Graf commitů
64512 Commity
Autor SHA1 Zpráva Datum
Bill(Shuzhou) Liu 84882db8fc Use Unified Changelog Template
The CHANGELOG.md is added to track changes.

Change-Id: I33547cb7f1596b4b8abf206aebdd664649d4d19f


[ROCm/rocm_smi_lib commit: b40933b895]
2023-02-21 14:27:55 -06:00
Bill(Shuzhou) Liu b40933b895 Use Unified Changelog Template
The CHANGELOG.md is added to track changes.

Change-Id: I33547cb7f1596b4b8abf206aebdd664649d4d19f
2023-02-21 14:27:55 -06:00
coleramos425 1deb399bf8 Comply to Python formatting
Signed-off-by: coleramos425 <colramos@amd.com>


[ROCm/rocprofiler-compute commit: 7c9d317527]
2023-02-21 13:02:43 -06:00
coleramos425 7c9d317527 Comply to Python formatting
Signed-off-by: coleramos425 <colramos@amd.com>
2023-02-21 13:02:43 -06:00
coleramos425 fdd5a54b4f Specify returncode for rocm-smi error (#85)
Signed-off-by: coleramos425 <colramos@amd.com>


[ROCm/rocprofiler-compute commit: c4eaa1d45b]
2023-02-21 12:40:41 -06:00
coleramos425 c4eaa1d45b Specify returncode for rocm-smi error (#85)
Signed-off-by: coleramos425 <colramos@amd.com>
2023-02-21 12:40:41 -06:00
Ajay 5fe52ac30f SWDEV-384100 - HIP support for CUDA 12.0
Apps are failing to build due to undefined deprecated texture APIs

Change-Id: I1fb64adc4bc0ba6ee6ecaa65d54b34da0327e6a3


[ROCm/hipother commit: 3fc34b8bf5]
2023-02-21 12:23:48 -05:00
Ajay 3fc34b8bf5 SWDEV-384100 - HIP support for CUDA 12.0
Apps are failing to build due to undefined deprecated texture APIs

Change-Id: I1fb64adc4bc0ba6ee6ecaa65d54b34da0327e6a3
2023-02-21 12:23:48 -05:00
Ajay a9675e0faa SWDEV-384100 - HIP support for CUDA 12.0
Apps are failing to build due to undefined deprecated texture APIs

Change-Id: I1fb64adc4bc0ba6ee6ecaa65d54b34da0327e6a3


[ROCm/clr commit: fc6a77a09f]
2023-02-21 12:23:48 -05:00
Ajay fc6a77a09f SWDEV-384100 - HIP support for CUDA 12.0
Apps are failing to build due to undefined deprecated texture APIs

Change-Id: I1fb64adc4bc0ba6ee6ecaa65d54b34da0327e6a3
2023-02-21 12:23:48 -05:00
Cole Ramos a368456cd1 Merge pull request #91 from AMDResearch/mi100-fix
Custom metrics for mi100

[ROCm/rocprofiler-compute commit: ab30dd6a21]
2023-02-21 10:10:24 -06:00
Cole Ramos ab30dd6a21 Merge pull request #91 from AMDResearch/mi100-fix
Custom metrics for mi100
2023-02-21 10:10:24 -06:00
coleramos425 b0bcc846aa Catch exceptions from rocm-smi (#85)
Signed-off-by: coleramos425 <colramos@amd.com>


[ROCm/rocprofiler-compute commit: f77173a2f1]
2023-02-21 09:47:45 -06:00
coleramos425 f77173a2f1 Catch exceptions from rocm-smi (#85)
Signed-off-by: coleramos425 <colramos@amd.com>
2023-02-21 09:47:45 -06:00
Jason Tang 562a086ff1 Merge "SWDEV-372613 - No need to link GLEW/GL for oclperf and oclruntime tests" into amd-staging
[ROCm/clr commit: cf2d46e97e]
2023-02-21 10:00:07 -05:00
Jason Tang cf2d46e97e Merge "SWDEV-372613 - No need to link GLEW/GL for oclperf and oclruntime tests" into amd-staging 2023-02-21 10:00:07 -05:00
Jenkins 4cf7d2ce98 SWDEV-299127 - Merge 'develop' into 'amd-staging'
Change-Id: Ic2aeb47a20cc1fc515228fc50e1bdb86a2620d1b


[ROCm/hip commit: 28c86d821c]
2023-02-21 12:11:16 +00:00
Jenkins 28c86d821c SWDEV-299127 - Merge 'develop' into 'amd-staging'
Change-Id: Ic2aeb47a20cc1fc515228fc50e1bdb86a2620d1b
2023-02-21 12:11:16 +00:00
Todd tiantuo Li 77a647b170 SWDEV-382501 - support new ASIC Revision for Phoenix2
Change-Id: I0225511960c9fbc50ec541511599ef87e509b567


[ROCm/clr commit: 189ce1ad05]
2023-02-21 02:54:35 -08:00
Todd tiantuo Li 189ce1ad05 SWDEV-382501 - support new ASIC Revision for Phoenix2
Change-Id: I0225511960c9fbc50ec541511599ef87e509b567
2023-02-21 02:54:35 -08:00
Rakesh Roy c506ec3604 SWDEV-377782 - Fix segmentation fault for hipLaunchKernel
- If fat binary doesn't contain code object for current gfx then inside FatBinaryInfo::ExtractFatBinary(), valid FatBinaryDeviceInfo address isn't stored in vector fatbin_dev_info_
- This raises segmentation fault during hipLaunchKernel

Change-Id: I21017338d91edbd5d9cc2d37277f66558198a129


[ROCm/clr commit: 348698ed30]
2023-02-20 09:54:44 -05:00
Rakesh Roy 348698ed30 SWDEV-377782 - Fix segmentation fault for hipLaunchKernel
- If fat binary doesn't contain code object for current gfx then inside FatBinaryInfo::ExtractFatBinary(), valid FatBinaryDeviceInfo address isn't stored in vector fatbin_dev_info_
- This raises segmentation fault during hipLaunchKernel

Change-Id: I21017338d91edbd5d9cc2d37277f66558198a129
2023-02-20 09:54:44 -05:00
Konstantin Zhuravlyov f789329113 SWDEV-1 - Add missing gpu cases in getProcName
Change-Id: I16103ab213cc70f388690df85e6a03e7a408384d


[ROCm/clr commit: caf7551dd7]
2023-02-17 15:55:09 -05:00
Konstantin Zhuravlyov caf7551dd7 SWDEV-1 - Add missing gpu cases in getProcName
Change-Id: I16103ab213cc70f388690df85e6a03e7a408384d
2023-02-17 15:55:09 -05:00
ROCm CI Service Account fb54929fc0 SWDEV-380145 - [GFX][CQE] Mathlibs and Blender build failed in both RT and CPL Jobs (#3169)
Remove --hip-device-lib-path from hip-config.cmake
and let clang determine device lib path

Change-Id: I81bfec0928480ff81ad2ef0dbbfb6feac9d96068


[ROCm/hip commit: 802e3f4397]
2023-02-17 12:16:00 -08:00
ROCm CI Service Account 802e3f4397 SWDEV-380145 - [GFX][CQE] Mathlibs and Blender build failed in both RT and CPL Jobs (#3169)
Remove --hip-device-lib-path from hip-config.cmake
and let clang determine device lib path

Change-Id: I81bfec0928480ff81ad2ef0dbbfb6feac9d96068
2023-02-17 12:16:00 -08:00
Charis Poag 2001f3620e [SWDEV-381630] Add reset partition functionality
Updates:
    * Added rsmi_dev_compute_partition_reset & rsmi_dev_nps_mode_reset
    * Added --resetcomputepartition and --resetnpsmode python smi calls
    * Added temp data files rocmsmi_boot_compute_partition_<device num>
      & rocmsmi_boot_nps_mode_partition_<device num>, writes UNKNOWN
      if data cannot be read or device does not support
    * Cleaned up NPS & compute API documentation
    * Added creation and reading of API temp files (used in reset
      functionality)
    * Cleaned up output of rocm_smi_example
    * Updated rocm_smi_example to check if running with sudo permission
      before executing write API calls (cleans up erroneous output)
    * Added template specialization for storing temp data, requires
      specific rsmi_type_t enums (restrics what data can be stored)
    * Added storage of temp data, if temp files do not exist
    * Updated google tests for NPS & compute to include reset API calls

Change-Id: I69895a466b97107617e6dbb355737b84499a76c9
Signed-off-by: Charis Poag <Charis.Poag@amd.com>


[ROCm/rocm_smi_lib commit: 77c950a4bf]
2023-02-17 12:55:08 -06:00
Charis Poag 77c950a4bf [SWDEV-381630] Add reset partition functionality
Updates:
    * Added rsmi_dev_compute_partition_reset & rsmi_dev_nps_mode_reset
    * Added --resetcomputepartition and --resetnpsmode python smi calls
    * Added temp data files rocmsmi_boot_compute_partition_<device num>
      & rocmsmi_boot_nps_mode_partition_<device num>, writes UNKNOWN
      if data cannot be read or device does not support
    * Cleaned up NPS & compute API documentation
    * Added creation and reading of API temp files (used in reset
      functionality)
    * Cleaned up output of rocm_smi_example
    * Updated rocm_smi_example to check if running with sudo permission
      before executing write API calls (cleans up erroneous output)
    * Added template specialization for storing temp data, requires
      specific rsmi_type_t enums (restrics what data can be stored)
    * Added storage of temp data, if temp files do not exist
    * Updated google tests for NPS & compute to include reset API calls

Change-Id: I69895a466b97107617e6dbb355737b84499a76c9
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
2023-02-17 12:55:08 -06:00
victzhan 46652f12a2 SWDEV-376995 - Fixed misplaced checking
Change-Id: Ia3bddabfa7fc76066541fb81723136ae6cffb0a4


[ROCm/clr commit: d6a08403d8]
2023-02-17 13:34:51 -05:00
victzhan d6a08403d8 SWDEV-376995 - Fixed misplaced checking
Change-Id: Ia3bddabfa7fc76066541fb81723136ae6cffb0a4
2023-02-17 13:34:51 -05:00
ROCm CI Service Account cdccbaec32 SWDEV-379927 - ignore --rocm-path option on nvidia platform (#3156)
Change-Id: I5e9ff0378c14dc0894a07d71aff4261c155ec1af


[ROCm/hip commit: f6f82b86da]
2023-02-17 10:34:51 -08:00
ROCm CI Service Account f6f82b86da SWDEV-379927 - ignore --rocm-path option on nvidia platform (#3156)
Change-Id: I5e9ff0378c14dc0894a07d71aff4261c155ec1af
2023-02-17 10:34:51 -08:00
Ranjith Ramakrishnan ee63dd0dc5 SWDEV-383843 - Use relative path for RUNPATH
Build script was passing relative path for setting the RUNPATH. But clinfo is overriding it and using absolute paths
Removed the same from source code

Change-Id: I505e7fdc84c53d147f3afbc6c4133f906de2110f


[ROCm/clr commit: 067beca608]
2023-02-17 11:31:48 -05:00
Ranjith Ramakrishnan 067beca608 SWDEV-383843 - Use relative path for RUNPATH
Build script was passing relative path for setting the RUNPATH. But clinfo is overriding it and using absolute paths
Removed the same from source code

Change-Id: I505e7fdc84c53d147f3afbc6c4133f906de2110f
2023-02-17 11:31:48 -05:00
German 73f02aa6dc SWDEV-382397 - Move VirtualGPU destruction back to the thread exit
OS can terminate unfinished queue thread from default stream at any
time. Potentially leaving the queue lock in a bad state and causing a
deadlock if runtime destroys VirtualGPU later from the host thread.

Change-Id: I247f102ee84e6b4dba947504933395071945c85d


[ROCm/clr commit: 28daf98f1f]
2023-02-17 10:05:49 -05:00
German 28daf98f1f SWDEV-382397 - Move VirtualGPU destruction back to the thread exit
OS can terminate unfinished queue thread from default stream at any
time. Potentially leaving the queue lock in a bad state and causing a
deadlock if runtime destroys VirtualGPU later from the host thread.

Change-Id: I247f102ee84e6b4dba947504933395071945c85d
2023-02-17 10:05:49 -05:00
Maneesh Gupta 4b8bf631d0 SWDEV-366831 - Revert "File reorg backward compatibility message changed to #error"
This reverts commit de1e612775.

Reason for revert: Need to wait for apps to make fixes first

Change-Id: I337a0b56d268093fc8689152002a6f8a8ceb258f


[ROCm/clr commit: c7576e2add]
2023-02-17 03:46:29 -05:00
Maneesh Gupta c7576e2add SWDEV-366831 - Revert "File reorg backward compatibility message changed to #error"
This reverts commit c86bd74c67.

Reason for revert: Need to wait for apps to make fixes first

Change-Id: I337a0b56d268093fc8689152002a6f8a8ceb258f
2023-02-17 03:46:29 -05:00
Anusha GodavarthySurya 06ffd15060 SWDEV-366636 - Fix performance drop in TF-RCCL models
Change-Id: Idc845bb0dab858b94b9d2720cae8308cac2e7328


[ROCm/clr commit: f98dcf9d7f]
2023-02-17 00:48:13 -05:00
Anusha GodavarthySurya f98dcf9d7f SWDEV-366636 - Fix performance drop in TF-RCCL models
Change-Id: Idc845bb0dab858b94b9d2720cae8308cac2e7328
2023-02-17 00:48:13 -05:00
Rakesh Roy 7323f7a4cf SWDEV-338733 - Implement hipArrayGet* APIs
- Add implementation for hipArrayGetInfo, hipArrayGetDescriptor &
  hipArray3DGetDescriptor APIs

Change-Id: I181a472066006bc3bd0d987408ea67e218310983


[ROCm/hipother commit: bb635fd5ff]
2023-02-16 23:04:29 -05:00
Rakesh Roy bb635fd5ff SWDEV-338733 - Implement hipArrayGet* APIs
- Add implementation for hipArrayGetInfo, hipArrayGetDescriptor &
  hipArray3DGetDescriptor APIs

Change-Id: I181a472066006bc3bd0d987408ea67e218310983
2023-02-16 23:04:29 -05:00
Rakesh Roy 9c51780b13 SWDEV-338733 - Implement hipArrayGet* APIs
- Add implementation for hipArrayGetInfo, hipArrayGetDescriptor &
  hipArray3DGetDescriptor APIs

Change-Id: I181a472066006bc3bd0d987408ea67e218310983


[ROCm/clr commit: 287d07ccd4]
2023-02-16 23:04:29 -05:00
Rakesh Roy 287d07ccd4 SWDEV-338733 - Implement hipArrayGet* APIs
- Add implementation for hipArrayGetInfo, hipArrayGetDescriptor &
  hipArray3DGetDescriptor APIs

Change-Id: I181a472066006bc3bd0d987408ea67e218310983
2023-02-16 23:04:29 -05:00
Ioannis Assiouras 5007a5f15d SWDEV-381402 - convert host_device to host_context of type amd::Context*
Change-Id: Ic1cb9a3fa64e16699fad7e9ec6679f1d34b14bef


[ROCm/clr commit: 8b87ccb8cc]
2023-02-16 17:49:12 -05:00
Ioannis Assiouras 8b87ccb8cc SWDEV-381402 - convert host_device to host_context of type amd::Context*
Change-Id: Ic1cb9a3fa64e16699fad7e9ec6679f1d34b14bef
2023-02-16 17:49:12 -05:00
jatang 6723b780ca SWDEV-372613 - No need to link GLEW/GL for oclperf and oclruntime tests
Change-Id: I0b9fe78f3fb6d93a4e7d0fe66ea279291716f30e


[ROCm/clr commit: 589988ee47]
2023-02-16 15:07:23 -05:00
jatang 589988ee47 SWDEV-372613 - No need to link GLEW/GL for oclperf and oclruntime tests
Change-Id: I0b9fe78f3fb6d93a4e7d0fe66ea279291716f30e
2023-02-16 15:07:23 -05:00
kjayapra-amd 9c479ada12 SWDEV-376697 - Pass uncached memory flag when hipDeviceMallocUncached memory is requested.
Change-Id: Idc12948047fdb69876e31edf34a8bcda46c9c303


[ROCm/clr commit: 193c5e387e]
2023-02-16 14:05:54 -05:00
kjayapra-amd 193c5e387e SWDEV-376697 - Pass uncached memory flag when hipDeviceMallocUncached memory is requested.
Change-Id: Idc12948047fdb69876e31edf34a8bcda46c9c303
2023-02-16 14:05:54 -05:00