All of lore.kernel.org
 help / color / mirror / Atom feed
From: "jianchao.wang" <jianchao.w.wang@oracle.com>
To: Dongli Zhang <dongli.zhang@oracle.com>
Cc: axboe@kernel.dk, linux-block@vger.kernel.org,
	jsmart2021@gmail.com, sagi@grimberg.me, josef@toxicpanda.com,
	linux-nvme@lists.infradead.org, linux-kernel@vger.kernel.org,
	keith.busch@intel.com, hare@suse.de, jthumshirn@suse.de,
	hch@lst.de, bvanassche@acm.org
Subject: Re: [PATCH V2 2/8] blk-mq: use static_rqs instead of rqs to iterate tags
Date: Mon, 25 Mar 2019 15:14:08 +0800	[thread overview]
Message-ID: <a34ae014-6e99-7c57-42bd-5061dd6ea900@oracle.com> (raw)
In-Reply-To: <58796005-2f96-ad5c-fa21-b3a0071deb06@oracle.com>

Hi Dongli

Thanks for your kindly response.

On 3/25/19 3:12 PM, Dongli Zhang wrote:
> Hi Jianchao,
> 
> On 3/25/19 1:38 PM, Jianchao Wang wrote:
>> tags->rqs[] will not been cleaned when free driver tag to avoid
>> an extra store on a shared area in the per io path. But there
>> is a window between get driver tag and write tags->rqs[], so we
>> may see stale rq in tags->rqs[] which may have been freed, as
>> following case,
>> blk_mq_get_request         blk_mq_queue_tag_busy_iter
>>   -> blk_mq_get_tag
> 
> Sorry I forgot to mention below when I was studying the v1 patchset.
> 
> If there is further review iteration, I would suggest clarify here that
> blk_mq_get_tag() would set the bit in sbitmap (e.g., via __sbitmap_get_word())
> that bt_for_each() would iterate. Otherwise, please ignore my message.
> 
> E.g.,
> 
> blk_mq_get_request
>   -> blk_mq_get_tag
>       -> taint sbitmap tag bit
>          without setting tags->rqs[tag]
> 
>                                           -> bt_for_each
>                                               -> iterate all dirty bit...
> 
> This would be much more friendly for people not working on block layer and when
> they are looking for the potential fix for a bug by simply searching over the
> commit messages. People not working on this would be able to quickly understand
> there is a window between setting bit and setting data.

Yes, I will do that.

> 
> 
>>                              -> bt_for_each
>>                                -> bt_iter
>>                                  -> rq = taags->rqs[]
> 
> %s/taags/tags/g

Yes.


Thanks
Jianchao

> 
>>                                  -> rq->q
>>   -> blk_mq_rq_ctx_init
>>     -> data->hctx->tags->rqs[rq->tag] = rq;
>>
> 
> Thank you very much!
> 
> Dongli Zhang
> 

WARNING: multiple messages have this Message-ID (diff)
From: jianchao.w.wang@oracle.com (jianchao.wang)
Subject: [PATCH V2 2/8] blk-mq: use static_rqs instead of rqs to iterate tags
Date: Mon, 25 Mar 2019 15:14:08 +0800	[thread overview]
Message-ID: <a34ae014-6e99-7c57-42bd-5061dd6ea900@oracle.com> (raw)
In-Reply-To: <58796005-2f96-ad5c-fa21-b3a0071deb06@oracle.com>

Hi Dongli

Thanks for your kindly response.

On 3/25/19 3:12 PM, Dongli Zhang wrote:
> Hi Jianchao,
> 
> On 3/25/19 1:38 PM, Jianchao Wang wrote:
>> tags->rqs[] will not been cleaned when free driver tag to avoid
>> an extra store on a shared area in the per io path. But there
>> is a window between get driver tag and write tags->rqs[], so we
>> may see stale rq in tags->rqs[] which may have been freed, as
>> following case,
>> blk_mq_get_request         blk_mq_queue_tag_busy_iter
>>   -> blk_mq_get_tag
> 
> Sorry I forgot to mention below when I was studying the v1 patchset.
> 
> If there is further review iteration, I would suggest clarify here that
> blk_mq_get_tag() would set the bit in sbitmap (e.g., via __sbitmap_get_word())
> that bt_for_each() would iterate. Otherwise, please ignore my message.
> 
> E.g.,
> 
> blk_mq_get_request
>   -> blk_mq_get_tag
>       -> taint sbitmap tag bit
>          without setting tags->rqs[tag]
> 
>                                           -> bt_for_each
>                                               -> iterate all dirty bit...
> 
> This would be much more friendly for people not working on block layer and when
> they are looking for the potential fix for a bug by simply searching over the
> commit messages. People not working on this would be able to quickly understand
> there is a window between setting bit and setting data.

Yes, I will do that.

> 
> 
>>                              -> bt_for_each
>>                                -> bt_iter
>>                                  -> rq = taags->rqs[]
> 
> %s/taags/tags/g

Yes.


Thanks
Jianchao

> 
>>                                  -> rq->q
>>   -> blk_mq_rq_ctx_init
>>     -> data->hctx->tags->rqs[rq->tag] = rq;
>>
> 
> Thank you very much!
> 
> Dongli Zhang
> 

  reply	other threads:[~2019-03-25  7:14 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-03-25  5:38 [PATCH V2 0/8]: blk-mq: use static_rqs to iterate busy tags Jianchao Wang
2019-03-25  5:38 ` Jianchao Wang
2019-03-25  5:38 ` [PATCH V2 1/8] blk-mq: get rid of the synchronize_rcu in __blk_mq_update_nr_hw_queues Jianchao Wang
2019-03-25  5:38   ` Jianchao Wang
2019-03-25  5:38 ` [PATCH V2 2/8] blk-mq: use static_rqs instead of rqs to iterate tags Jianchao Wang
2019-03-25  5:38   ` Jianchao Wang
2019-03-25  7:12   ` Dongli Zhang
2019-03-25  7:12     ` Dongli Zhang
2019-03-25  7:14     ` jianchao.wang [this message]
2019-03-25  7:14       ` jianchao.wang
2019-03-25  5:38 ` [PATCH V2 3/8] blk-mq: use blk_mq_queue_tag_inflight_iter in debugfs Jianchao Wang
2019-03-25  5:38   ` Jianchao Wang
2019-03-25  5:38 ` [PATCH V2 4/8] mtip32xx: use blk_mq_queue_tag_inflight_iter Jianchao Wang
2019-03-25  5:38   ` Jianchao Wang
2019-03-25  5:38 ` [PATCH V2 5/8] nbd: " Jianchao Wang
2019-03-25  5:38   ` Jianchao Wang
2019-03-25  5:38 ` [PATCH V2 6/8] skd: " Jianchao Wang
2019-03-25  5:38   ` Jianchao Wang
2019-03-25  5:38 ` [PATCH V2 7/8] nvme: " Jianchao Wang
2019-03-25  5:38   ` Jianchao Wang
2019-03-25 13:49   ` Keith Busch
2019-03-25 13:49     ` Keith Busch
2019-03-26  1:17     ` jianchao.wang
2019-03-26  1:17       ` jianchao.wang
2019-03-26  2:41       ` Ming Lei
2019-03-26  2:41         ` Ming Lei
2019-03-26  3:05         ` jianchao.wang
2019-03-26  3:05           ` jianchao.wang
2019-03-26 23:57           ` Keith Busch
2019-03-26 23:57             ` Keith Busch
2019-03-27  2:03             ` jianchao.wang
2019-03-27  2:03               ` jianchao.wang
2019-03-27  2:15               ` Keith Busch
2019-03-27  2:15                 ` Keith Busch
2019-03-27  2:27                 ` jianchao.wang
2019-03-27  2:27                   ` jianchao.wang
2019-03-27  2:33                   ` Keith Busch
2019-03-27  2:33                     ` Keith Busch
2019-03-27  2:45                     ` jianchao.wang
2019-03-27  2:45                       ` jianchao.wang
2019-03-27  6:51                       ` Keith Busch
2019-03-27  6:51                         ` Keith Busch
2019-03-27  7:18                         ` jianchao.wang
2019-03-27  7:18                           ` jianchao.wang
2019-03-25  5:38 ` [PATCH V2 8/8] blk-mq: remove blk_mq_tagset_busy_iter Jianchao Wang
2019-03-25  5:38   ` Jianchao Wang
2019-03-25  7:18   ` Hannes Reinecke
2019-03-25  7:18     ` Hannes Reinecke
2019-03-25  7:37     ` jianchao.wang
2019-03-25  7:37       ` jianchao.wang
2019-03-25  8:25       ` Hannes Reinecke
2019-03-25  8:25         ` Hannes Reinecke
2019-03-25  9:12         ` jianchao.wang
2019-03-25  9:12           ` jianchao.wang
2019-03-26 14:17         ` Jens Axboe
2019-03-26 14:17           ` Jens Axboe
  -- strict thread matches above, loose matches on Subject: below --
2019-03-25  5:28 [PATCH V2 0/8]: blk-mq: use static_rqs to iterate busy tags Jianchao Wang
2019-03-25  5:28 ` [PATCH V2 2/8] blk-mq: use static_rqs instead of rqs to iterate tags Jianchao Wang
2019-03-25  5:28   ` Jianchao Wang

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=a34ae014-6e99-7c57-42bd-5061dd6ea900@oracle.com \
    --to=jianchao.w.wang@oracle.com \
    --cc=axboe@kernel.dk \
    --cc=bvanassche@acm.org \
    --cc=dongli.zhang@oracle.com \
    --cc=hare@suse.de \
    --cc=hch@lst.de \
    --cc=josef@toxicpanda.com \
    --cc=jsmart2021@gmail.com \
    --cc=jthumshirn@suse.de \
    --cc=keith.busch@intel.com \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-nvme@lists.infradead.org \
    --cc=sagi@grimberg.me \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.