diff options
| author | Johannes Doerfert <johannes@jdoerfert.de> | 2020-04-09 22:10:16 -0500 |
|---|---|---|
| committer | Johannes Doerfert <johannes@jdoerfert.de> | 2020-04-09 22:10:31 -0500 |
| commit | 17d83342235f01d4b110dc5d4664fe96f6597f11 (patch) | |
| tree | 45c5560414f7a646130169a34f8c9e2652a4753c /llvm/lib/Support/OptimizedStructLayout.cpp | |
| parent | 03ff643d2e9ebbf319d71b3a17d2ed0320a6a25b (diff) | |
| download | llvm-17d83342235f01d4b110dc5d4664fe96f6597f11.zip llvm-17d83342235f01d4b110dc5d4664fe96f6597f11.tar.gz llvm-17d83342235f01d4b110dc5d4664fe96f6597f11.tar.bz2 | |
[OpenMP] Allow <math.h> to go first in C++-mode in target regions
If we are in C++ mode and include <math.h> (not <cmath>) first, we still
need to make sure <cmath> is read first. The problem otherwise is that
we haven't seen the declarations of the math.h functions when the system
math.h includes our cmath overlay. However, our cmath overlay, or better
the underlying overlay, e.g. CUDA, uses the math.h functions. Since we
haven't declared them yet we get errors. CUDA avoids this by eagerly
declaring all math functions (in the __device__ space) but we cannot do
this. Instead we break the dependence by forcing cmath to go first.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D77774
Diffstat (limited to 'llvm/lib/Support/OptimizedStructLayout.cpp')
0 files changed, 0 insertions, 0 deletions