aaron/ispc
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Fixed a number of issues related to memory alignment; a number of places

Browse Source 
were expecting vector-width-aligned pointers where in point of fact,
there's no guarantee that they would have been in general.

Removed the aligned memory allocation routines from some of the examples;
they're no longer needed.

No perf. difference on Core2/Core i5 CPUs; older CPUs may see some
regressions.

Still need to update the documentation for this change and finish reviewing
alignment issues in Load/Store instructions generated by .cpp files.
This commit is contained in:
Matt Pharr
2011-06-23 18:18:33 -07:00
parent d340dcbfcc
commit b84167dddd
11 changed files with 45 additions and 112 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
stdlib-avx.ll
View File

@@ -513,14 +513,14 @@ declare <8 x float> @llvm.x86.avx.blendvps(<8 x float>, <8 x float>,
define void @__masked_store_blend_32(<8 x i32>* nocapture, <8 x i32>,
<8 x i32>) nounwind alwaysinline {
%mask_as_float = bitcast <8 x i32> %2 to <8 x float>
%oldValue = load <8 x i32>* %0
%oldValue = load <8 x i32>* %0, align 4
%oldAsFloat = bitcast <8 x i32> %oldValue to <8 x float>
%newAsFloat = bitcast <8 x i32> %1 to <8 x float>
%blend = call <8 x float> @llvm.x86.avx.blendvps(<8 x float> %oldAsFloat,
<8 x float> %newAsFloat,
<8 x float> %mask_as_float)
%blendAsInt = bitcast <8 x float> %blend to <8 x i32>
store <8 x i32> %blendAsInt, <8 x i32>* %0
store <8 x i32> %blendAsInt, <8 x i32>* %0, align 4
ret void
}