On Fri, Aug 1, 2014 at 10:50 PM, Fengguang Wu <fengguang...@intel.com> wrote:
> Greetings,
>
> 0day kernel testing robot got the below dmesg and the first bad commit is
>
> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> commit 2c575026fae6e63771bd2a4c1d407214a8096a89
> Author: Hong Zhiguo <zhiguoh...@tencent.com>
> AuthorDate: Wed Nov 20 10:35:05 2013 -0700
> Commit: Jens Axboe <ax...@kernel.dk>
> CommitDate: Wed Nov 20 15:33:04 2013 -0700
>
> Update of blkg_stat and blkg_rwstat may happen in bh context.
> While u64_stats_fetch_retry is only preempt_disable on 32bit
> UP system. This is not enough to avoid preemption by bh and
> may read strange 64 bit value.
>
> Signed-off-by: Hong Zhiguo <zhiguoh...@tencent.com>
> Acked-by: Tejun Heo <t...@kernel.org>
> Cc: sta...@kernel.org
> Signed-off-by: Jens Axboe <ax...@kernel.dk>
>
> +-----------------------------------------------------------------------------+------------+------------+------------------+
> |
> | 82023bb7f7 | 2c575026fa | v3.16-rc7_080120 |
> +-----------------------------------------------------------------------------+------------+------------+------------------+
> | boot_successes
> | 270 | 22 | 7 |
> | boot_failures
> | 0 | 68 | 4 |
> | BUG:unable_to_handle_kernel_NULL_pointer_dereference
> | 0 | 1 | |
> | Oops
> | 0 | 1 | |
> | EIP_is_at_strlen
> | 0 | 1 | |
> | BUG:sleeping_function_called_from_invalid_context_at_kernel/locking/rwsem.c
> | 0 | 1 | |
> | INFO:lockdep_is_turned_off
> | 0 | 1 | |
> | backtrace:vfs_write
> | 0 | 1 | |
> | backtrace:SyS_write
> | 0 | 1 | |
> | WARNING:CPU:PID:at_kernel/softirq.c:_local_bh_enable_ip()
> | 0 | 67 | |
> | inconsistent_SOFTIRQ-ON-W-IN-SOFTIRQ-W_usage
> | 0 | 67 | 4 |
> | backtrace:redo_fd_request
> | 0 | 67 | |
> | backtrace:floppy_work_workfn
> | 0 | 0 | 4 |
> | backtrace:cpu_startup_entry
> | 0 | 0 | 4 |
> +-----------------------------------------------------------------------------+------------+------------+------------------+
>
> [ 2.848648] NX-protecting the kernel data: 5832k
> [ 2.903073] udev[122]: starting version 164
> [ 3.123339] ------------[ cut here ]------------
> [ 3.123386] WARNING: CPU: 0 PID: 85 at kernel/softirq.c:156
> _local_bh_enable_ip+0x34/0x97()
> [ 3.123386] Modules linked in:
> [ 3.123386] CPU: 0 PID: 85 Comm: kworker/u2:2 Not tainted
> 3.12.0-10878-g2c57502 #1
> [ 3.123386] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
> [ 3.123386] Workqueue: floppy redo_fd_request
> [ 3.123386] c0258bca 00000000 d277bce8 c0646df5 d277bd00 c022d793
> c0230f4d c04297b7
> [ 3.123386] d2746634 d277bd6c d277bd10 c022d81d 00000009 00000000
> d277bd1c c0230f4d
> [ 3.123386] d277bd4c d277bd24 c0230fc0 d277bd38 c04297b7 c001be10
> d2746400 d2743c00
> [ 3.123386] Call Trace:
> [ 3.123386] [<c0258bca>] ? console_unlock+0x355/0x382
> [ 3.123386] [<c0646df5>] dump_stack+0x16/0x18
> [ 3.123386] [<c022d793>] warn_slowpath_common+0x5f/0x76
> [ 3.123386] [<c0230f4d>] ? _local_bh_enable_ip+0x34/0x97
> [ 3.123386] [<c04297b7>] ? blkg_rwstat_read+0x1d/0x24
> [ 3.123386] [<c022d81d>] warn_slowpath_null+0x14/0x18
> [ 3.123386] [<c0230f4d>] _local_bh_enable_ip+0x34/0x97
> [ 3.123386] [<c0230fc0>] local_bh_enable+0x10/0x12
> [ 3.123386] [<c04297b7>] blkg_rwstat_read+0x1d/0x24
> [ 3.123386] [<c042ab0a>] __cfq_set_active_queue+0x65/0x165
> [ 3.123386] [<c025348b>] ? mark_lock+0x1e/0x1d3
> [ 3.123386] [<c025348b>] ? mark_lock+0x1e/0x1d3
> [ 3.123386] [<c0252132>] ? __lock_is_held+0x2d/0x43
> [ 3.123386] [<c021d2e1>] ? pvclock_clocksource_read+0xb1/0xda
> [ 3.123386] [<c021c888>] ? kvm_clock_read+0x14/0x1d
> [ 3.123386] [<c0206476>] ? paravirt_sched_clock+0x9/0xd
> [ 3.123386] [<c042d08d>] cfq_dispatch_requests+0x521/0x78a
> [ 3.123386] [<c0253f5c>] ? __lock_acquire+0x523/0xbdf
> [ 3.123386] [<c0253f5c>] ? __lock_acquire+0x523/0xbdf
> [ 3.123386] [<c025348b>] ? mark_lock+0x1e/0x1d3
> [ 3.123386] [<c02548e7>] ? lock_acquire+0xa3/0xfc
> [ 3.123386] [<c04f73a5>] ? redo_fd_request+0x3f/0xc0c
> [ 3.123386] [<c041a1a4>] blk_peek_request+0x59/0x211
> [ 3.123386] [<c041a36a>] blk_fetch_request+0xe/0x1e
> [ 3.123386] [<c04f73e0>] redo_fd_request+0x7a/0xc0c
> [ 3.123386] [<c02407ab>] ? process_one_work+0x13c/0x398
> [ 3.123386] [<c0240843>] process_one_work+0x1d4/0x398
> [ 3.123386] [<c0241092>] worker_thread+0x12b/0x1c8
> [ 3.123386] [<c0240f67>] ? rescuer_thread+0x1ff/0x1ff
> [ 3.123386] [<c0244b79>] kthread+0x8b/0x90
> [ 3.123386] [<c064ccb7>] ret_from_kernel_thread+0x1b/0x28
> [ 3.123386] [<c0244aee>] ? insert_kthread_work+0x5a/0x5a
> [ 3.123386] ---[ end trace 470268993472ede8 ]---
> [ 3.246919] end_request: I/O error, dev fd0, sector 0
>
> git bisect start v3.13 v3.12 --
> git bisect good 3bad8bb5cd3048a67df43ac6b1e2f191f19d9ff0 # 22:45 62+
> 0 Merge branch 'for-next' of git://git.samba.org/sfrench/cifs-2.6
> git bisect bad dd0508093b79141e0044ca02f0acb6319f69f546 # 22:59 0-
> 62 Merge branch 'sched-urgent-for-linus' of
> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
> git bisect bad 3f02ff5c2c69753666787ed125708d283a823ffb # 23:11 5-
> 26 Merge tag 'tty-3.13-rc3' of
> git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
> git bisect good 1ab231b274ba51a54acebec23c6aded0f3cdf54e # 23:19 62+
> 5 Merge branch 'timers-urgent-for-linus' of
> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
> git bisect bad 5ee540613db504a10e15fafaf4c08cac96aa1823 # 23:32 7-
> 5 Merge branch 'for-linus' of git://git.kernel.dk/linux-block
> git bisect good 53c6de50262a8edd6932bb59a32db7b9d92f8d67 # 23:39 90+
> 0 Merge branch 'x86/urgent' of
> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
> git bisect good 59fb2f0e9e30ad99a8bab0ff1efaf8f4a3b7105f # 00:12 90+
> 0 Merge tag 'fbdev-fixes-3.13' of
> git://git.kernel.org/pub/scm/linux/kernel/git/tomba/linux
> git bisect good ef1e4e32d595d3e6c9a6d3d2956f087d5886c5e5 # 01:00 90+
> 0 Merge tag 'for-linus' of
> git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs
> git bisect good 29be6345bbaec8502a70c4e2204d5818b48c4e8f # 01:08 90+
> 0 Merge tag 'nfs-for-3.13-3' of
> git://git.linux-nfs.org/projects/trondmy/linux-nfs
> git bisect bad e345d767f6530ec9cb0aabab7ea248072a9c6975 # 01:38 31-
> 32 Merge branch 'stable/for-jens-3.13-take-two' of
> git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip into for-linus
> git bisect bad c170bbb45febc03ac4d34ba2b8bb55e06104b7e7 # 01:43 28-
> 19 block: submit_bio_wait() conversions
> git bisect bad 2c575026fae6e63771bd2a4c1d407214a8096a89 # 02:04 6-
> 16 Update of blkg_stat and blkg_rwstat may happen in bh context. While
> u64_stats_fetch_retry is only preempt_disable on 32bit UP system. This is not
> enough to avoid preemption by bh and may read strange 64 bit value.
> # first bad commit: [2c575026fae6e63771bd2a4c1d407214a8096a89] Update of
> blkg_stat and blkg_rwstat may happen in bh context. While
> u64_stats_fetch_retry is only preempt_disable on 32bit UP system. This is not
> enough to avoid preemption by bh and may read strange 64 bit value.
> git bisect good 82023bb7f75b0052f40d3e74169d191c3e4e6286 # 02:29 270+
> 0 Merge tag 'pm+acpi-2-3.13-rc1' of
> git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
> git bisect bad f1f29248d675d4762e7b7f498be50d2fb9c294f3 # 02:30 0-
> 4 0day head guard for 'devel-hourly-2014080120'
> git bisect bad 6f0928036bcbd8dd2ebee9201580da8bcc0a81af # 03:31 20-
> 20 Merge tag 'pm+acpi-3.16-rc8' of
> git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
> git bisect bad 4198cf324ddea447daac3f83d3c2319ced774201 # 03:38 16-
> 17 Add linux-next specific files for 20140801
>
>
> This script may reproduce the error.
>
> ----------------------------------------------------------------------------
> #!/bin/bash
>
> kernel=$1
> initrd=yocto-minimal-i386.cgz
>
> wget --no-clobber
> https://github.com/fengguang/reproduce-kernel-bug/raw/master/initrd/$initrd
>
> kvm=(
> qemu-system-x86_64
> -enable-kvm
> -cpu Haswell,+smep,+smap
> -kernel $kernel
> -initrd $initrd
> -m 320
> -smp 1
> -net nic,vlan=1,model=e1000
> -net user,vlan=1
> -boot order=nc
> -no-reboot
> -watchdog i6300esb
> -rtc base=localtime
> -serial stdio
> -display none
> -monitor null
> )
>
> append=(
> hung_task_panic=1
> earlyprintk=ttyS0,115200
> debug
> apic=debug
> sysrq_always_enabled
> rcupdate.rcu_cpu_stall_timeout=100
> panic=10
> softlockup_panic=1
> nmi_watchdog=panic
> prompt_ramdisk=0
> console=ttyS0,115200
> console=tty0
> vga=normal
> root=/dev/ram0
> rw
> drbd.minor_count=8
> )
>
> "${kvm[@]}" --append "${append[*]}"
> ----------------------------------------------------------------------------
>
> Thanks,
> Fengguang
>
> _______________________________________________
> LKP mailing list
> l...@linux.intel.com
>
Hey Fenngguag,
I assume this is the commit your are talking about after the robot
runs git bisect.
commit 2c575026fae6e63771bd2a4c1d407214a8096a89
Author: Hong Zhiguo <zhiguoh...@tencent.com>
Date: Wed Nov 20 10:35:05 2013 -0700
Update of blkg_stat and blkg_rwstat may happen in bh context.
While u64_stats_fetch_retry is only preempt_disable on 32bit
UP system. This is not enough to avoid preemption by bh and
may read strange 64 bit value.
Signed-off-by: Hong Zhiguo <zhiguoh...@tencent.com>
Acked-by: Tejun Heo <t...@kernel.org>
Cc: sta...@kernel.org
Signed-off-by: Jens Axboe <ax...@kernel.dk>
diff --git a/block/blk-cgroup.h b/block/blk-cgroup.h
index 1610b22..86154ea 100644
--- a/block/blk-cgroup.h
+++ b/block/blk-cgroup.h
@@ -435,9 +435,9 @@ static inline uint64_t blkg_stat_read(struct
blkg_stat *stat)
uint64_t v;
do {
- start = u64_stats_fetch_begin(&stat->syncp);
+ start = u64_stats_fetch_begin_bh(&stat->syncp);
v = stat->cnt;
- } while (u64_stats_fetch_retry(&stat->syncp, start));
+ } while (u64_stats_fetch_retry_bh(&stat->syncp, start));
return v;
}
@@ -508,9 +508,9 @@ static inline struct blkg_rwstat
blkg_rwstat_read(struct blkg_rwstat *rwstat)
struct blkg_rwstat tmp;
do {
- start = u64_stats_fetch_begin(&rwstat->syncp);
+ start = u64_stats_fetch_begin_bh(&rwstat->syncp);
tmp = *rwstat;
- } while (u64_stats_fetch_retry(&rwstat->syncp, start));
+ } while (u64_stats_fetch_retry_bh(&rwstat->syncp, start));
return tmp;
}
After tracing i actually get the same call with irqs functions instead
of bottom halves,
maybe it's just the lxr I am using but it seems that this may be the issue we
are now calling in irqs, and not bottom halves. I am putting my function traces
below with the line numbers I am hitting in linus's tree after updating today.
Regards
static inline uint64_t blkg_stat_read(struct blkg_stat *stat)
434 {
435 unsigned int start;
436 uint64_t v;
437
438 do {
439 start = u64_stats_fetch_begin_irq(&stat->syncp);
440 v = stat->cnt;
441 } while (u64_stats_fetch_retry_irq(&stat->syncp, start));
442
443 return v;
444 }
506 static inline struct blkg_rwstat blkg_rwstat_read(struct
blkg_rwstat *rwstat)
507 {
508 unsigned int start;
509 struct blkg_rwstat tmp;
510
511 do {
512 start = u64_stats_fetch_begin_irq(&rwstat->syncp);
513 tmp = *rwstat;
514 } while (u64_stats_fetch_retry_irq(&rwstat->syncp, start));
515
516 return tmp;
517 }
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
