| 1 |
Ühel kenal päeval, K, 22.08.2018 kell 09:08, kirjutas Rich Freeman: |
| 2 |
> On Wed, Aug 22, 2018 at 8:26 AM Ben Kohler <bkohler@g.o> |
| 3 |
> wrote: |
| 4 |
> > |
| 5 |
> > 1) Adjust x86 profile defaults to drop the problematic -march=i686. |
| 6 |
> > This would be more in line with amd64 profiles (et al), which set |
| 7 |
> > no |
| 8 |
> > -march value so it can run on any hardware for this arch. |
| 9 |
> > |
| 10 |
> |
| 11 |
> My knee-jerk reaction was that this is a bad idea, but after a bit of |
| 12 |
> thought there are some arguments in favor of this: |
| 13 |
> |
| 14 |
> First, the argument against: i386 is VERY old. |
| 15 |
|
| 16 |
The topic is i486, not i386. |
| 17 |
i486 stages allows to more easily start up gentoo on i486 and i586 |
| 18 |
hardware, possibly also for some i686. |
| 19 |
E.g. if I'd ever get around to working on Geode graphics again, I would |
| 20 |
want a i486 or i586 stage to start with, as CMOV on the Geode is a |
| 21 |
performance penalty, not benefit, plus glibc i486 or i586 glibc |
| 22 |
intrinsics were faster than i686 ones as well. This includes a million |
| 23 |
or so OLPC XO-1's out there, not that they'd ever install Gentoo |
| 24 |
though, but just an example. I think some old VIAs are another? |
| 25 |
i686 builds sometimes wrongly make use of NOPL instruction as well, |
| 26 |
though hopefully that was fixed for good in binutils. |
| 27 |
|
| 28 |
Anyways, my point is that we are talking about being able to boot i486 |
| 29 |
and i586 here, not i386. |
| 30 |
Personally I can manage my potential own future needs without weekly |
| 31 |
stages. |
| 32 |
|
| 33 |
|
| 34 |
Mart |
Archives