[gentoo-dev] [summary] combining x86 and amd64

[gentoo-dev] [summary] combining x86 and amd64

[Chris White]

Ok, say yah, this thread is tooooo long, so I decided, "Hey, let's make a 
summary of the main important points".  That way everyone doesn't have to 
read threads that are longer than the combined code of portage.

So, let's get started

First off, this great guy named Grant decided it was a good idea for us to 
have a merger of the x86 and amd64 arch teams, but requested a glep for it.  
Not having a glep around, he suggested someone else bring one up for it.

http://article.gmane.org/gmane.linux.gentoo.devel/26749/match=glep+++++amd64
was suggested for quick reading by blubb, who didn't like the idea.

Flameeyes then came in noting the impossibilities due to binary 
incompatibility, ie., some things Just Work(tm) on x86 only.

Some comments were brought in about amd64 profile level masking, but blubb 
said he didn't like that, and noted the importance of ~amd64 keywording, as 
well as the issues of things ready for x86 stable being held back by amd64 
keywording.  Also a small note was made about the differences between 
sparc32/sparc64 and x86/amd64 (userland/kernel magic).

After this a short battle occured, and ferringb request that the main 
pros/cons be laid out.

more small exchanges go here, mentioning of mips as an example (one keyword 
for both 32 and 64 bit systems).

some points were brought up about 32 bit emulated libs under amd64 to which 
Flameeyes explained that there are too many differences to consider them true 
32 bit.

Complaints were made about not wanting multi-lib amd64 in some cases.  
However, explanations were made about a non-multi-lib profile as well as 
praise for the current multi-lib system.

Ok, some comments were made about the previous keywording arguments being 
untrue, and comments were made about current ebuilds in the tree utilizing 
systems to help prevent against such incompatibilities.  

Notes were made about using profile archs and use_expand as a solution.

bit of flames

wolf gets tired of flames, says to make an x86 arch team already and be done 
with it.

bit more misc stuff

geoman mentions that the reasoning for this is x86 is not "officially" 
supported and is hurting tree QA.

wolf mentions that we have an x86 security team, which can be considered to be 
a somewhat official x86 arch team.

ok, we're starting to think along the lines of x86 arch team creation.  This 
brings up the x86 security alias members, as well as:

"The people maintaining the x86 kernel should also join, as well as the
release maintainer (chris, is that you?), the grub/lilo maintainers,
etc... That would be a good start. 
We should also try to recruit one or two x86 arch testers, hparker has
offered to help."

People start agreeing that x86 arch team creation is a Good Idea(tm)

More misc talk

The end!

Ok, that should sum it up, and currently there's another thread about x86 arch 
team creation which seems to be going well.

Hopefully this makes the longggggggggggggg thread a bit more readable.

Chris White
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] [summary] combining x86 and amd64

[Olivier Crête]

[-- Attachment #1: Type: text/plain, Size: 339 bytes --]

On Fri, 2005-02-09 at 18:32 +0900, Chris White wrote:
> Ok, that should sum it up, and currently there's another thread about x86 arch 
> team creation which seems to be going well.

On top of that.. We are abusing bug #104525 as a discussion board on the
new x86 team.

-- 
Olivier Crête
tester@gentoo.org
x86 Security Liaison

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] [summary] combining x86 and amd64

[Henrik Brix Andersen]

[-- Attachment #1: Type: text/plain, Size: 515 bytes --]

On Fri, 2005-09-02 at 18:32 +0900, Chris White wrote:
> Ok, say yah, this thread is tooooo long, so I decided, "Hey, let's make a 
> summary of the main important points".  That way everyone doesn't have to 
> read threads that are longer than the combined code of portage.
[snip]
> Hopefully this makes the longggggggggggggg thread a bit more readable.

Thank you for doing such a summary :)

Sincerely,
Brix
-- 
Henrik Brix Andersen <brix@gentoo.org>
Gentoo Metadistribution | Mobile computing herd

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]