linux-kernel - Re: [PATCH] Revert "module, async: async_synchronize

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, 2 Feb 2022 17:11:20 -0800
From:   Luis Chamberlain <mcgrof@...nel.org>
To:     Tejun Heo <tj@...nel.org>, Aaron Tomlin <atomlin@...hat.com>
Cc:     Igor Pylypiv <ipylypiv@...gle.com>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        linux-modules@...r.kernel.org, linux-kernel@...r.kernel.org,
        Changyuan Lyu <changyuanl@...gle.com>
Subject: Re: [PATCH] Revert "module, async: async_synchronize_full() on
 module init iff async is used"

On Tue, Feb 01, 2022 at 07:50:08AM -1000, Tejun Heo wrote:
> On Tue, Feb 01, 2022 at 07:43:27AM -1000, Tejun Heo wrote:
> > That's quite a walk down the memory lane and I agree with your analysis. The
> > PF_USED_ASYNC is redundant for correctness with the removal of synchrnous
> > loading from iosched path and the WARN_ON guarantees that nothing in kernel
> > is creating a similar situation.
> > 
> > Acked-by: Tejun Heo <tj@...nel.org>
> 
> BTW, I can route this through workqueue tree but -mm or going to Linus's
> tree directly might be a better option. Any opinions?

FWIW, although I was not working on modules when the reverted commit was
merged, my take on this is concern for a regression, much more than
others. However I do agree that async probe can help for any slowdown,
and the idea was that folks would eventually strive slowly to make more
and more modules default to async probe as they can verify that's fine.

With that said,

Reviewed-by: Luis Chamberlain <mcgrof@...nel.org>

And I'd prefer this go through Linus' tree since this is a fix and
can be merged already. That would help with Aaron's kernel/module.c
refactor work (no functional changes) too as that will soon slice and
dice that to a kernel/module/*.c files.

 Luis