Discussion:
[OM Cooker] i586 uclibc compile problems
Per Øyvind Karlsen
2013-06-23 00:35:43 UTC
Permalink
does anyone know what is going on with uclibc builds on i586?
Busted compiler, for bero to fix...

On a related note(/package), could you please not push out releases of
packages just because you've made cosmetic changes to them such as this:
https://abf.rosalinux.ru/openmandriva/uClibc/commits/master ?

Also for the majority of your "cleanup" commits I've witnessed, they're
often just a matter of adjusting indenting of a few lines, or replacing
%patchX usage with %apply_patches, for which both of have too little to no
value for it to even be described as cleaning up, while possibly be at
conflict with personal preference and choices conciously made by others
previously.

Please stop blindly replacing %apply_patches everywhere, my decissions for
when to use or not are usually quite consciously made at least, making me
rather annoyed when I keep finding such commits labelled as "cleanups" with
this as sole change made to so many packages where I'd prefer for it to not
having been done to begin with..

And there certainly should be enough of far more urgent priorities to go
around for everyone and then some, what's the with the soft freeze you've
stressed so much about if this still remains as your primary focus?



--
Regards,
Per Øyvind
David Walser
2013-06-23 00:42:01 UTC
Permalink
--- On Sat, 6/22/13, Per Øyvind Karlsen <***@moondrake.net> wrote:
From: Per Øyvind Karlsen <***@moondrake.net>
Subject: Re: [Cooker] [OM Cooker] i586 uclibc compile problems
To: "Cooker OpenMandriva" <om-***@ml.openmandriva.org>, ***@zarb.org
Date: Saturday, June 22, 2013, 8:35 PM
Post by Per Øyvind Karlsen
does anyone know what is going on with uclibc builds on i586?
Busted compiler, for bero to fix...
On a related note(/package), could you please not push out releases of
packages just because you've made cosmetic changes to them such as
this: https://abf.rosalinux.ru/openmandriva/uClibc/commits/master ?
Certainly I've seen today, while looking at some packages, some commits from Matt that said cleanups, where some of the changes looked more like "unnecessary cosmetic changes." That being said, some of the cleanups are good, but wouldn't be needed if other people wouldn't make them "dirty" in the first place. Can someone please teach bero that macros that expand to a value should have braces on them? He's made quite a mess in some places by neglecting that. Thanks to Matt for cleaning it up where he sees it.
Per Øyvind Karlsen
2013-06-23 02:06:32 UTC
Permalink
Post by David Walser
Certainly I've seen today, while looking at some packages, some commits
from Matt that said cleanups, where some of the changes looked more like
"unnecessary cosmetic changes." That being said, some of the cleanups are
good, but wouldn't be needed if other people wouldn't make them "dirty" in
the first place. Can someone please teach bero that macros that expand to
a value should have braces on them? He's made quite a mess in some places
by neglecting that. Thanks to Matt for cleaning it up where he sees it.
Indeed. :)
More rpmlint checks etc. could prolly' be implemented & introduced to
better enforce such further as well, but not very good timing for giving
much priority currently..
Prolly' something worthy a broader discussion and focus with timing being
better at the beginning of next development cycle. :o)

--
Regards,
Per Øyvind
Per Øyvind Karlsen
2013-06-26 08:29:14 UTC
Permalink
On Sat, Jun 22, 2013 at 6:35 PM, Per Øyvind Karlsen <
Post by Per Øyvind Karlsen
does anyone know what is going on with uclibc builds on i586?
Busted compiler, for bero to fix...
\o/
Post by Per Øyvind Karlsen
On a related note(/package), could you please not push out releases of
https://abf.rosalinux.ru/openmandriva/uClibc/commits/master ?
Well.. I was trying to get a rebuild with a release bump to see if that
actually fixed the uclibc problem.
Post by Per Øyvind Karlsen
Also for the majority of your "cleanup" commits I've witnessed, they're
often just a matter of adjusting indenting of a few lines, or replacing
%patchX usage with %apply_patches, for which both of have too little to no
value for it to even be described as cleaning up, while possibly be at
conflict with personal preference and choices conciously made by others
previously.
And? I also catch a lot of legacy crap, convert tons of BRs to pkgconfig
provides, clean up many poorly crafted pkgs? I guess my work on libreoffice
is kinda crappy too huh?
And these things are more important than fixing actual bugs and maintaining
a focus on progress towards a release..?
Can I ask things from you too? I have a long list for POK.
By all means, ask away..
Post by Per Øyvind Karlsen
Please stop blindly replacing %apply_patches everywhere, my decissions
for when to use or not are usually quite consciously made at least, making
me rather annoyed when I keep finding such commits labelled as "cleanups"
with this as sole change made to so many packages where I'd prefer for it
to not having been done to begin with..
And? No, there is merit behind standardizing on %apply_patches. If you
wanna leave notes in the spec about the patch, leave it up by the header.
%apply_patches keeps lazy packing inline. It's easy to just comment out the
line where the line is applied. I am going to say there are always special
cases.
Well, one downside is that you cannot easily just temporarily disable
single patches or make them conditional..
I also often find myself doing things like adding an 'exit 1' after a
specific patch that I want to regenerate etc., when using %apply_patches
(which btw. is a distro specifc macro, not standardized upstream) you loose
the more fine grained control...
Post by Per Øyvind Karlsen
And there certainly should be enough of far more urgent priorities to go
around for everyone and then some, what's the with the soft freeze you've
stressed so much about if this still remains as your primary focus?
You keep talking to me like I'm the head honcho, making decisions on what
goes and what stay etc....
Well, you are the one who do restrict peoples' access and making the most
noise about others' work, callig attention to the soft freeze..
AFAIK, I'm not the release manager, and my time is purely volunteer. I
keep trying to say things of reason, ie the freeze. No one seems to listen
there so...
And what's the point of this freeze period fo you, if you don't actually
use it productively focusing on more important issues?
You were so persistent and yelling about this freeze period, but yet you
expect for the actual work on stabilizing the distro during this period to
be conducted by everyone else than yourself?
If you're not going to do any relevant of the relevant work on stabilizing
the distro as intended by this soft freeze period, then why do you call
attention to it so much?

Idk, as long as we don't hard freeze. idk why you are barking at me?
Because you are the one who yells the highest and the most eager to
restrict the access of other people who actually focus on moving forward
and doing work of greater importance (which spec cleanining is simply not
by comparison)..

--
Regards,
Per Øyvind

Loading...