2070 Commits

Author SHA1 Message Date
xuchen-amd ed877433f3 [rocprofiler-compute][TUI] Updates and refactor. (#703) 2025-08-22 11:02:32 -04:00
David Yat Sin a1597a358a rocr: Expose flag to allocate uncached memory (#674)
Add new flag for clients to directly request uncached memory
2025-08-22 09:52:39 -04:00
Todd tiantuo Li bbf0d34b48 SWDEV-550231 - fix hipGetDriverEntryPoint_common (#651) 2025-08-21 18:57:16 -07:00
Venkateshwar Reddy Kandula a77f7d5770 [aqlprofile] set gpu agent name while registering agent (#687)
* set gpu agent name, which will be used for gfx minor for pm4factory

* Update pm4_factory.cpp

* EOF.

* addr comments

---------

Co-authored-by: Venkateshwar Reddy Kandula <venkateshwar.kandula1306@gmail.com>
2025-08-21 15:16:28 -05:00
Julia Jiang d741a88f39 SWDEV-525231 - Update changelog for 7.0 on develop (#647) 2025-08-21 15:59:54 -04:00
David Galiffi 847580dd9e Update minimum_cmake_required to match version used in CI (#679)
- Update minimum_cmake_required to match version used in CI
  - We should match the minimum version that we test against

- Ensure ".S" files are treated as assembly.
2025-08-21 15:56:47 -04:00
systems-assistant[bot] c7b9533836 [Docs] Update using-pc-sampling (#157) 2025-08-21 11:14:16 -04:00
David Yat Sin 87b348c51d rocr: Fix hsa_amd_pointer_info regression (#638)
Fix regression when hsa_amd_pointer_info is called on a pointer that was
allocated using non-VMM APIs. The helper function VMemoryPtrInfo should
return error when the address is not found so that PtrInfo does the
lookup via Thunk.
2025-08-21 10:25:50 -04:00
xuchen-amd 6bf5648c6d [rocprofiler-compute] Handle amd-smi rocm==6.4.2 json output issue. (#667) 2025-08-21 10:06:04 -04:00
Hadi Naeisseh b2857b5db9 SWDEV-543981 Part 2 This is a new branch to avoid the many errors in the previous PR due to migration (#672)
Co-authored-by: hnaeisse_amdeng <hadi.naeisseh@amd.com>
2025-08-21 09:06:57 -04:00
David Galiffi 31c9d3424c Update README.md (#640)
* Update README.md

* Update README.md
2025-08-20 17:39:49 -04:00
xuchen-amd 0bf66a519c [rocprofiler-compute][TUI] Restructure Performance Metrics (#232) 2025-08-20 17:00:54 -04:00
Joseph Macaranas f5ac5efd79 Language Runtime README Changes (#650)
- Renaming old `README.md` files to keep their information intact.
- Default `README.md` files will have the deprecation notice to be mirrored back into the individual repos.
- Change ROCR-Runtime mirroring to `develop` branch.
2025-08-20 15:48:55 -04:00
vedithal-amd ed53090c25 Join SQC_* csv files as well (#669) 2025-08-20 15:33:48 -04:00
David Galiffi d111e9a297 [rocprofiler-systems] Add Debian 12 workflows (#402)
* Create CI dockers for debian 12

* Create Debian workflow

Signed-off-by: David Galiffi <David.Galiffi@amd.com>

* Fixing typo

* Apply suggestions from code review

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* Update default value for script's "VERSIONS" variable

* Fix Docker build warnings

LegacyKeyValueFormat: "ENV key=value" should be used instead of legacy "ENV key value" format

* Refactored the check for `pip install --break-system-packages`

---------

Signed-off-by: David Galiffi <David.Galiffi@amd.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2025-08-20 14:58:49 -04:00
systems-assistant[bot] 4d5db26646 SWDEV-549707 - Fixed memleak in hipMemImportFromShareableHandle (#442)
Co-authored-by: Ioannis Assiouras <Ioannis.Assiouras@amd.com>
2025-08-20 19:28:59 +01:00
systems-assistant[bot] 621da5410a SWDEV-465041 - Avoid wait in device enqueue (#443)
If we have PCIE atomics then we can avoid workaround in the scheduler, which requires an explicit wait on CPU
2025-08-20 12:46:47 -04:00
cfallows-amd 8bc0063f2e Update rocprofiler-compute workflows and CODEOWNERS (#221)
Update super repo codeowners- add Pratik to any doc changes for rocprofiler-compute.
Remove rocprofiler-compute-mi-rhel9.yml from super repo workflow folder- unused and out of date, coverage is already in RHEL8 workflow.
Remove CODEOWNERS, dependabot.yml, and all workflows from the projects/rocprofiler-compute/ project- confirmed these do not do anything anymore after being copied into super repo's main ./github/ section.

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>

* Add a branch filter for push events

* Update .github/CODEOWNERS

Use reviewers team instead of using the individual profiles.

Co-authored-by: David Galiffi <David.Galiffi@amd.com>

Update owners section for rocprofiler-compute codeowners, fix spacing

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>

---------

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>
Co-authored-by: David Galiffi <David.Galiffi@amd.com>
2025-08-20 11:17:25 -04:00
Danylo Lytovchenko f7338717ae SWDEV-470698 - fix formatting, add format check workflow (#657) 2025-08-20 19:58:06 +05:30
jamessiddeley-amd 5840940caa [rocprof-compute] Generalize Roofline (#325)
* per kernel analysis Roofline

* added per-kernel eval_metric calculation with display

* fixed typo

* updated tty.py show_all()

* formatting

* fixed ctest failures and updated equations

* formatting

* updated metric descriptoins

* review tweaks

* update docs

* added roofline gui analysis

* updated GUI docs

* updated print statement

* comment tweaks and ran ruff formatting
2025-08-20 09:58:08 -04:00
Rahul Manocha f4158ed3b7 SWDEV-546295 - fix param conversions for hipMemcpy3DBatchAsync (#641)
Co-authored-by: Rahul Manocha <rmanocha@amd.com>
2025-08-19 20:30:03 -07:00
systems-assistant[bot] 1f86010ca2 ROCpd support [Part 2] (#109)
* Rocpd part 2, caching

* Fix shadowed variables

* backward compatibility

* Fixed designated initializers

* Fix timemory include

* Remove benchmark & Fix build issues for rhel

* Add missing bracket

* Fix shadowing and pedantic

* Fix pedantic pt2

* Fix duplicated SDK calls

* Add decay in get_size_impl

* Rename sample cache to trace cache

* Add cache storage supported types

* Resolving track naming in sampling module

* fix sampling of flushing thread

* fix sampling of flushing thread 2

* throw exception upon store while buffer storage is not running

* Prevent fork crashing

* Fix rebase issue

* Applied suggestions from code review

* Change flushing thread to use PTL

* Fix agent creation order

* Fix stream id ci throw

* Remove force setup of rocprofiler-sdk

* Code cleanup

* Change initialization for agent

* Add missing namespace

* Fix the mismatch within the tool_agent->device_id

* Switch from using handle to use agent type index

* Fix pmc info comparator in metadata registry

---------

Co-authored-by: Aleksandar <aleksandar.djordjevic@amd.com>
Co-authored-by: Milan Radosavljevic <milan.radosavljevic@amd.com>
Co-authored-by: Marjan Antic <marantic@amd.com>
2025-08-19 22:01:04 -04:00
systems-assistant[bot] 351d598869 [Docs] Adding AQLprofile info (#150)
* Adding AQLprofile info

* update aqlprofile text

---------

Co-authored-by: Matt Williams <Matt.Williams+amdeng@amd.com>
2025-08-19 08:44:25 -07:00
systems-assistant[bot] c746ba6d31 Fix compilation with libc++ (#148)
`interop.cpp` uses `std::find_if` without a relevant include file, breaking compilation with Clang/libc++.

Co-authored-by: Sv. Lockal <lockalsash@gmail.com>
2025-08-19 08:42:03 -07:00
Giovanni Lenzi Baraldi ff43893902 Fix decoder description (#513) 2025-08-19 08:35:22 -07:00
Joseph Macaranas a22b6f3aca Merge commit 'ed7fae51d884a17cd3bd96c7973d1f1f78b5f3a5' into develop 2025-08-18 16:23:12 -04:00
systems-assistant[bot] bc455989d9 Merge commit 'b0519e8441273e20bbcbe1bb7b56cdf55369f514' into develop 2025-08-18 18:08:07 +00:00
systems-assistant[bot] c95526c57e Merge commit '55b54866fb50f264b064b20814679385ffd2847c' into develop 2025-08-18 18:08:05 +00:00
systems-assistant[bot] 23d4f23d7f Merge commit '0ac913e64c988e6b9ea8e86d74292cdf44c5d255' into develop 2025-08-18 18:08:03 +00:00
Jason Bonnell 24332bb65b [rocprofiler-systems] Update containers workflow to generate matrix from file (#369)
* Use matrix data from containers.yml in rocprofiler-systems-containers.yml

* add on pull_request to rocprofiler-systems-containers.yml

* update yq command in rocprofiler-systems-containers.yml

* ensure no indenting in json output

* update output definition in prepare_release_matrix

* added container-ci.yml for ci containers

* move ordering of steps in rocprofiler-systems-containers.yml

* Update projects/rocprofiler-systems/docker/containers-ci.yml

Co-authored-by: David Galiffi <David.Galiffi@amd.com>

* only push containers when running on schedule or push

* Formatting fix

Cleaning trailing whitespace.

---------

Co-authored-by: David Galiffi <David.Galiffi@amd.com>
2025-08-18 13:17:17 -04:00
Peter Park b4e336aef3 [rocm-smi-lib] docs: fix changelog heading ROCm 6.5.0 -> ROCm 7.0.0 (#363) 2025-08-18 11:46:02 -04:00
Aleksandar Djordjevic c58e3e92c1 Fix rocprof-sys-run --help command (#359)
Handle the `--help` parameter in the same way it is handled in other tools such as `rocprof-sys-sample` and `rocprof-sys-instrument`.
2025-08-15 17:28:08 -04:00
Kian Cossettini 947adacb3a [rocprofiler-systems] Fixed build-docker script failure on sparse checkout (#364)
Created `containers.yml` file in docker folder.
Updated `build-docker.sh` script to read from this file to determine supported OS + ROCm combinations.
2025-08-15 13:50:43 -04:00
hkasivis 53ba025a2e libhsakmt: Don't use MADV_DONTFORK for paged memory (#356)
Also advice parameter of madvise() system call is not a bitmask. So fix
that also

v2: Use MAP_SHARED instead of MAP_PRIVATE. This avoids MMU notifiers and
    evictions.

Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
2025-08-15 09:22:20 -04:00
jokim-amd 700afd2d17 Re-Enable IPC DMA Bufs by default
Let ROCr use the new IPC-DMA bufs path.
2025-08-14 18:49:09 -04:00
jamessiddeley-amd 2936a468aa [rocprof-compute] updated formatting workflow and docs (#348)
* updated formatting workflow and docs

* restored requirements.txt
2025-08-14 14:12:24 -04:00
systems-assistant[bot] 3fd8af5974 rocr: SvmPrefetch to a particular node (#294)
Previously regardless of hsa_agent passed the prefetch is always driven
to node 0, now the agent of interest may be properly prefetched.

Signed-off-by: Sunday Clement <Sunday.Clement@amd.com>
Co-authored-by: Sunday Clement <Sunday.Clement@amd.com>
Co-authored-by: systems-assistant[bot] <systems-assistant[bot]@users.noreply.github.com>
2025-08-14 09:52:45 -04:00
Giovanni Lenzi Baraldi 75eb06dc18 Add realtime clock agent info to ATT records (#319)
* Add agent info to ATT records

* Fix names

* Update projects/aqlprofile/src/pm4/sqtt_builder.h
2025-08-14 13:35:00 +02:00
Jonathan R. Madsen 9df2c1ec68 [rocprofiler-sdk] Fix formatting, linting, and CI workflows (#345)
* [rocprofiler-sdk] Fix formatting and lint workflows

- several formatting workflows were silently failing when listing files

* format metrics_test.h

* Improve formatting job robustness

* Source formatting workflow does not use container

* Use PyPi clang-format

* Format rocpd/source/csv.cpp source

* Fix rocprofiler-sdk CI workflow

- fix invalid context access

* Update run-ci.py

- fix ctest_update

* Update run-ci.py

- handle old checkout in ROCm/rocprofiler-sdk
2025-08-14 00:02:23 -05:00
Luca Bruni e28900793b Add Google logging initializer to OMPT start tool to prevent benign warning. (#337) 2025-08-13 18:39:49 -05:00
systems-assistant[bot] dd37d215fd Add documentation describing ROCPROFSYS_USE_RCCP (#110)
* Add documentation describing ROCPROFSYS_USE_RCCP

Signed-off-by: David Galiffi <David.Galiffi@amd.com>

* Update wordlist

Signed-off-by: David Galiffi <David.Galiffi@amd.com>

* Update CHANGELOGS.md

---------

Signed-off-by: David Galiffi <David.Galiffi@amd.com>
Co-authored-by: David Galiffi <David.Galiffi@amd.com>
2025-08-13 18:01:18 -04:00
vedithal-amd 80b7e6baee Backport CHANGELOG change from rc4 cherry pick (#342)
* Backport from https://github.com/ROCm/rocprofiler-compute/pull/860
2025-08-13 14:49:53 -04:00
David Yat Sin 875fb40a03 Dayatsin/develop vmm pointer info (#305)
* rocr: hsa_amd_pointer_info to support VMEM pointers

Extend hsa_amd_pointer_info to support virtual memory addresses.

If hsa_amd_pointer_info is called on an address that is reserved but not
mapped to memory, then the pointer type will be reported as
HSA_EXT_POINTER_TYPE_RESERVED_ADDR.

If hsa_amd_pointer_info is called on an address that is mapped, then the
pointer type will be reported as HSA_EXT_POINTER_TYPE_HSA_VMEM

* rocrtst: VirtMemory_Basic_Test test for pointer info

Extend rocrtstFunc.VirtMemory_Basic_Test to test for
hsa_amd_pointer_info

* rocrtst: Add SVM Memory Test
2025-08-13 14:21:47 -04:00
Dmitrii a2d3f4a0e0 rdc: Profiler - improve metrics path detection (#333)
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2025-08-13 12:33:17 -05:00
mat3ix c41050d01f rocr: SDMA improvements (#326)
- When SDMA queue gets full when copying 2GB or more it blocks async
copy api
- Improve/format logging
2025-08-13 10:25:29 -04:00
xuchen-amd 483315fd0a Fix amd_smi lookup logic to account for amd-smi>=26.0.0 (#320) 2025-08-13 10:21:45 -04:00
Ammar ELWazir 28977999ad Switching from RHEL 8 to RHEL 9 (#318)
* Switching from RHEL 8 to RHEL 9

* Switching from RHEL 8 to RHEL 9

* Switching from RHEL 8 to RHEL 9

* Fixing ROCPD for older Python Versions

* Formatting and checking if SQRT is available or not

* Formatting and checking if SQRT is available or not

* Formatting and checking if SQRT is available or not

* Formatting and checking if SQRT is available or not

* Formatting and checking if SQRT is available or not

* Update projects/rocprofiler-sdk/source/lib/python/rocpd/summary.py

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* Update summary.py

* Update summary.py

* Update projects/rocprofiler-sdk/source/lib/python/rocpd/summary.py

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* Update projects/rocprofiler-sdk/source/lib/python/rocpd/summary.py

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* Update summary.py

* Update summary.py

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2025-08-13 08:53:38 -05:00
David Galiffi 311e279596 Add "branch" requirement to "push" workflows. (#329)
* Add "branch" requirement to "push" workflows.

* Deleted obsolete workflow files

Signed-off-by: David Galiffi <David.Galiffi@amd.com>

* Fix path to the `check-copyright.sh` in `pre-commit-config.yaml`

---------

Signed-off-by: David Galiffi <David.Galiffi@amd.com>
2025-08-13 09:42:49 -04:00
systems-assistant[bot] d0a18e0eb9 [cmake] - Update search for LibElf (#256)
There is an issue with TheRock build currently. They have
a local source build of elfutils they want to use instead
of a system package. Currently, rocr uses it's own
FindLibElf.cmake module and this is inhibiting the build
from finding the libelf config built by TheRock.

Now we will first search in config mode and fallback to
module mode if nothing is found.

Authored-by: Ethan Stewart <ethan.stewart@amd.com>
Co-authored-by: systems-assistant[bot] <systems-assistant[bot]@users.noreply.github.com>
2025-08-13 09:13:45 -04:00
systems-assistant[bot] 76fd78d394 Merge commit '499e2ee044e15ebdf2c5cae1ee3ea282736ed66a' into develop 2025-08-13 12:09:50 +00:00