<div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">That strange can you send the xorg log. In the old org log you did give<br>it was using crtc 0 and i don't see how this
<br>would change anythings as they are handled the same way. For the 60Hz<br>this is because edid give this mode so<br>avivo use this one and likely simply ignore xorg provided mode. You can<br>try to add this mode with xrandr.
</blockquote><div><br>I've attached three sets of three files so you can compare them. For each set, I stared X and using the .xinitrc, ran "xrandr --verbose -q", ran "radeondump -d rdump", then exited and saved off the Xorg log file.
<br><br>All runs are with my CRT - which I believe does not support EDID. For me to run the avivo driver at all, I am using a patch to limit fb_size to 256M; this patch is included as limit_256M.patch.<br>- The "blanked.*" files are what I get when my CRT comes up blanked/power saving mode. (This should be similar to the Xorg log I sent before, but it is included here for completeness.
<br>- The "forced_crtc0.*" files are the results with the attached force_crtc0.patch applied. With this patch (which I mentioned in my last email), I can see and use the display normally.<br>- The third set is what I get with the fglrx driver instead. Notably, with fglrx the display is 1024x768 @ 75Hz instead of the 1024x768 @ 60Hz I see with the avivo driver.
<br> </div>I think I may have stumbled across a different problem, too. With my force_crtc0.patch applied, I tried changing the screen mode using xrandr, but every time I tried the display blanked/power save -- unlike the other blanking problem I'm seeing, the screen returns to normal when I exit X (with Ctrl-Alt-Backspace). Even when I used "xrandr -s 1024x768" while already in 1024x768 it blanked. The attached "mode_change.before" and "mode_change.after" are copies of the Xorg log file just before and after that xrandr command. When the mode change happens, lots of new modeline references appear in the log as well as a set where crtc(0) is being reconfigured. Those messages don't include the same references to the VGA connector as the earlier set -- is it possible the VGA connector is being disabled during the reconfiguration?
<br><br>Thanks,<br>Robert<br></div>