[gentoo-dev] "Slacking" arches - which are stable, which aren't?

[gentoo-dev] "Slacking" arches - which are stable, which aren't?

[Steev Klimaszewski]

Not wanting to start a huge war about what arches are slacking and
which aren't - I asked in -dev on IRC and was told to check out
profiles.desc - based on this information, I closed Bug 208917 which
was about stablizing dbus-glib-0.74.  The bug was opened on 04 Feb
2008, and as of today 05 Oct 2008, the only arches left to stable it
are arm, sh, and s390 (and according to profiles.desc, they are all
dev profiles) however hoffie said that didn't seem right since he
knows things get requested for stable for those arches.

So, IS there a definitive list somewhere of what arches are stable,
and which aren't, and if so, where can it be found?  I have no problem
re-opening the bug, but as I stated in -dev, its been almost 8 months
since the last *activity* on the bug, and I doubt that they are going
to be stabling it any time soon.

Thoughts? Helps?

Re: [gentoo-dev] "Slacking" arches - which are stable, which aren't?

[Friedrich Oslage]

[-- Attachment #1: Type: text/plain, Size: 913 bytes --]

Am Sonntag, den 05.10.2008, 16:26 -0500 schrieb Steev Klimaszewski:
>
> Thoughts? Helps?
> 

Afaik we have 3 types of arches:

- experimental
They are not CCed on stablization bugs and don't do stablizations at
all.

~mips, ~sparc-fbsd and ~x86-fbsd

- unsupported
They are CCed on stablizations bugs, but they are not supported by the
Gentoo Linux Security Project. It may take quite long until they
actually do the stablization. But I'm also wondering why some of their
profiles are marked as "dev".

arm, ia64, m68k, sh, s390

- supported
Most popular arches, supported by the Gentoo Linux Security Project,
they usually do your stablizations in time unless it requires some
exotic hardware(the devs/ats don't have) to test.

alpha, amd64, hppa, sparc, ppc, ppc64, x86

Sources:
- commits logs
- http://www.gentoo.org/security/en/vulnerability-policy.xml

Cheers,
Friedrich


[-- Attachment #2: Dies ist ein digital signierter Nachrichtenteil --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-dev] "Slacking" arches - which are stable, which aren't?

[Jeremy Olexa]

Friedrich Oslage wrote:
> Am Sonntag, den 05.10.2008, 16:26 -0500 schrieb Steev Klimaszewski:
>> Thoughts? Helps?
>>
> 
> Afaik we have 3 types of arches:
> 
> - experimental
> They are not CCed on stablization bugs and don't do stablizations at
> all.
> 
> ~mips, ~sparc-fbsd and ~x86-fbsd
> 
> - unsupported
> They are CCed on stablizations bugs, but they are not supported by the
> Gentoo Linux Security Project. It may take quite long until they
> actually do the stablization. But I'm also wondering why some of their
> profiles are marked as "dev".
> 
> arm, ia64, m68k, sh, s390
> 
> - supported
> Most popular arches, supported by the Gentoo Linux Security Project,
> they usually do your stablizations in time unless it requires some
> exotic hardware(the devs/ats don't have) to test.
> 
> alpha, amd64, hppa, sparc, ppc, ppc64, x86
> 
> Sources:
> - commits logs
> - http://www.gentoo.org/security/en/vulnerability-policy.xml

I would suggest moving all the "slacking" arches to "experimental" until 
there is desire from the dev community (read: manpower) to support a 
stable tree again. Until then, it seems pretty pointless to keep 
assigning bugs to these arches and they just keep rotting there because 
no one gets around to them.

2 cents,
-Jeremy

[gentoo-dev] Re: "Slacking" arches - which are stable, which aren't?

[Ryan Hill]

[-- Attachment #1: Type: text/plain, Size: 673 bytes --]

On Sun, 05 Oct 2008 20:44:51 -0500
Jeremy Olexa <darkside@gentoo.org> wrote:

> I would suggest moving all the "slacking" arches to "experimental"
> until there is desire from the dev community (read: manpower) to
> support a stable tree again. Until then, it seems pretty pointless to
> keep assigning bugs to these arches and they just keep rotting there
> because no one gets around to them.
> 
> 2 cents,
> -Jeremy

++ $473.57


-- 
gcc-porting,                                      by design, by neglect
treecleaner,                              for a fact or just for effect
wxwidgets @ gentoo     EFFD 380E 047A 4B51 D2BD C64F 8AA8 8346 F9A4 0662

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-dev] Re: "Slacking" arches - which are stable, which aren't?

[Steev Klimaszewski]

On Sun, Oct 5, 2008 at 9:07 PM, Ryan Hill <dirtyepic@gentoo.org> wrote:
> On Sun, 05 Oct 2008 20:44:51 -0500
> Jeremy Olexa <darkside@gentoo.org> wrote:
>
>> I would suggest moving all the "slacking" arches to "experimental"
>> until there is desire from the dev community (read: manpower) to
>> support a stable tree again. Until then, it seems pretty pointless to
>> keep assigning bugs to these arches and they just keep rotting there
>> because no one gets around to them.
>>
>> 2 cents,
>> -Jeremy
>
> ++ $473.57
>
>

My aim with the email wasn't to start up this discussion so much as to
figure out which arches are supported by stable keywords, and which
ones are okay to not request stable keywords so that bugs don't sit
around for months without action on them.  I know that vapier is
pretty much the only dev with an sh or s390 box, but I know there are
a couple of people with ARM, I was just hoping we had some sort of
official list somewhere.

Re: [gentoo-dev] Re: "Slacking" arches - which are stable, which aren't?

[Jeremy Olexa]

Steev Klimaszewski wrote:
> On Sun, Oct 5, 2008 at 9:07 PM, Ryan Hill <dirtyepic@gentoo.org> wrote:
>> On Sun, 05 Oct 2008 20:44:51 -0500
>> Jeremy Olexa <darkside@gentoo.org> wrote:
>>
>>> I would suggest moving all the "slacking" arches to "experimental"
>>> until there is desire from the dev community (read: manpower) to
>>> support a stable tree again. Until then, it seems pretty pointless to
>>> keep assigning bugs to these arches and they just keep rotting there
>>> because no one gets around to them.
>>>
>>> 2 cents,
>>> -Jeremy
>> ++ $473.57
>>
>>
> 
> My aim with the email wasn't to start up this discussion so much as to
> figure out which arches are supported by stable keywords, and which
> ones are okay to not request stable keywords so that bugs don't sit
> around for months without action on them.  I know that vapier is
> pretty much the only dev with an sh or s390 box, but I know there are
> a couple of people with ARM, I was just hoping we had some sort of
> official list somewhere.
> 

I wasn't trying to go down that road either but you should know that 
this discussion will be forced there if there is to be any conclusion to 
this topic. AFAIK, it is incorrect right now to exclude s390, arm, sh, 
etc on stablereqs right now..But, I ask this question to the dev 
community: "Why?" There are ~190 open bugs with s390 as assignee or on 
the CC list. Does it *really* matter if these under-staffed "odd" arches 
have a stable tree or not? If there is a problem with a stable package 
right now, will there be a new version marked stable in a reasonable 
amount of time? I think we can conclude that having a stable tree for 
understaffed arches might cause more harm than good.

To conclude my input on the topic:
It is my opinion that filing stablereqs against these arches is silly. 
However, I will continue to do so until requested otherwise. I respect 
that people may not have enough resources or time to keep up to x86 or 
amd64, so maybe there needs to be a policy change somehow..? (or maybe 
it just needs to be clarified better)

-Jeremy

[gentoo-dev] Re: "Slacking" arches - which are stable, which aren't?

[Duncan]

Jeremy Olexa <darkside@gentoo.org> posted 48EA6FF2.8020201@gentoo.org,
excerpted below, on  Mon, 06 Oct 2008 15:07:14 -0500:

> AFAIK, it is incorrect right now to exclude s390, arm, sh, etc on
> stablereqs right now..But, I ask this question to the dev community:
> "Why?" There are ~190 open bugs with s390 as assignee or on the CC list.
> Does it *really* matter if these under-staffed "odd" arches have a
> stable tree or not?

Having been an amd64 user back when it was much smaller, and having 
followed the previous discussion on this here, including the mips -> 
experimental move, yes, it does matter.  With the bugs there's at least 
some info on a package and its stabilization potential when/if someone 
gets around to doing something about it.  Without them, the job of 
bringing them back to unsupported and then to full supported, if there's 
suddenly a leap in interest, becomes much harder as there's that much 
less info on what /was/ stable at one point, and on anything in the ~arch 
versions that might need checked before they go stable again.

So it matters; there's a practical reason for it.  However, that's not 
the same as saying it's the overall best solution at this time.  I have 
no opinion on that, particularly as I /personally/ prefer ~arch in any 
case.

-- 
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] Re: "Slacking" arches - which are stable, which aren't?

[Santiago M. Mola]

[-- Attachment #1: Type: text/plain, Size: 1598 bytes --]

El lun, 06-10-2008 a las 23:13 +0000, Duncan escribió:
> Jeremy Olexa <darkside@gentoo.org> posted 48EA6FF2.8020201@gentoo.org,
> excerpted below, on  Mon, 06 Oct 2008 15:07:14 -0500:
> 
> > AFAIK, it is incorrect right now to exclude s390, arm, sh, etc on
> > stablereqs right now..But, I ask this question to the dev community:
> > "Why?" There are ~190 open bugs with s390 as assignee or on the CC list.
> > Does it *really* matter if these under-staffed "odd" arches have a
> > stable tree or not?
> 
> Having been an amd64 user back when it was much smaller, and having 
> followed the previous discussion on this here, including the mips -> 
> experimental move, yes, it does matter.  With the bugs there's at least 
> some info on a package and its stabilization potential when/if someone 
> gets around to doing something about it.  Without them, the job of 
> bringing them back to unsupported and then to full supported, if there's 
> suddenly a leap in interest, becomes much harder as there's that much 
> less info on what /was/ stable at one point, and on anything in the ~arch 
> versions that might need checked before they go stable again.
> 

I fully agree. I think bringing some understaffed arches back to ~arch
is an option, but should be avoided if possible.

I wonder how many of these 190 open bugs in s390 are actually bugs about
brokenness, and not just regular stabilizations...

And by the way, amd64 had a similar amount of open bugs by the end of
2007.

Regards,
-- 
Santiago Moisés Mola
Jabber: cooldwind@gmail.com | GPG: AAD203B5

[-- Attachment #2: Esta parte del mensaje está firmada digitalmente --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-dev] Re: "Slacking" arches - which are stable, which aren't?

[Mike Frysinger]

[-- Attachment #1: Type: text/plain, Size: 504 bytes --]

On Monday 06 October 2008 16:07:14 Jeremy Olexa wrote:
> Does it *really* matter if these under-staffed "odd" arches
> have a stable tree or not? If there is a problem with a stable package
> right now, will there be a new version marked stable in a reasonable
> amount of time? I think we can conclude that having a stable tree for
> understaffed arches might cause more harm than good.

how exactly do you propose we sanely build stages for such arches then ?  
short answer: you cant.
-mike

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 835 bytes --]