Calvin Spealman [EMAIL PROTECTED] writes:
but amd64 is itself an extension of the x86 architecture, isn't that the
case? amd64 chips run x86 code, i'm sure. do
they not support those extensions? I understand that for some reason the USE
flags are irrelevent here, but I don't
understand why
So its automatic detection of the features if possible, but USE flags
if otherwise. OK, now I get it. The way it had been said, I just got
the impression there was something different going on.
Thanks for clearing up the confusion.On 5/2/05, Graham Murray [EMAIL PROTECTED] wrote:
Calvin Spealman
Calvin Spealman wrote:
So its automatic detection of the features if possible, but USE flags if
otherwise.
No. There is no detection. Just assumption that AMD65 will support
certain extensions.
And please turn of HTML for posts in mailing lists.
Be lucky,
Neil
--
gentoo-user@gentoo.org mailing
On 5/2/05, Neil Walker [EMAIL PROTECTED] wrote:
So its automatic detection of the features if possible, but USE flags if
otherwise.
No. There is no detection. Just assumption that AMD65 will support
certain extensions.
In other words, -march=athlon64 implies those flags.
--
Bruno
but amd64 is itself an extension of the x86 architecture, isn't that
the case? amd64 chips run x86 code, i'm sure. do they not support those
extensions? I understand that for some reason the USE flags are
irrelevent here, but I don't understand why and it is starting to drive
me crazy because I'm
On 4/29/05, Ciaran McCreesh [EMAIL PROTECTED] wrote:
On Fri, 29 Apr 2005 16:54:57 + Calvin Spealman
[EMAIL PROTECTED] wrote:
| i dont understand whats going on here. I thought those flags enabled
| the use of extended sets of the x86 instruction set, and amd64 chips
| are in the x86
On Sat, 30 Apr 2005 10:27:15 -0600 Kiawud [EMAIL PROTECTED] wrote:
| I thought that if you 'cat /proc/cpuinfo' it'll show you the flags
| that are compatible with your cpu. Is that not accurate for the amd64
| chip?
Uh huh, but those flags don't correspond to the USE flags.
--
Ciaran McCreesh
What they're saying is that SSE/SSE2/MMX and so on are automatically
applied without the flags. The flags are only there because not every
x86 chip supports SSE, SSE2, and so on. On AMD64, the flags are ignored
and it's compiled with any of those processor optimizations that it
knowingly supports.
On April 28, 2005 11:14 pm, Paul Varner wrote:
On Thu, 2005-04-28 at 22:46 -0400, daniel wrote:
can someone please explain to me why these flags are being ignored and
what those brackets mean? if you're interested, here's my USE variable
in make.conf (i used app-portage/ufed):
What
On April 29, 2005 09:06 am, Ciaran McCreesh wrote:
On Fri, 29 Apr 2005 08:44:54 -0400 daniel
[EMAIL PROTECTED] wrote:
| oh that's what it means! where'd you find that out? does this mean
| that using an amd64 means i won't get support for any of those
| things? is this likely to change?
i dont understand whats going on here. I thought those flags enabled
the use of extended sets of the x86 instruction set, and amd64 chips
are in the x86 family, aren't they? and they support those extensions,
I'm sure, so what then do those flags do that the amd64 can't make use
of the flags?On
On Fri, 29 Apr 2005 16:54:57 + Calvin Spealman
[EMAIL PROTECTED] wrote:
| i dont understand whats going on here. I thought those flags enabled
| the use of extended sets of the x86 instruction set, and amd64 chips
| are in the x86 family, aren't they? and they support those
| extensions, I'm
i've been noticing lately that the output of emerge -pv is giving me some
interesting output:
$ emerge -pv mplayer
These are the packages that I would merge, in order:
Calculating dependencies ...done!
[ebuild R ] media-video/mplayer-1.0_pre6-r5 (-3dfx) (-3dnow) +3dnowext
+X
On Thu, 2005-04-28 at 22:46 -0400, daniel wrote:
i've been noticing lately that the output of emerge -pv is giving me some
interesting output:
$ emerge -pv mplayer
These are the packages that I would merge, in order:
Calculating dependencies ...done!
[ebuild R ]
14 matches
Mail list logo
