Skip to contentSkip to conversation

watchdog: Add sysctl knob hardlockup_panic

PATCH Superseded Don Zickus 3 participants 3 messages Oct 15, 2015

Subsystems (3)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore

Don Zickus

author
#0 lore raw

PATCH 1/1 watchdog: Add sysctl knob hardlockup_panic

The only way to enable a hardlockup to panic the machine is to set 'nmi_watchdog=panic' on the kernel command line.

This makes it awkward for end users and folks who want to run automate tests (like myself).

Mimic the softlockup_panic knob and create a /proc/sys/kernel/hardlockup_panic knob.

---
 Documentation/lockup-watchdogs.txt | 5 +++--
 include/linux/sched.h              | 1 +
 kernel/sysctl.c                    | 9 +++++++++
 kernel/watchdog.c                  | 2 +-
 4 files changed, 14 insertions(+), 3 deletions(-)
Documentation/lockup-watchdogs.txt +3 -2
index 22dd6af..4a6e33e 100644
@@ -20,8 +20,9 @@ kernel mode for more than 10 seconds (see "Implementation" below for
 details), without letting other interrupts have a chance to run.
 Similarly to the softlockup case, the current stack trace is displayed
 upon detection and the system will stay locked up unless the default
-behavior is changed, which can be done through a compile time knob,
-"BOOTPARAM_HARDLOCKUP_PANIC", and a kernel parameter, "nmi_watchdog"
+behavior is changed, which can be done through a sysctl,
+'hardlockup_panic', a compile time knob, "BOOTPARAM_HARDLOCKUP_PANIC",
+and a kernel parameter, "nmi_watchdog"
 (see "Documentation/kernel-parameters.txt" for details).
 
 The panic option can be used in combination with panic_timeout (this
include/linux/sched.h +1 -0
index b7b9501..5e65b14 100644
@@ -384,6 +384,7 @@ extern int proc_dowatchdog_thresh(struct ctl_table *table, int write,
 				  void __user *buffer,
 				  size_t *lenp, loff_t *ppos);
 extern unsigned int  softlockup_panic;
+extern unsigned int  hardlockup_panic;
 void lockup_detector_init(void);
 #else
 static inline void touch_softlockup_watchdog(void)
kernel/sysctl.c +9 -0
index efb0370..3e6a742 100644
@@ -887,6 +887,15 @@ static struct ctl_table kern_table[] = {
 		.extra1		= &zero,
 		.extra2		= &one,
 	},
+	{
+		.procname	= "hardlockup_panic",
+		.data		= &hardlockup_panic,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &zero,
+		.extra2		= &one,
+	},
 #ifdef CONFIG_SMP
 	{
 		.procname	= "softlockup_all_cpu_backtrace",
kernel/watchdog.c +1 -1
index f6b32b8..0a23125 100644
@@ -112,7 +112,7 @@ static unsigned long soft_lockup_nmi_warn;
  * Should we panic when a soft-lockup or hard-lockup occurs:
  */
 #ifdef CONFIG_HARDLOCKUP_DETECTOR
-static int hardlockup_panic =
+unsigned int __read_mostly hardlockup_panic =
 			CONFIG_BOOTPARAM_HARDLOCKUP_PANIC_VALUE;
 static unsigned long hardlockup_allcpu_dumped;
 /*
Signature · 1 line
1.8.4
#1 lore raw
reproduce: wget https://git.kernel.org/cgit/linux/kernel/git/wfg/lkp-tests.git/plain/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross # save the attached .config to linux build tree…

Hi Don,

[auto build test ERROR on next-20151013 -- if it's inappropriate base, please suggest rules for selecting the more suitable base]

url:    https://github.com/0day-ci/linux/commits/Don-Zickus/watchdog-Add-sysctl-knob-hardlockup_panic/20151015-104506
config: sparc64-defconfig (attached as .config)

reproduce:
wget https://git.kernel.org/cgit/linux/kernel/git/wfg/lkp-tests.git/plain/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross
# save the attached .config to linux build tree
make.cross ARCH=sparc64

All errors (new ones prefixed by >>):

>> kernel/built-in.o:(.data+0x24b8): undefined reference to `hardlockup_panic'

---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all Intel Corporation

Acked-by on v1 1/1
#2 lore raw
In reply to Don Zickus
Quoted 10 lines · Don Zickus

The only way to enable a hardlockup to panic the machine is to set 'nmi_watchdog=panic' on the kernel command line.

This makes it awkward for end users and folks who want to run automate tests (like myself).

Mimic the softlockup_panic knob and create a /proc/sys/kernel/hardlockup_panic knob.

Seems like with my previous stack-dumping patch + this one, we are finally covering all the possible combinations. Hence in principle

The build failure reported by Fengguang's bot of course needs to be fixed, but I think it's there for soft lockup detector as well, but Fengguang's .config has soft lockup detector enable, so that failure is invisible.

I guess the most straightfoward fix is to ...

[ ... snip ... ]

include/linux/sched.h extern int proc_dowatchdog_thresh(struct ctl_table *table, int write, · line 387

Quoted 8 lines

index b7b9501..5e65b14 100644

include/linux/sched.h +1 -0
@@ -384,6 +384,7 @@ extern int proc_dowatchdog_thresh(struct ctl_table *table, int write,
 				  void __user *buffer,
 				  size_t *lenp, loff_t *ppos);
 extern unsigned int  softlockup_panic;
+extern unsigned int  hardlockup_panic;

... those two be there only if appropriate config options are set.

Thanks,

Signature · 2 lines
Jiri Kosina
SUSE Labs