[prev in list] [next in list] [prev in thread] [next in thread] 

List:       mandrake-cooker
Subject:    Re: [Cooker] dup-invalid bugs
From:       Gary Greene <greeneg () student ! gvsu ! edu>
Date:       2003-03-11 7:43:19
[Download RAW message or body]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Monday 10 March 2003 2:45 pm, Buchan Milne wrote:
> Gary Greene wrote:
> > On Monday 10 March 2003 1:41 pm, Charles A Edwards wrote:
> >
> > I've pesonally been verifying that as newer versions of the packages
>
> came out
>
> > if the bugs ver made invalid or not. Please don't undermine the work
>
> that I
>
> > am doing before you consult with me as the bug filer! This kind of action
> > makes the entire bugzilla system useless if you are going to do such
>
> with out
>
> > even trying to contact and verify these things with the bug filer.
>
> It helps though to not file duplicates ... you did file at least two of
> that one AFAICS, so I have marked them as dupes ... but nothing else.
>
> Buchan

Thanks Buchan. I apologize for the tone of my post. I really should have 
followed my personal mantra about email: step away, reflect, then post. It 
however, felt like I was not getting any kind of confirmation that this bug 
is a real issue though. When another Cooker posted that the reports were 
invalid, it infuriated me since the bug is most definitely there. (with the 
most recent system.)

The most recent aspect of the 1160/3126 bug reports (which are slowing up 
determining the other problems) are that the module versions are wrong. I 
realize that there is very little chance that the DRI module could be THAT 
old however. Thus I think that the test for the module may be erroneous. Once 
that is fixed and I'll be able to load the DRI module again, I will continue 
to work on the DRM logout issue until the culprit is found and fixed.

I am trying very hard to test this issue thoroughly. I've been talking to the 
XFree86 project development members that are responsible for the radeon 
driver and the only things I'm getting from them is that the bug doesn't 
occur for them. However, the one who responded there isn't even running an 
ix86 machine to test on. :(

As near as I can tell, and please bear with me since I'm not a kernel hacker, 
the DRI module is to blame. I theorize that when the window manager tries to 
SIGHUP the running server, something in the code causes a race that bombs the 
kernel's IO subsystem, thus the reason that we aren't getting any opps or 
core information. I know that this problem DIDN'T occur under XFree86 4.2.1 
at any time with the 7500 R100 Radeons, however. The only time this occurs is 
when the DRI module is loaded and active.

- -- 
Gary L. Greene, Jr.
Sent from uriel.gvsu.edu
 01:53:57 up 2 days, 17:34,  4 users,  load average: 0.12, 0.05, 0.01
============================================================
Founder and president of the Grand Valley Linux Users Group.
  -=<http://www.gvlug.org>=-
PHONE : 331-0562
EMAIL : greeneg@student.gvsu.edu
============================================================
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE+bZOayPw381UL7WcRAiRWAJ0Y0bmTrV3Tlw+c0aIZdWiKwyyPKACdGDK2
CE7OFqkDReSaFvUJSHIe+70=
=IR0X
-----END PGP SIGNATURE-----



[prev in list] [next in list] [prev in thread] [next in thread]