Re: [gentoo-dev] On hosting self-produced distfiles

2011-01-19 Thread Matt Turner
On Thu, Jan 20, 2011 at 2:50 AM, Diego Elio Pettenò flamee...@gmail.com wrote:
 Il giorno mer, 19/01/2011 alle 21.44 -0500, Mike Frysinger ha scritto:

 you really need to start off discussions as here is the problem i
 perceive
 and here is a solution i think will address it.  shooting off e-mails
 from on
 high as an official edict without room for discussion is no way to run
 a team.

 There is nothing new or revolutionary in what I said — are you trying to
 challenge the need for traceability of distfiles?

Good grief. No.

He seems to be simply stating that you'd do better to say we have a
problem with $this and I think it'd be fixed with $that instead of
This is to be considered QA policy, so we're going to ask soon to
enforce this.

Not hosting distfiles from dev.g.o is a reasonable thing that I think
everyone understands and agrees with, but this will be official
policy jargon is annoying and doesn't really serve any purpose.

Matt



Re: [gentoo-dev] [warning] the bug queue has 118 bugs

2010-12-15 Thread Matt Turner
On Wed, Dec 15, 2010 at 12:25 PM, Paweł Hajdan, Jr.
phajdan...@gentoo.org wrote:
 By the way, we have a nice team of arch and herd testers - how about
 encouraging them to wrangle some bugs?

Yeah, I just came here to say this. One certainly doesn't need to have
completed the developer quizzes to sort bugs.

Matt



Re: [gentoo-dev] Move x86/amd64 CPU extensions USE flags to a new USE_EXPAND variable

2010-12-11 Thread Matt Turner
On Sat, Dec 11, 2010 at 5:57 PM, Jeroen Roovers j...@gentoo.org wrote:
[snip]

I agree that this could be better. To me, most of the problems with
this are due to users not knowing which of these should be set for
their particular CPU.

Instead of having defaults set by a profile, I'd like to figure out a
way we can have these flags set by default dependent on the user's
CPU. This might require some additional logic in portage; I don't
know.

Matt



Re: [gentoo-dev] Maintainer notes in metadata.xml?

2010-11-30 Thread Matt Turner
On Wed, Dec 1, 2010 at 1:25 AM, Nirbheek Chauhan nirbh...@gentoo.org wrote:
 On Wed, Dec 1, 2010 at 6:30 AM, Diego Elio Pettenò flamee...@gmail.com 
 wrote:
 Hi all,

 I was wondering if we have space already, or if others would feel
 strongly about making space for, maintainer notes in packages'
 metadata.xml.

 [snip]

 What I'm thinking of is having some sort of maintainernotes element,
 but not a passive one that has to be tested for, rather something that
 repoman would spit out on the terminal when doing a scan/full.

 Comments?

 Why don't we just encourage maintainers to add !-- -- comments to
 metadata.xml? I'd love to have a new element if the data to be stored
 in that element would need to be parsed/categorized by external
 programs, but otherwise xml comments would work just fine.

And have repoman scan/full print out all !-- -- comments? I think
that's why Diego is suggesting a new XML tag.

Matt



[gentoo-dev] IUSE=minimal seems like a bad idea

2010-11-20 Thread Matt Turner
matts...@sempron /usr/portage $ egrep -l 'IUSE=.*minimal' `find -name
'*.ebuild'`

^ shows lots of ebuilds with IUSE=minimal. Instead of having a
minimal use flag for these packages, shouldn't we have, possibly
local, use flags for whatever feature(s) the minimal flag turns off?

Thanks,
Matt



Re: [gentoo-dev] Re: Restabilizing MIPS

2010-11-15 Thread Matt Turner
On Mon, Nov 15, 2010 at 1:15 PM, Panagiotis Christopoulos
pchr...@gentoo.org wrote:
 On 11:20 Mon 15 Nov     , Christian Faulhammer wrote:
  I don't want to stop you, but you are relatively new to the real
 keywording business.  amd64 at a point had over 30 members and could
 not work on the backlog in a timely manner.  Think about
 ...

 I completely agree. I don't know even what's the status of ~mips in the
 main tree. It would be better, I believe, to start working on the ~mips
 system/world set, then start stabilizing only toolchain and/or
 @base-system packages. Please don't start keywording/stabilizing random
 packages without taking into *serious* consideration what Christian
 wrote. I was here when we dropped the stable mips keyword, I don't want
 to see it happening again.

I'm not planning an ad-hoc at-random stabilization/keywording spree.

I'd just like to have a stable and tested base system.



Re: [gentoo-dev] Re: Restabilizing MIPS

2010-11-12 Thread Matt Turner
On Thu, Nov 11, 2010 at 10:34 PM, Ryan Hill dirtye...@gentoo.org wrote:
 On Thu, 11 Nov 2010 20:37:51 -0500
 Matt Turner matts...@gentoo.org wrote:

 On Thu, Nov 11, 2010 at 8:28 PM, Ryan Hill dirtye...@gentoo.org wrote:
  On Thu, 11 Nov 2010 18:00:00 -0500
  Matt Turner matts...@gentoo.org wrote:
 
  Should we target package versions that aren't stabilized on other
  architectures yet, so that we'll have an extended testing period
  before they'll come up for stabilization? That is, can I plan to make
  gcc-4.5.1 or something the first restabilized version of gcc, go ahead
  and begin testing it, and be ready for stabilization when toolchain
  requests it?
 
  I'd work on getting it ~mips before you think about stabilizing. ;)  Last
  report I got it doesn't build.

 I've been using gcc-4.5.1 for the last two weeks or so. :)

 Should I add a ~mips keyword?

 Kumba told me about some weird behavior he saw where the stage 2 and 3
 comparison failed and the build restarted from scratch again and got into an
 endless loop.  Maybe a hardware problem?

I talked with him on IRC last night and he's fine with adding ~mips.
No idea what's the deal with his box.

 If you want to keyword 4.5.1 then you'll need to keyword dev-libs/mpc (bug
 #279851) and either keyword dev-libs/cloog-ppl or mask the graphite USE flag
 in your profiles (bug #269088).

Keyworded mpc-0.8.2, cloog-ppl-0.15.9, and gcc-4.5.1. Please check to
make sure I didn't screw anything up. I'm new at this you know. :)

Thanks,
Matt



[gentoo-dev] Restabilizing MIPS

2010-11-11 Thread Matt Turner
Hi,
I'd like to begin stabilizing packages on MIPS. I've gotten acks from
Redhatter, leio, and r0bertz, and Kumba doesn't really care.

What's the best method to go about doing this? Stabilize the system
packages, then remove ~mips from ACCEPT_KEYWORDS in the profiles?
Should we target package versions that aren't stabilized on other
architectures yet, so that we'll have an extended testing period
before they'll come up for stabilization? That is, can I plan to make
gcc-4.5.1 or something the first restabilized version of gcc, go ahead
and begin testing it, and be ready for stabilization when toolchain
requests it?

Thanks, and any advice is appreciated. :)

Matt



Re: [gentoo-dev] Restabilizing MIPS

2010-11-11 Thread Matt Turner
On Thu, Nov 11, 2010 at 6:13 PM, Robin H. Johnson robb...@gentoo.org wrote:
 On Thu, Nov 11, 2010 at 06:00:00PM -0500, Matt Turner wrote:
 I'd like to begin stabilizing packages on MIPS. I've gotten acks from
 Redhatter, leio, and r0bertz, and Kumba doesn't really care.
 Out of interest, what MIPS hardware do you have?

I have a Broadcom BCM91250a with a dual-core 800 MHz SiByte CPU and a
200 MHz O2 with 1 GB RAM.

See http://mattst88.com/computers/bcm91250a/ . I need to post some
pictures, but that'll have to be for later.

Matt



Re: [gentoo-dev] Re: Restabilizing MIPS

2010-11-11 Thread Matt Turner
On Thu, Nov 11, 2010 at 8:28 PM, Ryan Hill dirtye...@gentoo.org wrote:
 On Thu, 11 Nov 2010 18:00:00 -0500
 Matt Turner matts...@gentoo.org wrote:

 Should we target package versions that aren't stabilized on other
 architectures yet, so that we'll have an extended testing period
 before they'll come up for stabilization? That is, can I plan to make
 gcc-4.5.1 or something the first restabilized version of gcc, go ahead
 and begin testing it, and be ready for stabilization when toolchain
 requests it?

 I'd work on getting it ~mips before you think about stabilizing. ;)  Last
 report I got it doesn't build.

I've been using gcc-4.5.1 for the last two weeks or so. :)

Should I add a ~mips keyword?

 What's your target hardware for stabilization?  Are we still focusing on SGI
 stuff or moving on to newer platforms?

SGI stuff is going to become less and less interesting, but it's still
the most common MIPS hardware Gentoo users have [1].
STMicroelectronics MIPS systems (Lemote, Gdium, etc) are becoming more
common, and we should definitely do a better job supporting them. (I
should mention that I've been loaned a Yeelong by Daniel Clark, of
freedomincluded.com, to fix up the siliconmotion driver.)

I think we can reasonably support both. The Broadcom board I have can
with the switch of a jumper operate in big or little-endian mode, so I
can use it for both.

Matt

[1] 
http://archives.gentoo.org/gentoo-mips/msg_65e25b7dae79f7b897ed59199919a2a2.xml



Re: [gentoo-dev] LibreOffice project: request for contributors and mentoring

2010-11-11 Thread Matt Turner
On Thu, Nov 11, 2010 at 8:13 PM, Stuart Longland redhat...@gentoo.org wrote:
 On Thu, Nov 11, 2010 at 05:40:03PM +0800, David Nelson wrote:
 A number of Linux distributions have announced their intention to ship
 LibreOffice with their future releases. We know that they frequently
 do re-branding work to integrate their chosen office suite in lines
 with their project's thinking.

 So we are keen to involve you in our project branding and development,
 so that we ship releases that better fit your needs.

 Do we even brand OpenOffice?  I can't spot the difference between the
 self-built OpenOffice.org binary I have, and the official Sun binary I
 had previously.

 My concern with LibreOffice would be more to do with compiling it...
 it'd be a nice package to have on the Yeeloong, but AFAIK it needs
 Java..?  Something I've been trying to bootstrap unsuccessfully for the
 best part of two years now.  (gcj-jdk is a long way from usable, and
 there's a chicken-egg issue with icedtea6.)  It also needs _lots_ of RAM
 and disk space ... not a plentiful resource on MIPS.

gNewSense, which came installed on the Yeelong I have, has OpenJDK and
OpenOffice, so it's certainly possible.

Matt



Re: [gentoo-dev] Re: Restabilizing MIPS

2010-11-11 Thread Matt Turner
On Thu, Nov 11, 2010 at 9:55 PM, Stuart Longland redhat...@gentoo.org wrote:
 STMicroelectronics MIPS systems (Lemote, Gdium, etc) are becoming more
 common, and we should definitely do a better job supporting them. (I
 should mention that I've been loaned a Yeelong by Daniel Clark, of
 freedomincluded.com, to fix up the siliconmotion driver.)

 Interesting... I've found Zhang Le's overlay includes a quite workable
 siliconmotion driver which runs fine on my Yeeloong.

 The only catch is that one must compile it with -march=loongson2f in
 CFLAGS... -mips3 (my preference) won't do.

The main purpose of my project here is to write a kernel modesetting
driver. Apparently lots of FSF fanatics use Yeelongs entirely from the
terminal, so the potential for faster console scrolling speed is
somewhat appealing to them. ;)

 Out of interest... did you get around to those n32 stages at all?  I'd
 like to get some of my old SGI kit up and going, some of them will need
 a complete reinstall... so I may as well do that using n32 from the
 outset.

 My O2 can remain o32 for now since it was the one still standing after
 all this time.  The others, the userland is broken/stale to the point of
 uselessness.

I tried again last week to make n32 stages, but have had a terrible
time with catalyst.

The main problem I run into is that I can't get catalyst to
acknowledge any package.keywords files (which as I understand might be
by design), so I'm unable to put together a stage from the versions
I'd like to stabilize. Are your recent o32 stages straight-up ~mips?
Can you post your spec files somewhere?

Matt



<    1   2   3   4   5   6