diff options
author | Chris Lattner <sabre@nondot.org> | 2006-04-18 03:57:35 +0000 |
---|---|---|
committer | Chris Lattner <sabre@nondot.org> | 2006-04-18 03:57:35 +0000 |
commit | 19a815238e55458e95f99b4dad31ed053c9f635c (patch) | |
tree | d7a1e0c59ae8e186db779fc4787641557f9e81f9 /lib/Target/PowerPC/README_ALTIVEC.txt | |
parent | a637e589187ae92302fc33645df903507b33d262 (diff) | |
download | external_llvm-19a815238e55458e95f99b4dad31ed053c9f635c.zip external_llvm-19a815238e55458e95f99b4dad31ed053c9f635c.tar.gz external_llvm-19a815238e55458e95f99b4dad31ed053c9f635c.tar.bz2 |
Implement v16i8 multiply with this code:
vmuloub v5, v3, v2
vmuleub v2, v3, v2
vperm v2, v2, v5, v4
This implements CodeGen/PowerPC/vec_mul.ll. With this, v16i8 multiplies are
6.79x faster than before.
Overall, UnitTests/Vector/multiplies.c is now 2.45x faster with LLVM than with
GCC.
Remove the 'integer multiplies' todo from the README file.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27792 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'lib/Target/PowerPC/README_ALTIVEC.txt')
-rw-r--r-- | lib/Target/PowerPC/README_ALTIVEC.txt | 9 |
1 files changed, 0 insertions, 9 deletions
diff --git a/lib/Target/PowerPC/README_ALTIVEC.txt b/lib/Target/PowerPC/README_ALTIVEC.txt index 2d91986..9de4415 100644 --- a/lib/Target/PowerPC/README_ALTIVEC.txt +++ b/lib/Target/PowerPC/README_ALTIVEC.txt @@ -75,15 +75,6 @@ be constants. The verifier should enforce this constraint. //===----------------------------------------------------------------------===// -Implement multiply for vector integer types, to avoid the horrible scalarized -code produced by legalize. - -void test(vector int *X, vector int *Y) { - *X = *X * *Y; -} - -//===----------------------------------------------------------------------===// - extract_vector_elt of an arbitrary constant vector can be done with the following instructions: |