2009-09-11 20:01:28 +02:00
|
|
|
; RUN: opt < %s -instcombine -S | grep {GLOBAL.*align 16}
|
|
|
|
|
; RUN: opt < %s -instcombine -S | grep {tmp = load}
|
2009-11-03 16:29:06 +01:00
|
|
|
target datalayout = "E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-v64:64:64-v128:128:128"
|
when we see a unaligned load from an insufficiently aligned global or
alloca, increase the alignment of the load, turning it into an aligned load.
This allows us to compile:
#include <xmmintrin.h>
__m128i foo(__m128i x){
static const unsigned int c_0[4] = { 0, 0, 0, 0 };
__m128i v_Zero = _mm_loadu_si128((__m128i*)c_0);
x = _mm_unpacklo_epi8(x, v_Zero);
return x;
}
into:
_foo:
punpcklbw _c_0.5944, %xmm0
ret
.data
.lcomm _c_0.5944,16,4 # c_0.5944
instead of:
_foo:
movdqu _c_0.5944, %xmm1
punpcklbw %xmm1, %xmm0
ret
.data
.lcomm _c_0.5944,16,2 # c_0.5944
llvm-svn: 40971
2007-08-09 21:05:49 +02:00
|
|
|
|
2007-08-12 18:55:14 +02:00
|
|
|
@GLOBAL = internal global [4 x i32] zeroinitializer
|
when we see a unaligned load from an insufficiently aligned global or
alloca, increase the alignment of the load, turning it into an aligned load.
This allows us to compile:
#include <xmmintrin.h>
__m128i foo(__m128i x){
static const unsigned int c_0[4] = { 0, 0, 0, 0 };
__m128i v_Zero = _mm_loadu_si128((__m128i*)c_0);
x = _mm_unpacklo_epi8(x, v_Zero);
return x;
}
into:
_foo:
punpcklbw _c_0.5944, %xmm0
ret
.data
.lcomm _c_0.5944,16,4 # c_0.5944
instead of:
_foo:
movdqu _c_0.5944, %xmm1
punpcklbw %xmm1, %xmm0
ret
.data
.lcomm _c_0.5944,16,2 # c_0.5944
llvm-svn: 40971
2007-08-09 21:05:49 +02:00
|
|
|
|
|
|
|
|
define <16 x i8> @foo(<2 x i64> %x) {
|
|
|
|
|
entry:
|
2009-02-16 01:44:23 +01:00
|
|
|
%tmp = load <16 x i8>* bitcast ([4 x i32]* @GLOBAL to <16 x i8>*), align 1
|
when we see a unaligned load from an insufficiently aligned global or
alloca, increase the alignment of the load, turning it into an aligned load.
This allows us to compile:
#include <xmmintrin.h>
__m128i foo(__m128i x){
static const unsigned int c_0[4] = { 0, 0, 0, 0 };
__m128i v_Zero = _mm_loadu_si128((__m128i*)c_0);
x = _mm_unpacklo_epi8(x, v_Zero);
return x;
}
into:
_foo:
punpcklbw _c_0.5944, %xmm0
ret
.data
.lcomm _c_0.5944,16,4 # c_0.5944
instead of:
_foo:
movdqu _c_0.5944, %xmm1
punpcklbw %xmm1, %xmm0
ret
.data
.lcomm _c_0.5944,16,2 # c_0.5944
llvm-svn: 40971
2007-08-09 21:05:49 +02:00
|
|
|
ret <16 x i8> %tmp
|
|
|
|
|
}
|
|
|
|
|
|
