On Wed, Jul 18, 2007 at 04:08:58PM -0700, Andrew Morton wrote:
> On Mon, 16 Jul 2007 15:26:50 -0700
> Ravikiran G Thirumalai <[email protected]> wrote:
>
> > Kernel warns of softlockups if the softlockup thread is not able to run
> > on a CPU for 10s. It is useful to lower the softlockup warning
> > threshold in testing environments to catch potential lockups early.
> > Following patch adds a kernel parameter 'softlockup_lim' to control
> > the softlockup threshold.
> >
>
> Why not make it tunable at runtime?
Sure! Like a sysctl?
Here's a patch that does that (On top of Ingo's
softlockup-improve-debug-output.patch)
>
> >
> > Control the trigger limit for softlockup warnings. This is useful for
> > debugging softlockups, by lowering the softlockup_lim to identify
> > possible softlockups earlier.
>
> Please check your patches with scripts/checkpatch.pl.
Yep will-do.
(checkpatch emitted one warning for the patch below, but that was because
of a 'stylo' that already exists in include/linux/sysctl.h -- which probably
needs a style change patch by itself)
---
Control the trigger limit for softlockup warnings. This is useful for
debugging softlockups, by lowering the softlockup_thresh sysctl,
to identify possible softlockups earlier.
Patch also changes the softlockup printk to print the cpu softlockup time.
Signed-off-by: Ravikiran Thirumalai <[email protected]>
Signed-off-by: Shai Fultheim <[email protected]>
Index: linux-2.6.22/kernel/softlockup.c
===================================================================
--- linux-2.6.22.orig/kernel/softlockup.c 2007-07-18 11:15:18.506614500 -0700
+++ linux-2.6.22/kernel/softlockup.c 2007-07-18 21:39:20.498592750 -0700
@@ -23,6 +23,7 @@ static DEFINE_PER_CPU(unsigned long, pri
static DEFINE_PER_CPU(struct task_struct *, watchdog_task);
static int did_panic;
+int softlockup_thresh = 10;
static int
softlock_panic(struct notifier_block *this, unsigned long event, void *ptr)
@@ -101,7 +102,7 @@ void softlockup_tick(void)
wake_up_process(per_cpu(watchdog_task, this_cpu));
/* Warn about unreasonable 10+ seconds delays: */
- if (now <= (touch_timestamp + 10))
+ if (now <= (touch_timestamp + softlockup_thresh))
return;
regs = get_irq_regs();
@@ -109,8 +110,9 @@ void softlockup_tick(void)
per_cpu(print_timestamp, this_cpu) = touch_timestamp;
spin_lock(&print_lock);
- printk(KERN_ERR "BUG: soft lockup detected on CPU#%d! [%s:%d]\n",
- this_cpu, current->comm, current->pid);
+ printk(KERN_ERR "BUG: soft lockup - CPU#%d stuck for %lus! [%s:%d]\n",
+ this_cpu, now - touch_timestamp,
+ current->comm, current->pid);
if (regs)
show_regs(regs);
else
Index: linux-2.6.22/kernel/sysctl.c
===================================================================
--- linux-2.6.22.orig/kernel/sysctl.c 2007-07-08 16:32:17.000000000 -0700
+++ linux-2.6.22/kernel/sysctl.c 2007-07-18 21:05:57.877436750 -0700
@@ -78,6 +78,7 @@ extern int percpu_pagelist_fraction;
extern int compat_log;
extern int maps_protect;
extern int sysctl_stat_interval;
+extern int softlockup_thresh;
/* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */
static int maxolduid = 65535;
@@ -206,6 +207,10 @@ static ctl_table root_table[] = {
{ .ctl_name = 0 }
};
+/* Constants for kernel table minimum and maximum */
+static int one = 1;
+static int ten = 10;
+
static ctl_table kern_table[] = {
{
.ctl_name = KERN_PANIC,
@@ -615,6 +620,19 @@ static ctl_table kern_table[] = {
.proc_handler = &proc_dointvec,
},
#endif
+#ifdef CONFIG_DETECT_SOFTLOCKUP
+ {
+ .ctl_name = KERN_SOFTLOCKUP_THRESHOLD,
+ .procname = "softlockup_thresh",
+ .data = &softlockup_thresh,
+ .maxlen = sizeof(int),
+ .mode = 0644,
+ .proc_handler = &proc_dointvec_minmax,
+ .strategy = &sysctl_intvec,
+ .extra1 = &one,
+ .extra2 = &ten,
+ },
+#endif
{ .ctl_name = 0 }
};
Index: linux-2.6.22/include/linux/sysctl.h
===================================================================
--- linux-2.6.22.orig/include/linux/sysctl.h 2007-07-08 16:32:17.000000000 -0700
+++ linux-2.6.22/include/linux/sysctl.h 2007-07-18 21:41:56.584347500 -0700
@@ -165,6 +165,7 @@ enum
KERN_MAX_LOCK_DEPTH=74,
KERN_NMI_WATCHDOG=75, /* int: enable/disable nmi watchdog */
KERN_PANIC_ON_NMI=76, /* int: whether we will panic on an unrecovered */
+ KERN_SOFTLOCKUP_THRESHOLD=77, /* int: softlockup tolerance threshold */
};
Index: linux-2.6.22/Documentation/sysctl/kernel.txt
===================================================================
--- linux-2.6.22.orig/Documentation/sysctl/kernel.txt 2007-07-08 16:32:17.000000000 -0700
+++ linux-2.6.22/Documentation/sysctl/kernel.txt 2007-07-18 22:07:29.460146250 -0700
@@ -320,6 +320,14 @@ kernel. This value defaults to SHMMAX.
==============================================================
+softlockup_thresh:
+
+This value can be used to lower the softlockup tolerance
+threshold. The default threshold is 10s. If a cpu is locked up
+for 10s, the kernel complains. Valid values are 1-10s.
+
+==============================================================
+
tainted:
Non-zero if the kernel has been tainted. Numeric values, which
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
[Index of Archives]
[Kernel Newbies]
[Netfilter]
[Bugtraq]
[Photo]
[Stuff]
[Gimp]
[Yosemite News]
[MIPS Linux]
[ARM Linux]
[Linux Security]
[Linux RAID]
[Video 4 Linux]
[Linux for the blind]
[Linux Resources]
