Summary: | X.org 6.8.2/xorgconfig generates bad xorg.conf with SGI-1600SW-FP/#9 Rev.4 T2R4-FP monitor/card combo. | ||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Product: | xorg | Reporter: | Daniel Stone <daniel> | ||||||||||||||
Component: | App/xorgconfig | Assignee: | Adam Jackson <ajax> | ||||||||||||||
Status: | RESOLVED WONTFIX | QA Contact: | Xorg Project Team <xorg-team> | ||||||||||||||
Severity: | major | ||||||||||||||||
Priority: | high | Keywords: | regression | ||||||||||||||
Version: | 6.8.2 | ||||||||||||||||
Hardware: | x86 (IA32) | ||||||||||||||||
OS: | FreeBSD | ||||||||||||||||
Whiteboard: | |||||||||||||||||
i915 platform: | i915 features: | ||||||||||||||||
Attachments: |
|
Description
FreeDesktop Bugzilla Database Corruption Fix User
2005-06-03 10:35:07 UTC
Created attachment 2821 [details]
xorgconfig generated xorg.conf file
This is the file 'xorgconfig' generated. For some reason "1600x1024" mode is
not detected by 'xorgconfig'. Monitor and card combo work fine under
Windows2000 Pro, and worked in the past under FreeBSD4.3 and -4.5 running
XFree86.
Created attachment 2822 [details]
Xorg log file
Note the many "out of range" and "bad mode" errors in the log.
strange that DDC failed. do you have a log file from a successful run from a previous version of X? the easy fix here is to add a 1600x1024 modeline in the driver, but you'll still need to add it to your Modes list in xorg.conf. (In reply to comment #3) > strange that DDC failed. do you have a log file from a successful run from a > previous version of X? > > the easy fix here is to add a 1600x1024 modeline in the driver, but you'll still > need to add it to your Modes list in xorg.conf. No, I don't have any logs that old. The last time I _know_ it worked, was under XFree86-3.3.3 on FreeBSD4.5 (Stable, or Release, I can't recall which). I still have a copy of the config file for that, and will try it out. I haven't edited drivers before, so I'm uncertain if I can do that just with 'vi', or something else. I tried adding the "1600x1024" modes to the xorg.conf file and got some really ugly results when I ran 'startx': A weird blooming effect radiating out from the center of the screen, with the virtual terminal still showing through the "bloom"; multicolored lines were visible in the top corners of the screen, too. I just performed a clean install of FreeBSD4.11-Release on a spare HDD and the XFree86 configurator failed when I tried 'xf86cfg' and 'xf86cfg -text-mode' from the '/stand/sysinstall' screen. I'll try the old xf86config file I have archived and see if that works at all. -L. Created attachment 2843 [details]
XF86Config file for 'deathblow'
Here is a working config file for XFree86 3.3.6 on FreeBSD4.2-Release. 'Gnome'
and 'enlightenment' both work when I use this file. -Lance
Created attachment 2844 [details]
Output from 'X -version' on 'deathblow'
Here is the output of 'X -version' from 'deathblow'. -Lance
Created attachment 2845 [details]
'uname -a' out from 'deathblow'.
'uname -a' out from 'deathblow'. -Lance
Created attachment 2846 [details]
Some output from the X server on 'deathblow'.
Some output from the X server on 'deathblow'. -Lance
Comment on attachment 2846 [details]
Some output from the X server on 'deathblow'.
Now that I've got a working config file for XFree86, I will try it out on my
FreeBSD4.11-Release system. I will probably have to merge bits of this config
file with the default file. After that I'll see if I can use it with X.org,
which was my original complaint...
Even on the FreeBSD4.2 system error messages would show up VTY1 while 'Gnome'
and 'enlightenment' are running, though I can't seem to capture them. I tried
using commands like "startx >startx-output-deathblow" to redirect any error
output into a file; that didn't work. I was unable to scroll-back far enough on
VTY1 to capture all of the output, and couldn't paste the captured info I did
get into a file. (May try this again differently.) Any tips would be helpful.
Note: I'll update the full pathname for the file 'ServerOut-1.deathblow' in a
few minutes.
I hope some of this is useful.
-Lance
(In reply to comment #9) > (From update of attachment 2846 [details] [edit]) > Now that I've got a working config file for XFree86, I will try it out on my > FreeBSD4.11-Release system. I will probably have to merge bits of this config > file with the default file. After that I'll see if I can use it with X.org, > which was my original complaint... > > Even on the FreeBSD4.2 system error messages would show up VTY1 while 'Gnome' > and 'enlightenment' are running, though I can't seem to capture them. I tried > using commands like "startx >startx-output-deathblow" to redirect any error > output into a file; that didn't work. I was unable to scroll-back far enough on > VTY1 to capture all of the output, and couldn't paste the captured info I did > get into a file. (May try this again differently.) Any tips would be helpful. > > Note: I'll update the full pathname for the file 'ServerOut-1.deathblow' in a > few minutes. > > I hope some of this is useful. > > -Lance > In case it is important, the full path-name for the file noted above is "/tmp/. XF86Setup3351/12b61b8e-33ace106/ServerOut-1". -L. Created attachment 2862 [details]
Xorg.0.log
Ajax:
Attached is a photo of the SGI monitor illustrating how the window is
off-center using "1280x1024" mode.
One of the differences I noticed between Xorg and XFree86 is a lack of
accelerated mode in the config file:
# File generated by xf86config.
<snip>
# **********************************************************************
# Screen sections
# **********************************************************************
Section "Screen"
Driver "accel"
Device "9R4"
Monitor "1600SW"
Subsection "Display"
Depth 32
Modes "1600x1024d32"
ViewPort 0 0
EndSubsection
DefaultColorDepth 32
EndSection
# eof
Are the X.org drivers/servers accelerated?
-L.
Original comments and bug reporter e-mail lost in bugzilla disk death. xorg-team archives show: Summary: X.org 6.8.2/xorgconfig generates bad xorg.conf with SGI- 1600SW-FP/#9 Rev.4 T2R4-FP monitor/card combo. Product: xorg Version: 6.8.2 Platform: PC OS/Version: FreeBSD Status: NEW Keywords: regression Severity: major Priority: P2 Component: App/xorgconfig AssignedTo: xorg-team at lists.x.org ReportedBy: lanceharrison at mindspring.com First use of X.org on a clean install of FreeBSD5.4-Release generates an xorg. conf file that is non-optimal for the SGI-1600SW-FP (Flat Panel) and the #9 Revoultion 4 (Ticket 2 Ride 4) 32MB AGP card. During manual configuration with 'xorgconfig' there appears to be no mechanism to set the modes for wide-screen @ "1600x1024": it's not on the pick-list. I used 31.5-121.0 as the horizontal sync range and 60 as the vertical sync range. The X server appears to work in "1280x1024" mode but the desktop is off-center (right-shifted) and runs off the bottom of the screen. I have set [Option "Flat Panel" "True"] in the /root/xorg.conf.new file and this makes no difference. ------- Additional Comments From lanceharrison at mindspring.com 2005-06-03 10:48 ------- Created an attachment (id=2821) --> (https://bugs.freedesktop.org/attachment.cgi?id=2821&action=view) xorgconfig generated xorg.conf file This is the file 'xorgconfig' generated. For some reason "1600x1024" mode is not detected by 'xorgconfig'. Monitor and card combo work fine under Windows2000 Pro, and worked in the past under FreeBSD4.3 and -4.5 running XFree86. ------- Additional Comments From lanceharrison at mindspring.com 2005-06-03 10:53 ------- Created an attachment (id=2822) --> (https://bugs.freedesktop.org/attachment.cgi?id=2822&action=view) Xorg log file Note the many "out of range" and "bad mode" errors in the log. My account info got lost, too, it seems. At any rate: I have upgraded to an ATI DVI-supported card and am using an SGI-LVDS-to-DVI adapter, so if anyone is interested in using my old #9 AGP card for testing, please let me know. Otherwise, this bug can be closed out. Many thanks, Lance Harrison Sorry about the phenomenal bug spam, guys. Adding xorg-team@ to the QA contact so bugs don't get lost in future. i think the odds are low, so i'm closing this one. thanks. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.