| 1 |
On Fri, 2006-07-07 at 04:28 +0200, Diego 'Flameeyes' Pettenò wrote: |
| 2 |
> On Friday 07 July 2006 03:15, Mike Frysinger wrote: |
| 3 |
> > > x86_64 toolchain accepting 3dnow on a nocona arch? :) |
| 4 |
> > that isnt a cross-compile nor a different architecture |
| 5 |
> This is the whole point of my solution. |
| 6 |
> |
| 7 |
|
| 8 |
From what you discussed above, it sounds more like a problem due to |
| 9 |
short-sightedness on the amd64 team's part (no offence to amd64 team, |
| 10 |
just stating things as I see them) because they just enabled 3dnow for |
| 11 |
stuff that worked regardless. |
| 12 |
|
| 13 |
Stupid question though ... does the gcc test thingy list __3dNOW__ on |
| 14 |
nocona ? I would think that it does, as there is no -march=nocona (or |
| 15 |
whatever) yet. |
| 16 |
|
| 17 |
So now you want to instead of fixing the amd64 profiles to be more |
| 18 |
flexible, implement something that will give the green light to users on |
| 19 |
x86 to use flag combinations, especially on older gcc's that causes |
| 20 |
great pain for themselfs and developers ? |
| 21 |
|
| 22 |
Sure, nocona should have had CFLAGS="-march=k8 -mno-3dnow", but it would |
| 23 |
never have been an issue if the '3dnow' USE flag worked as expected on |
| 24 |
amd64 ;) |
| 25 |
|
| 26 |
Anyhow, just ranting - I understand the reasoning for doing it that way, |
| 27 |
but you should also see it from the x86 side where -msse could really |
| 28 |
mean a broken system, and maybe rethink your solution. |
| 29 |
|
| 30 |
|
| 31 |
-- |
| 32 |
Martin Schlemmer |
Archives