diff options
| author | Chris Lattner <sabre@nondot.org> | 2006-04-17 21:52:03 +0000 |
|---|---|---|
| committer | Chris Lattner <sabre@nondot.org> | 2006-04-17 21:52:03 +0000 |
| commit | 63a5cdc423f23a78a0a8f3f8a3eea0e5338f9ddb (patch) | |
| tree | 16a4db8ced4ed471b5ab823574f8292bcb66ef1b /llvm/lib | |
| parent | 6bd68ae81e64f823a293f616fd088b3552c86e30 (diff) | |
| download | bcm5719-llvm-63a5cdc423f23a78a0a8f3f8a3eea0e5338f9ddb.tar.gz bcm5719-llvm-63a5cdc423f23a78a0a8f3f8a3eea0e5338f9ddb.zip | |
remove done item
llvm-svn: 27778
Diffstat (limited to 'llvm/lib')
| -rw-r--r-- | llvm/lib/Target/PowerPC/README_ALTIVEC.txt | 21 |
1 files changed, 2 insertions, 19 deletions
diff --git a/llvm/lib/Target/PowerPC/README_ALTIVEC.txt b/llvm/lib/Target/PowerPC/README_ALTIVEC.txt index 2887b798b0e..2d91986f83f 100644 --- a/llvm/lib/Target/PowerPC/README_ALTIVEC.txt +++ b/llvm/lib/Target/PowerPC/README_ALTIVEC.txt @@ -5,8 +5,8 @@ registers, to generate better spill code. //===----------------------------------------------------------------------===// -Altivec support. The first should be a single lvx from the constant pool, the -second should be a xor/stvx: +The first should be a single lvx from the constant pool, the second should be +a xor/stvx: void foo(void) { int x[8] __attribute__((aligned(128))) = { 1, 1, 1, 17, 1, 1, 1, 1 }; @@ -39,23 +39,6 @@ a load/store/lve*x sequence. //===----------------------------------------------------------------------===// -There are a wide range of vector constants we can generate with combinations of -altivec instructions. - -Examples, these work with all widths: - Splat(+/- 16,18,20,22,24,28,30): t = vspliti I/2, r = t+t - Splat(+/- 17,19,21,23,25,29): t = vsplti +/-15, t2 = vsplti I-15, r=t + t2 - Splat(31): t = vsplti FB, r = srl t,t - Splat(256): t = vsplti 1, r = vsldoi t, t, 1 - -Lots more are listed here: -http://www.informatik.uni-bremen.de/~hobold/AltiVec.html - -This should be added to the ISD::BUILD_VECTOR case in -PPCTargetLowering::LowerOperation. - -//===----------------------------------------------------------------------===// - FABS/FNEG can be codegen'd with the appropriate and/xor of -0.0. //===----------------------------------------------------------------------===// |
