Attachment 150075 Details for Bug 206366 – RHEL5 fix for this issue

[patch] RHEL5 fix for this issue

softlockup.patch (text/plain), 9.03 KB, created by Prarit Bhargava on 2007-03-14 18:45:40 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Prarit Bhargava

Created: 2007-03-14 18:45:40 UTC

Size: 9.03 KB

patch

obsolete

>There are some situations when soft lockup warnings are expected in the
>kernel.  For example, when doing an alt-sysrq-t on a large number of processes,
>the dump to console can take a long time and spinlocks may be held over that
>period.  This results in a bogus soft lockup warning.
>
>This patch reworks touch_softlockup_watchdog to touch ALL cpu's
>touch_timestamp.  It also introduces touch_cpu_softlockup_watchdog to touch
>a single cpu's touch_timestamp.  This makes it functionally equivalent to
>touch_nmi_watchdog.
>
>touch_nmi_watchdog is not modified -- AFAICT, it was attempting to touch all
>cpu's softlockup watchdogs, not just a specific cpu.
>
>/drivers/ide/ide-iops.c does not need to call touch_softlockup_watchdog as it
>is done in the call to touch_nmi_watchdog.
>
>diff -urNp -X linux-2.6.18.ia64/Documentation/dontdiff linux-2.6.18.ia64.orig/arch/i386/kernel/nmi.c linux-2.6.18.ia64/arch/i386/kernel/nmi.c
>--- linux-2.6.18.ia64.orig/arch/i386/kernel/nmi.c	2007-03-12 05:47:49.000000000 -0400
>+++ linux-2.6.18.ia64/arch/i386/kernel/nmi.c	2007-03-14 13:19:34.000000000 -0400
>@@ -566,7 +566,7 @@ void touch_nmi_watchdog (void)
> 	 * Just reset the alert counters, (other CPUs might be
> 	 * spinning on locks we hold):
> 	 */
>-	for_each_possible_cpu(i)
>+	for_each_online_cpu(i)
> 		alert_counter[i] = 0;
> 
> 	/*
>diff -urNp -X linux-2.6.18.ia64/Documentation/dontdiff linux-2.6.18.ia64.orig/arch/i386/kernel/time-xen.c linux-2.6.18.ia64/arch/i386/kernel/time-xen.c
>--- linux-2.6.18.ia64.orig/arch/i386/kernel/time-xen.c	2007-03-12 05:47:49.000000000 -0400
>+++ linux-2.6.18.ia64/arch/i386/kernel/time-xen.c	2007-03-14 10:06:04.000000000 -0400
>@@ -711,7 +711,7 @@ irqreturn_t timer_interrupt(int irq, voi
> 
> 	/* Offlined for more than a few seconds? Avoid lockup warnings. */
> 	if (stolen > 5*HZ)
>-		touch_softlockup_watchdog();
>+		touch_cpu_softlockup_watchdog();
> 
> 	/* Local timer processing (see update_process_times()). */
> 	run_local_timers();
>diff -urNp -X linux-2.6.18.ia64/Documentation/dontdiff linux-2.6.18.ia64.orig/arch/ia64/kernel/uncached.c linux-2.6.18.ia64/arch/ia64/kernel/uncached.c
>--- linux-2.6.18.ia64.orig/arch/ia64/kernel/uncached.c	2007-03-12 05:47:48.000000000 -0400
>+++ linux-2.6.18.ia64/arch/ia64/kernel/uncached.c	2007-03-13 18:17:46.000000000 -0400
>@@ -254,7 +254,7 @@ static int __init uncached_build_memmap(
> 	struct gen_pool *pool = uncached_pools[nid].pool;
> 	size_t size = uc_end - uc_start;
> 
>-	touch_softlockup_watchdog();
>+	touch_cpu_softlockup_watchdog();
> 
> 	if (pool != NULL) {
> 		memset((char *)uc_start, 0, size);
>diff -urNp -X linux-2.6.18.ia64/Documentation/dontdiff linux-2.6.18.ia64.orig/drivers/ide/ide-iops.c linux-2.6.18.ia64/drivers/ide/ide-iops.c
>--- linux-2.6.18.ia64.orig/drivers/ide/ide-iops.c	2007-03-12 05:47:51.000000000 -0400
>+++ linux-2.6.18.ia64/drivers/ide/ide-iops.c	2007-03-14 09:27:41.000000000 -0400
>@@ -1243,7 +1243,6 @@ int ide_wait_not_busy(ide_hwif_t *hwif, 
> 		 */
> 		if (stat == 0xff)
> 			return -ENODEV;
>-		touch_softlockup_watchdog();
> 		touch_nmi_watchdog();
> 	}
> 	return -EBUSY;
>diff -urNp -X linux-2.6.18.ia64/Documentation/dontdiff linux-2.6.18.ia64.orig/drivers/ide/ide-taskfile.c linux-2.6.18.ia64/drivers/ide/ide-taskfile.c
>--- linux-2.6.18.ia64.orig/drivers/ide/ide-taskfile.c	2007-03-12 05:47:51.000000000 -0400
>+++ linux-2.6.18.ia64/drivers/ide/ide-taskfile.c	2007-03-13 18:18:04.000000000 -0400
>@@ -313,7 +313,7 @@ static void ide_pio_datablock(ide_drive_
> 	if (rq->bio)	/* fs request */
> 		rq->errors = 0;
> 
>-	touch_softlockup_watchdog();
>+	touch_cpu_softlockup_watchdog();
> 
> 	switch (drive->hwif->data_phase) {
> 	case TASKFILE_MULTI_IN:
>diff -urNp -X linux-2.6.18.ia64/Documentation/dontdiff linux-2.6.18.ia64.orig/drivers/mtd/nand/nand_base.c linux-2.6.18.ia64/drivers/mtd/nand/nand_base.c
>--- linux-2.6.18.ia64.orig/drivers/mtd/nand/nand_base.c	2007-03-12 05:47:50.000000000 -0400
>+++ linux-2.6.18.ia64/drivers/mtd/nand/nand_base.c	2007-03-13 18:18:14.000000000 -0400
>@@ -425,7 +425,7 @@ static void nand_wait_ready(struct mtd_i
> 	do {
> 		if (chip->dev_ready(mtd))
> 			break;
>-		touch_softlockup_watchdog();
>+		touch_cpu_softlockup_watchdog();
> 	} while (time_before(jiffies, timeo));
> 	led_trigger_event(nand_led_trigger, LED_OFF);
> }
>diff -urNp -X linux-2.6.18.ia64/Documentation/dontdiff linux-2.6.18.ia64.orig/drivers/mtd/onenand/onenand_base.c linux-2.6.18.ia64/drivers/mtd/onenand/onenand_base.c
>--- linux-2.6.18.ia64.orig/drivers/mtd/onenand/onenand_base.c	2007-03-12 05:47:50.000000000 -0400
>+++ linux-2.6.18.ia64/drivers/mtd/onenand/onenand_base.c	2007-03-13 18:18:20.000000000 -0400
>@@ -308,7 +308,7 @@ static int onenand_wait(struct mtd_info 
> 
> 		if (state != FL_READING)
> 			cond_resched();
>-		touch_softlockup_watchdog();
>+		touch_cpu_softlockup_watchdog();
> 	}
> 	/* To get correct interrupt status in timeout case */
> 	interrupt = this->read_word(this->base + ONENAND_REG_INTERRUPT);
>diff -urNp -X linux-2.6.18.ia64/Documentation/dontdiff linux-2.6.18.ia64.orig/drivers/xen/core/smpboot.c linux-2.6.18.ia64/drivers/xen/core/smpboot.c
>--- linux-2.6.18.ia64.orig/drivers/xen/core/smpboot.c	2007-03-12 05:47:50.000000000 -0400
>+++ linux-2.6.18.ia64/drivers/xen/core/smpboot.c	2007-03-13 18:18:32.000000000 -0400
>@@ -148,7 +148,7 @@ static void xen_smp_intr_exit(unsigned i
> void cpu_bringup(void)
> {
> 	cpu_init();
>-	touch_softlockup_watchdog();
>+	touch_cpu_softlockup_watchdog();
> 	preempt_disable();
> 	local_irq_enable();
> }
>diff -urNp -X linux-2.6.18.ia64/Documentation/dontdiff linux-2.6.18.ia64.orig/include/linux/sched.h linux-2.6.18.ia64/include/linux/sched.h
>--- linux-2.6.18.ia64.orig/include/linux/sched.h	2007-03-12 05:47:49.000000000 -0400
>+++ linux-2.6.18.ia64/include/linux/sched.h	2007-03-13 18:17:38.000000000 -0400
>@@ -215,6 +215,7 @@ extern void scheduler_tick(void);
> #ifdef CONFIG_DETECT_SOFTLOCKUP
> extern void softlockup_tick(void);
> extern void spawn_softlockup_task(void);
>+extern void touch_cpu_softlockup_watchdog(void);
> extern void touch_softlockup_watchdog(void);
> #else
> static inline void softlockup_tick(void)
>@@ -223,6 +224,9 @@ static inline void softlockup_tick(void)
> static inline void spawn_softlockup_task(void)
> {
> }
>+static inline void touch_cpu_softlockup_watchdog(void)
>+{
>+}
> static inline void touch_softlockup_watchdog(void)
> {
> }
>diff -urNp -X linux-2.6.18.ia64/Documentation/dontdiff linux-2.6.18.ia64.orig/kernel/panic.c linux-2.6.18.ia64/kernel/panic.c
>--- linux-2.6.18.ia64.orig/kernel/panic.c	2007-03-12 05:47:49.000000000 -0400
>+++ linux-2.6.18.ia64/kernel/panic.c	2007-03-13 18:18:42.000000000 -0400
>@@ -132,7 +132,7 @@ NORET_TYPE void panic(const char * fmt, 
> #endif
> 	local_irq_enable();
> 	for (i = 0;;) {
>-		touch_softlockup_watchdog();
>+		touch_cpu_softlockup_watchdog();
> 		i += panic_blink(i);
> 		mdelay(1);
> 		i++;
>diff -urNp -X linux-2.6.18.ia64/Documentation/dontdiff linux-2.6.18.ia64.orig/kernel/power/swsusp.c linux-2.6.18.ia64/kernel/power/swsusp.c
>--- linux-2.6.18.ia64.orig/kernel/power/swsusp.c	2007-03-12 05:47:49.000000000 -0400
>+++ linux-2.6.18.ia64/kernel/power/swsusp.c	2007-03-13 18:18:49.000000000 -0400
>@@ -274,7 +274,7 @@ int swsusp_resume(void)
> 	swsusp_free();
> 	restore_processor_state();
> 	restore_highmem();
>-	touch_softlockup_watchdog();
>+	touch_cpu_softlockup_watchdog();
> 	device_power_up();
> 	local_irq_enable();
> 	return error;
>diff -urNp -X linux-2.6.18.ia64/Documentation/dontdiff linux-2.6.18.ia64.orig/kernel/sched.c linux-2.6.18.ia64/kernel/sched.c
>--- linux-2.6.18.ia64.orig/kernel/sched.c	2007-03-12 05:47:49.000000000 -0400
>+++ linux-2.6.18.ia64/kernel/sched.c	2007-03-14 10:03:17.000000000 -0400
>@@ -4802,6 +4802,7 @@ void show_state(void)
> 		touch_nmi_watchdog();
> 		show_task(p);
> 	} while_each_thread(g, p);
>+	touch_softlockup_watchdog();
> 
> 	read_unlock(&tasklist_lock);
> 	debug_show_all_locks();
>diff -urNp -X linux-2.6.18.ia64/Documentation/dontdiff linux-2.6.18.ia64.orig/kernel/softlockup.c linux-2.6.18.ia64/kernel/softlockup.c
>--- linux-2.6.18.ia64.orig/kernel/softlockup.c	2007-03-12 05:47:49.000000000 -0400
>+++ linux-2.6.18.ia64/kernel/softlockup.c	2007-03-14 13:27:11.000000000 -0400
>@@ -8,6 +8,7 @@
>  */
> #include <linux/mm.h>
> #include <linux/cpu.h>
>+#include <linux/cpumask.h>
> #include <linux/init.h>
> #include <linux/delay.h>
> #include <linux/kthread.h>
>@@ -34,11 +35,19 @@ static struct notifier_block panic_block
> 	.notifier_call = softlock_panic,
> };
> 
>-void touch_softlockup_watchdog(void)
>+void touch_cpu_softlockup_watchdog(void)
> {
> 	__raw_get_cpu_var(touch_timestamp) = jiffies;
> }
>-EXPORT_SYMBOL(touch_softlockup_watchdog);
>+EXPORT_SYMBOL(touch_cpu_softlockup_watchdog);
>+
>+void touch_softlockup_watchdog(void)
>+{
>+	int cpu;
>+
>+	for_each_online_cpu(cpu)
>+		per_cpu(touch_timestamp, cpu) = jiffies;
>+}
> 
> /*
>  * This callback runs from the timer interrupt, and checks
>@@ -57,7 +66,7 @@ void softlockup_tick(void)
> 
> 	/* do not print during early bootup: */
> 	if (unlikely(system_state != SYSTEM_RUNNING)) {
>-		touch_softlockup_watchdog();
>+		touch_cpu_softlockup_watchdog();
> 		return;
> 	}
> 
>@@ -94,7 +103,7 @@ static int watchdog(void * __bind_cpu)
> 	 */
> 	while (!kthread_should_stop()) {
> 		set_current_state(TASK_INTERRUPTIBLE);
>-		touch_softlockup_watchdog();
>+		touch_cpu_softlockup_watchdog();
> 		schedule();
> 	}
>

Actions: View | Diff

Attachments on bug 206366: 136213 | 149958 | 149959 | 149961 | 149963 | 150074 | 150075 | 151651 | 161096 | 161097 | 161098 | 161099