github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-01-10 07:28:15 -05:00
Code Issues Packages Projects Releases Wiki Activity

KFD GEMM (#4221)

Browse Source 
added to benchmark CI and fixed duplicated filenames between cuda and ptx
This commit is contained in:
chenyu
2024-04-19 00:43:18 -04:00
committed by GitHub
parent 3f3af0fb85
commit a1133beb80
2 changed files with 9 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
.github/workflows/benchmark.yml vendored
View File

@@ -110,12 +110,12 @@ jobs:
run: CUDA=1 python3 test/external/external_model_benchmark.py
- name: Test speed vs torch
run: CUDA=1 BIG=2 TORCHCUDA=1 python3 test/test_speed_v_torch.py | tee torch_speed.txt
- name: Run Tensor Core GEMM(CUDA)
- name: Run Tensor Core GEMM (CUDA)
run: |
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
- name: Run Tensor Core GEMM(PTX)
run: CUDA=1 PTX=1 HALF=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul.txt
- name: Run Tensor Core GEMM (PTX)
run: CUDA=1 PTX=1 HALF=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul_ptx.txt
- name: Run LLaMA
run: |
CUDA=1 JIT=0 python3 examples/llama.py --gen 1 --prompt "Hello." --count 10 --temperature 0 --timing | tee llama_unjitted.txt
@@ -148,6 +148,7 @@ jobs:
torch_speed.txt
matmul.txt
matmul_bfloat16.txt
matmul_ptx.txt
llama_unjitted.txt
llama_jitted.txt
llama_beam.txt
@@ -192,8 +193,10 @@ jobs:
# run: |
# python3 -c "import torch; print(torch.__version__)"
# LD_PRELOAD="/opt/rocm/lib/libhsa-runtime64.so" HSA=1 BIG=2 TORCHCUDA=1 python3 test/test_speed_v_torch.py | tee torch_speed.txt
- name: Run Tensor Core GEMM
- name: Run Tensor Core GEMM (HSA)
run: HSA=1 HALF=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul.txt
- name: Run Tensor Core GEMM (KFD)
run: KFD=1 HALF=1 DEBUG=2 python3 extra/gemm/simple_matmul.py | tee matmul_kfd.txt
- name: Run Stable Diffusion
run: HSA=1 python3 examples/stable_diffusion.py --seed 0 --noshow --timing | tee sd.txt
- name: Run LLaMA 7B
@@ -229,6 +232,7 @@ jobs:
gpt2_unjitted.txt
gpt2_jitted.txt
matmul.txt
matmul_kfd.txt
sd.txt
mixtral.txt