Commit Graph

8920 Commits

Author SHA1 Message Date
Jose Santos 6db524e5fe Update test_analyze_workloads.py
Signed-off-by: Jose Santos <josantos@t007-004.hpcfund>
2024-02-23 10:00:23 -06:00
Jose Santos 40c8efc2ff update to new Omniperf workloads
Signed-off-by: Jose Santos <josantos@t007-004.hpcfund>
2024-02-23 10:00:23 -06:00
Jose Santos 989e29cf81 Moving workloads to old
Signed-off-by: Jose Santos <josantos@login1.hpcfund>
2024-02-23 10:00:23 -06:00
Karl W Schulz 76873dd0c4 more code formatting updates
Signed-off-by: Karl W Schulz <karl.schulz@amd.com>
2024-02-22 16:02:25 -06:00
Karl W Schulz 5a27189c82 more code formatting updates
Signed-off-by: Karl W Schulz <karl.schulz@amd.com>
2024-02-22 16:02:25 -06:00
Karl W Schulz 79b877d679 code formatting updates
Signed-off-by: Karl W Schulz <karl.schulz@amd.com>
2024-02-22 16:02:25 -06:00
Jonathan R. Madsen 7ce263b0e4 Update rocprofiler-register support
- add rocprofiler-register to CPACK_DEBIAN_BINARY_PACKAGE_DEPENDS when found
- add rocprofiler-register to CPACK_RPM_BINARY_PACKAGE_REQUIRES when found
- remove report_tool_load_failures_explicit_
- add HSA_TOOLS_DISABLE_REGISTER flag
- add HSA_TOOLS_REPORT_REGISTER_FAILURE
- use HSA_TOOLS_REPORT_REGISTER_FAILURE instead of HSA_TOOLS_REPORT_LOAD_FAILURE
- changed rocprofiler-register message to not include the word "error"

Change-Id: Ib7fd7f14c42758a54c347874018281bb1b5477a6
2024-02-22 11:55:25 -05:00
Giovanni LB fe00badc6e Adding late-replace envvar filenames to ATT
Change-Id: I5df51934ffa25358503013e1a230adb894d6ff94
2024-02-22 10:02:32 -05:00
Shweta Khatri 24633c7a85 Avoid releasing scratch for blit queues
At hsa_shutdown(), scratch_lock_ may be gone. Blit queues don't need it.

Change-Id: Ic132ac8a6be31fb2f0623137115608b0b222f077
2024-02-22 14:12:05 +00:00
Jonathan Kim 1f63ea3476 Fix export-close race during IPC attach request
If two attach requests to the same piece of shared memory occur,
a double export or premature dmabuf fd close can occur since the export
and close on demand calls are not atomic.

Use a reference counter on shared memory dmabuf FDs that have
already been opened to avoid this problem.

Change-Id: I14a59209c0385e32582af42a57b33b1c6838a9b1
2024-02-22 14:12:05 +00:00
David Yat Sin b77ade9c64 rocrtst: Add non-contiguous VMM map tests
Add rocrtst to test mapping non-contiguous memory to a
single VA range

Change-Id: Id2e57f83512f8b482456b2b1925586951ada7400
2024-02-22 14:12:05 +00:00
David Yat Sin 99e31e43aa rocrtst: Add test for GPU access to memory
Add test to verify whether GPU shaders can read memory created using VMM
APIs.
Split VMM rocrtst to two separate groups: Basic and Access tests

Change-Id: Iead8d46125580c71ccd582e967c8e2e891e75c5e
2024-02-22 14:12:05 +00:00
David Yat Sin 1f50219634 Fix compile error when using clang
Change-Id: Ibacf094934a9b489c052a18eeb6b26639aba3032
2024-02-22 14:12:05 +00:00
Gopesh Bhardwaj 2d71520953 Fixing opensuse compilation (#521)
* Fixing opensuse compilation

* source formatting (clang-format v11) (#526)

Co-authored-by: bgopesh <bgopesh@users.noreply.github.com>

* Update tests/tools/CMakeLists.txt

---------

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: bgopesh <bgopesh@users.noreply.github.com>
Co-authored-by: Jonathan R. Madsen <jrmadsen@users.noreply.github.com>
2024-02-22 01:10:23 -06:00
Jonathan R. Madsen 0d939edbba Updates/fixes for CI, docs, tests, samples, and common library (#528)
- .github/workflows/continuous_integration.yml
  - apt-get update before apt-get install
  - remove libgtest-dev
  - actions-comment-pull-request: v2.4.3 -> v2.5.0
- .github/workflows/formatting.yml
  - create-pull-request: v5 -> v6
- cmake/rocprofiler_options.cmake
  - remove unused ROCPROFILER_DEBUG_TRACE and ROCPROFILER_LD_AQLPROFILE options
- samples/counter_collection/callback_client.cpp
  - corr_id field renamed to correlation_id
- samples/counter_collection/client.cpp
  - corr_id field renamed to correlation_id
- include/rocprofiler-sdk/fwd.h
  - In rocprofiler_record_counter_t: rename corr_id field to correlation_id
  - doxygen fixes
- lib/common/utility.*
  - remove get_accurate_clock_id_impl
  - timestamp_ns() defaults to CLOCK_BOOTTIME
- lib/rocprofiler-sdk/counters/core.cpp
  - fix spelling mistake: extrenal -> external
  - corr_id field renamed to correlation_id
- lib/rocprofiler-sdk-tool/tool.cpp
  - fix destruction of static tool::output_file before finalization
- scripts/update-docs.sh
  - define PROJECT_NAME
- tests/async-copy-tracing/validate.py
  - init_time and fini_time checks
  - hip_api_traces, marker_api_tracing
- tests/common/serialization.hpp
  - fix save function for rocprofiler_record_counter_t following rename of corr_id to correlation_id
- tests/kernel-tracing/validate.py
  - init_time and fini_time checks
  - relax test_total_runtime range
- tests/rocprofv3/tracing/CMakeLists.txt
  - remove -M from rocprofv3-test-systrace-execute
  - exclude test_hsa_api_trace in rocprofv3-test-systrace-validate due to HIP API tracing
- tests/rocprofv3/tracing/validate.py
  - update test_kernel_trace to accept mangled or demangled
- tests/tools/json-tool.cpp
  - remove use of GLOG
  - include init_time and fini_time
  - write_json(...) function
2024-02-22 00:16:43 -06:00
Giovanni LB a1999fa366 SWDEV-446854: Fixing ATT crash on rocblas_sgemm
Change-Id: I932b92c4e188655500b1b150d9594ec663dd25c7
2024-02-20 14:55:27 -05:00
Galantsev, Dmitrii 6d5d9971c2 CMAKE - Find hsa-runtime64
Change-Id: Id877eb9cfcc61d81993a6a43703ef2e5f72e1e8f
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>
2024-02-19 23:49:38 -05:00
Benjamin Welton 7adffd5b22 Add rocprofiler_query_counter_info function (#452)
* Add rocprofiler_query_counter_info function

Replaces rocprofiler_query_counter_name. Allows for
querying other types of info from counters (such as
description) and gives us some flexibility to add
return data in the near future (if we have to).

* source formatting (clang-format v11) (#453)

Co-authored-by: bwelton <bwelton@users.noreply.github.com>

* Updated version fetching

* source formatting (clang-format v11) (#509)

Co-authored-by: bwelton <bwelton@users.noreply.github.com>

* Merged

---------

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: bwelton <bwelton@users.noreply.github.com>
2024-02-19 16:05:38 -08:00
Benjamin Welton 3638351b4c Callback based handler for counter collection (#506)
* Callback based handler for counter collection

* source formatting (clang-format v11) (#507)

Co-authored-by: bwelton <bwelton@users.noreply.github.com>

* cmake formatting (cmake-format) (#508)

Co-authored-by: bwelton <bwelton@users.noreply.github.com>

* Doc fix

* Minor doc fix

* More doc fixes

* More doc fixes

* More doc fixes

* Update CI

* Changes to the API per comments

* Mutex exception for HSA

* source formatting (clang-format v11) (#511)

Co-authored-by: bwelton <bwelton@users.noreply.github.com>

* Doc fix

---------

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: bwelton <bwelton@users.noreply.github.com>
2024-02-19 15:55:21 -08:00
Charis Poag 93ed5205f9 Merge amd-staging into amd-master 20240216
Change-Id: Id3e41507ab6143d08cb052710aa19c6f2e402fed
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
2024-02-16 20:03:19 -06:00
Karl W Schulz 4bcbc73080 enable formatting for 2.x branch
Signed-off-by: Karl W Schulz <karl.schulz@amd.com>
2024-02-16 15:13:51 -06:00
Karl W Schulz a05cb9757b remove non-used commented out bits from cmake config
Signed-off-by: Karl W Schulz <karl.schulz@amd.com>
2024-02-16 10:33:12 -06:00
Karl W Schulz 18251da534 adding to project markers
Signed-off-by: Karl W Schulz <karl.schulz@amd.com>
2024-02-16 10:33:12 -06:00
Karl W Schulz 98fd762d9e cmake updates for test_analyze_commands; map tests to newly defined
pytest markers and define a new "analyze_commands" label to support
parallel execution

Signed-off-by: Karl W Schulz <karl.schulz@amd.com>
2024-02-16 10:33:12 -06:00
Karl W Schulz fafff47822 demarcate analysis tests into groups
Signed-off-by: Karl W Schulz <karl.schulz@amd.com>
2024-02-16 10:33:12 -06:00
Oliveira, Daniel b4d37caa70 fix: [rocm/rocm_smi_lib] rsmi_dev_activity_metric_get gfx/memory activity does not update with GPU activity
Checks and forces rereading gpu metrics unconditionally

Code changes related to the following:
  * Device::dev_log_gpu_metrics()
  * Examples
  * Unit tests

Change-Id: Ic1c4f34a39f2bf197263f80ddbb84da26345807d
Signed-off-by: Oliveira, Daniel <daniel.oliveira@amd.com>
2024-02-16 09:47:45 -06:00
Gopesh Bhardwaj c9dfa9e617 Fixing AFAR builds for some GNU tool chains (#498) 2024-02-15 08:08:37 -08:00
Oliveira, Daniel ce36198cb1 fix: [rocm/rocm_smi_lib] header cleanup Remove non-unified headers
Cleans up individual gpu metric APIs which will be implemented according to 'unified-headers' standards

Code changes related to the following:
  * 'rsmi_dev_metrics_' APIs
  * Functional tests
  * Examples

Change-Id: I7d562a95889361ee6f8f7588f8a790f42c8eb262
Signed-off-by: Oliveira, Daniel <daniel.oliveira@amd.com>
2024-02-14 17:50:26 -06:00
Manjunath-Jakaraddi 8d705f8b96 SWDEV-445654: Adding metrics to MI300
Change-Id: I7c3d78d8739ad219fc05bd99d7fc11d1f023aadc
2024-02-14 13:17:15 -05:00
Giovanni LB 95b49ca593 Adding bank conflict and other MI300 missing metrics to rocprofv2
Change-Id: Ibb7fb6163aa3bd7742f199a03600c737720e3c61
2024-02-14 13:15:54 -05:00
Nick Curtis 852cc13f2a Fix clocks (#246)
Signed-off-by: Nicholas Curtis <nicurtis@amd.com>
Co-authored-by: Nicholas Curtis <nicurtis@amd.com>
2024-02-13 20:22:11 -06:00
Nick Curtis c60f031558 Fix hbmBW calculation, (#243)
* Fix hbmBW calculation, currently we're reading the wrong value of the mclk out of rocm-smi

Signed-off-by: Nicholas Curtis <nicurtis@amd.com>

* Fix for earlier smi

Signed-off-by: Nicholas Curtis <nicurtis@amd.com>

---------

Signed-off-by: Nicholas Curtis <nicurtis@amd.com>
Co-authored-by: Nicholas Curtis <nicurtis@amd.com>
2024-02-13 20:19:56 -06:00
colramos-amd b7d5d3b64b Modify yaml configs to reflect new combined Top Stats
Signed-off-by: colramos-amd <colramos@amd.com>
2024-02-13 19:56:46 -06:00
colramos-amd bd83f68412 Enable the revamped Top Stats section in standalone GUI
Signed-off-by: colramos-amd <colramos@amd.com>
2024-02-13 19:18:31 -06:00
colramos-amd 2526e14802 Combine kernel and dispatch stats in the Top Stats section of analysis
Signed-off-by: colramos-amd <colramos@amd.com>
2024-02-13 19:17:41 -06:00
Charis Poag 4b5ccb57f0 [SWDEV-423481/SWDEV-423393] Align all device identifier details
Updated:
 * [CLI] Fixed vram % - printf style formatting causes many data errors
   This fix updates to the recommended way of outputting formatted data.
   https://docs.python.org/3/library/stdtypes.html#printf-style-string-formatting
 * [API/CLI] Added gpu_id / GUID from kfd (rsmi_dev_guid_get)
       -> CLI name: "GUID"
       -> ROCm SMI calls: no arg, -i, --showhw, --showproduct
 * [API/CLI] Added node_id from kfd (rsmi_dev_node_get)
       -> CLI name: "Node"
       -> ROCm SMI calls: no arg, --showhw, --showproduct
 * [CLI] Added target gfx version from kfd
       -> CLI name: "GFX Version" or "GFX VER"
       -> ROCm SMI calls: --showhw, --showproduct
 * [CLI] Base ROCm CLI
       -> Removed - stacked id formatting:
	   This is to simplify identifiers helpful to users.
	   More identifiers can be found on -i --showhw, --showproduct
 * [CLI] Update -i, --showhw, --showproduct, w/out arg
      -> Card ID/DID/Model/SKU/VBIOS:
            All unsupported values now display "N/A" instead
            of "unknown" or "unsupported"
 * [CLI] Showhw now expands data based on content

Change-Id: Ifb8586f9f545892b8a5aa7903608273cdd77e075
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
2024-02-13 19:52:29 -05:00
Vladimir Stempen 677433b367 Fix [Not supported] status for get_compute_process_info_by_pid
On some systems [rocm-smi --showpids] reports
get_compute_process_info_by_pid, Not supported on the given system
[PID] [PROCESS NAME] 1 UNKNOWN UNKNOWN UNKNOWN

get_compute_process_info_by_pid fails because cu_occupancy debugfs method
is not provided on some graphics cards and GFX revisions by design

Proposing a change to return success status when only cu_occupancy debugfs method
is not found and provide cu_occupancy invalidation value to mark only
this parameter as UNKNOWN

Change-Id: Iae37070d9bd19483b4e6c8ee24c7d9a4c92f00d7
Signed-off-by: Vladimir Stempen <Vladimir.Stempen@amd.com>
Reviewed-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>
2024-02-13 18:17:47 -05:00
Galantsev, Dmitrii 32806681ca SWDEV-444700 - CMAKE - Fix RUNPATH
These RUNPATH changes make it so libraries can be found without setting
LD_LIBRARY_PATH.

Mostly tested on installed RDC binaries and libraries. The
build binaries should also work.

Change-Id: Ifd908a5b61d24dfcbb1d08d21b4ee830156d8643
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>
2024-02-13 16:56:28 -06:00
David Yat Sin 5b28a1bc17 Fix compile error on certain gcc versions
Change-Id: I8a4fab76d1dcc576eb7706ab45fc786c0cab274a
2024-02-13 15:25:34 -05:00
David Yat Sin 7e159e6d76 Add query for memory properties
Change-Id: I07c084c56b15c499ec564860b2a514e909ab7ca4
2024-02-13 14:28:30 -05:00
David Yat Sin ae16b3e14e Use sysconf pagesize for system pagesize
Provided by user huanggyizhi on github
https://github.com/RadeonOpenCompute/ROCR-Runtime/pull/124

Change-Id: Ia03c45f7a869ae2c804accf8163f8ae36c20dd5a
2024-02-13 14:28:10 -05:00
Gopesh Bhardwaj 2b1c11a4f3 Adding test for demangling (#467)
* Adding test for demangling

* moving -M option to existing one

* minor typo in script and test correction
2024-02-13 11:09:27 -06:00
Galantsev, Dmitrii de9eaaac8c CMAKE - Default to lib instead of lib64
Change-Id: Ib21d41018b091d92c2ed408ff0c4d28e6a74c903
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>
2024-02-12 20:16:28 -06:00
colramos-amd 7bb211a465 Fix logic for roofline field in sysinfo
Signed-off-by: colramos-amd <colramos@amd.com>
2024-02-12 17:07:55 -06:00
Benjamin Welton e9e7fc8e3f Add GFX1010 to counter data (#466)
NAVI 10 support.
2024-02-12 13:42:13 -06:00
Galantsev, Dmitrii d03061823a Merge amd-staging into amd-master 20240212
Change-Id: I662f2a470446550ba8c612aa1e5be911d7f7489f
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>
2024-02-12 11:30:04 -06:00
gobhardw 28871e661a node id to be reported as gpu id
Change-Id: I237281b33cfb5bdad67b7d6e8b69e9a86d62ca5d
2024-02-12 12:26:41 +05:30
Benjamin Welton a1d95daf84 Pin version to unblock builds (#474)
Merged to unblock runners on other PRs.
2024-02-10 12:30:31 -08:00
Galantsev, Dmitrii 81e3a78b1f Remove unsupported rocprofiler metrics
Change-Id: If6cfbcbe018227c591733471ab203fc6675d50af
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>
2024-02-09 15:18:54 -06:00
colramos-amd 2d8fb19652 Add numPipes field to test workloads
Signed-off-by: colramos-amd <colramos@amd.com>
2024-02-09 12:56:11 -06:00