* [PATCH] scsi/mvsas: Fix oops when ata command timeout.
@ 2012-08-01 7:14 majianpeng
2012-08-01 11:21 ` Xiangliang Yu
0 siblings, 1 reply; 3+ messages in thread
From: majianpeng @ 2012-08-01 7:14 UTC (permalink / raw)
To: JBottomley; +Cc: 于向亮, linux-scsi
Kernel message follows:
[ 511.712011] sd 11:0:0:0: [sdf] command ffff8800a4e81400 timed out
[ 511.712022] sas: Enter sas_scsi_recover_host busy: 1 failed: 1
[ 511.712024] sas: trying to find task 0xffff8800a4d24c80
[ 511.712026] sas: sas_scsi_find_task: aborting task 0xffff8800a4d24c80
[ 511.712029] drivers/scsi/mvsas/mv_sas.c 1631:mvs_abort_task() mvi=ffff8800b5300000 task=ffff8800a4d24c80 slot=ffff8800b5325038 slot_idx=x0
[ 511.712035] BUG: unable to handle kernel NULL pointer dereference at 0000000000000058
[ 511.712040] IP: [<ffffffff815f8c0c>] _raw_spin_lock_irqsave+0xc/0x30
[ 511.712047] PGD 0
[ 511.712049] Oops: 0002 [#1] SMP
[ 511.712052] Modules linked in: mvsas libsas scsi_transport_sas raid456 async_pq async_xor xor async_memcpy async_raid6_recov raid6_pq async_tx [last unloaded: mvsas]
[ 511.712062] CPU 3
[ 511.712066] Pid: 7322, comm: scsi_eh_11 Not tainted 3.5.0+ #106 To Be Filled By O.E.M. To Be Filled By O.E.M./To be filled by O.E.M.
[ 511.712068] RIP: 0010:[<ffffffff815f8c0c>] [<ffffffff815f8c0c>] _raw_spin_lock_irqsave+0xc/0x30
[ 511.712073] RSP: 0018:ffff880098d3bcb0 EFLAGS: 00010086
[ 511.712074] RAX: 0000000000000286 RBX: 0000000000000058 RCX: 00000000000000c3
[ 511.712076] RDX: 0000000000000100 RSI: 0000000000000046 RDI: 0000000000000058
[ 511.712078] RBP: ffff880098d3bcb0 R08: 000000000000000a R09: 0000000000000000
[ 511.712080] R10: 00000000000004e8 R11: 00000000000004e7 R12: ffff8800a4d24c80
[ 511.712082] R13: 0000000000000050 R14: ffff8800b5325038 R15: ffff8800a4eafe00
[ 511.712084] FS: 0000000000000000(0000) GS:ffff8800bdb80000(0000) knlGS:0000000000000000
[ 511.712086] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[ 511.712088] CR2: 0000000000000058 CR3: 00000000a4ce6000 CR4: 00000000000407e0
[ 511.712090] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 511.712091] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[ 511.712093] Process scsi_eh_11 (pid: 7322, threadinfo ffff880098d3a000, task ffff8800a61dde40)
[ 511.712095] Stack:
[ 511.712096] ffff880098d3bce0 ffffffff81060683 ffff880000000000 0000000000000000
[ 511.712099] ffff8800a4d24c80 ffff8800b5300000 ffff880098d3bcf0 ffffffffa0076a88
[ 511.712102] ffff880098d3bd50 ffffffffa0079bb5 ffff880000000000 ffff880000000018
[ 511.712106] Call Trace:
[ 511.712110] [<ffffffff81060683>] complete+0x23/0x60
[ 511.712115] [<ffffffffa0076a88>] mvs_tmf_timedout+0x18/0x20 [mvsas]
[ 511.712119] [<ffffffffa0079bb5>] mvs_slot_complete+0x765/0x7d0 [mvsas]
[ 511.712125] [<ffffffffa005a17d>] sas_scsi_recover_host+0x55d/0xdb0 [libsas]
[ 511.712128] [<ffffffff8106d600>] ? idle_balance+0xe0/0x130
[ 511.712133] [<ffffffff813b150c>] scsi_error_handler+0xcc/0x470
[ 511.712136] [<ffffffff815f7ad0>] ? __schedule+0x370/0x730
[ 511.712139] [<ffffffff8105f728>] ? __wake_up_common+0x58/0x90
[ 511.712142] [<ffffffff813b1440>] ? scsi_eh_get_sense+0x110/0x110
[ 511.712146] [<ffffffff810571be>] kthread+0x8e/0xa0
[ 511.712150] [<ffffffff816015f4>] kernel_thread_helper+0x4/0x10
[ 511.712153] [<ffffffff81057130>] ? flush_kthread_work+0x120/0x120
[ 511.712156] [<ffffffff816015f0>] ? gs_change+0xb/0xb
[ 511.712157] Code: 8a 00 01 00 00 89 d0 f0 66 0f b1 0f 66 39 d0 0f 94 c0 0f b6 c0 5d c3 0f 1f 84 00 00 00 00 00 55 48 89 e5 9c 58 fa ba 00 01 00 00 <f0> 66 0f c1 17 0f b6 ce 38 d1 74 11 0f 1f 84 00 00 00 00 00 f3
[ 511.712191] RIP [<ffffffff815f8c0c>] _raw_spin_lock_irqsave+0xc/0x30
[ 511.712194] RSP <ffff880098d3bcb0>
[ 511.712196] CR2: 0000000000000058
[ 511.712198] ---[ end trace a781c7b1e65db92c ]---
Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
---
drivers/scsi/mvsas/mv_sas.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/scsi/mvsas/mv_sas.c b/drivers/scsi/mvsas/mv_sas.c
index 4539d59..af6a1c5 100644
--- a/drivers/scsi/mvsas/mv_sas.c
+++ b/drivers/scsi/mvsas/mv_sas.c
@@ -1629,7 +1629,10 @@ int mvs_abort_task(struct sas_task *task)
mv_dprintk("mvs_abort_task() mvi=%p task=%p "
"slot=%p slot_idx=x%x\n",
mvi, task, slot, slot_idx);
- mvs_tmf_timedout((unsigned long)task);
+ if (task->slow_task != NULL)
+ mvs_tmf_timedout((unsigned long)task);
+ else
+ task->task_state_flags |= SAS_TASK_STATE_ABORTED;
mvs_slot_task_free(mvi, task, slot, slot_idx);
rc = TMF_RESP_FUNC_COMPLETE;
goto out;
--
1.7.9.5
^ permalink raw reply related [flat|nested] 3+ messages in thread* RE: [PATCH] scsi/mvsas: Fix oops when ata command timeout.
2012-08-01 7:14 [PATCH] scsi/mvsas: Fix oops when ata command timeout majianpeng
@ 2012-08-01 11:21 ` Xiangliang Yu
2012-08-02 5:13 ` majianpeng
0 siblings, 1 reply; 3+ messages in thread
From: Xiangliang Yu @ 2012-08-01 11:21 UTC (permalink / raw)
To: majianpeng, JBottomley; +Cc: linux-scsi
Hi, Jianpeng
I think MVSAS shouldn't care the completion sign here, just modify task state is ok.
> -----Original Message-----
> From: majianpeng [mailto:majianpeng@gmail.com]
> Sent: 2012年8月1日 15:15
> To: JBottomley
> Cc: Xiangliang Yu; linux-scsi
> Subject: [PATCH] scsi/mvsas: Fix oops when ata command timeout.
>
> Kernel message follows:
>
> [ 511.712011] sd 11:0:0:0: [sdf] command ffff8800a4e81400 timed out
> [ 511.712022] sas: Enter sas_scsi_recover_host busy: 1 failed: 1
> [ 511.712024] sas: trying to find task 0xffff8800a4d24c80
> [ 511.712026] sas: sas_scsi_find_task: aborting task 0xffff8800a4d24c80
> [ 511.712029] drivers/scsi/mvsas/mv_sas.c 1631:mvs_abort_task()
> mvi=ffff8800b5300000 task=ffff8800a4d24c80 slot=ffff8800b5325038
> slot_idx=x0
> [ 511.712035] BUG: unable to handle kernel NULL pointer dereference at
> 0000000000000058
> [ 511.712040] IP: [<ffffffff815f8c0c>] _raw_spin_lock_irqsave+0xc/0x30
> [ 511.712047] PGD 0
> [ 511.712049] Oops: 0002 [#1] SMP
> [ 511.712052] Modules linked in: mvsas libsas scsi_transport_sas raid456
> async_pq async_xor xor async_memcpy async_raid6_recov raid6_pq async_tx [last
> unloaded: mvsas]
> [ 511.712062] CPU 3
> [ 511.712066] Pid: 7322, comm: scsi_eh_11 Not tainted 3.5.0+ #106 To Be Filled
> By O.E.M. To Be Filled By O.E.M./To be filled by O.E.M.
> [ 511.712068] RIP: 0010:[<ffffffff815f8c0c>] [<ffffffff815f8c0c>]
> _raw_spin_lock_irqsave+0xc/0x30
> [ 511.712073] RSP: 0018:ffff880098d3bcb0 EFLAGS: 00010086
> [ 511.712074] RAX: 0000000000000286 RBX: 0000000000000058 RCX:
> 00000000000000c3
> [ 511.712076] RDX: 0000000000000100 RSI: 0000000000000046 RDI:
> 0000000000000058
> [ 511.712078] RBP: ffff880098d3bcb0 R08: 000000000000000a R09:
> 0000000000000000
> [ 511.712080] R10: 00000000000004e8 R11: 00000000000004e7 R12:
> ffff8800a4d24c80
> [ 511.712082] R13: 0000000000000050 R14: ffff8800b5325038 R15:
> ffff8800a4eafe00
> [ 511.712084] FS: 0000000000000000(0000) GS:ffff8800bdb80000(0000)
> knlGS:0000000000000000
> [ 511.712086] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> [ 511.712088] CR2: 0000000000000058 CR3: 00000000a4ce6000 CR4:
> 00000000000407e0
> [ 511.712090] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
> 0000000000000000
> [ 511.712091] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7:
> 0000000000000400
> [ 511.712093] Process scsi_eh_11 (pid: 7322, threadinfo ffff880098d3a000,
> task ffff8800a61dde40)
> [ 511.712095] Stack:
> [ 511.712096] ffff880098d3bce0 ffffffff81060683 ffff880000000000
> 0000000000000000
> [ 511.712099] ffff8800a4d24c80 ffff8800b5300000 ffff880098d3bcf0
> ffffffffa0076a88
> [ 511.712102] ffff880098d3bd50 ffffffffa0079bb5 ffff880000000000
> ffff880000000018
> [ 511.712106] Call Trace:
> [ 511.712110] [<ffffffff81060683>] complete+0x23/0x60
> [ 511.712115] [<ffffffffa0076a88>] mvs_tmf_timedout+0x18/0x20 [mvsas]
> [ 511.712119] [<ffffffffa0079bb5>] mvs_slot_complete+0x765/0x7d0 [mvsas]
> [ 511.712125] [<ffffffffa005a17d>] sas_scsi_recover_host+0x55d/0xdb0
> [libsas]
> [ 511.712128] [<ffffffff8106d600>] ? idle_balance+0xe0/0x130
> [ 511.712133] [<ffffffff813b150c>] scsi_error_handler+0xcc/0x470
> [ 511.712136] [<ffffffff815f7ad0>] ? __schedule+0x370/0x730
> [ 511.712139] [<ffffffff8105f728>] ? __wake_up_common+0x58/0x90
> [ 511.712142] [<ffffffff813b1440>] ? scsi_eh_get_sense+0x110/0x110
> [ 511.712146] [<ffffffff810571be>] kthread+0x8e/0xa0
> [ 511.712150] [<ffffffff816015f4>] kernel_thread_helper+0x4/0x10
> [ 511.712153] [<ffffffff81057130>] ? flush_kthread_work+0x120/0x120
> [ 511.712156] [<ffffffff816015f0>] ? gs_change+0xb/0xb
> [ 511.712157] Code: 8a 00 01 00 00 89 d0 f0 66 0f b1 0f 66 39 d0 0f 94 c0 0f
> b6 c0 5d c3 0f 1f 84 00 00 00 00 00 55 48 89 e5 9c 58 fa ba 00 01 00 00 <f0>
> 66 0f c1 17 0f b6 ce 38 d1 74 11 0f 1f 84 00 00 00 00 00 f3
> [ 511.712191] RIP [<ffffffff815f8c0c>] _raw_spin_lock_irqsave+0xc/0x30
> [ 511.712194] RSP <ffff880098d3bcb0>
> [ 511.712196] CR2: 0000000000000058
> [ 511.712198] ---[ end trace a781c7b1e65db92c ]---
>
> Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
> ---
> drivers/scsi/mvsas/mv_sas.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/scsi/mvsas/mv_sas.c b/drivers/scsi/mvsas/mv_sas.c
> index 4539d59..af6a1c5 100644
> --- a/drivers/scsi/mvsas/mv_sas.c
> +++ b/drivers/scsi/mvsas/mv_sas.c
> @@ -1629,7 +1629,10 @@ int mvs_abort_task(struct sas_task *task)
> mv_dprintk("mvs_abort_task() mvi=%p task=%p "
> "slot=%p slot_idx=x%x\n",
> mvi, task, slot, slot_idx);
> - mvs_tmf_timedout((unsigned long)task);
> + if (task->slow_task != NULL)
> + mvs_tmf_timedout((unsigned long)task);
> + else
> + task->task_state_flags |= SAS_TASK_STATE_ABORTED;
> mvs_slot_task_free(mvi, task, slot, slot_idx);
> rc = TMF_RESP_FUNC_COMPLETE;
> goto out;
> --
> 1.7.9.5
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: RE: [PATCH] scsi/mvsas: Fix oops when ata command timeout.
2012-08-01 11:21 ` Xiangliang Yu
@ 2012-08-02 5:13 ` majianpeng
0 siblings, 0 replies; 3+ messages in thread
From: majianpeng @ 2012-08-02 5:13 UTC (permalink / raw)
To: 于向亮, scsi maintainers; +Cc: linux-scsi
On 2012-08-01 19:21 Xiangliang Yu <yuxiangl@marvell.com> Wrote:
>Hi, Jianpeng
>
>I think MVSAS shouldn't care the completion sign here, just modify task state is ok.
>
>
>
[snip]
>>
>> Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
>> ---
>> drivers/scsi/mvsas/mv_sas.c | 5 ++++-
>> 1 file changed, 4 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/scsi/mvsas/mv_sas.c b/drivers/scsi/mvsas/mv_sas.c
>> index 4539d59..af6a1c5 100644
>> --- a/drivers/scsi/mvsas/mv_sas.c
>> +++ b/drivers/scsi/mvsas/mv_sas.c
>> @@ -1629,7 +1629,10 @@ int mvs_abort_task(struct sas_task *task)
>> mv_dprintk("mvs_abort_task() mvi=%p task=%p "
>> "slot=%p slot_idx=x%x\n",
>> mvi, task, slot, slot_idx);
>> - mvs_tmf_timedout((unsigned long)task);
>> + if (task->slow_task != NULL)
>> + mvs_tmf_timedout((unsigned long)task);
>> + else
>> + task->task_state_flags |= SAS_TASK_STATE_ABORTED;
>> mvs_slot_task_free(mvi, task, slot, slot_idx);
>> rc = TMF_RESP_FUNC_COMPLETE;
>> goto out;
>> --
>> 1.7.9.5
Thanks, I'll modify the code and resend.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2012-08-02 5:11 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-08-01 7:14 [PATCH] scsi/mvsas: Fix oops when ata command timeout majianpeng
2012-08-01 11:21 ` Xiangliang Yu
2012-08-02 5:13 ` majianpeng
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).