76333 Коммитов
Автор SHA1 Сообщение Дата
anusha GodavarthySurya ef1ec6ffde SWDEV-240806 - hipGraph performance create new graph commands for every launch
Change-Id: Ifd4a373d6a76118ae0946238b29accfacbe32937
2021-11-19 00:09:47 -05:00
Stanley Tsang 5b442ffdc1 Set ROCM_PATH CMake variable in install script (#470)
* Fixing cmake_install_prefix search to include /opt/rocm-xxxx

* Removing all hard references to /opt/rocm with ROCM_PATH

* Setting ROCM_PATH CMake variable in install script

[ROCm/rccl commit: 7b8b54955b]
2021-11-18 14:44:19 -07:00
Stanley Tsang 7b8b54955b Set ROCM_PATH CMake variable in install script (#470)
* Fixing cmake_install_prefix search to include /opt/rocm-xxxx

* Removing all hard references to /opt/rocm with ROCM_PATH

* Setting ROCM_PATH CMake variable in install script
2021-11-18 14:44:19 -07:00
Jeremy Newton 7649cd862e Fix packaging of license file
CPACK doesn't have proper logic for installing the license as described
by CPACK_RESOURCE_FILE_LICENSE.

For Debian packaging, the license is expected to be installed as:

/usr/share/doc/PACKAGENAME/copyright

To do this, I've added a bit of logic for CPACK to copy this into the
package using CPACK_INSTALL_COMMANDS to prep the directory, and
CPACK_INSTALLED_DIRECTORIES to add it to the package. This applies to
both RPM and DEB, so I've added some logic to the spec file to exclude
this file (note that CPACK_RPM_EXCLUDE_FROM_AUTO_FILELIST_ADDITION does
not work for files installed with CPACK_INSTALLED_DIRECTORIES).

For RPM install, I've just added a small bit of logic to the spec file
to handle it. The file needs to be copied into the spec working
directory, then a macro is used to handle the rest. Note the license
macro does not work on EL6, but I don't think we want to support this.

Change-Id: I06ce63d300419893cb8274bc504a15633e304d91
Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com>
2021-11-18 16:41:48 -05:00
Jeremy Newton b93ee2fe7a Fix packaging of license file
CPACK doesn't have proper logic for installing the license as described
by CPACK_RESOURCE_FILE_LICENSE.

For Debian packaging, the license is expected to be installed as:

/usr/share/doc/PACKAGENAME/copyright

To do this, I've added a bit of logic for CPACK to copy this into the
package using CPACK_INSTALL_COMMANDS to prep the directory, and
CPACK_INSTALLED_DIRECTORIES to add it to the package. This applies to
both RPM and DEB, so I've added some logic to the spec file to exclude
this file (note that CPACK_RPM_EXCLUDE_FROM_AUTO_FILELIST_ADDITION does
not work for files installed with CPACK_INSTALLED_DIRECTORIES).

For RPM install, I've just added a small bit of logic to the spec file
to handle it. The file needs to be copied into the spec working
directory, then a macro is used to handle the rest. Note the license
macro does not work on EL6, but I don't think we want to support this.

Change-Id: I06ce63d300419893cb8274bc504a15633e304d91
Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com>


[ROCm/ROCR-Runtime commit: 7649cd862e]
2021-11-18 16:41:48 -05:00
German Andryeyev c116411e00 SWDEV-294669 - Avoid queue drain
Use slot wait logic for direct dispatch

Change-Id: I431ba1418eb4aa066b9881934f4055b3d338ce3a


[ROCm/clr commit: 8e4101b4fd]
2021-11-18 13:06:12 -05:00
German Andryeyev 8e4101b4fd SWDEV-294669 - Avoid queue drain
Use slot wait logic for direct dispatch

Change-Id: I431ba1418eb4aa066b9881934f4055b3d338ce3a
2021-11-18 13:06:12 -05:00
Chauncey Hui a3e56d08b7 SWDEV-2 - Change OpenCL version number from 3391 to 3392
[ROCm/clr commit: 874513375f]
2021-11-18 03:00:10 -05:00
Chauncey Hui 874513375f SWDEV-2 - Change OpenCL version number from 3391 to 3392 2021-11-18 03:00:10 -05:00
Jeremy Newton 529c96c08b Fix to previous commit
I used the binary directory instead of the source directory to specify
the spec.in path, which passed local testing since these directories
are in the same location. This is not guarenteed to be true.

Change-Id: I1b49ca8453b9c074a947104c26fb39667d728a8e
Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com>
2021-11-17 17:47:53 -05:00
Jeremy Newton c0397d4a44 Fix to previous commit
I used the binary directory instead of the source directory to specify
the spec.in path, which passed local testing since these directories
are in the same location. This is not guarenteed to be true.

Change-Id: I1b49ca8453b9c074a947104c26fb39667d728a8e
Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com>


[ROCm/ROCR-Runtime commit: 529c96c08b]
2021-11-17 17:47:53 -05:00
Sreekant Somasekharan 70be1fab11 Modify bool variable to true in if condition of src=dst
Change-Id: Ie2024b3a6ad68e48384bb3472fe8785bcd643665


[ROCm/rocm_smi_lib commit: 3f27dcc1ac]
2021-11-17 12:53:40 -05:00
Sreekant Somasekharan 3f27dcc1ac Modify bool variable to true in if condition of src=dst
Change-Id: Ie2024b3a6ad68e48384bb3472fe8785bcd643665
2021-11-17 12:53:40 -05:00
Sreekant Somasekharan d48617a959 Modify bool variable to true in if condition of src=dst
Change-Id: Ie2024b3a6ad68e48384bb3472fe8785bcd643665


[ROCm/amdsmi commit: 3f27dcc1ac]
2021-11-17 12:53:40 -05:00
Jeremy Newton 3c0e4fee0f Implement RPM recommends for libdrm-amdgpu
CPack does not support recommends for RPM generation, so I've generated
a template RPM SPEC files in order to make modifications to allow for
support of recommends.

The spec.in file was generated using the cpack option
"CPACK_RPM_GENERATE_USER_BINARY_SPECFILE_TEMPLATE" and was modified very
sparingly to avoid any maintanance burden, e.g. can be easily
regenerated. The CPACK_RPM_USER_BINARY_SPECFILE is then used to specify
the customized template file, instead of using the cmake's template.

From what I understand, the point of these two options is to allow
developers to tailor the specfile to their desire, since rpm spec files
are much more advanced then the equilivent debian file.

Change-Id: I80c69be58a3c57729ed997fd2ce01f5d16b9e9b9
Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com>
2021-11-17 11:57:55 -05:00
Jeremy Newton 04144c6d0d Implement RPM recommends for libdrm-amdgpu
CPack does not support recommends for RPM generation, so I've generated
a template RPM SPEC files in order to make modifications to allow for
support of recommends.

The spec.in file was generated using the cpack option
"CPACK_RPM_GENERATE_USER_BINARY_SPECFILE_TEMPLATE" and was modified very
sparingly to avoid any maintanance burden, e.g. can be easily
regenerated. The CPACK_RPM_USER_BINARY_SPECFILE is then used to specify
the customized template file, instead of using the cmake's template.

From what I understand, the point of these two options is to allow
developers to tailor the specfile to their desire, since rpm spec files
are much more advanced then the equilivent debian file.

Change-Id: I80c69be58a3c57729ed997fd2ce01f5d16b9e9b9
Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com>


[ROCm/ROCR-Runtime commit: 3c0e4fee0f]
2021-11-17 11:57:55 -05:00
pghafari e7df603b2a SWDEV-297142 - HIP-Vulkan - linux interop buffer
Change-Id: I0278e56bba632024c214beb9e1758587ccba0927


[ROCm/clr commit: e38a200bf7]
2021-11-17 06:06:58 -05:00
pghafari e38a200bf7 SWDEV-297142 - HIP-Vulkan - linux interop buffer
Change-Id: I0278e56bba632024c214beb9e1758587ccba0927
2021-11-17 06:06:58 -05:00
anusha GodavarthySurya 239cc6b437 SWDEV-240806 - Added support for host graph node
Change-Id: I8b3088d34d0b4e8080b55b5da80d821c5f29d8a4


[ROCm/clr commit: ce58dae704]
2021-11-17 01:10:18 -08:00
anusha GodavarthySurya ce58dae704 SWDEV-240806 - Added support for host graph node
Change-Id: I8b3088d34d0b4e8080b55b5da80d821c5f29d8a4
2021-11-17 01:10:18 -08:00
anusha GodavarthySurya f675e61387 SWDEV-308983 - Fix hipGraphKernelNodeSetParams
Change-Id: I8a57a4e33048931955b7ba5ed567631703bd6c5d


[ROCm/clr commit: 9c8b6b65e0]
2021-11-17 01:02:50 -08:00
anusha GodavarthySurya 9c8b6b65e0 SWDEV-308983 - Fix hipGraphKernelNodeSetParams
Change-Id: I8a57a4e33048931955b7ba5ed567631703bd6c5d
2021-11-17 01:02:50 -08:00
anusha GodavarthySurya 71215f1231 SWDEV-240806 - hip stream capture handling for memory APIs
Change-Id: Ie29493cd09421175c3ce5a198f8c2c299739f997


[ROCm/clr commit: 3e2c2dfb71]
2021-11-17 00:15:13 -08:00
anusha GodavarthySurya 3e2c2dfb71 SWDEV-240806 - hip stream capture handling for memory APIs
Change-Id: Ie29493cd09421175c3ce5a198f8c2c299739f997
2021-11-17 00:15:13 -08:00
anusha GodavarthySurya abb6f92138 SWDEV-240806 - hipGraph performance create new graph commands for every launch
Change-Id: I7360b48055f67a91c23169ca532a19124d40e98b


[ROCm/clr commit: ec6b8d82b5]
2021-11-17 00:13:52 -08:00
anusha GodavarthySurya ec6b8d82b5 SWDEV-240806 - hipGraph performance create new graph commands for every launch
Change-Id: I7360b48055f67a91c23169ca532a19124d40e98b
2021-11-17 00:13:52 -08:00
Chauncey Hui 0e39d6f41a SWDEV-2 - Change OpenCL version number from 3390 to 3391
[ROCm/clr commit: 13ef90af61]
2021-11-17 03:00:15 -05:00
Chauncey Hui 13ef90af61 SWDEV-2 - Change OpenCL version number from 3390 to 3391 2021-11-17 03:00:15 -05:00
German Andryeyev 95c01935b8 SWDEV-305016 - Correct timeout logic
Timeout logic should be applied always even if the wait is active.

Change-Id: I2e5db7ac8a0f9a0355ad7b40e4227d76fb002aa0


[ROCm/clr commit: 9877fc9dbf]
2021-11-16 23:04:34 -05:00
German Andryeyev 9877fc9dbf SWDEV-305016 - Correct timeout logic
Timeout logic should be applied always even if the wait is active.

Change-Id: I2e5db7ac8a0f9a0355ad7b40e4227d76fb002aa0
2021-11-16 23:04:34 -05:00
Wenkai Du a14a57ea71 gtest: dynamically generate tests based on test machine's GPU count (#467)
* gtest: dynamically generate tests based on test machine's GPU count

* Adjust test element size and bfloat16 threshold for up to 16 GPUs

[ROCm/rccl commit: 03a830293c]
2021-11-16 10:28:26 -08:00
Wenkai Du 03a830293c gtest: dynamically generate tests based on test machine's GPU count (#467)
* gtest: dynamically generate tests based on test machine's GPU count

* Adjust test element size and bfloat16 threshold for up to 16 GPUs
2021-11-16 10:28:26 -08:00
Jeremy Newton 86c27a7af8 Use recommends for libdrm-amdgpu-amdgpu1
For the use of libdrm-amdgpu-amdgpu1 and libdrm-amdgpu, we should use
recommends, as we want these packages installed with a strong dependency
but avoid a strict dependency, since this is enhancement feature.

Using the newer libdrm, which is build for amdgpu-dkms, is ideal since
it will produce more correct marketing names, but should not be mandated
due to two reasons:
- A user may not want to install both libdrms on their system
- The system might not have the newer libdrm available

This patch only fixes the Ubuntu/debian package since recommends is not
properly implemented for the RPM generator for CPACK. For now,
"suggests" will have to do, since it's the closest option we have. I
will investigate if we can get around this issue.

Change-Id: I33a90c3ead235bbbe265238c026933688ea63fe3
Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com>
2021-11-16 11:08:40 -05:00
Jeremy Newton 5d8bcd03db Use recommends for libdrm-amdgpu-amdgpu1
For the use of libdrm-amdgpu-amdgpu1 and libdrm-amdgpu, we should use
recommends, as we want these packages installed with a strong dependency
but avoid a strict dependency, since this is enhancement feature.

Using the newer libdrm, which is build for amdgpu-dkms, is ideal since
it will produce more correct marketing names, but should not be mandated
due to two reasons:
- A user may not want to install both libdrms on their system
- The system might not have the newer libdrm available

This patch only fixes the Ubuntu/debian package since recommends is not
properly implemented for the RPM generator for CPACK. For now,
"suggests" will have to do, since it's the closest option we have. I
will investigate if we can get around this issue.

Change-Id: I33a90c3ead235bbbe265238c026933688ea63fe3
Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com>


[ROCm/ROCR-Runtime commit: 86c27a7af8]
2021-11-16 11:08:40 -05:00
Wenkai Du db075a1fab Add missing hipStreamDestroy at test exit
[ROCm/rccl-tests commit: 602b745ff4]
2021-11-16 07:50:18 -08:00
Wenkai Du 602b745ff4 Add missing hipStreamDestroy at test exit 2021-11-16 07:50:18 -08:00
Icarus Sparry 39a11d7dc4 Merge "Allow creation of just deb or rpm" into amd-staging 2021-11-15 23:03:57 -05:00
Icarus Sparry 0f272c9815 Merge "Allow creation of just deb or rpm" into amd-staging
[ROCm/rocprofiler commit: 39a11d7dc4]
2021-11-15 23:03:57 -05:00
Maneesh Gupta 634e465644 Bump HIP version to 5.0.0
[ROCm/hip commit: 3384fbdfe5]
2021-11-16 08:21:00 +05:30
Maneesh Gupta 3384fbdfe5 Bump HIP version to 5.0.0 2021-11-16 08:21:00 +05:30
Rahul Garg 417851d666 SWDEV-289558 - Generate hip-config.cmake on Windows
Change-Id: I00b553bcba596e882573b99cfb6894c310cf7c2a


[ROCm/clr commit: bcfac99cea]
2021-11-15 20:58:13 -05:00
Rahul Garg bcfac99cea SWDEV-289558 - Generate hip-config.cmake on Windows
Change-Id: I00b553bcba596e882573b99cfb6894c310cf7c2a
2021-11-15 20:58:13 -05:00
Sean Keely fc75731034 Add comments to GetPcieBlit.
Comments call out the specific operation being selected since the
ternary nest is a bit hard to read.

Change-Id: If033dbaa6cba132e96196ad3fc6d5572042041f4
2021-11-15 19:34:03 -06:00
Sean Keely e462118b6e Add comments to GetPcieBlit.
Comments call out the specific operation being selected since the
ternary nest is a bit hard to read.

Change-Id: If033dbaa6cba132e96196ad3fc6d5572042041f4


[ROCm/ROCR-Runtime commit: fc75731034]
2021-11-15 19:34:03 -06:00
Sean Keely b198016949 Fix leak in hsa_amd_interop_map_buffer.
Agent temp array could have leaked if one of the given agent
handles was invalid.

Change-Id: I9e638b3a4f6bb917a4e3209ad81a1253bb603365
2021-11-15 19:22:20 -06:00
Sean Keely 01c7c9856c Fix leak in hsa_amd_interop_map_buffer.
Agent temp array could have leaked if one of the given agent
handles was invalid.

Change-Id: I9e638b3a4f6bb917a4e3209ad81a1253bb603365


[ROCm/ROCR-Runtime commit: b198016949]
2021-11-15 19:22:20 -06:00
Sean Keely f48a786662 Correct order of argument check and default assignment in lock APIs.
Argument must be checked for nullptr before being dereferenced and
filled with the default return value.

Change-Id: I9ff366f066a5e18c78129bf59cc3ba00fca3ef18
2021-11-15 19:22:02 -06:00
Sean Keely 289cc7b6b4 Correct order of argument check and default assignment in lock APIs.
Argument must be checked for nullptr before being dereferenced and
filled with the default return value.

Change-Id: I9ff366f066a5e18c78129bf59cc3ba00fca3ef18


[ROCm/ROCR-Runtime commit: f48a786662]
2021-11-15 19:22:02 -06:00
Sean Keely 322588a60e Add missing return in ScopeGuard::operator=.
This omission did not cause problems earlier due to having not been
instanced.

Change-Id: I7a54f82e06c299902f3bf6b4d3737cc5e30961ad
2021-11-15 18:50:46 -06:00
Sean Keely a7dc6d7802 Add missing return in ScopeGuard::operator=.
This omission did not cause problems earlier due to having not been
instanced.

Change-Id: I7a54f82e06c299902f3bf6b4d3737cc5e30961ad


[ROCm/ROCR-Runtime commit: 322588a60e]
2021-11-15 18:50:46 -06:00