- Apr 17, 2006
-
-
Chris Lattner authored
llvm-svn: 27743
-
Chris Lattner authored
llvm-svn: 27742
-
Chris Lattner authored
llvm-svn: 27741
-
Chris Lattner authored
llvm-svn: 27740
-
Chris Lattner authored
of various 4-element vectors. llvm-svn: 27739
-
Chris Lattner authored
llvm-svn: 27738
-
Chris Lattner authored
llvm-svn: 27737
-
Chris Lattner authored
Altivec vectors. llvm-svn: 27736
-
- Apr 16, 2006
-
-
Evan Cheng authored
llvm-svn: 27735
-
Evan Cheng authored
llvm-svn: 27734
-
Evan Cheng authored
llvm-svn: 27733
-
Evan Cheng authored
address has to be 16-byte aligned but the values aren't spilled to 128-bit locations. llvm-svn: 27732
-
Chris Lattner authored
one type (v4i32) so that we don't have to write patterns for each type, and so that more CSE opportunities are exposed. llvm-svn: 27731
-
Chris Lattner authored
to write one pattern for vector stores instead of 4. llvm-svn: 27730
-
Chris Lattner authored
Remove some done items from the todo list. llvm-svn: 27729
-
Chris Lattner authored
Make the insert/extract elt -> shuffle code more aggressive. This fixes CodeGen/PowerPC/vec_shuffle.ll llvm-svn: 27728
-
Chris Lattner authored
llvm-svn: 27727
-
Chris Lattner authored
llvm-svn: 27726
-
Chris Lattner authored
go away when I start using evan's binop type canonicalizer llvm-svn: 27725
-
Chris Lattner authored
bitconverted from some other type. llvm-svn: 27724
-
Chris Lattner authored
buildvectors. llvm-svn: 27723
-
- Apr 15, 2006
-
-
Evan Cheng authored
llvm-svn: 27722
-
Evan Cheng authored
llvm-svn: 27721
-
Evan Cheng authored
llvm-svn: 27720
-
Evan Cheng authored
llvm-svn: 27719
-
Evan Cheng authored
llvm-svn: 27718
-
Chris Lattner authored
maximal shuffles out of them where possible. llvm-svn: 27717
-
Evan Cheng authored
llvm-svn: 27716
-
Evan Cheng authored
llvm-svn: 27715
-
Chris Lattner authored
llvm-svn: 27714
-
Chris Lattner authored
llvm-svn: 27713
-
Chris Lattner authored
These instructions always return a packed vector. Improve the class definitions to expose this fact. llvm-svn: 27712
-
- Apr 14, 2006
-
-
Evan Cheng authored
llvm-svn: 27711
-
Chris Lattner authored
insert/extractelement operations. This implements Transforms/ScalarRepl/vector_promote.ll llvm-svn: 27710
-
Chris Lattner authored
void test(vector float *F, float f) { vector float G = *F + *F; *((float*)&G) = f; *F = G + G; } void test2(vector float *F, float f) { vector float G = *F + *F; ((float*)&G)[2] = f; *F = G + G; } void test3(vector float *F, float *f) { vector float G = *F + *F; *f = ((float*)&G)[2]; } void test4(vector float *F, float *f) { vector float G = *F + *F; *f = *((float*)&G); } into insert/extract element operations with no memory traffic. llvm-svn: 27709
-
Tanya Lattner authored
llvm-svn: 27704
-
Tanya Lattner authored
llvm-svn: 27703
-
Reid Spencer authored
that they are "Constant" as they derive from GlobalValue. Also, fix some of the wording where it mentions this. Patch inspired by Nai Xia. llvm-svn: 27701
-
Reid Spencer authored
of the external tests. This turns into --with-externals option to configure. llvm-svn: 27700
-
Evan Cheng authored
llvm-svn: 27699
-