Received-SPF: pass (google.com: domain of linux-ext4-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Fri, 9 Dec 2022 01:53:02 -0500
From:   Sasha Levin <sashal@kernel.org>
To:     Thorsten Leemhuis <regressions@leemhuis.info>
Cc:     Theodore Ts'o <tytso@mit.edu>, Jan Kara <jack@suse.cz>,
        Andreas Dilger <adilger.kernel@dilger.ca>,
        linux-ext4@vger.kernel.org, stable@vger.kernel.org,
        Thilo Fromm <t-lo@linux.microsoft.com>,
        Jeremi Piotrowski <jpiotrowski@linux.microsoft.com>,
        Andreas Gruenbacher <agruenba@redhat.com>
Subject: Re: [PATCH] ext4: Fix deadlock due to mbcache entry corruption
Message-ID: <Y5LbTkjORxVhgpKy@sashalap>
References: <20221123193950.16758-1-jack@suse.cz>
 <20221201151021.GA18380@linuxonhyperv3.guj3yctzbm1etfxqx2vob5hsef.xx.internal.cloudapp.net>
 <9c414060-989d-55bb-9a7b-0f33bf103c4f@leemhuis.info>
 <Y5F8ayz4gEtKn0LF@mit.edu>
 <20221208091523.t6ka6tqtclcxnsrp@quack3>
 <Y5IFR4K9hO8ax1Y0@mit.edu>
 <e2a77778-7a2b-2811-95ff-be67a44afceb@leemhuis.info>
 <Y5LR2ffwz39donWu@mit.edu>
 <cf8401c8-d9cb-c0be-890b-6aa14d06c1d2@leemhuis.info>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Disposition: inline
In-Reply-To: <cf8401c8-d9cb-c0be-890b-6aa14d06c1d2@leemhuis.info>
Precedence: bulk

On Fri, Dec 09, 2022 at 07:31:03AM +0100, Thorsten Leemhuis wrote:
>On 09.12.22 07:12, Theodore Ts'o wrote:
>> On Thu, Dec 08, 2022 at 06:16:02PM +0100, Thorsten Leemhuis wrote:
>>>
>>> Maybe I should talk to Greg again to revert backported changes like
>>> 1be97463696c until fixes for them are ready.
>>
>> The fix is in the ext4 git tree, and it's ready to be pushed to Linus
>> when the merge window opens --- presumably, on Sunday.
>
>Thx!
>
>> So it's probably not worth it to revert the backported change, only to
>> reapply immediately afterwards.
>
>Definitely agreed, I was more taking in the general sense (sorry, should
>have been clearer), as it's not the first time some backport exposes
>existing problems that take a while to get analyzed and fixed in
>mainline. Which is just how it is sometimes, hence a revert and a
>reapply of that backport (once the fix is available) in stable/longterm
>sounds appropriate to me to prevent users from running into known problems.

It's a balancing act: reverting a fix would mean that we reintroduce an
issue that was previously fixed back to users. It's not always the right
thing to do, and sometimes we won't.

-- 
Thanks,
Sasha