Bug 117212 - Position and size settings through sax2 test utility are not being saved.
Summary: Position and size settings through sax2 test utility are not being saved.
Status: VERIFIED WONTFIX
Alias: None
Product: SUSE LINUX 10.0
Classification: openSUSE
Component: X.Org (show other bugs)
Version: RC 4
Hardware: Other All
: P3 - Medium : Normal
Target Milestone: ---
Assignee: Stefan Dirsch
QA Contact: E-mail List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-09-15 13:40 UTC by Joe Harmon
Modified: 2008-05-22 18:18 UTC (History)
2 users (show)

See Also:
Found By: Other
Services Priority:
Business Priority:
Blocker: ---
Marketing QA Status: ---
IT Deployment: ---


Attachments
hwinfo --monitor (759 bytes, text/plain)
2005-09-15 13:40 UTC, Joe Harmon
Details
sax log (71.02 KB, text/plain)
2005-09-16 16:44 UTC, Joe Harmon
Details
xfine file (88 bytes, text/plain)
2005-09-19 16:13 UTC, Joe Harmon
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Joe Harmon 2005-09-15 13:40:14 UTC
I have RC4 installed and am using GNOME as my desktop. I then go into yast and
select hardware and graphic card... to get into the sax2 utility. At the end of
any changes (when finish is clicked) there is the option to save or test. I
choose test so that I can adjust the position and/or size. After adjusting the
postion or size I choose save. I then receive a dialog box explaining that my
changes will take affect the next time X is restarted. I reboot and my settings
are still the same as before I made the change.

There is a second part to this that I think is related. I am using a 19' Sony
flat panel. If I adjust the monitor to have the proper position / size, it will
work until the next time that I reboot. After I reboot the monitor will need to
be adjusted again.

Both of these problems didn't occur on all previous version of suse (9.3 and below).
Comment 1 Joe Harmon 2005-09-15 13:40:46 UTC
Created attachment 50020 [details]
hwinfo --monitor
Comment 2 Marcus Schaefer 2005-09-16 13:19:38 UTC
Hmm I need a bit more information. If you don't mind do the following:  
  
- switch to runlevel 3 and login as root  
- run   
  
    sax2 -r  
  
- click on "change configuration" and do exactly the same as you did  
  in your report. After this "test" your configuration and adapt the screen  
  as you did before.  
  
- save this configuration and exit sax  
  
- send me the log file  
  
    /var/log/SaX.log  
  
Thanks for helping 
Comment 3 Joe Harmon 2005-09-16 16:44:02 UTC
Created attachment 50186 [details]
sax log
Comment 4 Joe Harmon 2005-09-16 16:44:30 UTC
Let me know if you need anything else.
Comment 5 Marcus Schaefer 2005-09-19 15:42:07 UTC
Hmm I cannot see a problem and I cannot reproduce this. The program doing 
the adjustment can be called like this: 
 
  /usr/sbin/xfine 
 
you must be root to do this. Once the application appears please adjust as 
you did before and click on "Save". xfine will store the information in 
 
  /var/cache/xfine/ 
 
would be fine if you could send the file located there. 
 
Thanks 
Comment 6 Joe Harmon 2005-09-19 16:13:53 UTC
Created attachment 50320 [details]
xfine file
Comment 7 Joe Harmon 2005-09-19 16:15:07 UTC
I have attached the file, and after a reboot, the same problem happened (ie, the
changes were not saved)
Comment 8 Marcus Schaefer 2005-09-19 16:43:19 UTC
Ok, now I know what is going on. The configuration file provides a valid 
modeline accepted by the NV driver: 
 
  (**) NV(0): *Mode "1280x1024": 108.9 MHz, 63.6 kHz, 60.0 Hz 
  (II) NV(0): Modeline "1280x1024"  108.88  1280 1360 1496 1712  1024 1025 
1028 1060 
 
The X-Server starts and SaX runs the xfine program which asks the X-Server 
for the current active timing values which have to be the same as the 
log told us but the timings are different: 
 
   1280 1328 1440 1688 1024 1025 1028 1066 
 
As you can see the NV driver internally is using other timings which are not 
part of the configuration or the log file. xfine is searching for this 
"wrong" reference modeline to exchange it with the new mode you created while 
using xfine, but this is impossible because the reference modeline doesn't 
exist in the xorg.conf file. The result: no changes can be saved 
 
 
 
Comment 9 Stefan Dirsch 2005-09-28 19:59:55 UTC
Hmm ... I thought these timings were changed by xfine? So why should these 
still be the same as the one in the logfile when the Xserver starts? 
Comment 10 Marcus Schaefer 2005-09-29 09:33:23 UTC
xfine is first asking the X-server for the current mode timings. These 
information is used as reference and if these timings aren't the same as 
in the config file xfine has no change to save the new mode 
 
Well we tested with a NVidia Quadro FX 3000 and it works for us like 
it should so I'm sorry but we cannot reproduce that problem here 
Comment 11 Joe Harmon 2005-09-29 14:05:28 UTC
(In reply to comment #10)

> Well we tested with a NVidia Quadro FX 3000 and it works for us like 
> it should so I'm sorry but we cannot reproduce that problem here 

Is there any more information I can gather on my side to help troubleshoot this
issue?

Comment 12 Stefan Dirsch 2005-10-08 15:27:07 UTC
Raymund, in case you can get the Quadro2 MXR from Heiko, we would like to 
have it for a small test (see above). 
Comment 13 Stefan Dirsch 2005-10-11 10:05:29 UTC
Finally I could try to verify this problem on a Quadro2 MXR/MX/Go (10de:0113,
10de:0070), but I can't reproduce. I tried exactly the same modeline as
mentioned above.

  (Modeline "1280x1024"  108.88  1280 1360 1496 1712  1024 1025 1028 1060)

reconfigured this resolution with SaX2 (xfine) and configuration has been saved
by SaX2 and works. Need to close this one as WORKSFORME.


Comment 14 Joe Harmon 2005-10-11 14:29:51 UTC
As I stated before. I am more than willing to get more troubleshooting
information. I can reproduce this here, so please let me know what you need next
and I will provide it.
Comment 15 Stefan Dirsch 2005-10-11 14:36:29 UTC
I don't think that any more information I would request in this case will help.
It's unlikely that we're able to fix this issue for you. :-(
Comment 16 Stefan Dirsch 2005-10-12 12:59:36 UTC
Trying to close this as WONTFIX.
Comment 17 Joe Harmon 2005-10-12 14:49:18 UTC
(In reply to comment #15)
> I don't think that any more information I would request in this case will help.
> It's unlikely that we're able to fix this issue for you. :-(

What if I setup the exact same machine in a lab here in Provo and gave you full
access to it. Would that help? Seems if you can't duplicate it and I can provide
a workstation to you for the duplication that we could move forward on this.