Re: [gentoo-dev] include files question

2013-09-21 Thread Peter Stuge
Michał Górny wrote:
 Putting another includedir is even worse kind of band-aid.

By using the expression band-aid I wasn't commenting on the virtues
of having the include file in /usr/include.

The expression rip off the band-aid is an idiom in english which
refers to minimizing the duration of unpleasantness by making a
change quickly and with determination.


 If we're to put them in a directory, I'd rather require
 'more complete' includes, like:
 
   #include openrc/rc.h

Consider installs into non-/usr prefixes. It's true that this may not
be so much of a concern since openrc is mostly Gentoo-specific though.


 Otherwise, you're just fighting conflicts in the scope of a single
 application.

There are two benefits of mandating use of pkg-config; one is that
openrc will just worktm on odd platforms such as OS X package
managers where the install prefix isn't included in the platform
compiler's include search path, the other is that the API can be
versioned trivially, allowing different versions to be installed
at the same time.


//Peter


pgpohyTAaZB4b.pgp
Description: PGP signature


[gentoo-dev] sys-kernel/tuxonice-sources up for grabs

2013-09-21 Thread Pacho Ramos
I don't have time for it for a long time (since I switched to official
suspend time ago and moved back to gentoo-sources then), updated patches
are periodically generated and put at:
http://tuxonice.nigelcunningham.com.au/downloads/all/?C=MO=D

Feel free to get it if you want




[gentoo-dev] Re: sys-kernel/tuxonice-sources up for grabs

2013-09-21 Thread Pacho Ramos
El sáb, 21-09-2013 a las 14:42 +0200, Pacho Ramos escribió:
 I don't have time for it for a long time (since I switched to official
 suspend time ago and moved back to gentoo-sources then), updated patches
 are periodically generated and put at:
 http://tuxonice.nigelcunningham.com.au/downloads/all/?C=MO=D
 
 Feel free to get it if you want

It goes with tuxonice-userui




[gentoo-dev] Re: RFC: FEATURES=binchecks strip for plain file packages e.g. latex, perl, python

2013-09-21 Thread Michael Palimaka

On 21/09/2013 10:49, Duncan wrote:

In addition to what the others have said, FEATURES= ??  AFAIK, FEATURES
is PM-implementation-specific and not part of PMS.  It's not something
ebuilds/eclasses should be messing with or care about at all, as it's PM-
private-implementation domain, not ebuild domain.
Presumably he is talking about FEATURES in the context of adding them to 
RESTRICT, which is discussed in PMS.


(Not that it really matters anyway - there are plenty of non-PMS 
variables used in ebuilds. That's fine, they control portage features 
and will be ignored by other package managers.)





[gentoo-dev] Re: RFC: FEATURES=binchecks strip for plain file packages e.g. latex, perl, python

2013-09-21 Thread Duncan
Michael Palimaka posted on Sat, 21 Sep 2013 23:27:07 +1000 as excerpted:

 On 21/09/2013 10:49, Duncan wrote:
 In addition to what the others have said, FEATURES= ??  AFAIK, FEATURES
 is PM-implementation-specific and not part of PMS.  It's not something
 ebuilds/eclasses should be messing with or care about at all, as it's
 PM-private-implementation domain, not ebuild domain.
 Presumably he is talking about FEATURES in the context of adding them to
 RESTRICT, which is discussed in PMS.

That /does/ make a bit more sense, and I see others presumed it as well, 
but it was entirely missing from the OP, which really mixed me up, even 
after seeing discussion of RESTRICT in the replies, since that seemed to 
come from nowhere.

 (Not that it really matters anyway - there are plenty of non-PMS
 variables used in ebuilds. That's fine, they control portage features
 and will be ignored by other package managers.)

True.  I guess it's nothing to worry about as long as the ebuild isn't 
depending on the implementation-specific var for anything vital to the 
package, only optional bits.

Thanks.  Clearer now.  =:^)

-- 
Duncan - List replies preferred.   No HTML msgs.
Every nonfree program has a lord, a master --
and if you use the program, he is your master.  Richard Stallman




Re: [gentoo-dev] RFC: RESTRICT=binchecks strip for plain file packages e.g. latex, perl, python

2013-09-21 Thread Justin (jlec)
On 20/09/13 19:58, Justin (jlec) wrote:
 Hi,
 
 what is your opinion to set
 
 FEATURES=binchecks strip

Of course this is wrong, I meant

RESTRICT=binchecks strip

justin



signature.asc
Description: OpenPGP digital signature


Re: [gentoo-dev] RFC: FEATURES=binchecks strip for plain file packages e.g. latex, perl, python

2013-09-21 Thread Michał Górny
Dnia 2013-09-20, o godz. 20:51:46
Paweł Hajdan, Jr. phajdan...@gentoo.org napisał(a):

 On 9/20/13 12:24 PM, Diego Elio Pettenò wrote:
  As Michał said.
 
 Please consider updating the documentation - ebuild(5):
 
  RESTRICT = [strip,mirror,fetch,userpriv]
 This  should  be a space delimited list of portage features to
 restrict.  You may use conditional syntax to vary restrictions as
 seen above in DEPEND.
 binchecks
Disable all QA checks for binaries.  This should ONLY be
used in packages  for which binary checks make no sense
(linux-headers and kernel-sources, for example, can safely
be skipped since they have no binaries).  If the binary
checks need  to  be skipped for other reasons (such as
proprietary binaries), see the QA CONTROL VARIABLES section
for more specific exemptions.
 
 This is currently used at least by sys-apps/man-pages, as well as
 packages in x11-themes and sys-kernel.

This is just another of portage's deviations from the standard. PMS
doesn't list RESTRICT=binchecks, and I suppose this makes using it in
the tree at least a little wrong.

-- 
Best regards,
Michał Górny


signature.asc
Description: PGP signature


[gentoo-dev] Re: RFC: FEATURES=binchecks strip for plain file packages e.g. latex, perl, python

2013-09-21 Thread Michael Palimaka

On 22/09/2013 00:26, Michał Górny wrote:

This is just another of portage's deviations from the standard. PMS
doesn't list RESTRICT=binchecks, and I suppose this makes using it in
the tree at least a little wrong.

It does however say: Package managers may recognise other tokens, but 
ebuilds may not rely upon them being supported.


Since it's just another variable that doesn't actually affect the ebuild 
or other package managers, it's no problem.





[gentoo-dev] News item: GRUB2 migration

2013-09-21 Thread Mike Gilbert
GRUB2 will be stabilized soon (bug 455544). Here's a draft of a news
item to hopefully prevent any confusion. Please review.



Title: GRUB2 migration
Author: Mike Gilbert flop...@gentoo.org
Content-Type: text/plain
Posted: 2013-09-22
Revision: 1
News-Item-Format: 1.0
Display-If-Installed: sys-boot/grub-1

A newer version of GRUB (sys-boot/grub) is now stable. There are now
two available slots:

sys-boot/grub:0 - Known as GRUB Legacy
sys-boot/grub:2 - Known as GRUB2

GRUB2 uses a different configuration format, and requires a manual
migration before your system will actually use it. A guide [1] is
available on the gentoo.org website.

If you would prefer not to migrate at this time, you do not need to
take any action: GRUB Legacy will remain functional in /boot. To
prevent any associated files (documentation) from being removed, add
sys-boot/grub:0 to your world file. For example:

emerge --noreplace sys-boot/grub:0

References:

[1] http://www.gentoo.org/doc/en/grub2-migration.xml



signature.asc
Description: OpenPGP digital signature


Re: [gentoo-dev] News item: GRUB2 migration

2013-09-21 Thread Paweł Hajdan, Jr.
On 9/21/13 8:42 AM, Mike Gilbert wrote:
 GRUB2 will be stabilized soon (bug 455544). Here's a draft of a news
 item to hopefully prevent any confusion. Please review.

Great news! Thanks for working on this, and it basically works
(chainloaded for now, see below) on my x86 system.

Some general remarks:

1. When using chainloaded GRUB2, neither Linux kernel boot messages nor
openrc boot messages would show up. Also, at shutdown openrc shutdown
messages were also not visible. Everything is fine if I boot the kernel
using GRUB legacy. This is inside VMWare Fusion.

2. At least /etc/default/grub refers to /boot/grub2, as opposed to
/boot/grub from Gentoo docs.

 GRUB2 uses a different configuration format, and requires a manual
 migration before your system will actually use it. A guide [1] is
 available on the gentoo.org website.
 
 [...]
 
 [1] http://www.gentoo.org/doc/en/grub2-migration.xml

The grub postinst message links to
http://wiki.gentoo.org/wiki/GRUB2_Quick_Start instead - and that wiki
page has a link to http://www.gentoo.org/doc/en/grub2-migration.xml .

IMHO it would be elegant for the news item to match the postinst
message: either make the grub2-migration.xml point to wiki as possibly
containing more information (and switch grub postint to point to
grub2-migration.xml), or make both link to the wiki.

Paweł




signature.asc
Description: OpenPGP digital signature


Re: [gentoo-dev] News item: GRUB2 migration

2013-09-21 Thread Mike Gilbert
On Sat, Sep 21, 2013 at 2:07 PM, Paweł Hajdan, Jr.
phajdan...@gentoo.org wrote:
 On 9/21/13 8:42 AM, Mike Gilbert wrote:
 GRUB2 will be stabilized soon (bug 455544). Here's a draft of a news
 item to hopefully prevent any confusion. Please review.

 Great news! Thanks for working on this, and it basically works
 (chainloaded for now, see below) on my x86 system.

 Some general remarks:

 1. When using chainloaded GRUB2, neither Linux kernel boot messages nor
 openrc boot messages would show up. Also, at shutdown openrc shutdown
 messages were also not visible. Everything is fine if I boot the kernel
 using GRUB legacy. This is inside VMWare Fusion.


Hmm... I don't really have a clue here. I will try to reproduce it in qemu.

 2. At least /etc/default/grub refers to /boot/grub2, as opposed to
 /boot/grub from Gentoo docs.


Thanks for catching that; fixed in the tree.

 GRUB2 uses a different configuration format, and requires a manual
 migration before your system will actually use it. A guide [1] is
 available on the gentoo.org website.

 [...]

 [1] http://www.gentoo.org/doc/en/grub2-migration.xml

 The grub postinst message links to
 http://wiki.gentoo.org/wiki/GRUB2_Quick_Start instead - and that wiki
 page has a link to http://www.gentoo.org/doc/en/grub2-migration.xml .

 IMHO it would be elegant for the news item to match the postinst
 message: either make the grub2-migration.xml point to wiki as possibly
 containing more information (and switch grub postint to point to
 grub2-migration.xml), or make both link to the wiki.


How about linking to both pages from the news item?

For the postinst message, I could link to the migration guide if
sys-boot/grub:0 is installed. If grub:0 is not installed, there should
be no need to migrate.



Re: [gentoo-dev] News item: GRUB2 migration

2013-09-21 Thread Michael Orlitzky
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 09/21/2013 11:42 AM, Mike Gilbert wrote:
 GRUB2 will be stabilized soon (bug 455544). Here's a draft of a
 news item to hopefully prevent any confusion. Please review.

The FAQ / Known Problems / Gotchas section of the guide is still
empty. Maybe fill this in before pushing this stable?

My contribution:

About half of our servers failed at the install phase with the error
(copy/pasted from Google, might not be exact),

  /usr/sbin/grub2-install: warn: Your embedding area is unusually
  small. core.img won't fit in it.

  /usr/sbin/grub2-install: warn: Embedding is not possible. GRUB can
  only be installed in this setup by using blocklists. However,
  blocklists are UNRELIABLE and their use is discouraged.

  /usr/sbin/grub2-install: error: will not proceed with blocklists.

To paraphrase, there ain't enough space before your first partition.

Generally this means that you are screwed, and that you'll need to
boot something like the gparted liveCD and resize all of your
partitions. It would be nice to list some options on the page.

If the user is lucky (thankfully, I was), his swap partition is the
first partition on the disk. In that case, he can simply `swapoff` the
thing, mangle it with fdisk, and then `mkswap foo`  `swapon foo`.

The fdisk commands were,

  1. 'x' - extra functionality (experts only)

  2. 'b' - move beginning of data in a partition

  3. '1' - Choose the first partition

  4. '4096' - New beginning of data (1-8000369, default 63)

Something smaller might work, but 4096 definitely does. Better play it
safe in case grub3 wants to play HD video.

For what it's worth, the following wiki pages both provided useful
information (at the time, the g.o page didn't exist):

  * https://wiki.gentoo.org/wiki/GRUB2_Quick_Start

  * https://wiki.gentoo.org/wiki/GRUB2


-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.20 (GNU/Linux)
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=Xm1p
-END PGP SIGNATURE-



Re: [gentoo-dev] News item: GRUB2 migration

2013-09-21 Thread Mike Gilbert
On Sat, Sep 21, 2013 at 2:22 PM, Michael Orlitzky mich...@orlitzky.com wrote:
 -BEGIN PGP SIGNED MESSAGE-
 Hash: SHA1

 On 09/21/2013 11:42 AM, Mike Gilbert wrote:
 GRUB2 will be stabilized soon (bug 455544). Here's a draft of a
 news item to hopefully prevent any confusion. Please review.

 The FAQ / Known Problems / Gotchas section of the guide is still
 empty. Maybe fill this in before pushing this stable?


It would probably be better to just link to the GRUB2 article on the
wiki. If you could update that with any relevant information, that
would be great.



Re: [gentoo-dev] News item: GRUB2 migration

2013-09-21 Thread Mike Gilbert
On 09/21/2013 02:18 PM, Mike Gilbert wrote:
 On Sat, Sep 21, 2013 at 2:07 PM, Paweł Hajdan, Jr.
 The grub postinst message links to
 http://wiki.gentoo.org/wiki/GRUB2_Quick_Start instead - and that wiki
 page has a link to http://www.gentoo.org/doc/en/grub2-migration.xml .

 IMHO it would be elegant for the news item to match the postinst
 message: either make the grub2-migration.xml point to wiki as possibly
 containing more information (and switch grub postint to point to
 grub2-migration.xml), or make both link to the wiki.

 
 How about linking to both pages from the news item?
 

Here's a second draft including that change.



Title: GRUB2 stabilization
Author: Mike Gilbert flop...@gentoo.org
Content-Type: text/plain
Posted: 2013-09-22
Revision: 1
News-Item-Format: 1.0
Display-If-Installed: sys-boot/grub-1

A newer version of GRUB (sys-boot/grub) is now stable. There are now
two available slots:

sys-boot/grub:0 - Known as GRUB Legacy
sys-boot/grub:2 - Known as GRUB2

GRUB2 uses a different configuration format, and requires a manual
migration before your system will actually use it. A guide [1] is
available on the gentoo.org website, and the Gentoo wiki [2][3] has
additional information.

If you would prefer not to migrate at this time, you do not need to
take any action: GRUB Legacy will remain functional in /boot. To
prevent any associated files (documentation) from being removed, add
sys-boot/grub:0 to your world file. For example:

emerge --noreplace sys-boot/grub:0

References:

[1] http://www.gentoo.org/doc/en/grub2-migration.xml
[2] https://wiki.gentoo.org/wiki/GRUB2_Quick_Start
[3] https://wiki.gentoo.org/wiki/GRUB2



signature.asc
Description: OpenPGP digital signature


[gentoo-dev] Re: rfc: status of OpenRC's public API

2013-09-21 Thread William Hubbs
All,

this is a followup to the original message that started this thread.

A case has been made for librc, but not libeinfo. There could be reasons
to allow the librc functionality to stay around, but I'm not convinced
wrt libeinfo, especially since there are no consumers.

Does anyone see a reason we should keep the einfo/eerror/etc c functions
in a public API?

William



signature.asc
Description: Digital signature


Re: [gentoo-dev] Re: rfc: status of OpenRC's public API

2013-09-21 Thread Ian Stakenvicius
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

On 21/09/13 03:06 PM, William Hubbs wrote:
 All,
 
 this is a followup to the original message that started this 
 thread.
 
 A case has been made for librc, but not libeinfo. There could be 
 reasons to allow the librc functionality to stay around, but I'm 
 not convinced wrt libeinfo, especially since there are no 
 consumers.
 
 Does anyone see a reason we should keep the einfo/eerror/etc c 
 functions in a public API?
 
 William
 

IIRC we still don't have an openrc-replacement script in the tree for
the /etc/init.d/function.sh symlink to target.  Since libeinfo is
already public, why not instead of making it private we go the other
way -- keep it public, package it out separately in the tree, and make
openrc (and others from bug 373219 and elsewhere) depend on it?

grobian already has a fork of libeinfo with an independent build
system and a functions.sh wrapper; we could leverage that for anything
missing from a standalone package in portage.  Probably the whole deal
could be done in under an hour, and it's already long-proven code
since it's what openrc and prefix has been using for years.

Out of curiosity, what is the reasoning behind making these libs private?
-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.20 (GNU/Linux)

iF4EAREIAAYFAlI98aMACgkQ2ugaI38ACPCTcQD9HIqOTlhia/ktPFANAZdJbfEv
DqOh7CUCULZw+FqkOpQBAISPbWdsg+flecvnv5OfWGsnLqnYK6GPG4e23KwDyz1e
=OCdp
-END PGP SIGNATURE-



Re: [gentoo-dev] News item: GRUB2 migration

2013-09-21 Thread Mike Gilbert
On Sat, Sep 21, 2013 at 2:18 PM, Mike Gilbert flop...@gentoo.org wrote:
 On Sat, Sep 21, 2013 at 2:07 PM, Paweł Hajdan, Jr.
 phajdan...@gentoo.org wrote:
 On 9/21/13 8:42 AM, Mike Gilbert wrote:
 GRUB2 will be stabilized soon (bug 455544). Here's a draft of a news
 item to hopefully prevent any confusion. Please review.

 Great news! Thanks for working on this, and it basically works
 (chainloaded for now, see below) on my x86 system.

 Some general remarks:

 1. When using chainloaded GRUB2, neither Linux kernel boot messages nor
 openrc boot messages would show up. Also, at shutdown openrc shutdown
 messages were also not visible. Everything is fine if I boot the kernel
 using GRUB legacy. This is inside VMWare Fusion.


 Hmm... I don't really have a clue here. I will try to reproduce it in qemu.


I wasn't able to reproduce this in qemu. It is possible that GRUB's
graphical terminal driver has some issues with VMWare.

However, have a Gentoo box at work that runs GRUB2 on VMWare ESX, and
I don't recall having any display issues on the console. But I'm also
not chainloading it there.



Re: [gentoo-dev] Re: sys-kernel/tuxonice-sources up for grabs

2013-09-21 Thread Rick Zero_Chaos Farina
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 09/21/2013 08:44 AM, Pacho Ramos wrote:
 El sáb, 21-09-2013 a las 14:42 +0200, Pacho Ramos escribió:
 I don't have time for it for a long time (since I switched to official
 suspend time ago and moved back to gentoo-sources then), updated patches
 are periodically generated and put at:
 http://tuxonice.nigelcunningham.com.au/downloads/all/?C=MO=D

 Feel free to get it if you want
 
 It goes with tuxonice-userui
 
 
 
Is any of this really needed anymore?  I suspend/hibernate/resume with
gentoo-sources and hardened-sources Does it really make sense to
hold on to tuxonice still?

- -Zero
-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.20 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=tZda
-END PGP SIGNATURE-