Updates:
* Fixed rsmi_dev_compute_partition_get
& rsmi_dev_nps_mode_get to properly check
for invalid arguments
* Updated compute partition & NPS mode tests
- Now properly confirms the invalid
argument is seen
- Spacing for multiple devices is added
to better see distinction between
separate device's tests (for verbose output)
- Changed expect to assert calls, so errors
are observed faster for test failures
- Fixed multiple device testing where a
variable should have been unset, but
having multiple devices caused it to
set
- Updated multiple device testing to iterate
accross all devices (previously returned,
instead of continuing checking support
after RSMI_STATUS_NOT_SUPPORTED detected)
- Fixed a few spelling errors & verbose output
Change-Id: Ieba9e5b46763c6cd880fbf27fcdf58be8ecbc683
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
Updates:
* Added RSMI_STATUS_SETTING_UNAVAILABLE for
rsmi_dev_compute_partition_set - gives users
better error output when attempting to set
compute partition to values not listed in
available_compute_partition SYSFS
* Updated python --setcomputepartition to
provide better output when receiving
RSMI_STATUS_SETTING_UNAVAILABLE
* Updated all test & example files to check for
RSMI_STATUS_SETTING_UNAVAILABLE when doing
rsmi_dev_compute_partition_set
Change-Id: Ida5d54880d9b9b6e4a0468cdb962fdc0c18d6257
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
Updates:
* Added rsmi_dev_compute_partition_reset & rsmi_dev_nps_mode_reset
* Added --resetcomputepartition and --resetnpsmode python smi calls
* Added temp data files rocmsmi_boot_compute_partition_<device num>
& rocmsmi_boot_nps_mode_partition_<device num>, writes UNKNOWN
if data cannot be read or device does not support
* Cleaned up NPS & compute API documentation
* Added creation and reading of API temp files (used in reset
functionality)
* Cleaned up output of rocm_smi_example
* Updated rocm_smi_example to check if running with sudo permission
before executing write API calls (cleans up erroneous output)
* Added template specialization for storing temp data, requires
specific rsmi_type_t enums (restrics what data can be stored)
* Added storage of temp data, if temp files do not exist
* Updated google tests for NPS & compute to include reset API calls
Change-Id: I69895a466b97107617e6dbb355737b84499a76c9
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
Updates:
* Added rsmi_dev_nps_mode_set and rsmi_dev_nps_mode_get
* Added ability to set multiple SYSFS files in debug build
* Added ability to see user's env variables set for debug build
* Added tests for rsmi_dev_nps_mode_set and rsmi_dev_nps_mode_get
* Added ability to restart AMD GPU driver, used in nps_mode_set
* Updated ROCm_SMI_Manual.pdf to include new APIs
* Added progress bar for long running python_smi_tools, used
in setting nps_mode if runs longer than .1 seconds
Change-Id: I6d61bedd28d7cba6aff432ad2d127ba741b7d15a
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
Original updates:
* Added .gitignore to help with future commits
* Updated/added copyrights on modified or added files
* Updated rocm_smi.h/.cc
- Added 3 new SMI API functions:
rsmi_dev_compute_partition_set &
rsmi_dev_compute_partition_get
- Added helpful maps/enums used in
new get/set compute_partition API calls
* Updated rocm_smi.py
- Added --showcomputepartition
- Added --setcomputepartition
- Fixed a few mistypes
* Updated rsmiBindings.py - added helpful class/dict/list
* Updated rocm_smi_example.cc
- Added helpful MACRO to detect if api is not supported.
- Added current_compute_partition set/get rocm lib calls
- Added helpful macro to discover future RSMI errors
- Commented out test_set_freq, was having permission issues
on a Navi21
* Updated rocm_smi_main.cc
- Added helpful map to debug API calls, left in for future use
- Added comment to better understand a non-class function returns
* Added computepartition_read_write.cc/.h
- Added get/set compute partition API test calls
- Confirmed on devices that do not support the API calls, tests pass
* Updated rocm_smi_test/main.cc
- Calls new compute partition gtests
Added following updates from review feedback:
* Updated rocm_smi.h/cc
- Removed C++ API calls, adding support for both C/C++
API calls could cause confusion and adds extra work for us
- rsmi_dev_compute_partition_get -> Fixed an edge case where
user gives a small buffer length size (smaller than data
received), but does not receive the partial buffer back.
google Tests are updated to reflect this find.
* Updated rocm_smi_example.cc
- Fixed test_set_freq, issue was that file was not writable.
We now indicate this warning, so prior errors make sense.
- General test code cleanup. Removed extra code,
by creating loops for tests.
* Updated rocm_smi_main.cc
- Moved and got rid of an external reference to a map used
for debugging RSMI enums, now is a const public reference.
* Updated rocm_smi.py
- Updated python code to identify NOT_SUPPORTED due to
(currently) only a few GPU support the feature
Change-Id: I4a567acbb59d6771fb64df08d19175fe3604fd1b
Devices with CPU XGMI iolink do not support PCIe peer access. Therefore,
they should not be reported as accessible links in the topology.
Signed-off-by: Alex Sierra <Alex.Sierra@amd.com>
Change-Id: I3ee51796945dc0966200dee03886510e8f1846b7
1. Memory allocated for handle was not deleted
when no variant, subvariant or supported function
was found
2. handle->func_id_iter address was set to 0
before delete[]
Signed-off-by: Divya Shikre <DivyaUday.Shikre@amd.com>
Change-Id: Iab50fdfbe03eec8e6fd0e84e03bd2c47e645b3d8
Show an optional debug log (RSMI_DEBUG_BITFIELD=2) to
the user in the following scenarios:
1. If more than one current frequency is found
2. If frequencies are not read in increasing order of
their value
If current frequency is not available, index for it is
set to -1, values will not have * next to it in the
output. This will also be handled in rocm_smi.py.
Signed-off-by: Divya Shikre <DivyaUday.Shikre@amd.com>
Change-Id: I477ec065f7513c8045d6392f12ef6cb835a6b8f6
This patch adds the following 4 missing GPU blocks to the SMI LIB:
-RSMI_GPU_BLOCK_MMHUB
-RSMI_GPU_BLOCK_PCIE_BIF
-RSMI_GPU_BLOCK_HDP
-RSMI_GPU_BLOCK_XGMI_WAFL
Signed-off-by: Ori Messinger <Ori.Messinger@amd.com>
Change-Id: Ia1ec6f53e195f4bf7b8f073d6bed4fdb6572e546
When an application call the library in a system without amdgpu,
it may always print out "rsmi_init() failed". Suppress the error
message in the library.
Change-Id: Ice63dd3a764b221a6935536bff1bfa6aa3e51a46
The (temperature == nullptr) check happens only when HBM temperature is retrieved.
This check needs to apply in other cases as well, hence moving this outside the HBM condition.
This should return RSMI_STATUS_INVALID_ARGS consistently in all cases when nullptr is passed through rsmitst.
Signed-off-by: Divya Shikre <DivyaUday.Shikre@amd.com>
Change-Id: Iea3cec75312a0a669c7da27e15e9782e6a885c5f
Driver mem fills in 0xFF for all for the metrices not supported for that ASIC.
So if 0xFF is detected, return RSMI_STATUS_NOT_SUPPORTED
Signed-off-by: Divya Shikre <DivyaUday.Shikre@amd.com>
Change-Id: I86a38148c7a288ea0db94893f685560eaac098ab
Driver mem fills in 0xFF for all for the metrices not supported for that ASIC.
So if 0xFF is detected, return RSMI_STATUS_NOT_SUPPORTED
Signed-off-by: Divya Shikre <DivyaUday.Shikre@amd.com>
Change-Id: Iacb6474486e3732f2aa824ff447c17f8243b65cd
Implements rsmi_is_p2p_accessible API.
The function returns True if P2P is possible between two nodes.
Signed-off-by: Ori Messinger <Ori.Messinger@amd.com>
Change-Id: Ic7316eebcec4480175c7ad04c21a42b2e1a4c454
rocm-smi --showproductname will not show "Card series" in its output if
product_name exported by Kernel is empty string. This has been raised a
regression by customer.
BUG: SWDEV-297228
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Change-Id: I9aae24778e2d3a30aa661d8f338278c1666590fb
Implement default GPU power cap functionality in the LIB.
Signed-off-by: Ori Messinger <Ori.Messinger@amd.com>
Change-Id: Ia6b3420beb0e4df5559c3e6d11d0667972590b53
The rsmi_dev_temp_metric_get() can also support the HBM
temperatures which is retrieved from gpu_metrics.
Change-Id: I96b979296e90cf881523627b41b1a02849676416
Previously, during the rsmi_init discovery process, the existence
of an hwmon# directory was used to distinguish between gpus nodes
and non-gpu nodes. This isn't reliable in some scenarios. Instead,
the existence of the vbios_version file is used as an
indicator that the node is indeed a gpu.
Change-Id: Icfbe5c42ed0970077b05f25c3d209308a31bec85
The purpose of this patch is to implement GPU reset functionality
in the LIB, and to call it from the rocm_smi python CLI.
Signed-off-by: Ori Messinger <Ori.Messinger@amd.com>
Change-Id: Iaf525f7016f8354a7fd93af0209ca2e97ef4fd56
Some rsmi apps fail without much explanation when
rsmi_init() fails. This patch hopes to provide some clues to
the reason for the failure.
Change-Id: Id51308dc327b9871d537dd3e709b677db4ef10bc
Previously, when a process holding a shared mutex was killed,
the next time an RSMI application was started, it would not be
able to obtain the mutex--the application would have to exit.
This fix uses pthread_mutexattr_setrobust() to detect this
situation and act accordingingly.
Also, add some missing, needed mutexes and move mutexes
closer to where the protect resource is used.
Change-Id: Icfdc3a246f4cfa3fd008e3f13472199abd76fd35
A special mode of operation to achieve minimal performance variation by letting
the user have the ability to provide the desired frequency to be set as the soft limit.
The user can control the entry and exit to the mode via rocm-smi a mechanism to
enter / exit performance determinism mode as below.
Enter performance determinism mode:
- hold a lock
- write performance_determinism to power_dpm_force_performance_level
- write input clk_freq to pp_dpm_sclk
- release lock
Exit performance determinism_mode:
- hold a lock
- write auto to power_dpm_force_performance_level
- release lock
Signed-off-by: Divya Shikre <DivyaUday.Shikre@amd.com>
Change-Id: Ia24e27954cdf1c4337ffc83d8948fbdfaf4552d2
WKS and server cards have an FRU with product information, so try to use
that for product name and product SKU if it exists.
Signed-off-by: Kent Russell <kent.russell@amd.com>
Change-Id: I40bbd3bf62f4cb02e96015ed1630112691cacbc3
Event bitmask in KFD SMI event is now replaced with event index in
the SMI event message. Sending a event bitmask, which was a 64-bit
field with only 1 bit set, was quite wasteful of memory and also
potentially limiting to 64 events. Instead the kernel would send
event index in the SMI event message. As a result, update the
KFD SMI event handling to expect the event index in the message.
Change-Id: I3e74620788d3c1f7c0bdaa69e9d9ab3d1aba2c92
Also:
* fix doxygen manual generation that was altered during
OAM refactor
* quiet some compile warnings.
Change-Id: I548a3cf00eb887bea3dbf58e362ca6dfe90bde28
Some systems have kfd sysfs properties entries that
are unreadable--for example, when a multi-gpu system is
dividing the gpus among containers, each container may
only be able to access certain gpus.
Previously, all kfd topology node properties entries were
assumed to be valid. Now, we check for readability before
declaring them "valid".
Fixes SWDEV-240169
Also:
* remove an assertion that would happen when no pcie
device identifier files are found on the system.
* fix cpplint issues
Change-Id: I74321b685159dd2628c890b33c39ad82988cb9dd
This corrects issues that arose after OAM reorganization.
It should address SWDEV-243294.
Also, fix some compile warnings that show up on RHEL.
Change-Id: Id14d444905da35cd7346bcfbcd82b6d0572708c4
A message is output in debug builds when pthread_unlock
returns an error. However, in most cases, it should return
EPERM. In fact, if it doesn't return EPERM, it is an
indication of a problem. This commit adjusts accordingly.
Change-Id: Ia5cad89aa6e68e79c1291ea21adffb0fa68f2300
Charis Poag
Bill(Shuzhou) Liu
Alex Sierra
Sreekant Somasekharan
Divya Shikre
Ori Messinger
Elena Sakhnovitch
Harish Kasiviswanathan
Chris Freehill
Kent Russell
Mukul Joshi