-
Notifications
You must be signed in to change notification settings - Fork 3.8k
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.
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
Enable MM_PREFETCH and MM_MALLOC on aarch64 #4124
Comments
Closed in favor of being in #2302. We decided to keep all feature requests in one place. Welcome to contribute this feature! Please re-open this issue (or post a comment if you are not a topic starter) if you are actively working on implementing this feature. |
See the error at https://clang.llvm.org/doxygen/xmmintrin_8h_source.html.
This can be reproduced like this: // conftest.cpp
#include <xmmintrin.h>
void main() {
int a = 0;
_mm_prefetch(&a, _MM_HINT_NTA);
return 0;
} clang++ -arch arm64 -std=gnu++17 -o conftest conftest.cpp |
I'm not an expert in all these things at all, but it looks like Arm Compiler supports prefetching via UPD: ... and gcc: https://github.com/gcc-mirror/gcc/blob/master/gcc/testsuite/gcc.target/aarch64/vect-prefetch-drop.c |
Linking #4331 and #6514 (comment). |
Refer to #3948 (comment).
Related source codes:
LightGBM/CMakeLists.txt
Lines 243 to 271 in e98da99
LightGBM/include/LightGBM/meta.h
Lines 16 to 23 in e98da99
LightGBM/include/LightGBM/utils/common.h
Lines 44 to 58 in e98da99
The text was updated successfully, but these errors were encountered: