linux-kernel - Re: [PATCH] btrfs: fix race between balance and cancel/pause

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]

Date:   Wed, 9 Aug 2023 09:16:00 -0400
From:   Josef Bacik <josef@...icpanda.com>
To:     xiaoshoukui <xiaoshoukui@...il.com>
Cc:     dsterba@...e.cz, clm@...com, dsterba@...e.com,
        linux-btrfs@...r.kernel.org, linux-kernel@...r.kernel.org,
        xiaoshoukui@...jie.com.cn
Subject: Re: [PATCH] btrfs: fix race between balance and cancel/pause

On Mon, Aug 07, 2023 at 10:47:48PM -0400, xiaoshoukui wrote:
> I think this patch does not fully fix the issue.
> 
> This patch just fix assertion panic, but in the race situation, the ioctl pause 
> request still returns an incorrect value 0 to the user which mislead the user the
> pause request finished successfully. In fact, the balance request has not been paused.
> 
> Test results and analysis are as follows:
> https://lore.kernel.org/linux-btrfs/20230726030617.109018-1-xiaoshoukui@gmail.com/T/#me125d17fa59e9e671149cc76d410ced747f488b1

They're just two different issues.  My patch is concerned with the panic, yours
is concerned with getting the correct return value out to the user.

Rebase your patch ontop of Sterba's tree with my fix and send it along, getting
an accurate errno out to the user is a reasonable goal.  Thanks,

Josef