[FP8] Expand MaskedSelect to float8 #141928
Conversation
Needed for printing those...
malfet
added
release notes: python_frontend
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/141928
Note: Links to docs will display an error until the docs builds have been completed. ❌ 3 New FailuresAs of commit 22b9821 with merge base 4959784 ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
pytorch-bot
bot
added
the
module: cpu
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Merge failedReason: 1 mandatory check(s) failed. The first few are: Dig deeper by viewing the failures on hud |
|
Dispatch based on element size makes sense IMO... especially for copy/select/fill/comparisons... |
|
@pytorchbot merge -i |
Merge startedYour change will be merged while ignoring the following 3 checks: pull / linux-focal-cuda12.4-py3.10-gcc9-sm89 / test (default, 1, 5, lf.linux.g6.4xlarge.experimental.nvidia.gpu), pull / linux-focal-cuda12.4-py3.10-gcc9-sm89 / test (default, 4, 5, lf.linux.g6.4xlarge.experimental.nvidia.gpu), trunk / linux-focal-cuda12.4-py3.10-gcc9-sm86 / test (default, 5, 5, lf.linux.g5.4xlarge.nvidia.gpu) Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Needed for printing those. Though I wonder if better solution would be to change those ops to use element size rather than actual type (to extend them automatically to unsigned integral types as well) Pull Request resolved: pytorch#141928 Approved by: https://github.com/ezyang, https://github.com/jgong5
Needed for printing those. Though I wonder if better solution would be to change those ops to use element size rather than actual type (to extend them automatically to unsigned integral types as well) Pull Request resolved: pytorch#141928 Approved by: https://github.com/ezyang, https://github.com/jgong5
5 participants
Needed for printing those.
Though I wonder if better solution would be to change those ops to use element size rather than actual type (to extend them automatically to unsigned integral types as well)
cc @jgong5 @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10