-
Notifications
You must be signed in to change notification settings - Fork 8
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
Worse performance than pyerfa on Windows #57
Comments
Which compiler flags did you use? |
Looking to the logs of |
I used Visual Studio 2019 and set the target to 'x64-Clang-Release'. PSB a snippet of the ninja build output (build.ninja). Not sure if its helpful; I dont have much experience with build systems and compilation. build src\CMakeFiles\erfa.dir\apcg13.c.obj: C_COMPILER__erfa_RelWithDebInfo ......\src\apcg13.c || cmake_object_order_depends_target_erfa |
Thanks for posting the build log @dcherrie-dstl. I can see that your binaries were compiled with Out of curiosity, if you use the DLL from Clang and run the tests ( |
Yes, it's possible that Clang is doing a better job at building more performant binaries, but the factor 2-3x looks a bit surprising. Maybe GCC is being very conservative about what to optimise? I think that trying to build with |
Unfortunately, the CMake build system has not been merged, yet. I will give it a try locally on macOS. |
Ugh, right, I missed that 😅 |
Running the tests on macOS is virtually identical with |
On windows I have found ~2-3X worse performance compared to pyerfa. Using cmakelists (liberfa/erfa#75) and a clang compiler I was able to substitute the dll resulting in comparative (slightly better) performance.
The text was updated successfully, but these errors were encountered: