- User Since
- Nov 1 2016, 4:55 AM (232 w, 2 h)
Oct 26 2020
Oct 23 2020
Oct 22 2020
Oct 21 2020
I've successfully applied your change in Cuda.cpp to the following clang versions: 10.0.0, 10.0.1, 11.0.0, and 12.0.0git; and have created the working patches (#206).
Oct 20 2020
Oct 29 2019
Oct 28 2019
It looks like CUDA doesn't support double argument for device function __isnan on FreeBSD.
It's actually the opposite -- FreeBSD does not provide *host*-side __isnan(double) -- the error complains that it's the host code that tried to use __isnan and failed when overload resolution produced a device variant.
Aug 27 2019
Aug 23 2019
Aug 22 2019
Apr 18 2019
Apr 17 2019
Apr 5 2019
Oooh, sorry, but I've just pushed the fix. But with the following words: "Add missing long double device functions' declarations. Provide only declarations to prevent any use of long double on the device side, because CUDA does not support long double on the device side."
Provide only declarations for missing long double device functions to prevent any use of long double on the device side, because CUDA does not support long double on the device side.
Apr 3 2019
Mar 15 2019
The intent is to avoid unintentional clashes with the preprocessor macros the user may have defined.
Use of __ prefix is reserved for internal compiler use.
It is generally known. What is unknown, at least for me, is why in those clang's headers __ prefix is used so heavily?
As long as _sptr is also a keyword reserved by Microsoft, I agree, that float *__s and float *__c are a good choice. Moreover, there are also double *__b and int *__c presented as device functions' arguments.
Perhaps for consistency sake it would be better to replace __sptr -> __s and __cptr -> __c.
Well, it came from NVIDIA code, you know, I mean all those double underscores. As long as _sptr is also a keyword reserved by Microsoft, I think __s and __c are a good choice. Moreover, there are also double *__b and int *__c presented as device functions' arguments.
Mar 14 2019
Sep 25 2017
Sep 21 2017
Sep 5 2017
Sep 1 2017
Aug 30 2017
Ooops, you right, you've already fixed it with:
Aug 28 2017
Aug 24 2017
Test is added.
Aug 23 2017