Improve support for DirectNIC (CX8)
* Add support for XDR speed detection.
* When DirectNIC is enabled, report only the RDMA interfaces.
Extend the P2C (PXN over C2C) support to send/receive operations.
Support compilation with GCC 14 (Issues #1743, #1751).
Fix the unloading of network plugins that also provide tuner capability.
Fix the change of the current device across the calls to ncclCommDestroy()
and ncclCommAbort().
A note for users on MNNVL systems: please ensure an adequate stack size for
NCCL threads. While the default Linux stack size limit of 8192 KB is known
to be sufficient, we've seen crashes if the limit is changed to
"unlimited", as it causes the glibc library to unexpectedly *decrease* the
stack size of NCCL's background threads to just 2048 KB. Use "ulimit -s"
in bash to print the current limit; if needed, reset it to 8192 KB using
"ulimit -s 8192" (one also needs to ensure that the new setting is
propagated to other nodes when launching a multi-node NCCL job).
Move the wallclock frequency query from GpuAgent to driver layer to improve
code organization and support multiple driver types. This change:
1. Add GetWallclockFrequency API to KFD/XDNA drivers
2. Move libdrm GPU info query from GpuAgent to driver implementation
3. Update GpuAgent to use the new driver API
Signed-off-by: Honglei Huang <Honglei1.Huang@amd.com>
Move the wallclock frequency query from GpuAgent to driver layer to improve
code organization and support multiple driver types. This change:
1. Add GetWallclockFrequency API to KFD/XDNA drivers
2. Move libdrm GPU info query from GpuAgent to driver implementation
3. Update GpuAgent to use the new driver API
Signed-off-by: Honglei Huang <Honglei1.Huang@amd.com>
[ROCm/ROCR-Runtime commit: 412e386b50]
- Implemented GetTileConfig in KfdDriver to retrieve tile configuration for
a specific node.
- Added a stub implementation of GetTileConfig in XdnaDriver.
- Updated driver.h to include a virtual GetTileConfig method.
- Extended hsa_internal.h with a new hsa_get_tile_config function.
- Integrated hsa_get_tile_config into hsa.cpp to call the driver-specific
implementation.
- Updated driver headers to declare the new GetTileConfig method.
Signed-off-by: Honglei Huang <Honglei1.Huang@amd.com>
- Implemented GetTileConfig in KfdDriver to retrieve tile configuration for
a specific node.
- Added a stub implementation of GetTileConfig in XdnaDriver.
- Updated driver.h to include a virtual GetTileConfig method.
- Extended hsa_internal.h with a new hsa_get_tile_config function.
- Integrated hsa_get_tile_config into hsa.cpp to call the driver-specific
implementation.
- Updated driver headers to declare the new GetTileConfig method.
Signed-off-by: Honglei Huang <Honglei1.Huang@amd.com>
[ROCm/ROCR-Runtime commit: 9bc38e2ee6]
This commit introduces a new GetClockCounters API to the driver interface.
- Implemented GetClockCounters in KfdDriver to fetch clock counters
using hsaKmtGetClockCounters.
- Added a stub implementation of GetClockCounters in XdnaDriver that
returns HSA_STATUS_ERROR.
- Modified GpuAgent to use driver().GetClockCounters instead of
directly calling hsaKmtGetClockCounters.
Signed-off-by: Honglei Huang <Honglei1.Huang@amd.com>
This commit introduces a new GetClockCounters API to the driver interface.
- Implemented GetClockCounters in KfdDriver to fetch clock counters
using hsaKmtGetClockCounters.
- Added a stub implementation of GetClockCounters in XdnaDriver that
returns HSA_STATUS_ERROR.
- Modified GpuAgent to use driver().GetClockCounters instead of
directly calling hsaKmtGetClockCounters.
Signed-off-by: Honglei Huang <Honglei1.Huang@amd.com>
[ROCm/ROCR-Runtime commit: 8d077dba3b]
This commit introduces a new GetDeviceHandle API to the driver
interface, allowing retrieval of the device handle for a
specific node.
- Implemented GetDeviceHandle in KfdDriver to fetch the AMD GPU
device handle using hsaKmtGetAMDGPUDeviceHandle.
- Added a stub implementation of GetDeviceHandle in XdnaDriver
that returns HSA_STATUS_ERROR.
- Modified GpuAgent::InitLibDrm to use driver().GetDeviceHandle
instead of directly calling hsaKmtGetAMDGPUDeviceHandle.
Signed-off-by: Honglei Huang <Honglei1.Huang@amd.com>
This commit introduces a new GetDeviceHandle API to the driver
interface, allowing retrieval of the device handle for a
specific node.
- Implemented GetDeviceHandle in KfdDriver to fetch the AMD GPU
device handle using hsaKmtGetAMDGPUDeviceHandle.
- Added a stub implementation of GetDeviceHandle in XdnaDriver
that returns HSA_STATUS_ERROR.
- Modified GpuAgent::InitLibDrm to use driver().GetDeviceHandle
instead of directly calling hsaKmtGetAMDGPUDeviceHandle.
Signed-off-by: Honglei Huang <Honglei1.Huang@amd.com>
[ROCm/ROCR-Runtime commit: 05b83e72d9]
This change improves code maintainability and error handling by
centralizing DMABuf export functionality in the driver interface.
- Replace direct hsaKmtExportDMABufHandle calls with driver's ExportDMABuf method
- Improve error handling with more specific error status returns
- Add explicit invalid parameter checks and assertions
- Consolidate DMABuf export logic in IPC and VMemory paths
- Propagate detailed error status from driver layer
Signed-off-by: Honglei Huang <Honglei1.Huang@amd.com>
This change improves code maintainability and error handling by
centralizing DMABuf export functionality in the driver interface.
- Replace direct hsaKmtExportDMABufHandle calls with driver's ExportDMABuf method
- Improve error handling with more specific error status returns
- Add explicit invalid parameter checks and assertions
- Consolidate DMABuf export logic in IPC and VMemory paths
- Propagate detailed error status from driver layer
Signed-off-by: Honglei Huang <Honglei1.Huang@amd.com>
[ROCm/ROCR-Runtime commit: 837fd044d0]
* addressing issues
* doc fix
* test fix
* fix
* fix formatting issue and doc update
* fix column size
* fix
* fix formatting in output
* tests fix
* test fix
* add new line
* add new line
* fix new line
* fixing typo in using-rocprofv3-avail.rst
[ROCm/rocprofiler-sdk commit: 3aaffc42da]
* addressing issues
* doc fix
* test fix
* fix
* fix formatting issue and doc update
* fix column size
* fix
* fix formatting in output
* tests fix
* test fix
* add new line
* add new line
* fix new line
* fixing typo in using-rocprofv3-avail.rst
* Fix roofline rocm version bug
* Fix utils bug
* Remove unnecessary tests
* Do not check textual-fspicker package in cmake build
* Use rocprofv3 to test MI 100 and fix tests
[ROCm/rocprofiler-compute commit: 000fd4f5b2]
* Fix roofline rocm version bug
* Fix utils bug
* Remove unnecessary tests
* Do not check textual-fspicker package in cmake build
* Use rocprofv3 to test MI 100 and fix tests
* Disabled violation argument for monitor on guests as it is supported on BM only.
* Added `-v` and `--violation` args to metric along with `throttle` due to legacy behavior.
* Supressed metric throttle arg and do not show in help text
---------
Signed-off-by: Bindhiya Kanangot Balakrishnan <Bindhiya.KanangotBalakrishnan@amd.com>
* Disabled violation argument for monitor on guests as it is supported on BM only.
* Added `-v` and `--violation` args to metric along with `throttle` due to legacy behavior.
* Supressed metric throttle arg and do not show in help text
---------
Signed-off-by: Bindhiya Kanangot Balakrishnan <Bindhiya.KanangotBalakrishnan@amd.com>
[ROCm/amdsmi commit: f6b854b4ed]
Modified the file used to fetch process name so that complete name with path can be displayed.
Changes:
amd-smi monitor -q
- human readable format will output only the process name
- csv and json formats will print the full path
amd-smi process
- name will always be the full path to the process
amd-smi (default output)
- name will always be truncated.
---------
Signed-off-by: Bindhiya Kanangot Balakrishnan <Bindhiya.KanangotBalakrishnan@amd.com>
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
Modified the file used to fetch process name so that complete name with path can be displayed.
Changes:
amd-smi monitor -q
- human readable format will output only the process name
- csv and json formats will print the full path
amd-smi process
- name will always be the full path to the process
amd-smi (default output)
- name will always be truncated.
---------
Signed-off-by: Bindhiya Kanangot Balakrishnan <Bindhiya.KanangotBalakrishnan@amd.com>
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
[ROCm/amdsmi commit: 514517e536]
Kamil Iskra
Bertan Dogancay
Honglei Huang
Nilesh M Negi
GunaShekar, Ajay
Tony Gutierrez
Nagaraj, Sriraksha
Justin Williams
David Yat Sin
Mustafa Abduljabbar
vedithal-amd
xuchen-amd
Atul Kulkarni
Elwazir, Ammar
Kanangot Balakrishnan, Bindhiya
AL Musaffar, Yazen
josnarlo
mberenjk