linux-kernel - Re: [PATCH v9 00/29] Generic `Allocator` support for Rust

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] [day] [month] [year] [list]

Message-ID: <20241116142102.GA1484376@robin.jannau.net>
Date: Sat, 16 Nov 2024 15:21:02 +0100
From: Janne Grunau <j@...nau.net>
To: Alice Ryhl <aliceryhl@...gle.com>
Cc: Asahi Lina <lina@...hilina.net>, Danilo Krummrich <dakr@...nel.org>,
	ojeda@...nel.org, alex.gaynor@...il.com, wedsonaf@...il.com,
	boqun.feng@...il.com, gary@...yguo.net, bjorn3_gh@...tonmail.com,
	benno.lossin@...ton.me, a.hindborg@...sung.com,
	akpm@...ux-foundation.org, daniel.almeida@...labora.com,
	faith.ekstrand@...labora.com, boris.brezillon@...labora.com,
	mcanal@...lia.com, zhiw@...dia.com, cjia@...dia.com,
	jhubbard@...dia.com, airlied@...hat.com, ajanulgu@...hat.com,
	lyude@...hat.com, linux-kernel@...r.kernel.org,
	rust-for-linux@...r.kernel.org, linux-mm@...ck.org,
	asahi@...ts.linux.dev
Subject: Re: [PATCH v9 00/29] Generic `Allocator` support for Rust

On Fri, Nov 15, 2024 at 03:00:43PM +0100, Alice Ryhl wrote:
> On Fri, Nov 15, 2024 at 12:33 PM Asahi Lina <lina@...hilina.net> wrote:
> >
> > We (Asahi Linux) noticed that this doesn't support unsized types, which
> > is a regression from the alloc Box. drm/asahi uses boxed trait objects
> > to generalize across certain object types like render commands, so this
> > breaks the driver.
> >
> > Is there a plan to add this back in? AIUI this would need the
> > coerce_unsized unstable feature, but is there any other blocker?
> 
> We're already using coerce_unsized for Arc, so I see no reason we
> can't just add that for KBox too. Soon, we should be able to use the
> derive macro [1] I proposed to do this without unstable language
> features. The limitation that the RFC mentions for custom allocators
> does not apply to KBox because it uses PhantomData<A> instead of A for
> the allocator field.

I have branch rebased onto rust-next and Danilo's "Device / Driver PCI /
Platform Rust abstractions". In addition to coerce_unsized for kbox
kvec misses the unproblematic methods clear and truncate and
array_try_from_vec. drain() as taken from the rust library looks like
more effort considering kvec's IntoIter.

Janne