AI/rocm-systems
2
0 Форкнуть 0
Код Задачи Запросы на слияние Действия Пакеты Проекты Релизы Вики Активность
76 333 коммитов 3 Ветки 0 Теги
develop
Граф коммитов
76333 Коммитов
This Branch
This Branch
Все ветки
Автор SHA1 Сообщение Дата
Edgar Gabriel 7148c0aa7b add binary tree 
In addition, introduce the ability to have 2 trees at the same time.
Only for allreduce at the moment.


[ROCm/rccl commit: 65e2ae20e5]
 2022-09-13 20:52:32 +00:00
Edgar Gabriel 65e2ae20e5 add binary tree 
In addition, introduce the ability to have 2 trees at the same time.
Only for allreduce at the moment.
 2022-09-13 20:52:32 +00:00
Ranjith Ramakrishnan 4f08cdd1ea SWDEV-345870 - Correct install interface for new directory layout 
Install interface should provide /opt/rocm-ver/include as the include path
Path /opt/rocm-ver/rocm_smi/include should be used only as  part of backward compatibility support

Change-Id: Idc1f663069356c6b1fbd492f45ef4637fc90e4eb


[ROCm/rocm_smi_lib commit: 9a650b1378]
 2022-09-13 10:48:21 -07:00
Ranjith Ramakrishnan 9a650b1378 SWDEV-345870 - Correct install interface for new directory layout 
Install interface should provide /opt/rocm-ver/include as the include path
Path /opt/rocm-ver/rocm_smi/include should be used only as  part of backward compatibility support

Change-Id: Idc1f663069356c6b1fbd492f45ef4637fc90e4eb
 2022-09-13 10:48:21 -07:00
Ranjith Ramakrishnan 7438ed187c SWDEV-345870 - Correct install interface for new directory layout 
Install interface should provide /opt/rocm-ver/include as the include path
Path /opt/rocm-ver/rocm_smi/include should be used only as  part of backward compatibility support

Change-Id: Idc1f663069356c6b1fbd492f45ef4637fc90e4eb


[ROCm/amdsmi commit: 9a650b1378]
 2022-09-13 10:48:21 -07:00
Laurent Morichetti e8cb732660 Cleanup the include files 
Use #include "header" instead of #include <header> so that the header
files are found when the application #includes <roctracer/roctracer.h>
with -I /opt/rocm/include.

Change-Id: I24feac9a5030d3600aee98084340e246c3990db5


[ROCm/roctracer commit: 05ee3ff973]
 2022-09-13 08:23:40 -07:00
Laurent Morichetti 05ee3ff973 Cleanup the include files 
Use #include "header" instead of #include <header> so that the header
files are found when the application #includes <roctracer/roctracer.h>
with -I /opt/rocm/include.

Change-Id: I24feac9a5030d3600aee98084340e246c3990db5
 2022-09-13 08:23:40 -07:00
Laurent Morichetti 14c153601d SWDEV-355896 - Fix a data corruption error in post processing 
The post-processing script cannot handle HIP ops without a correlation
ID. The correlation ID is needed to connect the record to a HIP stream
and originating thread.

This issue was exposed by a change to the tracer API to report
asynchronous activities even if their originating synchronous API
activity (callback) is not enabled. This was a flow in the API.

Also fix an issue with the API filtering. Undefined API names should
not cause an exception, they should be ignored.

Change-Id: Iab2221af6180ade2b9c2eb10c256c3a73d872e9f


[ROCm/roctracer commit: 4856d33959]
 2022-09-13 08:21:13 -07:00
Laurent Morichetti 4856d33959 SWDEV-355896 - Fix a data corruption error in post processing 
The post-processing script cannot handle HIP ops without a correlation
ID. The correlation ID is needed to connect the record to a HIP stream
and originating thread.

This issue was exposed by a change to the tracer API to report
asynchronous activities even if their originating synchronous API
activity (callback) is not enabled. This was a flow in the API.

Also fix an issue with the API filtering. Undefined API names should
not cause an exception, they should be ignored.

Change-Id: Iab2221af6180ade2b9c2eb10c256c3a73d872e9f
 2022-09-13 08:21:13 -07:00
Jonathan R. Madsen 6227c25220 Fix setup-env + hsa/rocm/ompt serialization + testing + misc (#156) 
- Fix setup-env.sh
  - Closes #149 
- omnitrace exe color
- test-install.sh script
- if config variable is updated in config or env, include in generated
config
- metadata for hsa, rocm, and ompt
  - Closes #153 
  - Closes #154

[ROCm/rocprofiler-systems commit: 15e6e6d979]
 2022-09-13 08:11:48 -05:00
Jonathan R. Madsen 15e6e6d979 Fix setup-env + hsa/rocm/ompt serialization + testing + misc (#156) 
- Fix setup-env.sh
  - Closes #149 
- omnitrace exe color
- test-install.sh script
- if config variable is updated in config or env, include in generated
config
- metadata for hsa, rocm, and ompt
  - Closes #153 
  - Closes #154
 2022-09-13 08:11:48 -05:00
Jonathan R. Madsen 63a999481c Individual perfetto debug annotations (#157) 
- instead of args="<list-of-string>" in perfetto, each argument is added
individually, enabling matching pointer values, etc.

## Previous behavior

Previously, all function arguments were wrapped into a single string,
e.g.:

![Screen Shot 2022-09-12 at 11 52 11
PM](https://user-images.githubusercontent.com/6001865/189811896-3b13d5a4-e884-40b3-b6b4-f84f3e16f603.png)

## New Behavior

With the exception of the HIP API (whose args are provided as a single
string via `hipApiString`), all functions from MPI, RCCL, pthreads, etc.
have individual arguments in perfetto, e.g.:

![Screen Shot 2022-09-12 at 11 52 01
PM](https://user-images.githubusercontent.com/6001865/189812137-63afba72-170d-42b3-b3bc-ae74e32ceadf.png)

In the above, previously, this would have been:

|  |  |
| - | - |
| args | `pthread_rwlock_t*=0x1c1cc50` |

The key benefit enabled is the ability to find slices with same arg
values:

<img width="753" alt="Screen Shot 2022-09-12 at 11 59 18 PM"
src="https://user-images.githubusercontent.com/6001865/189812915-0342f841-e5ce-4f8e-8169-0cb52f3425b5.png">

Previously, the entire "args" field would have had to match, which
essentially never happened if the pointer was used in two different
functions with different function signatures

## Example

![Screen Shot 2022-09-12 at 11 49 18
PM](https://user-images.githubusercontent.com/6001865/189811621-5eed0008-f340-438e-938d-a140f836b583.png)

![Screen Shot 2022-09-12 at 11 49 45
PM](https://user-images.githubusercontent.com/6001865/189811584-a983ed1c-095f-4d54-a790-9ef3adbe4e08.png)

[ROCm/rocprofiler-systems commit: 4ed8f8f762]
 2022-09-13 02:05:03 -05:00
Jonathan R. Madsen 4ed8f8f762 Individual perfetto debug annotations (#157) 
- instead of args="<list-of-string>" in perfetto, each argument is added
individually, enabling matching pointer values, etc.

## Previous behavior

Previously, all function arguments were wrapped into a single string,
e.g.:

![Screen Shot 2022-09-12 at 11 52 11
PM](https://user-images.githubusercontent.com/6001865/189811896-3b13d5a4-e884-40b3-b6b4-f84f3e16f603.png)

## New Behavior

With the exception of the HIP API (whose args are provided as a single
string via `hipApiString`), all functions from MPI, RCCL, pthreads, etc.
have individual arguments in perfetto, e.g.:

![Screen Shot 2022-09-12 at 11 52 01
PM](https://user-images.githubusercontent.com/6001865/189812137-63afba72-170d-42b3-b3bc-ae74e32ceadf.png)

In the above, previously, this would have been:

|  |  |
| - | - |
| args | `pthread_rwlock_t*=0x1c1cc50` |

The key benefit enabled is the ability to find slices with same arg
values:

<img width="753" alt="Screen Shot 2022-09-12 at 11 59 18 PM"
src="https://user-images.githubusercontent.com/6001865/189812915-0342f841-e5ce-4f8e-8169-0cb52f3425b5.png">

Previously, the entire "args" field would have had to match, which
essentially never happened if the pointer was used in two different
functions with different function signatures

## Example

![Screen Shot 2022-09-12 at 11 49 18
PM](https://user-images.githubusercontent.com/6001865/189811621-5eed0008-f340-438e-938d-a140f836b583.png)

![Screen Shot 2022-09-12 at 11 49 45
PM](https://user-images.githubusercontent.com/6001865/189811584-a983ed1c-095f-4d54-a790-9ef3adbe4e08.png)
 2022-09-13 02:05:03 -05:00
ROCm CI Service Account 767ec039e2 SWDEV-340649 - Test case not valid anymore (#2928) 
Change-Id: I87bf19a3f4fe9966b66fad10c7371d60c3e25469


[ROCm/hip-tests commit: a8c0fa145d]
 2022-09-12 21:06:32 -07:00
ROCm CI Service Account a8c0fa145d SWDEV-340649 - Test case not valid anymore (#2928) 
Change-Id: I87bf19a3f4fe9966b66fad10c7371d60c3e25469
 2022-09-12 21:06:32 -07:00
ROCm CI Service Account 267bba96de SWDEV-340649 - Test case not valid anymore (#2928) 
Change-Id: I87bf19a3f4fe9966b66fad10c7371d60c3e25469


[ROCm/hip commit: 176e64898a]
 2022-09-12 21:06:32 -07:00
ROCm CI Service Account 176e64898a SWDEV-340649 - Test case not valid anymore (#2928) 
Change-Id: I87bf19a3f4fe9966b66fad10c7371d60c3e25469
 2022-09-12 21:06:32 -07:00
Jenkins 274d423b99 SWDEV-299127 - Merge 'develop' into 'amd-staging' 
Change-Id: I16a09979c78ba47e1b3e8e0a30e08301a001b72a


[ROCm/hip commit: a33182a9da]
 2022-09-12 19:10:35 -04:00
Jenkins a33182a9da SWDEV-299127 - Merge 'develop' into 'amd-staging' 
Change-Id: I16a09979c78ba47e1b3e8e0a30e08301a001b72a
 2022-09-12 19:10:35 -04:00
Jonathan R. Madsen 3e69264431 GOTCHA wrappers for NUMA functions (#152) 
- GOTCHA wrappers for:
  - mbind
  - migrate_pages
  - move_pages
  - numa_migrate_pages
  - numa_move_pages
  - numa_alloc
  - numa_alloc_local
  - numa_alloc_interleaved
  - numa_alloc_onnode
  - numa_realloc
  - numa_free
- bumped version to 1.6.0
- updated categories
- hardware_counters -> kernel_hardware_counters and
thread_hardware_counters
  - simplified mapping category structs to perfetto category names
- updated timemory submodule

[ROCm/rocprofiler-systems commit: 64afa49193]
 2022-09-12 17:44:27 -05:00
Jonathan R. Madsen 64afa49193 GOTCHA wrappers for NUMA functions (#152) 
- GOTCHA wrappers for:
  - mbind
  - migrate_pages
  - move_pages
  - numa_migrate_pages
  - numa_move_pages
  - numa_alloc
  - numa_alloc_local
  - numa_alloc_interleaved
  - numa_alloc_onnode
  - numa_realloc
  - numa_free
- bumped version to 1.6.0
- updated categories
- hardware_counters -> kernel_hardware_counters and
thread_hardware_counters
  - simplified mapping category structs to perfetto category names
- updated timemory submodule
 2022-09-12 17:44:27 -05:00
David Yat Sin d9935e6fba Add .kd to symbol kernel name for Binary Search sample 
Fix Binary Search sample code as kernel symbol name has a .kd
extension.

Change-Id: Id21d2e432faa40bcd5cf343345502e823678fd0f
 2022-09-12 16:17:04 -04:00
David Yat Sin 434b412a72 Add .kd to symbol kernel name for Binary Search sample 
Fix Binary Search sample code as kernel symbol name has a .kd
extension.

Change-Id: Id21d2e432faa40bcd5cf343345502e823678fd0f


[ROCm/ROCR-Runtime commit: d9935e6fba]
 2022-09-12 16:17:04 -04:00
Victor Zhang ae3d266843 Merge "SWDEV-343905 - Updated headers to match the khronos repository" into amd-staging 
[ROCm/clr commit: fe8acef64e]
 2022-09-12 14:32:42 -04:00
Victor Zhang fe8acef64e Merge "SWDEV-343905 - Updated headers to match the khronos repository" into amd-staging 2022-09-12 14:32:42 -04:00
nives-vukovic 41e09a2681 EXSWHTEC-14 - Implement tests for hipEventCreateWithFlags (#2902) 
- Add hipEventCreateWithFlags.cc with a simple positive unit test
- Divide Unit_hipEventCreate_NullCheck into two separate tests

Signed-off-by: Nives Vukovic <nives.vukovic@htecgroup.com>

[ROCm/hip-tests commit: 613301ce68]
 2022-09-12 20:29:42 +05:30
nives-vukovic 613301ce68 EXSWHTEC-14 - Implement tests for hipEventCreateWithFlags (#2902) 
- Add hipEventCreateWithFlags.cc with a simple positive unit test
- Divide Unit_hipEventCreate_NullCheck into two separate tests

Signed-off-by: Nives Vukovic <nives.vukovic@htecgroup.com>
 2022-09-12 20:29:42 +05:30
nives-vukovic d2157cb085 EXSWHTEC-14 - Implement tests for hipEventCreateWithFlags (#2902) 
- Add hipEventCreateWithFlags.cc with a simple positive unit test
- Divide Unit_hipEventCreate_NullCheck into two separate tests

Signed-off-by: Nives Vukovic <nives.vukovic@htecgroup.com>

[ROCm/hip commit: 758a6b863c]
 2022-09-12 20:29:42 +05:30
nives-vukovic 758a6b863c EXSWHTEC-14 - Implement tests for hipEventCreateWithFlags (#2902) 
- Add hipEventCreateWithFlags.cc with a simple positive unit test
- Divide Unit_hipEventCreate_NullCheck into two separate tests

Signed-off-by: Nives Vukovic <nives.vukovic@htecgroup.com>
 2022-09-12 20:29:42 +05:30
Maneesh Gupta 06593a072f SWDEV-352502 - SWDEV-355630 - Revert "SWDEV-299940 - Query global memory available on device with HSA attribute HSA_AMD_AGENT_INFO_MEMORY_AVAIL" 
This reverts commit 1b2f5fbce1.

Change-Id: I86175ccf2b543c3ff3dffec6ccae410976972f22


[ROCm/clr commit: 4afca0647e]
 2022-09-12 07:34:32 -04:00
Maneesh Gupta 4afca0647e SWDEV-352502 - SWDEV-355630 - Revert "SWDEV-299940 - Query global memory available on device with HSA attribute HSA_AMD_AGENT_INFO_MEMORY_AVAIL" 
This reverts commit 73021582d3.

Change-Id: I86175ccf2b543c3ff3dffec6ccae410976972f22
 2022-09-12 07:34:32 -04:00
ROCm CI Service Account 723f0c9f5c SWDEV-353807 - Fix deviceLib tests build when build/execution node is same (#2933) 
Change-Id: I4eccc4b6dce755cd49e8eff6df6393d2a8c552f5

[ROCm/hip-tests commit: 0f7572b4c9]
 2022-09-12 11:53:15 +05:30
ROCm CI Service Account 0f7572b4c9 SWDEV-353807 - Fix deviceLib tests build when build/execution node is same (#2933) 
Change-Id: I4eccc4b6dce755cd49e8eff6df6393d2a8c552f5
 2022-09-12 11:53:15 +05:30
ROCm CI Service Account 1200b6c30a SWDEV-353807 - Fix deviceLib tests build when build/execution node is same (#2933) 
Change-Id: I4eccc4b6dce755cd49e8eff6df6393d2a8c552f5

[ROCm/hip commit: 997f7aec04]
 2022-09-12 11:53:15 +05:30
ROCm CI Service Account 997f7aec04 SWDEV-353807 - Fix deviceLib tests build when build/execution node is same (#2933) 
Change-Id: I4eccc4b6dce755cd49e8eff6df6393d2a8c552f5
 2022-09-12 11:53:15 +05:30
Gilbert Lee 1d24c476f4 Merge branch 'develop' into 2.13.4 
[ROCm/rccl commit: 009e79623f]
 2022-09-09 23:07:04 +00:00
Gilbert Lee 009e79623f Merge branch 'develop' into 2.13.4 2022-09-09 23:07:04 +00:00
gilbertlee-amd 35872115f8 Updating stream caching (#614) 
- Adding non-captured hipStream for use in setup

[ROCm/rccl commit: dd56135a9a]
 2022-09-09 16:30:15 -06:00
gilbertlee-amd dd56135a9a Updating stream caching (#614) 
- Adding non-captured hipStream for use in setup
 2022-09-09 16:30:15 -06:00
gilbertlee-amd af71be44f1 GraphBench (#613) 
Adding simple GraphBench tool for comparing RCCL hipGraph performance

[ROCm/rccl commit: 65d78e9a1d]
 2022-09-09 12:12:25 -06:00
gilbertlee-amd 65d78e9a1d GraphBench (#613) 
Adding simple GraphBench tool for comparing RCCL hipGraph performance
 2022-09-09 12:12:25 -06:00
Wenkai Du 7874a99c75 Merge remote-tracking branch 'nccl/master' into develop 
[ROCm/rccl commit: a79d9e3586]
 2022-09-09 16:05:38 +00:00
Wenkai Du a79d9e3586 Merge remote-tracking branch 'nccl/master' into develop 2022-09-09 16:05:38 +00:00
sdashmiz 44c74209ed SWDEV-351274 - add proper checks for set attribute 
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I9243e6180f3040341396b7159622f2b96e143ac2


[ROCm/clr commit: 39fc6605ec]
 2022-09-09 10:49:33 -04:00
sdashmiz 39fc6605ec SWDEV-351274 - add proper checks for set attribute 
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I9243e6180f3040341396b7159622f2b96e143ac2
 2022-09-09 10:49:33 -04:00
Graham Sider 14aa475905 kfdtest: Update exclude for gfx1101 
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I78c1130c4a85a98a265a090e40390e56d3be2819
 2022-09-09 09:40:23 -04:00
Graham Sider df1259df37 kfdtest: Update exclude for gfx1101 
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I78c1130c4a85a98a265a090e40390e56d3be2819


[ROCm/ROCR-Runtime commit: 14aa475905]
 2022-09-09 09:40:23 -04:00
Graham Sider 82a41c7e4d kfdtest: Remove unnecessary v_ ops in IterateIsa 
IterateIsa had some leftover instructions from when the shader was
getting updated for KFDCWSRTest.BasicTest.

Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I41ae7b7948cbe2aff8bf61b170b9a7d498b836a3
 2022-09-09 09:40:04 -04:00
Graham Sider 77bdfc1389 kfdtest: Remove unnecessary v_ ops in IterateIsa 
IterateIsa had some leftover instructions from when the shader was
getting updated for KFDCWSRTest.BasicTest.

Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I41ae7b7948cbe2aff8bf61b170b9a7d498b836a3


[ROCm/ROCR-Runtime commit: 82a41c7e4d]
 2022-09-09 09:40:04 -04:00
Laurent Morichetti 1d0f27e2c8 Fix the symbol name for deprecated functions 
Change-Id: I53c0af1d1f6a3998992bdaa737e9b10829e5abc3


[ROCm/roctracer commit: 900d5e0a64]
 2022-09-08 21:04:41 -07:00