Skip to content

Speedup Warmup on ROCm #24

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
michaelmckinsey1 wants to merge 6 commits into
base: main
Choose a base branch
from
Draft

Speedup Warmup on ROCm #24

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
32b2fd9
Update scaffold-tuolumne.job
michaelmckinsey1 Mar 9, 2026
73c6d12
Update scaffold-tuolumne-torchpypi.job
michaelmckinsey1 Mar 9, 2026
3a8f5e0
Update scaffold-tuolumne.job
michaelmckinsey1 Mar 13, 2026
2cc5932
Update scaffold-tuolumne-torchpypi.job
michaelmckinsey1 Mar 13, 2026
69aa4ff
Update scaffold-tuolumne-torchpypi.job
michaelmckinsey1 Mar 20, 2026
263b4a9
Update scaffold-tuolumne.job
michaelmckinsey1 Mar 20, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 10 additions & 0 deletions scripts/scaffold-tuolumne-torchpypi.job
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,16 @@ ml cce/21.0.0 cray-mpich/9.1.0 rocm/7.1.0 rccl/fast-env-slows-mpi
# Use ccl plugin that we manually built with install-rccl.sh
export NCCL_NET_PLUGIN=../aws-ofi-nccl.git/install/lib/librccl-net.so

# Disable direct convolution benchmarking (should speedup warmup by a significant amount)
# export MIOPEN_DEBUG_CONV_DIRECT=0

# Disable direct naive convolution benchmarking (naive_conv_ab_nonpacked_fwd_ndhwc_half_double_half.kd)
export MIOPEN_DEBUG_CONV_DIRECT_NAIVE_CONV_FWD=0
# Disable naive_conv_ab_nonpacked_bwd_ndhwc_half_double_half.kd
export MIOPEN_DEBUG_CONV_DIRECT_NAIVE_CONV_BWD=0
# Disable naive_conv_ab_nonpacked_wrw_ndhwc_half_double_half.kd
export MIOPEN_DEBUG_CONV_DIRECT_NAIVE_CONV_WRW=0

torchrun-hpc -N 1 -n 1 $(which scaffold) generate_fractals -c $(pwd)/ScaFFold/configs/benchmark_default.yml

# Uncomment if you want torch profiling
Expand Down
10 changes: 10 additions & 0 deletions scripts/scaffold-tuolumne.job
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,16 @@ ml cce/21.0.0 cray-mpich/9.1.0 rocm/7.1.0 rccl/fast-env-slows-mpi
# (2) Removing libmpi may cause segfault on mpi4py import
export LD_PRELOAD="/opt/rocm-7.1.0/llvm/lib/libomp.so /opt/cray/pe/mpich/9.1.0/ofi/gnu/11.2/lib/libmpi_gnu.so.12"

# Disable direct convolution benchmarking (should speedup warmup by a significant amount)
# export MIOPEN_DEBUG_CONV_DIRECT=0

# Disable direct naive convolution benchmarking (naive_conv_ab_nonpacked_fwd_ndhwc_half_double_half.kd)
export MIOPEN_DEBUG_CONV_DIRECT_NAIVE_CONV_FWD=0
# Disable naive_conv_ab_nonpacked_bwd_ndhwc_half_double_half.kd
export MIOPEN_DEBUG_CONV_DIRECT_NAIVE_CONV_BWD=0
# Disable naive_conv_ab_nonpacked_wrw_ndhwc_half_double_half.kd
export MIOPEN_DEBUG_CONV_DIRECT_NAIVE_CONV_WRW=0

torchrun-hpc -N 1 -n 1 $(which scaffold) generate_fractals -c $(pwd)/ScaFFold/configs/benchmark_default.yml

# Uncomment if you want torch profiling
Expand Down
Loading