With the 2.1.0 verision of the intel driver, I get dropped into a 39 Hz mode (if I understand the log file correctly) wheras I get 60 Hz using i810. Defining Modelines in the xorg.conf doesn't help one bit. They seem to be ignored. The same goes for setting the acceptable HorizSync rates in the "Monitor" section. :-(
Created attachment 11064 [details] xorg.conf
Created attachment 11065 [details] Xorg.0.log with ModesDebug=true
Created attachment 11069 [details] [review] use bios mode instead of probed mode Looks like we're misprobing the clocks, which is our workaround for the usual 855 failure of the bios tables being munged. Attached workaround will likely work, but I need to either take a look at why we're misdetecting the clock, or figure out why we fail on other peoples' bios tables.
Eric, thanks for your quick analysis, maybe I'll take the time this week-end to compile the driver from source to verify your patch fixes the issue for my machine. However, I presume that there is a possibility to override automatically detected ModeLines in the config files? So far including a specific mode line in the Monitor section as I have done in the config file attached to this bug hasn't helped at all. Surely this is some problem on my part, so I'll try to fiddle with that some more.
Eric, do you think my comment in bug 11932 also makes sense to fix this?
Eric, sorry for taking so long :( I can confirm that the patch does indeed fix the issue for me. Thanks for your help. If you need me to do further testing, I'll be glad to comply.
Eric, I can confirm that this also fixes the problem for me. If needed, I can assist in solving the mis-probing problem as the company I work for has a number of these units that I can easily test on.
I think we're pretty clear at this point that we want to move to using BIOS modes if they can be found, but we really need to solve the incorrect BIOS decode on 855 first (the rate of reports for the BIOS decode issue was far higher than for the variations on this issue).
*** Bug 11932 has been marked as a duplicate of this bug. ***
Please try the patch in bug #10773 comment #16, I think your problem is the same. We are mis-calcing the dotclock of mode. *** This bug has been marked as a duplicate of bug 10773 ***
It looks strange to me that bug #11932 got fixed as duplicate of this one and now this one is fixed as duplicate of #10773 since: * #10773 is about VGA mode flickering * #11932 is about LVDS panel size being incorrectly detected I asked my downstream users to test the patch anyway, we'll see what happens.
(In reply to comment #11) > It looks strange to me that bug #11932 got fixed as duplicate of this one and > now this one is fixed as duplicate of #10773 since: > * #10773 is about VGA mode flickering Actually it is flickering on LVDS display on their 855GM laptop. > * #11932 is about LVDS panel size being incorrectly detected > I asked my downstream users to test the patch anyway, we'll see what happens. > From comment #1 of this bug, we mis-calc the mode clock when trying to find the panel fix mode, which causes the user has a 39Hz mode. Using the BIOS vbt mode can fix this problem, but using patch in bug 10773 can also fix the problem (we fixed the mode before we go to vbios vbt parsing). Thanks, Hong
Hello Hong, yes when I apply the patch from bug #10773 comment #16, I no longer get the flickering display. Thanks a lot for your help! Bye Felix
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.