-
Notifications
You must be signed in to change notification settings - Fork 661
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[LLVMGPU] Add a verifier for tile sizes. #19906
base: main
Are you sure you want to change the base?
[LLVMGPU] Add a verifier for tile sizes. #19906
Conversation
Signed-off-by: erman-gurses <[email protected]>
Signed-off-by: erman-gurses <[email protected]>
Signed-off-by: erman-gurses <[email protected]>
Signed-off-by: erman-gurses <[email protected]>
Signed-off-by: erman-gurses <[email protected]>
Signed-off-by: erman-gurses <[email protected]>
a462b4f
to
ae7e5a6
Compare
0a7e8fd
to
8a9f54d
Compare
Signed-off-by: erman-gurses <[email protected]>
8a9f54d
to
09f2633
Compare
getLoweringConfig<IREE::Codegen::LoweringConfigAttr>(op); | ||
if (!loweringConfig) | ||
auto gpuLoweringConfig = | ||
getLoweringConfig<IREE::GPU::LoweringConfigAttr>(op); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Rather than casting and passing two lowering config attributes, it would be cleaner to add an additional template parameter to this function (template <typename F, typename ConfigTy>
) and only cast to the config type that works for F
. Then rather than combining the verifiers for LLVMGPUMatmulTensorCoreMmaSync
and LLVMGPUVectorDistribute
, we have two separate verifyGPUMatmulPipeline
functions, one for the GPU lowering config and one for the Codegen lowering config.
Add a verifier for tile sizes.