AMD ROCm™ Software
Official@rocm
AMD ROCm software provides specialized kernel authoring, performance profiling, and telemetry analysis for high-performance computing on AMD CDNA GPU architectures.
Agent Skills by AMD ROCm™ Software
Showing 17 vetted skills indexed across 2 GitHub repositories.
ROCmjob-report
Generate HPC job summary reports from Omnistat GPU telemetry.
ROCmjob-analysis
Parse Omnistat inspect data to diagnose GPU and host performance issues.
ROCmopen-database
Loads Omnistat telemetry data into VictoriaMetrics for querying and visualization.
ROCmport-to-layout-api
Port FlyDSL GPU kernels from raw buffer_ops to the layout API.
ROCmformat-code
Format Python and C/C++ files by removing unused imports and variables.
ROCmbuild-rocm-image
Build a ROCm development Docker image on a remote host via SSH.
ROCmflydsl-kernel-authoring
Author AMDGPU kernels with FlyDSL layout algebra and MLIR lowering.
ROCmbuild-flydsl
Automate building and installing FlyDSL from source on remote hosts or containers.
ROCmbisect-perf-regression
Automate Git bisect to locate the commit introducing GPU kernel performance regressions.
ROCmdebug-flydsl-kernel
Diagnose incorrect FlyDSL GPU kernel outputs, NaNs, infs, and crashes.
ROCmadd-target-atom-op
Add target-specific MmaOp/CopyOp to a FlyDSL backend dialect.
ROCmcapture-kernel-trace
Capture and analyze GPU kernel ATT traces using rocprofv3.
ROCmkernel-trace-analysis
Analyze ATT traces to profile GPU kernels and identify stall hotspots.
ROCmlds-optimization
Diagnose LDS bank conflicts and lgkmcnt stalls in FlyDSL GPU kernels.
ROCmprefetch-data-load
Introduce loop-carried prefetch to overlap data loads with compute in FlyDSL GPU kernel loops.
ROCmgemm-optimization
Optimizes GEMM kernels on AMD CDNA GPUs using FlyDSL tiling, memory layout, MFMA scheduling.
ROCmFlyDSL Tile Programming
Generate and validate FlyDSL GPU kernels from kernel specifications using tile programming.
Frequently Asked Questions About AMD ROCm™ Software
FAQPage SchemaWhat specific tasks can engineers perform using these capabilities?▼
Engineers can author GPU kernels using FlyDSL, optimize memory layouts to resolve LDS bank conflicts, perform performance regression bisecting, and analyze ATT traces to identify stall hotspots. Additionally, the suite enables the ingestion of Omnistat telemetry data into VictoriaMetrics for large-scale cluster performance monitoring and reporting.
Which target personas benefit from these technical capabilities?▼
These capabilities are designed for HPC systems engineers, GPU kernel developers, and performance architects working on AMD CDNA-based infrastructure. The functionality specifically targets those responsible for low-level hardware optimization, kernel debugging, and large-scale GPU cluster telemetry analysis.
What are the prerequisites and dependencies for utilizing these functions?▼
Utilization requires an AMD CDNA-based GPU environment, the ROCm software stack, and access to the FlyDSL compiler infrastructure. Specific tasks like trace analysis require rocprofv3, while telemetry ingestion depends on the presence of Omnistat data and a configured VictoriaMetrics instance for storage and visualization.