76333 Cometimentos
Autor(a) SHA1 Mensagem Data
Wenkai Du 661b1351a3 Limit max channels for ring graph on single node Rome (#347)
* Limit max channels for ring graph on single node Rome
* Partially revert "Use non-temporal access for streaming data (#341)"

[ROCm/rccl commit: a79f74082e]
2021-04-14 10:14:54 -07:00
Wenkai Du a79f74082e Limit max channels for ring graph on single node Rome (#347)
* Limit max channels for ring graph on single node Rome
* Partially revert "Use non-temporal access for streaming data (#341)"
2021-04-14 10:14:54 -07:00
sumanthtg 455fcfd330 SWDEV-278967 - Changes to handle dtest failures on cuda runtime version 11.2
Change-Id: I79c87d40b77b4e4beb7e3d0a9ea396ad918a81e8


[ROCm/clr commit: ee8b053a3e]
2021-04-14 12:38:46 -04:00
sumanthtg ee8b053a3e SWDEV-278967 - Changes to handle dtest failures on cuda runtime version 11.2
Change-Id: I79c87d40b77b4e4beb7e3d0a9ea396ad918a81e8
2021-04-14 12:38:46 -04:00
sumanthtg 578ce42bc5 SWDEV-278967 - Changes to handle dtest failures on cuda runtime version 11.2
Change-Id: I79c87d40b77b4e4beb7e3d0a9ea396ad918a81e8


[ROCm/hip commit: 5476c9877b]
2021-04-14 12:38:46 -04:00
sumanthtg 5476c9877b SWDEV-278967 - Changes to handle dtest failures on cuda runtime version 11.2
Change-Id: I79c87d40b77b4e4beb7e3d0a9ea396ad918a81e8
2021-04-14 12:38:46 -04:00
Vladislav Sytchenko c585ae92a7 SWDEV-280473 - Support HSAIL shared library build
This change makes HSAIL usage similar to that of Comgr. By default, the
runtime will statically link against it, however if HSAIL_DYN_DLL is
defined, then the runtime will try to dynamically load HSAIL.

Currently stick to statically linking to HSAIL. In a feature patch the
dynamic loading behaviour will be enabled.

Change-Id: I6a78a4375975cf847f236b200404c8cf941d012b


[ROCm/clr commit: c7b50bb890]
2021-04-14 12:25:54 -04:00
Vladislav Sytchenko c7b50bb890 SWDEV-280473 - Support HSAIL shared library build
This change makes HSAIL usage similar to that of Comgr. By default, the
runtime will statically link against it, however if HSAIL_DYN_DLL is
defined, then the runtime will try to dynamically load HSAIL.

Currently stick to statically linking to HSAIL. In a feature patch the
dynamic loading behaviour will be enabled.

Change-Id: I6a78a4375975cf847f236b200404c8cf941d012b
2021-04-14 12:25:54 -04:00
Wenkai DueEiden Yoshida 0f4d497edc Add gfx90a target (#344)
* Add gfx90a target

* Support gfx90a topology

Co-authored-by: Eiden Yoshida <eiden.yoshida@amd.com>

[ROCm/rccl commit: 1fe031402a]
2021-04-14 09:29:00 -06:00
Wenkai DueEiden Yoshida 1fe031402a Add gfx90a target (#344)
* Add gfx90a target

* Support gfx90a topology

Co-authored-by: Eiden Yoshida <eiden.yoshida@amd.com>
2021-04-14 09:29:00 -06:00
Rachida Kebichi a5571976e5 SWDEV-281008 replace hcc_detail by amd_detail
Change-Id: I180b18f9e1fae40c923d6210901f06cba14e8f13


[ROCm/roctracer commit: 6267df36e4]
2021-04-14 11:18:38 -04:00
Rachida Kebichi 6267df36e4 SWDEV-281008 replace hcc_detail by amd_detail
Change-Id: I180b18f9e1fae40c923d6210901f06cba14e8f13
2021-04-14 11:18:38 -04:00
Bill(Shuzhou) Liu 6218ae6fa9 Unit tests for coarse grain utilization counters
The unit tests for GFX and Memory activity counters.

Change-Id: I968dabc9ef6de9d335d7f751b290fb713b51a79c


[ROCm/rocm_smi_lib commit: 6340176b99]
2021-04-14 10:53:55 -04:00
Bill(Shuzhou) Liu 6340176b99 Unit tests for coarse grain utilization counters
The unit tests for GFX and Memory activity counters.

Change-Id: I968dabc9ef6de9d335d7f751b290fb713b51a79c
2021-04-14 10:53:55 -04:00
Bill(Shuzhou) Liu 62bef2b6c4 Unit tests for coarse grain utilization counters
The unit tests for GFX and Memory activity counters.

Change-Id: I968dabc9ef6de9d335d7f751b290fb713b51a79c


[ROCm/amdsmi commit: 6340176b99]
2021-04-14 10:53:55 -04:00
Bill(Shuzhou) Liu c3c7019436 Add energy accumulator counter
The energy accumulator counter tracks all energy consumed.

Change-Id: I5b25f817b7802d81c477361447f0ecd7ec02fc61


[ROCm/rocm_smi_lib commit: 8eec0a7d36]
2021-04-14 10:43:01 -04:00
Bill(Shuzhou) Liu 8eec0a7d36 Add energy accumulator counter
The energy accumulator counter tracks all energy consumed.

Change-Id: I5b25f817b7802d81c477361447f0ecd7ec02fc61
2021-04-14 10:43:01 -04:00
Bill(Shuzhou) Liu 919364871d Add energy accumulator counter
The energy accumulator counter tracks all energy consumed.

Change-Id: I5b25f817b7802d81c477361447f0ecd7ec02fc61


[ROCm/amdsmi commit: 8eec0a7d36]
2021-04-14 10:43:01 -04:00
Bill(Shuzhou) Liu dfa31bb5c4 Add coarse grain utilization counter
The coarse grain utilization counter includes GFX and Memory activity.

Change-Id: I5d09976792d3f4a1c1081651fa24ff857016d4c0


[ROCm/rocm_smi_lib commit: 9bfb9ac297]
2021-04-14 10:40:19 -04:00
Bill(Shuzhou) Liu 9bfb9ac297 Add coarse grain utilization counter
The coarse grain utilization counter includes GFX and Memory activity.

Change-Id: I5d09976792d3f4a1c1081651fa24ff857016d4c0
2021-04-14 10:40:19 -04:00
Bill(Shuzhou) Liu 38ddf00856 Add coarse grain utilization counter
The coarse grain utilization counter includes GFX and Memory activity.

Change-Id: I5d09976792d3f4a1c1081651fa24ff857016d4c0


[ROCm/amdsmi commit: 9bfb9ac297]
2021-04-14 10:40:19 -04:00
Chauncey Hui ba2d6da261 SWDEV-2 - Change OpenCL version number from 3291 to 3292
[ROCm/clr commit: 759cf19600]
2021-04-14 03:00:05 -04:00
Chauncey Hui 759cf19600 SWDEV-2 - Change OpenCL version number from 3291 to 3292 2021-04-14 03:00:05 -04:00
Aaron Enye Shi eb7035a7c1 SWDEV-261391 - Add Cache Coherency Tests for gfx90a
Add a test for fine-grained device to device coherency.
Add a test for fine-grained host to device coherency.

Change-Id: I62482cae917fa19feaa17adb53f3084527ad8fda


[ROCm/clr commit: 49696c568f]
2021-04-13 16:13:56 -04:00
Aaron Enye Shi 49696c568f SWDEV-261391 - Add Cache Coherency Tests for gfx90a
Add a test for fine-grained device to device coherency.
Add a test for fine-grained host to device coherency.

Change-Id: I62482cae917fa19feaa17adb53f3084527ad8fda
2021-04-13 16:13:56 -04:00
Aaron Enye Shi c7cc093f1a SWDEV-261391 - Add Cache Coherency Tests for gfx90a
Add a test for fine-grained device to device coherency.
Add a test for fine-grained host to device coherency.

Change-Id: I62482cae917fa19feaa17adb53f3084527ad8fda


[ROCm/hip commit: eed3fe6da9]
2021-04-13 16:13:56 -04:00
Aaron Enye Shi eed3fe6da9 SWDEV-261391 - Add Cache Coherency Tests for gfx90a
Add a test for fine-grained device to device coherency.
Add a test for fine-grained host to device coherency.

Change-Id: I62482cae917fa19feaa17adb53f3084527ad8fda
2021-04-13 16:13:56 -04:00
Julia Jiang 7bb189c4c5 SWDEV-268186 - OCL ReBar optimization
Change-Id: I69d8bce8d48a5b6f94a05272c83ee91fbec1688c


[ROCm/clr commit: aef4ab1fc8]
2021-04-13 15:08:32 -04:00
Julia Jiang aef4ab1fc8 SWDEV-268186 - OCL ReBar optimization
Change-Id: I69d8bce8d48a5b6f94a05272c83ee91fbec1688c
2021-04-13 15:08:32 -04:00
Vladislav Sytchenko 8edf4d9cd3 SWDEV-280473 - Fix Windows build
Change-Id: Id1acb5ad07a6717c5866065776f9f2b3bf85be8e


[ROCm/clr commit: 2ae1b17b7c]
2021-04-13 11:38:17 -04:00
Vladislav Sytchenko 2ae1b17b7c SWDEV-280473 - Fix Windows build
Change-Id: Id1acb5ad07a6717c5866065776f9f2b3bf85be8e
2021-04-13 11:38:17 -04:00
Mike Li d077606e22 Get GPU cache information from KFD
Signed-off-by: Mike Li <Tianxinmike.Li@amd.com>
Change-Id: I8dc8c97ae81c3747b7cd88cf2cdb7a9e4694a88d
2021-04-13 10:29:34 -04:00
Mike Li 3258d72d3b Get GPU cache information from KFD
Signed-off-by: Mike Li <Tianxinmike.Li@amd.com>
Change-Id: I8dc8c97ae81c3747b7cd88cf2cdb7a9e4694a88d


[ROCm/ROCR-Runtime commit: d077606e22]
2021-04-13 10:29:34 -04:00
Kent Russell 17bdc065a1 rocm_smi.py: Don't try to reset non-AMD GPUs
This won't work for obvious reasons, so exit with an error instead of
trying to access a file that doesn't exist and segfaulting

Change-Id: Id1230922fa6e9a19e9394280faad88a43c7d2e34


[ROCm/rocm_smi_lib commit: c7c2ac5559]
2021-04-13 08:00:17 -04:00
Kent Russell c7c2ac5559 rocm_smi.py: Don't try to reset non-AMD GPUs
This won't work for obvious reasons, so exit with an error instead of
trying to access a file that doesn't exist and segfaulting

Change-Id: Id1230922fa6e9a19e9394280faad88a43c7d2e34
2021-04-13 08:00:17 -04:00
Kent Russell ffbe481241 rocm_smi.py: Don't try to reset non-AMD GPUs
This won't work for obvious reasons, so exit with an error instead of
trying to access a file that doesn't exist and segfaulting

Change-Id: Id1230922fa6e9a19e9394280faad88a43c7d2e34


[ROCm/amdsmi commit: c7c2ac5559]
2021-04-13 08:00:17 -04:00
Chauncey Hui c9fec26190 SWDEV-2 - Change OpenCL version number from 3290 to 3291
[ROCm/clr commit: 4711005dfd]
2021-04-13 03:00:08 -04:00
Chauncey Hui 4711005dfd SWDEV-2 - Change OpenCL version number from 3290 to 3291 2021-04-13 03:00:08 -04:00
Sylvain Jeaugey 20da390b96 2.9.6-1
Add support for CUDA graphs.
Fuse BCM Gen4 switches to avoid suboptimal performance on some platforms. Issue #439.
Fix bootstrap issue caused by connection reordering.
Fix CPU locking block.
Improve CollNet algorithm.
Improve performance on DGX A100 for communicators with only one GPU per node.


[ROCm/rccl commit: a46ea10583]
2021-04-12 16:00:46 -07:00
Sylvain Jeaugey a46ea10583 2.9.6-1
Add support for CUDA graphs.
Fuse BCM Gen4 switches to avoid suboptimal performance on some platforms. Issue #439.
Fix bootstrap issue caused by connection reordering.
Fix CPU locking block.
Improve CollNet algorithm.
Improve performance on DGX A100 for communicators with only one GPU per node.
2021-04-12 16:00:46 -07:00
Tao Sang 73df8d8dd9 SWDEV-273138 - Rearrange device attributes
Change-Id: Iab5902c9a2702456fbe437fdefc40541c48abb74


[ROCm/clr commit: 2c0b63059b]
2021-04-12 17:05:27 -04:00
Tao Sang 2c0b63059b SWDEV-273138 - Rearrange device attributes
Change-Id: Iab5902c9a2702456fbe437fdefc40541c48abb74
2021-04-12 17:05:27 -04:00
Tao Sang 8e4eab2f4d SWDEV-273138 - Rearrange device attributes
Change-Id: Iab5902c9a2702456fbe437fdefc40541c48abb74


[ROCm/hip commit: ca50ac83c7]
2021-04-12 17:05:27 -04:00
Tao Sang ca50ac83c7 SWDEV-273138 - Rearrange device attributes
Change-Id: Iab5902c9a2702456fbe437fdefc40541c48abb74
2021-04-12 17:05:27 -04:00
Vladislav Sytchenko 7b3014ec69 SWDEV-280473 - Remove HSAIL support from the ROCm backend
In adition to removing the HSAIL logic from the ROCm backend, guard all
of the HSAIL includes in the common layer behind the WITH_COMPILER_LIB
define. This is to avoid including HSAIL headers when building with
no support for it.

In common logic replace the use of the aclType enum with the new
Program::file_type_t enum. This is essentially a local copy of the HSAIL
enum to avoid including any HSAIL headers.

Change-Id: Ica0651d1b29dfccc255cc584eb82a5cb35e1b520


[ROCm/clr commit: cbeb372e46]
2021-04-12 14:55:06 -04:00
Vladislav Sytchenko cbeb372e46 SWDEV-280473 - Remove HSAIL support from the ROCm backend
In adition to removing the HSAIL logic from the ROCm backend, guard all
of the HSAIL includes in the common layer behind the WITH_COMPILER_LIB
define. This is to avoid including HSAIL headers when building with
no support for it.

In common logic replace the use of the aclType enum with the new
Program::file_type_t enum. This is essentially a local copy of the HSAIL
enum to avoid including any HSAIL headers.

Change-Id: Ica0651d1b29dfccc255cc584eb82a5cb35e1b520
2021-04-12 14:55:06 -04:00
Wenkai Du 27f33208e3 Remove link to NUMA lib as it is no longer needed (#346)
[ROCm/rccl commit: 3f18540f50]
2021-04-12 09:53:17 -07:00
Wenkai Du 3f18540f50 Remove link to NUMA lib as it is no longer needed (#346) 2021-04-12 09:53:17 -07:00
Kent Russell cd21f8fdb3 CMakeLists: Add python3 to required packages
Signed-off-by: Kent Russell <kent.russell@amd.com>
Change-Id: I434b24d12e92d2f6a6928b7450e74c3898303a44


[ROCm/rocm_smi_lib commit: b016a8269a]
2021-04-12 11:33:39 -04:00
Kent Russell b016a8269a CMakeLists: Add python3 to required packages
Signed-off-by: Kent Russell <kent.russell@amd.com>
Change-Id: I434b24d12e92d2f6a6928b7450e74c3898303a44
2021-04-12 11:33:39 -04:00