update rocprofiler-sdk with stuff in 0.8.0 by cj401-amd · Pull Request #513 · ROCm/xla

cj401-amd · 2026-01-09T15:21:38Z

Motivation

backporting rocprofiler-sdk from 0.8.0 update rocprofiler-sdk (v3) and roctracer (v1) #473
still no kernel details in the trace file when building from rocm-jax.
running python3 profiler_test.py from jax ci_cj_profiler_test_rocm-jaxlib-v0.8.0, which requires xprof (it may be needed for CI later)

[ RUN      ] ProfilerTest.test_rocm_gpu_events_present_for_many_matmul_shapes
W0109 15:19:23.492738 140051182680960 raw_to_tool_data.py:103] Received old tool format: trace_viewer@^; mapped to new format: trace_viewer@
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
W0000 00:00:1767971963.496396 1297529 derived_timeline.cc:590] Found derived XLine with clashing display ID: 1. This will cause rendering issues in Trace Viewer.
W0000 00:00:1767971963.498741 1297529 trace_events.cc:433] Remapping device id 1for host 1 to 1001
W0000 00:00:1767971963.498752 1297529 trace_events.cc:433] Remapping device id 701for host 1 to 1701
W0109 15:19:27.411699 140051182680960 raw_to_tool_data.py:103] Received old tool format: trace_viewer@^; mapped to new format: trace_viewer@
W0000 00:00:1767971967.413357 1297529 derived_timeline.cc:590] Found derived XLine with clashing display ID: 1. This will cause rendering issues in Trace Viewer.
W0000 00:00:1767971967.415660 1297529 trace_events.cc:433] Remapping device id 701for host 1 to 1701
W0000 00:00:1767971967.415669 1297529 trace_events.cc:433] Remapping device id 1for host 1 to 1001
W0109 15:19:32.673717 140051182680960 raw_to_tool_data.py:103] Received old tool format: trace_viewer@^; mapped to new format: trace_viewer@
W0000 00:00:1767971972.678285 1297529 derived_timeline.cc:590] Found derived XLine with clashing display ID: 1. This will cause rendering issues in Trace Viewer.
W0000 00:00:1767971972.680584 1297529 trace_events.cc:433] Remapping device id 701for host 1 to 1701
W0000 00:00:1767971972.680594 1297529 trace_events.cc:433] Remapping device id 1for host 1 to 1001
W0109 15:19:36.813895 140051182680960 raw_to_tool_data.py:103] Received old tool format: trace_viewer@^; mapped to new format: trace_viewer@
W0000 00:00:1767971976.815375 1297529 derived_timeline.cc:590] Found derived XLine with clashing display ID: 1. This will cause rendering issues in Trace Viewer.
W0000 00:00:1767971976.818270 1297529 trace_events.cc:433] Remapping device id 701for host 1 to 1701
W0000 00:00:1767971976.818290 1297529 trace_events.cc:433] Remapping device id 1for host 1 to 1001
W0109 15:19:42.465533 140051182680960 raw_to_tool_data.py:103] Received old tool format: trace_viewer@^; mapped to new format: trace_viewer@
W0000 00:00:1767971982.467098 1297529 derived_timeline.cc:590] Found derived XLine with clashing display ID: 1. This will cause rendering issues in Trace Viewer.
W0000 00:00:1767971982.470008 1297529 trace_events.cc:433] Remapping device id 1for host 1 to 1001
W0000 00:00:1767971982.470021 1297529 trace_events.cc:433] Remapping device id 701for host 1 to 1701
W0109 15:19:46.431867 140051182680960 raw_to_tool_data.py:103] Received old tool format: trace_viewer@^; mapped to new format: trace_viewer@
W0000 00:00:1767971986.433316 1297529 derived_timeline.cc:590] Found derived XLine with clashing display ID: 1. This will cause rendering issues in Trace Viewer.
W0000 00:00:1767971986.435200 1297529 trace_events.cc:433] Remapping device id 1for host 1 to 1001
W0000 00:00:1767971986.435211 1297529 trace_events.cc:433] Remapping device id 701for host 1 to 1701
W0109 15:19:50.702192 140051182680960 raw_to_tool_data.py:103] Received old tool format: trace_viewer@^; mapped to new format: trace_viewer@
W0000 00:00:1767971990.704211 1297529 derived_timeline.cc:590] Found derived XLine with clashing display ID: 1. This will cause rendering issues in Trace Viewer.
W0000 00:00:1767971990.707054 1297529 trace_events.cc:433] Remapping device id 701for host 1 to 1701
W0000 00:00:1767971990.707074 1297529 trace_events.cc:433] Remapping device id 1for host 1 to 1001
[       OK ] ProfilerTest.test_rocm_gpu_events_present_for_many_matmul_shapes
----------------------------------------------------------------------
Ran 21 tests in 63.537s

i-chaochen · 2026-01-10T16:40:32Z

+build:ci_multi_gpu --experimental_guard_against_concurrent_changes
+build:ci_multi_gpu --test_env=HIP_VISIBLE_DEVICES=0,1,2,3
+build:ci_multi_gpu --strategy=TestRunner=local
+


why your rocprof-sdk PR has this change? I don't think this is relevant and don't put any others in this profiling backport PR

why your rocprof-sdk PR has this change? I don't think this is relevant and don't put any others in this profiling backport PR

those were mainly for CI test. Otherwise, it failed straightaway as ci_single/mult_gpu requires those definition. it seems we still got failures that are not related to the backported stuff.

If this is only for CI test due to #453 and f6785ef , please create another PR

cc @alekstheod

six failed ci_multi_gpu tests on a local MI300

bazel --bazelrc=/work/xla/build_tools/rocm/rocm_xla.bazelrc test \ --config=rocm_ci \ --config=ci_multi_gpu \ --test_output=errors \ --spawn_strategy=local \ --strategy=TestRunner=local \ --repo_env=TF_ROCM_AMDGPU_TARGETS=gfx942,gfx90a \ //xla/tests:array_elementwise_ops_test_amdgpu_any

//xla/tests:array_elementwise_ops_test_amdgpu_any \ //xla/tests:convert_test_amdgpu_any \ //xla/tests:dot_operation_test_autotune_disabled_amdgpu_any \ //xla/tests:iota_test_amdgpu_any \ //xla/service/gpu/transforms:command_buffer_scheduling_test_amdgpu_any \ //xla/tests:local_client_execute_test_amdgpu_any

//xla/tests:array_elementwise_ops_test_amdgpu_any

INFO: Found 1 test target... Target //xla/tests:array_elementwise_ops_test_amdgpu_any up-to-date: bazel-bin/xla/tests/array_elementwise_ops_test_amdgpu_any INFO: Elapsed time: 240.038s, Critical Path: 237.85s INFO: 12 processes: 7 internal, 5 local. INFO: Build completed successfully, 12 total actions //xla/tests:array_elementwise_ops_test_amdgpu_any PASSED in 212.8s Executed 1 out of 1 test: 1 test passes.

//xla/tests:convert_test_amdgpu_any

-- Test timed out at 2026-01-12 18:13:37 UTC -- ================================================================================ INFO: Found 1 test target... Target //xla/tests:convert_test_amdgpu_any up-to-date: bazel-bin/xla/tests/convert_test_amdgpu_any INFO: Elapsed time: 928.377s, Critical Path: 681.32s INFO: 8616 processes: 75 internal, 8541 local. INFO: Build completed successfully, 8616 total actions //xla/tests:convert_test_amdgpu_any FLAKY, failed in 1 out of 2 in 300.7s Stats over 2 runs: max = 300.7s, min = 222.3s, avg = 261.5s, dev = 39.2s /root/.cache/bazel/_bazel_root/ea1efa0977f8828bf242d5b6a382af7f/execroot/xla/bazel-out/k8-opt/testlogs/xla/tests/convert_test_amdgpu_any/test_attempts/attempt_1.log Executed 1 out of 1 test: 1 test passes.

//xla/tests:dot_operation_test_autotune_disabled_amdgpu_any

INFO: Found 1 test target... Target //xla/tests:dot_operation_test_autotune_disabled_amdgpu_any up-to-date: bazel-bin/xla/tests/dot_operation_test_autotune_disabled_amdgpu_any INFO: Elapsed time: 88.008s, Critical Path: 86.59s INFO: 16 processes: 9 internal, 7 local. INFO: Build completed successfully, 16 total actions Executed 1 out of 1 test: 1 test passes.disabled_amdgpu_any PASSED in 63.4s

//xla/tests:iota_test_amdgpu_any

[ RUN ] IotaR2TestInstantiation/IotaR2Test.DoIt/1171 I0000 00:00:1768218999.961985 3016214 se_gpu_pjrt_client.cc:1381] Using BFC allocator. I0000 00:00:1768218999.962074 3016214 gpu_helpers.cc:136] XLA backend allocating 16491332239 bytes on device 0 for BFCAllocator. I0000 00:00:1768218999.962097 3016214 gpu_helpers.cc:136] XLA backend allocating 16491332239 bytes on device 1 for BFCAllocator. I0000 00:00:1768218999.962110 3016214 gpu_helpers.cc:136] XLA backend allocating 16491332239 bytes on device 2 for BFCAllocator. I0000 00:00:1768218999.962119 3016214 gpu_helpers.cc:136] XLA backend allocating 16491332239 bytes on device 3 for BFCAllocator. I0000 00:00:1768218999.962128 3016214 gpu_helpers.cc:177] XLA backend will use up to 189650320752 bytes on device 0 for CollectiveBFCAllocator. I0000 00:00:1768218999.962136 3016214 gpu_helpers.cc:177] XLA backend will use up to 189650320752 bytes on device 1 for CollectiveBFCAllocator. I0000 00:00:1768218999.962144 3016214 gpu_helpers.cc:177] XLA backend will use up to 189650320752 bytes on device 2 for CollectiveBFCAllocator. I0000 00:00:1768218999.962151 3016214 gpu_helpers.cc:177] XLA backend will use up to 189650320752 bytes on device 3 for CollectiveBFCAllocator. -- Test timed out at 2026-01-12 11:56:40 UTC -- WARNING: Build options --action_env, --run_under, and --test_env have changed, discarding analysis cache (this can be expensive, see https://bazel.build/advanced/performance/iteration-speed). INFO: Analyzed target //xla/tests:iota_test_amdgpu_any (376 packages loaded, 53666 targets configured). INFO: Found 1 test target... Target //xla/tests:iota_test_amdgpu_any up-to-date: bazel-bin/xla/tests/iota_test_amdgpu_any INFO: Elapsed time: 6654.189s, Critical Path: 301.90s INFO: 8663 processes: 77 internal, 8586 local. INFO: Build completed successfully, 8663 total actions //xla/tests:iota_test_amdgpu_any PASSED in 126.7s Stats over 50 runs: max = 126.7s, min = 123.1s, avg = 124.5s, dev = 0.7s Executed 1 out of 1 test: 1 test passes. There were tests whose specified size is too big. Use the --test_verbose_timeout_warnings command line option to see which ones these are.

//xla/service/gpu/transforms:command_buffer_scheduling_test_amdgpu_any

[ RUN ] CommandBufferSchedulingTest.DynamicSliceFusionWithDynamicAddressesNotACommand 2026-01-12 12:34:57.785081: W ./xla/service/compiler.h:234] Ignoring the buffer assignment proto provided. xla/service/gpu/transforms/command_buffer_scheduling_test.cc:1477: Failure Value of: RunAndCompareTwoModulesReplicated(std::move(m_ref), std::move(m), true, true, std::nullopt) Actual: false (UNIMPLEMENTED: Empty nodes are not supported on ROCM.) Expected: true [ FAILED ] CommandBufferSchedulingTest.DynamicSliceFusionWithDynamicAddressesNotACommand (338 ms) [ RUN ] CommandBufferSchedulingTest.AllGatherStartFollowedByDone [ OK ] CommandBufferSchedulingTest.AllGatherStartFollowedByDone (3 ms) [ RUN ] CommandBufferSchedulingTest.MoveGTEs [ OK ] CommandBufferSchedulingTest.MoveGTEs (3 ms) [ RUN ] CommandBufferSchedulingTest.SingleCommandBuffer [ OK ] CommandBufferSchedulingTest.SingleCommandBuffer (1 ms) [----------] 29 tests from CommandBufferSchedulingTest (3838 ms total) [----------] Global test environment tear-down [==========] 30 tests from 2 test suites ran. (4605 ms total) [ PASSED ] 27 tests. [ SKIPPED ] 2 tests, listed below: [ SKIPPED ] CommandBufferSchedulingTest.Conditional [ SKIPPED ] CommandBufferSchedulingTest.While [ FAILED ] 1 test, listed below: [ FAILED ] CommandBufferSchedulingTest.DynamicSliceFusionWithDynamicAddressesNotACommand 1 FAILED TEST

//xla/tests:local_client_execute_test_amdgpu_any

[ RUN ] LocalClientExecuteTest.CompilePartitionedExecutable 2026-01-12 12:38:10.563499: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? xla/tests/local_client_execute_test.cc:767: Failure Expected equality of these values: 2 executables.size() Which is: 1 [ FAILED ] LocalClientExecuteTest.CompilePartitionedExecutable (34 ms) [ RUN ] LocalClientExecuteTest.AddArraysWithDifferentInputLayouts 2026-01-12 12:38:10.597713: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.AddArraysWithDifferentInputLayouts (57 ms) [ RUN ] LocalClientExecuteTest.Constant 2026-01-12 12:38:10.655456: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.Constant (8 ms) [ RUN ] LocalClientExecuteTest.SizeOfGeneratedCodeInBytes 2026-01-12 12:38:10.664086: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.SizeOfGeneratedCodeInBytes (34 ms) [ RUN ] LocalClientExecuteTest.InfeedOutfeedTest 2026-01-12 12:38:10.698750: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.InfeedOutfeedTest (29 ms) [ RUN ] LocalClientExecuteTest.ValidateMemoryFittingLevel 2026-01-12 12:38:10.728554: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.ValidateMemoryFittingLevel (30 ms) [ RUN ] LocalClientExecuteTest.ShapeBufferToLiteralConversion 2026-01-12 12:38:10.759389: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.ShapeBufferToLiteralConversion (2 ms) [ RUN ] LocalClientExecuteTest.AddScalars 2026-01-12 12:38:10.762242: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.AddScalars (29 ms) [ RUN ] LocalClientExecuteTest.ValidateOptimizationLevel 2026-01-12 12:38:10.792232: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.ValidateOptimizationLevel (23 ms) [ RUN ] LocalClientExecuteTest.TupleArguments 2026-01-12 12:38:10.815470: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.TupleArguments (31 ms) [ RUN ] LocalClientExecuteTest.LargeNestedTuple 2026-01-12 12:38:10.846715: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.LargeNestedTuple (5281 ms) [ RUN ] LocalClientExecuteTest.ValidateExecTimeOptimizationEffort 2026-01-12 12:38:16.128208: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.ValidateExecTimeOptimizationEffort (28 ms) [ RUN ] LocalClientExecuteTest.RunOnStreamForWrongPlatform 2026-01-12 12:38:16.156942: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.RunOnStreamForWrongPlatform (5 ms) [ RUN ] LocalClientExecuteTest.DeepTuple 2026-01-12 12:38:16.162605: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.DeepTuple (131 ms) [ RUN ] LocalClientExecuteTest.ValidateDeviceMemorySize 2026-01-12 12:38:16.294098: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.ValidateDeviceMemorySize (24 ms) [ RUN ] LocalClientExecuteTest.ValidateFDOProfile 2026-01-12 12:38:16.318581: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? 2026-01-12 12:38:16.324076: I xla/service/gpu/gpu_hlo_schedule.cc:342] Attempting to parse as a binary proto. 2026-01-12 12:38:16.324101: I xla/service/gpu/gpu_hlo_schedule.cc:347] Not a binary proto, attempt to parse it as a text proto. WARNING: All log messages before absl::InitializeLog() is called are written to STDERR E0000 00:00:1768221496.324192 3356912 text_format.cc:378] Error parsing text-format tensorflow.profiler.ProfiledInstructionsProto: 1:8: Message type "tensorflow.profiler.ProfiledInstructionsProto" has no field named "Testing". 2026-01-12 12:38:16.324219: E xla/service/gpu/gpu_hlo_schedule.cc:356] Unable to parse fdo_profile: not a valid text or binary ProfiledInstructionsProto [ OK ] LocalClientExecuteTest.ValidateFDOProfile (26 ms) [ RUN ] LocalClientExecuteTest.AddArraysWithDifferentOutputLayouts 2026-01-12 12:38:16.344869: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.AddArraysWithDifferentOutputLayouts (58 ms) [ RUN ] LocalClientExecuteTest.RunOnAllDeviceOrdinals 2026-01-12 12:38:16.403325: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? [ OK ] LocalClientExecuteTest.RunOnAllDeviceOrdinals (34 ms) [----------] 37 tests from LocalClientExecuteTest (11475 ms total) [----------] Global test environment tear-down [==========] 37 tests from 1 test suite ran. (11475 ms total) [ PASSED ] 36 tests. [ FAILED ] 1 test, listed below: [ FAILED ] LocalClientExecuteTest.CompilePartitionedExecutable 1 FAILED TEST

cherry-pick from 926af0e to 0.7.1

The following two need further investigation

//xla/service/gpu/transforms:command_buffer_scheduling_test_amdgpu_any \ //xla/tests:local_client_execute_test_amdgpu_any

//xla/service/gpu/transforms:command_buffer_scheduling_test_amdgpu_any

[ SKIPPED ] CommandBufferSchedulingTest.Conditional (0 ms) [ RUN ] CommandBufferSchedulingTest.CollectCommandBufferSequence [ OK ] CommandBufferSchedulingTest.CollectCommandBufferSequence (0 ms) [ RUN ] CommandBufferSchedulingTest.DynamicSliceFusionWithDynamicAddressesNotACommand 2026-01-12 17:55:42.615535: W ./xla/service/compiler.h:234] Ignoring the buffer assignment proto provided. xla/service/gpu/transforms/command_buffer_scheduling_test.cc:1477: Failure Value of: RunAndCompareTwoModulesReplicated(std::move(m_ref), std::move(m), true, true, std::nullopt) Actual: false (UNIMPLEMENTED: Empty nodes are not supported on ROCM.) Expected: true [ FAILED ] CommandBufferSchedulingTest.DynamicSliceFusionWithDynamicAddressesNotACommand (386 ms) [ RUN ] CommandBufferSchedulingTest.While xla/service/gpu/transforms/command_buffer_scheduling_test.cc:962: Skipped Not supported for ROCm! [ SKIPPED ] CommandBufferSchedulingTest.While (1 ms) [ RUN ] CommandBufferSchedulingTest.CollectivePermuteStartFollowedByAnotherStart [ OK ] CommandBufferSchedulingTest.CollectivePermuteStartFollowedByAnotherStart (3 ms) [ RUN ] CommandBufferSchedulingTest.ReduceScatterStartFollowedByDone [ OK ] CommandBufferSchedulingTest.ReduceScatterStartFollowedByDone (1 ms) [----------] 29 tests from CommandBufferSchedulingTest (5484 ms total) [----------] Global test environment tear-down [==========] 30 tests from 2 test suites ran. (7164 ms total) [ PASSED ] 27 tests. [ SKIPPED ] 2 tests, listed below: [ SKIPPED ] CommandBufferSchedulingTest.Conditional [ SKIPPED ] CommandBufferSchedulingTest.While [ FAILED ] 1 test, listed below: [ FAILED ] CommandBufferSchedulingTest.DynamicSliceFusionWithDynamicAddressesNotACommand 1 FAILED TEST

//xla/tests:local_client_execute_test_amdgpu_any

[ RUN ] LocalClientExecuteTest.CompilePartitionedExecutable 2026-01-12 17:59:44.042897: I xla/service/platform_util.cc:84] platform Host present but no XLA compiler available: could not find registered compiler for platform Host -- was support for that platform linked in? 2026-01-12 17:59:45.376245: I xla/service/service.cc:163] XLA service 0x55bfd2d516d0 initialized for platform ROCM (this does not guarantee that XLA will be used). Devices: 2026-01-12 17:59:45.376623: I xla/service/service.cc:171] StreamExecutor device (0): gfx942:sramecc+:xnack-, AMDGPU ISA version: gfx942:sramecc+:xnack- 2026-01-12 17:59:45.376629: I xla/service/service.cc:171] StreamExecutor device (1): gfx942:sramecc+:xnack-, AMDGPU ISA version: gfx942:sramecc+:xnack- 2026-01-12 17:59:45.376634: I xla/service/service.cc:171] StreamExecutor device (2): gfx942:sramecc+:xnack-, AMDGPU ISA version: gfx942:sramecc+:xnack- 2026-01-12 17:59:45.376638: I xla/service/service.cc:171] StreamExecutor device (3): gfx942:sramecc+:xnack-, AMDGPU ISA version: gfx942:sramecc+:xnack- 2026-01-12 17:59:45.376642: I xla/service/service.cc:171] StreamExecutor device (4): gfx942:sramecc+:xnack-, AMDGPU ISA version: gfx942:sramecc+:xnack- 2026-01-12 17:59:45.376645: I xla/service/service.cc:171] StreamExecutor device (5): gfx942:sramecc+:xnack-, AMDGPU ISA version: gfx942:sramecc+:xnack- 2026-01-12 17:59:45.376649: I xla/service/service.cc:171] StreamExecutor device (6): gfx942:sramecc+:xnack-, AMDGPU ISA version: gfx942:sramecc+:xnack- 2026-01-12 17:59:45.376653: I xla/service/service.cc:171] StreamExecutor device (7): gfx942:sramecc+:xnack-, AMDGPU ISA version: gfx942:sramecc+:xnack- xla/tests/local_client_execute_test.cc:767: Failure Expected equality of these values: 2 executables.size() Which is: 1 [ FAILED ] LocalClientExecuteTest.CompilePartitionedExecutable (1427 ms) [----------] 1 test from LocalClientExecuteTest (1427 ms total) [----------] Global test environment tear-down [==========] 1 test from 1 test suite ran. (1427 ms total) [ PASSED ] 0 tests. [ FAILED ] 1 test, listed below: [ FAILED ] LocalClientExecuteTest.CompilePartitionedExecutable 1 FAILED TEST

i-chaochen

If these are tests are CI's quirks and irrelevent to your PR, then you can merge it to unblock newer 0.7.1 release.

update rocprofiler-sdk with stuff in 0.8.0

aa7c126

cj401-amd requested review from JehandadKhan, alekstheod, i-chaochen, mminutoli and pemeliya January 9, 2026 15:21

cj401-amd added 4 commits January 9, 2026 10:22

update rocprofiler-sdk with stuff in 0.8.0

4597dea

update rocprofiler-sdk with stuff in 0.8.0

d4a576f

update rocprofiler-sdk with stuff in 0.8.0

11d49b5

update rocprofiler-sdk with stuff in 0.8.0

bef28b6

i-chaochen reviewed Jan 10, 2026

View reviewed changes

i-chaochen approved these changes Jan 12, 2026

View reviewed changes

i-chaochen merged commit 00c7948 into rocm-jaxlib-v0.7.1 Jan 12, 2026
6 of 9 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

update rocprofiler-sdk with stuff in 0.8.0#513

update rocprofiler-sdk with stuff in 0.8.0#513
i-chaochen merged 5 commits into
rocm-jaxlib-v0.7.1from
ci_cj_rocprofv3_v1_rocm-jaxlib-v0.7.1

cj401-amd commented Jan 9, 2026 •

edited

Loading

Uh oh!

i-chaochen Jan 10, 2026

Uh oh!

cj401-amd Jan 10, 2026

Uh oh!

i-chaochen Jan 11, 2026 •

edited

Loading

Uh oh!

cj401-amd Jan 12, 2026 •

edited

Loading

Uh oh!

cj401-amd Jan 12, 2026

Uh oh!

i-chaochen left a comment •

edited

Loading

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

cj401-amd commented Jan 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Motivation

Uh oh!

i-chaochen Jan 10, 2026

Choose a reason for hiding this comment

Uh oh!

cj401-amd Jan 10, 2026

Choose a reason for hiding this comment

Uh oh!

i-chaochen Jan 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

cj401-amd Jan 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

cj401-amd Jan 12, 2026

Choose a reason for hiding this comment

Uh oh!

i-chaochen left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

cj401-amd commented Jan 9, 2026 •

edited

Loading

i-chaochen Jan 11, 2026 •

edited

Loading

cj401-amd Jan 12, 2026 •

edited

Loading

i-chaochen left a comment •

edited

Loading