Initial code merge of Hopper support (#2036)

The initial code merge of Nvidia Hopper features support. Please be
aware that the code merge is not finished yet and the trouble-shooting
is still ongoing. The new hardware features (GMMA, TMA, STMATRIX etc.)
and automatic warp-specialization are experimental for now and turned
off by default. It is recommended for a trial when version 3.0 is
released.

The work is contributed by:
ben-zhang-609, bealwang, donproc, qliu93, jsh20, allatit23, LyricZhao,
ivanyinwz, goostavz & yangjunpro
from Nvidia, in cooperation with:
ptillet, Jokeren, ThomasRaoux & zahimoud
from OpenAI.

Co-authored-by: Goostav Zhu <gzhu@nvidia.com>
This commit is contained in:
goostavz
2023-08-07 09:53:04 +08:00
committed by GitHub
parent 5df904233c
commit f1512bded1
220 changed files with 28448 additions and 2295 deletions

View File

@@ -50,6 +50,8 @@ jobs:
if: ${{(matrix.runner[0] == 'self-hosted') && (matrix.runner[1] == 'V100' || matrix.runner[1] == 'A100' || matrix.runner[1] == 'H100')}}
run: |
echo "BACKEND=CUDA" >> "${GITHUB_ENV}"
echo "ENABLE_TMA=0" >> "${GITHUB_ENV}"
echo "ENABLE_MMA_V3=0" >> "${GITHUB_ENV}"
- name: Clear cache
run: |
@@ -79,8 +81,32 @@ jobs:
fi
lit -v "${LIT_TEST_DIR}"
- name: Run python tests on CUDA
if: ${{ env.BACKEND == 'CUDA'}}
- name: Enable MMAV3 and TMA
if: ${{(matrix.runner[0] == 'self-hosted') && (matrix.runner[1] == 'H100')}}
run: |
echo "ENABLE_TMA=1" >> "${GITHUB_ENV}"
echo "ENABLE_MMA_V3=1" >> "${GITHUB_ENV}"
- name: Run python tests on CUDA with ENABLE_TMA=1 and ENABLE_MMA_V3=1
if: ${{ env.BACKEND == 'CUDA' && env.ENABLE_TMA == '1' && env.ENABLE_MMA_V3 == '1'}}
run: |
cd python/test/unit
python3 -m pytest -n 8 --ignore=runtime
# run runtime tests serially to avoid race condition with cache handling.
python3 -m pytest runtime/
- name: Disable MMAV3 and TMA
if: ${{(matrix.runner[0] == 'self-hosted') && (matrix.runner[1] == 'H100')}}
run: |
echo "ENABLE_TMA=0" >> "${GITHUB_ENV}"
echo "ENABLE_MMA_V3=0" >> "${GITHUB_ENV}"
- name: Clear cache
run: |
rm -rf ~/.triton
- name: Run python tests on CUDA with ENABLE_TMA=0 and ENABLE_MMA_V3=0
if: ${{ env.BACKEND == 'CUDA' && env.ENABLE_TMA == '0' && env.ENABLE_MMA_V3 == '0'}}
run: |
cd python/test/unit
python3 -m pytest -n 8 --ignore=runtime