[rocroller] Enable azure testing #1406
Open
+115
−21
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Generated Documentation |
Performance Report for gfx12Results@@ Significant (p-val <0.05) Performance Diffs @@
====================================================================================================
+ 0.20% | p=2.5347e-02
| 3. FloatsGEMM(M: 7680, N: 8448, K: 8192, alpha: 2, beta: 0.5, types: {'type_A': 'bf16', 'type_B': 'bf16', 'type_C': 'bf16', 'type_D': 'bf16', 'type_acc': 'bf16', 'trans_A': 'T', 'trans_B': 'N', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 64, mac_n: 64, mac_k: 64, wave_m: 16, wave_n: 16, wave_k: 16, wave_b: 1, workgroup_size_x: 64, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Sequential, prefetch: False, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: False, numWGs: 0, streamKTwoTile: False, architecture: {'ArchString': 'gfx1201', 'Xnack': False, 'Sramecc': False}, matchMemoryAccess: True, version: ) | GEMMProblem(M=7680, N=8448, K=8192, alpha=2, beta=0.5, types={'type_A': 'bf16', 'type_B': 'bf16', 'type_C': 'bf16', 'type_D': 'bf16', 'type_acc': 'bf16', 'trans_A': 'T', 'trans_B': 'N', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=64, mac_n=64, mac_k=64, wave_m=16, wave_n=16, wave_k=16, wave_b=1, workgroup_size_x=64, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Sequential', prefetch=False, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=False, numWGs=0, streamKTwoTile=False, architecture={'ArchString': 'gfx1201', 'Xnack': False, 'Sramecc': False}, matchMemoryAccess=True, version='')
+ 0.16% | p=1.7451e-03
| 3. FloatsGEMM(M: 7680, N: 8448, K: 8192, alpha: 2, beta: 0.5, types: {'type_A': 'bf8', 'type_B': 'fp8', 'type_C': 'float', 'type_D': 'float', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'N', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 64, mac_n: 64, mac_k: 64, wave_m: 16, wave_n: 16, wave_k: 16, wave_b: 1, workgroup_size_x: 64, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Sequential, prefetch: False, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: False, numWGs: 0, streamKTwoTile: False, architecture: {'ArchString': 'gfx1201', 'Xnack': False, 'Sramecc': False}, matchMemoryAccess: True, version: ) | GEMMProblem(M=7680, N=8448, K=8192, alpha=2, beta=0.5, types={'type_A': 'bf8', 'type_B': 'fp8', 'type_C': 'float', 'type_D': 'float', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'N', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=64, mac_n=64, mac_k=64, wave_m=16, wave_n=16, wave_k=16, wave_b=1, workgroup_size_x=64, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Sequential', prefetch=False, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=False, numWGs=0, streamKTwoTile=False, architecture={'ArchString': 'gfx1201', 'Xnack': False, 'Sramecc': False}, matchMemoryAccess=True, version='')
- 0.14% | p=2.5347e-02
| 3. FloatsGEMM(M: 7680, N: 8448, K: 8192, alpha: 2, beta: 0.5, types: {'type_A': 'fp8', 'type_B': 'fp8', 'type_C': 'float', 'type_D': 'float', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 64, mac_n: 64, mac_k: 64, wave_m: 16, wave_n: 16, wave_k: 16, wave_b: 1, workgroup_size_x: 64, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Priority, prefetch: False, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: False, numWGs: 0, streamKTwoTile: False, architecture: {'ArchString': 'gfx1201', 'Xnack': False, 'Sramecc': False}, matchMemoryAccess: True, version: ) | GEMMProblem(M=7680, N=8448, K=8192, alpha=2, beta=0.5, types={'type_A': 'fp8', 'type_B': 'fp8', 'type_C': 'float', 'type_D': 'float', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=64, mac_n=64, mac_k=64, wave_m=16, wave_n=16, wave_k=16, wave_b=1, workgroup_size_x=64, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Priority', prefetch=False, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=False, numWGs=0, streamKTwoTile=False, architecture={'ArchString': 'gfx1201', 'Xnack': False, 'Sramecc': False}, matchMemoryAccess=True, version='')
- 1.88% | p=1.7451e-03
| CodeGen(instCount: 40000, instructions: comments)| CodeGen() | CodeGen(instCount: 40000, instructions: comments)
- 2.33% | p=5.6994e-05
| CodeGen(instCount: 40000, instructions: simple_mi)| CodeGen() | CodeGen(instCount: 40000, instructions: simple_mi)
Full table of results
|
Performance Report for gfx950-rocrollerResults@@ Significant (p-val <0.05) Performance Diffs @@
====================================================================================================
+ 0.87% | p=2.6741e-11
| 3. FloatsGEMM(M: 4096, N: 4096, K: 32768, alpha: 2, beta: 0, types: {'type_A': 'fp4', 'type_B': 'fp4', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'T', 'trans_B': 'N', 'scale_A': 'Separate', 'scaleType_A': 'E8M0', 'scale_B': 'Separate', 'scaleType_B': 'E8M0', 'scaleBlockSize': 32, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: 0, workgroupMappingValue: -1)| GEMM(mac_m: 256, mac_n: 256, mac_k: 128, wave_m: 16, wave_n: 16, wave_k: 128, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: False, betaInFma: True, direct2LDS_A: True, direct2LDS_B: True, scheduler: Priority, prefetch: True, prefetchInFlight: 4, prefetchLDSFactor: 1, prefetchMixMemOps: True, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: True, prefetchScale: True, streamK: False, numWGs: 0, streamKTwoTile: False, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=4096, N=4096, K=32768, alpha=2, beta=0, types={'type_A': 'fp4', 'type_B': 'fp4', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'T', 'trans_B': 'N', 'scale_A': 'Separate', 'scaleType_A': 'E8M0', 'scale_B': 'Separate', 'scaleType_B': 'E8M0', 'scaleBlockSize': 32, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=0, workgroupMappingValue=-1)GEMMSolution(mac_m=256, mac_n=256, mac_k=128, wave_m=16, wave_n=16, wave_k=128, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=False, betaInFma=True, direct2LDS_A=True, direct2LDS_B=True, scheduler='Priority', prefetch=True, prefetchInFlight=4, prefetchLDSFactor=1, prefetchMixMemOps=True, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=True, prefetchScale=True, streamK=False, numWGs=0, streamKTwoTile=False, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
+ 0.60% | p=1.1377e-13
| 3. FloatsGEMM(M: 4096, N: 4096, K: 32768, alpha: 2, beta: 0, types: {'type_A': 'fp4', 'type_B': 'fp4', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'T', 'trans_B': 'N', 'scale_A': 'Separate', 'scaleType_A': 'E8M0', 'scale_B': 'Separate', 'scaleType_B': 'E8M0', 'scaleBlockSize': 32, 'scaleSkipPermlane': True, 'scaleShuffleTileA': [64, 4, 2], 'scaleShuffleTileB': [64, 4, 2]}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: 0, workgroupMappingValue: -1)| GEMM(mac_m: 256, mac_n: 256, mac_k: 128, wave_m: 32, wave_n: 32, wave_k: 64, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: False, betaInFma: True, direct2LDS_A: True, direct2LDS_B: True, scheduler: Priority, prefetch: True, prefetchInFlight: 4, prefetchLDSFactor: 1, prefetchMixMemOps: True, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: True, prefetchScale: True, streamK: False, numWGs: 0, streamKTwoTile: False, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=4096, N=4096, K=32768, alpha=2, beta=0, types={'type_A': 'fp4', 'type_B': 'fp4', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'T', 'trans_B': 'N', 'scale_A': 'Separate', 'scaleType_A': 'E8M0', 'scale_B': 'Separate', 'scaleType_B': 'E8M0', 'scaleBlockSize': 32, 'scaleSkipPermlane': True, 'scaleShuffleTileA': [64, 4, 2], 'scaleShuffleTileB': [64, 4, 2]}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=0, workgroupMappingValue=-1)GEMMSolution(mac_m=256, mac_n=256, mac_k=128, wave_m=32, wave_n=32, wave_k=64, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=False, betaInFma=True, direct2LDS_A=True, direct2LDS_B=True, scheduler='Priority', prefetch=True, prefetchInFlight=4, prefetchLDSFactor=1, prefetchMixMemOps=True, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=True, prefetchScale=True, streamK=False, numWGs=0, streamKTwoTile=False, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
+ 1.36% | p=1.0433e-37
| 3. FloatsGEMM(M: 4096, N: 4096, K: 32768, alpha: 2, beta: 0.5, types: {'type_A': 'fp4', 'type_B': 'fp4', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'T', 'trans_B': 'N', 'scale_A': 'Separate', 'scaleType_A': 'E8M0', 'scale_B': 'Separate', 'scaleType_B': 'E8M0', 'scaleBlockSize': 32, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 256, mac_n: 256, mac_k: 128, wave_m: 32, wave_n: 32, wave_k: 64, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: True, direct2LDS_B: True, scheduler: Priority, prefetch: True, prefetchInFlight: 2, prefetchLDSFactor: 1, prefetchMixMemOps: False, loadLDSScale_A: True, loadLDSScale_B: True, swizzleScale: False, prefetchScale: False, streamK: False, numWGs: 0, streamKTwoTile: False, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=4096, N=4096, K=32768, alpha=2, beta=0.5, types={'type_A': 'fp4', 'type_B': 'fp4', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'T', 'trans_B': 'N', 'scale_A': 'Separate', 'scaleType_A': 'E8M0', 'scale_B': 'Separate', 'scaleType_B': 'E8M0', 'scaleBlockSize': 32, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=256, mac_n=256, mac_k=128, wave_m=32, wave_n=32, wave_k=64, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=True, direct2LDS_B=True, scheduler='Priority', prefetch=True, prefetchInFlight=2, prefetchLDSFactor=1, prefetchMixMemOps=False, loadLDSScale_A=True, loadLDSScale_B=True, swizzleScale=False, prefetchScale=False, streamK=False, numWGs=0, streamKTwoTile=False, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
- 0.61% | p=6.9031e-07
| 3. FloatsGEMM(M: 4096, N: 4096, K: 32768, alpha: 2, beta: 0.5, types: {'type_A': 'fp4', 'type_B': 'fp4', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'T', 'trans_B': 'N', 'scale_A': 'Separate', 'scaleType_A': 'E8M0', 'scale_B': 'Separate', 'scaleType_B': 'E8M0', 'scaleBlockSize': 32, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: 0, workgroupMappingValue: -1)| GEMM(mac_m: 256, mac_n: 256, mac_k: 128, wave_m: 32, wave_n: 32, wave_k: 64, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: True, direct2LDS_B: True, scheduler: Priority, prefetch: True, prefetchInFlight: 2, prefetchLDSFactor: 1, prefetchMixMemOps: False, loadLDSScale_A: True, loadLDSScale_B: True, swizzleScale: False, prefetchScale: False, streamK: False, numWGs: 0, streamKTwoTile: False, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=4096, N=4096, K=32768, alpha=2, beta=0.5, types={'type_A': 'fp4', 'type_B': 'fp4', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'T', 'trans_B': 'N', 'scale_A': 'Separate', 'scaleType_A': 'E8M0', 'scale_B': 'Separate', 'scaleType_B': 'E8M0', 'scaleBlockSize': 32, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=0, workgroupMappingValue=-1)GEMMSolution(mac_m=256, mac_n=256, mac_k=128, wave_m=32, wave_n=32, wave_k=64, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=True, direct2LDS_B=True, scheduler='Priority', prefetch=True, prefetchInFlight=2, prefetchLDSFactor=1, prefetchMixMemOps=False, loadLDSScale_A=True, loadLDSScale_B=True, swizzleScale=False, prefetchScale=False, streamK=False, numWGs=0, streamKTwoTile=False, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
- 1.28% | p=1.7451e-03
| 3. FloatsGEMM(M: 4096, N: 4096, K: 8192, alpha: 2, beta: 0.5, types: {'type_A': 'fp8', 'type_B': 'fp8', 'type_C': 'float', 'type_D': 'float', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'N', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 64, mac_n: 64, mac_k: 128, wave_m: 16, wave_n: 16, wave_k: 128, wave_b: 1, workgroup_size_x: 256, workgroup_size_y: 1, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Priority, prefetch: True, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: False, numWGs: 0, streamKTwoTile: False, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=4096, N=4096, K=8192, alpha=2, beta=0.5, types={'type_A': 'fp8', 'type_B': 'fp8', 'type_C': 'float', 'type_D': 'float', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'N', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=64, mac_n=64, mac_k=128, wave_m=16, wave_n=16, wave_k=128, wave_b=1, workgroup_size_x=256, workgroup_size_y=1, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Priority', prefetch=True, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=False, numWGs=0, streamKTwoTile=False, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
+ 4.16% | p=5.6994e-05
| 3. FloatsGEMM(M: 4096, N: 4096, K: 8192, alpha: 2, beta: 0.5, types: {'type_A': 'fp8', 'type_B': 'fp8', 'type_C': 'float', 'type_D': 'float', 'type_acc': 'float', 'trans_A': 'T', 'trans_B': 'N', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 128, mac_n: 128, mac_k: 64, wave_m: 32, wave_n: 32, wave_k: 64, wave_b: 1, workgroup_size_x: 256, workgroup_size_y: 1, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Priority, prefetch: True, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: False, numWGs: 0, streamKTwoTile: False, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=4096, N=4096, K=8192, alpha=2, beta=0.5, types={'type_A': 'fp8', 'type_B': 'fp8', 'type_C': 'float', 'type_D': 'float', 'type_acc': 'float', 'trans_A': 'T', 'trans_B': 'N', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=128, mac_n=128, mac_k=64, wave_m=32, wave_n=32, wave_k=64, wave_b=1, workgroup_size_x=256, workgroup_size_y=1, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Priority', prefetch=True, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=False, numWGs=0, streamKTwoTile=False, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
+ 0.47% | p=1.7451e-03
| 3. FloatsGEMM(M: 7680, N: 8448, K: 8192, alpha: 2, beta: 0.5, types: {'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'T', 'trans_B': 'N', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 64, mac_n: 64, mac_k: 64, wave_m: 32, wave_n: 32, wave_k: 8, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Priority, prefetch: True, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: False, numWGs: 0, streamKTwoTile: False, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=7680, N=8448, K=8192, alpha=2, beta=0.5, types={'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'T', 'trans_B': 'N', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=64, mac_n=64, mac_k=64, wave_m=32, wave_n=32, wave_k=8, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Priority', prefetch=True, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=False, numWGs=0, streamKTwoTile=False, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
- 0.33% | p=1.7451e-03
| 3. FloatsGEMM(M: 7680, N: 8448, K: 8448, alpha: 2, beta: 0.5, types: {'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 128, mac_n: 128, mac_k: 16, wave_m: 32, wave_n: 32, wave_k: 8, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Priority, prefetch: False, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: True, numWGs: 256, streamKTwoTile: True, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=7680, N=8448, K=8448, alpha=2, beta=0.5, types={'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=128, mac_n=128, mac_k=16, wave_m=32, wave_n=32, wave_k=8, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Priority', prefetch=False, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=True, numWGs=256, streamKTwoTile=True, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
+ 3.95% | p=5.6994e-05
| 3. FloatsGEMM(M: 7680, N: 8448, K: 8448, alpha: 2, beta: 0.5, types: {'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 128, mac_n: 64, mac_k: 16, wave_m: 32, wave_n: 32, wave_k: 8, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Priority, prefetch: False, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: True, numWGs: 256, streamKTwoTile: True, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=7680, N=8448, K=8448, alpha=2, beta=0.5, types={'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=128, mac_n=64, mac_k=16, wave_m=32, wave_n=32, wave_k=8, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Priority', prefetch=False, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=True, numWGs=256, streamKTwoTile=True, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
- 1.84% | p=5.6994e-05
| 3. FloatsGEMM(M: 7680, N: 8448, K: 8448, alpha: 2, beta: 0.5, types: {'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 128, mac_n: 64, mac_k: 32, wave_m: 32, wave_n: 32, wave_k: 8, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Priority, prefetch: False, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: True, numWGs: 256, streamKTwoTile: True, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=7680, N=8448, K=8448, alpha=2, beta=0.5, types={'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=128, mac_n=64, mac_k=32, wave_m=32, wave_n=32, wave_k=8, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Priority', prefetch=False, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=True, numWGs=256, streamKTwoTile=True, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
+ 1.24% | p=5.6994e-05
| 3. FloatsGEMM(M: 7680, N: 8448, K: 8448, alpha: 2, beta: 0.5, types: {'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 64, mac_n: 128, mac_k: 32, wave_m: 32, wave_n: 32, wave_k: 8, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Priority, prefetch: False, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: True, numWGs: 256, streamKTwoTile: False, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=7680, N=8448, K=8448, alpha=2, beta=0.5, types={'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=64, mac_n=128, mac_k=32, wave_m=32, wave_n=32, wave_k=8, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Priority', prefetch=False, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=True, numWGs=256, streamKTwoTile=False, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
+ 9.45% | p=1.7451e-03
| 3. FloatsGEMM(M: 7680, N: 8448, K: 8448, alpha: 2, beta: 0.5, types: {'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 64, mac_n: 256, mac_k: 16, wave_m: 32, wave_n: 32, wave_k: 8, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Priority, prefetch: False, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: True, numWGs: 256, streamKTwoTile: False, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=7680, N=8448, K=8448, alpha=2, beta=0.5, types={'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=64, mac_n=256, mac_k=16, wave_m=32, wave_n=32, wave_k=8, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Priority', prefetch=False, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=True, numWGs=256, streamKTwoTile=False, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
+ 0.53% | p=1.7451e-03
| 3. FloatsGEMM(M: 7680, N: 8448, K: 8448, alpha: 2, beta: 0.5, types: {'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 64, mac_n: 256, mac_k: 16, wave_m: 32, wave_n: 32, wave_k: 8, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Priority, prefetch: False, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: True, numWGs: 256, streamKTwoTile: True, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=7680, N=8448, K=8448, alpha=2, beta=0.5, types={'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=64, mac_n=256, mac_k=16, wave_m=32, wave_n=32, wave_k=8, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Priority', prefetch=False, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=True, numWGs=256, streamKTwoTile=True, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
+ 0.75% | p=2.5347e-02
| 3. FloatsGEMM(M: 7680, N: 8448, K: 8448, alpha: 2, beta: 0.5, types: {'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 64, mac_n: 256, mac_k: 32, wave_m: 32, wave_n: 32, wave_k: 8, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Priority, prefetch: False, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: True, numWGs: 256, streamKTwoTile: True, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=7680, N=8448, K=8448, alpha=2, beta=0.5, types={'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=64, mac_n=256, mac_k=32, wave_m=32, wave_n=32, wave_k=8, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Priority', prefetch=False, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=True, numWGs=256, streamKTwoTile=True, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
+ 0.15% | p=1.7451e-03
| 3. FloatsGEMM(M: 7680, N: 8448, K: 8448, alpha: 2, beta: 0.5, types: {'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 64, mac_n: 64, mac_k: 16, wave_m: 32, wave_n: 32, wave_k: 8, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Priority, prefetch: False, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: True, numWGs: 256, streamKTwoTile: False, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=7680, N=8448, K=8448, alpha=2, beta=0.5, types={'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=64, mac_n=64, mac_k=16, wave_m=32, wave_n=32, wave_k=8, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Priority', prefetch=False, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=True, numWGs=256, streamKTwoTile=False, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
+ 0.75% | p=5.6994e-05
| 3. FloatsGEMM(M: 7680, N: 8448, K: 8448, alpha: 2, beta: 0.5, types: {'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 64, mac_n: 64, mac_k: 16, wave_m: 32, wave_n: 32, wave_k: 8, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Priority, prefetch: False, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: True, numWGs: 256, streamKTwoTile: True, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=7680, N=8448, K=8448, alpha=2, beta=0.5, types={'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=64, mac_n=64, mac_k=16, wave_m=32, wave_n=32, wave_k=8, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Priority', prefetch=False, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=True, numWGs=256, streamKTwoTile=True, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
+ 0.63% | p=5.6994e-05
| 3. FloatsGEMM(M: 7680, N: 8448, K: 8448, alpha: 2, beta: 0.5, types: {'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 64, mac_n: 64, mac_k: 64, wave_m: 32, wave_n: 32, wave_k: 8, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Priority, prefetch: False, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: True, numWGs: 256, streamKTwoTile: True, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=7680, N=8448, K=8448, alpha=2, beta=0.5, types={'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=64, mac_n=64, mac_k=64, wave_m=32, wave_n=32, wave_k=8, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Priority', prefetch=False, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=True, numWGs=256, streamKTwoTile=True, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
+ 1.80% | p=1.7451e-03
| 3. FloatsGEMM(M: 8448, N: 8448, K: 128, alpha: 2, beta: 0.5, types: {'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A: 1, scaleValue_B: 1, workgroupMappingDim: -1, workgroupMappingValue: -1)| GEMM(mac_m: 128, mac_n: 256, mac_k: 16, wave_m: 32, wave_n: 32, wave_k: 8, wave_b: 1, workgroup_size_x: 128, workgroup_size_y: 2, workgroupRemapXCC: False, workgroupRemapXCCValue: -1, unroll_x: 0, unroll_y: 0, loadLDS_A: True, loadLDS_B: True, storeLDS_D: True, betaInFma: True, direct2LDS_A: False, direct2LDS_B: False, scheduler: Priority, prefetch: True, prefetchInFlight: 2, prefetchLDSFactor: 0, prefetchMixMemOps: False, loadLDSScale_A: False, loadLDSScale_B: False, swizzleScale: False, prefetchScale: False, streamK: False, numWGs: 0, streamKTwoTile: False, architecture: {'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess: True, version: ) | GEMMProblem(M=8448, N=8448, K=128, alpha=2, beta=0.5, types={'type_A': 'half', 'type_B': 'half', 'type_C': 'half', 'type_D': 'half', 'type_acc': 'float', 'trans_A': 'N', 'trans_B': 'T', 'scale_A': 'None', 'scaleType_A': 'None', 'scale_B': 'None', 'scaleType_B': 'None', 'scaleBlockSize': -1, 'scaleSkipPermlane': False, 'scaleShuffleTileA': [], 'scaleShuffleTileB': []}, scaleValue_A=1, scaleValue_B=1, workgroupMappingDim=-1, workgroupMappingValue=-1)GEMMSolution(mac_m=128, mac_n=256, mac_k=16, wave_m=32, wave_n=32, wave_k=8, wave_b=1, workgroup_size_x=128, workgroup_size_y=2, workgroupRemapXCC=False, workgroupRemapXCCValue=-1, unroll_x=0, unroll_y=0, loadLDS_A=True, loadLDS_B=True, storeLDS_D=True, betaInFma=True, direct2LDS_A=False, direct2LDS_B=False, scheduler='Priority', prefetch=True, prefetchInFlight=2, prefetchLDSFactor=0, prefetchMixMemOps=False, loadLDSScale_A=False, loadLDSScale_B=False, swizzleScale=False, prefetchScale=False, streamK=False, numWGs=0, streamKTwoTile=False, architecture={'ArchString': 'gfx950', 'Xnack': False, 'Sramecc': True}, matchMemoryAccess=True, version='')
+ 1.62% | p=5.6994e-05
| CodeGen(instCount: 40000, instructions: comments)| CodeGen() | CodeGen(instCount: 40000, instructions: comments)
+ 1.20% | p=5.6994e-05
| CodeGen(instCount: 40000, instructions: simple_mi)| CodeGen() | CodeGen(instCount: 40000, instructions: simple_mi)
Full table of results
|
Code Coverage Report for gfx942Summary
Artifacts
Commit Hashes |
awhittle3
reviewed
Sep 8, 2025
shared/rocroller/CMakeLists.txt
Outdated
|
|
||
| if(ROCROLLER_ENABLE_YAML_CPP) | ||
| find_package(yaml-cpp 0.8.0) | ||
| find_package(yaml-cpp) |
There was a problem hiding this comment.
yaml-cpp v0.7.0 and v0.8.0 are really different in terms of CMake setup. Your change here to use ${YAML_CPP_LIBRARIES} suggests to me that you might be trying to get this to work with v0.7.0. Note that we should maintain v0.8.0 support, and there might be a need to add more complex logic here if we want to support both versions.
There was a problem hiding this comment.
The yaml cpp libraries variable seems to bridge the gap between the old/new build systems.
awhittle3
reviewed
Sep 8, 2025
shared/rocroller/CMakeLists.txt
Outdated
| endif() | ||
|
|
||
| find_package(fmt 11.1.3 QUIET) | ||
| find_package(fmt REQUIRED) |
There was a problem hiding this comment.
We'll need to re-add libfmt-dev to our Dockerfiles if we don't want CMake to fetch it. Adding REQUIRED here causes CMake to hard fail on failure to find fmt and prevents all fetching of this package.
There was a problem hiding this comment.
Yeah I wanted it to hard fail while testing. This will change back before merging.
awhittle3
reviewed
Sep 8, 2025
shared/rocroller/CMakeLists.txt
Outdated
| endif() | ||
| endif() | ||
|
|
||
| find_package(Boost REQUIRED) # is 1.74.0 required? |
There was a problem hiding this comment.
Note that v1.74.0 is the minimum version of Boost that RocRoller requires, as it is one of the first with CMake support. Please make sure that this works on systems like RHEL and SLES as desired (SLES 15.5 has Boost v1.66 iirc) before you commit this change.
There was a problem hiding this comment.
(NB: I think v1.73.0 added CMake support, but I think we tested it and v1.74.0 was the earliest one that worked for us)
There was a problem hiding this comment.
Our Dockerfiles will also need to install this, too
There was a problem hiding this comment.
Also testing behavior on almalinux8 which has 1.66. The issue isn't cmake so much as it is a header was added in 1.74. I am not certain but it seems like we are only using boost headers.
There was a problem hiding this comment.
PR #1714 will remove the Boost dependency in RocRoller and will be so nice 😄
davidd-amd
force-pushed
the
users/davidd-amd/rocroller-drop-boost-fetch
branch
from
0d84d19 to
82a29c5
Compare
davidd-amd
force-pushed
the
users/davidd-amd/rocroller-drop-boost-fetch
branch
from
82a29c5 to
e9c355f
Compare
davidd-amd
commented
Oct 13, 2025
davidd-amd
commented
Oct 13, 2025
davidd-amd
changed the title
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Motivation
rocroller isn't currently tested under azure CI which limits some important test coverage.
Technical Details
Test Plan
Tested in both azure ci and matchci.