AI/rocm-systems
2
0
Fork 0
Code Issues Pull-Requests Actions Pakete Projekte Releases Wiki Aktivität
12.819 Commits 3 Branches 0 Tags
f44ca70bb7420927912d93c6837c2d76a10d6484
Commit graph

12819 Commits

Dieser Branch
Dieser Branch
Alle Branches
Autor SHA1 Nachricht Datum
Julia Jiang f44ca70bb7 Revert "SWDEV-444098 - remove rocm-ocl-icd packaging" 
This reverts commit 5f68a4531a.

Reason for revert: due to windows staging branch using Opencl-icd-loader master

Change-Id: I9cca7564a21de1733665a34da6f0322aa3b886e7
 2024-03-22 16:59:23 -04:00
Jaydeep Patel 0be92b8f09 SWDEV-452299 - Pass dst pitch while capturing hipMemcpyParam2DAsync & elementSize should be 1 as width is in bytes while capturing hipMemset2DAsync. 
Change-Id: I8f9122a30cba0a07c097dfd7609432090caab142
 2024-03-21 12:49:34 -04:00
Ioannis Assiouras ad32e604c7 SWDEV-449146 - Call hipDeviceSynchronize before removeFatBinary 
Added call to hipDeviceSynchronize in __hipUnregisterFatBinary
to ensure that all HSA async signal handlers complete before removeFatBinary

Change-Id: I756fecca1c2a5eae092613d8079de266399e5685
 2024-03-20 23:40:46 +00:00
Matt Arsenault 67473ba345 SWDEV-1 - Do not pass daz_opt to comgr link options 
This no longer does anything.

Change-Id: I0643198a46a534a76454a5b461d010ed1776a89a
 2024-03-19 22:55:44 -04:00
Ioannis Assiouras be2bdabb76 SWDEV-430437,SWDEV-434702 - Split the streamset per device 
Change-Id: If1bcca45825c9899462bb95ed6f637f5af806cc8
 2024-03-18 19:08:52 -04:00
German Andryeyev 1239309c90 SWDEV-449558 - Update barrier's logic 
PAL optimized the logic for the barriers, which caused failures with CP DMA on Navi4x.
Change barrier's code to match the most recent PAL optimizations.

Change-Id: I55eeab20f51eb8e920bcbb4b55fbe3c7f77fd3fa
 2024-03-18 10:52:32 -04:00
Sameer Sahasrabuddhe 4ec3a977b2 SWDEV-445460: Changelog for warp sync builtins in HIP 6.2. 
Change-Id: I79de9ce945a4d9413a8fbb8bbe162883ffd3b63c
 2024-03-18 00:51:28 -04:00
Julia Jiang 5f68a4531a SWDEV-444098 - remove rocm-ocl-icd packaging 
Change-Id: If0234ff03cf86e1f0435e743c1df75a3e179d48c

Jenifer helped trigger a combined PSDB for the below three changes
https://gerrit-git.amd.com/c/compute/ec/prototype/+/1020194
https://gerrit-git.amd.com/c/compute/ec/clr/+/1003601
https://gerrit-git.amd.com/c/compute/ec/OpenCL-ICD-Loader/+/1012373

Combined PSDB was successful:
http://rocm-ci.amd.com/job/compute-psdb-staging/27382/
Change-Id: If0234ff03cf86e1f0435e743c1df75a3e179d48c
 2024-03-15 10:46:07 -05:00
Ioannis Assiouras 96f5c44851 SWDEV-451166 - Disable kernel args for non-XGMI if HDP flush register is invalid 
Change-Id: I227e046e2b9cb25476a50240f5d070adbd558f21
 2024-03-15 05:27:52 -04:00
Anusha GodavarthySurya 476dc4c87c SWDEV-436126 - hipStreamDestroy to remove stream from capture list 
Change-Id: Ie735f27b9c5df4faf39bd4be10fbe4401fd88cdb
 2024-03-14 11:32:10 -04:00
David Salinas 383ed4cedf roc-obj uses incorrect path to find ROCm's llvm-objdump 
SWDEV-448278 - [LLNLA-260] (ELCAP-546) roc-obj uses bad path

Change-Id: I1a19f1fea29b301cfc183018fa050a8b8aeaaf02
 2024-03-14 10:55:25 -04:00
Satyanvesh Dittakavi 684fd60c8f SWDEV-444021 - Implement hipGetFuncBySymbol 
Change-Id: I7ef13d02c5b5c6ce2386ccb92b5602d005b35988
 2024-03-14 02:56:30 -04:00
Jaydeep Patel 43e2bb6126 SWDEV-447384 - Handle double free case. 
Change-Id: I7a1d45c2f7510180911507d56ec863982b161f6c
 2024-03-13 23:08:27 -04:00
Jacob Lambert 6311ed8a8e SWDEV-371628 Use integrated Comgr action to link ROCm device libs 
Previously, we used the following approach and Comgr actions
for device lib linking:

    AMD_COMGR_COMPILE_SOURCE_TO_BC (compile with clang driver)
    AMD_COMGR_ADD_DEVICE_LIBRARIES (link in device libs with
      llvm-link API)

However, the clang driver can link in device libraries as part
of compilation, assuming a --rocm-path is set. In this context,
this is accomplished by using the following Comgr action instead:

  AMD_COMGR_COMPILE_SOURCE_WITH_DEVICE_LIBS_TO_BC (compile and
    link in device libs with clang driver)

Change-Id: Ie0bbee7d9a12672536b6d751056a941128ed58be
 2024-03-13 20:34:59 -04:00
jiabaxie 14bcbcc25b SWDEV-450204 - hipGetProcAddress check for null corrected 
Change-Id: Ic34b0d561d3d70f3411a60924a64c3b89954394f
 2024-03-13 16:44:08 -04:00
pghafari c13864abed SWDEV-430356 - updated VArange to use SVM 
Change-Id: Ice3582732dd6f43af6d566e5420234a581a1c194
 2024-03-13 11:47:36 -04:00
Jaydeep Patel 79162c244b SWDEV-447525 - Fix merge conflict. 
Change-Id: I9dc852e634fb702a0d17ace3fad53e14178d4da0
 2024-03-13 11:52:28 +00:00
Jaydeep Patel b72d8da1bd SWDEV-447525 - maxSize param for Mem pool props. 
Change-Id: Ib743611a66921aeaa449b6622b933cc8b9190d87
 2024-03-13 00:58:48 -04:00
Saleel Kudchadker f1adecd186 SWDEV-301667 - Use right macros and level for logging 
- Sometimes we want to mask out kernel names, use right level for kernel
logging

Change-Id: Ideae9647c57b86ae390ff2f4131f6d8c6df5c086
 2024-03-12 19:00:03 -04:00
kjayapra-amd f5ca620baa SWDEV-423835 - Fixing kernel launch issues on Virtual Memory Management path. 
Change-Id: I9f5e8a3d83af3809b2c50b21a10697e26113dd23
 2024-03-12 17:22:07 -04:00
taosang2 dd43dc930d SWDEV-448792 - Fix clinfo local buffer type 
Change-Id: I0228c5e97ce7c466983fd4af71d25e7f8161c2c8
 2024-03-12 13:48:59 -04:00
taosang2 4c82df817f SWDEV-450225 - Add missing targets in hipRtc 
Change-Id: Ifccdddfc8ac72585f6ba145578e5bbb7e3cd1643
 2024-03-12 10:34:08 -04:00
German Andryeyev 1d8562eb7d SWDEV-311271 - Correct the error codes in mempool 
Change-Id: Iacf8ad2cc454dfe53ccdb47c08d871b24ecf4107
 2024-03-12 10:18:44 -04:00
Satyanvesh Dittakavi 1b25484f0f SWDEV-447405 - Reset the last SDMA engine after every few copies 
The copies can get blocked if the last SDMA engine is used by another
copy and this can lead to perf drop in some of the tests like Gromacs.
Resetting the last engine by checking the engine status and fetching the
new mask after few copies can avoid this.

Change-Id: I8fe8ea678db508d291c6242f3741fa9215e99921
 2024-03-12 02:10:27 -04:00
sdashmiz 315d0fc46e SWDEV-447458 - remove check for image support 
- remove the check from hipmallocpitch

Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: Ife3a95381be703b1c1b7492abfd4432c4a8141c8
 2024-03-11 11:39:14 -04:00
German Andryeyev 0f3391b93e SWDEV-311271 - Enable mempool under Windows 
Change-Id: Ifa4cac4a8d52e031d63f62515439ca09efe7b4cb
 2024-03-11 10:45:51 -04:00
Shilei Tian a09204388a SWDEV-442724 - Remove uses of -amdgpu-early-inline-all to enable function calls 
Currently we force inlining everything for HIP. Now we'd like to enable function
supports. The first step is to remove uses of `-amdgpu-early-inline-all` in
various places. This patch is to remove all of them from clr.

Change-Id: Ib0cad1f586714c9989778b00746aa4c47a4eec95
 2024-03-11 10:02:12 -04:00
Sameer Sahasrabuddhe 0421a35f71 SWDEV-295298: hide warp sync builtins in ROCm 6.2 (absent in 6.1) 
Change-Id: I6d59e4f3a133dce5ea0521fc9d09f93f9fd0f376
 2024-03-11 02:44:20 -04:00
Ioannis Assiouras 5e294f89a8 SWDEV-444670 - Register the Runtime::tearDown function to be called at exit 
Change-Id: I2200b8154ce6806bf99150d0d68f6f20138fa3dc
 2024-03-08 18:46:59 -05:00
Jaydeep Patel 66f6a3e255 SWDEV-446219 - Validate option value. 
Change-Id: Ifccdddfc8ac71585f6ba045578e5bbb7e3cd1649
 2024-03-07 12:53:34 -05:00
Anusha GodavarthySurya 4feb1f9337 SWDEV-448586 - Added implementation for new API hipStreamBeginCaptureToGraph 
Change-Id: I1ce802102cef2b66c92d3375f769983841de793f
 2024-03-07 05:24:49 +00:00
German Andryeyev 17d0c166d2 SWDEV-311271 - Add dependency tracking for streams 
Mempool has capability to track dependency between streams for
faster memory reuse. Enable that capability.

Change-Id: I28266a7e38d0fc4c5d027b9542d3719653840821
 2024-03-05 12:53:09 -05:00
German Andryeyev 11a6be1ede SWDEV-311271 - Return different errors 
Match errors with the tests

Change-Id: I32db83843e45e0f09359149ea9fd7a532c881e16
 2024-03-04 17:02:30 -05:00
Rahul Manocha e5e7687fd8 SWDEV-448281 - Fix for CG Tile Partition Dynamic Size 
Change-Id: Ibdccec45470caf55a8c00194e6da0dde73e4b180
 2024-03-04 13:00:24 -05:00
Vikram 6f390f5af9 SWDEV-424956 - Fix OpenCL printf bug while printing vectors of half type 
OpenCL printf handling did not process vector of half precision floats properly
 (mainly because compiler packs 2 halfs into a dword and runtime failed to extract the
 individual parts).

 This patch fixes the issue.

Change-Id: Ia1f15ccfb5db52b71c43cfd588dd38f551ee5277
 2024-03-04 03:53:18 -05:00
Ioannis Assiouras 74cae705ae SWDEV-445960 - Fix __umul64hi __mul64hi on windows 
Change-Id: I4857d269bc80fcc0e6b77df941f24ff1964a4d42
 2024-03-03 12:42:42 -05:00
cadolphe 47e3ed545d Revert "SWDEV-374668 - Observed one subtest failure in ocltst32" 
This reverts commit 60d413a5b9.

Change-Id: I42d6248204107f741a462101f9608e01a7ba91d5
 2024-03-02 14:05:44 -05:00
Saleel Kudchadker 984c86f407 SWDEV-301667 - Better log 
- Print SWq for AQL packets, this helps correlating a stream to the HWq
mapped

Change-Id: I610430c0872a1abc6636027c00163ec46983cd65
 2024-03-01 16:43:06 -05:00
Jatin Chaudhary 7f9651d4bb SWDEV-441629 - add some missing hiprtc link options 
Change-Id: I10019823e212c6d4de6df30c0fe1c20d9e992f29
 2024-03-01 07:49:57 -05:00
ChingShihLi 6723277ad4 SWDEV-413377 - Segfault for hipLaunchKernel with multi GPUs 
- Using runtime unbundler, no any gfx device can load fat binary,
if there is any device without available code object.

- Extract available code object to corresponding gfx devices. So
users can work ROCm with those ready devices without segmentation
fault.

Change-Id: I9f14c65ecebf2d3c4b127a007cb434a3ae98c450
 2024-03-01 02:24:22 -05:00
jujiang 574d946cfb SWDEV-444835 - Porting changelog from 6.1 to staging 
Change-Id: I85b2993485da797df4ac3e146481c341680c4595
 2024-02-29 23:44:02 -05:00
Saleel Kudchadker 94c7004df8 SWDEV-301667 - Increase default signal pool to 4096 
Change-Id: I4ab23b0f87e295b40ab76ad6e96249d11b8ad04d
 2024-02-29 22:52:02 +00:00
Saleel Kudchadker 68f40f78dd SWDEV-443760 - Enable device kernel args for MI300 
- Enable Device kernel args for MI300* for now.
- Fix a perf issue which impacts graph instantiate when dev kernel args
are enabled.

Change-Id: I962e58fd9d8dd1a8db95e601cb03a8e9c7bac97f
 2024-02-28 19:10:04 -05:00
Rahul Manocha 2c0fa829b4 [SWDEV-448077][SWDEV-448067] - Changes to Kernel Node Attribute related APIs 
Change-Id: Ibbf773fd5f134a62b7ce04f6956b10c1086b1782
 2024-02-28 18:14:41 -05:00
Anusha GodavarthySurya e0e63eb04d SWDEV-447545 - Fix Enable/Disable node with hipGraph 
Node can be enabled/disabled only for kernel, memcpy and memset nodes.
If the node is disabled it becomes empty node.
To maintain ordering just enqueue marker with respective node dependencies.

Change-Id: I710f3e88ab4e76c81f6f86a40a7dc61fd4c7e440
 2024-02-28 17:34:03 -05:00
Anusha GodavarthySurya ea66946399 SWDEV-444988 - Fix hiddenheap variable initialization 
Change-Id: Icdb32ef4718b8ca0beedd569a4aaf40a9f4fb49d
 2024-02-28 17:22:32 -05:00
German Andryeyev d808ae6782 SWDEV-311271 - Update mempool errors 
Add extra validations for negative tests.

Change-Id: I7b2879b8840b5d1b6c5cb0d919a651cd868b7529
 2024-02-28 17:13:49 -05:00
Sourabh Betigeri 3fdd46ae59 SWDEV-425640 - An instantiated graphExec should retain a copy of every reference in the source graph 
Change-Id: Idf6b224449ca642af2860b33dc739f51a6248e4c
 2024-02-28 12:04:53 -05:00
Satyanvesh Dittakavi 15a058cb60 SWDEV-446781 - Check validity of hiprtc link state before using it 
Change-Id: I864e8d1a79bb28d9df0e4ee528cb50a96bc245cc
 2024-02-28 00:27:17 -05:00
Anusha GodavarthySurya 4b4ec7fc52 SWDEV-445981 - Handle hipGraphExecUpdate to update graph kernel node params with graph performance optimizations 
Change-Id: I3b05c6bfc83404152bcae9b31cfdf56af7cc61a4
 2024-02-27 15:20:23 -05:00