genpatches-2.6.16-2 release

This is an automated email announcing the release of
genpatches-2.6.16-2


CHANGES SINCE 2.6.16-1
-----------------------

Revision 341: 
Use sparsemem for ultrasparc machines, thanks to Gustavo
Zacarias (dsd)
Added: 1710_sparc-sparsemem.patch

Revision 343: 
sky2 update, hopefully solves stability issues (dsd)
Added: 2400_sky2-1.1.patch

Revision 344: 
skge update, hopefully solves performance issues (dsd)
Added: 2405_skge-1.5.patch

Revision 345: 
input.h header fix for splashutils compile (dsd)
Added: 2600_input-h-bits-per-long.patch

Revision 346: 
Fix disconnect when renewing keys with hostap driver (dsd)
Added: 2410_hostap-eapol-encryption.patch

Revision 347: 
Linux 2.6.16.1 thanks to Francisco Lloret in bug #127838
(dsd)
Added: 1000_linux-2.6.16.1.patch

Revision 348: 
2.6.16-2 release (dsd)


PATCHES
-------

When the website updates, the complete patch list and
split-out patches will be
available here:
http://dev.gentoo.org/~dsd/genpatches/patches-2.6.16-2
.htm
http://dev.gentoo.org/~dsd/genpat
ches/tarballs/genpatches-2.6.16-2.base.tar.bz2
http://dev.gentoo.org/~dsd/genp
atches/tarballs/genpatches-2.6.16-2.extras.tar.bz2


ABOUT GENPATCHES
----------------

genpatches is the patchset applied to some kernels available
in Portage.

For more information, see the genpatches homepage:
http://dev.gent
oo.org/~dsd/genpatches

For a simple example of how to use genpatches in your kernel
ebuild, look at a
recent gentoo-sources-2.6.* ebuild.
-- 
gentoo-kernelgentoo.org mailing list

On Tue, Mar 28, 2006 at 02:19:22PM +0100, Daniel Drake 
<dsdgentoo.org> wrote:
> Revision 347: 
> Linux 2.6.16.1 thanks to Francisco Lloret in bug
#127838 (dsd)
> Added: 1000_linux-2.6.16.1.patch

Since were bunlding these whole now, should we just start
naming the
ebuilds correctly instead?

gentoo-sources-2.6.16.1.ebuild perhaps in this case?
At least that way it will save a marginally amount on
downloaded data,
we can revision bump stable releases and we can also tie in
better with
portage versioning.

-- 
Role:            Gentoo Linux Kernel Lead
Gentoo Linux:    http://www.gentoo.org
Public Key:      gpg --recv-keys 9C745515
Key fingerprint: A0AF F3C8 D699 A05A EC5C  24F7 95AA 241D
9C74 5515

John Mylchreest wrote:
> On Tue, Mar 28, 2006 at 02:19:22PM +0100, Daniel Drake 
<dsdgentoo.org> wrote:
>> Revision 347: 
>> Linux 2.6.16.1 thanks to Francisco Lloret in bug
#127838 (dsd)
>> Added: 1000_linux-2.6.16.1.patch
> 
> Since were bunlding these whole now, should we just
start naming the
> ebuilds correctly instead?

I don't think so. 2 -stable releases in the 2.6.15 cycle
didn't have any 
immediate corresponding gentoo-sources bump since the
patches were 
already included. The fact that we include patches alongside
-stable 
means that using their notation isn't entirely accurate in
our situation.

Daniel

-- 
gentoo-kernelgentoo.org mailing list

On Tue, Mar 28, 2006 at 05:48:13PM +0100, Daniel Drake
wrote:
> John Mylchreest wrote:
> >Since were bunlding these whole now, should we just
start naming the
> >ebuilds correctly instead?
> 
> I don't think so. 2 -stable releases in the 2.6.15
cycle didn't have any 
> immediate corresponding gentoo-sources bump since the
patches were 
> already included. The fact that we include patches
alongside -stable 
> means that using their notation isn't entirely
accurate in our situation.

Both naming schemes have their advantages, but I see more
disadvantages with using the 2.6.x.y naming scheme in
gentoo-sources
than I see advantages (primarily the reason Daniel stated
above).

I suggest we keep the current naming scheme for
gentoo-sources, as it
seems most correct.

Regards,
Brix
-- 
Henrik Brix Andersen <brixgentoo.org>
Gentoo Metadistribution | Mobile computing herd

Henrik Brix Andersen wrote:
> On Tue, Mar 28, 2006 at 05:48:13PM +0100, Daniel Drake
wrote:
>> John Mylchreest wrote:
>>> Since were bunlding these whole now, should we
just start naming the
>>> ebuilds correctly instead?
>> I don't think so. 2 -stable releases in the 2.6.15
cycle didn't have any 
>> immediate corresponding gentoo-sources bump since
the patches were 
>> already included. The fact that we include patches
alongside -stable 
>> means that using their notation isn't entirely
accurate in our situation.
> 
> Both naming schemes have their advantages, but I see
more
> disadvantages with using the 2.6.x.y naming scheme in
gentoo-sources
> than I see advantages (primarily the reason Daniel
stated above).
> 
> I suggest we keep the current naming scheme for
gentoo-sources, as it
> seems most correct.
> 
> Regards,
> Brix

fwiw, I've got mips-sources modified to follow the upstream
kernel.org notation 
of 2.6.x.y (even though our primary upstream,
linux-mips.org, doesn't utilize 
this notation at all).

There's logic in the ebuild that can switch between using
-rc, point releases 
(the .y ones) and standard versions, and it modifies the
logic to fetch the 
appropriate patch(es) from upstream to save on what we need
to upload to the 
gentoo mirrors.

Been following this since ~2.6.12, and we haven't had any
issues with the 
versioning in portage or reported from users.  mips-headers
even uses it now.


--Kumba

-- 
Gentoo/MIPS Team Lead
Gentoo Foundation Board of Trustees

"Such is oft the course of deeds that move the wheels
of the world: small hands 
do them because they must, while the eyes of the great are
elsewhere."  --Elrond
-- 
gentoo-kernelgentoo.org mailing list

Hi Kumba,

On Tue, Mar 28, 2006 at 11:37:15PM -0500, Kumba wrote:
> fwiw, I've got mips-sources modified to follow the
upstream kernel.org 
> notation of 2.6.x.y (even though our primary upstream,
linux-mips.org, 
> doesn't utilize this notation at all).

mips-sources doesn't use genpatches, so the two can not
really be
compared.

> There's logic in the ebuild that can switch between
using -rc, point 
> releases (the .y ones) and standard versions, and it
modifies the logic to 
> fetch the appropriate patch(es) from upstream to save
on what we need to 
> upload to the gentoo mirrors.

The logic for doing so in genpatches based sources are in
kernel-2.eclass.

Regards,
Brix
-- 
Henrik Brix Andersen <brixgentoo.org>
Gentoo Metadistribution | Mobile computing herd

On Tue, Mar 28, 2006 at 11:37:15PM -0500, Kumba
<kumbagentoo.org> wrote:
> Henrik Brix Andersen wrote:
> fwiw, I've got mips-sources modified to follow the
upstream kernel.org 
> notation of 2.6.x.y (even though our primary upstream,
linux-mips.org, 
> doesn't utilize this notation at all).
> 
> There's logic in the ebuild that can switch between
using -rc, point 
> releases (the .y ones) and standard versions, and it
modifies the logic to 
> fetch the appropriate patch(es) from upstream to save
on what we need to 
> upload to the gentoo mirrors.
> 
> Been following this since ~2.6.12, and we haven't had
any issues with the 
> versioning in portage or reported from users. 
mips-headers even uses it 
> now.

On that note, whats the chances of getting
mips-sources/headers to use
kernel-2? ALl the logic exists already in there anyways (has
done for
ages) and it would reduce that 500 odd line ebuild down to
~50 

-- 
Role:            Gentoo Linux Kernel Lead
Gentoo Linux:    http://www.gentoo.org
Public Key:      gpg --recv-keys 9C745515
Key fingerprint: A0AF F3C8 D699 A05A EC5C  24F7 95AA 241D
9C74 5515