AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
76,333 Commits 3 Branches 0 Tags
develop
Commit Graph

76333 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Vladana Stojiljkovic a5522a55ef SWDEV-478767 - Implement tests for hipTexRefSetAddress2D 
Change-Id: Id5b946c216097c606a1e4b839514f68d351adc19


[ROCm/hip-tests commit: b24e61671a]
 2024-10-25 14:43:45 -04:00
Vladana Stojiljkovic b24e61671a SWDEV-478767 - Implement tests for hipTexRefSetAddress2D 
Change-Id: Id5b946c216097c606a1e4b839514f68d351adc19
 2024-10-25 14:43:45 -04:00
Vladana Stojiljkovic 61948cf01d SWDEV-465142 - Enable disabled Unit_hipGraphAddMemAllocNode_Positive_FreeSeparateGraph test 
Change-Id: Ie59a9ab8b62c38f3a527a31309827633b4a16768


[ROCm/hip-tests commit: c647226ed7]
 2024-10-25 14:42:15 -04:00
Vladana Stojiljkovic c647226ed7 SWDEV-465142 - Enable disabled Unit_hipGraphAddMemAllocNode_Positive_FreeSeparateGraph test 
Change-Id: Ie59a9ab8b62c38f3a527a31309827633b4a16768
 2024-10-25 14:42:15 -04:00
Vladana Stojiljkovic fdacd9a9bd SWDEV-454681 Enable disabled Unit_hipDrvGraphAddMemsetNode_Negative_Parameters test 
Change-Id: Ie20cefe7e422234c19e2e52cea67d170bfb41090


[ROCm/hip-tests commit: 6591d26a7d]
 2024-10-25 14:38:41 -04:00
Vladana Stojiljkovic 6591d26a7d SWDEV-454681 Enable disabled Unit_hipDrvGraphAddMemsetNode_Negative_Parameters test 
Change-Id: Ie20cefe7e422234c19e2e52cea67d170bfb41090
 2024-10-25 14:38:41 -04:00
German Andryeyev 0fc7369ba5 rocr: Disable WaitAny() in AsyncEventsLoop() 
- Add the new path to avoid WaitAny() calls  in AsyncEventsLoopp() with
HSA_WAIT_ANY_DEBUG key. The new path is selected by default.
The optimizaiton combines all logic of WaitAny() in a single processing loop
and avoids extra memory allocations or ref counting.  Also it won't spin
on the CPU if all events are busy.

Change-Id: I197ce60d0d023fbb672f700d6e87702686f1f55a
 2024-10-25 14:37:02 -04:00
German Andryeyev 6617af10e6 rocr: Disable WaitAny() in AsyncEventsLoop() 
- Add the new path to avoid WaitAny() calls  in AsyncEventsLoopp() with
HSA_WAIT_ANY_DEBUG key. The new path is selected by default.
The optimizaiton combines all logic of WaitAny() in a single processing loop
and avoids extra memory allocations or ref counting.  Also it won't spin
on the CPU if all events are busy.

Change-Id: I197ce60d0d023fbb672f700d6e87702686f1f55a


[ROCm/ROCR-Runtime commit: 0fc7369ba5]
 2024-10-25 14:37:02 -04:00
David Yat Sin d90fbee9c4 rocr: find first dispatch pkt that needs scratch 
On GPUs where EOP is handled in asic, the read_dispatch_id is not always
updated after each packet. Look for the first dispatch packet that needs
scratch memory before allocating scratch.

Change-Id: Ibf4b4b485f99bf2fabfe48e9609ca99111fdafbe
 2024-10-25 14:36:40 -04:00
David Yat Sin 7d84abbc3b rocr: find first dispatch pkt that needs scratch 
On GPUs where EOP is handled in asic, the read_dispatch_id is not always
updated after each packet. Look for the first dispatch packet that needs
scratch memory before allocating scratch.

Change-Id: Ibf4b4b485f99bf2fabfe48e9609ca99111fdafbe


[ROCm/ROCR-Runtime commit: d90fbee9c4]
 2024-10-25 14:36:40 -04:00
Aryan Salmanpour c6635a9de5 VideoDecode samples - Set the default display_delay to 1 (#441) 
[ROCm/rocdecode commit: 81e341b786]
 2024-10-25 14:20:33 -04:00
Aryan Salmanpour 81e341b786 VideoDecode samples - Set the default display_delay to 1 (#441) 2024-10-25 14:20:33 -04:00
Ajay ae11292a4b SWDEV-1 - complex hiprtc negative tests 
Change-Id: I22a200b1e762651689659a7a404b5f27f1ab01a7


[ROCm/hip-tests commit: c025def8ac]
 2024-10-25 12:34:04 -04:00
Ajay c025def8ac SWDEV-1 - complex hiprtc negative tests 
Change-Id: I22a200b1e762651689659a7a404b5f27f1ab01a7
 2024-10-25 12:34:04 -04:00
Jonathan R. Madsen 07f698a532 CMake: Consistently name CMake Targets (#1082) 
* Change all rocprofiler-X target names to rocprofiler-sdk-X

* Update rocprofiler-sdk-config.cmake

- fix install tree target names
- simplify logic for using find w/ components and find w/o components

* Update rocprofiler-sdk-roctx-config.cmake

- simplify logic for using find w/ components and find w/o components

* Update samples/intercept_table/CMakeLists.txt

- demonstrate/test use of `find_package(rocprofiler-sdk ... COMPONENTS ...)`

[ROCm/rocprofiler-sdk commit: 74facf87a6]
 2024-10-25 11:17:34 -05:00
Jonathan R. Madsen 74facf87a6 CMake: Consistently name CMake Targets (#1082) 
* Change all rocprofiler-X target names to rocprofiler-sdk-X

* Update rocprofiler-sdk-config.cmake

- fix install tree target names
- simplify logic for using find w/ components and find w/o components

* Update rocprofiler-sdk-roctx-config.cmake

- simplify logic for using find w/ components and find w/o components

* Update samples/intercept_table/CMakeLists.txt

- demonstrate/test use of `find_package(rocprofiler-sdk ... COMPONENTS ...)`
 2024-10-25 11:17:34 -05:00
Ioannis Assiouras 0976bd0249 SWDEV-480400 - Add unit test for hipExtHostAlloc 
Change-Id: Ica92d1ae91bdc4feef52b7b07d55da4dff8d7542


[ROCm/hip-tests commit: 359aed1348]
 2024-10-25 11:51:42 -04:00
Ioannis Assiouras 359aed1348 SWDEV-480400 - Add unit test for hipExtHostAlloc 
Change-Id: Ica92d1ae91bdc4feef52b7b07d55da4dff8d7542
 2024-10-25 11:51:42 -04:00
Sameer Sahasrabuddhe ccb09057d8 SWDEV-490198: _sync() will be enabled by default in 6.4 
Change-Id: Id029424a9c0f6b144a7aa0e96fe8acc4a138ec51


[ROCm/clr commit: 556390f9c7]
 2024-10-25 09:54:40 -04:00
Sameer Sahasrabuddhe 556390f9c7 SWDEV-490198: _sync() will be enabled by default in 6.4 
Change-Id: Id029424a9c0f6b144a7aa0e96fe8acc4a138ec51
 2024-10-25 09:54:40 -04:00
Sameer Sahasrabuddhe 82b8d5ba27 SWDEV-490198 - _sync() will be enabled by default in 6.4 
Change-Id: I70a2c86695acf7c9b269b4576f24e0fa41781b81


[ROCm/hip commit: 174cb78806]
 2024-10-25 09:54:14 -04:00
Sameer Sahasrabuddhe 174cb78806 SWDEV-490198 - _sync() will be enabled by default in 6.4 
Change-Id: I70a2c86695acf7c9b269b4576f24e0fa41781b81
 2024-10-25 09:54:14 -04:00
Julia Jiang ec7ae95977 SWDEV-488396,489257 - Fixed the regression in CTS pipes sub-test failure 
Change-Id: Id4004f0d6da5754b12c9a21038de50472cb1fee5


[ROCm/clr commit: 9f2f6a8aa7]
 2024-10-25 05:58:46 -04:00
Julia Jiang 9f2f6a8aa7 SWDEV-488396,489257 - Fixed the regression in CTS pipes sub-test failure 
Change-Id: Id4004f0d6da5754b12c9a21038de50472cb1fee5
 2024-10-25 05:58:46 -04:00
Jonathan R. Madsen 651e28f7f0 rocprofv3: support specifying HW counters via command line (#1130) 
* rocprofv3: support specifying PMC counters via command line

- E.g. `rocprofv3 --pmc SQ_WAVES -- <app>`

* Update CHANGELOG

* updated rocprofv3 help and documentation

---------

Co-authored-by: Gopesh Bhardwaj <gopesh.bhardwaj@amd.com>

[ROCm/rocprofiler-sdk commit: d93d4d9463]
 2024-10-25 02:49:30 -05:00
Jonathan R. Madsen d93d4d9463 rocprofv3: support specifying HW counters via command line (#1130) 
* rocprofv3: support specifying PMC counters via command line

- E.g. `rocprofv3 --pmc SQ_WAVES -- <app>`

* Update CHANGELOG

* updated rocprofv3 help and documentation

---------

Co-authored-by: Gopesh Bhardwaj <gopesh.bhardwaj@amd.com>
 2024-10-25 02:49:30 -05:00
Giovanni Baraldi a6328a1481 Revert "SWDEV-310289: Adding SPI pipe selection" 
This reverts commit e5e2c6041d.

Reason for revert: Requires priv_cp_queues=1

Change-Id: Ia6c78ac25b88d7ef4703654075d54e672a6e320c
 2024-10-25 02:57:52 -04:00
Giovanni Baraldi dc25330772 Revert "SWDEV-310289: Adding SPI pipe selection" 
This reverts commit 896404efb3.

Reason for revert: Requires priv_cp_queues=1

Change-Id: Ia6c78ac25b88d7ef4703654075d54e672a6e320c


[ROCm/rocprofiler commit: a6328a1481]
 2024-10-25 02:57:52 -04:00
Giovanni Lenzi Baraldi dbad56597a Fix for clang tidy (#1155) 
[ROCm/rocprofiler-sdk commit: b5f1d70105]
 2024-10-24 20:37:05 -03:00
Giovanni Lenzi Baraldi b5f1d70105 Fix for clang tidy (#1155) 2024-10-24 20:37:05 -03:00
German Andryeyev 584c9c1ee1 SWDEV-440746 - Fix a typo with GPU_PINNED_XFER_SIZE 
Change-Id: I8fdbfb4e1c6b1274206c28a529eee9ebeaaa26fb


[ROCm/clr commit: dceb320ba7]
 2024-10-24 18:33:14 -04:00
German Andryeyev dceb320ba7 SWDEV-440746 - Fix a typo with GPU_PINNED_XFER_SIZE 
Change-Id: I8fdbfb4e1c6b1274206c28a529eee9ebeaaa26fb
 2024-10-24 18:33:14 -04:00
Sourabh Betigeri 78d068f153 SWDEV-450052 - Return if numDevices is more than device count on the platform 
Change-Id: I538106d1b02084df9cd06b41427629207312e76f


[ROCm/clr commit: 64e1b15551]
 2024-10-24 17:07:11 -04:00
Sourabh Betigeri 64e1b15551 SWDEV-450052 - Return if numDevices is more than device count on the platform 
Change-Id: I538106d1b02084df9cd06b41427629207312e76f
 2024-10-24 17:07:11 -04:00
Edgar Gabriel e21e9a8fb6 Merge pull request #32 from edgargabriel/topic/to_all_first_version 
ipc/to_all: add direct allreduce algorithm

[ROCm/rocshmem commit: 7ae0a54550]
 2024-10-24 15:59:24 -05:00
Edgar Gabriel 7ae0a54550 Merge pull request #32 from edgargabriel/topic/to_all_first_version 
ipc/to_all: add direct allreduce algorithm
 2024-10-24 15:59:24 -05:00
Giovanni Lenzi Baraldi f1c7a1a368 SWDEV-489158: Optimizing counter collection performance (#1150) 
* SWDEV-489158: Optimizing counter collection performance

* Static initializer fix

* adding sched_yield+sleep

[ROCm/rocprofiler-sdk commit: aef18896dd]
 2024-10-24 16:58:07 -03:00
Giovanni Lenzi Baraldi aef18896dd SWDEV-489158: Optimizing counter collection performance (#1150) 
* SWDEV-489158: Optimizing counter collection performance

* Static initializer fix

* adding sched_yield+sleep
 2024-10-24 16:58:07 -03:00
spolifroni-amd 3f746445a7 updated the changelog for 6.3 (#439) 
[ROCm/rocdecode commit: 8636edce54]
 2024-10-24 14:14:25 -04:00
spolifroni-amd 8636edce54 updated the changelog for 6.3 (#439) 2024-10-24 14:14:25 -04:00
Julia Jiang 8034b85e7f SWDEV-452448 - Adding more information for inline_asm sample 
Change-Id: I52b3bd804cd323eb862473da9e48a1d4cf58da07


[ROCm/hip-tests commit: edecba21f2]
 2024-10-24 13:54:42 -04:00
Julia Jiang edecba21f2 SWDEV-452448 - Adding more information for inline_asm sample 
Change-Id: I52b3bd804cd323eb862473da9e48a1d4cf58da07
 2024-10-24 13:54:42 -04:00
Julia Jiang 858a34670e SWDEV-452448 - Adding more information for inline assembly in HIP kernels 
Change-Id: I29c6ac9f3aa22d9d69bb4d545ff9250fbfe52a10


[ROCm/hip commit: e0d0dad697]
 2024-10-24 13:52:35 -04:00
Julia Jiang e0d0dad697 SWDEV-452448 - Adding more information for inline assembly in HIP kernels 
Change-Id: I29c6ac9f3aa22d9d69bb4d545ff9250fbfe52a10
 2024-10-24 13:52:35 -04:00
Aryan Salmanpour 896998cc14 Revert "Allow overriding CMAKE_CXX_COMPILER (#436)" (#440) 
This reverts commit 5b9907034e.

[ROCm/rocdecode commit: b2149abcef]
 2024-10-24 13:28:50 -04:00
Aryan Salmanpour b2149abcef Revert "Allow overriding CMAKE_CXX_COMPILER (#436)" (#440) 
This reverts commit 07ecb5e7d6.
 2024-10-24 13:28:50 -04:00
Sylvain Jeaugey 637acfba05 Merge pull request #259 from NVIDIA/fix-ncclstringtotype 
Future-proof ncclstringtotype

[ROCm/rccl-tests commit: 8dfeab9eb9]
 2024-10-24 10:28:02 -07:00
Sylvain Jeaugey 8dfeab9eb9 Merge pull request #259 from NVIDIA/fix-ncclstringtotype 
Future-proof ncclstringtotype
 2024-10-24 10:28:02 -07:00
zichguan-amd 3328cf8d6c Enable baseline comparison between workloads from the same path (#447) 
* Enable baseline comparison between workloads from the same path

Signed-off-by: zichguan-amd <zichuan.guan@amd.com>

* Formatting

Signed-off-by: zichguan-amd <zichuan.guan@amd.com>

* Change test to verify baseline works with same path

Signed-off-by: zichguan-amd <zichuan.guan@amd.com>

* Add three-way comparison test case

Signed-off-by: zichguan-amd <zichuan.guan@amd.com>

---------

Signed-off-by: zichguan-amd <zichuan.guan@amd.com>

[ROCm/rocprofiler-compute commit: 13a06f5131]
 2024-10-24 12:00:36 -05:00
zichguan-amd 13a06f5131 Enable baseline comparison between workloads from the same path (#447) 
* Enable baseline comparison between workloads from the same path

Signed-off-by: zichguan-amd <zichuan.guan@amd.com>

* Formatting

Signed-off-by: zichguan-amd <zichuan.guan@amd.com>

* Change test to verify baseline works with same path

Signed-off-by: zichguan-amd <zichuan.guan@amd.com>

* Add three-way comparison test case

Signed-off-by: zichguan-amd <zichuan.guan@amd.com>

---------

Signed-off-by: zichguan-amd <zichuan.guan@amd.com>
 2024-10-24 12:00:36 -05:00