Grafico dei commit

64512 Commit

Autore SHA1 Messaggio Data
Lancelot SIX 6f828d8609 Revert "trap_handler: Set status.skip_export when halting a wave"
This reverts commit c5db063b2f.  This
change is required for the runtime to generate reliable core dump files,
but this feature has been disabled for now by
5e3be9c28a.  Until it is needed, revert
the ABI change in the trap handler to maintain compatibility with older
debugger.

Change-Id: I77a1562dc7962befe2bf88442df858e2d2b1c5ab
2024-01-16 15:55:59 +00:00
Lancelot SIX 9317d0fbc0 Revert "trap_handler: Set status.skip_export when halting a wave"
This reverts commit 4c8a849772.  This
change is required for the runtime to generate reliable core dump files,
but this feature has been disabled for now by
816b46868a.  Until it is needed, revert
the ABI change in the trap handler to maintain compatibility with older
debugger.

Change-Id: I77a1562dc7962befe2bf88442df858e2d2b1c5ab


[ROCm/ROCR-Runtime commit: 6f828d8609]
2024-01-16 15:55:59 +00:00
kjayapra-amd b457d0ba82 SWDEV-413997 - Physical address size could be more than mapped size, only virtual address needs to be match the size of unmap request.
Change-Id: If0e973149d8ce1b6f01000ab951cf6e9f4b38e97


[ROCm/clr commit: 0f9c624f4c]
2024-01-16 07:09:50 -05:00
kjayapra-amd 0f9c624f4c SWDEV-413997 - Physical address size could be more than mapped size, only virtual address needs to be match the size of unmap request.
Change-Id: If0e973149d8ce1b6f01000ab951cf6e9f4b38e97
2024-01-16 07:09:50 -05:00
Maneesh Gupta ad499128d6 SWDEV-1 - Temporarily disable rocprofiler-register support
Change-Id: I2e4b4c5f6c0640fc4e1378dd920b080b88583921


[ROCm/clr commit: 4b6912031b]
2024-01-16 04:15:05 +00:00
Maneesh Gupta 4b6912031b SWDEV-1 - Temporarily disable rocprofiler-register support
Change-Id: I2e4b4c5f6c0640fc4e1378dd920b080b88583921
2024-01-16 04:15:05 +00:00
Sam Wu 5efc1132fc SWDEV-422771 - Update rocm-docs-core documentation requirement
ROCDOC-203

For new doc theme

Change-Id: I66061fa696ff769b127e4b4a09012e2f3127c7eb


[ROCm/hip commit: 15d8076f40]
2024-01-15 17:23:32 -05:00
Sam Wu 15d8076f40 SWDEV-422771 - Update rocm-docs-core documentation requirement
ROCDOC-203

For new doc theme

Change-Id: I66061fa696ff769b127e4b4a09012e2f3127c7eb
2024-01-15 17:23:32 -05:00
Saleel Kudchadker 34fd1b7fe5 SWDEV-301667 - Remove resetFenceDirty
Dont track the status of fence_dirty_ flag on the host, instead clear it
when we submit a barrier on the respective stream.

Change-Id: I4d98dbf20c81379c9c5da9f5b67629a8f9f6dfcd


[ROCm/clr commit: 0b0df605d4]
2024-01-15 15:43:14 -05:00
Saleel Kudchadker 0b0df605d4 SWDEV-301667 - Remove resetFenceDirty
Dont track the status of fence_dirty_ flag on the host, instead clear it
when we submit a barrier on the respective stream.

Change-Id: I4d98dbf20c81379c9c5da9f5b67629a8f9f6dfcd
2024-01-15 15:43:14 -05:00
Jatin Chaudhary 06292cfc6a SWDEV-431399 - fix issues highlighted by address sanitizer
Change-Id: I927009a8d408bdf13677ce5cc926908fb4552441


[ROCm/hip-tests commit: a87786f060]
2024-01-15 04:01:36 -05:00
Jatin Chaudhary a87786f060 SWDEV-431399 - fix issues highlighted by address sanitizer
Change-Id: I927009a8d408bdf13677ce5cc926908fb4552441
2024-01-15 04:01:36 -05:00
Jonathan R. Madsen 3c47bc534a Update tests/apps/* to support multigpu in single process (#374)
- CPU threads will spread out across multiple devices

[ROCm/rocprofiler-sdk commit: a24b08c2db]
2024-01-14 21:54:13 -06:00
Jonathan R. Madsen a24b08c2db Update tests/apps/* to support multigpu in single process (#374)
- CPU threads will spread out across multiple devices
2024-01-14 21:54:13 -06:00
Jonathan R. Madsen c4a749d99e Remove directly linking to HIP and HSA runtimes (#373)
- librocprofiler-sdk.so no longer directly links to HIP runtime library (libamdhip64.so)
- librocprofiler-sdk.so no longer directly links to HSA runtime library (libhsa-runtime64.so)
- librocprofiler-sdk.so does directly link to AQL profile library (libhsa-amd-aqlprofile64.so) so HSA runtime library is still in DT_NEEDED
  - This needs to be fixed eventually so rocprofiler-sdk library can be independent of HSA and HIP runtimes

[ROCm/rocprofiler-sdk commit: 57113e98c9]
2024-01-14 19:05:10 -06:00
Jonathan R. Madsen 57113e98c9 Remove directly linking to HIP and HSA runtimes (#373)
- librocprofiler-sdk.so no longer directly links to HIP runtime library (libamdhip64.so)
- librocprofiler-sdk.so no longer directly links to HSA runtime library (libhsa-runtime64.so)
- librocprofiler-sdk.so does directly link to AQL profile library (libhsa-amd-aqlprofile64.so) so HSA runtime library is still in DT_NEEDED
  - This needs to be fixed eventually so rocprofiler-sdk library can be independent of HSA and HIP runtimes
2024-01-14 19:05:10 -06:00
Jonathan R. Madsen b3afc6c1ea Support documentation build + install + packaging (#369)
* Support documentation build + install + packaging

* Fix NUM_ROCPROFILER_PACKAGING_COMPONENTS check

[ROCm/rocprofiler-sdk commit: 84499b17c6]
2024-01-14 08:10:14 -06:00
Jonathan R. Madsen 84499b17c6 Support documentation build + install + packaging (#369)
* Support documentation build + install + packaging

* Fix NUM_ROCPROFILER_PACKAGING_COMPONENTS check
2024-01-14 08:10:14 -06:00
Benjamin Welton 60f2d204d2 Tests for agent and aql packet generation (#365)
* Tests for agent and aql packet generation

Test for agent and fixing test problems with aql packet that caused test
to not run.

* cmake formatting (cmake-format) (#366)

Co-authored-by: bwelton <bwelton@users.noreply.github.com>

* source formatting (clang-format v11) (#367)

Co-authored-by: bwelton <bwelton@users.noreply.github.com>

* Minor tweak

* source formatting (clang-format v11) (#368)

Co-authored-by: bwelton <bwelton@users.noreply.github.com>

* Add gfx900 to basic_counters

* Update samples/counter_collection/client.cpp

- fix data race by flushing buffer during tool_fini

* Fix data race for output stream destruction

---------

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: bwelton <bwelton@users.noreply.github.com>
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>

[ROCm/rocprofiler-sdk commit: 802e79b113]
2024-01-14 04:46:27 -06:00
Benjamin Welton 802e79b113 Tests for agent and aql packet generation (#365)
* Tests for agent and aql packet generation

Test for agent and fixing test problems with aql packet that caused test
to not run.

* cmake formatting (cmake-format) (#366)

Co-authored-by: bwelton <bwelton@users.noreply.github.com>

* source formatting (clang-format v11) (#367)

Co-authored-by: bwelton <bwelton@users.noreply.github.com>

* Minor tweak

* source formatting (clang-format v11) (#368)

Co-authored-by: bwelton <bwelton@users.noreply.github.com>

* Add gfx900 to basic_counters

* Update samples/counter_collection/client.cpp

- fix data race by flushing buffer during tool_fini

* Fix data race for output stream destruction

---------

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: bwelton <bwelton@users.noreply.github.com>
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>
2024-01-14 04:46:27 -06:00
coleramos425 f3e1ff4ed5 Catch failed rocprof subprocess
Signed-off-by: coleramos425 <colramos@amd.com>


[ROCm/rocprofiler-compute commit: e534e0ab3b]
2024-01-13 11:31:55 -06:00
coleramos425 e534e0ab3b Catch failed rocprof subprocess
Signed-off-by: coleramos425 <colramos@amd.com>
2024-01-13 11:31:55 -06:00
Jonathan R. Madsen 3c5da3e666 Add rocprofiler-register support
- Add API table versioning defines to hip_api_trace.hpp
- Add rocprofiler-register implementation in hip_api_trace.cpp
  - created ToolInit implementation
  - moved some local functions into anonymous namespace
  - renamed some function to be overloads
  - made most of the initialization code templated (reduced code duplication)
  - enforce ABI
- Updated hipamd/src/CMakeLists.txt
  - find rocprofiler-register package (enabled by default but not required)
- Address review comments
  - simplify size calculation for dispatch table
  - remove setting .size field in GetDispatchTableImpl
- Fix windows build
- Do not enforce ABI on Windows

Change-Id: I08766e8a083528a1236996274bf4522e0e8a1b9f


[ROCm/clr commit: 0f1fdc7794]
2024-01-13 01:02:59 -06:00
Jonathan R. Madsen 0f1fdc7794 Add rocprofiler-register support
- Add API table versioning defines to hip_api_trace.hpp
- Add rocprofiler-register implementation in hip_api_trace.cpp
  - created ToolInit implementation
  - moved some local functions into anonymous namespace
  - renamed some function to be overloads
  - made most of the initialization code templated (reduced code duplication)
  - enforce ABI
- Updated hipamd/src/CMakeLists.txt
  - find rocprofiler-register package (enabled by default but not required)
- Address review comments
  - simplify size calculation for dispatch table
  - remove setting .size field in GetDispatchTableImpl
- Fix windows build
- Do not enforce ABI on Windows

Change-Id: I08766e8a083528a1236996274bf4522e0e8a1b9f
2024-01-13 01:02:59 -06:00
Galantsev, Dmitrii 673087e6a0 Remove word 'error' from non-error message
This simplifies grep lookup

Change-Id: I46cd13e0ab414791655fd93e8dcf270a946a6687
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>


[ROCm/rocm_smi_lib commit: 147af192b5]
2024-01-12 15:18:55 -06:00
Galantsev, Dmitrii 147af192b5 Remove word 'error' from non-error message
This simplifies grep lookup

Change-Id: I46cd13e0ab414791655fd93e8dcf270a946a6687
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>
2024-01-12 15:18:55 -06:00
German 339523c475 SWDEV-440746 - Limit WG for compute P2P
Use only 16 workgroups for compute P2P copies.
That should be enough to utilize XGMI bandwidth.

Change-Id: I60dfe019279bb95f93c8874244c1738aad1896d8


[ROCm/clr commit: 31101c6219]
2024-01-12 14:56:29 -05:00
German 31101c6219 SWDEV-440746 - Limit WG for compute P2P
Use only 16 workgroups for compute P2P copies.
That should be enough to utilize XGMI bandwidth.

Change-Id: I60dfe019279bb95f93c8874244c1738aad1896d8
2024-01-12 14:56:29 -05:00
sdashmiz 29eba6df09 SWDEV-421027 - Add hipGraphAddNode
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: Ic8cf293ff483ee2547b52d2975062bcb9a6f5d17


[ROCm/clr commit: d23835ffbe]
2024-01-12 11:36:30 -05:00
sdashmiz d23835ffbe SWDEV-421027 - Add hipGraphAddNode
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: Ic8cf293ff483ee2547b52d2975062bcb9a6f5d17
2024-01-12 11:36:30 -05:00
Jatin Chaudhary a65d07302f SWDEV-438181 - check stream associated with event
...before we dereference it. It might have been deleted.

Change-Id: Ief832ee0907658a40ca42b9d78d19658153a05dd


[ROCm/clr commit: 2cfb8bc3c1]
2024-01-12 10:39:20 -05:00
Jatin Chaudhary 2cfb8bc3c1 SWDEV-438181 - check stream associated with event
...before we dereference it. It might have been deleted.

Change-Id: Ief832ee0907658a40ca42b9d78d19658153a05dd
2024-01-12 10:39:20 -05:00
Julia Jiang c23d97cdd8 SWDEV-410522 - Update doxygen config file to handle default parameter
Change-Id: I92afc697ecd772743d2200b621d19e9911d3d9f8


[ROCm/hip commit: 355b16b473]
2024-01-12 10:08:57 -05:00
Julia Jiang 355b16b473 SWDEV-410522 - Update doxygen config file to handle default parameter
Change-Id: I92afc697ecd772743d2200b621d19e9911d3d9f8
2024-01-12 10:08:57 -05:00
Ioannis Assiouras 5d5d8e52e1 SWDEV-430116 - Re-enabled fixed tests on Windows
Unit_hipFreeAsync_negative
Unit_hipLaunchHostFunc_multistreams
Unit_hipGraphMem_Alloc_Free_NodeGetParams_Negative

Change-Id: Ic7b00b688ded737d1762b2c8795c0772107dbb64


[ROCm/hip-tests commit: 7939517b03]
2024-01-12 15:25:42 +05:30
Ioannis Assiouras 7939517b03 SWDEV-430116 - Re-enabled fixed tests on Windows
Unit_hipFreeAsync_negative
Unit_hipLaunchHostFunc_multistreams
Unit_hipGraphMem_Alloc_Free_NodeGetParams_Negative

Change-Id: Ic7b00b688ded737d1762b2c8795c0772107dbb64
2024-01-12 15:25:42 +05:30
Benjamin Welton cf296be224 Fix init order for getMetricIdMap (#341)
* Fix init order for getMetricIdMap

Ensure that getMetricIdMap() stays valid until the destruction
of the counting service. Adds a test to ensure that this behavior
is maintained (this test fails without this change).

* source formatting (clang-format v11) (#343)

Co-authored-by: bwelton <bwelton@users.noreply.github.com>

* cmake formatting (cmake-format) (#342)

Co-authored-by: bwelton <bwelton@users.noreply.github.com>

* Remove threading

* Update usage of rocprofiler::counters::get{Metric,MetricId}Map()

- uses common::static_object

* Update counter init_order testing

---------

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: bwelton <bwelton@users.noreply.github.com>
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>

[ROCm/rocprofiler-sdk commit: ef71cc38c1]
2024-01-11 23:35:12 -08:00
Benjamin Welton ef71cc38c1 Fix init order for getMetricIdMap (#341)
* Fix init order for getMetricIdMap

Ensure that getMetricIdMap() stays valid until the destruction
of the counting service. Adds a test to ensure that this behavior
is maintained (this test fails without this change).

* source formatting (clang-format v11) (#343)

Co-authored-by: bwelton <bwelton@users.noreply.github.com>

* cmake formatting (cmake-format) (#342)

Co-authored-by: bwelton <bwelton@users.noreply.github.com>

* Remove threading

* Update usage of rocprofiler::counters::get{Metric,MetricId}Map()

- uses common::static_object

* Update counter init_order testing

---------

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: bwelton <bwelton@users.noreply.github.com>
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>
2024-01-11 23:35:12 -08:00
Sameer Sahasrabuddhe 2dba215643 SWDEV-295298: introduce warp sync functions
The following builtins from the CUDA spec are implemented:

  - __all_sync, __any_sync, __ballot_sync and __activemask
  - __match_any_sync and __match_all_sync
  - __shfl_sync, __shfl_up_sync, __shfl_down_sync, and __shfl_xor_sync

The following builtins are NOT implemented, pending support in the compiler:

  - __reduce_add_sync, __reduce_min_sync, __reduce_max_sync
  - __reduce_and_sync, __reduce_or_sync, __reduce_xor_sync

Change-Id: I07dedbbfe5449f4b5c9b040bed59f5603ccec8c3


[ROCm/clr commit: c5ab5680b4]
2024-01-11 22:31:27 -05:00
Sameer Sahasrabuddhe c5ab5680b4 SWDEV-295298: introduce warp sync functions
The following builtins from the CUDA spec are implemented:

  - __all_sync, __any_sync, __ballot_sync and __activemask
  - __match_any_sync and __match_all_sync
  - __shfl_sync, __shfl_up_sync, __shfl_down_sync, and __shfl_xor_sync

The following builtins are NOT implemented, pending support in the compiler:

  - __reduce_add_sync, __reduce_min_sync, __reduce_max_sync
  - __reduce_and_sync, __reduce_or_sync, __reduce_xor_sync

Change-Id: I07dedbbfe5449f4b5c9b040bed59f5603ccec8c3
2024-01-11 22:31:27 -05:00
Saleel Kudchadker 9864d6e56c SWDEV-301667 - Missing marker for hipExtLaunch* optimization
- Use isFenceDirty() check in hipStreamQuery.
In case if there is a stopEvent for hipExtLaunch* API we avoid system
scope if there are consecutive hipExtLaunch* and replace them with agent
scope, but then if one uses hipStreamQuery it should detect we need a
barrier with system scope.

Change-Id: I539810f659d4e1c49104e3a0da6ff2803c6b3dee


[ROCm/clr commit: 4b3c4709ff]
2024-01-11 18:19:30 -05:00
Saleel Kudchadker 4b3c4709ff SWDEV-301667 - Missing marker for hipExtLaunch* optimization
- Use isFenceDirty() check in hipStreamQuery.
In case if there is a stopEvent for hipExtLaunch* API we avoid system
scope if there are consecutive hipExtLaunch* and replace them with agent
scope, but then if one uses hipStreamQuery it should detect we need a
barrier with system scope.

Change-Id: I539810f659d4e1c49104e3a0da6ff2803c6b3dee
2024-01-11 18:19:30 -05:00
jiabaxie 5def5b2ec5 SWDEV-437702 - implemented hipGetProcAddress
Change-Id: I70498841aefae85593b8ae1abf34d1cec8f232bc


[ROCm/hip commit: a4e5fd1ae1]
2024-01-11 18:04:43 -05:00
jiabaxie a4e5fd1ae1 SWDEV-437702 - implemented hipGetProcAddress
Change-Id: I70498841aefae85593b8ae1abf34d1cec8f232bc
2024-01-11 18:04:43 -05:00
Sam Wu c785a58e99 [ROCDOC-95] Standardize documentation for ReadtheDocs
Apply the following changes to project documentation for ReadtheDocs:

add version number to documentation left navigation bar and page title
add an "About" section with a license page
enable htmlzip, pdf, epub formats when publishing on Read the Docs
set pdf title, author, copyright, and version
rename .sphinx/.doxygen to sphinx/doxygen
remove docBin from URL
update rocm-docs-core dependency
update dependabot config

Change-Id: Ife8c89a2e9323f436b3e54ef2a9e013c19b3b228


[ROCm/rocm_smi_lib commit: 67dc4b0f2a]
2024-01-11 17:47:58 -05:00
Sam Wu 67dc4b0f2a [ROCDOC-95] Standardize documentation for ReadtheDocs
Apply the following changes to project documentation for ReadtheDocs:

add version number to documentation left navigation bar and page title
add an "About" section with a license page
enable htmlzip, pdf, epub formats when publishing on Read the Docs
set pdf title, author, copyright, and version
rename .sphinx/.doxygen to sphinx/doxygen
remove docBin from URL
update rocm-docs-core dependency
update dependabot config

Change-Id: Ife8c89a2e9323f436b3e54ef2a9e013c19b3b228
2024-01-11 17:47:58 -05:00
Rahul Manocha ee3f99f59f SWDEV-421025 - Graph Kernel node priority attribute
Change-Id: I6a4f08020e8aa1ff6fb06eabd376aa266b9c04bc


[ROCm/hip commit: f653cd037a]
2024-01-11 16:57:12 -05:00
Rahul Manocha f653cd037a SWDEV-421025 - Graph Kernel node priority attribute
Change-Id: I6a4f08020e8aa1ff6fb06eabd376aa266b9c04bc
2024-01-11 16:57:12 -05:00
jujiang 898e647f84 SWDEV-410522 - Fix a doxygen warning in a header in AMD platform
Change-Id: I29cd222af453c42a5d472081747bacf0c3f8d17c


[ROCm/clr commit: 2a2fea6c07]
2024-01-11 13:40:25 -05:00
jujiang 2a2fea6c07 SWDEV-410522 - Fix a doxygen warning in a header in AMD platform
Change-Id: I29cd222af453c42a5d472081747bacf0c3f8d17c
2024-01-11 13:40:25 -05:00