linux-kernel - Re: [PATCH v2] rust: Fix build error

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: <ZyzoR4JLPOm9Pi_z@Boquns-Mac-mini.local>
Date: Thu, 7 Nov 2024 08:18:15 -0800
From: Boqun Feng <boqun.feng@...il.com>
To: Eder Zulian <ezulian@...hat.com>
Cc: rust-for-linux@...r.kernel.org, linux-kernel@...r.kernel.org,
	miguel.ojeda.sandonis@...il.com, tglx@...utronix.de,
	williams@...hat.com, ojeda@...nel.org, alex.gaynor@...il.com,
	gary@...yguo.net, bjorn3_gh@...tonmail.com, benno.lossin@...ton.me,
	a.hindborg@...nel.org, aliceryhl@...gle.com, tmgross@...ch.edu,
	jlelli@...hat.com
Subject: Re: [PATCH v2] rust: Fix build error

On Thu, Nov 07, 2024 at 08:15:15AM +0100, Eder Zulian wrote:
[...]
> > > Fixes: 876346536c1b ("rust: kbuild: split up helpers.c")
> > 
> > I'm not sure this is the correct "Fixes" tag, that commit is a code
> > move, so it's unlikely introducing issue itself. Moreover, we really
> > need PREEMPT_RT being able to trigger the issue, so I think the correct
> 
> One may argue that we need 'RUST=y' in order to trigger the issue.
> 

But RUST support was in mainline earlier than PREEMPT_RT enablement
(again I know we have RT code quite earlier than Rust support, but we
are talking about mainline and potential stable backporting here), so
when the lock support in Rust was added, although the code was missing
RT support, but it's fine from a mainline PoV, and when we really
enabled PREEMPT_RT, we should have added the missing piece.

In other words, would we want to backport this fix into an early version
(say 6.6 stable) where RT has not been enabled? Would there be users who
want to use RT and Rust in that version?

Regards,
Boqun

> > "Fixes" tag is:
> > 
> > Fixes: d2d6422f8bd1 ("x86: Allow to enable PREEMPT_RT.")
> > 
> > (yes, I know PREEMPT_RT is a long existing project, but it was until
> > that commit, you can build a kernel with PREEMPT_RT=y IIUC)
> > 
> > This will help stable maintainers for backport decisions.
> > 
> 
> Perhaps omitting the 'Fixes:' tag would be a solution. Is that an option?
> 
> > 
> > The rest of patch looks good to me (we could maybe provide a
> > __spin_lock_init() for !RT build as well, but that's more of a
> > cleanup)
> > 
> > Regards,
> > Boqun
> > 
> 
> Thanks,
> Eder
> 
[...]