linux-kernel - Re: [PATCH] workqueue: rust: add creation of workqueues

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: <aAFmO0FhHrzG_F8H@cassiopeiae>
Date: Thu, 17 Apr 2025 22:36:11 +0200
From: Danilo Krummrich <dakr@...nel.org>
To: Alice Ryhl <aliceryhl@...gle.com>
Cc: Tejun Heo <tj@...nel.org>, phasta@...nel.org,
	Miguel Ojeda <ojeda@...nel.org>,
	Lai Jiangshan <jiangshanlai@...il.com>,
	Boqun Feng <boqun.feng@...il.com>, Gary Guo <gary@...yguo.net>,
	Björn Roy Baron <bjorn3_gh@...tonmail.com>,
	Benno Lossin <benno.lossin@...ton.me>,
	Andreas Hindborg <a.hindborg@...nel.org>,
	Trevor Gross <tmgross@...ch.edu>,
	Daniel Almeida <daniel.almeida@...labora.com>,
	Tamir Duberstein <tamird@...il.com>, rust-for-linux@...r.kernel.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH] workqueue: rust: add creation of workqueues

On Thu, Apr 17, 2025 at 10:26:04PM +0200, Alice Ryhl wrote:
> On Thu, Apr 17, 2025 at 9:28 AM Tejun Heo <tj@...nel.org> wrote:
> >
> > On Thu, Apr 17, 2025 at 09:22:40AM +0200, Philipp Stanner wrote:
> > > I assume you, ultimately, mean that the list of delayed_work's would be
> > > accessible through workqueue_struct, correct?
> > >
> > > And then destroy_workqueue() could loop over all of them with
> > > cancel_delayed_work_sync()?
> >
> > Yeap, I was thinking flush_delayed_work() but maybe
> > cancel_delayed_work_sync() is better.
> 
> But doesn't that have a cleanup problem? If the work item owns an
> allocation or a refcount that's cleared by the work item's run
> function, then using cancel_delayed_work_sync() will fail to clean
> that up. Whereas flush_delayed_work() avoids this problem.

I also think it may be a bit unexpected to users if pending "normal" work "will
be done first", but delayed work is canceled instead.

That sounds like a good source for wrong use. :(