Openembedded Core Discussions
 help / color / mirror / Atom feed
From: "Mathieu Dubois-Briand" <mathieu.dubois-briand@bootlin.com>
To: <sundeep.kokkonda@windriver.com>,
	<openembedded-core@lists.openembedded.org>
Cc: <Randy.MacLeod@windriver.com>
Subject: Re: [OE-core] [PATCH] rust: Upgrade 1.96.1 -> 1.97.0
Date: Fri, 17 Jul 2026 15:03:52 +0200	[thread overview]
Message-ID: <DK0V8ES17T70.PO85FXGM6YP8@bootlin.com> (raw)
In-Reply-To: <20260713085833.3698049-1-sundeep.kokkonda@windriver.com>

On Mon Jul 13, 2026 at 10:58 AM CEST, Sundeep KOKKONDA via lists.openembedded.org wrote:
> From: Sundeep KOKKONDA <sundeep.kokkonda@windriver.com>
>
> Rust Changes:
> https://github.com/rust-lang/rust/releases/tag/1.97.0
>
> Cargo Changes:
> https://doc.rust-lang.org/nightly/cargo/CHANGELOG.html#cargo-197-2026-07-09
>
> Release highlights:
> - v0 symbol mangling scheme now default
> - New allow-by-default `dead_code_pub_in_binary` lint
> - Cargo: stabilized `build.warnings` and `resolver.lockfile-path` configs
> - Multiple new stabilized APIs (isolate_highest_one, bit_width, etc)
>
> Drop merged patches:
> 0001-riscv32-Define-plain-syscalls-as-their-time64-varian.patch
>   (merged upstream in vendored libc 0.2.186)
> 0003-explicit-tail-calls-disable-two-tests.patch
>   (backport from fb9e970c1f93 and 250beb317422)
> 0004-explicit-tail-calls-ignore-tests-on-unsupported-llvm.patch
>   (backport from ded0aaba6ce2)
>
> New patches:
> 0003-tests-assembly-llvm-issue-141649-fix-expected-stack.patch
>   Fix expected stack allocation size for aarch64 with LLVM 22.1.8.
>   LLVM 22.1.8 optimizes non-overlapping scope stack reuse more
>   aggressively (32 bytes instead of 48). This test is new in 1.97.0
>   and was validated against bundled LLVM 22.1.6.
>
> Updated files:
> rust-source.inc: updated sha256sum, removed merged patches, added new patches
> rust-snapshot.inc: updated SNAPSHOT_VERSION 1.95.0 -> 1.96.0 with all arch checksums
> tcmode-default.inc: updated RUSTVERSION 1.96.1% -> 1.97.0%
> rust_1.97.0.bb: remove vendored .a files from installed sources
>   (fix dwarfsrcfiles failure on non-ELF wit-bindgen-0.57.1 archive)
>
> Selftest Changes:
> Exclude tests/ui/codegen/huge-stacks.rs (fails in QEMU emulated
>   environments due to resource limits — new test in 1.97.0.
>   Upstream discussion - https://github.com/rust-lang/rust/issues/158557)
> Fix stale exclusion path: tests/assembly-llvm/c-variadic-arm.rs ->
>   tests/assembly-llvm/c-variadic/arm.rs (file moved to subdirectory)
>
> Excluded tests:
> +---------+---------+---------+
> | Before  |  After  | Skipped |
> +---------+---------+---------+
> |   41    |   42    |   +1    |
> +---------+---------+---------+
>
> rust v1.97.0 (from autobuilder)
> +-------------+---------+---------+
> | Machine     | Passed  | Skipped |
> +-------------+---------+---------+
> | qemux86-64  |  22752  |   762   |
> | qemux86     |  22505  |  1009   |
> | qemuarm64   |  22579  |   935   |
> | qemuarm     |  22486  |  1028   |
> | qemuriscv64 |  22528  |   986   |
> +-------------+---------+---------+
>
> Test results difference (1.97.0 - 1.96.0):
>
> +-------------+--------+---------+
> | Machine     | Passed | Skipped |
> +-------------+--------+---------+
> | qemux86-64  |  +316  |   +30   |
> | qemux86     |  +315  |   +31   |
> | qemuarm64   |  +317  |   +29   |
> | qemuarm     |  +314  |   +32   |
> | qemuriscv64 |  +312  |   +34   |
> +-------------+--------+---------+
>
> Reproducibility: PASSED (x86-64)
>
> Assisted-by: Kiro
> Signed-off-by: Sundeep KOKKONDA <sundeep.kokkonda@windriver.com>
> ---

Hi Sundeep,

I've got this upgrade in my branch for a few days, and I believe it is
the source of some intermittent reproducibility issue.

Some examples:

https://autobuilder.yoctoproject.org/valkyrie/#/builders/37/builds/4393
https://valkyrie.yocto.io/pub/repro-fail/oe-reproducible-20260716-z8el_9yb/packages/diff-html/

https://autobuilder.yoctoproject.org/valkyrie/#/builders/37/builds/4396
https://valkyrie.yocto.io/pub/repro-fail/oe-reproducible-20260716-111q2w5w/packages/diff-html/

I saw you sent a new version upgrading to 1.97.1. I will test it during
the week-end, but I do expect similar issues.

Thanks,
Mathieu

-- 
Mathieu Dubois-Briand, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com



      parent reply	other threads:[~2026-07-17 13:04 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-07-13  8:58 [PATCH] rust: Upgrade 1.96.1 -> 1.97.0 sundeep.kokkonda
2026-07-16 10:11 ` [OE-core] " Alexander Kanavin
2026-07-17  4:16   ` Sundeep KOKKONDA
2026-07-17 13:03 ` Mathieu Dubois-Briand [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=DK0V8ES17T70.PO85FXGM6YP8@bootlin.com \
    --to=mathieu.dubois-briand@bootlin.com \
    --cc=Randy.MacLeod@windriver.com \
    --cc=openembedded-core@lists.openembedded.org \
    --cc=sundeep.kokkonda@windriver.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox