linux-kernel - Re: remove dead bdi congestion leftovers

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20200701182538.GU25523@casper.infradead.org>
Date:   Wed, 1 Jul 2020 19:25:38 +0100
From:   Matthew Wilcox <willy@...radead.org>
To:     Mike Snitzer <snitzer@...hat.com>
Cc:     Christoph Hellwig <hch@....de>, Jens Axboe <axboe@...nel.dk>,
        linux-raid@...r.kernel.org, linux-mm@...ck.org,
        linux-bcache@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-block@...r.kernel.org, drbd-dev@...n.linbit.com,
        dm-devel@...hat.com, Tejun Heo <tj@...nel.org>,
        cgroups@...r.kernel.org, linux-btrfs@...r.kernel.org
Subject: Re: remove dead bdi congestion leftovers

On Wed, Jul 01, 2020 at 06:57:47PM +0100, Matthew Wilcox wrote:
> On Wed, Jul 01, 2020 at 12:41:03PM -0400, Mike Snitzer wrote:
> > On Wed, Jul 01 2020 at  5:06am -0400,
> > Christoph Hellwig <hch@....de> wrote:
> > 
> > > Hi Jens,
> > > 
> > > we have a lot of bdi congestion related code that is left around without
> > > any use.  This series removes it in preparation of sorting out the bdi
> > > lifetime rules properly.
> > 
> > I could do some git archeology to see what the fs, mm and block core
> > changes were to stop using bdi congested but a pointer to associated
> > changes (or quick recap) would save me some time.
> > 
> > Also, curious to know how back-pressure should be felt back up the IO
> > stack now? (apologies if these are well worn topics, I haven't been
> > tracking this area of development).
> 
> It isn't.  Jens declared the implementation was broken, and broke it
> more.  So we're just living with stupid broken timeouts.

Here's a thread about it.  This would have been a discussion topic at
LSFMM2020, but COVID.

https://lore.kernel.org/linux-mm/20190917115824.16990-1-linf@wangsu.com/T/#u