User Details
- User Since
- Dec 14 2019, 8:54 PM (132 w, 4 d)
Sun, Jun 26
Please add a test case as well for the new op.
Sat, Jun 25
Wed, Jun 22
@herhut - would you be able to also review this?
Mon, Jun 13
Sun, Jun 5
Thu, Jun 2
May 27 2022
This is a great improvement - thanks!
May 25 2022
LGTM - should have definitely been a regular pass. Some minor comments.
May 18 2022
May 17 2022
LGTM - thanks. There aren't indeed any use cases to attach Values to local identifiers. Some minor comments.
May 14 2022
May 13 2022
May 12 2022
LGTM
May 8 2022
Looking mostly good. Some minor comments and a question to under the async dep related guard.
Apr 29 2022
Apr 22 2022
Apr 21 2022
Adjust error message.
Apr 20 2022
Rebase.
Apr 19 2022
Any more comments here @csigg ?
Any other comments here: @rriddle?
Apr 17 2022
Update cmd line flag name and added test case.
Apr 16 2022
Apr 15 2022
Title: spec -> doc
Fix typos.
Apr 14 2022
Switch to {}.
Apr 13 2022
Add a couple more test cases for the outlining pass.
Missed updates for gpu.launch -> gpu.launch_func.
Fix semantics and syntax to allow async without any deps.
This is looking good to me. Thank you for contributing these. Mostly minor comments/requests for documentation.
Apr 12 2022
/vol/worker/mlir-nvidia/mlir-nvidia/llvm.obj/bin/mlir-cpu-runner: symbol lookup error: /vol/worker/mlir-nvidia/mlir-nvidia/llvm.obj/lib/libmlir_cuda_runtime.so: undefined symbol: cuMemAllocAsync
The bot has an old version of CUDA that doesn't support async alloc. We already guarded it with >= 11.2 above, and I double-checked again that a 11.2 or higher version should have had that method. Looking at the logs, the bot has CUDA_VERSION=10.2.89 in its env (not sure what the preprocessor sees). Either the format is different which is tripping the macro check or the build is being compiled with newer headers but is being linked with older libraries. Someone with access to the bot will have to see what CUDA_VERSION is being set to (for the build preprocessor).
Apr 11 2022
Base it on the right commit.
Rebase.
Sorted order.
Add AsyncOpInterface to gpu.launch and move common methods in GPUDialect.cpp up.
Update commit summary.