linux-kernel - Re: [PATCH 0/4] Fix issues in check_irq

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:   Mon, 21 Jun 2021 15:28:53 +0800
From:   Boqun Feng <boqun.feng@...il.com>
To:     Johannes Berg <johannes@...solutions.net>
Cc:     Peter Zijlstra <peterz@...radead.org>,
        Ingo Molnar <mingo@...hat.com>, Will Deacon <will@...nel.org>,
        Waiman Long <longman@...hat.com>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 0/4] Fix issues in check_irq_usage()

On Fri, Jun 18, 2021 at 07:36:26PM +0200, Johannes Berg wrote:
> Hi Boqun,
> 

Hello,

> Great, thanks! I'll ask the folks who could reproduce this issue to do
> so as soon as possible.
> 
> Not sure if you saw my previous posting:
> https://lore.kernel.org/lkml/8a61ecda99843307018e3e71a5540682436443fc.camel@sipsolutions.net/T/#u
> 

I just replied that thread on the particular deadlock scenario there.

> That was with patch 3 of this set already applied.
> 
> 
> If I understand correctly, then you're basically saying that if we apply
> all the 3 (or 4) patches here, it'll just change the report (that you
> can see at the link above) to actually say something that I can
> understand to see where the issue is?
> 

Sort of, these patches will provide you a correct lock dependency path
on the deadlock possibility along with the correct call stacks. However,
currently I don't think it's easy for lockdep to print the 4-CPU
scenario that I post in the other email, so it will still take some
effort to decode the lockdep, so you will still need some lockdep
knowledge to understand the real issue ;-(

Regards,
Boqun

> Thanks,
> johannes
> 
>