From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from aws-us-west-2-korg-lkml-1.web.codeaurora.org (localhost.localdomain [127.0.0.1]) by smtp.lore.kernel.org (Postfix) with ESMTP id 38ABDE9A02C for ; Wed, 18 Feb 2026 17:26:13 +0000 (UTC) Subject: Re: [PATCH] rust: Enable dynamic linking with llvm To: openembedded-core@lists.openembedded.org From: "Dora, Sunil Kumar" X-Originating-Location: Bengaluru, Karnataka, IN (106.51.216.76) X-Originating-Platform: Windows Chrome 144 User-Agent: GROUPS.IO Web Poster MIME-Version: 1.0 Date: Wed, 18 Feb 2026 09:26:11 -0800 References: <20260212165035.1806068-1-sunilkumar.dora@windriver.com> <675892.1771012298240808740@lists.openembedded.org> <5f3097df-5e24-41f8-9293-db01248e383c@windriver.com> In-Reply-To: <5f3097df-5e24-41f8-9293-db01248e383c@windriver.com> Message-ID: <976428.1771435571945028979@lists.openembedded.org> Content-Type: multipart/alternative; boundary="XzRH7EdURpM3u0dLm0JL" List-Id: X-Webhook-Received: from 45-33-107-173.ip.linodeusercontent.com [45.33.107.173] by aws-us-west-2-korg-lkml-1.web.codeaurora.org with HTTPS for ; Wed, 18 Feb 2026 17:26:13 -0000 X-Groupsio-URL: https://lists.openembedded.org/g/openembedded-core/message/231312 --XzRH7EdURpM3u0dLm0JL Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable On Tue, Feb 17, 2026 at 07:54 PM, Randy MacLeod wrote: >=20 > I'm still a bit uncomfortable with changing the way that target Rust is > built to deal with > a Rust-native problem. Also there's the question about dynamic linking > increasing runtime > by a small amount. Finally, I worry that few people/distros change the > default linking to by dynamic. >=20 > So I won't object to this work-around especially if we can easily limit i= t > to -native. Have you tried that? Hi Randy, I=E2=80=99ve now limited the change as you suggested: - rust-native links with LLVM dynamically - rust-target links with LLVM statically This keeps the workaround only in -native and avoids changing how the targe= t toolchain is built. I reran the relevant tests and didn=E2=80=99t see any regressions or runtim= e issues. I=E2=80=99ll send V3 shortly with this approach. Thanks, Sunil Dora --XzRH7EdURpM3u0dLm0JL Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: quoted-printable
On Tue, Feb 17, 2026 at 07:54 PM, Randy MacLeod wrote:
I'm still a bit uncomfortable with changing the way that target= Rust is built to deal with
a Rust-native problem. Also there's the qu= estion about dynamic linking increasing runtime
by a small amount. Fin= ally, I worry that few people/distros change the default linking to by dyna= mic.

So I won't object to this work-around especially if we can = easily limit it to -native. Have you tried that?
Hi Randy,
 
I’ve now limited the change as you suggested:
  - rust-native links with LLVM dynamically
  - rust-target links with LLVM statically

This keeps the workaround only in -native and avoids changing ho= w the target toolchain is built.
I reran the relevant tests and didn’t see any regressions or run= time issues.
I’ll send V3 shortly with this approach.

Thanks,
Sunil Dora
--XzRH7EdURpM3u0dLm0JL--