Checks and forces rereading gpu metrics unconditionally
Code changes related to the following:
* Device::dev_log_gpu_metrics()
* amdsmi_get_gpu_metrics_header_info()
Removed unintentionally during work on 'header cleanup Remove non-unified headers'
* Examples
* Unit tests
Change-Id: I83710e173c0f7102d0b7f865c18474c979a95cd8
Signed-off-by: Oliveira, Daniel <daniel.oliveira@amd.com>
[ROCm/amdsmi commit: 78074d7d77]
Renamed structs to be more conistent with what they are calling
Signed-off-by: Maisam Arif <maisarif@amd.com>
Change-Id: I6f2be2fcb76f004aa592f0dad8545565700ccd4b
[ROCm/amdsmi commit: f831cf49f7]
The @platform alias will describe which platform (for example,
gpu_baremetal or/and host) an API can be used.
The get_platform.py is a tool to compare APIs in different platforms.
Change-Id: I902bc4fea048269eace6e9f3f4a8e93f3adb7f87
[ROCm/amdsmi commit: 86d025daaa]
- Updated header and source files
- Updated python interface
- Generated python wrapper for updated header
- Updated the CLI to have cpu family & cpu model
as part of metric table
Change-Id: Iea440251797270d5d29ffe883b0ad6db790be658
[ROCm/amdsmi commit: 6f7273fda5]
The current code assume err_count sysfs only have 2 lines, which is
changed for umc_err_count by adding extra line for defer errors.
The code is changed to relax such check.
Change-Id: I7129bfcf9ef47346bca3c5a9535bc8c16c670f88
[ROCm/amdsmi commit: b125f3ee8f]
Cleans up individual gpu metric APIs which will be implemented according to 'unified-headers' standards
Code changes related to the following:
* '_get_gpu_metrics_' APIs
* Functional tests
Change-Id: I2dd2ecde11c1d77e343e0ae0e10aeb9120ae9b99
Signed-off-by: Oliveira, Daniel <daniel.oliveira@amd.com>
[ROCm/amdsmi commit: 55734d2d7a]
Changes:
* [CLI] Provide fix for "/opt/rocm/bin/amd-smi metric
TypeError: '>' not supported between instances of 'str' and 'i"
--> Python API was updated, CLI needed to reflect these changes
* [API] Updated amdsmi.h's with ROCm SMI
--> Incorrectly added mem_bandwidth_acc & mem_max_bandwidth
--> Realigned wrapper with updates
* [Test] Added metrics not shown in gpu_metrics_read.cc
Change-Id: Ia3a172377fd5a582254dd5a46d81dbec7e763cd9
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
[ROCm/amdsmi commit: 34bd26c68e]
- [CLI] Refactored cache info to display
cache flags as "cache_properties" names.
Names are displayed as a list of comma-separated
cache type strings. Previously, values
were shown one by one as ENABLED.
ex.
CACHE_PROPERTIES = <a,b,c>
- [JSON] mirrors CLI fields.
No longer display "cache_flags", renamed
field as "cache_properties" dictionary. This
allows users to better understand the
list of names provided.
- [Python API] Updated amdsmi_get_gpu_cache_info
to mirror Host return.
README.md - updated to reflect all changes.
Change-Id: Ife2ef5adcef30058937d1376efb01749e45c02fb
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
[ROCm/amdsmi commit: fe86afed8c]
Update our change log to reflect a few of the major updates
for ROCm 6.0.
Change-Id: I82157fcfad22e63b62d2409bdc979b312356abe8
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
[ROCm/amdsmi commit: c260819003]
- Update the README.md file in amdsmi_cli folder to include information
for CPU's and CORE's along with the GPU's
Change-Id: I7670811696bc5299a287a6bc8883afe40eeeb557
[ROCm/amdsmi commit: 994b956d5e]
- Add checks for no gpu devices, cpu and core devices
- Update units for core energy and cpu energy
Change-Id: Ieea43f1bb7fc303ebbbdf72f1ab22644a28df25c
[ROCm/amdsmi commit: 18d8087711]
- Fix issues in cpu API "cpu_lclk_dpm_level"
- Fix issue for invalid core id
- Update the error message for invalid devices
Change-Id: I71216ff72f89cfe0c86928ae3dce1f88eae91665
[ROCm/amdsmi commit: 256907989b]
Oliveira, Daniel
Maisam Arif
Bill(Shuzhou) Liu
Deepak Mewar
khashaik
Galantsev, Dmitrii
Charis Poag