All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ming Lei <ming.lei@redhat.com>
To: Hannes Reinecke <hare@suse.de>
Cc: Jens Axboe <axboe@kernel.dk>, Hannes Reinecke <hare@suse.com>,
	Bart van Assche <bvanassche@acm.org>,
	Ming Lei <ming.lei@gmail.com>,
	linux-nvme@lists.infradead.org, linux-block@vger.kernel.org,
	Christoph Hellwig <hch@lst.de>,
	neilb@suse.com
Subject: Re: [PATCH] block: use static bio_set for bio_split() calls
Date: Thu, 18 Apr 2019 22:34:30 +0800	[thread overview]
Message-ID: <20190418143429.GA19175@ming.t460p> (raw)
In-Reply-To: <20190418140632.60606-1-hare@suse.de>

Hi Hannes,

On Thu, Apr 18, 2019 at 04:06:32PM +0200, Hannes Reinecke wrote:
> When calling blk_queue_split() it will be using the per-queue
> bioset to allocate the split bio from. However, blk_steal_bios()
> might move the bio to another queue, _and_ the original queue
> might be removed completely (nvme is especially prone to do so).

Could you explain a bit how the original queue is removed in case
that blk_steal_bios() is involved?

> That leaves the bvecs of the split bio with a missing / destroyed
> mempool, and a really fun crash in bio_endio().

per-queue bioset is used originally for avoiding deadlock, are you
sure the static bioset is safe?


Thanks,
Ming

WARNING: multiple messages have this Message-ID (diff)
From: ming.lei@redhat.com (Ming Lei)
Subject: [PATCH] block: use static bio_set for bio_split() calls
Date: Thu, 18 Apr 2019 22:34:30 +0800	[thread overview]
Message-ID: <20190418143429.GA19175@ming.t460p> (raw)
In-Reply-To: <20190418140632.60606-1-hare@suse.de>

Hi Hannes,

On Thu, Apr 18, 2019@04:06:32PM +0200, Hannes Reinecke wrote:
> When calling blk_queue_split() it will be using the per-queue
> bioset to allocate the split bio from. However, blk_steal_bios()
> might move the bio to another queue, _and_ the original queue
> might be removed completely (nvme is especially prone to do so).

Could you explain a bit how the original queue is removed in case
that blk_steal_bios() is involved?

> That leaves the bvecs of the split bio with a missing / destroyed
> mempool, and a really fun crash in bio_endio().

per-queue bioset is used originally for avoiding deadlock, are you
sure the static bioset is safe?


Thanks,
Ming

  reply	other threads:[~2019-04-18 14:34 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-04-18 14:06 [PATCH] block: use static bio_set for bio_split() calls Hannes Reinecke
2019-04-18 14:06 ` Hannes Reinecke
2019-04-18 14:34 ` Ming Lei [this message]
2019-04-18 14:34   ` Ming Lei
2019-04-18 15:02   ` Hannes Reinecke
2019-04-18 15:02     ` Hannes Reinecke
2019-04-18 18:26     ` Edmund Nadolski (Microsoft)
2019-04-18 18:26       ` Edmund Nadolski (Microsoft)
2019-04-24 17:20   ` Sagi Grimberg
2019-04-24 17:20     ` Sagi Grimberg
2019-04-24 18:56     ` Hannes Reinecke
2019-04-24 18:56       ` Hannes Reinecke
2019-04-24 18:58       ` Sagi Grimberg
2019-04-24 18:58         ` Sagi Grimberg
2019-04-24 22:14     ` Ming Lei
2019-04-24 22:14       ` Ming Lei
2019-04-25  0:41       ` Ming Lei
2019-04-25  0:41         ` Ming Lei
2019-04-25 14:32         ` Hannes Reinecke
2019-04-25 14:32           ` Hannes Reinecke
2019-04-25 15:36           ` Ming Lei
2019-04-25 15:36             ` Ming Lei
2019-04-30 11:48             ` Hannes Reinecke
2019-04-30 11:48               ` Hannes Reinecke
2019-04-24 19:49 ` Bart Van Assche
2019-04-24 19:49   ` Bart Van Assche
2019-04-25  6:06   ` Hannes Reinecke
2019-04-25  6:06     ` 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=20190418143429.GA19175@ming.t460p \
    --to=ming.lei@redhat.com \
    --cc=axboe@kernel.dk \
    --cc=bvanassche@acm.org \
    --cc=hare@suse.com \
    --cc=hare@suse.de \
    --cc=hch@lst.de \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-nvme@lists.infradead.org \
    --cc=ming.lei@gmail.com \
    --cc=neilb@suse.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 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.