On modern system e.g. render nodes are made accessible via the udev
uaccess functionality, which adds the logged in user to the ACL of the
device. This means just checking for user and group is bound to give
false positives. Instead use os.access as a first check
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Add new implementation of direct send/recv reduce scatter
* Resolved conflicts
* Add multiple channels support to the reduction kernel of direct reduce scatter and adjust offset into buffer to utilize multiple channels.
* Resolve validation issue when number of elements is not divisible by number of channels leaving elements unaccount for in reduction.
* fix proxy hang
* set maxSrcs to 64 in reduceCopy
* optimize multi-channel code
* fix validation issue in single node MI300
* Tune the message size range for 2,4, and 8 Nodes
* Move Direct RS into separate kernel
* Add Copyright
* resolve review comments
* resolve review comments
* fix merge build issue
* revert move Direct RS into separate kernel
* address review comments
* address review comments
---------
Co-authored-by: KawtharShafie <kawtharshafie@gmail.com>
Co-authored-by: Ghadeer Alabandi <abandiga@gmail.com>
Co-authored-by: systems-assistant[bot] <systems-assistant[bot]@users.noreply.github.com>
This enhances libhsakmt's capabilities for multi-context KFD support by implementing per-context topology management.
Changes:
* Add hsaKmtGetClockCountersCtx for multi-context support
- Add context-aware version of hsaKmtGetClockCounters
- Original API is retained as a wrapper calling the ctx-version with primary context
* Enable independent debug sessions across multiple KFD contexts
-Create hsa_kfd_debug_context, introduce context-aware debug APIs, shift debug state to per-context
* Add perf sub-context for per-context performance counter management
- Introduce hsa_kfd_perf_context, move counter properties, add context - aware perf APIs, and update initialization
* Refactor FMM for per-context resource management
- Refactor multiple global variables related to FMM, including
GPU ID arrays , svm, cpuvm_aperture, and mem_handle_aperture to hsa_kfd_fmm_context
* Implement per-context topology for complete context isolation
- Migrate global topology data (g_system, g_props, map_user_to_sysfs_node_id)
to per-context hsa_kfd_topology_context structure
- Update all topology functions to accept HsaKFDContext parameter for
context-aware operations (validate_nodeid, get_node_props, get_iolink_props, etc.)
- Refactor topology snapshot management for per-context isolation
- Add context-aware PMC trace access APIs
Signed-off-by: Junhua Shen <Junhua.Shen@amd.com>
* Update aqlprofile-continuous_integration.yml to use rocm-7.2.0
* Update rocprofiler-sdk-continuous_integration.yml to use rocm-7.2.0
* Update rocprofiler-sdk-docs.yml to use rocm-7.2.0
* clr: Update signal count and pool size for staging buffer
* Change to naming of variables etc
---------
Co-authored-by: Rahul Manocha <rmanocha@amd.com>
* Fix channel overuse for 1 rank comms
* limit channels when warpSpeed is enabled but not used
* enable std::min check against # of CUs for maxChannels computation when warpSpeed is enabled
---------
Co-authored-by: Mustafa Abduljabbar <muabdulj@amd.com>
Co-authored-by: isaki001 <ioannissakiotis@gmail.com>
Co-authored-by: Corey Derochie <161367113+corey-derochie-amd@users.noreply.github.com>
* [rocprofiler-sdk] Fix domain_ops_padding for 515+ HIP operations
The HIP runtime API now has 515+ operations (as of ROCm 7.x), but
domain_ops_padding was set to 512. This caused std::out_of_range
exceptions when checking operations >= 512 via std::bitset::test().
Changes:
- Increase domain_ops_padding from 512 to 1024
- Add compile-time static_assert to validate padding is sufficient
for all API domains (HIP, HSA, marker, RCCL, rocDecode, rocJPEG)
Co-Authored-By: Claude (claude-opus-4.5) <noreply@anthropic.com>
* Update projects/rocprofiler-sdk/source/lib/rocprofiler-sdk/context/domain.cpp
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* [rocprofiler-sdk] Apply clang-format-11 to domain.cpp
Co-Authored-By: Claude (claude-opus-4.5) <noreply@anthropic.com>
* Rework implementation to ensure coverage of all operation enums
* Fix compiler error in unit test for enum_string.cpp
* Fix data types of domain_ops_padding values
* Revert some changes in domain.cpp
---------
Co-authored-by: Claude (claude-opus-4.5) <noreply@anthropic.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>
* SWDEV-561708 Counted queue size from env var
* use counted_queue_size for test
* remove rocrtst changes; add a const for default queue size
* Remove env var from test; use queue->size
* Improve env var documentation
* Correct type
Donato Capitella
Pedram Alizadeh
ammallya
systems-assistant[bot]
Shadi Dashmiz
xuchen-amd
cfreeamd
Sjoerd Simons
David Yat Sin
Alysa Liu
Jin Jung
Alexandra Sidorova
Jan Stephan
Junhua Shen
vedithal-amd
Venkateshwar Reddy Kandula
pghoshamd
Mark Meserve
Rahul Manocha
Benjamin Welton
Yiannis Papadopoulos