Repository navigation
Register fallback NVTX instruction titles and support compact XLA domain ranges. - #3671
Open
copybara-service[bot] wants to merge 1 commit into
Open
copybara-service[bot] wants to merge 1 commit into
copybara-service[bot] wants to merge 1 commit into
Conversation
copybara-service
Bot
force-pushed
the
test_1007265476
branch
from
October 26, 2026 15:33
a8905dc to
eb567bc
Compare
…ain ranges. `GetInstructionAnnotation` previously passed an unregistered `std::string` to `ScopedAnnotation` outside `ScopedModuleAnnotations`, preventing Nsight Systems `--capture-range=nvtx` from matching fallback thunk ranges. Also, `pgo_nsys_converter.py` used a greedy `hlo_op=(.*)#` regex that captured trailing `,unique_hlo_op_id=...` metadata. Add `tsl::profiler::RangePush(ProfilerDomainHandle, StringHandle)` and cache registered fallback handles in `GetInstructionAnnotation`. Add `tsl::profiler::XlaProfilerDomain()` and `NvtxRangeFormat::kHloOpName` (selected when `XLA_GPU_NVTX_DOMAIN=XLA`) to emit compact `<hlo_op>` NVTX titles in the `XLA` domain while keeping `xprof_name()` in `Thunk:#...#` format. Update `pgo_nsys_converter.py` to stop at `,` or `#` and accept `XLA:<hlo_op>` ranges. PiperOrigin-RevId: 1007265476
copybara-service
Bot
force-pushed
the
test_1007265476
branch
from
October 26, 2026 17:08
eb567bc to
116a8ac
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Register fallback NVTX instruction titles and support compact XLA domain ranges.
GetInstructionAnnotationpreviously passed an unregisteredstd::stringtoScopedAnnotationoutsideScopedModuleAnnotations, preventing Nsight Systems--capture-range=nvtxfrom matching fallback thunk ranges. Also,pgo_nsys_converter.pyused a greedyhlo_op=(.*)#regex that captured trailing,unique_hlo_op_id=...metadata.Add
tsl::profiler::RangePush(ProfilerDomainHandle, StringHandle)and cache registered fallback handles inGetInstructionAnnotation. Addtsl::profiler::XlaProfilerDomain()andNvtxRangeFormat::kHloOpName(selected whenXLA_GPU_NVTX_DOMAIN=XLA) to emit compact<hlo_op>NVTX titles in theXLAdomain while keepingxprof_name()inThunk:#...#format. Updatepgo_nsys_converter.pyto stop at,or#and acceptXLA:<hlo_op>ranges.