* clr: Update signal count and pool size for staging buffer
* Change to naming of variables etc
---------
Co-authored-by: Rahul Manocha <rmanocha@amd.com>
This change resolves some of the warnings generated during clr builds.
Quiet regular output of doxygen.
Disable non-documented warnings of doxygen.
Signed-off-by: Sebastian Luzynski <Sebastian.Luzynski@amd.com>
* SWDEV-558836, SWDEV-558837 - Add hipMemSetMemPool and hipMemGetMemPool implementation
* Add managed allocation type for mem pools
* Update rocprofiler-sdk with APis declaration
Integrates rocm-kpack runtime library for loading device code from
external kpack archives at HIP initialization time.
Changes:
- Add kpack_params_ optional to FatBinaryInfo for HIPK metadata
- Parse HIPK magic (0x4B504948) in digestFatBinary to detect kpack'd binaries
- Add ExtractKpackBinary() to load code objects via kpack_load_code_object()
- Wire up kpack cache lifecycle in hip_global.cpp
- Track kpack allocations for proper cleanup
- Support multi-TU binaries via bundle_index (co_index parameter)
The ROCM_KPACK_ENABLED cmake flag controls whether kpack support is compiled
in. When disabled, HIPK binaries return hipErrorNotSupported.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
* clr: Implement dynamic stream to HW queue assignment
This change implements dynamic stream to hardware queue (HWq) mapping
with the following features:
* Queue depth heuristics with weights for optimal HWq assignment
* Make last used queue sticky for better locality
* Use pipe HWq to pipe mapping - gfx9 follows a round-robin queue to
pipe mapping based on creation order (single process per device only,
as pipe ID is statically assigned by runtime)
* More aggressive heuristic usage for better queue distribution
* Extend dynamic queues support for all stream priorities
Environment variables:
* DEBUG_HIP_DYNAMIC_QUEUE: 0 - disabled, 1 - Depth heuristics 2 -
Depth+Pipe heuristics
* DEBUG_HIP_IGNORE_STREAM_PRIORITY=1: ignore priority stream creation
* clr: Clean up last_used_queue_
* SWDEV-555889 - Support mipmap on rocr
Support mipmap in hip-rt on rocr backend.
Enable all mipmap tests in Windows.
Some other minor improvement.
Add some SRD logs that will be removed finally.
* Add sampler.mipFilter to fix sampler issues on mipmap in rocr.
Fix format issues of view of leveled image and mipmap image in blit kernel in rocr.
Enabled disabled mipmap tests.
* Rewrite view logic
* Set word4.f.PITCH = 0 for mipmap SRD on navi31 to fix unstable test issues.
Reset last error in nagative tests.
* Remove SRD dump log from hip-rt
Let Rocr mipmap log be in condition.
* minor format chang
* Exclude mipmap tests for mi200+ which don't support mipmap.
Updated to convert flags correctly
Added ObjectRegistry to track registered and mapped resources and incorporated it into hip_gl.
Added mip level check
Made functions static in-line
Reworked validation to be more clear.
* SWDEV-549518 - Enable logging dynamically through HIP APIS.
* SWDEV-549518 - Adding ROCProfiler related new API changes.
* rocprofiler-sdk changes for hip api additions.
---------
Co-authored-by: Venkateshwar Reddy Kandula <venkateshwar.kandula1306@gmail.com>
Co-authored-by: jainprad <92369414+jainprad@users.noreply.github.com>
GCC does not support anonymous structs with members that have non-trivial constructors. This commit changes the header to remove the union when compiling with gcc. This should be a non-breaking change for other compilers.
* Add hipDeviceAttributeExpertSchedMode
---------
Co-authored-by: Stefan Sokolovic <stefan.sokolovic2@amd.com>
* Update hipDeviceAttributeExpertSchedMode unit test
* Move check to ROCr from thunk interface
* Revert unrelated whitespace changes
* Revert version bump
---------
Co-authored-by: Stefan Sokolovic <stefan.sokolovic2@amd.com>
AMD_LOG_LEVEL_SIZE is being used in a global variable.
This always uses the default value of 2048 because the
HIP runtime doesn't have the opportunity to load
environment variables at the point where global variables
are initialized.
The solution is to use AMD_LOG_LEVEL_SIZE inside
truncate_log_file() function.
Problem:
The existing SDMA engine selection logic had several issues:
1. Same VirtualGPU/stream could use different SDMA engines for consecutive
async copies since copy_engine_status may report engines as busy
2. Busy and Preferred engine check for every copy
3. No global tracking of which VirtualGPU uses which engine, leading to
suboptimal resource allocation
Solution:
Implemented a global SDMA engine allocator with per-stream affinity:
- Added Device::SdmaEngineAllocator to manage VirtualGPU → engine assignments
* Maintains global map of active assignments
* Enforces exclusivity: different streams use different engines (except
inter-GPU copies where preferred engines are prioritized for optimal
hardware paths like XGMI links)
* Thread-safe allocation/release with Monitor lock
- Modified VirtualGPU to cache assigned engine locally (assigned_sdma_engine_)
for fast lookup without map access on hot path
- Refactored rocrCopyBuffer() to:
1. Check local cached engine first → use if assigned
2. Call AllocateSdmaEngine() if not assigned → cache result
- Moved HSA API queries (memory_copy_engine_status, memory_get_preferred_copy_engine)
into AllocateEngine() for cleaner separation of concerns
- Engine release on HostQueue::finish() instead of only VirtualGPU destruction
* Improves engine utilization by releasing earlier
* Added virtual ReleaseSdmaEngines() method to device::VirtualDevice
- Added future path for simple round-robin allocation (kUseSimpleRR) for
next-gen GPUs with uniform SDMA bandwidth (disabled by default)
Cleanup:
- Removed selectSdmaEngine() helper (logic moved to allocator)
- Removed getSdmaRWMasks() (allocator accesses maxSdmaReadMask_/WriteMask_ directly)
- Removed unused sdmaEngineReadMask_/WriteMask_ member variables from DmaBlitManager
Benefits:
- Ensures consistent per-stream SDMA engine usage
- Prevents cross-stream contention and engine thrashing
- Prioritizes hardware-optimal paths for inter-GPU transfers
- Better resource utilization through earlier release
- Cleaner, more maintainable code structure
* SWDEV-508225 - do not assert() after calling digestFatBinary() if it fails. Otherwise this causes assertions to trigger easily in systems that have an APU and a discrete GPU and the code was compiled for the discrete one
* SWDEV-508225 - fix that when using a non-existent ordinal in HIP_VISIBLE_DEVICES, getCurrentArch() would crash
Jin Jung
Alexandra Sidorova
Rahul Manocha
German Andryeyev
SaleelK
Shadi Dashmiz
sluzynsk-amd
Ioannis Assiouras
vstojilj
marandje
Stella Laurenzo
Xie, AlexBin
Tao Sang
Sam Ruscica
Karthik Jayaprakash
systems-assistant[bot]
Fábio Mestre
Filip Jankovic
David Yat Sin
Luca Bruni
AidanBeltonS
Andrei Kochin
Godavarthy Surya, Anusha
Gerardo Hernandez
Jimbo
Sourabh U Betigeri
Matt Arsenault