[gentoo-dev] Re: xorg-x11-6.8.2-r2 status on sparc
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Sat, 11 Jun 2005, Donnie Berkholz wrote: --[PinePGP]--[begin]-- Ferris McCormick wrote: You asked for patches. Both of these are incorporated in the 6.8.99 series, and have been around long enough that they should go into 6.8.2-r2 if it is going mainstream: https://bugs.freedesktop.org/show_bug.cgi?id=1259 https://bugs.freedesktop.org/show_bug.cgi?id=1891 (They are also part of the attachment 53024 at https://bugs.gentoo.org/show_bug.cgi?id=79467) I'd prefer to avoid adding patches that are feature enhancements instead of small fixes, if possible. Is it a problem if we pass on those? Not really. They are not in 6.8.2-r1, and the way for anyone wanting them into 6.8.2-r2 is the same as for 6.8.2-r1. Also, unless I am missing something, for sparc, 6.8.2-r2 and 6.8.2-r1 are the same. At least, when 6.8.2-r2 first came out, that was what I inferred from the ChangeLog notes. That's why I never bothered with ~sparc-ing it until yesterday in response to your status request. The reason I mentioned them is that if you anticipate a long and stable life for 6.8.2-r2, it makes things a little cleaner sparc-ishly. Otherwise, there's not much point. So, unless there is some compelling reason I am unaware of, I'm inclined never to move 6.8.2-r2 ~sparc-->sparc. When asked, I suggest unmasking the 6.8.99.xx series and test them instead, anyway. (I've kept them ~sparc all along because they work fine except for some eccentricities in, I believe, Mesa vtxvector, but I need feedback from other people to track it down. I plan to get to it, but more pressing tasks keep pushing it down on my priority list.) Done witfinals yet? :) Regards, Ferris - -- Ferris McCormick (P44646, MI) <[EMAIL PROTECTED]> Developer, Gentoo Linux (sparc) -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iD8DBQFCrDdPQa6M3+I///cRAs1XAKChw01mgvK1cdgw3t63zTDY9seYewCgskzV KLCktBof9fGt8xH7Ny7Kqrw= =yhly -END PGP SIGNATURE- -- gentoo-dev@gentoo.org mailing list
[gentoo-dev] Even More Portage Bashrc Fun
Hello,
So a long time ago solar wrote a bashrc for portage, and posted it on
this mailing list for everyone to see. I took it, and started extending
it with various things of my own design, and some contributions from
others. I've since updated it with the things solar's been adding to
his bashrc, plus adding the functionality from ChrisWhites bashrc.
Well, adding all of that stuff made the bashrc quite large. In addition
I was using CVS for a while, and wrote a lot of hooks to work with CVS,
when I went back to 2.0.51, I didnt want to lose all of those, so I
backported in a way the hooks from CVS (I mean in a way because the post
hooks arent really post hooks, they just run before the pre hook of the
next phase), but that meant the bashrc was huge.
So, sometime last night or this morning I set out to create a modular
portage bashrc [1], and I have done so. I'll admit, the code is ugly,
and it probably could've been done better. But it does lay a good
framework for future extensions.
The bashrc file is just a skeleton that loads the modules, and handles
the pseudo-hook implementation. The real magic happens in the
bashrc-modules subdirectory. core-functions.bmod contains the basic
functions and handles loading other modules. hooks.bmod defines the
hooks used and defines functions used to add new function calls to the
hooks. The other bmod files define some sort of library for other
modules (like an eclass) or functions that get loaded into the hooks. I
don't have any documentation written up for that, so read through the
source and you should get an idea of how it all works.
The BASHRC_DYN_MODULES variable can be defined in make.conf to limit the
modules that are loaded by the bashrc. This is useful if you only want
to use a subset of the functionality available in the modules, and dont
want to load them all. The default behavior is to load all of the
modules under ${ROOT}/etc/portage/bashrc-modules/. To extend the
bashrc, for example, to add another feature, you simply create a new
bmod following the examples given, then either let it load automatically
or add it to BASHRC_DYN_MODULES.
I've done some thorough testing and beu did some testing as well, so
there shouldnt be any major bugs, but if you find some, please e-mail me
with them. I'm especially interested in command not found errors. This
is some really ugly bash code, so if you run into obscure errors, dont
freak out, theyre a result of how I deal with the infrastructure.
Overall though, it appears to be stable, and I'm currently using it on
my system.
Finally, I'd like to thank the people who directly or indirectly helped
with this bashrc and the bashrc system: solar, ChrisWhite, beu,
bluefoxicy, and anyone else who I forgot to name. Enjoy everyone!
Mike Tindal
PS: Heed the warning given in the setup phase, you *cannot* modify
variables that affect depends because the environment the bashrc
modifies isnt picked up by depends. Be very careful with what you do
with category.use, since that can very easily break builds.
[1]
http://dev.gentoo.org/~urilith/portage-tools/bashrc-2.0.51-modular-20050612.tar.bz2
I've got some sample files in that dir for the random files the bashrc
supports.
signature.asc
Description: OpenPGP digital signature
[gentoo-dev] Last rites for glukalka, junkie, napshare, gv4l, dyndnsupdate
The following packages will be removed from Portage in 48 hours, unless someone steps up to fix the corresponding security bugs - and takeover upstream when it's dead :) app-emulation/glukalka (bug 70666) No upstream and vulnerable to multiple tempfile vulns and race conditions, this package has been masked since 2004-11-26, with noone missing it. net-ftp/junkie (bug 74696) No upstream and vulnerable to remote execution of code, this package has been masked since 2004-12-30, no complaints received. net-p2p/napshare (bug 74703) Vulnerable to remote execution of code, the new version does not compile correctly and no maintainer stepped up. This package has been masked since 2005-01-07, apparently noone is missing it. media-video/gv4l (bug 82631) No upstream answer, and vulnerable to various local problems, this package has been masked since 2005-03-18, nobody asked for it. net-misc/dyndnsupdate (bug 84659) No upstream and vulnerable to remote execution of code, this package has been masked since 2005-03-21, with everyone switching to more secure alternatives. -- Thierry Carrez (Koon) Operational Manager, Gentoo Linux Security signature.asc Description: OpenPGP digital signature
Re: [gentoo-dev] where goes Gentoo?
> So to sum it up, it's not really (for me maybe) about enterprise v. hobbyist, > it's about moving ANYONE over to Linux, period. Actually, I rather like to think that Gentoo is one of the very few distributions that cares more about meeting existing Linux [power]users' needs rather than getting any new users to Linux. I mean lets face it, a distro that's largely DIY is hardly a good first Linux, but an excellent second Linux and indeed thats the very reason why I use Gentoo. Let Redhat/Fedora/Mandrake do the initial user grab, that's what they're good at. Athul -- gentoo-dev@gentoo.org mailing list
Re: [gentoo-dev] where goes Gentoo?
Athul Acharya wrote: >>So to sum it up, it's not really (for me maybe) about enterprise v. hobbyist, >>it's about moving ANYONE over to Linux, period. > > > Actually, I rather like to think that Gentoo is one of the very few > distributions that cares more about meeting existing Linux > [power]users' needs rather than getting any new users to Linux. I > mean lets face it, a distro that's largely DIY is hardly a good first > Linux, but an excellent second Linux and indeed thats the very reason > why I use Gentoo. Let Redhat/Fedora/Mandrake do the initial user > grab, that's what they're good at. > > Athul > OTOH, most computer users are unable to or uninterested in installing/configuring an OS. All they need is someone to setup Gentoo for them and they can basically use it like they would an MS Windows "appliance". Zac -- gentoo-dev@gentoo.org mailing list
Re: [gentoo-dev] Re: xorg-x11-6.8.2-r2 status on sparc
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Ferris McCormick wrote: > On Sat, 11 Jun 2005, Donnie Berkholz wrote: >>> I'd prefer to avoid adding patches that are feature enhancements instead >>> of small fixes, if possible. Is it a problem if we pass on those? >>> > > Not really. They are not in 6.8.2-r1, and the way for anyone wanting > them into 6.8.2-r2 is the same as for 6.8.2-r1. Also, unless I am > missing something, for sparc, 6.8.2-r2 and 6.8.2-r1 are the same. At > least, > when 6.8.2-r2 first came out, that was what I inferred from the > ChangeLog notes. That's why I never bothered with ~sparc-ing it until > yesterday in response to your status request. Here's a reasonably complete list: * Crash bug in Xft (Bugs: Gentoo #85153, fd.o #1958, fd.o #1951) * Fix major slowdowns because of bogus /usr/share/fonts/fonts circular symlink, from weird interactions in the migration to /usr functions. This still needs a little more work, however. * I'm not sure whether sparc has any issues with lib64 now or in the future, but that's cleaner on -r2 * GCC-4 compile fixes * Cleaner use of opengl-update, avoiding a number of problems with stale symlinks etc * Less problems with USE=doc build * The constant errors for the `find` syntax are fixed, in the ebuild and the xfs init script * xdm init script cleanly shuts down, instead of erroring out * fc-cache isn't forced, which allows cleaner unmerges and stops overwriting other font packages' files * Hardened builds more consistently because of use of has_hardened instead of USE flag > So, unless there is some compelling reason I am unaware of, I'm inclined > never to move 6.8.2-r2 ~sparc-->sparc. When asked, I suggest unmasking > the 6.8.99.xx series and test them instead, anyway. Not gonna happen, ever. I'm not interested in dealing with all the crap that unmasked weekly un-QA'd, untested snapshots will bring. > Done witfinals yet? :) Yep. Thanks, Donnie -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iD8DBQFCrMdyXVaO67S1rtsRAtvwAKC2Dbc2BCPS414TdKuRfG4e+Uk85wCg/R0W 1Q6TtZr1hliwzaMP9XyPQaY= =XK76 -END PGP SIGNATURE- -- gentoo-dev@gentoo.org mailing list
[gentoo-dev] Request for testing revdep-rebuild (gentoolkit-0.2.1_pre3)
All: I have bumped gentoolkit to 0.2.1_pre3 and package masked it for architecture testing and general testing of the new improved revdep-rebuild. This version contains lots of bug fixes and I would like give it a workout before unmasking. There are some major changes in identifying the broken links and since I only have access to x86, I want to make sure that I have not broken portability to other achitectures. Because of this, I have keyworded this ebuild with "-* ~x86". I have CC'ed the architecture teams on bug #62644. However, if you are interested in helping to test, please add gentoolkit to your /etc/portage/package unmask and let me know of your results. Regards, Paul -- gentoo-dev@gentoo.org mailing list
[gentoo-dev] Proposal: New Bugzilla resolution: NEEDMAINTAINER
Hi gang, I just (again) noticed I'm currently the only active developer of the desktop-misc herd, which is going to change as soon as the recruitment process is reopened again. About 80% of all bugs assigned to desktop-misc are new ebuild submissions for software which either falls into no other graphical application herds responsibility or applications that are not wanted by the relevant herds. Which is just fine, I don't want to hear any whining here, it's one of the purposes of the desktop-misc herd. Anyway, as I've already said I'm the _currently_ the only active herd member and simply cannot take the responsibility of about 100 new packages. Therefor I suggest introducing a new keyword to bugzilla's resolution list, called NEEDMAINTAINER. With this new keyword everyone could compile a list of currently unmaintained packages and just take the ones he likes. Additionally this would be a great opportunity for non-developers for starting their carrier within Gentoo. PS: If I'm repeating my self.. It's 4:30am and I'm awake for 68 hours, so.. Regards, Markus -- (o_ Markus Nigbur //\ Gentoo GNU/Linux Developer [ ]/_ http://www.gentoo.org GPG Fingerprint: 9E31 187A 6C78 210C F248 EDBB F2FB A77F 707D F8C1 signature.asc Description: This is a digitally signed message part
[gentoo-dev] [Fwd: Re: [gentoo-core] Proposal: New Bugzilla resolution: NEEDMAINTAINER]
--- Begin Message --- -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Would, RESOLVED LATER, with 'need maintainer' in the status whiteboard not be suitable? for the purpose of pulling queries, it's just as affective. Robert Markus Nigbur wrote: > Hi gang, > > I just (again) noticed I'm currently the only active developer of the > desktop-misc herd, which is going to change as soon as the recruitment > process is reopened again. > About 80% of all bugs assigned to desktop-misc are new ebuild > submissions for software which either falls into no other graphical > application herds responsibility or applications that are not wanted by > the relevant herds. Which is just fine, I don't want to hear any whining > here, it's one of the purposes of the desktop-misc herd. > > Anyway, as I've already said I'm the _currently_ the only active herd > member and simply cannot take the responsibility of about 100 new > packages. > Therefor I suggest introducing a new keyword to bugzilla's resolution > list, called NEEDMAINTAINER. > With this new keyword everyone could compile a list of currently > unmaintained packages and just take the ones he likes. Additionally this > would be a great opportunity for non-developers for starting their > carrier within Gentoo. > > PS: If I'm repeating my self.. It's 4:30am and I'm awake for 68 hours, > so.. > > Regards, > Markus > -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (Darwin) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD4DBQFCrPG/ZwjIiODIZ4oRAvyWAJ4wEnWoaLoKiBYEAkn0VKSh2hx8egCWLX4C niG6zUelJ6dloAgmIJxY7w== =jMPD -END PGP SIGNATURE- -- [EMAIL PROTECTED] mailing list --- End Message ---
Re: [gentoo-dev] [Fwd: Re: [gentoo-core] Proposal: New Bugzilla resolution: NEEDMAINTAINER]
On Monday 13 June 2005 04:42, Robert Paskowitz wrote: > Would, RESOLVED LATER, with 'need maintainer' in the status whiteboard > not be suitable? for the purpose of pulling queries, it's just as > affective. > > Robert Probably, yes. It'd be just quicker/more organized way of doing it. -- (o_ Markus Nigbur //\ Gentoo GNU/Linux Developer [ ]/_ http://www.gentoo.org GPG Fingerprint: 9E31 187A 6C78 210C F248 EDBB F2FB A77F 707D F8C1 pgp05wXb0BzBQ.pgp Description: PGP signature
