github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-04-07 03:00:26 -04:00
Code Issues Packages Projects Releases Wiki Activity

Revert "fix TF32 tensor core dropped in tc_sm89 (#9798)"

Browse Source 
This reverts commit 7c9a96824f.
This commit is contained in:
George Hotz
2025-04-09 12:27:39 +08:00
parent 1ed4eae510
commit 14928fecff
3 changed files with 15 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
.github/workflows/benchmark.yml vendored
View File

@@ -67,11 +67,11 @@ jobs:
DEBUG=2 CPU=1 AMX=1 python3.11 test/test_linearizer.py TestLinearizer.test_tensor_cores TestLinearizer.test_tensor_cores_padded
DEBUG=2 LLVM=1 AMX=1 python3.11 test/test_linearizer.py TestLinearizer.test_tensor_cores TestLinearizer.test_tensor_cores_padded
- name: Run Tensor Core GEMM (float)
run: DEBUG=2 SHOULD_USE_TC=1 python3.11 extra/gemm/simple_matmul.py | tee matmul.txt
run: DEBUG=2 python3.11 extra/gemm/simple_matmul.py | tee matmul.txt
- name: Run Tensor Core GEMM (half)
run: DEBUG=2 SHOULD_USE_TC=1 HALF=1 python3.11 extra/gemm/simple_matmul.py | tee matmul_half.txt
run: DEBUG=2 HALF=1 python3.11 extra/gemm/simple_matmul.py | tee matmul_half.txt
- name: Run Tensor Core GEMM (bfloat16)
run: DEBUG=2 SHOULD_USE_TC=1 BFLOAT16=1 python3.11 extra/gemm/simple_matmul.py | tee matmul_bfloat16.txt
run: DEBUG=2 BFLOAT16=1 python3.11 extra/gemm/simple_matmul.py | tee matmul_bfloat16.txt
- name: Fuzz Padded Tensor Core GEMM
run: METAL=1 M_START=6 M_STOP=10 M_STEP=1 N_START=6 N_STOP=10 N_STEP=1 K_START=6 K_STOP=24 K_STEP=1 TC_OPT=2 DEBUG=2 python3.11 ./extra/gemm/fuzz_matmul.py
- name: Run LLaMA
@@ -178,13 +178,13 @@ jobs:
PTX=1 ALLOW_TF32=1 NV=1 python3 test/test_linearizer.py TestLinearizer.test_tensor_cores TestLinearizer.test_tensor_cores_padded
- name: Run Tensor Core GEMM (CUDA)
run: |
CUDA=1 SHOULD_USE_TC=1 HALF=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul.txt
CUDA=1 SHOULD_USE_TC=1 BFLOAT16=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul_bfloat16.txt
CUDA=1 SHOULD_USE_TC=1 ALLOW_TF32=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul_tf32.txt
CUDA=1 HALF=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul.txt
CUDA=1 BFLOAT16=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul_bfloat16.txt
CUDA=1 ALLOW_TF32=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul_tf32.txt
- name: Run Tensor Core GEMM (PTX)
run: NV=1 PTX=1 SHOULD_USE_TC=1 HALF=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul_ptx.txt
run: NV=1 PTX=1 HALF=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul_ptx.txt
- name: Run Tensor Core GEMM (NV)
run: NV=1 SHOULD_USE_TC=1 HALF=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul_nv.txt
run: NV=1 HALF=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul_nv.txt
- name: Test NV=1
run: DEBUG=2 NV=1 python -m pytest -rA test/test_tiny.py
- name: Test CUDA=1
@@ -371,9 +371,9 @@ jobs:
- name: Test tensor cores
run: |
AMD=1 python3 test/test_linearizer.py TestLinearizer.test_tensor_cores TestLinearizer.test_tensor_cores_padded
AMD=1 SHOULD_USE_TC=1 BFLOAT16=1 DEBUG=2 python3 extra/gemm/simple_matmul.py
AMD=1 BFLOAT16=1 DEBUG=2 python3 extra/gemm/simple_matmul.py
- name: Run Tensor Core GEMM (AMD)
run: AMD=1 SHOULD_USE_TC=1 HALF=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul_amd.txt
run: AMD=1 HALF=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul_amd.txt
- name: Test AMD=1
run: DEBUG=2 AMD=1 python -m pytest -rA test/test_tiny.py
- name: Test HIP=1