Fix learning_rate precision in SGD GPU kernel to use MT instead of T #25876
Triggered via pull request
April 23, 2026 03:26
zhanghonggeng
Status
Failure
Total duration
2h 13m 11s
Artifacts
–
CI.yml
on: pull_request
Check file paths
1m 7s
Distribute-stable-build
/
...
/
Check bypass
8s
Linux-IXUCA
/
Build and Test
10m 28s
Mac-CPU
/
Build and test
20m 57s
Distribute-stable-build
/
Build
31m 24s
PR-CI-SOT
/
Build and Test
54m 40s
Linux-DCU
/
Test
1h 32m
Linux-XPU
/
Test
29m 37s
Linux-NPU
/
Test
7m 32s
Annotations
7 errors and 3 warnings
|
Linux-IXUCA / Build and Test
Executing the custom container implementation failed. Please contact your self hosted runner administrator.
|
|
Linux-IXUCA / Build and Test
Process completed with exit code 1.
|
|
Linux-IXUCA / Build and Test
Error: failed to run script step: TypeError: Cannot read properties of null (reading 'jobPod')
|
|
Linux-IXUCA / Build and Test
Executing the custom container implementation failed. Please contact your self hosted runner administrator.
|
|
Linux-IXUCA / Build and Test
Process completed with exit code 1.
|
|
Linux-IXUCA / Build and Test
Error: pod failed to come online with error: Error: Pod iluvatar-gpu-2-4bfnv-runner-jmmmx-workflow is unhealthy with phase status Pending
|
|
Linux-DCU / Test
Process completed with exit code 8.
|
|
Mac-CPU / Build and test
libtool 2.5.4 is already installed and up-to-date.
To reinstall 2.5.4, run:
brew reinstall libtool
|
|
Mac-CPU / Build and test
automake 1.18.1 is already installed and up-to-date.
To reinstall 1.18.1, run:
brew reinstall automake
|
|
Mac-CPU / Build and test
autoconf 2.73 is already installed and up-to-date.
To reinstall 2.73, run:
brew reinstall autoconf
|