Prev: fs: optimize mpage_bio_submit()
Next: [PATCH 3/3] perf hist: fix objdump output parsing
From: Arnaldo Carvalho de Melo on 29 May 2010 09:50
Em Sat, May 29, 2010 at 09:29:48AM +0200, Sam Ravnborg escreveu:
> On Fri, May 28, 2010 at 03:53:57PM -0300, Arnaldo Carvalho de Melo wrote:
> > Hi Michal, Sam,

> > Are you OK with with this change?

> No - I do not like it :-)

> But I like the concept.

I saw the scripts/package, but felt it wasn't quite there where it fits,
because there all existing cases build binary packages and this is
purely a source package, guess I was wrong then :-)

> We have some minimal infrastructure to support packaging
> and this seems like an OK fit for this too.
> Only drawback is that you need to configure your kernel
> before you can pack the source tarball.
>
> Moving the packaging support to scripts/package/
> avoiding any changes to the top-level Makefile.
>
> Appended is my attemt to add it to said file.
> But my git did not support the '-o' option so it is untested.
>
> To apply it you need your MANIFEST file (obviously).
>
> Help will say:
> $ make help | grep perf
> perf-pkg - Build perf-2.6.34.tar.gz source tarball
>
>
> You may consider to restructure cmd_perf a little so
> it exits on error. I just did the quick version here.

I'll try to do that later today, thanks a lot!

> Sam
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Arnaldo Carvalho de Melo on 31 May 2010 11:50
Em Sat, May 29, 2010 at 09:29:48AM +0200, Sam Ravnborg escreveu:
> On Fri, May 28, 2010 at 03:53:57PM -0300, Arnaldo Carvalho de Melo wrote:
> > Hi Michal, Sam,
> >
> > Are you OK with with this change?
>
> No - I do not like it :-)
>
> But I like the concept.
>
> We have some minimal infrastructure to support packaging
> and this seems like an OK fit for this too.
> Only drawback is that you need to configure your kernel
> before you can pack the source tarball.

Which is really annoying, can't we lift that requirement for source
tarballs? I'm looking into this now.

- Arnaldo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Michal Marek on 31 May 2010 16:20
On 31.5.2010 19:42, Arnaldo Carvalho de Melo wrote:
> +git archive --prefix=$(perf-tar)/ HEAD^{tree} \

If you use plain "HEAD" (a commit-ish) instead if HEAD^{tree}, then
git archive will store the commit id in the archive metadata and the
user can then use git get-tar-commit-id to extract it.


> + $$(cat $(srctree)/tools/perf/MANIFEST) -o $(perf-tar).tar; \
> +mkdir -p $(perf-tar); \
> +git rev-parse HEAD > $(perf-tar)/HEAD; \
> +tar rf $(perf-tar).tar $(perf-tar)/HEAD; \

.... then this HEAD file might not even be necessary (even if you keep
it, it would be nice not to break git get-tar-commit-id).

Michal
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Michal Marek on 1 Jun 2010 06:30
On 1.6.2010 01:06, Arnaldo Carvalho de Melo wrote:
> Em Mon, May 31, 2010 at 10:11:24PM +0200, Michal Marek escreveu:
>> On 31.5.2010 19:42, Arnaldo Carvalho de Melo wrote:
>>> +git archive --prefix=$(perf-tar)/ HEAD^{tree} \
>>
>> If you use plain "HEAD" (a commit-ish) instead if HEAD^{tree}, then
>> git archive will store the commit id in the archive metadata and the
>> user can then use git get-tar-commit-id to extract it.
>
> I just used what was in git.git and then when talking with a git
> enthusiast, he showed me this:
>
> [acme(a)emilia git]$ git show 9cd625b79babaf50f50a0e5d96903eaacb1ee600
> commit 9cd625b79babaf50f50a0e5d96903eaacb1ee600
> Author: Rene Scharfe <rene.scharfe(a)lsrfire.ath.cx>
> Date: Sun Jun 18 15:25:33 2006 +0200
>
> Make release tarballs friendlier to older tar versions

....

Ah, ok.


> Well, what about the other aspects of the patch, is this the only issue
> you have found with it?

The patch is OK.

Acked-by: Michal Marek <mmarek(a)suse.cz>

Michal
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Michal Marek on 7 Jun 2010 06:20
On 4.6.2010 19:35, Arnaldo Carvalho de Melo wrote:
> Em Tue, Jun 01, 2010 at 12:28:58PM +0200, Michal Marek escreveu:
>> On 1.6.2010 01:06, Arnaldo Carvalho de Melo wrote:
>>> Well, what about the other aspects of the patch, is this the only issue
>>> you have found with it?
>>
>> The patch is OK.
>>
>> Acked-by: Michal Marek <mmarek(a)suse.cz>
>
> Argh, there was one last detail, so that we don't require that the
> kernel be configured first, and as we don't use KERNELRELEASE at all in
> the -src-pkg targets, would that be OK with you too?
>
> [acme(a)emilia linux-2.6-tip]$ git diff Makefile
> diff --git a/Makefile b/Makefile
> index 0ab0c6f..6e186a1 100644
> --- a/Makefile
> +++ b/Makefile
> @@ -1215,6 +1215,8 @@ distclean: mrproper
> # rpm target kept for backward compatibility
> package-dir := $(srctree)/scripts/package
>
> +%src-pkg: FORCE
> + $(Q)$(MAKE) $(build)=$(package-dir) $@
> %pkg: include/config/kernel.release FORCE
> $(Q)$(MAKE) $(build)=$(package-dir) $@
> rpm: include/config/kernel.release FORCE
> [acme(a)emilia linux-2.6-tip]$
>
> On a make mrproper'ed kernel we get:
>
> a linux-2.6-tip]$ LANG= make perf-tarbz2-src-pkg
> /bin/sh: include/config/kernel.release: No such file or directory
> make: *** [include/config/kernel.release] Error 1
> [acme(a)emilia linux-2.6-tip]$
>
> :-\
>
> If you're ok with it, I'll add it as a follow on patch that I'll send
> via Ingo.

Looks ok to me, any future *-src-pkg target will also just package
sources and thus not depend on the configuration.

Acked-by: Michal Marek <mmarek(a)suse.cz>

Michal
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
 | 
Pages: 1
Prev: fs: optimize mpage_bio_submit()
Next: [PATCH 3/3] perf hist: fix objdump output parsing