* Read the ids_flags when fetching GPU info
The ids_flags contains the flags that can help identify if a GPU
is a dGPU or an APU.
* Show correct memory pool for APUs
The kernel policy for APUs will be to choose the bigger pool of
memory (GTT or VRAM) for KFD work. Adjust the policy for the monitor
and default commands to show the right memory pool when using an APU.
* Fix powercap default to enum for sensor_ind
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
* [SWDEV-559965] Refactor amdsmi set power cap
Modified power cap set to accept args with
optional power_cap type. Added power_cap helper
validate_and_set_power_cap(). Fixed JSON output
format.
Signed-off-by: Bindhiya Kanangot Balakrishnan <Bindhiya.KanangotBalakrishnan@amd.com>
---------
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
Signed-off-by: Bindhiya Kanangot Balakrishnan <Bindhiya.KanangotBalakrishnan@amd.com>
Co-authored-by: Bindhiya Kanangot Balakrishnan <Bindhiya.KanangotBalakrishnan@amd.com>
* Added Python & C API's for new node devices. Currently these are functional for node 0 only.
- amdsmi_get_node_handle
- amdsmi_get_npm_info
* Added `amd-smi node` CLI for Node Power Management
---------
Signed-off-by: Bindhiya Kanangot Balakrishnan <Bindhiya.KanangotBalakrishnan@amd.com>
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
Co-authored-by: Maisam Arif <Maisam.Arif@amd.com>
[ROCm/amdsmi commit: f8e4771363]
Changes:
- Simplified reset calls
- Updated static limit N/A values to all possible data
(helps csv format be consistent)
- Unit format was broken on static
- get_power_cap() had min/max values swapped, and the return
was missing two fields
- Updated changelog to reflect all changes
Change-Id: I23713471b984f52085372486c6e6ff852e2f42f8
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
[ROCm/amdsmi commit: 00a893d299]
- Updated python integration test to account for PPT1 support changes
- Updated set/reset power-cap input format
- Adjusted python API and updated C++ API test
Signed-off-by: gabrpham_amdeng <Gabriel.Pham@amd.com>
Change-Id: Ia9d02868b6e91c88c10a9772d9e2d9f37c3c352f
[ROCm/amdsmi commit: 18faddf6f3]
The argparse 'choices' parameter was receiving a comma-separated string
instead of a list, causing it to treat individual characters as valid
choices rather than complete tokens like 'SPX', 'DPX', etc.
Fixed by removing the unnecessary join() operation in
get_accelerator_choices_types_indices() to return the list directly.
This matches the pattern used by get_memory_partition_types().
Now 'amd-smi set -C DPX' and other partition commands work correctly.
[ROCm/amdsmi commit: f6b1cb9024]
* Changes:
- Modified attempting to open files to check
permissions -> check read access only.
Do not try to open all paths, may cause driver issues.
Read access is sufficient to check permissions.
Reason: GPUs which support partitioning (memory/compute),
logical devices will not be valid until configured.
See `sudo amd-smi set -h` or applicable APIs
to configure on supported hardware.
Example error dmesg output:
[965358.883112] amdgpu 0000:15:00.0: amdgpu: renderD153 partition 1 not valid!
[965358.883283] amdgpu 0000:15:00.0: amdgpu: renderD154 partition 2 not valid!
[965358.883438] amdgpu 0000:15:00.0: amdgpu: renderD155 partition 3 not valid!
[965358.883594] amdgpu 0000:15:00.0: amdgpu: renderD156 partition 4 not valid!
[965358.883749] amdgpu 0000:15:00.0: amdgpu: renderD157 partition 5 not valid!
[965358.883904] amdgpu 0000:15:00.0: amdgpu: renderD158 partition 6 not valid!
[965358.884060] amdgpu 0000:15:00.0: amdgpu: renderD159 partition 7 not valid!
---------
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
[ROCm/amdsmi commit: d73726698b]
* Updated groups printing
* added parameters to check_required_groups
* two device groups since kfd and render require the same group
---------
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
Co-authored-by: Maisam Arif <Maisam.Arif@amd.com>
[ROCm/amdsmi commit: ee1445e2cc]
Added the following API's to amdsmi_interface.py.
amdsmi_get_cpu_handle()
amdsmi_get_esmi_err_msg()
amdsmi_get_gpu_event_notification()
amdsmi_get_processor_count_from_handles()
amdsmi_get_processor_handles_by_type()
amdsmi_gpu_validate_ras_eeprom()
amdsmi_init_gpu_event_notification()
amdsmi_set_gpu_event_notification_mask()
amdsmi_stop_gpu_event_notification()
amdsmi_get_gpu_busy_percent()
Added additional return value to API amdsmi_get_xgmi_plpd().
The entry policies is added to the end of the dictionary to match API definition.
The entry plpds is marked for deprecation as it has the same information as policies.
---------
Signed-off-by: josnarlo <Joseph.Narlo@amd.com>
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
[ROCm/amdsmi commit: 7decbc67a1]
* Update amd-smi doc with examples of CPER and AFID API usage.
---------
Signed-off-by: Oosman Saeed <oossaeed@amd.com>
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
[ROCm/amdsmi commit: fd5e37a07e]
Changes:
- Update violation status logic and metric naming for XCP/XCC metrics (thrm/thm consistency)
- Added XCP identifier in monitor to allow partition metrics to be shown with applicable APIs
(Violation Status is the first example of this in monitor)
- Improve CLI monitor output:
support multiple GPU lines per GPU, add new columns, and better formatting
- Refactor helpers and logger for flexible unit formatting and table rendering
- Add examples for amdsmi_get_gpu_pm_metrics_info()/amdsmi_get_gpu_reg_table_info()
new metrics APIs in C++ example
- Sync Python/C++ interface and structures for new metrics fields and naming
- Remove deprecated/unused RSMI activity APIs, documentation not needed since
the APIs no longer exist in ROCm SMI either.
- Cleanup metric violations + fix handle watch arguments
- Provide better handling/doc for average_flattened_ints()
- Group xcp metrics with brackets in human readable + adjust output size
Signed-off-by: Poag, Charis <Charis.Poag@amd.com>
[ROCm/amdsmi commit: e2e4fc65c1]
Description:
- Added a new API `amdsmi_gpu_driver_reload()` to reload the AMD GPU driver independently.
- Updated CLI (`sudo amd-smi reset -r`) and Python bindings to support driver reload functionality.
- Removed automatic driver reload from `amdsmi_set_gpu_memory_partition()` and `amdsmi_set_gpu_memory_partition_mode()`.
- Enhanced CLI and test cases to allow users to control when the driver reload occurs.
- Updated documentation and changelog to reflect the new driver reload process.
- Improved error handling and logging for driver reload operations.
- Added progress bar and user confirmation prompts for driver reload commands.
* Update build/test strategy to only allow one test execution at a time
* Modify API verbage + modify systemctl error output
- Systemctl is typically not enabled on docker.
- And is an edge case for gpu being active process/etc for display devices.
* Remove AMDSMI_STATUS_AMDGPU_RESTART_ERR from the return values
* Move driver reload to after we save original compute partitions
---------
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
[ROCm/amdsmi commit: d24dc7ef89]
The bug was reproduced like this.
In terminal #1, run command:
sudo amd-smi ras --cper --gpu 6 --severity all --folder /tmp/cper_dump --follow
In terminal #2, inject errors:
while true; do sudo amdgpuras -b 7 -s 1 -m 6 -t 2; sleep 2; done
The terminal #1 starts dumping cper entry information that it captures. After 20 entries have been captured, open terminal #3 and run same command as terminal #1:
sudo amd-smi ras --cper --gpu 6 --severity all --folder /tmp/cper_dump --follow
From terminal #3, there will be no output, even when terminal #1 continues capturing and printing information.
The fix:
Since we already have more than 20 CPER entries available in the GPU buffer, when we run the command from terminal #3 to start capturing from the beginning and pass 20 buffers to copy entries to, the C++ API returns a code saying there is more data available.
The Python CLI should not treat this as an error, but should continue to print what the API returned.
---------
Signed-off-by: Oosman Saeed <oossaeed@amd.com>
[ROCm/amdsmi commit: 5b95d227bc]
* Adjusted help text
* Adjusted --afid to run only with --cper-file
* Fixed interface return error
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
Change-Id: I2b96f4515c85f3b9dd84ba5c2d819729a997141b
[ROCm/amdsmi commit: ac63f410c2]
* fix endian problem
* use hw_revision and flags_mask from cper section instead of hardcoded values
---------
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
[ROCm/amdsmi commit: 815e0252b1]
* show afids on each line of printout
* clean up afids and cper code
---------
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
Co-authored-by: Maisam Arif <Maisam.Arif@amd.com>
[ROCm/amdsmi commit: fab13c5b60]
(#280)
- Fixed potential issue with min/max values when only one frequency is available
- Improve error handling in GPU frequency range detection
- Refactor clock frequency range detection for better readability
- Added special handling for current frequency indicator (*) in DPM output
- Added comments explaining special case handling for current frequency
- Cleaned up incorrect definitions in hsmp metric table definition
---------
Signed-off-by: Juan Castillo <juan.castillo@amd.com>
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
Co-authored-by: Maisam Arif <Maisam.Arif@amd.com>
[ROCm/amdsmi commit: 4d92dea079]
* Changes:
- Updates to DRM renderD* / card* pathing for partition
- Now use KFD to discover AMD devices and populate accordingly
Device MUST have an accessible KFD node (via cgroups)
- Updated serveral AMD SMI CLI outputs to handle SYSFS files
which are not accessible on partition nodes
- Tests are updated to handle not supported features
- Added new method to help get card/drm info
(rsmi_dev_device_identifiers_get) from ROCm SMI
- Renamed device->get_card_id() & device->get_drm_render_minor()
These can now be used on internal AMD SMI calls.
- Removed warnings shown in build
Change-Id: Ice882fd9b97fb625a5bd4ef327f3ceaf247dc570
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
[ROCm/amdsmi commit: 4782528770]
Mario Limonciello
Yazen AL Musaffar
Maisam Arif
Adam Pryor
systems-assistant[bot]
Ramalingam, Muthusamy
Kanangot Balakrishnan, Bindhiya
Charis Poag
gabrpham_amdeng
Saitarun Movva
Pryor, Adam
Narlo, Joseph
Saeed, Oosman
Oosman Saeed
Deepak Mewar
Castillo, Juan