AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge amd-staging into amd-master 20230601

Browse Source 
Signed-off-by: Hao Zhou <Hao.Zhou@amd.com>
Change-Id: I05594273afa23ac4a10ac0e9eff9788e31b73f3c
This commit is contained in:
Hao Zhou
2023-06-01 13:26:07 +08:00
parent b7320bc2fb c5e06b4040
commit c3395134e5
24 changed files with 310 additions and 839 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.github/dependabot.yml
+12
View File
@@ -0,0 +1,12 @@
# To get started with Dependabot version updates, you'll need to specify which
# package ecosystems to update and where the package manifests are located.
# Please see the documentation for all configuration options:
# https://docs.github.com/github/administering-a-repository/configuration-options-for-dependency-updates
version: 2
updates:
- package-ecosystem: "pip" # See documentation for possible values
directory: "/docs/sphinx" # Location of package manifests
open-pull-requests-limit: 10
schedule:
interval: "daily"
.gitignore
+6
View File
@@ -17,3 +17,9 @@ docs/*.pdf
__pycache__/
*.py[cod]
*.egg-*
# documentation artifacts
_toc.yml
_build/
_doxygen/
docBin/
.readthedocs.yaml
+18
View File
@@ -0,0 +1,18 @@
# Read the Docs configuration file
# See https://docs.readthedocs.io/en/stable/config-file/v2.html for details
version: 2
build:
os: ubuntu-22.04
tools:
python: "3.8"
sphinx:
configuration: docs/conf.py
formats: [htmlzip, pdf, epub]
python:
install:
- requirements: docs/sphinx/requirements.txt
docs/conf.py
+26
View File
@@ -0,0 +1,26 @@
# Configuration file for the Sphinx documentation builder.
#
# This file only contains a selection of the most common options. For a full
# list see the documentation:
# https://www.sphinx-doc.org/en/master/usage/configuration.html
import subprocess
from rocm_docs import ROCmDocs
name = "AMD SMI"
get_version = r'sed -n -e "s/^get_package_version_number(.*\"\([0-9\.]\{1,\}\).*/\1/p" ../CMakeLists.txt'
version = subprocess.getoutput(get_version)
if len(version) > 0:
name = f"{name} {version}"
external_toc_path = "./sphinx/_toc.yml"
docs_core = ROCmDocs(f"{name} Documentation")
docs_core.run_doxygen(doxygen_root="doxygen", doxygen_path="doxygen/docBin/xml")
docs_core.enable_api_reference()
docs_core.setup()
for sphinx_var in ROCmDocs.SPHINX_VARS:
globals()[sphinx_var] = getattr(docs_core, sphinx_var)
docs/amd_smi_doxygen.cfg → docs/doxygen/Doxyfile
+19 -14
View File
@@ -32,7 +32,7 @@ DOXYFILE_ENCODING = UTF-8
# title of most generated pages and in a few other places.
# The default value is: My Project.
PROJECT_NAME = "AMDSMI"
PROJECT_NAME = AMD SMI
# The PROJECT_NUMBER tag can be used to enter a project or revision number. This
# could be handy for archiving the generated documentation or if some version
@@ -58,7 +58,7 @@ PROJECT_LOGO =
# entered, it will be relative to the location where doxygen was started. If
# left blank the current directory will be used.
OUTPUT_DIRECTORY =
OUTPUT_DIRECTORY = docBin
# If the CREATE_SUBDIRS tag is set to YES then doxygen will create 4096 sub-
# directories (in 2 levels) under the output directory of each output format and
@@ -759,8 +759,8 @@ WARN_LOGFILE =
# spaces. See also FILE_PATTERNS and EXTENSION_MAPPING
# Note: If this tag is empty the current directory is searched.
INPUT = @CMAKE_CURRENT_SOURCE_DIR@/../README.md \
@CMAKE_CURRENT_SOURCE_DIR@/../include/amd_smi/amdsmi.h
INPUT = ../../README.md \
../../include/amd_smi/amdsmi.h
# This tag can be used to specify the character encoding of the source files
# that doxygen parses. Internally doxygen uses the UTF-8 encoding. Doxygen uses
@@ -1086,7 +1086,7 @@ HTML_FILE_EXTENSION = .html
# of the possible markers and block names see the documentation.
# This tag requires that the tag GENERATE_HTML is set to YES.
HTML_HEADER =
HTML_HEADER = ../_doxygen/header.html
# The HTML_FOOTER tag can be used to specify a user-defined HTML footer for each
# generated HTML page. If the tag is left blank doxygen will generate a standard
@@ -1096,7 +1096,7 @@ HTML_HEADER =
# that doxygen normally uses.
# This tag requires that the tag GENERATE_HTML is set to YES.
HTML_FOOTER =
HTML_FOOTER = ../_doxygen/footer.html
# The HTML_STYLESHEET tag can be used to specify a user-defined cascading style
# sheet that is used by each HTML page. It can be used to fine-tune the look of
@@ -1108,7 +1108,7 @@ HTML_FOOTER =
# obsolete.
# This tag requires that the tag GENERATE_HTML is set to YES.
HTML_STYLESHEET =
HTML_STYLESHEET = ../_doxygen/stylesheet.css
# The HTML_EXTRA_STYLESHEET tag can be used to specify additional user-defined
# cascading style sheets that are included after the standard style sheets
@@ -1118,10 +1118,15 @@ HTML_STYLESHEET =
# Doxygen will copy the style sheet files to the output directory.
# Note: The order of the extra style sheet files is of importance (e.g. the last
# style sheet in the list overrules the setting of the previous ones in the
# list). For an example see the documentation.
# list).
# Note: Since the styling of scrollbars can currently not be overruled in
# Webkit/Chromium, the styling will be left out of the default doxygen.css if
# one or more extra stylesheets have been specified. So if scrollbar
# customization is desired it has to be added explicitly. For an example see the
# documentation.
# This tag requires that the tag GENERATE_HTML is set to YES.
HTML_EXTRA_STYLESHEET =
HTML_EXTRA_STYLESHEET = ../_doxygen/extra_stylesheet.css
# The HTML_EXTRA_FILES tag can be used to specify one or more extra images or
# other source files which should be copied to the HTML output directory. Note
@@ -1521,7 +1526,7 @@ MATHJAX_CODEFILE =
# The default value is: YES.
# This tag requires that the tag GENERATE_HTML is set to YES.
SEARCHENGINE = YES
SEARCHENGINE = NO
# When the SERVER_BASED_SEARCH tag is enabled the search engine will be
# implemented using a web server instead of a web client using Javascript. There
@@ -1875,7 +1880,7 @@ MAN_LINKS = NO
# captures the structure of the code including all documentation.
# The default value is: NO.
GENERATE_XML = NO
GENERATE_XML = YES
# The XML_OUTPUT tag is used to specify where the XML pages will be put. If a
# relative path is entered the value of OUTPUT_DIRECTORY will be put in front of
@@ -2072,7 +2077,7 @@ TAGFILES =
# tag file that is based on the input files it reads. See section "Linking to
# external documentation" for more information about the usage of tag files.
GENERATE_TAGFILE =
GENERATE_TAGFILE = docBin/html/tagfile.xml
# If the ALLEXTERNALS tag is set to YES, all external class will be listed in
# the class index. If set to NO, only the inherited external classes will be
@@ -2307,7 +2312,7 @@ DIRECTORY_GRAPH = YES
# The default value is: png.
# This tag requires that the tag HAVE_DOT is set to YES.
DOT_IMAGE_FORMAT = png
DOT_IMAGE_FORMAT = svg
# If DOT_IMAGE_FORMAT is set to svg, then this option can be set to YES to
# enable generation of interactive SVG images that allow zooming and panning.
@@ -2319,7 +2324,7 @@ DOT_IMAGE_FORMAT = png
# The default value is: NO.
# This tag requires that the tag HAVE_DOT is set to YES.
INTERACTIVE_SVG = NO
INTERACTIVE_SVG = YES
# The DOT_PATH tag can be used to specify the path where the dot tool can be
# found. If left blank, it is assumed the dot tool can be found in the path.
docs/index.rst
+2
View File
@@ -0,0 +1,2 @@
.. include:: ../README.md
:parser: myst_parser.sphinx_
docs/license.rst
+6
View File
@@ -0,0 +1,6 @@
=======
License
=======
.. include:: ../LICENSE
:literal:
docs/sphinx/_toc.yml.in
+14
View File
@@ -0,0 +1,14 @@
# Anywhere {branch} is used, the branch name will be substituted.
# These comments will also be removed.
defaults:
numbered: False
maxdepth: 6
root: index
subtrees:
- caption: API
entries:
- file: doxygen/docBin/html/index
title: API
- caption: About
entries:
- file: license
docs/sphinx/requirements.in
+1
View File
@@ -0,0 +1 @@
rocm-docs-core[api_reference]==0.10.3
docs/sphinx/requirements.txt
+172
View File
@@ -0,0 +1,172 @@
#
# This file is autogenerated by pip-compile with Python 3.8
# by the following command:
#
# pip-compile sphinx/requirements.in
#
accessible-pygments==0.0.4
# via pydata-sphinx-theme
alabaster==0.7.13
# via sphinx
babel==2.12.1
# via
# pydata-sphinx-theme
# sphinx
beautifulsoup4==4.12.2
# via pydata-sphinx-theme
breathe==4.35.0
# via rocm-docs-core
certifi==2023.5.7
# via requests
cffi==1.15.1
# via
# cryptography
# pynacl
charset-normalizer==3.1.0
# via requests
click==8.1.3
# via
# click-log
# doxysphinx
# sphinx-external-toc
click-log==0.4.0
# via doxysphinx
cryptography==40.0.2
# via pyjwt
deprecated==1.2.13
# via pygithub
docutils==0.19
# via
# breathe
# myst-parser
# pydata-sphinx-theme
# sphinx
doxysphinx==3.3.3
# via rocm-docs-core
gitdb==4.0.10
# via gitpython
gitpython==3.1.31
# via rocm-docs-core
idna==3.4
# via requests
imagesize==1.4.1
# via sphinx
importlib-metadata==6.6.0
# via sphinx
importlib-resources==5.12.0
# via rocm-docs-core
jinja2==3.1.2
# via
# myst-parser
# sphinx
libsass==0.22.0
# via doxysphinx
linkify-it-py==1.0.3
# via myst-parser
lxml==4.9.2
# via doxysphinx
markdown-it-py==2.2.0
# via
# mdit-py-plugins
# myst-parser
markupsafe==2.1.2
# via jinja2
mdit-py-plugins==0.3.5
# via myst-parser
mdurl==0.1.2
# via markdown-it-py
mpire==2.7.1
# via doxysphinx
myst-parser[linkify]==1.0.0
# via rocm-docs-core
packaging==23.1
# via
# pydata-sphinx-theme
# sphinx
pycparser==2.21
# via cffi
pydata-sphinx-theme==0.13.3
# via
# rocm-docs-core
# sphinx-book-theme
pygithub==1.58.2
# via rocm-docs-core
pygments==2.15.1
# via
# accessible-pygments
# mpire
# pydata-sphinx-theme
# sphinx
pyjson5==1.6.2
# via doxysphinx
pyjwt[crypto]==2.7.0
# via pygithub
pynacl==1.5.0
# via pygithub
pyparsing==3.0.9
# via doxysphinx
pytz==2023.3
# via babel
pyyaml==6.0
# via
# myst-parser
# sphinx-external-toc
requests==2.30.0
# via
# pygithub
# sphinx
rocm-docs-core[api_reference]==0.10.3
# via -r sphinx/requirements.in
smmap==5.0.0
# via gitdb
snowballstemmer==2.2.0
# via sphinx
soupsieve==2.4.1
# via beautifulsoup4
sphinx==5.3.0
# via
# breathe
# myst-parser
# pydata-sphinx-theme
# rocm-docs-core
# sphinx-book-theme
# sphinx-copybutton
# sphinx-design
# sphinx-external-toc
# sphinx-notfound-page
sphinx-book-theme==1.0.1
# via rocm-docs-core
sphinx-copybutton==0.5.2
# via rocm-docs-core
sphinx-design==0.4.1
# via rocm-docs-core
sphinx-external-toc==0.3.1
# via rocm-docs-core
sphinx-notfound-page==0.8.3
# via rocm-docs-core
sphinxcontrib-applehelp==1.0.4
# via sphinx
sphinxcontrib-devhelp==1.0.2
# via sphinx
sphinxcontrib-htmlhelp==2.0.1
# via sphinx
sphinxcontrib-jsmath==1.0.1
# via sphinx
sphinxcontrib-qthelp==1.0.3
# via sphinx
sphinxcontrib-serializinghtml==1.1.5
# via sphinx
tqdm==4.65.0
# via mpire
typing-extensions==4.5.0
# via pydata-sphinx-theme
uc-micro-py==1.0.2
# via linkify-it-py
urllib3==2.0.2
# via requests
wrapt==1.15.0
# via deprecated
zipp==3.15.0
# via
# importlib-metadata
# importlib-resources
example/amd_smi_drm_example.cc
+3 -3
View File
@@ -59,7 +59,7 @@
const char *err_str; \
std::cout << "AMDSMI call returned " << RET << " at line " \
<< __LINE__ << std::endl; \
amdsmi_status_string(RET, &err_str); \
amdsmi_status_code_to_string(RET, &err_str); \
std::cout << err_str << std::endl; \
return RET; \
} \
@@ -478,9 +478,9 @@ int main() {
// Get ECC error counts
amdsmi_error_count_t err_cnt_info = {};
ret = amdsmi_get_gpu_ecc_error_count(processor_handles[j], &err_cnt_info);
ret = amdsmi_get_gpu_total_ecc_count(processor_handles[j], &err_cnt_info);
CHK_AMDSMI_RET(ret)
printf(" Output of amdsmi_get_gpu_ecc_error_count:\n");
printf(" Output of amdsmi_get_gpu_total_ecc_count:\n");
printf("\tCorrectable errors: %lu\n", err_cnt_info.correctable_count);
printf("\tUncorrectable errors: %lu\n\n",
err_cnt_info.uncorrectable_count);
example/amd_smi_nodrm_example.cc
+3 -3
View File
@@ -57,7 +57,7 @@
const char *err_str; \
std::cout << "AMDSMI call returned " << RET << " at line " \
<< __LINE__ << std::endl; \
amdsmi_status_string(RET, &err_str); \
amdsmi_status_code_to_string(RET, &err_str); \
std::cout << err_str << std::endl; \
return RET; \
} \
@@ -269,9 +269,9 @@ int main() {
// Get ECC error counts
amdsmi_error_count_t err_cnt_info = {};
ret = amdsmi_get_gpu_ecc_error_count(processor_handles[j], &err_cnt_info);
ret = amdsmi_get_gpu_total_ecc_count(processor_handles[j], &err_cnt_info);
CHK_AMDSMI_RET(ret)
printf(" Output of amdsmi_get_gpu_ecc_error_count:\n");
printf(" Output of amdsmi_get_gpu_total_ecc_count:\n");
printf("\tCorrectable errors: %lu\n", err_cnt_info.correctable_count);
printf("\tUncorrectable errors: %lu\n\n",
err_cnt_info.uncorrectable_count);
include/amd_smi/amdsmi.h
+11 -418
View File
@@ -433,16 +433,6 @@ typedef enum {
* @brief Available clock types.
*/
/**
* @brief Software components
*/
typedef enum {
AMDSMI_SW_COMP_FIRST = 0x0,
AMDSMI_SW_COMP_DRIVER = AMDSMI_SW_COMP_FIRST, //!< Driver
AMDSMI_SW_COMP_LAST = AMDSMI_SW_COMP_DRIVER
} amdsmi_sw_component_t;
/**
* Event counter types
@@ -1064,41 +1054,10 @@ typedef struct {
uint32_t cu_occupancy; //!< Compute Unit usage in percent
} amdsmi_process_info_t;
/**
* @brief Opaque handle to function-support object
*/
typedef struct amdsmi_func_id_iter_handle * amdsmi_func_id_iter_handle_t;
//! Place-holder "variant" for functions that have don't have any variants,
//! but do have monitors or sensors.
#define AMDSMI_DEFAULT_VARIANT 0xFFFFFFFFFFFFFFFF
/**
* @brief This union holds the value of an ::amdsmi_func_id_iter_handle_t. The
* value may be a function name, or an ennumerated variant value of types
* such as ::amdsmi_memory_type_t, ::amdsmi_temperature_metric_t, etc.
*/
typedef union {
uint64_t id; //!< uint64_t representation of value
const char *name; //!< name string (applicable to functions only)
union {
//!< Used for ::amdsmi_memory_type_t variants
amdsmi_memory_type_t memory_type;
//!< Used for ::amdsmi_temperature_metric_t variants
amdsmi_temperature_metric_t temp_metric;
//!< Used for ::amdsmi_event_type_t variants
amdsmi_event_type_t evnt_type;
//!< Used for ::amdsmi_event_group_t variants
amdsmi_event_group_t evnt_group;
//!< Used for ::amdsmi_clk_type_t variants
amdsmi_clk_type_t clk_type;
//!< Used for ::amdsmi_fw_block_t variants
amdsmi_fw_block_t fw_block;
//!< Used for ::amdsmi_gpu_block_t variants
amdsmi_gpu_block_t gpu_block_type;
};
} amdsmi_func_id_value_t;
/*****************************************************************************/
/** @defgroup InitShutAdmin Initialization and Shutdown
* These functions are used for initialization of AMD SMI and clean up when done.
@@ -1273,7 +1232,7 @@ amdsmi_status_t amdsmi_get_processor_handle_from_bdf(amdsmi_bdf_t bdf, amdsmi_pr
* @p id. This ID is an identification of the type of device, so calling this
* function for different devices will give the same value if they are kind
* of device. Consequently, this function should not be used to distinguish
* one device from another. amdsmi_get_gpu_pci_id() should be used to get a
* one device from another. admsmi_get_gpu_bdf_id() should be used to get a
* unique identifier.
*
* @param[in] processor_handle a processor handle
@@ -1403,22 +1362,6 @@ amdsmi_status_t amdsmi_get_gpu_subsystem_id(amdsmi_processor_handle processor_ha
amdsmi_status_t
amdsmi_get_gpu_subsystem_name(amdsmi_processor_handle processor_handle, char *name, size_t len);
/**
* @brief Get the drm minor number associated with this device
*
* @details Given a processor handle @p processor_handle, find its render device file
* /dev/dri/renderDN where N corresponds to its minor number.
*
* @param[in] processor_handle a processor handle
*
* @param[in,out] minor a pointer to a uint32_t into which minor number will
* be copied
*
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t
amdsmi_get_gpu_drm_render_minor(amdsmi_processor_handle processor_handle, uint32_t *minor);
/** @} End IDQuer */
/*****************************************************************************/
@@ -1478,7 +1421,7 @@ amdsmi_get_gpu_pci_bandwidth(amdsmi_processor_handle processor_handle, amdsmi_pc
*
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t amdsmi_get_gpu_pci_id(amdsmi_processor_handle processor_handle, uint64_t *bdfid);
amdsmi_status_t admsmi_get_gpu_bdf_id(amdsmi_processor_handle processor_handle, uint64_t *bdfid);
/**
* @brief Get the NUMA node associated with a device
@@ -1732,12 +1675,13 @@ amdsmi_get_gpu_memory_usage(amdsmi_processor_handle processor_handle, amdsmi_mem
uint64_t *used);
/**
* @brief The first call to this API returns the number of bad pages which
* should be used to allocate the buffer that should contain the bad page
* records.
* @brief Get the bad pages of a processor.
* @details This call will query the device @p processor_handle for the
* number of bad pages (written to @p num_pages address). The results are
* written to address held by the @p info pointer.
* The first call to this API returns the number of bad pages which
* should be used to allocate the buffer that should contain the bad page
* records.
* @param[in] processor_handle a processor handle
* @param[out] num_pages Number of bad page records.
* @param[out] info The results will be written to the
@@ -1772,27 +1716,6 @@ amdsmi_status_t
amdsmi_get_gpu_ras_block_features_enabled(amdsmi_processor_handle processor_handle, amdsmi_gpu_block_t block,
amdsmi_ras_err_state_t *state);
/**
* @brief Get percentage of time any device memory is being used
*
* @details Given a processor handle @p processor_handle, this function returns the
* percentage of time that any device memory is being used for the specified
* device.
*
* @param[in] processor_handle a processor handle
*
* @param[in,out] busy_percent a pointer to the uint32_t to which the busy
* percent will be written
* If this parameter is nullptr, this function will return
* ::AMDSMI_STATUS_INVAL if the function is supported with the provided,
* arguments and ::AMDSMI_STATUS_NOT_SUPPORTED if it is not supported with the
* provided arguments.
*
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t
amdsmi_get_gpu_memory_busy_percent(amdsmi_processor_handle processor_handle, uint32_t *busy_percent);
/**
* @brief Get information about reserved ("retired") memory pages
*
@@ -2024,28 +1947,6 @@ amdsmi_status_t amdsmi_set_gpu_fan_speed(amdsmi_processor_handle processor_handl
* @{
*/
/**
* @brief Get percentage of time device is busy doing any processing
*
* @details Given a processor handle @p processor_handle, this function returns the
* percentage of time that the specified device is busy. The device is
* considered busy if any one or more of its sub-blocks are working, and idle
* if none of the sub-blocks are working.
*
* @param[in] processor_handle a processor handle
*
* @param[in,out] busy_percent a pointer to the uint32_t to which the busy
* percent will be written
* If this parameter is nullptr, this function will return
* ::AMDSMI_STATUS_INVAL if the function is supported with the provided,
* arguments and ::AMDSMI_STATUS_NOT_SUPPORTED if it is not supported with the
* provided arguments.
*
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t
amdsmi_get_busy_percent(amdsmi_processor_handle processor_handle, uint32_t *busy_percent);
/**
* @brief Get coarse grain utilization counter of the specified device
*
@@ -2393,9 +2294,6 @@ amdsmi_status_t
* @brief Set the PowerPlay performance level associated with the device with
* provided processor handle with the provided value.
*
* @deprecated :: amdsmi_set_gpu_perf_level_v1() is preferred, with an
* interface that more closely matches the rest of the amd_smi API.
*
* @details Given a processor handle @p processor_handle and an ::amdsmi_dev_perf_level_t @p
* perf_level, this function will set the PowerPlay performance level for the
* device to the value @p perf_lvl.
@@ -2411,32 +2309,10 @@ amdsmi_status_t
amdsmi_status_t
amdsmi_set_gpu_perf_level(amdsmi_processor_handle processor_handle, amdsmi_dev_perf_level_t perf_lvl);
/**
* @brief Set the PowerPlay performance level associated with the device with
* provided processor handle with the provided value.
*
* @details Given a processor handle @p processor_handle and an ::amdsmi_dev_perf_level_t @p
* perf_level, this function will set the PowerPlay performance level for the
* device to the value @p perf_lvl.
*
* @note This function requires root access
*
* @param[in] processor_handle a processor handle
*
* @param[in] perf_lvl the value to which the performance level should be set
*
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t
amdsmi_set_gpu_perf_level_v1(amdsmi_processor_handle processor_handle, amdsmi_dev_perf_level_t perf_lvl);
/**
* @brief Set the overdrive percent associated with the device with provided
* processor handle with the provided value. See details for WARNING.
*
* @deprecated This function is deprecated. :: amdsmi_set_gpu_overdrive_level_v1
* has the same functionaltiy, with an interface that more closely
* matches the rest of the amd_smi API.
*
* @details Given a processor handle @p processor_handle and an overdrive level @p od,
* this function will set the overdrive level for the device to the value
@@ -2471,45 +2347,6 @@ amdsmi_status_t
*/
amdsmi_status_t amdsmi_set_gpu_overdrive_level(amdsmi_processor_handle processor_handle, uint32_t od);
/**
* @brief Set the overdrive percent associated with the device with provided
* processor handle with the provided value. See details for WARNING.
*
* @details Given a processor handle @p processor_handle and an overdrive level @p od,
* this function will set the overdrive level for the device to the value
* @p od. The overdrive level is an integer value between 0 and 20, inclusive,
* which represents the overdrive percentage; e.g., a value of 5 specifies
* an overclocking of 5%.
*
* The overdrive level is specific to the gpu system clock.
*
* The overdrive level is the percentage above the maximum Performance Level
* to which overclocking will be limited. The overclocking percentage does
* not apply to clock speeds other than the maximum. This percentage is
* limited to 20%.
*
* ******WARNING******
* Operating your AMD GPU outside of official AMD specifications or outside of
* factory settings, including but not limited to the conducting of
* overclocking (including use of this overclocking software, even if such
* software has been directly or indirectly provided by AMD or otherwise
* affiliated in any way with AMD), may cause damage to your AMD GPU, system
* components and/or result in system failure, as well as cause other problems.
* DAMAGES CAUSED BY USE OF YOUR AMD GPU OUTSIDE OF OFFICIAL AMD SPECIFICATIONS
* OR OUTSIDE OF FACTORY SETTINGS ARE NOT COVERED UNDER ANY AMD PRODUCT
* WARRANTY AND MAY NOT BE COVERED BY YOUR BOARD OR SYSTEM MANUFACTURER'S
* WARRANTY. Please use this utility with caution.
*
* @note This function requires root access
*
* @param[in] processor_handle a processor handle
*
* @param[in] od the value to which the overdrive level should be set
*
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t amdsmi_set_gpu_overdrive_level_v1(amdsmi_processor_handle processor_handle, uint32_t od);
/**
* @brief Control the set of allowed frequencies that can be used for the
* specified clock.
@@ -2566,34 +2403,7 @@ amdsmi_status_t amdsmi_set_clk_freq(amdsmi_processor_handle processor_handle,
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t
amdsmi_get_version(amdsmi_version_t *version);
/**
* @brief Get the driver version string for the current system.
*
* @details Given a software component @p component, a pointer to a char
* buffer, @p ver_str, this function will write the driver version string
* (up to @p len characters) for the current system to @p ver_str. The caller
* must ensure that it is safe to write at least @p len characters to @p
* ver_str.
*
* @param[in] component The component for which the version string is being
* requested
*
* @param[in,out] ver_str A pointer to a buffer of char's to which the version
* of @p component will be written
*
* @param[in] len the length of the caller provided buffer @p name.
*
* @note ::AMDSMI_STATUS_INSUFFICIENT_SIZE is returned if @p len bytes is not
* large enough to hold the entire name. In this case, only @p len bytes will
* be written.
*
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t
amdsmi_get_version_str(amdsmi_sw_component_t component, char *ver_str,
uint32_t len);
amdsmi_get_lib_version(amdsmi_version_t *version);
/** @} End VersQuer */
@@ -2693,7 +2503,7 @@ amdsmi_status_t amdsmi_get_gpu_ecc_status(amdsmi_processor_handle processor_han
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t
amdsmi_status_string(amdsmi_status_t status, const char **status_string);
amdsmi_status_code_to_string(amdsmi_status_t status, const char **status_string);
/** @} End ErrQuer */
@@ -3125,7 +2935,7 @@ amdsmi_topo_get_link_weight(amdsmi_processor_handle processor_handle_src, amdsmi
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t
amdsmi_get_minmax_bandwidth(amdsmi_processor_handle processor_handle_src, amdsmi_processor_handle processor_handle_dst,
admsmi_get_minmax_bandwith_between_processors(amdsmi_processor_handle processor_handle_src, amdsmi_processor_handle processor_handle_dst,
uint64_t *min_bandwidth, uint64_t *max_bandwidth);
/**
@@ -3179,223 +2989,6 @@ amdsmi_is_P2P_accessible(amdsmi_processor_handle processor_handle_src, amdsmi_pr
/** @} End HWTopo */
/*****************************************************************************/
/** @defgroup APISupport Supported Functions
* API function support varies by both GPU type and the version of the
* installed ROCm stack. The functions described in this section can be used
* to determine, up front, which functions are supported for a given device
* on a system. If such "up front" knowledge of support for a function is not
* needed, alternatively, one can call a device related function and check the
* return code.
*
* Some functions have several variations ("variants") where some variants are
* supported and others are not. For example, on a given device,
* :: amdsmi_get_temp_metric may support some types of temperature metrics
* (e.g., ::AMDSMI_TEMP_CRITICAL_HYST), but not others
* (e.g., ::AMDSMI_TEMP_EMERGENCY).
*
* In addition to a top level of variant support for a function, a function
* may have varying support for monitors/sensors. These are considered
* "sub-variants" in functions described in this section. Continuing the
* :: amdsmi_get_temp_metric example, if variant
* ::AMDSMI_TEMP_CRITICAL_HYST is supported, perhaps
* only the sub-variant sensors ::AMDSMI_TEMP_TYPE_EDGE
* and ::AMDSMI_TEMP_TYPE_EDGE are supported, but not
* ::AMDSMI_TEMP_TYPE_MEMORY.
*
* In cases where a function takes in a sensor id parameter but does not have
* any "top level" variants, the functions in this section will indicate a
* default "variant", ::AMDSMI_DEFAULT_VARIANT, for the top level variant, and
* the various monitor support will be sub-variants of this.
*
* The functions in this section use the "iterator" concept to list which
* functions are supported; to list which variants of the supported functions
* are supported; and finally which monitors/sensors are supported for a
* variant.
*
* Here is example code that prints out all supported functions, their
* supported variants and sub-variants. Please see the related descriptions
* functions and AMDSMI types.
* @latexonly
* \pagebreak
* @endlatexonly
* @code{.cpp}
* amdsmi_func_id_iter_handle_t iter_handle, var_iter, sub_var_iter;
* amdsmi_func_id_value_t value;
* amdsmi_status_t err;
* amdsmi_processor_handle device;
*
* // Get the processor handle via amdsmi_get_processor_handles()
* // ... ...
*
* std::cout << "Supported AMDSMI Functions:" << std::endl; *
* err = amdsmi_open_supported_func_iterator(device, &iter_handle);
*
* while (1) {
* err = amdsmi_get_func_iter_value(iter_handle, &value);
* std::cout << "Function Name: " << value.name << std::endl;
*
* err = amdsmi_open_supported_variant_iterator(iter_handle, &var_iter);
* if (err != AMDSMI_STATUS_NO_DATA) {
* std::cout << "\tVariants/Monitors: ";
* while (1) {
* err = amdsmi_get_func_iter_value(var_iter, &value);
* if (value.id == AMDSMI_DEFAULT_VARIANT) {
* std::cout << "Default Variant ";
* } else {
* std::cout << value.id;
* }
* std::cout << " (";
*
* err =
* amdsmi_open_supported_variant_iterator(var_iter, &sub_var_iter);
* if (err != AMDSMI_STATUS_NO_DATA) {
*
* while (1) {
* err = amdsmi_get_func_iter_value(sub_var_iter, &value);
* std::cout << value.id << ", ";
*
* err = amdsmi_next_func_iter(sub_var_iter);
*
* if (err == AMDSMI_STATUS_NO_DATA) {
* break;
* }
* }
* err = amdsmi_close_supported_func_iterator(&sub_var_iter);
* }
*
* std::cout << "), ";
*
* err = amdsmi_next_func_iter(var_iter);
*
* if (err == AMDSMI_STATUS_NO_DATA) {
* break;
* }
* }
* std::cout << std::endl;
*
* err = amdsmi_close_supported_func_iterator(&var_iter);
* }
*
* err = amdsmi_next_func_iter(iter_handle);
*
* if (err == AMDSMI_STATUS_NO_DATA) {
* break;
* }
* }
* err = amdsmi_close_supported_func_iterator(&iter_handle);
* }
* @endcode
*
* @{
*/
/**
* @brief Get a function name iterator of supported AMDSMI functions for a device
*
* @details Given a processor handle @p processor_handle, this function will write a function
* iterator handle to the caller-provided memory pointed to by @p handle. This
* handle can be used to iterate through all the supported functions.
*
* Note that although this function takes in @p processor_handle as an argument,
* ::amdsmi_open_supported_func_iterator itself will not be among the
* functions listed as supported. This is because
* ::amdsmi_open_supported_func_iterator does not depend on hardware or
* driver support and should always be supported.
*
* @param[in] processor_handle a processor handle of device for which support information is
* requested
*
* @param[in,out] handle A pointer to caller-provided memory to which the
* function iterator will be written.
*
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t
amdsmi_open_supported_func_iterator(amdsmi_processor_handle processor_handle,
amdsmi_func_id_iter_handle_t *handle);
/**
* @brief Get a variant iterator for a given handle
*
* @details Given a ::amdsmi_func_id_iter_handle_t @p obj_h, this function will
* write a function iterator handle to the caller-provided memory pointed to
* by @p var_iter. This handle can be used to iterate through all the supported
* variants of the provided handle. @p obj_h may be a handle to a function
* object, as provided by a call to ::amdsmi_open_supported_func_iterator, or
* it may be a variant itself (from a call to
* ::amdsmi_open_supported_variant_iterator), it which case @p var_iter will
* be an iterator of the sub-variants of @p obj_h (e.g., monitors).
*
* This call allocates a small amount of memory to @p var_iter. To free this memory
* ::amdsmi_close_supported_func_iterator should be called on the returned
* iterator handle @p var_iter when it is no longer needed.
*
* @param[in] obj_h an iterator handle for which the variants are being requested
*
* @param[in,out] var_iter A pointer to caller-provided memory to which the
* sub-variant iterator will be written.
*
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t
amdsmi_open_supported_variant_iterator(amdsmi_func_id_iter_handle_t obj_h,
amdsmi_func_id_iter_handle_t *var_iter);
/**
* @brief Advance a function identifer iterator
*
* @details Given a function id iterator handle (::amdsmi_func_id_iter_handle_t)
* @p handle, this function will increment the iterator to point to the next
* identifier. After a successful call to this function, obtaining the value
* of the iterator @p handle will provide the value of the next item in the
* list of functions/variants.
*
* If there are no more items in the list, ::AMDSMI_STATUS_NO_DATA is returned.
*
* @param[in] handle A pointer to an iterator handle to be incremented
*
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t
amdsmi_next_func_iter(amdsmi_func_id_iter_handle_t handle);
/**
* @brief Close a variant iterator handle
*
* @details Given a pointer to an ::amdsmi_func_id_iter_handle_t @p handle, this
* function will free the resources being used by the handle
*
* @param[in] handle A pointer to an iterator handle to be closed
*
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t
amdsmi_close_supported_func_iterator(amdsmi_func_id_iter_handle_t *handle);
/**
* @brief Get the value associated with a function/variant iterator
*
* @details Given an ::amdsmi_func_id_iter_handle_t @p handle, this function
* will write the identifier of the function/variant to the user provided
* memory pointed to by @p value.
*
* @p value may point to a function name, a variant id, or a monitor/sensor
* index, depending on what kind of iterator @p handle is
*
* @param[in] handle An iterator for which the value is being requested
*
* @param[in,out] value A pointer to an ::amdsmi_func_id_value_t provided by the
* caller to which this function will write the value assocaited with @p handle
*
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t
amdsmi_get_func_iter_value(amdsmi_func_id_iter_handle_t handle,
amdsmi_func_id_value_t *value);
/** @} End APISupport */
/*****************************************************************************/
/** @defgroup EvntNotif Event Notification Functions
* These functions are used to configure for and get asynchronous event
@@ -3799,7 +3392,7 @@ amdsmi_get_gpu_process_info(amdsmi_processor_handle processor_handle, amdsmi_pro
*/
/**
* @brief Returns the number of ECC errors (correctable and
* @brief Returns the total number of ECC errors (correctable and
* uncorrectable) in the given GPU.
*
* @param[in] processor_handle Device which to query
@@ -3810,7 +3403,7 @@ amdsmi_get_gpu_process_info(amdsmi_processor_handle processor_handle, amdsmi_pro
* @return ::amdsmi_status_t | ::AMDSMI_STATUS_SUCCESS on success, non-zero on fail
*/
amdsmi_status_t
amdsmi_get_gpu_ecc_error_count(amdsmi_processor_handle processor_handle, amdsmi_error_count_t *ec);
amdsmi_get_gpu_total_ecc_count(amdsmi_processor_handle processor_handle, amdsmi_error_count_t *ec);
/** @} End eccinfo */
rocm_smi/src/rocm_smi_gpu_metrics.cc
+2 -2
View File
@@ -129,8 +129,8 @@ typedef struct {
/* Voltage (mV) */
uint16_t voltage_soc;
uint16_t voltage_gfx;
uint16_t voltage_mem;
uint16_t gfx_voltage;
uint16_t mem_voltage;
uint16_t padding1;
src/CMakeLists.txt
-36
View File
@@ -138,39 +138,3 @@ install(
# DESTINATION libexec/${AMD_SMI})
#install(FILES ${CMAKE_CURRENT_BINARY_DIR}/bin/rocm-smi
# DESTINATION bin)
# Generate Doxygen documentation
find_package(Doxygen)
find_package(LATEX COMPONENTS PDFLATEX)
if(DOXYGEN_FOUND AND LATEX_FOUND)
set(ASMI_MANUAL_NAME "AMD_SMI_Manual")
configure_file(${PROJECT_SOURCE_DIR}/docs/amd_smi_doxygen.cfg ${CMAKE_CURRENT_BINARY_DIR}/Doxyfile @ONLY)
add_custom_command(
OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/latex/refman.tex
COMMAND ${DOXYGEN_EXECUTABLE} ${CMAKE_CURRENT_BINARY_DIR}/Doxyfile
DEPENDS ${PROJECT_SOURCE_DIR}/docs/amd_smi_doxygen.cfg "${PROJECT_SOURCE_DIR}/include/amd_smi/amdsmi.h"
WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR})
add_custom_command(
OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/latex/refman.pdf
COMMAND make > /dev/null
COMMAND cp ${CMAKE_CURRENT_BINARY_DIR}/latex/refman.pdf ${PROJECT_SOURCE_DIR}/docs/${ASMI_MANUAL_NAME}_new.pdf
DEPENDS ${CMAKE_CURRENT_BINARY_DIR}/latex/refman.tex
WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}/latex)
add_custom_target(docs DEPENDS ${CMAKE_CURRENT_BINARY_DIR}/latex/refman.pdf)
add_dependencies(${AMD_SMI_TARGET} docs)
install(
FILES ${CMAKE_CURRENT_BINARY_DIR}/latex/refman.pdf
DESTINATION share/doc/${AMD_SMI}
RENAME ${ASMI_MANUAL_NAME}.pdf
COMPONENT dev)
install(
FILES ${PROJECT_SOURCE_DIR}/README.md
DESTINATION share/doc/${AMD_SMI}
COMPONENT dev)
else()
message("Doxygen or Latex is not found. Will not generate documents.")
endif(DOXYGEN_FOUND AND LATEX_FOUND)
src/amd_smi/amd_smi.cc
+6 -194
View File
@@ -138,7 +138,7 @@ amdsmi_shut_down() {
}
amdsmi_status_t
amdsmi_status_string(amdsmi_status_t status, const char **status_string) {
amdsmi_status_code_to_string(amdsmi_status_t status, const char **status_string) {
switch (status) {
case AMDSMI_STATUS_FAIL_LOAD_MODULE:
*status_string = "FAIL_LOAD_MODULE: Fail to load module.";
@@ -667,7 +667,7 @@ amdsmi_topo_get_link_weight(amdsmi_processor_handle processor_handle_src, amdsmi
}
amdsmi_status_t
amdsmi_get_minmax_bandwidth(amdsmi_processor_handle processor_handle_src, amdsmi_processor_handle processor_handle_dst,
admsmi_get_minmax_bandwith_between_processors(amdsmi_processor_handle processor_handle_src, amdsmi_processor_handle processor_handle_dst,
uint64_t *min_bandwidth, uint64_t *max_bandwidth) {
AMDSMI_CHECK_INIT();
@@ -743,145 +743,6 @@ amdsmi_reset_gpu_xgmi_error(amdsmi_processor_handle processor_handle) {
return rsmi_wrapper(rsmi_dev_xgmi_error_reset, processor_handle);
}
amdsmi_status_t
amdsmi_open_supported_func_iterator(amdsmi_processor_handle processor_handle,
amdsmi_func_id_iter_handle_t *handle) {
AMDSMI_CHECK_INIT();
if (handle == nullptr)
return AMDSMI_STATUS_INVAL;
return rsmi_wrapper(rsmi_dev_supported_func_iterator_open, processor_handle,
reinterpret_cast<rsmi_func_id_iter_handle_t*>(handle));
}
amdsmi_status_t
amdsmi_open_supported_variant_iterator(amdsmi_func_id_iter_handle_t obj_h,
amdsmi_func_id_iter_handle_t *var_iter) {
AMDSMI_CHECK_INIT();
if (var_iter == nullptr)
return AMDSMI_STATUS_INVAL;
auto r = rsmi_dev_supported_variant_iterator_open(
reinterpret_cast<rsmi_func_id_iter_handle_t>(obj_h),
reinterpret_cast<rsmi_func_id_iter_handle_t*>(var_iter));
return amd::smi::rsmi_to_amdsmi_status(r);
}
amdsmi_status_t
amdsmi_next_func_iter(amdsmi_func_id_iter_handle_t handle) {
AMDSMI_CHECK_INIT();
auto r = rsmi_func_iter_next(
reinterpret_cast<rsmi_func_id_iter_handle_t>(handle));
return amd::smi::rsmi_to_amdsmi_status(r);
}
amdsmi_status_t
amdsmi_close_supported_func_iterator(amdsmi_func_id_iter_handle_t *handle) {
AMDSMI_CHECK_INIT();
if (handle == nullptr)
return AMDSMI_STATUS_INVAL;
auto r = rsmi_dev_supported_func_iterator_close(
reinterpret_cast<rsmi_func_id_iter_handle_t*>(handle));
return amd::smi::rsmi_to_amdsmi_status(r);
}
amdsmi_status_t
amdsmi_get_func_iter_value(amdsmi_func_id_iter_handle_t handle,
amdsmi_func_id_value_t *value) {
AMDSMI_CHECK_INIT();
if (value == nullptr)
return AMDSMI_STATUS_INVAL;
static const std::map<std::string, const char*> rsmi_2_amdsmi = {
{"rsmi_dev_vram_vendor_get", "amdsmi_get_gpu_vram_vendor"},
{"rsmi_dev_id_get", "amdsmi_get_gpu_id"},
{"rsmi_dev_vendor_id_get", "amdsmi_get_gpu_asic_info"},
{"rsmi_dev_name_get", "amdsmi_get_gpu_board_info"},
{"rsmi_dev_sku_get", "amdsmi_get_gpu_board_info"},
{"rsmi_dev_brand_get", "amdsmi_get_gpu_asic_info"},
{"rsmi_dev_vendor_name_get", "amdsmi_get_gpu_vendor_name"},
{"rsmi_dev_serial_number_get", "amdsmi_get_gpu_asic_info"},
{"rsmi_dev_subsystem_id_get", "amdsmi_get_gpu_subsystem_id"},
{"rsmi_dev_subsystem_name_get", "amdsmi_get_gpu_subsystem_name"},
{"rsmi_dev_drm_render_minor_get", "amdsmi_get_gpu_drm_render_minor"},
{"rsmi_dev_subsystem_vendor_id_get", "amdsmi_get_gpu_asic_info"},
{"rsmi_dev_unique_id_get", "amdsmi_get_gpu_board_info"},
{"rsmi_dev_pci_bandwidth_get", "amdsmi_get_gpu_pci_bandwidth"},
{"rsmi_dev_pci_id_get", "amdsmi_get_gpu_pci_id"},
{"rsmi_dev_pci_throughput_get", "amdsmi_get_gpu_pci_throughput"},
{"rsmi_dev_pci_replay_counter_get", " amdsmi_get_gpu_pci_replay_counter"},
{"rsmi_dev_pci_bandwidth_set", " amdsmi_set_gpu_pci_bandwidth"},
{"rsmi_dev_power_profile_set", " amdsmi_set_gpu_power_profile"},
{"rsmi_dev_memory_busy_percent_get", "amdsmi_get_gpu_memory_busy_percent"},
{"rsmi_dev_busy_percent_get", "amdsmi_get_busy_percent"},
{"rsmi_dev_memory_reserved_pages_get", "amdsmi_get_gpu_memory_reserved_pages"},
{"rsmi_dev_overdrive_level_get", "amdsmi_get_gpu_overdrive_level"},
{"rsmi_dev_power_profile_presets_get", " amdsmi_get_gpu_power_profile_presets"},
{"rsmi_dev_perf_level_set", " amdsmi_set_gpu_perf_level"},
{"rsmi_dev_perf_level_set_v1", " amdsmi_set_gpu_perf_level_v1"},
{"rsmi_dev_perf_level_get", "amdsmi_get_gpu_perf_level"},
{"rsmi_perf_determinism_mode_set", "amdsmi_set_gpu_perf_determinism_mode"},
{"rsmi_dev_overdrive_level_set", " amdsmi_set_gpu_overdrive_level"},
{"rsmi_dev_vbios_version_get", "amdsmi_get_gpu_vbios_info"},
{"rsmi_dev_od_volt_info_get", " amdsmi_get_gpu_od_volt_info"},
{"rsmi_dev_od_volt_info_set", " amdsmi_set_gpu_od_volt_info"},
{"rsmi_dev_od_volt_curve_regions_get", " amdsmi_get_gpu_od_volt_curve_regions"},
{"rsmi_dev_ecc_enabled_get", " amdsmi_get_gpu_ecc_enabled"},
{"rsmi_dev_ecc_status_get", " amdsmi_get_gpu_ecc_status"},
{"rsmi_dev_counter_group_supported", "amdsmi_gpu_counter_group_supported"},
{"rsmi_dev_counter_create", "amdsmi_gpu_create_counter"},
{"rsmi_dev_xgmi_error_status", "amdsmi_gpu_xgmi_error_status"},
{"rsmi_dev_xgmi_error_reset", "amdsmi_reset_gpu_xgmi_error"},
{"rsmi_dev_memory_reserved_pages_get", "amdsmi_get_gpu_memory_reserved_pages"},
{"rsmi_topo_numa_affinity_get", "amdsmi_get_gpu_topo_numa_affinity"},
{"rsmi_dev_gpu_metrics_info_get", " amdsmi_get_gpu_metrics_info"},
{"rsmi_dev_gpu_reset", "amdsmi_reset_gpu"},
{"rsmi_dev_memory_total_get", "amdsmi_get_gpu_memory_total"},
{"rsmi_dev_memory_usage_get", "amdsmi_get_gpu_memory_usage"},
{"rsmi_dev_gpu_clk_freq_get", " amdsmi_get_clk_freq"},
{"rsmi_dev_gpu_clk_freq_set", " amdsmi_set_clk_freq"},
{"rsmi_dev_firmware_version_get", "amdsmi_get_fw_info"},
{"rsmi_dev_ecc_count_get", " amdsmi_get_gpu_ecc_count"},
{"rsmi_counter_available_counters_get", " amdsmi_get_gpu_available_counters"},
{"rsmi_dev_power_cap_get", "amdsmi_get_power_cap_info"},
{"rsmi_dev_power_cap_default_get", "amdsmi_get_power_cap_info"},
{"rsmi_dev_power_cap_range_get", "amdsmi_get_power_cap_info"},
{"rsmi_dev_power_cap_set", " amdsmi_set_power_cap"},
{"rsmi_dev_fan_rpms_get", "amdsmi_get_gpu_fan_rpms"},
{"rsmi_dev_fan_speed_get", "amdsmi_get_gpu_fan_speed"},
{"rsmi_dev_fan_speed_max_get", "amdsmi_get_gpu_fan_speed_max"},
{"rsmi_dev_temp_metric_get", " amdsmi_get_temp_metric"},
{"rsmi_dev_fan_reset", "amdsmi_reset_gpu_fan"},
{"rsmi_dev_fan_speed_set", "amdsmi_set_gpu_fan_speed"},
{"rsmi_dev_volt_metric_get", " amdsmi_get_gpu_volt_metric"}
};
auto rocm_func_handle =
reinterpret_cast<rsmi_func_id_iter_handle_t>(handle);
auto r = rsmi_func_iter_value_get(
rocm_func_handle,
reinterpret_cast<rsmi_func_id_value_t*>(value));
if ( r != RSMI_STATUS_SUCCESS )
return amd::smi::rsmi_to_amdsmi_status(r);
// Only change the function name, FUNC_ITER == 0
if (rocm_func_handle->id_type != 0)
return amd::smi::rsmi_to_amdsmi_status(r);
auto iter = rsmi_2_amdsmi.find(value->name);
if (iter != rsmi_2_amdsmi.end()) {
value->name = (*iter).second;
}
return amd::smi::rsmi_to_amdsmi_status(r);
}
amdsmi_status_t
amdsmi_get_gpu_compute_process_info(amdsmi_process_info_t *procs, uint32_t *num_items) {
AMDSMI_CHECK_INIT();
@@ -947,16 +808,6 @@ amdsmi_status_t amdsmi_get_gpu_ecc_status(amdsmi_processor_handle processor_han
reinterpret_cast<rsmi_ras_err_state_t*>(state));
}
amdsmi_status_t
amdsmi_get_busy_percent(amdsmi_processor_handle processor_handle,
uint32_t *busy_percent) {
AMDSMI_CHECK_INIT();
if (busy_percent == nullptr)
return AMDSMI_STATUS_INVAL;
return rsmi_wrapper(rsmi_dev_busy_percent_get, processor_handle,
busy_percent);
}
amdsmi_status_t amdsmi_get_gpu_metrics_info(
amdsmi_processor_handle processor_handle,
amdsmi_gpu_metrics_t *pgpu_metrics) {
@@ -1076,13 +927,6 @@ amdsmi_status_t amdsmi_get_gpu_perf_level(amdsmi_processor_handle processor_hand
amdsmi_status_t
amdsmi_set_gpu_perf_level(amdsmi_processor_handle processor_handle,
amdsmi_dev_perf_level_t perf_lvl) {
return rsmi_wrapper(rsmi_dev_perf_level_set, processor_handle,
static_cast<rsmi_dev_perf_level_t>(perf_lvl));
}
amdsmi_status_t
amdsmi_set_gpu_perf_level_v1(amdsmi_processor_handle processor_handle,
amdsmi_dev_perf_level_t perf_lvl) {
return rsmi_wrapper(rsmi_dev_perf_level_set_v1, processor_handle,
static_cast<rsmi_dev_perf_level_t>(perf_lvl));
}
@@ -1186,7 +1030,7 @@ amdsmi_status_t amdsmi_get_gpu_overdrive_level(
amdsmi_status_t amdsmi_set_gpu_overdrive_level(
amdsmi_processor_handle processor_handle, uint32_t od) {
return rsmi_wrapper(rsmi_dev_overdrive_level_set, processor_handle, od);
return rsmi_wrapper(rsmi_dev_overdrive_level_set_v1, processor_handle, od);
}
amdsmi_status_t amdsmi_get_gpu_pci_replay_counter(
amdsmi_processor_handle processor_handle, uint64_t *counter) {
@@ -1245,13 +1089,6 @@ amdsmi_status_t amdsmi_set_gpu_clk_range(amdsmi_processor_handle processor_handl
static_cast<rsmi_clk_type_t>(clkType));
}
amdsmi_status_t amdsmi_set_gpu_overdrive_level_v1(
amdsmi_processor_handle processor_handle,
uint32_t od) {
return rsmi_wrapper(rsmi_dev_overdrive_level_set_v1, processor_handle,
od);
}
amdsmi_status_t amdsmi_reset_gpu(amdsmi_processor_handle processor_handle) {
return rsmi_wrapper(rsmi_dev_gpu_reset, processor_handle);
}
@@ -1264,12 +1101,6 @@ amdsmi_status_t amdsmi_get_utilization_count(amdsmi_processor_handle processor_h
reinterpret_cast<rsmi_utilization_counter_t*>(utilization_counters),
count, timestamp);
}
amdsmi_status_t amdsmi_get_gpu_memory_busy_percent(
amdsmi_processor_handle processor_handle,
uint32_t *busy_percent) {
return rsmi_wrapper(rsmi_dev_memory_busy_percent_get, processor_handle,
busy_percent);
}
amdsmi_status_t amdsmi_get_energy_count(amdsmi_processor_handle processor_handle,
uint64_t *power, float *counter_resolution, uint64_t *timestamp) {
@@ -1277,13 +1108,7 @@ amdsmi_status_t amdsmi_get_energy_count(amdsmi_processor_handle processor_handle
power, counter_resolution, timestamp);
}
amdsmi_status_t amdsmi_get_gpu_drm_render_minor(
amdsmi_processor_handle processor_handle, uint32_t *minor) {
return rsmi_wrapper(rsmi_dev_drm_render_minor_get, processor_handle,
minor);
}
amdsmi_status_t amdsmi_get_gpu_pci_id(
amdsmi_status_t admsmi_get_gpu_bdf_id(
amdsmi_processor_handle processor_handle, uint64_t *bdfid) {
return rsmi_wrapper(rsmi_dev_pci_id_get, processor_handle,
bdfid);
@@ -1295,7 +1120,7 @@ amdsmi_status_t amdsmi_get_gpu_topo_numa_affinity(
numa_node);
}
amdsmi_status_t amdsmi_get_version(amdsmi_version_t *version) {
amdsmi_status_t amdsmi_get_lib_version(amdsmi_version_t *version) {
AMDSMI_CHECK_INIT();
if (version == nullptr)
@@ -1306,19 +1131,6 @@ amdsmi_status_t amdsmi_get_version(amdsmi_version_t *version) {
return amd::smi::rsmi_to_amdsmi_status(rstatus);
}
amdsmi_status_t amdsmi_get_version_str(amdsmi_sw_component_t component,
char *ver_str,
uint32_t len) {
AMDSMI_CHECK_INIT();
if (ver_str == nullptr)
return AMDSMI_STATUS_INVAL;
auto status = rsmi_version_str_get(
static_cast<rsmi_sw_component_t>(component), ver_str, len);
return amd::smi::rsmi_to_amdsmi_status(status);
}
amdsmi_status_t
amdsmi_get_gpu_vbios_info(amdsmi_processor_handle processor_handle, amdsmi_vbios_info_t *info) {
AMDSMI_CHECK_INIT();
@@ -1491,7 +1303,7 @@ amdsmi_get_gpu_bad_page_info(amdsmi_processor_handle processor_handle, uint32_t
}
amdsmi_status_t
amdsmi_get_gpu_ecc_error_count(amdsmi_processor_handle processor_handle, amdsmi_error_count_t *ec) {
amdsmi_get_gpu_total_ecc_count(amdsmi_processor_handle processor_handle, amdsmi_error_count_t *ec) {
AMDSMI_CHECK_INIT();
if (ec == nullptr) {
tests/amd_smi_test/functional/api_support_read.cc
-89
View File
@@ -96,93 +96,4 @@ void TestAPISupportRead::Run(void) {
return;
}
amdsmi_func_id_iter_handle_t iter_handle, var_iter, sub_var_iter;
amdsmi_func_id_value_t value;
for (uint32_t x = 0; x < num_iterations(); ++x) {
for (uint32_t i = 0; i < num_monitor_devs(); ++i) {
IF_VERB(STANDARD) {
PrintDeviceHeader(processor_handles_[i]);
std::cout << "Supported AMDSMI Functions:" << std::endl;
std::cout << "\tVariants (Monitors)" << std::endl;
}
err = amdsmi_open_supported_func_iterator(processor_handles_[i], &iter_handle);
CHK_ERR_ASRT(err)
while (1) {
err = amdsmi_get_func_iter_value(iter_handle, &value);
CHK_ERR_ASRT(err)
IF_VERB(STANDARD) {
std::cout << "Function Name: " << value.name << std::endl;
}
err = amdsmi_open_supported_variant_iterator(iter_handle, &var_iter);
if (err != AMDSMI_STATUS_NO_DATA) {
CHK_ERR_ASRT(err)
IF_VERB(STANDARD) {
std::cout << "\tVariants/Monitors: ";
}
while (1) {
err = amdsmi_get_func_iter_value(var_iter, &value);
CHK_ERR_ASRT(err)
IF_VERB(STANDARD) {
if (value.id == AMDSMI_DEFAULT_VARIANT) {
std::cout << "Default Variant ";
} else {
std::cout << value.id;
}
std::cout << " (";
}
err =
amdsmi_open_supported_variant_iterator(var_iter, &sub_var_iter);
if (err != AMDSMI_STATUS_NO_DATA) {
CHK_ERR_ASRT(err)
while (1) {
err = amdsmi_get_func_iter_value(sub_var_iter, &value);
CHK_ERR_ASRT(err)
IF_VERB(STANDARD) {
std::cout << value.id << ", ";
}
err = amdsmi_next_func_iter(sub_var_iter);
if (err == AMDSMI_STATUS_NO_DATA) {
break;
}
CHK_ERR_ASRT(err)
}
err = amdsmi_close_supported_func_iterator(&sub_var_iter);
CHK_ERR_ASRT(err)
}
IF_VERB(STANDARD) {
std::cout << "), ";
}
err = amdsmi_next_func_iter(var_iter);
if (err == AMDSMI_STATUS_NO_DATA) {
break;
}
CHK_ERR_ASRT(err)
}
IF_VERB(STANDARD) {
std::cout << std::endl;
}
err = amdsmi_close_supported_func_iterator(&var_iter);
CHK_ERR_ASRT(err)
}
err = amdsmi_next_func_iter(iter_handle);
if (err == AMDSMI_STATUS_NO_DATA) {
break;
}
CHK_ERR_ASRT(err)
// err = amdsmi_open_supported_variant_iterator(iter_handle, &var_iter);
//
}
err = amdsmi_close_supported_func_iterator(&iter_handle);
CHK_ERR_ASRT(err)
}
}
}
tests/amd_smi_test/functional/gpu_busy_read.cc
-23
View File
@@ -94,27 +94,4 @@ void TestGPUBusyRead::Run(void) {
std::cout << "** SetUp Failed for this test. Skipping.**" << std::endl;
return;
}
for (uint32_t x = 0; x < num_iterations(); ++x) {
for (uint32_t i = 0; i < num_monitor_devs(); ++i) {
PrintDeviceHeader(processor_handles_[i]);
err = amdsmi_get_busy_percent(processor_handles_[i], &val_ui32);
if (err != AMDSMI_STATUS_SUCCESS) {
if (err == AMDSMI_STATUS_FILE_ERROR) {
IF_VERB(STANDARD) {
std::cout << "\t**GPU Busy Percent: Not supported on this machine"
<< std::endl;
}
} else {
CHK_ERR_ASRT(err)
}
} else {
IF_VERB(STANDARD) {
std::cout << "\t**GPU Busy Percent (Percent Idle):" << std::dec <<
val_ui32 << " (" << 100 - val_ui32 << ")" << std::endl;
}
}
}
}
}
tests/amd_smi_test/functional/id_info_read.cc
+2 -16
View File
@@ -145,20 +145,6 @@ void TestIdInfoRead::Run(void) {
}
}
err = amdsmi_get_gpu_drm_render_minor(processor_handles_[i], &drm_render_minor);
if (err == AMDSMI_STATUS_NOT_SUPPORTED) {
// Verify api support checking functionality is working
err = amdsmi_get_gpu_drm_render_minor(processor_handles_[i], nullptr);
ASSERT_EQ(err, AMDSMI_STATUS_NOT_SUPPORTED);
} else {
CHK_ERR_ASRT(err)
IF_VERB(STANDARD) {
std::cout << "\t**DRM Render Minor: " << drm_render_minor << std::endl;
}
// Verify api support checking functionality is working
err = amdsmi_get_gpu_drm_render_minor(processor_handles_[i], nullptr);
ASSERT_EQ(err, AMDSMI_STATUS_INVAL);
}
err = amdsmi_get_gpu_vendor_name(processor_handles_[i], buffer, kBufferLen);
if (err == AMDSMI_STATUS_NOT_SUPPORTED) {
std::cout << "\t**Device Vendor name string not found on this system." <<
@@ -231,7 +217,7 @@ void TestIdInfoRead::Run(void) {
ASSERT_EQ(err, AMDSMI_STATUS_INVAL);
}
err = amdsmi_get_gpu_pci_id(processor_handles_[i], &val_ui64);
err = admsmi_get_gpu_bdf_id(processor_handles_[i], &val_ui64);
// Don't check for AMDSMI_STATUS_NOT_SUPPORTED since this should always be
// supported. It is not based on a sysfs file.
CHK_ERR_ASRT(err)
@@ -240,7 +226,7 @@ void TestIdInfoRead::Run(void) {
std::cout << " (" << std::dec << val_ui64 << ")" << std::endl;
}
// Verify api support checking functionality is working
err = amdsmi_get_gpu_pci_id(processor_handles_[i], nullptr);
err = admsmi_get_gpu_bdf_id(processor_handles_[i], nullptr);
ASSERT_EQ(err, AMDSMI_STATUS_INVAL);
}
}
tests/amd_smi_test/functional/mem_util_read.cc
-11
View File
@@ -118,17 +118,6 @@ void TestMemUtilRead::Run(void) {
for (uint32_t i = 0; i < num_monitor_devs(); ++i) {
PrintDeviceHeader(processor_handles_[i]);
#if 0
err = amdsmi_get_gpu_memory_busy_percent(i, &mem_busy_percent);
err_chk("amdsmi_get_gpu_memory_busy_percent()");
if (err != AMDSMI_STATUS_SUCCESS) {
return;
}
IF_VERB(STANDARD) {
std::cout << "\t**" << "GPU Memory Busy %: " << mem_busy_percent <<
std::endl;
}
#endif
for (uint32_t mem_type = AMDSMI_MEM_TYPE_FIRST;
mem_type <= AMDSMI_MEM_TYPE_LAST; ++mem_type) {
err = amdsmi_get_gpu_memory_total(processor_handles_[i],
tests/amd_smi_test/functional/mutual_exclusion.cc
+2 -7
View File
@@ -215,7 +215,7 @@ void TestMutualExclusion::Run(void) {
CHECK_RET(ret, AMDSMI_STATUS_BUSY);
ret = amdsmi_get_gpu_subsystem_id(processor_handles_[0], &dmy_ui16);
CHECK_RET(ret, AMDSMI_STATUS_BUSY);
ret = amdsmi_get_gpu_pci_id(processor_handles_[0], &dmy_ui64);
ret = admsmi_get_gpu_bdf_id(processor_handles_[0], &dmy_ui64);
CHECK_RET(ret, AMDSMI_STATUS_BUSY);
ret = amdsmi_get_gpu_pci_throughput(processor_handles_[0], &dmy_ui64, &dmy_ui64, &dmy_ui64);
CHECK_RET(ret, AMDSMI_STATUS_BUSY);
@@ -245,8 +245,6 @@ void TestMutualExclusion::Run(void) {
CHECK_RET(ret, AMDSMI_STATUS_BUSY);
ret = amdsmi_get_gpu_od_volt_curve_regions(processor_handles_[0], &dmy_ui32, &dmy_vlt_reg);
CHECK_RET(ret, AMDSMI_STATUS_BUSY);
ret = amdsmi_set_gpu_overdrive_level_v1(processor_handles_[0], dmy_i32);
CHECK_RET(ret, AMDSMI_STATUS_BUSY);
ret = amdsmi_set_clk_freq(processor_handles_[0], CLK_TYPE_SYS, 0);
CHECK_RET(ret, AMDSMI_STATUS_BUSY);
ret = amdsmi_get_gpu_ecc_count(processor_handles_[0], AMDSMI_GPU_BLOCK_UMC, &dmy_err_cnt);
@@ -258,7 +256,6 @@ void TestMutualExclusion::Run(void) {
/* Other functions holding device mutexes. Listed for reference.
amdsmi_dev_sku_get
amdsmi_set_gpu_perf_level_v1
amdsmi_set_gpu_od_clk_info
amdsmi_set_gpu_od_volt_info
amdsmi_dev_firmware_version_get
@@ -267,7 +264,6 @@ void TestMutualExclusion::Run(void) {
amdsmi_dev_brand_get
amdsmi_get_gpu_vram_vendor
amdsmi_get_gpu_subsystem_name
amdsmi_get_gpu_drm_render_minor
amdsmi_get_gpu_vendor_name
amdsmi_get_gpu_pci_bandwidth
amdsmi_set_gpu_pci_bandwidth
@@ -283,6 +279,7 @@ void TestMutualExclusion::Run(void) {
amdsmi_get_gpu_metrics_info
amdsmi_get_gpu_od_volt_curve_regions
amdsmi_dev_power_max_get
amdsmi_get_power_ave
amdsmi_dev_power_cap_get
amdsmi_dev_power_cap_range_get
amdsmi_set_power_cap
@@ -290,8 +287,6 @@ void TestMutualExclusion::Run(void) {
amdsmi_set_gpu_power_profile
amdsmi_get_gpu_memory_total
amdsmi_get_gpu_memory_usage
amdsmi_get_gpu_memory_busy_percent
amdsmi_get_busy_percent
amdsmi_dev_vbios_version_get
amdsmi_dev_serial_number_get
amdsmi_get_gpu_pci_replay_counter
tests/amd_smi_test/functional/sys_info_read.cc
+3 -3
View File
@@ -128,14 +128,14 @@ void TestSysInfoRead::Run(void) {
}
}
err = amdsmi_get_gpu_pci_id(processor_handles_[i], &val_ui64);
err = admsmi_get_gpu_bdf_id(processor_handles_[i], &val_ui64);
CHK_ERR_ASRT(err)
IF_VERB(STANDARD) {
std::cout << "\t**PCI ID (BDFID): 0x" << std::hex << val_ui64;
std::cout << " (" << std::dec << val_ui64 << ")" << std::endl;
}
// Verify api support checking functionality is working
err = amdsmi_get_gpu_pci_id(processor_handles_[i], nullptr);
err = admsmi_get_gpu_bdf_id(processor_handles_[i], nullptr);
ASSERT_EQ(err, AMDSMI_STATUS_INVAL);
err = amdsmi_get_gpu_topo_numa_affinity(processor_handles_[i], &val_ui32);
@@ -177,7 +177,7 @@ void TestSysInfoRead::Run(void) {
}
}
err = amdsmi_get_version(&ver);
err = amdsmi_get_lib_version(&ver);
CHK_ERR_ASRT(err)
ASSERT_TRUE(ver.major != 0xFFFFFFFF && ver.minor != 0xFFFFFFFF &&
tests/amd_smi_test/functional/version_read.cc
+1 -19
View File
@@ -86,11 +86,6 @@ void TestVersionRead::Close() {
static const uint32_t kVerMaxStrLen = 80;
static const std::map<uint32_t, const char *>
kComponentNameMap = {
{AMDSMI_SW_COMP_DRIVER, "Driver Version"},
};
void TestVersionRead::Run(void) {
amdsmi_status_t err;
amdsmi_version_t ver = {0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, nullptr};
@@ -101,7 +96,7 @@ void TestVersionRead::Run(void) {
return;
}
err = amdsmi_get_version(&ver);
err = amdsmi_get_lib_version(&ver);
CHK_ERR_ASRT(err)
ASSERT_TRUE(ver.major != 0xFFFFFFFF && ver.minor != 0xFFFFFFFF &&
@@ -110,17 +105,4 @@ void TestVersionRead::Run(void) {
std::cout << "\t**AMD SMI Library version: " << ver.major << "." <<
ver.minor << "." << ver.patch << " (" << ver.build << ")" << std::endl;
}
char ver_str[kVerMaxStrLen];
for (uint32_t cmp = AMDSMI_SW_COMP_FIRST; cmp <= AMDSMI_SW_COMP_LAST; ++cmp) {
err = amdsmi_get_version_str(static_cast<amdsmi_sw_component_t>(cmp),
ver_str, kVerMaxStrLen);
CHK_ERR_ASRT(err)
IF_VERB(STANDARD) {
std::cout << "\t**" << kComponentNameMap.at(cmp) << ": " <<
ver_str << std::endl;
}
}
}
tests/amd_smi_test/test_common.h
+1 -1
View File
@@ -77,7 +77,7 @@ void DumpMonitorInfo(const TestBase *test);
if (RET != AMDSMI_STATUS_SUCCESS) { \
const char *err_str; \
std::cout << "\t===> ERROR: AMDSMI call returned " << (RET) << std::endl; \
amdsmi_status_string((RET), &err_str); \
amdsmi_status_code_to_string((RET), &err_str); \
std::cout << "\t===> (" << err_str << ")" << std::endl; \
std::cout << "\t===> at " << __FILE__ << ":" << std::dec << __LINE__ << \
std::endl; \