On 20-01-2024 14:19, Richard Purdie wrote:
> CAUTION: This email comes from a non Wind River email account!
> Do not click links or open attachments unless you recognize the sender and know the content is safe.
>
> On Sat, 2024-01-20 at 08:42 +0000, Richard Purdie via
> lists.openembedded.org wrote:
>> On Fri, 2024-01-19 at 07:09 -0800, Shinde, Yash via
>> lists.openembedded.org wrote:
>>> From: Yash Shinde<Yash.Shinde@windriver.com>
>>>
>>> Fixes: Exception: no cargo executable found at
>>>         `${B}/rustc-1.74.1-src/build/x86_64-unknown-linux-gnu/stage0/bin/cargo`
>>>
>>> Fix the cargo binary path error on oe-selftest and path set to rust-snapshot dir.
>>>
>>> Patch sent to upstream-https://github.com/rust-lang/rust/pull/120125
>>>
>>> Signed-off-by: Yash Shinde<Yash.Shinde@windriver.com>
>>> ---
>>>   .../rust/files/cargo-path.patch               | 37 +++++++++++++++++++
>>>   1 file changed, 37 insertions(+)
>>>   create mode 100644 meta/recipes-devtools/rust/files/cargo-path.patch
>> The series failed on mips in testing:
>>
>> https://autobuilder.yoctoproject.org/typhoon/#/builders/147/builds/1196/steps/12/logs/stdio
>>
>> Sadly the error redirection is to a file and even in the failed build,
>> we do not have that file as far as I can see so it is hard to know what
>> went wrong.
> I was able to find the file:
>
> https://autobuilder.yocto.io/pub/failed-builds-data/summary.txt
>
> (buildbot-venv) [pokybuild@rocky9-ty-1 build-st-1051644]$ cp ./tmp/work/mips32r2-poky-linux/rust/1.74.1/rustc-1.74.1-src/summary.txt /srv/autobuilder/autobuilder.yocto.io/pub/failed-builds-data/
>
> ---- [ui] tests/ui/issues/issue-32805.rs stdout ----
>
> .
> .
> .
> .
>
> failures:
>      [ui] tests/ui/issues/issue-32805.rs
>      [ui] tests/ui/numbers-arithmetic/i128.rs
>      [ui] tests/ui/numbers-arithmetic/saturating-float-casts.rs
>      [ui] tests/ui/runtime/backtrace-debuginfo.rs
>
> test result: FAILED. 15246 passed; 4 failed; 291 ignored; 0 measured; 21 filtered out; finished in 693.24s
>
> Some tests failed in compiletest suite=ui mode=ui host=x86_64-unknown-linux-gnu target=mips-poky-linux-gnu
> Testing stage1 compiletest suite=run-pass-valgrind mode=run-pass-valgrind (x86_64-unknown-linux-gnu -> mips-poky-linux-gnu)
>
It is observed that "mips(32 bits)" target is pretty unstable with rust 
tests. Thus, we decided unanimously to skip mips(32 bits) target in the 
initial implementation of the rust oe-selftest. I did check the 
testsuite for the mips target and many more failures were found.

According to the rust doc, the mips target variants are classified into 
tier 3 targets( for which the Rust codebase has support for, but which 
the Rust project does not build or test automatically, so they may or 
may not work. 
<https://doc.rust-lang.org/nightly/rustc/platform-support.html#:~:text=Tier%203%20targets%20are%20those%20which%20the%20Rust%20codebase%20has%20support%20for%2C%20but%20which%20the%20Rust%20project%20does%20not%20build%20or%20test%20automatically%2C%20so%20they%20may%20or%20may%20not%20work.>), 
which might be the reason for failures.

https://doc.rust-lang.org/nightly/rustc/platform-support.html#tier-3:~:text=mips%2Dunknown%2Dlinux%2Dgnu,kernel%204.4%2C%20glibc%202.23)


<https://doc.rust-lang.org/nightly/rustc/platform-support.html#tier-3:~:text=mips%2Dunknown%2Dlinux%2Dgnu,kernel%204.4%2C%20glibc%202.23)>Regards,
Yash

>
> Cheers,
>
> Richard