- Jun 24, 2010
-
-
Jakob Stoklund Olesen authored
CoalescerPair can determine if a copy can be coalesced, and which register gets merged away. The old logic in SimpleRegisterCoalescing had evolved into something a bit too convoluted. This second attempt fixes some crashes that only occurred Linux. llvm-svn: 106769
-
Dan Gohman authored
llvm-svn: 106768
-
Jakob Stoklund Olesen authored
[L]oad, [u]se, [d]ef, or [S]tore slots. This makes it easier to see if two indices refer to the same instruction, avoiding mental mod 4 calculations. llvm-svn: 106766
-
Dan Gohman authored
which trivially fold away. llvm-svn: 106765
-
Dan Gohman authored
llvm-svn: 106764
-
Dan Gohman authored
enough special case, and it theoretically allows more folding because it works even when x is unanalyzable. llvm-svn: 106763
-
Bob Wilson authored
form so they can be narrowed to 16-bit instructions. llvm-svn: 106762
-
Bob Wilson authored
None of the existing implementations of commuteInstruction create new instructions unless the NewMI parameter is true, but the comment had implied otherwise. findCommutedOpIndices returns false, not true, when it doesn't know how to commute the instruction. llvm-svn: 106761
-
Dan Gohman authored
was over-complicated. llvm-svn: 106760
-
Dan Gohman authored
llvm-svn: 106759
-
Fariborz Jahanian authored
Thanks Daniel D. llvm-svn: 106758
-
Dan Gohman authored
handling of pointer types. llvm-svn: 106757
-
Dan Gohman authored
llvm-svn: 106756
-
Ted Kremenek authored
llvm-svn: 106755
-
Dan Gohman authored
instead of a StringRef, avoiding the need to copy the string in the common case. llvm-svn: 106754
-
Jakob Stoklund Olesen authored
Also keep track of the original DstREg before subregister adjustments. llvm-svn: 106753
-
Gabor Greif authored
llvm-svn: 106752
-
Jakob Stoklund Olesen authored
In this case it is essential that the kill is real because the spiller will decide to omit a spill if it thinks there is a later kill. llvm-svn: 106751
-
Gabor Greif authored
llvm-svn: 106750
-
Dan Gohman authored
as many tests as possible. llvm-svn: 106749
-
Dan Gohman authored
when the condition is constant. This optimization shouldn't be necessary, because codegen shouldn't be able to find dead control paths that the IR-level optimizer can't find. And it's undesirable, because it encourages bugpoint to leave "br i1 false" branches in its output. And it wasn't updating the CFG. I updated all the tests I could, but some tests are too reduced and I wasn't able to meaningfully preserve them. llvm-svn: 106748
-
Gabor Greif authored
llvm-svn: 106747
-
Dan Gohman authored
llvm-svn: 106746
-
Gabor Greif authored
llvm-svn: 106745
-
Gabor Greif authored
llvm-svn: 106744
-
Gabor Greif authored
llvm-svn: 106743
-
Zhongxing Xu authored
llvm-svn: 106742
-
Zhongxing Xu authored
llvm-svn: 106741
-
Gabor Greif authored
llvm-svn: 106740
-
Zhongxing Xu authored
llvm-svn: 106738
-
Gabor Greif authored
llvm-svn: 106737
-
Gabor Greif authored
llvm-svn: 106736
-
Gabor Greif authored
llvm-svn: 106734
-
Benjamin Kramer authored
The constant was copied from clang's limits.h. llvm-svn: 106732
-
Gabor Greif authored
llvm-svn: 106731
-
Gabor Greif authored
llvm-svn: 106730
-
Gabor Greif authored
llvm-svn: 106729
-
Gabor Greif authored
llvm-svn: 106728
-
Argyrios Kyrtzidis authored
llvm-svn: 106727
-
Argyrios Kyrtzidis authored
llvm-svn: 106726
-