diff options
| author | Justin Lebar <jlebar@google.com> | 2016-05-19 22:49:13 +0000 |
|---|---|---|
| committer | Justin Lebar <jlebar@google.com> | 2016-05-19 22:49:13 +0000 |
| commit | 2e4ecfdebe8fa73ab4ed6f738307339ee9586418 (patch) | |
| tree | a8eb6c205f01da79d919bb3f573cb92f11100a40 /clang/lib/CodeGen/CGCXXABI.cpp | |
| parent | b926bdac4c18e0f31d827dec482f207856e88e1e (diff) | |
| download | bcm5719-llvm-2e4ecfdebe8fa73ab4ed6f738307339ee9586418.tar.gz bcm5719-llvm-2e4ecfdebe8fa73ab4ed6f738307339ee9586418.zip | |
[CUDA] Implement __ldg using intrinsics.
Summary:
Previously it was implemented as inline asm in the CUDA headers.
This change allows us to use the [addr+imm] addressing mode when
executing ld.global.nc instructions. This translates into a 1.3x
speedup on some benchmarks that call this instruction from within an
unrolled loop.
Reviewers: tra, rsmith
Subscribers: jhen, cfe-commits, jholewinski
Differential Revision: http://reviews.llvm.org/D19990
llvm-svn: 270150
Diffstat (limited to 'clang/lib/CodeGen/CGCXXABI.cpp')
0 files changed, 0 insertions, 0 deletions
