Displaying 3 results from an estimated 3 matches for "59b586c6".
2013 Nov 12
2
[LLVMdev] Limit loop vectorizer to SSE
The AVX + JIT bug is hitting more frequently now. On a AVX machine the
loop vectorizer goes for a vector length of 8 for some of my functions
which in turn causes a SEGFAULT.
Is there a way to limit the loop vectorizer to a certain vector length,
say 4, such that I can work around the bug?
Frank
2013 Nov 12
0
[LLVMdev] Limit loop vectorizer to SSE
I am asking because the option 'force-vector-width' is too restrictive.
I would like to leave open the possibility to use vector width 2.
Frank
On 12/11/13 10:05, Frank Winter wrote:
> The AVX + JIT bug is hitting more frequently now. On a AVX machine the
> loop vectorizer goes for a vector length of 8 for some of my functions
> which in turn causes a SEGFAULT.
>
> Is
2013 Nov 12
2
[LLVMdev] Limit loop vectorizer to SSE
...JIT, I suppose setting the Target
attributes would be enough. Nor I know what CPU string limits support to
SSE, but that should do it.
cheers,
--renato
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20131112/59b586c6/attachment.html>