linux-kernel - Re: [REGRESSION] ioprio performance hangs, bisected

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 for Android: free password hash cracker in your pocket

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

Message-ID: <09db95ba-b396-4734-9ff0-9331579c92b7@kernel.dk>
Date: Sat, 30 Nov 2024 09:00:20 -0700
From: Jens Axboe <axboe@...nel.dk>
To: David Wang <00107082@....com>, chris.bainbridge@...il.com
Cc: bvanassche@....org, hch@....de, linux-block@...r.kernel.org,
 linux-kernel@...r.kernel.org, regressions@...ts.linux.dev,
 semen.protsenko@...aro.org
Subject: Re: [REGRESSION] ioprio performance hangs, bisected

On 11/29/24 11:09 PM, David Wang wrote:
> Would fix/revert reach 6.13-rc1? I think this regression has
> significant nagative impact on daily usage. From time to time,
> my system would *stuck* for seconds, sometimes even when
>  just `cat` some small file it would take seconds, and once,
> my desktop failed to resume from a suspend, I had to power cycle the system;
> Polling /proc/diskstats,  I noticed Block-IO read latency,
> delta(# of milliseconds spent reading)/delta(# of reads completed),
> is very high, ~200ms average, and frequently reaches 10s.
>  (Strangely block-io write latency seems normal.)
> My bisect also land on this commit, revert or apply this patch would
> fix it.
> 
> Kind of think that 6.13-rc1 would be very unpleseant to use without
> a fix/revert for this.

The pull including this fix has been sent off to Linus yesterday,
it should make -rc1.

-- 
Jens Axboe