2018-11-26 23:35 GMT+01:00, Michael Niedermayer :
> On Mon, Nov 26, 2018 at 12:45:26AM +0100, Carl Eugen Hoyos wrote:
>> 2018-11-25 16:17 GMT+01:00, Lauri Kasanen :
>> > Hi,
>> >
>> > The lone power8 fate failing test seems like an aliasing issue.
>> > I've isolated it into the attached standalone
On Mon, Nov 26, 2018 at 12:45:26AM +0100, Carl Eugen Hoyos wrote:
> 2018-11-25 16:17 GMT+01:00, Lauri Kasanen :
> > Hi,
> >
> > The lone power8 fate failing test seems like an aliasing issue.
> > I've isolated it into the attached standalone test case. Compiling it
> > with
> > gcc -std=c11
On Mon, 26 Nov 2018 00:45:26 +0100
Carl Eugen Hoyos wrote:
> 2018-11-25 16:17 GMT+01:00, Lauri Kasanen :
> > Hi,
> >
> > The lone power8 fate failing test seems like an aliasing issue.
> > I've isolated it into the attached standalone test case. Compiling it
> > with
> > gcc -std=c11 -maltivec
2018-11-25 16:17 GMT+01:00, Lauri Kasanen :
> Hi,
>
> The lone power8 fate failing test seems like an aliasing issue.
> I've isolated it into the attached standalone test case. Compiling it
> with
> gcc -std=c11 -maltivec -mabi=altivec -mvsx -O3 -fno-tree-vectorize
> -o test test.c
>
> reproduces
On Sun, 25 Nov 2018 17:17:58 +0200
Lauri Kasanen wrote:
> This code would probably crash on systems where unaligned access is
> prohibited, I think the incoming block is just 16-bit aligned.
I see the block comes from aligned malloc, so scratch that part, it's at
least 128-bit aligned.
- Lauri
Hi,
The lone power8 fate failing test seems like an aliasing issue.
I've isolated it into the attached standalone test case. Compiling it
with
gcc -std=c11 -maltivec -mabi=altivec -mvsx -O3 -fno-tree-vectorize
-o test test.c
reproduces on gcc 8.2.0, dropping the optimization level fixes it. This