Граф коммитов
64512 Коммитов
Автор SHA1 Сообщение Дата
Xiaogang Chen cc4fb2d1a9 libhsakmt: Revert "libhsakmt: Update FD creation logic"
This reverts commit fd48f14ceb.
Current amdgpu exposes one render node for one gpu node/partition,
revert to previous way to open render node at Thunk.

Signed-off-by: Xiaogang Chen<Xiaogang.Chen@amd.com>
Change-Id: I436be74f8e872a7ab5c4a1420b4ea884f5a00e57
2023-05-29 00:15:14 -04:00
Xiaogang Chen 51392afedb libhsakmt: Revert "libhsakmt: Update FD creation logic"
This reverts commit 89ce41694f.
Current amdgpu exposes one render node for one gpu node/partition,
revert to previous way to open render node at Thunk.

Signed-off-by: Xiaogang Chen<Xiaogang.Chen@amd.com>
Change-Id: I436be74f8e872a7ab5c4a1420b4ea884f5a00e57


[ROCm/ROCR-Runtime commit: cc4fb2d1a9]
2023-05-29 00:15:14 -04:00
Jenkins 4199cce0da Merge 'develop' into 'amd-staging'
Change-Id: I6079c9f3a2edf4e42b8d6f7f3902322cfca02cf6


[ROCm/hip-tests commit: 4448ae9f2a]
2023-05-28 23:10:34 +00:00
Jenkins 4448ae9f2a Merge 'develop' into 'amd-staging'
Change-Id: I6079c9f3a2edf4e42b8d6f7f3902322cfca02cf6
2023-05-28 23:10:34 +00:00
Jenkins 046156784d SWDEV-299127 - Merge 'develop' into 'amd-staging'
Change-Id: I3ea1fb084d5b2f3b2bc861aebdb4edfdd348a54b


[ROCm/hip commit: adcb387c30]
2023-05-28 23:10:30 +00:00
Jenkins adcb387c30 SWDEV-299127 - Merge 'develop' into 'amd-staging'
Change-Id: I3ea1fb084d5b2f3b2bc861aebdb4edfdd348a54b
2023-05-28 23:10:30 +00:00
David Yat Sin b290d65ec9 Bump interface versions due to hsa_amd_memory_async_copy_on_engine added
Change-Id: Iff36719e800280d58217647bb70d3b5d5fcc91fe
2023-05-26 12:04:06 +00:00
David Yat Sin 3661d76c74 Bump interface versions due to hsa_amd_memory_async_copy_on_engine added
Change-Id: Iff36719e800280d58217647bb70d3b5d5fcc91fe


[ROCm/ROCR-Runtime commit: b290d65ec9]
2023-05-26 12:04:06 +00:00
Vikram 5ba0fb5802 SWDEV-354050 - Runtime support for HIP non hostcall printf
Relates to https://reviews.llvm.org/D150427,

Each printf call populates buffer with following data
1. Control DWord - contains info regarding stream, format string constness and size of data frame
   (see http://gerrit-git.amd.com/c/lightning/ec/device-libs/+/857722 for more info)
2. Hash of the format string (if constant) else the format string itself
3. Printf arguments (each aligned to 8 byte boundary)

Change-Id: I7e320deb343921b4b4cfaf08a2be2883e0bc1f65


[ROCm/clr commit: 7b6a8f1702]
2023-05-26 01:55:20 -04:00
Vikram 7b6a8f1702 SWDEV-354050 - Runtime support for HIP non hostcall printf
Relates to https://reviews.llvm.org/D150427,

Each printf call populates buffer with following data
1. Control DWord - contains info regarding stream, format string constness and size of data frame
   (see http://gerrit-git.amd.com/c/lightning/ec/device-libs/+/857722 for more info)
2. Hash of the format string (if constant) else the format string itself
3. Printf arguments (each aligned to 8 byte boundary)

Change-Id: I7e320deb343921b4b4cfaf08a2be2883e0bc1f65
2023-05-26 01:55:20 -04:00
Jeremy Newton b1266e557e SWDEV-361988 - Fix hip library install permissions
"FILES" installs files as 644, but we want libraries to be 755, which
we can do with "PROGRAMS".

Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com>
Change-Id: I155ed77482839ff6d71f90239a014d239e20f4b8


[ROCm/clr commit: 2cda949920]
2023-05-25 22:47:32 -04:00
Jeremy Newton 2cda949920 SWDEV-361988 - Fix hip library install permissions
"FILES" installs files as 644, but we want libraries to be 755, which
we can do with "PROGRAMS".

Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com>
Change-Id: I155ed77482839ff6d71f90239a014d239e20f4b8
2023-05-25 22:47:32 -04:00
coleramos425 c034f1e41c Detect illogical counter values and throw warning (#103)
Signed-off-by: coleramos425 <colramos@amd.com>


[ROCm/rocprofiler-compute commit: 70405f184f]
2023-05-25 15:45:05 -05:00
coleramos425 70405f184f Detect illogical counter values and throw warning (#103)
Signed-off-by: coleramos425 <colramos@amd.com>
2023-05-25 15:45:05 -05:00
Kent Russell 478a68d49c kfdtest: Test XNACK on and off for SVM tests
Add parameterization for KFDSVM tests so that we test with both XNACK
enabled and XNACK disabled. This will be overridden by HSA_XNACK, if set

Change-Id: Ie96eb61c03115f947e08cfa076ac459f7440f5d8
2023-05-25 12:08:16 -04:00
Kent Russell 8bbcfca082 kfdtest: Test XNACK on and off for SVM tests
Add parameterization for KFDSVM tests so that we test with both XNACK
enabled and XNACK disabled. This will be overridden by HSA_XNACK, if set

Change-Id: Ie96eb61c03115f947e08cfa076ac459f7440f5d8


[ROCm/ROCR-Runtime commit: 478a68d49c]
2023-05-25 12:08:16 -04:00
coleramos425 cf76ed6546 Remove submodule checkout from CI
Signed-off-by: coleramos425 <colramos@amd.com>


[ROCm/rocprofiler-compute commit: 8af6efc0be]
2023-05-25 10:26:59 -05:00
coleramos425 8af6efc0be Remove submodule checkout from CI
Signed-off-by: coleramos425 <colramos@amd.com>
2023-05-25 10:26:59 -05:00
gobhardw 7ef30a17c8 Fix async copy test and logic flaw in tracer test
Change-Id: I046f12cc7513f7b555b4c148d5ec34bc2354280e
2023-05-25 09:46:42 -04:00
gobhardw 9d491815d7 Fix async copy test and logic flaw in tracer test
Change-Id: I046f12cc7513f7b555b4c148d5ec34bc2354280e


[ROCm/rocprofiler commit: 7ef30a17c8]
2023-05-25 09:46:42 -04:00
Giovanni LB cd9ad067fa SWDEV-401887: Removed output folder sent to att.py
Change-Id: I3e09c1ead9c24a5fd03cd26e4bee18d580a2f377
2023-05-25 07:52:01 -03:00
Giovanni LB 11acda6290 SWDEV-401887: Removed output folder sent to att.py
Change-Id: I3e09c1ead9c24a5fd03cd26e4bee18d580a2f377


[ROCm/rocprofiler commit: cd9ad067fa]
2023-05-25 07:52:01 -03:00
Satyanvesh Dittakavi 790f51e1f3 SWDEV-401850 - Fix hipCreateChannelDescHalf APIs
- Address ChannelDescHalf1 and ChannelDescHalf2 APIs not returning
correct Channel Descriptors in amd headers
- Add missing hipCreateChannelDescHalf APIs in nvidia headers

Change-Id: I558847425c2459d3cde329ca2e926d882bb0a9dd


[ROCm/hipother commit: e5e4a16996]
2023-05-25 04:36:24 -04:00
Satyanvesh Dittakavi e5e4a16996 SWDEV-401850 - Fix hipCreateChannelDescHalf APIs
- Address ChannelDescHalf1 and ChannelDescHalf2 APIs not returning
correct Channel Descriptors in amd headers
- Add missing hipCreateChannelDescHalf APIs in nvidia headers

Change-Id: I558847425c2459d3cde329ca2e926d882bb0a9dd
2023-05-25 04:36:24 -04:00
Satyanvesh Dittakavi 094b2bfa9e SWDEV-401850 - Fix hipCreateChannelDescHalf APIs
- Address ChannelDescHalf1 and ChannelDescHalf2 APIs not returning
correct Channel Descriptors in amd headers
- Add missing hipCreateChannelDescHalf APIs in nvidia headers

Change-Id: I558847425c2459d3cde329ca2e926d882bb0a9dd


[ROCm/clr commit: 586731c2cc]
2023-05-25 04:36:24 -04:00
Satyanvesh Dittakavi 586731c2cc SWDEV-401850 - Fix hipCreateChannelDescHalf APIs
- Address ChannelDescHalf1 and ChannelDescHalf2 APIs not returning
correct Channel Descriptors in amd headers
- Add missing hipCreateChannelDescHalf APIs in nvidia headers

Change-Id: I558847425c2459d3cde329ca2e926d882bb0a9dd
2023-05-25 04:36:24 -04:00
Hao Zhou cad1008c3e Revert "Merge amd-staging into amd-master 20230524"
This reverts commit 11be79e5fa.

Reason for revert: <commit break mainline rocm builds - RHEL-9>

Change-Id: Ideed94313b15bc8af2bd9855535d490fd84b5c3b


[ROCm/rocm_smi_lib commit: 7d508f75ab]
2023-05-25 01:53:24 -04:00
Hao Zhou 7d508f75ab Revert "Merge amd-staging into amd-master 20230524"
This reverts commit 3629c4d5ca.

Reason for revert: <commit break mainline rocm builds - RHEL-9>

Change-Id: Ideed94313b15bc8af2bd9855535d490fd84b5c3b
2023-05-25 01:53:24 -04:00
Ammar ELWazir 3368b7483f Fixing ROCSYS
Change-Id: I907e2d0dae0413ebda4f83227dfe99e8a8f86d12
2023-05-25 00:26:01 -04:00
Ammar ELWazir af2247a018 Fixing ROCSYS
Change-Id: I907e2d0dae0413ebda4f83227dfe99e8a8f86d12


[ROCm/rocprofiler commit: 3368b7483f]
2023-05-25 00:26:01 -04:00
Ammar ELWazir 078cb99d2c Removing Grafana
Change-Id: I73c84daa6f8a9da6de217b81665087662d25f4c5
2023-05-25 00:25:27 -04:00
Ammar ELWazir 6ead732610 Removing Grafana
Change-Id: I73c84daa6f8a9da6de217b81665087662d25f4c5


[ROCm/rocprofiler commit: 078cb99d2c]
2023-05-25 00:25:27 -04:00
Ammar ELWazir 6b07b139c4 Fixing Timestamps for Tracing for the trace buffer results
Change-Id: I89b36daacb6531840ac334ba4629c73a7b8f1e79
2023-05-25 01:59:54 +00:00
Ammar ELWazir e45d633356 Fixing Timestamps for Tracing for the trace buffer results
Change-Id: I89b36daacb6531840ac334ba4629c73a7b8f1e79


[ROCm/rocprofiler commit: 6b07b139c4]
2023-05-25 01:59:54 +00:00
ROCm CI Service Account 7b3a57c9b9 SWDEV-399104 - Fix hipMemcpy2DAsync_PinnedMemory.tst (#3231)
failure

- As per spec, hipMemset2D is async with respect to host when target memory is device memory
- Hence need to call hipDeviceSynchronize before changing device

Change-Id: I58e6efc9466233d80b8ba2c58c0c691c68827128



[ROCm/hip commit: 856d4b5075]
2023-05-24 17:44:22 -07:00
ROCm CI Service Account 856d4b5075 SWDEV-399104 - Fix hipMemcpy2DAsync_PinnedMemory.tst (#3231)
failure

- As per spec, hipMemset2D is async with respect to host when target memory is device memory
- Hence need to call hipDeviceSynchronize before changing device

Change-Id: I58e6efc9466233d80b8ba2c58c0c691c68827128
2023-05-24 17:44:22 -07:00
Ammar ELWazir 1bba393b1b SWDEV-389161: Fixing CTF plugin to work with the TraceBuffer
Change-Id: I4bd9f21bb91d6bd7cee1417d88a81d9d3be6ea9b
2023-05-24 22:46:40 +00:00
Ammar ELWazir c81ec8a710 SWDEV-389161: Fixing CTF plugin to work with the TraceBuffer
Change-Id: I4bd9f21bb91d6bd7cee1417d88a81d9d3be6ea9b


[ROCm/rocprofiler commit: 1bba393b1b]
2023-05-24 22:46:40 +00:00
Sriraksha Nagaraj cdc74d31db SWDEV-389161:Adding fix for tensorflow app hang with rocprofv2
Change-Id: I48ade2e853468940aa5e1fca58ba22f686d89f26
2023-05-24 22:46:40 +00:00
Sriraksha Nagaraj 4ba25a5c39 SWDEV-389161:Adding fix for tensorflow app hang with rocprofv2
Change-Id: I48ade2e853468940aa5e1fca58ba22f686d89f26


[ROCm/rocprofiler commit: cdc74d31db]
2023-05-24 22:46:40 +00:00
Cole Ramos 7e6f69dfc8 Merge pull request #132 from arghdos/fix_merge_on_grid
Fix merge on grid

[ROCm/rocprofiler-compute commit: 435ba2f54b]
2023-05-24 16:58:01 -05:00
Cole Ramos 435ba2f54b Merge pull request #132 from arghdos/fix_merge_on_grid
Fix merge on grid
2023-05-24 16:58:01 -05:00
Nicholas Curtis 6434ab2194 reformat
[ROCm/rocprofiler-compute commit: b48fb1e4cb]
2023-05-24 16:08:09 -04:00
Nicholas Curtis b48fb1e4cb reformat 2023-05-24 16:08:09 -04:00
Bill(Shuzhou) Liu 269e0e23f9 Fallback to gpu_metrics if the sysfs is not available
The gpu_metrics may have required PCI link width and speed.

Change-Id: I939d733f5f6a71088545ba042345eb1b6ad20ee5


[ROCm/rocm_smi_lib commit: a6467c4083]
2023-05-24 14:51:43 -05:00
Bill(Shuzhou) Liu a6467c4083 Fallback to gpu_metrics if the sysfs is not available
The gpu_metrics may have required PCI link width and speed.

Change-Id: I939d733f5f6a71088545ba042345eb1b6ad20ee5
2023-05-24 14:51:43 -05:00
Nicholas Curtis 9512b7479d fix column name
[ROCm/rocprofiler-compute commit: 2e403232db]
2023-05-24 15:29:29 -04:00
Nicholas Curtis 2e403232db fix column name 2023-05-24 15:29:29 -04:00
Graham Sider 0772e8d618 rocrtst: Throw on LocateKernelFile open() failures
Throw runtime error instead of returning empty string when open() fails
in LocateKernelFile()

Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: Iafa360fbc2d3c9b01b9fe7ea4c11d70bd254ccce
2023-05-24 14:31:26 -04:00
Graham Sider f0eeb60222 rocrtst: Throw on LocateKernelFile open() failures
Throw runtime error instead of returning empty string when open() fails
in LocateKernelFile()

Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: Iafa360fbc2d3c9b01b9fe7ea4c11d70bd254ccce


[ROCm/ROCR-Runtime commit: 0772e8d618]
2023-05-24 14:31:26 -04:00