Граф коммитов
64512 Коммитов
Автор SHA1 Сообщение Дата
Graham Sider 60831e86b2 kfdtest: Update GFX11 blacklists
Remove BLACKLIST_GFX10_NV2X from GFX11 blacklists, update
BLACKLIST_GFX11 as needed.

Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I84bd91ba20a5d3df27478fb4c97afa12f8a3e76a
2023-02-23 09:11:27 -05:00
Graham Sider 179ddc4870 kfdtest: Update GFX11 blacklists
Remove BLACKLIST_GFX10_NV2X from GFX11 blacklists, update
BLACKLIST_GFX11 as needed.

Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I84bd91ba20a5d3df27478fb4c97afa12f8a3e76a


[ROCm/ROCR-Runtime commit: 60831e86b2]
2023-02-23 09:11:27 -05:00
Hao Zhou 1ebe5b4722 Merge amd-staging into amd-master 20230223
Signed-off-by: Hao Zhou <Hao.Zhou@amd.com>
Change-Id: If82fc6ac361e6ef5a68ae9dfe333dd8685a02fb5


[ROCm/rocm_smi_lib commit: d93a307f4b]
2023-02-23 18:13:38 +08:00
Hao Zhou d93a307f4b Merge amd-staging into amd-master 20230223
Signed-off-by: Hao Zhou <Hao.Zhou@amd.com>
Change-Id: If82fc6ac361e6ef5a68ae9dfe333dd8685a02fb5
2023-02-23 18:13:38 +08:00
Sarbojit Sarkar cd6c8fc2fb SWDEV-293749 - Fix for operator mixup
Change-Id: I61d1fea5a6ed2176dd92050c6d96cee1af3a39fb


[ROCm/clr commit: d0920c5959]
2023-02-23 04:45:23 -05:00
Sarbojit Sarkar d0920c5959 SWDEV-293749 - Fix for operator mixup
Change-Id: I61d1fea5a6ed2176dd92050c6d96cee1af3a39fb
2023-02-23 04:45:23 -05:00
Anusha GodavarthySurya d650f7a50c SWDEV-345571 - Added support for half/half2 data types for warp shuffle functions
Change-Id: I276a55129d6527b73292d75eb18403acd2623c8a


[ROCm/clr commit: 32a9249d3e]
2023-02-22 22:40:01 -05:00
Anusha GodavarthySurya 32a9249d3e SWDEV-345571 - Added support for half/half2 data types for warp shuffle functions
Change-Id: I276a55129d6527b73292d75eb18403acd2623c8a
2023-02-22 22:40:01 -05:00
David Yat Sin 37b5b421b3 Revert "Enforce uncached memory on AllocatePCIeRW request"
This reverts commit ed0a1be2c3.

Change-Id: I5a7fe9e99685f589f95dd89eacf04d44e5587f2f
2023-02-22 21:55:48 -05:00
David Yat Sin a2e4ba149c Revert "Enforce uncached memory on AllocatePCIeRW request"
This reverts commit 36da397f96.

Change-Id: I5a7fe9e99685f589f95dd89eacf04d44e5587f2f


[ROCm/ROCR-Runtime commit: 37b5b421b3]
2023-02-22 21:55:48 -05:00
coleramos425 106dcd2eaa Correct filtering in standalone roofline
Signed-off-by: coleramos425 <colramos@amd.com>


[ROCm/rocprofiler-compute commit: 0705db024c]
2023-02-22 14:37:20 -06:00
coleramos425 0705db024c Correct filtering in standalone roofline
Signed-off-by: coleramos425 <colramos@amd.com>
2023-02-22 14:37:20 -06:00
Cole Ramos d6bc463136 Update citation
Signed-off-by: Cole Ramos <colramos@amd.com>

[ROCm/rocprofiler-compute commit: 6c54bdad3b]
2023-02-22 11:10:55 -06:00
Cole Ramos 6c54bdad3b Update citation
Signed-off-by: Cole Ramos <colramos@amd.com>
2023-02-22 11:10:55 -06:00
David Yat Sin cc48dfdbff Use mwaitx when busy-waiting signals
Use mwaitx instructions when busy waiting for signals to reduce CPU
energy usage.
This can be disabled by setting HSA_ENABLE_MWAITX=0

Change-Id: Ic207895a491b2bf6dacba47ef0921df3faad5b5a
2023-02-22 16:55:43 +00:00
David Yat Sin d022746cb7 Use mwaitx when busy-waiting signals
Use mwaitx instructions when busy waiting for signals to reduce CPU
energy usage.
This can be disabled by setting HSA_ENABLE_MWAITX=0

Change-Id: Ic207895a491b2bf6dacba47ef0921df3faad5b5a


[ROCm/ROCR-Runtime commit: cc48dfdbff]
2023-02-22 16:55:43 +00:00
David Yat Sin 0ed1568afc Add function for parse CPUID information
Used to detect whether mwaitx instruction is supported

Change-Id: I66fe906325aa523c8815133cf782df3a17a7edab
2023-02-22 16:55:42 +00:00
David Yat Sin 72e7fe7aec Add function for parse CPUID information
Used to detect whether mwaitx instruction is supported

Change-Id: I66fe906325aa523c8815133cf782df3a17a7edab


[ROCm/ROCR-Runtime commit: 0ed1568afc]
2023-02-22 16:55:42 +00:00
Cole Ramos 6a06b2bf05 Update README
Signed-off-by: Cole Ramos <colramos@amd.com>

[ROCm/rocprofiler-compute commit: 81460ea7db]
2023-02-22 09:10:44 -06:00
Cole Ramos 81460ea7db Update README
Signed-off-by: Cole Ramos <colramos@amd.com>
2023-02-22 09:10:44 -06:00
Cole Ramos 977885fa4d Update VERSION
Signed-off-by: Cole Ramos <colramos@amd.com>

[ROCm/rocprofiler-compute commit: e73026fe65]
2023-02-22 09:08:37 -06:00
Cole Ramos e73026fe65 Update VERSION
Signed-off-by: Cole Ramos <colramos@amd.com>
2023-02-22 09:08:37 -06:00
Saleel Kudchadker 9fdbbe53dc SWDEV-364604 - Add support for hipEventDisableSystemFence
Change-Id: I1a6451c873fb22729ac61e4e80f8531251e990f0


[ROCm/clr commit: 7fc5ae2226]
2023-02-21 23:11:40 -05:00
Saleel Kudchadker 7fc5ae2226 SWDEV-364604 - Add support for hipEventDisableSystemFence
Change-Id: I1a6451c873fb22729ac61e4e80f8531251e990f0
2023-02-21 23:11:40 -05:00
pghafari 24bf83f808 SWDEV-366520 - DLL_PROCESS_DETACH update
making sure the current thread is not null before calling ihipDestroyDevice

Change-Id: Ib75e3f9ee1c4e0065986e3fed1065907220f4379


[ROCm/clr commit: c221406681]
2023-02-21 23:02:24 -05:00
pghafari c221406681 SWDEV-366520 - DLL_PROCESS_DETACH update
making sure the current thread is not null before calling ihipDestroyDevice

Change-Id: Ib75e3f9ee1c4e0065986e3fed1065907220f4379
2023-02-21 23:02:24 -05:00
Jaydeep Patel ac593a9dca SWDEV-383397 - Set large bar false for windows.
Change-Id: Iaef70e477afab7ca8694aa55d1f704e769e24ed9


[ROCm/clr commit: f364ab7680]
2023-02-21 22:18:58 -05:00
Jaydeep Patel f364ab7680 SWDEV-383397 - Set large bar false for windows.
Change-Id: Iaef70e477afab7ca8694aa55d1f704e769e24ed9
2023-02-21 22:18:58 -05:00
Jaydeep Patel 013905b593 SWDEV-319526 - SE ID size incresed.
Change-Id: Ia7df94d172102b321f12c8caa821e3a7e4447a21


[ROCm/clr commit: 0ad47eb3e5]
2023-02-21 22:13:57 -05:00
Jaydeep Patel 0ad47eb3e5 SWDEV-319526 - SE ID size incresed.
Change-Id: Ia7df94d172102b321f12c8caa821e3a7e4447a21
2023-02-21 22:13:57 -05:00
Saleel Kudchadker 858e311f34 SWDEV-364604 - Add ROCclr support for hipEventDisableSystemFence
Change-Id: I6127b432a8759359359a1890fda85bc401be6a56


[ROCm/clr commit: 3e603d986a]
2023-02-21 19:07:35 -05:00
Saleel Kudchadker 3e603d986a SWDEV-364604 - Add ROCclr support for hipEventDisableSystemFence
Change-Id: I6127b432a8759359359a1890fda85bc401be6a56
2023-02-21 19:07:35 -05:00
Jaydeep Patel ae81589d28 SWDEV-382664 - Remove WaveLimiter init
The wavelimiter init was unintentionally added in one change.
It wasn't supposed to perform any logic, since LC doesn't support it and
the number of waves can be overwritten only with an environment variable.

Change-Id: I447bd1ad685800f874b6a6fb7409dc67e43640ff


[ROCm/clr commit: 4830dd168c]
2023-02-21 18:12:08 -05:00
Jaydeep Patel 4830dd168c SWDEV-382664 - Remove WaveLimiter init
The wavelimiter init was unintentionally added in one change.
It wasn't supposed to perform any logic, since LC doesn't support it and
the number of waves can be overwritten only with an environment variable.

Change-Id: I447bd1ad685800f874b6a6fb7409dc67e43640ff
2023-02-21 18:12:08 -05:00
Ioannis Assiouras 2633e62801 SWDEV-381402 - Derive hip::Stream from amd::HostQueue
Change-Id: I6c1aca5eb350c32d974ae4ffcc725705355956d8


[ROCm/clr commit: e3633dc8f4]
2023-02-21 18:12:03 -05:00
Ioannis Assiouras e3633dc8f4 SWDEV-381402 - Derive hip::Stream from amd::HostQueue
Change-Id: I6c1aca5eb350c32d974ae4ffcc725705355956d8
2023-02-21 18:12:03 -05:00
Cole Ramos f8ca7a7023 Merge pull request #92 from AMDResearch/dev
Release v1.0.7

[ROCm/rocprofiler-compute commit: f251c55f62]
2023-02-21 16:41:46 -06:00
Cole Ramos f251c55f62 Merge pull request #92 from AMDResearch/dev
Release v1.0.7
2023-02-21 16:41:46 -06:00
Ioannis Assiouras b91c27aef1 SWDEV-384362 - Added explicit cast from __fp16 to _Float16 in rcph
Change-Id: Ibbb11e928a80126ed7a8c7cd8bddebc74c38986d


[ROCm/clr commit: aaf1c19609]
2023-02-21 17:32:59 -05:00
Ioannis Assiouras aaf1c19609 SWDEV-384362 - Added explicit cast from __fp16 to _Float16 in rcph
Change-Id: Ibbb11e928a80126ed7a8c7cd8bddebc74c38986d
2023-02-21 17:32:59 -05:00
Xie,AlexBin 4d6ec86edf SWDEV-378367 - Observed performance drop for Geekbench5
Change-Id: I0beabc6e3bec095574c8168fcf52af1e94105792


[ROCm/clr commit: 97243ff755]
2023-02-21 17:18:16 -05:00
Xie,AlexBin 97243ff755 SWDEV-378367 - Observed performance drop for Geekbench5
Change-Id: I0beabc6e3bec095574c8168fcf52af1e94105792
2023-02-21 17:18:16 -05:00
Cole Ramos 545aa535d3 Update CHANGES
[ROCm/rocprofiler-compute commit: 04e3f5277f]
2023-02-21 15:53:35 -06:00
Cole Ramos 04e3f5277f Update CHANGES 2023-02-21 15:53:35 -06:00
Cole Ramos d9f021e52f Add v1.07 issues to changelog
[ROCm/rocprofiler-compute commit: 8108a13b7f]
2023-02-21 15:46:35 -06:00
Cole Ramos 8108a13b7f Add v1.07 issues to changelog 2023-02-21 15:46:35 -06:00
Karl W. Schulz 45e0659ed3 Updating changelog with example v1.0.7 change
[ROCm/rocprofiler-compute commit: 4b1a41bbc0]
2023-02-21 14:44:03 -06:00
Karl W. Schulz 4b1a41bbc0 Updating changelog with example v1.0.7 change 2023-02-21 14:44:03 -06:00
Yaxun (Sam) Liu e660f913b5 SWDEV-380145 - [GFX][CQE] Mathlibs and Blender build failed in both RT and CPL Jobs
Remove --hip-device-lib-path from hip-config.cmake
and let clang determine device lib path

Change-Id: Ice009875624692bc2895020c9577b94ddbd6840f


[ROCm/clr commit: cbb3937196]
2023-02-21 15:33:22 -05:00
Yaxun (Sam) Liu cbb3937196 SWDEV-380145 - [GFX][CQE] Mathlibs and Blender build failed in both RT and CPL Jobs
Remove --hip-device-lib-path from hip-config.cmake
and let clang determine device lib path

Change-Id: Ice009875624692bc2895020c9577b94ddbd6840f
2023-02-21 15:33:22 -05:00