On Wed, Sep 12, 2007 at 02:32:57PM -0500, Gordon Pedersen wrote: > On 9/12/07, Andrew Sackville-West <andrew@farwestbilliards.com> wrote: > > On Tue, Sep 11, 2007 at 04:21:31PM -0500, Gordon Pedersen wrote: > > > On 9/11/07, Andrew Sackville-West <andrew@farwestbilliards.com> wrote: > > > > On Tue, Sep 11, 2007 at 02:11:25PM -0500, Gordon Pedersen wrote: > > > > > > > > log. Quick check verifies that there is just the builtin 82845G > > > > > chipset video - no extra video card - and certainly no second > > > > > monitor(!!). > > > > > > I reinstalled xserver-xorg then followed these suggestions: > > > . Added to xorg.conf: > > > . Section "Device" Option "Monitor-VGA" "Hitachi 803" > > > . Section "Monitor" > > > Modeline "1600x1200" 202.50 1600 1664 1856 2160 1200 1201 1204 > > > 1250 +hsync +vsync > > > > > > > > More good news - X does seem to use the 1600x1200 Modeline specified > > > in xorg.conf Monitor section. > > > > again, does your monitor support that resolution? It probably won't > > keep xorg from starting but I'd hate for you to fry the monitor... > > I was pretty sure but I doublechecked: > see http://www.monitorworld.com/Monitors/nsahitachi/superscansupreme803cm803mu.html > > Max Resolution: 1600 x 1280 okay. > > > > But still X does not start. See errors immediately below, followed > > > by xorg.conf and Xorg.0.log. > > > > I assume its the same thing as before? back to the cli? > > Yes, same symptom - back to the cli. okay. ... > > > > one think that I'm having trouble with, there is no error message that > > indicates a failure. Typically, you'll see something like: > > > > (EE) intel(0): No valid modes. > > (EE) Screen(s) found, but none have a usable configuration. > > > > Fatal server error: > > no screens found > > > > and either you're snipping that or its not there. > > When I report those (EE) messages I use grep to gather all of them. > I might forget on occasion to look at very bottom of the Xorg.0.log > file, I'm getting so bored with it, so it's good that you prompt me to > be sure to do so!! well, still, there is no "Fatal server error:" message at the end of your Xorg.0.log which I find baffling and is the reason I'm barking up the tree below... > > > > > Lets review plese how you are starting this system. I know you are > > using some startx incantation. What config file are you using to > > control that xsession, and what are the contents of that file? > > > > The reason I ask is, lacking those error messages above, I think you > > are *may* be successfully starting an x session and then automatically > > killing it. If you combine that with a mode that your monitor doesn't > > support, and you won't even see the typical X grey screen with "x" cursor". > > I issue a "startx -- -logverbose" command every time during these trials. per man startx, the lack of a client before the -- means that it tries to read ~/.xinitrc and then the xinitrc in xinit's library directory, which seems to be /etc/X11/xinit/xinitrc whic subsequently sources /etc/X11/Xsession which then goes on to source a whole bunch of other stuff. Lets try removing some of the guesswork here and either do: startx x-terminal-emulator -- -logverbose or edit ~/.xinitrc to include just x-terminal-emulator and then try again. Also, is there anything in your ~/.xsession-errors? > > xorg.conf and complete Xorg.0.log come after the following comments. > > Interesting to me that it still starts in 1600x1200 even though I am > now issuing a Modeline using 1024x768 in xorg.conf. that's probably because of the way xrandr seems to parse through the available modes. Your supplied modeline probably shows up farther down the list than the probed 1600x1200 and thus doesn't get selected. I really don't understand how xrandr works and so thats all assumption based on my wild guesses. > > After repetitive "Vblank Pipe Setup Failed" errors, > googled: i830 vblank pipe setup failed > > I won't try to summarize what I saw from Google, which is to me not > meaningful, and maybe inconclusive, maybe not. One bug report on > freedesktop.org does point to a "Wang Zhenyu" > zhenyu.z.wang at intel.com who was earlier this year fixing bugs in > this driver. None that I saw had outright failure to start as a > symptom but it wasn't an exhaustive survey of the bugs. ;-) > > Still sounds to me like a driver with known problems, and possibly > unworkable for my hardware right at this moment in time. well, I think that's a fair assumption. how about the relevant lspci -v output, just for kicks and grins. > > In separate post to come I will give detail on the failed vesa > driver X session. Different errors, and maybe solvable. bummer. vesa "Always" works. > > Did I ever say that this hardware setup works just fine in an > earlier testing version I had installed last year? Using an earlier > version of all xorg, I believe. Would have to reboot to verify > exactly what. one thing to think about doing is starting over. with vesa not working, and I know you've only partially upgraded your system (am I right? or did you bring the whole thing up to lenny...). Anyway, its probably a worthwhile endeavour at some point to really purge all of X and bring it back in just to start over from scratch. Maybe doing that with just the vesa driver might get you runnning. A ps. your determination is commendable... -- current song: Weezer - My Name Is Jonas
Attachment:
signature.asc
Description: Digital signature