The touch_nmi_watchdog() routine on x86 ultimately calls touch_softlockup_watchdog(). The problem is that to touch the softlockup watchdog, the cpu_clock code has to be called which could involve multiple cpu locks and can lead to a hard hang if one of the locks is held by a processor that is not going to return anytime soon (such as could be the case with kgdb or perhaps even with some other kind of exception). This patch causes the public version of the touch_softlockup_watchdog() to defer the cpu clock access to a later point. The test case for this problem is to use the following kernel config options: CONFIG_KGDB_TESTS=y CONFIG_KGDB_TESTS_ON_BOOT=y CONFIG_KGDB_TESTS_BOOT_STRING="V1F100I100000" It should be noted that kgdb test suite and these options were not available until 2.6.26-rc2, so it was necessary to patch the kgdb test suite during the bisection. I would consider this patch a regression fix because the problem first appeared in commit 27ec4407790d075c325e1f4da0a19c56953cce23 when some logic was added to try to periodically sync the clocks. It was possible to work around this particular problem by simply not performing the sync anytime the system was in a critical context. This was ok until commit 3e51f33fcc7f55e6df25d15b55ed10c8b4da84cd, which added config option CONFIG_HAVE_UNSTABLE_SCHED_CLOCK and some multi-cpu locks to sync the clocks. It became clear that accessing this code from an nmi was the source of the lockups. Avoiding the access to the low level clock code from an code inside the NMI processing also fixed the problem with the 27ec44... commit.
Unified · Split
| @@ -49,12 +49,17 @@ static unsigned long get_timestamp(int this_cpu) · 48 unmodified lines | ||
| 49 | 49 | return cpu_clock(this_cpu) >> 30LL; /* 2^30 ~= 10^9 */ |
| 50 | 50 | } |
| 51 | 51 | |
| 52 | −void touch_softlockup_watchdog(void) | |
| 52 | +static void __touch_softlockup_watchdog(void) | |
| 53 | 53 | { |
| 54 | 54 | int this_cpu = raw_smp_processor_id(); |
| 55 | 55 | |
| 56 | 56 | __raw_get_cpu_var(touch_timestamp) = get_timestamp(this_cpu); |
| 57 | 57 | } |
| 58 | + | |
| 59 | +void touch_softlockup_watchdog(void) | |
| 60 | +{ | |
| 61 | + __raw_get_cpu_var(touch_timestamp) = 0; | |
| 62 | +} | |
| 58 | 63 | EXPORT_SYMBOL(touch_softlockup_watchdog); |
| 59 | 64 | |
| 60 | 65 | void touch_all_softlockup_watchdogs(void) |
| @@ -80,7 +85,7 @@ void softlockup_tick(void) · 19 unmodified lines | ||
| 80 | 85 | unsigned long now; |
| 81 | 86 | |
| 82 | 87 | if (touch_timestamp == 0) { |
| 83 | − touch_softlockup_watchdog(); | |
| 88 | + __touch_softlockup_watchdog(); | |
| 84 | 89 | return; |
| 85 | 90 | } |
| 86 | 91 | |
| @@ -95,7 +100,7 @@ void softlockup_tick(void) · 8 unmodified lines | ||
| 95 | 100 | |
| 96 | 101 | /* do not print during early bootup: */ |
| 97 | 102 | if (unlikely(system_state != SYSTEM_RUNNING)) { |
| 98 | − touch_softlockup_watchdog(); | |
| 103 | + __touch_softlockup_watchdog(); | |
| 99 | 104 | return; |
| 100 | 105 | } |
| 101 | 106 | |
| @@ -214,7 +219,7 @@ static int watchdog(void *__bind_cpu) · 112 unmodified lines | ||
| 214 | 219 | sched_setscheduler(current, SCHED_FIFO, ¶m); |
| 215 | 220 | |
| 216 | 221 | /* initialize timestamp */ |
| 217 | − touch_softlockup_watchdog(); | |
| 222 | + __touch_softlockup_watchdog(); | |
| 218 | 223 | |
| 219 | 224 | set_current_state(TASK_INTERRUPTIBLE); |
| 220 | 225 | /* |
| @@ -223,7 +228,7 @@ static int watchdog(void *__bind_cpu) · 2 unmodified lines | ||
| 223 | 228 | * debug-printout triggers in softlockup_tick(). |
| 224 | 229 | */ |
| 225 | 230 | while (!kthread_should_stop()) { |
| 226 | − touch_softlockup_watchdog(); | |
| 231 | + __touch_softlockup_watchdog(); | |
| 227 | 232 | schedule(); |
| 228 | 233 | |
| 229 | 234 | if (kthread_should_stop()) |