Skip to content

Conversation

@bb-sycl
Copy link
Contributor

@bb-sycl bb-sycl commented Dec 5, 2021

Uplift GPU RT version for Linux to 21.48.21782

@yanfeng3721
Copy link
Contributor

The same Regression detected as #5037

[2021-12-05T20:52:43.225Z] # command stderr:
[2021-12-05T20:52:43.225Z] 'ADLS' is not a recognized processor for this target (ignoring processor)
[2021-12-05T20:52:43.225Z] 'ADLS' is not a recognized processor for this target (ignoring processor)
[2021-12-05T20:52:43.225Z] 'ADLS' is not a recognized processor for this target (ignoring processor)
[2021-12-05T20:52:43.225Z] 'ADLS' is not a recognized processor for this target (ignoring processor)
[2021-12-05T20:52:43.225Z] error: GenXLegalization failed for: <  %call.i.i.i.i = select <1 x i1> %cmp.i.i85.i, <1 x i64> %sub.i.i, <1 x i64> %call.i.i.i27.i.i>: 'i64' data type is not supported by this target <ADLS>
[2021-12-05T20:52:43.225Z] llvm-foreach: 
[2021-12-05T20:52:43.225Z] clang-14: error: gen compiler command failed with exit code 1 (use -v to see invocation)
[2021-12-05T20:52:43.225Z] 
[2021-12-05T20:52:43.225Z] error: command failed with exit status: 1

@bader
Copy link
Contributor

bader commented Dec 7, 2021

@kbobrovs, FYI.

@kbobrovs
Copy link
Contributor

kbobrovs commented Dec 8, 2021

@bader, mandelbrot and dgetrf are already re-enabled by intel/llvm-test-suite#600 2 days ago. patch for aot_mixed: intel/llvm-test-suite#605

@vladimirlaz
Copy link
Contributor

/verify with intel/llvm-test-suite#605

@vladimirlaz
Copy link
Contributor

/summary:run

@vladimirlaz vladimirlaz self-assigned this Dec 8, 2021
@vladimirlaz
Copy link
Contributor

/verify with intel/llvm-test-suite#605

@bader
Copy link
Contributor

bader commented Dec 17, 2021

@vladimirlaz, ping.

@vladimirlaz
Copy link
Contributor

/verify with intel/llvm-test-suite#605

@bader
Copy link
Contributor

bader commented Dec 20, 2021

@vladimirlaz, why do we pushing outdated version?
Shouldn't we commit #5129 instead?

@vladimirlaz
Copy link
Contributor

@vladimirlaz, why do we pushing outdated version? Shouldn't we commit #5129 instead?

you are right. closing obsolete PR

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants