| 1 |
On Thursday 06 May 2010 12:15:18 Thomas Kahle wrote: |
| 2 |
> Hi, |
| 3 |
> |
| 4 |
> > I think this is an error we already had and solved before - please send |
| 5 |
> > us your emerge --info and uname -a on the pc you compiled sage-core. You |
| 6 |
> > may also |
| 7 |
> |
| 8 |
> > try to change the line in src_prepare() in sage-core: |
| 9 |
> I'm on x86 mostly stable (glibc-2.11.1 is installed for archtesting...), |
| 10 |
> and indeed -fno-strict-aliasing solves this. |
| 11 |
|
| 12 |
Excellent - fix in overlay. |
| 13 |
|
| 14 |
> |
| 15 |
> denkmatte bin # uname -a |
| 16 |
> Linux denkmatte 2.6.33-gentoo #1 SMP PREEMPT Mon Mar 29 21:48:35 EDT |
| 17 |
> 2010 i686 Intel(R) Core(TM)2 Duo CPU L7500 @ 1.60GHz GenuineIntel GNU/Linux |
| 18 |
|
| 19 |
By the way: I always thought -fstrict-aliasing would trigger erroneous |
| 20 |
behavior only on non x86 architectures - good to know that its not the case. |
| 21 |
|
| 22 |
Christopher |
Archives