Re: stats for pure64 vs. gcc-3.4

2004-11-13 Thread Rafael Rodríguez
Just asking... where can i learn about all this gcc-3.4 vs pure64 stuff? I 
read the debian-amd64-howto and installed my system, but never heard about 
this two-branches stuff 

thx in advance,

Rafael Rodríguez

El Viernes, 12 de Noviembre de 2004 23:32, Kurt Roeckx escribió:
> On Fri, Nov 12, 2004 at 02:09:30PM -0800, kristian kvilekval wrote:
> > Are there any online stats or recommendations
> > when to use pure64 vs.  gcc-3.4 repositories?
> >
> > I've noticed that some pure64 missing applications (evolution2)
> > are showing up in the gcc-3.4.
>
> evolution 2.0.2-3 is in pure64 since today.  The reason this
> took longer to get in pure64 than gcc-3.4 is that I (pure64)
> always wait for "upstream debian" to fix it in sid.  It was
> waiting for evolution-data-server which was fixed in yesterdays
> upload which made building evolution (and a few other related)
> package possible today.  I'm trying to get the difference between
> upstream sources and ours as small as possible.
>
> The gcc-3.4 archive on the other hand applies many more patches
> itself, basicly because it has to.  There are alot of packages
> that fail to build using gcc-3.4, and upstream debian isn't
> always that fast with applying patches.
>
> For the stats part.  They just started their daily build cycle,
> so I don't know the result of what is going to be installed when
> they're done, but here are the currents stats for both archives:
>
> There are stats available in the "amd64.txt" file and in the
> "dists/sid" version of each archive.
>
> Pure64:
> Installed   : 5531 ()
> Installed+all   : 8466 ()
> Needs-Build : 43
> Building: 1 ()
> Uploaded: 16 ()
> Failed  : 120
> Dep-Wait: 69 ()
> Failed-Removed  : 0
> Dep-Wait-Removed: 0
> Not-For-Us  : 94
> total   : 5675
> total+all   : 8715
>
> gcc-3.4:
> Installed   : 5600 ()
> Installed+all   : 8606 ()
> Needs-Build : 64
> Building: 0 ()
> Uploaded: 0 ()
> Failed  : 99
> Dep-Wait: 24 ()
> Failed-Removed  : 0
> Dep-Wait-Removed: 0
> Not-For-Us  : 67
> total   : 5747
> total+all   : 8794
>
>
> Now, the most interesting part is why are they different?
>
> The first thing you'll notice is that gcc-3.4 seems to have 69
> pakcakges more installed.  And 71 arch all packages more
> installed, having a total of 140 packages more installed.
>
> The arch all package all seem to be related to java.  I don't
> know if kaffe works on the gcc-3.4 archive, but there are
> currently problems with it on i386 and pure64.
>
> The other are mostly packages that just fail to build and have a
> patch available in the bug tracking system, but isn't uploaded
> yet.
>
> Then there are some packages that shouldn't be build on amd64 in
> the first place.  For instance there are mono (C#) and ada
> pacakges in the gcc-3.4 archive for which upstream says amd64 is
> not supported.
>
>
> Kurt




Re: stats for pure64 vs. gcc-3.4

2004-11-12 Thread Kurt Roeckx
On Fri, Nov 12, 2004 at 02:09:30PM -0800, kristian kvilekval wrote:
> Are there any online stats or recommendations
> when to use pure64 vs.  gcc-3.4 repositories?
> 
> I've noticed that some pure64 missing applications (evolution2)
> are showing up in the gcc-3.4.

evolution 2.0.2-3 is in pure64 since today.  The reason this
took longer to get in pure64 than gcc-3.4 is that I (pure64)
always wait for "upstream debian" to fix it in sid.  It was
waiting for evolution-data-server which was fixed in yesterdays
upload which made building evolution (and a few other related)
package possible today.  I'm trying to get the difference between
upstream sources and ours as small as possible.

The gcc-3.4 archive on the other hand applies many more patches
itself, basicly because it has to.  There are alot of packages
that fail to build using gcc-3.4, and upstream debian isn't
always that fast with applying patches.

For the stats part.  They just started their daily build cycle,
so I don't know the result of what is going to be installed when
they're done, but here are the currents stats for both archives:

There are stats available in the "amd64.txt" file and in the
"dists/sid" version of each archive.

Pure64:
Installed   : 5531 ()
Installed+all   : 8466 ()
Needs-Build : 43
Building: 1 ()
Uploaded: 16 ()
Failed  : 120
Dep-Wait: 69 ()
Failed-Removed  : 0
Dep-Wait-Removed: 0
Not-For-Us  : 94
total   : 5675
total+all   : 8715

gcc-3.4:
Installed   : 5600 ()
Installed+all   : 8606 ()
Needs-Build : 64
Building: 0 ()
Uploaded: 0 ()
Failed  : 99
Dep-Wait: 24 ()
Failed-Removed  : 0
Dep-Wait-Removed: 0
Not-For-Us  : 67
total   : 5747
total+all   : 8794


Now, the most interesting part is why are they different?

The first thing you'll notice is that gcc-3.4 seems to have 69
pakcakges more installed.  And 71 arch all packages more
installed, having a total of 140 packages more installed.

The arch all package all seem to be related to java.  I don't
know if kaffe works on the gcc-3.4 archive, but there are
currently problems with it on i386 and pure64.

The other are mostly packages that just fail to build and have a
patch available in the bug tracking system, but isn't uploaded
yet.

Then there are some packages that shouldn't be build on amd64 in
the first place.  For instance there are mono (C#) and ada
pacakges in the gcc-3.4 archive for which upstream says amd64 is
not supported.


Kurt




stats for pure64 vs. gcc-3.4

2004-11-12 Thread kristian kvilekval
Are there any online stats or recommendations
when to use pure64 vs.  gcc-3.4 repositories?

I've noticed that some pure64 missing applications (evolution2)
are showing up in the gcc-3.4.

Has gcc-3.4 cought up with pure64 in terms of packages and stability?

-- 
kristian kvilekval <[EMAIL PROTECTED]>