76333 Commits
Author SHA1 Message Date
sdashmiz a88c64f86d SWDEV-367877 - Detect cycle in graph
- detect cycle when graph is instantiated

- remove level calculation from add/remove node

Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I3f7432f91f70aec8e4fd866b2766256f8a9a0cfe

graph-cycle-corrections

Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I8a3cec9a5a503aac6ea1e85ff3dd2b972790fb1d


[ROCm/clr commit: 8578da8a3d]
2023-05-18 09:44:39 -04:00
sdashmiz 8578da8a3d SWDEV-367877 - Detect cycle in graph
- detect cycle when graph is instantiated

- remove level calculation from add/remove node

Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I3f7432f91f70aec8e4fd866b2766256f8a9a0cfe

graph-cycle-corrections

Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I8a3cec9a5a503aac6ea1e85ff3dd2b972790fb1d
2023-05-18 09:44:39 -04:00
Ranjith Ramakrishnan 1fa9649637 For testing purpose removing the dependency of rocm-core-asan with rocm-core
The change is made temporarily for testing the ASAN jenkins build.
Need to revert this change after testing

Change-Id: Icbfcfec423170114becdc82696afcf54b6d79d48


[ROCm/rocm-core commit: 8ce1529027]
2023-05-18 05:34:35 -07:00
Ranjith Ramakrishnan 8ce1529027 For testing purpose removing the dependency of rocm-core-asan with rocm-core
The change is made temporarily for testing the ASAN jenkins build.
Need to revert this change after testing

Change-Id: Icbfcfec423170114becdc82696afcf54b6d79d48
2023-05-18 05:34:35 -07:00
gobhardw 36353266bf Fixing broken profiler v2 from build directory
Change-Id: Ice4f49519953a4cce0ac116535f402ff94adf9d1
2023-05-18 15:03:48 +05:30
gobhardw 6963f82c84 Fixing broken profiler v2 from build directory
Change-Id: Ice4f49519953a4cce0ac116535f402ff94adf9d1


[ROCm/rocprofiler commit: 36353266bf]
2023-05-18 15:03:48 +05:30
Jaydeep Patel f7c3760996 SWDEV-389199 - Add GL Memory object to memory map so that it is acknowledged by getMemoryObject.
Change-Id: Ia58bc30ba8be016c1f004f6af01dd767f45f7905


[ROCm/clr commit: 97dc73f581]
2023-05-18 01:10:33 -04:00
Jaydeep Patel 97dc73f581 SWDEV-389199 - Add GL Memory object to memory map so that it is acknowledged by getMemoryObject.
Change-Id: Ia58bc30ba8be016c1f004f6af01dd767f45f7905
2023-05-18 01:10:33 -04:00
Charis Poag 99bc3fb502 [SWDEV-398070] Adding logging to ROCm SMI (by default off)
Updates:
    * [rocm-smi] Provide a thread-safe logging feature
    * [rocm-smi] Adding logrotation into install/upgrade/remove
      scripts
    * [rocm-smi] Updated cmake lists to include rocm_smi_logger
    * [rocm-smi] Updated DEB/RPM install/remove logging file &
      folder with all users having r/w privledges for
      /var/log/rocm_smi_lib/ROCm-SMI-lib.log
    * [rocm-smi] Added ability to do a glob search for multiple files
      (globFileExists), assists doing file searches with * strings
    * [rocm-smi] Added ability to log system details when RSMI_LOGGING
      is turned on (getSystemDetails())
    * [rocm-smi] Added logging to provide which ROCm API is being called
      when RSMI_LOGGING is on
    * [rocm-smi] Added logging to provide SYSFS path and read value,
      when RSMI_LOGGING is on. Provides error reponse on failure.
    * [rocm-smi] Added logging to provide SYSFS path and read value,
      when RSMI_LOGGING is on. Provides error reponse on failure.
    * [rocm-smi] Added environment variable RSMI_LOGGING to control
      when logging is enabled or disabled. By default, by not
      setting this env. variable, logging is turned off. When
      setting RSMI_LOGGING=<any value>, logging is enabled
      which is placed in /var/log/rocm_smi_lib/ROCm-SMI-lib.log file.
      Setting RSMI_LOGGING is allowed in both debug and release builds.
    * [rocm-smi] Removed an initialize procedure which keeps
      debug_inf_loop. Seems this feature is not being used.

Change-Id: I79b48387609c6233c6f05b04fb8bba66b68c2399
Signed-off-by: Charis Poag <Charis.Poag@amd.com>


[ROCm/rocm_smi_lib commit: c3a095a180]
2023-05-17 21:18:52 -05:00
Charis Poag c3a095a180 [SWDEV-398070] Adding logging to ROCm SMI (by default off)
Updates:
    * [rocm-smi] Provide a thread-safe logging feature
    * [rocm-smi] Adding logrotation into install/upgrade/remove
      scripts
    * [rocm-smi] Updated cmake lists to include rocm_smi_logger
    * [rocm-smi] Updated DEB/RPM install/remove logging file &
      folder with all users having r/w privledges for
      /var/log/rocm_smi_lib/ROCm-SMI-lib.log
    * [rocm-smi] Added ability to do a glob search for multiple files
      (globFileExists), assists doing file searches with * strings
    * [rocm-smi] Added ability to log system details when RSMI_LOGGING
      is turned on (getSystemDetails())
    * [rocm-smi] Added logging to provide which ROCm API is being called
      when RSMI_LOGGING is on
    * [rocm-smi] Added logging to provide SYSFS path and read value,
      when RSMI_LOGGING is on. Provides error reponse on failure.
    * [rocm-smi] Added logging to provide SYSFS path and read value,
      when RSMI_LOGGING is on. Provides error reponse on failure.
    * [rocm-smi] Added environment variable RSMI_LOGGING to control
      when logging is enabled or disabled. By default, by not
      setting this env. variable, logging is turned off. When
      setting RSMI_LOGGING=<any value>, logging is enabled
      which is placed in /var/log/rocm_smi_lib/ROCm-SMI-lib.log file.
      Setting RSMI_LOGGING is allowed in both debug and release builds.
    * [rocm-smi] Removed an initialize procedure which keeps
      debug_inf_loop. Seems this feature is not being used.

Change-Id: I79b48387609c6233c6f05b04fb8bba66b68c2399
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
2023-05-17 21:18:52 -05:00
Charis Poag 14b2f93d48 [SWDEV-398070] Adding logging to ROCm SMI (by default off)
Updates:
    * [rocm-smi] Provide a thread-safe logging feature
    * [rocm-smi] Adding logrotation into install/upgrade/remove
      scripts
    * [rocm-smi] Updated cmake lists to include rocm_smi_logger
    * [rocm-smi] Updated DEB/RPM install/remove logging file &
      folder with all users having r/w privledges for
      /var/log/rocm_smi_lib/ROCm-SMI-lib.log
    * [rocm-smi] Added ability to do a glob search for multiple files
      (globFileExists), assists doing file searches with * strings
    * [rocm-smi] Added ability to log system details when RSMI_LOGGING
      is turned on (getSystemDetails())
    * [rocm-smi] Added logging to provide which ROCm API is being called
      when RSMI_LOGGING is on
    * [rocm-smi] Added logging to provide SYSFS path and read value,
      when RSMI_LOGGING is on. Provides error reponse on failure.
    * [rocm-smi] Added logging to provide SYSFS path and read value,
      when RSMI_LOGGING is on. Provides error reponse on failure.
    * [rocm-smi] Added environment variable RSMI_LOGGING to control
      when logging is enabled or disabled. By default, by not
      setting this env. variable, logging is turned off. When
      setting RSMI_LOGGING=<any value>, logging is enabled
      which is placed in /var/log/rocm_smi_lib/ROCm-SMI-lib.log file.
      Setting RSMI_LOGGING is allowed in both debug and release builds.
    * [rocm-smi] Removed an initialize procedure which keeps
      debug_inf_loop. Seems this feature is not being used.

Change-Id: I79b48387609c6233c6f05b04fb8bba66b68c2399
Signed-off-by: Charis Poag <Charis.Poag@amd.com>


[ROCm/amdsmi commit: c3a095a180]
2023-05-17 21:18:52 -05:00
gobhardw 70a6c26704 fixing ci test issues for v1
Change-Id: I6be62c83a04b6a1a9f7b128086762dcf5ad79fb4
2023-05-17 21:32:12 -04:00
gobhardw d869b23a4f fixing ci test issues for v1
Change-Id: I6be62c83a04b6a1a9f7b128086762dcf5ad79fb4


[ROCm/rocprofiler commit: 70a6c26704]
2023-05-17 21:32:12 -04:00
David Yat Sin a397373cea Add HSA_ENABLE_PEER_SDMA env variable
Add support for HSA_ENABLE_PEER_SDMA env variable that can be used to
disable use of SDMA engines for device-to-device transfers. Note that
setting HSA_ENABLE_SDMA=0 will disable all SDMA transfers and override
HSA_ENABLE_PEER_SDMA values.

Change-Id: I737b3c2b2efcf3ff237f98bc748f49b8252ed24a
2023-05-18 00:10:20 +00:00
David Yat Sin 14052ab9d0 Add HSA_ENABLE_PEER_SDMA env variable
Add support for HSA_ENABLE_PEER_SDMA env variable that can be used to
disable use of SDMA engines for device-to-device transfers. Note that
setting HSA_ENABLE_SDMA=0 will disable all SDMA transfers and override
HSA_ENABLE_PEER_SDMA values.

Change-Id: I737b3c2b2efcf3ff237f98bc748f49b8252ed24a


[ROCm/ROCR-Runtime commit: a397373cea]
2023-05-18 00:10:20 +00:00
pghafari 9eda493469 SWDEV-399734, SWDEV-399733 - update maxGridDim for y and z
Change-Id: Id5b1dd2491798dd20a0672362f445ab4de164f6d


[ROCm/clr commit: fb82b292c2]
2023-05-17 19:03:04 -04:00
pghafari fb82b292c2 SWDEV-399734, SWDEV-399733 - update maxGridDim for y and z
Change-Id: Id5b1dd2491798dd20a0672362f445ab4de164f6d
2023-05-17 19:03:04 -04:00
Jacob Lambert 2e664d2492 SWDEV-371628 - Shift device lib linking into clang driver
Previously, we used the following approach and Comgr actions
for device lib linking:

    AMD_COMGR_COMPILE_SOURCE_TO_BC (compile with clang driver)
    AMD_COMGR_ADD_DEVICE_LIBRARIES (link in device libs with
      llvm-link API)

However, the clang driver can link in device libraries as part
of compilation, assuming a --rocm-path is set. In this context,
this is accomplished by using the following Comgr action instead:

  AMD_COMGR_COMPILE_SOURCE_WITH_DEVICE_LIBS_TO_BC (compile and
    link in device libs with clang driver)

Change-Id: I661465865365afecc44aa15d4df91bfab361af8d


[ROCm/clr commit: a4c5c44008]
2023-05-17 17:40:56 -04:00
Jacob Lambert a4c5c44008 SWDEV-371628 - Shift device lib linking into clang driver
Previously, we used the following approach and Comgr actions
for device lib linking:

    AMD_COMGR_COMPILE_SOURCE_TO_BC (compile with clang driver)
    AMD_COMGR_ADD_DEVICE_LIBRARIES (link in device libs with
      llvm-link API)

However, the clang driver can link in device libraries as part
of compilation, assuming a --rocm-path is set. In this context,
this is accomplished by using the following Comgr action instead:

  AMD_COMGR_COMPILE_SOURCE_WITH_DEVICE_LIBS_TO_BC (compile and
    link in device libs with clang driver)

Change-Id: I661465865365afecc44aa15d4df91bfab361af8d
2023-05-17 17:40:56 -04:00
gobhardw c3107fc258 SWDEV-398161, SWDEV-398764 Fixed --hsa-api for v2
Change-Id: I1902ff463851a3485b48f7b9400f7116a073365a
2023-05-17 17:22:44 -04:00
gobhardw 614fd47cd3 SWDEV-398161, SWDEV-398764 Fixed --hsa-api for v2
Change-Id: I1902ff463851a3485b48f7b9400f7116a073365a


[ROCm/rocprofiler commit: c3107fc258]
2023-05-17 17:22:44 -04:00
Philip Yang 5df82e3d14 kfdtest: Enable KFDEvictTest and KFDSVMEvictTest on aqua_vanjaram
For aqua_vanjaram APU mode, KFDEvictTest and KFDSVMEvictTest are
skipped. Those tests passed on dGPU mode with memory reporting partition
support on GFX 9.4.3.

Change-Id: I56357843c6743b01b807359dbb37b32391fd9a25
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
2023-05-17 17:17:46 -04:00
Philip Yang 3795454c1b kfdtest: Enable KFDEvictTest and KFDSVMEvictTest on aqua_vanjaram
For aqua_vanjaram APU mode, KFDEvictTest and KFDSVMEvictTest are
skipped. Those tests passed on dGPU mode with memory reporting partition
support on GFX 9.4.3.

Change-Id: I56357843c6743b01b807359dbb37b32391fd9a25
Signed-off-by: Philip Yang <Philip.Yang@amd.com>


[ROCm/ROCR-Runtime commit: 5df82e3d14]
2023-05-17 17:17:46 -04:00
sdashmiz 915677e9a0 SWDEV-396533 - correct _shfl function to match cuda
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I311419fd25c055339f25fe0c7a132ec9ee225600


[ROCm/clr commit: 23e99dbb07]
2023-05-17 14:19:46 -04:00
sdashmiz 23e99dbb07 SWDEV-396533 - correct _shfl function to match cuda
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I311419fd25c055339f25fe0c7a132ec9ee225600
2023-05-17 14:19:46 -04:00
Wen-Heng (Jack) Chung 932822fc46 Address review feedbacks and make the flag be disabled by default.
[ROCm/rccl commit: ca4a1dfd67]
2023-05-17 17:50:25 +00:00
Wen-Heng (Jack) Chung ca4a1dfd67 Address review feedbacks and make the flag be disabled by default. 2023-05-17 17:50:25 +00:00
Bing Ma 1e6d728730 libhsakmt: Add support functions for ASAN
Add support functions to remap the first page of device memory (GPU/GTT)
to share host ASAN logic.

Signed-off-by: David Yat Sin <David.YatSin@amd.com>
Change-Id: I4c27d5417ba80a172dccb0a079a597c5dc1c8f85
2023-05-17 13:38:19 -04:00
Bing Ma a868d0972b libhsakmt: Add support functions for ASAN
Add support functions to remap the first page of device memory (GPU/GTT)
to share host ASAN logic.

Signed-off-by: David Yat Sin <David.YatSin@amd.com>
Change-Id: I4c27d5417ba80a172dccb0a079a597c5dc1c8f85


[ROCm/ROCR-Runtime commit: 1e6d728730]
2023-05-17 13:38:19 -04:00
AravindanC 27616948ac SWDEV-336531 - Fix for relative path issue in debuginfo
Change-Id: Ie36d1a93f40ddc3a42823eb83e041af8c55bcdee


[ROCm/clr commit: 3f4b70cafd]
2023-05-17 13:06:26 -04:00
AravindanC 3f4b70cafd SWDEV-336531 - Fix for relative path issue in debuginfo
Change-Id: Ie36d1a93f40ddc3a42823eb83e041af8c55bcdee
2023-05-17 13:06:26 -04:00
Wen-Heng (Jack) Chung a7fbe68535 Skip doneEvent inside MSCCL by default.
Added a RCCL_MSCCL_ENABLE_DONE_EVENT env var, set it be 0 by default.

The env var is to control whether to use doneEvent when invoking MSCCL
kernels.

Skipping doneEvent would cause the firmware to skip L2 cache flush,
resulting in overall performance improvement.


[ROCm/rccl commit: 12dba425de]
2023-05-17 16:49:42 +00:00
Wen-Heng (Jack) Chung 12dba425de Skip doneEvent inside MSCCL by default.
Added a RCCL_MSCCL_ENABLE_DONE_EVENT env var, set it be 0 by default.

The env var is to control whether to use doneEvent when invoking MSCCL
kernels.

Skipping doneEvent would cause the firmware to skip L2 cache flush,
resulting in overall performance improvement.
2023-05-17 16:49:42 +00:00
gobhardw d6f481e570 Fixing standalone sample compilation and CHANGELOG
Change-Id: I4b8cf46b32e3d76bcedeadb12357fb6eed1ee213
2023-05-17 11:46:24 -04:00
gobhardw 39e398bfd7 Fixing standalone sample compilation and CHANGELOG
Change-Id: I4b8cf46b32e3d76bcedeadb12357fb6eed1ee213


[ROCm/rocprofiler commit: d6f481e570]
2023-05-17 11:46:24 -04:00
Ammar ELWazir ac6a76d718 Fixing rocprofv2 script
Change-Id: I96ee69e284eb51dc1fbd23682d3d88699a298dae
2023-05-17 11:41:16 -04:00
Ammar ELWazir 32f2e4c2e4 Fixing rocprofv2 script
Change-Id: I96ee69e284eb51dc1fbd23682d3d88699a298dae


[ROCm/rocprofiler commit: ac6a76d718]
2023-05-17 11:41:16 -04:00
Ranjith Ramakrishnan ad002f1e7b Use the RUNPATH provided by build scripts
RUNPATH in libraries will be : $ORIGIN
RUNPATH in binaries will be : $ORIGIN/../lib

Change-Id: Iafa66a8e02cc8c5783903d40927b63652042d2f1
2023-05-17 09:10:50 -04:00
Ranjith Ramakrishnan 82b4216e40 Use the RUNPATH provided by build scripts
RUNPATH in libraries will be : $ORIGIN
RUNPATH in binaries will be : $ORIGIN/../lib

Change-Id: Iafa66a8e02cc8c5783903d40927b63652042d2f1


[ROCm/ROCR-Runtime commit: ad002f1e7b]
2023-05-17 09:10:50 -04:00
Dalibor Stanisavljevic ca7f965018 SWDEV-384797 - Renamed measure to info
Change-Id: I2397ed189fe0171ed29bd6440f8fa0bb210b95a5
Signed-off-by: Dalibor Stanisavljevic <Dalibor.Stanisavljevic@amd.com>
2023-05-17 05:10:58 -04:00
Dalibor Stanisavljevic c76586e482 SWDEV-384797 - Renamed measure to info
Change-Id: I2397ed189fe0171ed29bd6440f8fa0bb210b95a5
Signed-off-by: Dalibor Stanisavljevic <Dalibor.Stanisavljevic@amd.com>


[ROCm/amdsmi commit: ca7f965018]
2023-05-17 05:10:58 -04:00
David Yat Sin 39feb83b88 Update documentation for hsa_amd_pointer_info
Update documentation for hsa_amd_pointer_info to clarify which fields
are invalid when the allocation type is HSA_EXT_POINTER_TYPE_UNKNOWN.

Change-Id: Idaed985962c4a98d281ebe01bef8ec2459da3985
2023-05-16 18:36:54 -04:00
David Yat Sin b8e97a8d1b Update documentation for hsa_amd_pointer_info
Update documentation for hsa_amd_pointer_info to clarify which fields
are invalid when the allocation type is HSA_EXT_POINTER_TYPE_UNKNOWN.

Change-Id: Idaed985962c4a98d281ebe01bef8ec2459da3985


[ROCm/ROCR-Runtime commit: 39feb83b88]
2023-05-16 18:36:54 -04:00
Cordell Bloor f7a7997c66 SWDEV-346940 - Remove default AMDGPU_TARGETS
hipcc and clang++ both have logic to detect the installed hardware
and to automatically select the appropriate AMDGPU target when it is
left unspecified. When the AMDGPU_TARGETS property is initialized with
a set of default values, it results in the addition of an explicit set
of --offload-arch flags being passed. These explicit architecture flags
disable the architecture autodetection in the compiler.

The resulting behaviour from setting fixed defaults makes it unpleasant
to compile with CMake because they increase the build times for projects
unless they are overriden (as most users do not need to build for all
five default architectures). The fixed defaults are also troublesome for
users with hardware not included in the default set (e.g., gfx1011,
gfx1031, gfx1100).

A possible alternative might be to detect the architecture within
hip-config.cmake rather than running the detection logic on each
compiler invocation. However, this approach is simpler.

Change-Id: I9495d766b7eed03852eb4dc72b0aabe4100bc32c
Signed-off-by: Cordell Bloor <Cordell.Bloor@amd.com>


[ROCm/clr commit: e1bed6f354]
2023-05-16 16:35:09 -06:00
Cordell Bloor e1bed6f354 SWDEV-346940 - Remove default AMDGPU_TARGETS
hipcc and clang++ both have logic to detect the installed hardware
and to automatically select the appropriate AMDGPU target when it is
left unspecified. When the AMDGPU_TARGETS property is initialized with
a set of default values, it results in the addition of an explicit set
of --offload-arch flags being passed. These explicit architecture flags
disable the architecture autodetection in the compiler.

The resulting behaviour from setting fixed defaults makes it unpleasant
to compile with CMake because they increase the build times for projects
unless they are overriden (as most users do not need to build for all
five default architectures). The fixed defaults are also troublesome for
users with hardware not included in the default set (e.g., gfx1011,
gfx1031, gfx1100).

A possible alternative might be to detect the architecture within
hip-config.cmake rather than running the detection logic on each
compiler invocation. However, this approach is simpler.

Change-Id: I9495d766b7eed03852eb4dc72b0aabe4100bc32c
Signed-off-by: Cordell Bloor <Cordell.Bloor@amd.com>
2023-05-16 16:35:09 -06:00
Cole Ramos 49e3ee6916 Merge pull request #125 from AMDResearch/separate-pmc-perf
Manually join rocprof runs

[ROCm/rocprofiler-compute commit: c5a19ca06a]
2023-05-16 16:51:46 -05:00
Cole Ramos c5a19ca06a Merge pull request #125 from AMDResearch/separate-pmc-perf
Manually join rocprof runs
2023-05-16 16:51:46 -05:00
coleramos425 0809575cc3 Enhance logging and warning reporting
Signed-off-by: coleramos425 <colramos@amd.com>


[ROCm/rocprofiler-compute commit: 44edef4635]
2023-05-16 15:39:45 -05:00
coleramos425 44edef4635 Enhance logging and warning reporting
Signed-off-by: coleramos425 <colramos@amd.com>
2023-05-16 15:39:45 -05:00
Ammar ELWazir 256875a69f SWDEV-387702 & SWDEV-387711: Fixing the location of the rocprofilerv2 tool
Change-Id: I5a7439351f1153c4c9f29b3fb4990fcc4de69c65
2023-05-16 18:10:49 +00:00