Commit Graph

64521 Commits

Author SHA1 Message Date
sdashmiz ee6bbb0004 FEAT-60269 - Test for large LDS
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I8f476f1b44324fffcc25423209a632caeb48cb99
2024-07-08 12:07:14 -04:00
Giovanni LB 7e199e6b7a SWDEV-470113: Fixing hang on navi33
Change-Id: I5c1d30547039a6f23ab5974f43ea63f971573108
2024-07-05 20:39:56 -04:00
Giovanni LB e4e1e4857e SWDEV-470113: Fixing hang on navi33
Change-Id: I5c1d30547039a6f23ab5974f43ea63f971573108


[ROCm/rocprofiler commit: 7e199e6b7a]
2024-07-05 20:39:56 -04:00
jujiang 386c8a80b0 SWDEV-455106 - Make a correction in clr readme
Change-Id: I218e2b3ad22424b6a1c564498bbaa18c14e8a751


[ROCm/clr commit: e3114e8f38]
2024-07-04 13:32:29 -04:00
jujiang e3114e8f38 SWDEV-455106 - Make a correction in clr readme
Change-Id: I218e2b3ad22424b6a1c564498bbaa18c14e8a751
2024-07-04 13:32:29 -04:00
Jaydeep Patel 130b72302a SWDEV-457316 - Return invalid val from add mem free node if corresponding mem alloc node is missing.
Change-Id: Ib0c346a439fc38ebfd106bcbdf75bd10bfd2f090


[ROCm/clr commit: cf0320a0b9]
2024-07-04 13:09:46 -04:00
Jaydeep Patel cf0320a0b9 SWDEV-457316 - Return invalid val from add mem free node if corresponding mem alloc node is missing.
Change-Id: Ib0c346a439fc38ebfd106bcbdf75bd10bfd2f090
2024-07-04 13:09:46 -04:00
jujiang 1975687a7e SWDEV-455106 - Adding new API support in 6.2
Change-Id: I0170fc1ab2b0a988aee78489bc140434464f0a21


[ROCm/clr commit: 149f393991]
2024-07-04 12:01:11 -04:00
jujiang 149f393991 SWDEV-455106 - Adding new API support in 6.2
Change-Id: I0170fc1ab2b0a988aee78489bc140434464f0a21
2024-07-04 12:01:11 -04:00
Tao Sang b8cf863eaa SWDEV-458943 - Implement std::mutex based monitor
Implement std::mutex based monitor that has much
simpler logics than legacy monitor.
Create DEBUG_CLR_USE_STDMUTEX_IN_AMD_MONITOR to
toggle them.
If DEBUG_CLR_USE_STDMUTEX_IN_AMD_MONITOR = false
  (by default), use legacy monitor;
If DEBUG_CLR_USE_STDMUTEX_IN_AMD_MONITOR = true,
  use std::mutex based monitor.
If no perf drop of stl::mutex based monitor,
legacy one will be removed later.

Change-Id: I1d21368ff462477d3238d71e4e2a1a7d6b9167ad


[ROCm/clr commit: 73c02041e1]
2024-07-04 11:50:46 -04:00
Tao Sang 73c02041e1 SWDEV-458943 - Implement std::mutex based monitor
Implement std::mutex based monitor that has much
simpler logics than legacy monitor.
Create DEBUG_CLR_USE_STDMUTEX_IN_AMD_MONITOR to
toggle them.
If DEBUG_CLR_USE_STDMUTEX_IN_AMD_MONITOR = false
  (by default), use legacy monitor;
If DEBUG_CLR_USE_STDMUTEX_IN_AMD_MONITOR = true,
  use std::mutex based monitor.
If no perf drop of stl::mutex based monitor,
legacy one will be removed later.

Change-Id: I1d21368ff462477d3238d71e4e2a1a7d6b9167ad
2024-07-04 11:50:46 -04:00
Giovanni Lenzi Baraldi 19329c3a97 General fixes to ATT, packets and event ID retrieval (#960)
* General fixes to ATT, packets and event ID retrieval

* Update source/lib/rocprofiler-sdk/hsa/aql_packet.hpp

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

---------

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

[ROCm/rocprofiler-sdk commit: 4e2144dbfa]
2024-07-04 03:58:45 -03:00
Giovanni Lenzi Baraldi 4e2144dbfa General fixes to ATT, packets and event ID retrieval (#960)
* General fixes to ATT, packets and event ID retrieval

* Update source/lib/rocprofiler-sdk/hsa/aql_packet.hpp

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

---------

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
2024-07-04 03:58:45 -03:00
Ioannis Assiouras a774b89a43 SWDEV-470787 - Fixed undefined symbols for flags in static build
Change-Id: I7812c8924396d0df9ab331f9a1844aabbf5a9211


[ROCm/clr commit: fa07c33cba]
2024-07-04 02:57:22 -04:00
Ioannis Assiouras fa07c33cba SWDEV-470787 - Fixed undefined symbols for flags in static build
Change-Id: I7812c8924396d0df9ab331f9a1844aabbf5a9211
2024-07-04 02:57:22 -04:00
Zhang Ava fed31f08b4 Merge amd-staging into amd-master 20240628
Signed-off-by: Zhang Ava <niandong.zhang@amd.com>
Change-Id: I9493cdf35b64cfa0a99de017e2d6b521af71cf14


[ROCm/rocm_smi_lib commit: 4c0ce45912]
2024-07-04 14:19:02 +08:00
Zhang Ava 4c0ce45912 Merge amd-staging into amd-master 20240628
Signed-off-by: Zhang Ava <niandong.zhang@amd.com>
Change-Id: I9493cdf35b64cfa0a99de017e2d6b521af71cf14
2024-07-04 14:19:02 +08:00
Vladimir Indic 1c98e94488 PC sampling: integration test with instruction decoding (#929)
* PC sampling: integration test with instruction decoding

* PC sampling: verifying internal and external CIDs

The PC sampling integration test has been extended
to verify internal and external correlation IDs.

* tmp solution of using Instructions as keys

* wrapper for HIP call

* PCS integration test: ld_addr as instruction id

For the sake of the integration test, use  as the
instruction identifier. To support code object unloading
and relocations, use  as the identifier
(the change in the decoder is required).

* PCS integration test: removing shared_ptr

Completely removing usage of shared pointers.

* PCS integration test: removing decoder

When a code object has been unloaded, ensure all PC samples
corresponding to that object are decoded, prior to removing
the decoder.

* PCS integration test: fixing build flags and imports

* PCS integration test: fixing labels

* PCS integration test: cmake flags fix

* PC sampling cmake labels renamed

* PCS integration test refactoring

* PCS integration test: minimize usage of raw pointers

* PCS integration test: at least one sample should be delivered.

* PC sampling lables: pc-sampling

[ROCm/rocprofiler-sdk commit: 498b1f2bd7]
2024-07-04 11:19:27 +05:30
Vladimir Indic 498b1f2bd7 PC sampling: integration test with instruction decoding (#929)
* PC sampling: integration test with instruction decoding

* PC sampling: verifying internal and external CIDs

The PC sampling integration test has been extended
to verify internal and external correlation IDs.

* tmp solution of using Instructions as keys

* wrapper for HIP call

* PCS integration test: ld_addr as instruction id

For the sake of the integration test, use  as the
instruction identifier. To support code object unloading
and relocations, use  as the identifier
(the change in the decoder is required).

* PCS integration test: removing shared_ptr

Completely removing usage of shared pointers.

* PCS integration test: removing decoder

When a code object has been unloaded, ensure all PC samples
corresponding to that object are decoded, prior to removing
the decoder.

* PCS integration test: fixing build flags and imports

* PCS integration test: fixing labels

* PCS integration test: cmake flags fix

* PC sampling cmake labels renamed

* PCS integration test refactoring

* PCS integration test: minimize usage of raw pointers

* PCS integration test: at least one sample should be delivered.

* PC sampling lables: pc-sampling
2024-07-04 11:19:27 +05:30
Jaydeep Patel 44e6cc3e7a SWDEV-462616 - Sample 19 is disabled temporarily due to ASAN failure.
Change-Id: I6528095909da1f91dbe7383752deca780a38a231


[ROCm/hip-tests commit: ba83484a8d]
2024-07-04 05:41:15 +00:00
Jaydeep Patel ba83484a8d SWDEV-462616 - Sample 19 is disabled temporarily due to ASAN failure.
Change-Id: I6528095909da1f91dbe7383752deca780a38a231
2024-07-04 05:41:15 +00:00
Saleel Kudchadker ff12be080d SWDEV-465602 - Fix random segfault
- Introduce a lock when checking isUserObjectValid. We need a lock
here as one can remove the userObject T2, leading to buffer overflow
when checking ranges in T1.

Change-Id: I058144b8cc463c90ab6bf5cf96bf937897742917


[ROCm/clr commit: 6ac67afdd5]
2024-07-03 18:57:23 -04:00
Saleel Kudchadker 6ac67afdd5 SWDEV-465602 - Fix random segfault
- Introduce a lock when checking isUserObjectValid. We need a lock
here as one can remove the userObject T2, leading to buffer overflow
when checking ranges in T1.

Change-Id: I058144b8cc463c90ab6bf5cf96bf937897742917
2024-07-03 18:57:23 -04:00
Chris Freehill 9830e32e7a Fix hsakmt-roct-devel REQUIRES/PROVIDES CPack handling
Change-Id: If8bac85f2f7a23bce1b967fcec25216603b4c7bd
2024-07-03 16:46:00 -05:00
Chris Freehill 4bcdb4a2db Fix hsakmt-roct-devel REQUIRES/PROVIDES CPack handling
Change-Id: If8bac85f2f7a23bce1b967fcec25216603b4c7bd


[ROCm/ROCR-Runtime commit: 9830e32e7a]
2024-07-03 16:46:00 -05:00
Maneesh Gupta 539dc7cee2 SWDEV-1 - Update CHANGELOG
Change-Id: I715ea6e30fdf7d561654acf9fc47fc4108d1ce0f


[ROCm/clr commit: 933b083360]
2024-07-03 14:47:24 -04:00
Maneesh Gupta 933b083360 SWDEV-1 - Update CHANGELOG
Change-Id: I715ea6e30fdf7d561654acf9fc47fc4108d1ce0f
2024-07-03 14:47:24 -04:00
Giovanni LB 70fedc5195 SWDEV-470113: Swapping test order and resetting queue.cpp variables
Change-Id: I20f3fa3cbf389ab89ba8ea8e430819c4ef837ad0
2024-07-03 15:41:37 -03:00
Giovanni LB d29580c94e SWDEV-470113: Swapping test order and resetting queue.cpp variables
Change-Id: I20f3fa3cbf389ab89ba8ea8e430819c4ef837ad0


[ROCm/rocprofiler commit: 70fedc5195]
2024-07-03 15:41:37 -03:00
jujiang a589005d7b SWDEV-430321 - Correct the name for HIP library
Change-Id: I9a4d00be528a786275cee47b429e22d45a9bb3a4


[ROCm/hip commit: 818dca2c0b]
2024-07-03 14:26:32 -04:00
jujiang 818dca2c0b SWDEV-430321 - Correct the name for HIP library
Change-Id: I9a4d00be528a786275cee47b429e22d45a9bb3a4
2024-07-03 14:26:32 -04:00
Julia Jiang 12a307219a SWDEV-470017 - Changed the return value for the API hipMemGetAdressRange
Change-Id: I6ccc627d3d233610f4509e0323cfef5bf92422a6


[ROCm/hip commit: 9e407373f4]
2024-07-03 14:25:58 -04:00
Julia Jiang 9e407373f4 SWDEV-470017 - Changed the return value for the API hipMemGetAdressRange
Change-Id: I6ccc627d3d233610f4509e0323cfef5bf92422a6
2024-07-03 14:25:58 -04:00
sdashmiz 0eaf547f0b SWDEV-458069 - Fix return of local variable
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I8964f988383f61e35363490b669a416341150979


[ROCm/hipother commit: 1287d7e4a6]
2024-07-03 09:50:42 -04:00
sdashmiz 1287d7e4a6 SWDEV-458069 - Fix return of local variable
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I8964f988383f61e35363490b669a416341150979
2024-07-03 09:50:42 -04:00
Jonathan R. Madsen e0fb6a2335 Executables append omnitrace library directory to LD_LIBRARY_PATH (#356)
- omnitrace-run, omnitrace-sample, and omnitrace-causal now automatically append the LD_LIBRARY_PATH with the directory containing the omnitrace libraries
  - this helps ensure that binary rewritten exes can resolve omnitrace-rt library location

[ROCm/rocprofiler-systems commit: 18833a0a5e]
2024-07-02 18:55:37 -05:00
Jonathan R. Madsen 18833a0a5e Executables append omnitrace library directory to LD_LIBRARY_PATH (#356)
- omnitrace-run, omnitrace-sample, and omnitrace-causal now automatically append the LD_LIBRARY_PATH with the directory containing the omnitrace libraries
  - this helps ensure that binary rewritten exes can resolve omnitrace-rt library location
2024-07-02 18:55:37 -05:00
Jonathan R. Madsen cd6a94dcb5 Fix kernel trace gaps (#961)
- source/lib/rocprofiler-sdk/hsa/queue.cpp
  - Optimize WriteInterceptor to eliminate extra barrier packets causing gaps between kernels in kernel tracing
  - increase timeout_hint in hsa_signal_wait in set_profiler_active_on_queue
  - misc logging improvements
- source/lib/rocprofiler-sdk/counters/agent_profiling.cpp
  - increase timeout_hint in hsa_signal_wait in set_profiler_active_on_queue
- tests/rocprofv3/hsa-queue-dependency/CMakeLists.txt
  - add TIMEOUT for rocprofv3-test-hsa-multiqueue-execute

[ROCm/rocprofiler-sdk commit: 64b8f8370e]
2024-07-02 18:49:04 -05:00
Jonathan R. Madsen 64b8f8370e Fix kernel trace gaps (#961)
- source/lib/rocprofiler-sdk/hsa/queue.cpp
  - Optimize WriteInterceptor to eliminate extra barrier packets causing gaps between kernels in kernel tracing
  - increase timeout_hint in hsa_signal_wait in set_profiler_active_on_queue
  - misc logging improvements
- source/lib/rocprofiler-sdk/counters/agent_profiling.cpp
  - increase timeout_hint in hsa_signal_wait in set_profiler_active_on_queue
- tests/rocprofv3/hsa-queue-dependency/CMakeLists.txt
  - add TIMEOUT for rocprofv3-test-hsa-multiqueue-execute
2024-07-02 18:49:04 -05:00
Saleel Kudchadker bbef85714e SWDEV-470008 - Fix AMD_SERIALIZE_KERNEL
- awaitCompletion code may do a endless spin wait for cases where we
dont submit a handler. One such case can be the hipExt*Launch API which
takes a stop event. In that case we optimize the stop event by attaching
a signal to the dispatch packet but dont submit a handler when we attach
the signal. That means if awaitCompletion() is called after that, we
would keep on waiting on command status on the host rather than simply
checking signal value.

Change-Id: Ie8bf175aeefa3f9e4299b1ae7ae9108dad67e283


[ROCm/clr commit: 561fb8a459]
2024-07-02 19:05:05 -04:00
Saleel Kudchadker 561fb8a459 SWDEV-470008 - Fix AMD_SERIALIZE_KERNEL
- awaitCompletion code may do a endless spin wait for cases where we
dont submit a handler. One such case can be the hipExt*Launch API which
takes a stop event. In that case we optimize the stop event by attaching
a signal to the dispatch packet but dont submit a handler when we attach
the signal. That means if awaitCompletion() is called after that, we
would keep on waiting on command status on the host rather than simply
checking signal value.

Change-Id: Ie8bf175aeefa3f9e4299b1ae7ae9108dad67e283
2024-07-02 19:05:05 -04:00
Jatin Chaudhary 58d27f6f47 SWDEV-460834 - add unsafe atomic add for fp16 and bf16
Change-Id: I6de5c2c425c9f8ac7f6c4e5c83c8b8b7ac8fe4cb


[ROCm/clr commit: ecd812b2d8]
2024-07-02 16:45:17 -04:00
Jatin Chaudhary ecd812b2d8 SWDEV-460834 - add unsafe atomic add for fp16 and bf16
Change-Id: I6de5c2c425c9f8ac7f6c4e5c83c8b8b7ac8fe4cb
2024-07-02 16:45:17 -04:00
Sourabh Betigeri 524c4663a6 SWDEV-470703 - Avoids a potential segfault caused by nullptr dereferencing
Change-Id: If80b00b41869076c18651995c46f89095e7266f9


[ROCm/clr commit: 9d628a4a3d]
2024-07-02 12:22:29 -04:00
Sourabh Betigeri 9d628a4a3d SWDEV-470703 - Avoids a potential segfault caused by nullptr dereferencing
Change-Id: If80b00b41869076c18651995c46f89095e7266f9
2024-07-02 12:22:29 -04:00
Julia Jiang 3c63ca66e4 SWDEV-337797 - correct doxygen for the API hipIpcOpenEventHandle
Change-Id: I45158be773d9a3baf5128590e3df7218d4fbbd7f


[ROCm/hip commit: fe808eb26f]
2024-07-02 11:03:39 -04:00
Julia Jiang fe808eb26f SWDEV-337797 - correct doxygen for the API hipIpcOpenEventHandle
Change-Id: I45158be773d9a3baf5128590e3df7218d4fbbd7f
2024-07-02 11:03:39 -04:00
Vladana Stojiljkovic 5d51ecfd8b SWDEV-456699 - Fix Unit_hipDrvGraphAddMemcpyNode_Negative_Parameters test case
Change-Id: Ia27362115bbf0e96a7eac5a5ae21c2d504419bc8


[ROCm/hip-tests commit: a3bc534a5f]
2024-07-02 05:03:44 -04:00
Vladana Stojiljkovic a3bc534a5f SWDEV-456699 - Fix Unit_hipDrvGraphAddMemcpyNode_Negative_Parameters test case
Change-Id: Ia27362115bbf0e96a7eac5a5ae21c2d504419bc8
2024-07-02 05:03:44 -04:00
Ranjith Ramakrishnan c08a36f06a Revert "SWDEV-469045 - Replace the rocm-llvm package dependency with rocm-llvm-static for static builds"
This reverts commit 7d021f8196.

Reason for revert: <rocm-llvm package name change not required for static builds>

Change-Id: Ib2214a74162e5b015b096dc286151ecbd3ca0a80


[ROCm/clr commit: 0670697ab4]
2024-07-02 04:12:38 -04:00