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

[x86] Depthwise conv2d #6745

Merged
merged 14 commits into from
Sep 10, 2021
Merged

[x86] Depthwise conv2d #6745

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
14 commits
Select commit Hold shift + click to select a range
b1fd731
add x86 depthwise_conv 3x3s1p1 3x3s2p1 test=develop
lxwlaq Aug 20, 2021
4da6c5e
merge from develop test=develop
lxwlaq Aug 20, 2021
08e2df6
Fix pre-commit error test=develop
lxwlaq Aug 22, 2021
71c805c
Fix lite/kernels/x86/CMakeLists.txt confict test=develop
lxwlaq Aug 23, 2021
e27c25f
fix lite/kernels/x86/CMakeLists.txt conflict test=develop
lxwlaq Aug 23, 2021
351522a
optimize mask about left and right processing test=develop
lxwlaq Aug 28, 2021
5037411
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle-Lite…
lxwlaq Aug 28, 2021
6716de2
fix pre-commit
lxwlaq Aug 29, 2021
e3530e1
fix lite/kernels/x86/conv_depthwise.cc bug
lxwlaq Aug 29, 2021
d627821
fix CMakeLists.txt bug
lxwlaq Sep 1, 2021
c4bc9ef
fix pre-commit bug
lxwlaq Sep 2, 2021
1061670
fix pre-commmit bug
lxwlaq Sep 2, 2021
4a3bffa
Merge branch 'develop' into depthwise_conv2d
lxwlaq Sep 6, 2021
7a86db4
fix conflict
lxwlaq Sep 7, 2021
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
9 changes: 9 additions & 0 deletions lite/backends/x86/math/avx/avx_mathfuns.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,6 +32,15 @@ void sincos256_ps(v8sf x, v8sf *s, v8sf *c);
// FMA support
#ifndef __AVX2__
#define _mm256_fmadd_ps(a, b, c) _mm256_add_ps(c, _mm256_mul_ps(a, b))
#define _mm256_permutevar8x32_ps(a, b) \
_mm256_setr_ps(*((float *)(&a) + *((int *)(&b))), \
*((float *)(&a) + *((int *)(&b) + 1)), \
*((float *)(&a) + *((int *)(&b) + 2)), \
*((float *)(&a) + *((int *)(&b) + 3)), \
*((float *)(&a) + *((int *)(&b) + 4)), \
*((float *)(&a) + *((int *)(&b) + 5)), \
*((float *)(&a) + *((int *)(&b) + 6)), \
*((float *)(&a) + *((int *)(&b) + 7)))
#endif

} // namespace math
Expand Down
Loading