Sebastian Hofmann wrote: > Brice Goglin wrote: > >> If 2.3.0 is broken and 2.2.0 works fine, you should do: >> git bisect start >> git bisect good xf86-video-intel-2.2.0 >> git bisect bad xf86-video-intel-2.3.0 >> >> Then git will choose a commit to test (in the middle of the above >> range). Run autogen, make and make install. If it works, you run "git >> bisect good", if it breaks, run "git bisect bad". Git will propose >> another commit in the middle of the remaining range. Do the same test, >> ... again and again until git finds the exact commit breaking things. >> >> When it is done, git bisect --reset >> > > The result of git bisect is: > > commit 5a06388bf83c97f49565ad482161645f996492c7 > Author: Jesse Barnes <[EMAIL PROTECTED]> > Date: Thu Mar 13 18:12:00 2008 -0700 > > Initial panel fitting changes > > Basic support for panel fitting. > (cherry picked from commit 05cf07071e25f84dec3476a9bed7235ed50cd249) > > > But I wonder that in the bad cases the screen remained completely black > and did not react to any keystroke, so I had to restart the computer. I > could not reproduce the behaviour I originally reported while bisect... >
I see some related commits in latest upstream git. Can you try intel 2.3.2 from unstable? If it does not help, can you try to build the upstream git? Brice -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
