Skip to content
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

Jump to bottom

patching nvfuser conv cudnn test numerics mismatch #2048

Merged
merged 2 commits into from
Oct 11, 2022
Merged

patching nvfuser conv cudnn test numerics mismatch #2048

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
8535126
patching nvfuser conv cudnn test numerics mismatch
jjsjann123 Oct 10, 2022
76a8ce3
lintrunner
jjsjann123 Oct 11, 2022
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
1 change: 1 addition & 0 deletions torch/csrc/jit/codegen/cuda/test/test_gpu_shift.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2976,6 +2976,7 @@ TEST_F(NVFuserTest, FusionConv2D_CUDA) {
TEST_F(NVFuserTest, FusionConv2DNoPadding_CUDA) {
Fusion fusion;
FusionGuard fg(&fusion);
ContextCudnnTF32Disabled disabling_tf32_cudnn;

// Input: [C, H, W]
auto inp = makeSymbolicTensor(3);
Expand Down
16 changes: 16 additions & 0 deletions torch/csrc/jit/codegen/cuda/test/test_utils.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,6 +8,7 @@
#include <torch/csrc/jit/codegen/cuda/lower_magic_zero.h>
#include <torch/csrc/jit/codegen/cuda/transform_replay.h>

#include <ATen/Context.h>
#include <ATen/cuda/CUDAContext.h>
#include <c10/cuda/CUDACachingAllocator.h>
#include <torch/torch.h>
Expand Down Expand Up @@ -340,6 +341,21 @@ struct TransformPropagatorWithCheck : public TransformPropagator {

} // namespace

class ContextCudnnTF32Disabled {
public:
ContextCudnnTF32Disabled() {
flag_ = at::globalContext().allowTF32CuDNN();
at::globalContext().setAllowTF32CuDNN(false);
}

~ContextCudnnTF32Disabled() {
at::globalContext().setAllowTF32CuDNN(flag_);
}

private:
bool flag_;
};

// Fixture class must be uniquely identified, i.e., can't be in an
// anonymous namespace
class NVFuserTest : public ::testing::Test {
Expand Down