[PATCH] gcov: fix softlockups

From: Andrey Ryabinin
Date: Thu Apr 02 2015 - 06:49:49 EST


gcov profiling if enabled with other heavy compile-time
instrumentation like KASan could trigger following softlockups:

[ 72.460059] NMI watchdog: BUG: soft lockup - CPU#0 stuck for 22s! [swapper/0:1]
[ 72.460068] Modules linked in:
[ 72.460068] irq event stamp: 22823276
[ 72.460068] hardirqs last enabled at (22823275): [<ffffffff86e8d10d>] mutex_lock_nested+0x7d9/0x930
[ 72.460068] hardirqs last disabled at (22823276): [<ffffffff86e9521d>] apic_timer_interrupt+0x6d/0x80
[ 72.460068] softirqs last enabled at (22823172): [<ffffffff811ed969>] __do_softirq+0x4db/0x729
[ 72.460068] softirqs last disabled at (22823167): [<ffffffff811edfcf>] irq_exit+0x7d/0x15b
[ 72.460068] CPU: 0 PID: 1 Comm: swapper/0 Tainted: G W 3.19.0-05245-gbb33326-dirty #3
[ 72.460068] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5.1-0-g8936dbb-20141113_115728-nilsson.home.kraxel.org 04/01/2014
[ 72.460068] task: ffff88006cba8000 ti: ffff88006cbb0000 task.ti: ffff88006cbb0000
[ 72.460068] RIP: 0010:[<ffffffff8146822a>] [<ffffffff8146822a>] kasan_mem_to_shadow+0x1e/0x1f
[ 72.460068] RSP: 0000:ffff88006cbb3cb0 EFLAGS: 00000207
[ 72.460068] RAX: fffffbfff1331380 RBX: ffffffff81468fc2 RCX: ffff88006d600006
[ 72.460068] RDX: dffffc0000000000 RSI: ffffffff89989bfa RDI: 1ffffffff1331380
[ 72.460068] RBP: ffff88006cbb3cf8 R08: 00000000037178af R09: 0000000003714ae1
[ 72.460068] R10: ffffed000c8d0b1f R11: 00000000000000cf R12: ffffffff8d8c2ba0
[ 72.460068] R13: ffff88006d640780 R14: ffffffff81269ad5 R15: ffff88006cbb3c58
[ 72.460068] FS: 0000000000000000(0000) GS:ffff88006d600000(0000) knlGS:0000000000000000
[ 72.460068] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[ 72.460068] CR2: 0000000000000000 CR3: 000000000a229000 CR4: 00000000000006f0
[ 72.460068] Stack:
[ 72.460068] ffffffff81468a10 dffffc0000000000 ffffffff8abcfac0 ffff88006cbb3d28
[ 72.460068] ffffffff81468fc2 ffff88006cbb3d38 ffffffff81468fc2 dffffc0000000000
[ 72.460068] ffffffff89989c05 ffff88006cbb3d28 ffffffff8212fea0 ffff880063049398
[ 72.460068] Call Trace:
[ 72.460068] [<ffffffff81468a10>] ? __asan_load1+0x66/0xbb
[ 72.460068] [<ffffffff81468fc2>] ? __asan_load8+0x6d/0x10c
[ 72.460068] [<ffffffff81468fc2>] ? __asan_load8+0x6d/0x10c
[ 72.460068] [<ffffffff8212fea0>] strcmp+0x28/0x70
[ 72.460068] [<ffffffff813228af>] get_node_by_name+0x66/0x99
[ 72.460068] [<ffffffff81323879>] gcov_event+0x4f/0x69e
[ 72.460068] [<ffffffff86e90477>] ? mutex_unlock+0x15/0x1e
[ 72.460068] [<ffffffff8ca18cb9>] ? gcov_persist_setup+0x77/0x77
[ 72.460068] [<ffffffff8ca18cb9>] ? gcov_persist_setup+0x77/0x77
[ 72.460068] [<ffffffff813227a6>] gcov_enable_events+0x54/0x7b
[ 72.460068] [<ffffffff8ca18db1>] gcov_fs_init+0xf8/0x134
[ 72.460068] [<ffffffff810022ca>] do_one_initcall+0x1b2/0x288
[ 72.460068] [<ffffffff81468fc2>] ? __asan_load8+0x6d/0x10c
[ 72.460068] [<ffffffff8c9c8f02>] kernel_init_freeable+0x467/0x580
[ 72.460068] [<ffffffff86dd9a48>] ? rest_init+0x23b/0x23b
[ 72.460068] [<ffffffff86dd9a5d>] kernel_init+0x15/0x18b
[ 72.460068] [<ffffffff86e93f3c>] ret_from_fork+0x7c/0xb0
[ 72.460068] [<ffffffff86dd9a48>] ? rest_init+0x23b/0x23b
[ 72.460068] Code: ff 48 ff 05 61 ec c2 0c 48 89 e5 5d c3 55 48 c1 ef 03 48 ba 00 00 00 00 00 fc ff df 48 8d 04 17 48 ff 05 da f5 c2 0c 48 89 e5 5d <c3> 55 48 ff 05 f5 fe c2 0c 48 89 e5 5d c3 55 48 ff 05 f0 fe c2
[ 72.460068] Kernel panic - not syncing: softlockup: hung tasks

Fix this by sticking cond_resched() in gcov_enable_events().

Signed-off-by: Andrey Ryabinin <a.ryabinin@xxxxxxxxxxx>
Reported-by: Fengguang Wu <fengguang.wu@xxxxxxxxx>
---
kernel/gcov/base.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/kernel/gcov/base.c b/kernel/gcov/base.c
index b358a80..a744098 100644
--- a/kernel/gcov/base.c
+++ b/kernel/gcov/base.c
@@ -18,6 +18,7 @@
#include <linux/init.h>
#include <linux/module.h>
#include <linux/mutex.h>
+#include <linux/sched.h>
#include "gcov.h"

static int gcov_events_enabled;
@@ -107,8 +108,10 @@ void gcov_enable_events(void)
gcov_events_enabled = 1;

/* Perform event callback for previously registered entries. */
- while ((info = gcov_info_next(info)))
+ while ((info = gcov_info_next(info))) {
gcov_event(GCOV_ADD, info);
+ cond_resched();
+ }

mutex_unlock(&gcov_lock);
}
--
2.3.5

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/