linux-kernel - Re: DM Regression in 4.16-rc1 - read() returns data when it shouldn't

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: <c312f639-5fe6-d47f-2395-3985c80c0fc6@leemhuis.info>
Date:   Mon, 26 Feb 2018 18:31:37 +0100
From:   Thorsten Leemhuis <regressions@...mhuis.info>
To:     NeilBrown <neilb@...e.com>, Mike Snitzer <snitzer@...hat.com>
Cc:     Milan Broz <gmazyland@...il.com>,
        device-mapper development <dm-devel@...hat.com>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: DM Regression in 4.16-rc1 - read() returns data when it shouldn't

On 26.02.2018 12:01, NeilBrown wrote:
> On Mon, Feb 26 2018, Thorsten Leemhuis wrote:
>> Hi Mike! On 19.02.2018 18:15, Mike Snitzer wrote:
>>> On Mon, Feb 19 2018 at  8:44am -0500,
>>> Thorsten Leemhuis <regressions@...mhuis.info> wrote:
>>>> JFYI: This issues is tracked in the regression reports for Linux 4.16
>>>> (http://bit.ly/lnxregrep416 ) with this id:
>>>> Linux-Regression-ID: lr#9e195f
>>>> Please include this line in the comment section of patches that are
>>> […]
>>> The fix was already merged by Linus on Friday, see:
>>> git.kernel.org/linus/8dd601fa8317243be887458c49f6c29c2f3d719f
>> Ohh, thx for the pointer. Could you please next time add a tag like
>> Fixes: 18a25da84354 ("dm: ensure bio submission follows a depth-first
>> tree walk")
> 
> The thing is... it didn't fix that commit.  That commit was fine.
> If fixed something else much further back, which that commit just made
> more problematic.
> That is why I added the Cc: stable with the earliest version that needed
> fixing.
> 
> Unfortunately, reality isn't always neat and tidy :-(  when it is, I do
> use Fixes:

Ha, okay, sorry for the noise then.

Side note:  At the same time this is might be a reason why a
"Linux-Regression-ID" or something like that might make sense, because
it shows that even a bisected commit sometimes isn't enough to build a
obvious connection between a regression report and the commit with the
fix...
 Ciao, Thorsten