_counts' defined in .data section in kernel/built-in.o kernel/exit.c:78: relocation truncated to fit: R_FRV_GPREL12 against symbol `per_cpu__process_counts' defined in .data section in kernel/built-in.o To fix this, DECLARE_PER_CPU() should simply apply the same section attribute as does DEFINE_PER_CPU(). However, this is made slightly more complex by virtue of the fact that there are several variants on DEFINE, so these need to be matched by variants on DECLARE. Signed-off-by: David Howells Signed-off-by: Linus Torvalds --- arch/alpha/include/asm/percpu.h | 2 +- arch/ia64/include/asm/smp.h | 2 +- arch/x86/include/asm/desc.h | 2 +- arch/x86/include/asm/hardirq.h | 2 +- arch/x86/include/asm/processor.h | 6 +++--- arch/x86/include/asm/tlbflush.h | 2 +- include/asm-generic/percpu.h | 43 ++++++++++++++++++++++++++++++++++++++-- include/linux/percpu.h | 24 ---------------------- net/rds/rds.h | 2 +- 9 files changed, 50 insertions(+), 35 deletions(-) diff --git a/arch/alpha/include/asm/percpu.h b/arch/alpha/include/asm/percpu.h index 3495e8e00d70..e9e0bb5a23bf 100644 --- a/arch/alpha/include/asm/percpu.h +++ b/arch/alpha/include/asm/percpu.h @@ -73,6 +73,6 @@ extern unsigned long __per_cpu_offset[NR_CPUS]; #endif /* SMP */ -#define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu_var(name) +#include #endif /* __ALPHA_PERCPU_H */ diff --git a/arch/ia64/include/asm/smp.h b/arch/ia64/include/asm/smp.h index 598408336251..d217d1d4e051 100644 --- a/arch/ia64/include/asm/smp.h +++ b/arch/ia64/include/asm/smp.h @@ -58,7 +58,7 @@ extern struct smp_boot_data { extern char no_int_routing __devinitdata; extern cpumask_t cpu_core_map[NR_CPUS]; -DECLARE_PER_CPU(cpumask_t, cpu_sibling_map); +DECLARE_PER_CPU_SHARED_ALIGNED(cpumask_t, cpu_sibling_map); extern int smp_num_siblings; extern void __iomem *ipi_base_addr; extern unsigned char smp_int_redirect; diff --git a/arch/x86/include/asm/desc.h b/arch/x86/include/asm/desc.h index 5623c50d67b2..c45f415ce315 100644 --- a/arch/x86/include/asm/desc.h +++ b/arch/x86/include/asm/desc.h @@ -37,7 +37,7 @@ extern gate_desc idt_table[]; struct gdt_page { struct desc_struct gdt[GDT_ENTRIES]; } __attribute__((aligned(PAGE_SIZE))); -DECLARE_PER_CPU(struct gdt_page, gdt_page); +DECLARE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page); static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu) { diff --git a/arch/x86/include/asm/hardirq.h b/arch/x86/include/asm/hardirq.h index 039db6aa8e02..37555e52f980 100644 --- a/arch/x86/include/asm/hardirq.h +++ b/arch/x86/include/asm/hardirq.h @@ -26,7 +26,7 @@ typedef struct { #endif } ____cacheline_aligned irq_cpustat_t; -DECLARE_PER_CPU(irq_cpustat_t, irq_stat); +DECLARE_PER_CPU_SHARED_ALIGNED(irq_cpustat_t, irq_stat); /* We can have at most NR_VECTORS irqs routed to a cpu at a time */ #define MAX_HARDIRQS_PER_CPU NR_VECTORS diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index fcf4d92e7e04..c2cceae709c8 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -138,7 +138,7 @@ extern struct tss_struct doublefault_tss; extern __u32 cleared_cpu_caps[NCAPINTS]; #ifdef CONFIG_SMP -DECLARE_PER_CPU(struct cpuinfo_x86, cpu_info); +DECLARE_PER_CPU_SHARED_ALIGNED(struct cpuinfo_x86, cpu_info); #define cpu_data(cpu) per_cpu(cpu_info, cpu) #define current_cpu_data __get_cpu_var(cpu_info) #else @@ -270,7 +270,7 @@ struct tss_struct { } ____cacheline_aligned; -DECLARE_PER_CPU(struct tss_struct, init_tss); +DECLARE_PER_CPU_SHARED_ALIGNED(struct tss_struct, init_tss); /* * Save the original ist values for checking stack pointers during debugging @@ -393,7 +393,7 @@ union irq_stack_union { }; }; -DECLARE_PER_CPU(union irq_stack_union, irq_stack_union); +DECLARE_PER_CPU_FIRST(union irq_stack_union, irq_stack_union); DECLARE_INIT_PER_CPU(irq_stack_union); DECLARE_PER_CPU(char *, irq_stack_ptr); diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h index d3539f998f88..16a5c84b0329 100644 --- a/arch/x86/include/asm/tlbflush.h +++ b/arch/x86/include/asm/tlbflush.h @@ -152,7 +152,7 @@ struct tlb_state { struct mm_struct *active_mm; int state; }; -DECLARE_PER_CPU(struct tlb_state, cpu_tlbstate); +DECLARE_PER_CPU_SHARED_ALIGNED(struct tlb_state, cpu_tlbstate); static inline void reset_lazy_tlbstate(void) { diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h index b0e63c672ebd..af47b9e10064 100644 --- a/include/asm-generic/percpu.h +++ b/include/asm-generic/percpu.h @@ -73,11 +73,50 @@ extern void setup_per_cpu_areas(void); #endif /* SMP */ +#ifndef PER_CPU_BASE_SECTION +#ifdef CONFIG_SMP +#define PER_CPU_BASE_SECTION ".data.percpu" +#else +#define PER_CPU_BASE_SECTION ".data" +#endif +#endif + +#ifdef CONFIG_SMP + +#ifdef MODULE +#define PER_CPU_SHARED_ALIGNED_SECTION "" +#else +#define PER_CPU_SHARED_ALIGNED_SECTION ".shared_aligned" +#endif +#define PER_CPU_FIRST_SECTION ".first" + +#else + +#define PER_CPU_SHARED_ALIGNED_SECTION "" +#define PER_CPU_FIRST_SECTION "" + +#endif + #ifndef PER_CPU_ATTRIBUTES #define PER_CPU_ATTRIBUTES #endif -#define DECLARE_PER_CPU(type, name) extern PER_CPU_ATTRIBUTES \ - __typeof__(type) per_cpu_var(name) +#define DECLARE_PER_CPU_SECTION(type, name, section) \ + extern \ + __attribute__((__section__(PER_CPU_BASE_SECTION section))) \ + PER_CPU_ATTRIBUTES __typeof__(type) per_cpu__##name + +#define DECLARE_PER_CPU(type, name) \ + DECLARE_PER_CPU_SECTION(type, name, "") + +#define DECLARE_PER_CPU_SHARED_ALIGNED(type, name) \ + DECLARE_PER_CPU_SECTION(type, name, PER_CPU_SHARED_ALIGNED_SECTION) \ + ____cacheline_aligned_in_smp + +#define DECLARE_PER_CPU_PAGE_ALIGNED(type, name) \ + DECLARE_PER_CPU_SECTION(type, name, ".page_aligned") + +#define DECLARE_PER_CPU_FIRST(type, name) \ + DECLARE_PER_CPU_SECTION(type, name, PER_CPU_FIRST_SECTION) #endif /* _ASM_GENERIC_PERCPU_H_ */ diff --git a/include/linux/percpu.h b/include/linux/percpu.h index cfda2d5ad319..f052d8184993 100644 --- a/include/linux/percpu.h +++ b/include/linux/percpu.h @@ -9,30 +9,6 @@ #include -#ifndef PER_CPU_BASE_SECTION -#ifdef CONFIG_SMP -#define PER_CPU_BASE_SECTION ".data.percpu" -#else -#define PER_CPU_BASE_SECTION ".data" -#endif -#endif - -#ifdef CONFIG_SMP - -#ifdef MODULE -#define PER_CPU_SHARED_ALIGNED_SECTION "" -#else -#define PER_CPU_SHARED_ALIGNED_SECTION ".shared_aligned" -#endif -#define PER_CPU_FIRST_SECTION ".first" - -#else - -#define PER_CPU_SHARED_ALIGNED_SECTION "" -#define PER_CPU_FIRST_SECTION "" - -#endif - #define DEFINE_PER_CPU_SECTION(type, name, section) \ __attribute__((__section__(PER_CPU_BASE_SECTION section))) \ PER_CPU_ATTRIBUTES __typeof__(type) per_cpu__##name diff --git a/net/rds/rds.h b/net/rds/rds.h index 619f0a30a4e5..71794449ca4e 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -638,7 +638,7 @@ struct rds_message *rds_send_get_message(struct rds_connection *, void rds_rdma_unuse(struct rds_sock *rs, u32 r_key, int force); /* stats.c */ -DECLARE_PER_CPU(struct rds_statistics, rds_stats); +DECLARE_PER_CPU_SHARED_ALIGNED(struct rds_statistics, rds_stats); #define rds_stats_inc_which(which, member) do { \ per_cpu(which, get_cpu()).member++; \ put_cpu(); \ -- cgit v1.2.3 From 5028eaa97dd1dab9cd7c30c4d38f71c708ca64bc Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 21 Apr 2009 23:00:29 +0100 Subject: PERCPU: Collect the DECLARE/DEFINE declarations together Collect the DECLARE/DEFINE declarations together in linux/percpu-defs.h so that they're in one place, and give them descriptive comments, particularly the SHARED_ALIGNED variant. It would be nice to collect these in linux/percpu.h, but that's not possible without sorting out the severe #include recursion between the x86 arch headers and the general headers (and possibly other arches too). Signed-off-by: David Howells Signed-off-by: Linus Torvalds --- include/asm-generic/percpu.h | 26 ++------------ include/linux/percpu-defs.h | 84 ++++++++++++++++++++++++++++++++++++++++++++ include/linux/percpu.h | 20 ----------- 3 files changed, 86 insertions(+), 44 deletions(-) create mode 100644 include/linux/percpu-defs.h diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h index af47b9e10064..d7d50d7ee51e 100644 --- a/include/asm-generic/percpu.h +++ b/include/asm-generic/percpu.h @@ -1,13 +1,9 @@ #ifndef _ASM_GENERIC_PERCPU_H_ #define _ASM_GENERIC_PERCPU_H_ + #include #include - -/* - * Determine the real variable name from the name visible in the - * kernel sources. - */ -#define per_cpu_var(var) per_cpu__##var +#include #ifdef CONFIG_SMP @@ -101,22 +97,4 @@ extern void setup_per_cpu_areas(void); #define PER_CPU_ATTRIBUTES #endif -#define DECLARE_PER_CPU_SECTION(type, name, section) \ - extern \ - __attribute__((__section__(PER_CPU_BASE_SECTION section))) \ - PER_CPU_ATTRIBUTES __typeof__(type) per_cpu__##name - -#define DECLARE_PER_CPU(type, name) \ - DECLARE_PER_CPU_SECTION(type, name, "") - -#define DECLARE_PER_CPU_SHARED_ALIGNED(type, name) \ - DECLARE_PER_CPU_SECTION(type, name, PER_CPU_SHARED_ALIGNED_SECTION) \ - ____cacheline_aligned_in_smp - -#define DECLARE_PER_CPU_PAGE_ALIGNED(type, name) \ - DECLARE_PER_CPU_SECTION(type, name, ".page_aligned") - -#define DECLARE_PER_CPU_FIRST(type, name) \ - DECLARE_PER_CPU_SECTION(type, name, PER_CPU_FIRST_SECTION) - #endif /* _ASM_GENERIC_PERCPU_H_ */ diff --git a/include/linux/percpu-defs.h b/include/linux/percpu-defs.h new file mode 100644 index 000000000000..8f921d74f49f --- /dev/null +++ b/include/linux/percpu-defs.h @@ -0,0 +1,84 @@ +#ifndef _LINUX_PERCPU_DEFS_H +#define _LINUX_PERCPU_DEFS_H + +/* + * Determine the real variable name from the name visible in the + * kernel sources. + */ +#define per_cpu_var(var) per_cpu__##var + +/* + * Base implementations of per-CPU variable declarations and definitions, where + * the section in which the variable is to be placed is provided by the + * 'section' argument. This may be used to affect the parameters governing the + * variable's storage. + * + * NOTE! The sections for the DECLARE and for the DEFINE must match, lest + * linkage errors occur due the compiler generating the wrong code to access + * that section. + */ +#define DECLARE_PER_CPU_SECTION(type, name, section) \ + extern \ + __attribute__((__section__(PER_CPU_BASE_SECTION section))) \ + PER_CPU_ATTRIBUTES __typeof__(type) per_cpu__##name + +#define DEFINE_PER_CPU_SECTION(type, name, section) \ + __attribute__((__section__(PER_CPU_BASE_SECTION section))) \ + PER_CPU_ATTRIBUTES __typeof__(type) per_cpu__##name + +/* + * Variant on the per-CPU variable declaration/definition theme used for + * ordinary per-CPU variables. + */ +#define DECLARE_PER_CPU(type, name) \ + DECLARE_PER_CPU_SECTION(type, name, "") + +#define DEFINE_PER_CPU(type, name) \ + DEFINE_PER_CPU_SECTION(type, name, "") + +/* + * Declaration/definition used for per-CPU variables that must come first in + * the set of variables. + */ +#define DECLARE_PER_CPU_FIRST(type, name) \ + DECLARE_PER_CPU_SECTION(type, name, PER_CPU_FIRST_SECTION) + +#define DEFINE_PER_CPU_FIRST(type, name) \ + DEFINE_PER_CPU_SECTION(type, name, PER_CPU_FIRST_SECTION) + +/* + * Declaration/definition used for per-CPU variables that must be cacheline + * aligned under SMP conditions so that, whilst a particular instance of the + * data corresponds to a particular CPU, inefficiencies due to direct access by + * other CPUs are reduced by preventing the data from unnecessarily spanning + * cachelines. + * + * An example of this would be statistical data, where each CPU's set of data + * is updated by that CPU alone, but the data from across all CPUs is collated + * by a CPU processing a read from a proc file. + */ +#define DECLARE_PER_CPU_SHARED_ALIGNED(type, name) \ + DECLARE_PER_CPU_SECTION(type, name, PER_CPU_SHARED_ALIGNED_SECTION) \ + ____cacheline_aligned_in_smp + +#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name) \ + DEFINE_PER_CPU_SECTION(type, name, PER_CPU_SHARED_ALIGNED_SECTION) \ + ____cacheline_aligned_in_smp + +/* + * Declaration/definition used for per-CPU variables that must be page aligned. + */ +#define DECLARE_PER_CPU_PAGE_ALIGNED(type, name) \ + DECLARE_PER_CPU_SECTION(type, name, ".page_aligned") + +#define DEFINE_PER_CPU_PAGE_ALIGNED(type, name) \ + DEFINE_PER_CPU_SECTION(type, name, ".page_aligned") + +/* + * Intermodule exports for per-CPU variables. + */ +#define EXPORT_PER_CPU_SYMBOL(var) EXPORT_SYMBOL(per_cpu__##var) +#define EXPORT_PER_CPU_SYMBOL_GPL(var) EXPORT_SYMBOL_GPL(per_cpu__##var) + + +#endif /* _LINUX_PERCPU_DEFS_H */ diff --git a/include/linux/percpu.h b/include/linux/percpu.h index f052d8184993..1581ff235c7e 100644 --- a/include/linux/percpu.h +++ b/include/linux/percpu.h @@ -9,26 +9,6 @@ #include -#define DEFINE_PER_CPU_SECTION(type, name, section) \ - __attribute__((__section__(PER_CPU_BASE_SECTION section))) \ - PER_CPU_ATTRIBUTES __typeof__(type) per_cpu__##name - -#define DEFINE_PER_CPU(type, name) \ - DEFINE_PER_CPU_SECTION(type, name, "") - -#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name) \ - DEFINE_PER_CPU_SECTION(type, name, PER_CPU_SHARED_ALIGNED_SECTION) \ - ____cacheline_aligned_in_smp - -#define DEFINE_PER_CPU_PAGE_ALIGNED(type, name) \ - DEFINE_PER_CPU_SECTION(type, name, ".page_aligned") - -#define DEFINE_PER_CPU_FIRST(type, name) \ - DEFINE_PER_CPU_SECTION(type, name, PER_CPU_FIRST_SECTION) - -#define EXPORT_PER_CPU_SYMBOL(var) EXPORT_SYMBOL(per_cpu__##var) -#define EXPORT_PER_CPU_SYMBOL_GPL(var) EXPORT_SYMBOL_GPL(per_cpu__##var) - /* enough to cover all DEFINE_PER_CPUs in modules */ #ifdef CONFIG_MODULES #define PERCPU_MODULE_RESERVE (8 << 10) -- cgit v1.2.3 From 5dd559f020c98a2a4b3e063f09c0e4bc771ed838 Mon Sep 17 00:00:00 2001 From: Jonathan Corbet Date: Tue, 21 Apr 2009 16:30:32 -0600 Subject: Trivial: fix a typo in slow-work.h Fix a comment typo in slow-work.h ...a trivial mistake, but it will mess up kerneldoc if nothing else. Signed-off-by: Jonathan Corbet Signed-off-by: Linus Torvalds --- include/linux/slow-work.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/slow-work.h b/include/linux/slow-work.h index 85958277f83d..b65c8881f07a 100644 --- a/include/linux/slow-work.h +++ b/include/linux/slow-work.h @@ -67,7 +67,7 @@ static inline void slow_work_init(struct slow_work *work, } /** - * slow_work_init - Initialise a very slow work item + * vslow_work_init - Initialise a very slow work item * @work: The work item to initialise * @ops: The operations to use to handle the slow work item * -- cgit v1.2.3 From d4d5291c8cd499b1b590336059d5cc3e24c1ced6 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Tue, 21 Apr 2009 13:32:54 -0700 Subject: driver synchronization: make scsi_wait_scan more advanced There is currently only one way for userspace to say "wait for my storage device to get ready for the modules I just loaded": to load the scsi_wait_scan module. Expectations of userspace are that once this module is loaded, all the (storage) devices for which the drivers were loaded before the module load are present. Now, there are some issues with the implementation, and the async stuff got caught in the middle of this: The existing code only waits for the scsy async probing to finish, but it did not take into account at all that probing might not have begun yet. (Russell ran into this problem on his computer and the fix works for him) This patch fixes this more thoroughly than the previous "fix", which had some bad side effects (namely, for kernel code that wanted to wait for the scsi scan it would also do an async sync, which would deadlock if you did it from async context already.. there's a report about that on lkml): The patch makes the module first wait for all device driver probes, and then it will wait for the scsi parallel scan to finish. Signed-off-by: Arjan van de Ven Tested-by: Russell King Signed-off-by: Linus Torvalds --- drivers/base/dd.c | 1 + drivers/scsi/scsi_scan.c | 2 -- drivers/scsi/scsi_wait_scan.c | 11 +++++++++++ include/linux/device.h | 1 + 4 files changed, 13 insertions(+), 2 deletions(-) diff --git a/drivers/base/dd.c b/drivers/base/dd.c index f17c3266a0e0..742cbe6b042b 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -179,6 +179,7 @@ void wait_for_device_probe(void) wait_event(probe_waitqueue, atomic_read(&probe_count) == 0); async_synchronize_full(); } +EXPORT_SYMBOL_GPL(wait_for_device_probe); /** * driver_probe_device - attempt to bind device & driver together diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index a14d245a66b8..6f51ca485f35 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -180,8 +180,6 @@ int scsi_complete_async_scans(void) spin_unlock(&async_scan_lock); kfree(data); - /* Synchronize async operations globally */ - async_synchronize_full(); return 0; } diff --git a/drivers/scsi/scsi_wait_scan.c b/drivers/scsi/scsi_wait_scan.c index 2f21af21269a..74708fcaf82f 100644 --- a/drivers/scsi/scsi_wait_scan.c +++ b/drivers/scsi/scsi_wait_scan.c @@ -11,10 +11,21 @@ */ #include +#include #include static int __init wait_scan_init(void) { + /* + * First we need to wait for device probing to finish; + * the drivers we just loaded might just still be probing + * and might not yet have reached the scsi async scanning + */ + wait_for_device_probe(); + /* + * and then we wait for the actual asynchronous scsi scan + * to finish. + */ scsi_complete_async_scans(); return 0; } diff --git a/include/linux/device.h b/include/linux/device.h index 2918c0e8fdfd..6a69caaac18a 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -551,6 +551,7 @@ extern int (*platform_notify_remove)(struct device *dev); extern struct device *get_device(struct device *dev); extern void put_device(struct device *dev); +extern void wait_for_device_probe(void); /* drivers/base/power/shutdown.c */ extern void device_shutdown(void); -- cgit v1.2.3 From 091069740304c979f957ceacec39c461d0192158 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 21 Apr 2009 20:07:00 -0700 Subject: Linux 2.6.30-rc3 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 6f4208ff2d10..9e5dc8f0ef47 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 30 -EXTRAVERSION = -rc2 +EXTRAVERSION = -rc3 NAME = Temporary Tasmanian Devil # *DOCUMENTATION* -- cgit v1.2.3