github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-04-29 03:00:14 -04:00
Code Issues Packages Projects Releases Wiki Activity

add support for "shared_mem" for LLVM (#10093)

Browse Source 
* init llvm shared

* add test_tensor_cores_emulation run for llvm
This commit is contained in:
Ignacio Sica
2025-04-29 09:56:36 -03:00
committed by GitHub
parent ad7546c931
commit 58cf8cd493
2 changed files with 9 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
.github/workflows/benchmark.yml vendored
View File

@@ -63,11 +63,10 @@ jobs:
run: BIG=2 MPS=1 python3.11 test/test_speed_v_torch.py | tee torch_speed.txt
- name: Test tensor cores
run: METAL=1 python3.11 test/test_linearizer.py TestLinearizer.test_tensor_cores TestLinearizer.test_tensor_cores_emulation TestLinearizer.test_tensor_cores_padded TestLinearizer.test_tensor_cores_padded_uops
# TODO: add TestLinearizer.test_tensor_cores_emulation for llvm (#10093)
- name: Test AMX tensor cores
run: |
DEBUG=2 CPU=1 AMX=1 python3.11 test/test_linearizer.py TestLinearizer.test_tensor_cores TestLinearizer.test_tensor_cores_emulation TestLinearizer.test_tensor_cores_padded TestLinearizer.test_tensor_cores_padded_uops
DEBUG=2 LLVM=1 AMX=1 python3.11 test/test_linearizer.py TestLinearizer.test_tensor_cores TestLinearizer.test_tensor_cores_padded TestLinearizer.test_tensor_cores_padded_uops
DEBUG=2 LLVM=1 AMX=1 python3.11 test/test_linearizer.py TestLinearizer.test_tensor_cores TestLinearizer.test_tensor_cores_emulation TestLinearizer.test_tensor_cores_padded TestLinearizer.test_tensor_cores_padded_uops
- name: Run Tensor Core GEMM (float)
run: DEBUG=2 SHOULD_USE_TC=1 python3.11 extra/gemm/simple_matmul.py | tee matmul.txt
- name: Run Tensor Core GEMM (half)