[llvm-commits] CVS: llvm/lib/Target/X86/README-SSE.txt
Chris Lattner
sabre at nondot.org
Sun Dec 10 17:01:18 PST 2006
Changes in directory llvm/lib/Target/X86:
README-SSE.txt updated: 1.10 -> 1.11
---
Log message:
this is done
---
Diffs of the changes: (+0 -27)
README-SSE.txt | 27 ---------------------------
1 files changed, 27 deletions(-)
Index: llvm/lib/Target/X86/README-SSE.txt
diff -u llvm/lib/Target/X86/README-SSE.txt:1.10 llvm/lib/Target/X86/README-SSE.txt:1.11
--- llvm/lib/Target/X86/README-SSE.txt:1.10 Tue Dec 5 12:25:10 2006
+++ llvm/lib/Target/X86/README-SSE.txt Sun Dec 10 19:01:03 2006
@@ -527,33 +527,6 @@
//===---------------------------------------------------------------------===//
-Implement some missing insert/extract element operations without going through
-the stack. Testcase here:
-CodeGen/X86/vec_ins_extract.ll
-corresponds to this C code:
-
-typedef float vectorfloat __attribute__((vector_size(16)));
-void test(vectorfloat *F, float f) {
- vectorfloat G = *F + *F;
- *((float*)&G) = f;
- *F = G + G;
-}
-void test2(vectorfloat *F, float f) {
- vectorfloat G = *F + *F;
- ((float*)&G)[2] = f;
- *F = G + G;
-}
-void test3(vectorfloat *F, float *f) {
- vectorfloat G = *F + *F;
- *f = ((float*)&G)[2];
-}
-void test4(vectorfloat *F, float *f) {
- vectorfloat G = *F + *F;
- *f = *((float*)&G);
-}
-
-//===---------------------------------------------------------------------===//
-
Apply the same transformation that merged four float into a single 128-bit load
to loads from constant pool.
More information about the llvm-commits
mailing list