From: Nikos Tsironis <ntsironis@arrikto.com>
To: Mike Snitzer <snitzer@redhat.com>
Cc: ejt@redhat.com, dm-devel@redhat.com, agk@redhat.com
Subject: Re: [PATCH v2] dm thin: Fix bug wrt FUA request completion
Date: Fri, 15 Feb 2019 19:21:53 +0200 [thread overview]
Message-ID: <57cc8449-e12d-f588-f16d-47cbfde93437@arrikto.com> (raw)
In-Reply-To: <20190215151655.GB20900@redhat.com>
On 2/15/19 5:16 PM, Mike Snitzer wrote:
> On Fri, Feb 15 2019 at 9:33am -0500,
> Nikos Tsironis <ntsironis@arrikto.com> wrote:
>
>> On 2/15/19 3:54 PM, Joe Thornber wrote:
>>> Ack.
>>>
>>> Thanks for this I was under the mistaken impression that FUA requests got split
>>> by core dm into separate payload and PREFLUSH requests.
>>>
>>> I've audited dm-cache and that looks ok.
>>>
>>> How did you test this patch? That missing bio_list_init() in V1 must
>>> have caused memory corruption?
>>>
>>> - Joe
>>
>> Hi Joe,
>>
>> bio_list_init() initializes the bio list's head and tail pointers to
>> NULL and pool_create() allocates the struct pool structure using
>> kzalloc() so the bio list was implicitly correctly initialized and no
>> memory corruption occurred.
>
> Yes, exactly right. v1 tested fine for me, so when I saw v2 I reasoned
> through why the bio_list_init() wasn't an issue and it is like you've
> said (kzalloc() saved us).
>
> Can you help us understand how you identified this issue? Did you have
> corruption after crash/powerfail and got to looking closer?
>
> Thanks,
> Mike
Hi Mike,
I identified the issue through code inspection. My job involves working
with device mapper extensively, so I have been studying its code for
quite some time now. I have already submitted a couple of fixes/performance
improvements to dm-snapshot. I will be following them up with a more
extensive patch set.
In this particular case I was trying to understand how to properly
handle REQ_FUA and REQ_PREFLUSH in the context of periodic metadata
commits, like dm-thin and dm-cache do, and I stumbled on this bug while
reading dm-thin's code.
Nikos
prev parent reply other threads:[~2019-02-15 17:21 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-14 23:21 [PATCH v2] dm thin: Fix bug wrt FUA request completion Nikos Tsironis
2019-02-15 0:10 ` Mike Snitzer
2019-02-15 13:54 ` Joe Thornber
2019-02-15 14:33 ` Nikos Tsironis
2019-02-15 15:16 ` Mike Snitzer
2019-02-15 17:21 ` Nikos Tsironis [this message]
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=57cc8449-e12d-f588-f16d-47cbfde93437@arrikto.com \
--to=ntsironis@arrikto.com \
--cc=agk@redhat.com \
--cc=dm-devel@redhat.com \
--cc=ejt@redhat.com \
--cc=snitzer@redhat.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.