From: Jason Yan <yanaijie@huawei.com>
To: Hannes Reinecke <hare@suse.de>,
"Martin K. Petersen" <martin.petersen@oracle.com>
Cc: Christoph Hellwig <hch@lst.de>,
James Bottomley <james.bottomley@hansenpartnership.com>,
linux-scsi@vger.kernel.org, Yves-Alexis Perez <corsac@debian.org>,
Hannes Reinecke <hare@suse.com>
Subject: Re: [PATCHv2] libsas: Check for completed commands before calling lldd_abort_task()
Date: Tue, 9 Jan 2018 12:09:25 +0800 [thread overview]
Message-ID: <5A544075.4010705@huawei.com> (raw)
In-Reply-To: <1515413086-34256-1-git-send-email-hare@suse.de>
Hannes,
On 2018/1/8 20:04, Hannes Reinecke wrote:
> The abort handler might be racing with command completion, so the
> task might already be NULL by the time the abort handler is called.
>
> Signed-off-by: Hannes Reinecke <hare@suse.com>
> ---
> drivers/scsi/libsas/sas_scsi_host.c | 22 +++++++++++++++++++---
> 1 file changed, 19 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/scsi/libsas/sas_scsi_host.c b/drivers/scsi/libsas/sas_scsi_host.c
> index 58476b7..08203fb 100644
> --- a/drivers/scsi/libsas/sas_scsi_host.c
> +++ b/drivers/scsi/libsas/sas_scsi_host.c
> @@ -486,18 +486,34 @@ static int sas_queue_reset(struct domain_device *dev, int reset_type,
>
> int sas_eh_abort_handler(struct scsi_cmnd *cmd)
> {
> - int res;
> - struct sas_task *task = TO_SAS_TASK(cmd);
> + int res = TMF_RESP_FUNC_COMPLETE;
> + struct sas_task *task;
> struct Scsi_Host *host = cmd->device->host;
> struct sas_internal *i = to_sas_internal(host->transportt);
> + struct domain_device *dev = cmd_to_domain_dev(cmd);
> + struct sas_ha_struct *ha = dev->port->ha;
> + unsigned long flags;
>
> if (!i->dft->lldd_abort_task)
> return FAILED;
>
> - res = i->dft->lldd_abort_task(task);
> + /* Avoid sas_scsi_task_done() interfering */
> + spin_lock_irqsave(&dev->done_lock, flags);
> + task = TO_SAS_TASK(cmd);
> + if (test_bit(SAS_HA_FROZEN, &ha->state)) {
> + res = TMF_RESP_FUNC_FAILED;
> + task = NULL;
> + } else
> + ASSIGN_SAS_TASK(cmd, NULL);
> + spin_unlock_irqrestore(&dev->done_lock, flags);
> + if (task)
> + res = i->dft->lldd_abort_task(task);
> if (res == TMF_RESP_FUNC_SUCC || res == TMF_RESP_FUNC_COMPLETE)
> return SUCCESS;
>
> + spin_lock_irqsave(&dev->done_lock, flags);
> + ASSIGN_SAS_TASK(cmd, task);
> + spin_unlock_irqrestore(&dev->done_lock, flags);
Why do you assign task back? As I remember, when this cmd dispatch
again, we will create a new task and assign to it again. So should we
end this task here?
> return FAILED;
> }
> EXPORT_SYMBOL_GPL(sas_eh_abort_handler);
>
next prev parent reply other threads:[~2018-01-09 4:11 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-01-08 12:04 [PATCHv2] libsas: Check for completed commands before calling lldd_abort_task() Hannes Reinecke
2018-01-08 17:11 ` Yves-Alexis Perez
2018-01-09 9:30 ` Hannes Reinecke
2018-01-09 14:13 ` Yves-Alexis Perez
2018-01-09 4:09 ` Jason Yan [this message]
2018-01-09 7:34 ` Hannes Reinecke
2018-01-09 9:04 ` Jason Yan
2018-01-09 9:27 ` Hannes Reinecke
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=5A544075.4010705@huawei.com \
--to=yanaijie@huawei.com \
--cc=corsac@debian.org \
--cc=hare@suse.com \
--cc=hare@suse.de \
--cc=hch@lst.de \
--cc=james.bottomley@hansenpartnership.com \
--cc=linux-scsi@vger.kernel.org \
--cc=martin.petersen@oracle.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox