Thanks to visit codestin.com
Credit goes to github.com

Skip to content

DNN: try to fix the sporadic crashes in perf_dnn #23511

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

Merged
merged 1 commit into from
Apr 22, 2023
Merged

DNN: try to fix the sporadic crashes in perf_dnn #23511

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
5 changes: 3 additions & 2 deletions modules/dnn/src/layers/cpu_kernels/convolution.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -456,7 +456,8 @@ void runFastConv(InputArray _input, OutputArray _output, const Ptr<FastConv>& co
int dilation_d = conv->dilation_d, dilation_h = conv->dilation_h, dilation_w = conv->dilation_w;

int ksize = Dk*Hk*Wk;
bool fast_1x1 = ksize == 1 && stride_d == 1 && stride_w == 1 && stride_h == 1;
bool fast_1x1 = ksize == 1 && stride_d == 1 && stride_w == 1 && stride_h == 1
&& pad_front == 0 && pad_left == 0 && pad_top == 0;
int DkHkWkCg = Dk*Hk*Wk*Cg;

std::vector<int> ofstab_(Hk*Wk*Dk*4, 0);
Expand Down Expand Up @@ -502,7 +503,7 @@ void runFastConv(InputArray _input, OutputArray _output, const Ptr<FastConv>& co
}
}

int MAX_STRIPES = (56 + CONV_NR - 1)/CONV_NR;
int MAX_STRIPES = conv->conv_type == CONV_TYPE_DEPTHWISE_REMAIN ? 1 : (56 + CONV_NR - 1)/CONV_NR;
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

(56 + CONV_NR - 1)/CONV_NR

BTW, there is divUp() call for that


// Friendly to L1 cache
const int K_BLOCK_SIZE = conv->conv_type == CONV_TYPE_DEPTHWISE_REMAIN ? 1 : 32;
Expand Down
1 change: 1 addition & 0 deletions modules/dnn/test/test_onnx_importer.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -256,6 +256,7 @@ TEST_P(Test_ONNX_layers, Convolution3D_bias)
applyTestTag(CV_TEST_TAG_DNN_SKIP_CUDA_FP16);
}
testONNXModels("conv3d_bias");
testONNXModels("conv3d_depthwise_bias"); // kernel 1x1
}

TEST_P(Test_ONNX_layers, Two_convolution)
Expand Down