projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 295d4e9) | patch
Speed up getABITypeSize by turning a i64 mul and div into an
authorChris Lattner <sabre@nondot.org>
Mon, 8 Dec 2008 06:50:51 +0000 (06:50 +0000)
committerChris Lattner <sabre@nondot.org>
Mon, 8 Dec 2008 06:50:51 +0000 (06:50 +0000)
commit0a3ba71a0576fbde5b8ce1ff276f63f3f3d16229
tree8d4e141adf5312daa138c60f561a00ad0bdf8e93tree | snapshot
parent295d4e953a1214f60632220b9fcb31c1af8b0c27commit | diff
Speed up getABITypeSize by turning a i64 mul and div into an
AND.  This is speedup on any reasonable target, but particularly
on 32-bit targets where this often turns into a libcall like udivdi3.

We know that alignments are a power of two but the compiler doesn't.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@60688 91177308-0d34-0410-b5e6-96231b3b80d8
include/llvm/Target/TargetData.h diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors