Hi Jonas, > sorry for all the noise, I promise this is the last message I'm sending > for today.
Well, it's very useful noise. You've pretty much figured it out... > The bug doesn't happen either with the current upstream version of > xf86-video-intel. i just upgraded to the git version and i don't > experience it anymore. > > i suspect thus, it is actually in xf86-video-intel and got fixed between > version 2.15.0-1 and current upstream. If this is confirmed by others, we're probably in good shape! I suspect it might have to do with http://cgit.freedesktop.org/xorg/driver/xf86-video-intel/commit/?id=3145530feed879082bcfab11ffc8e7fd0911c920 (This is a commit from 3 days ago, fixing an issue with flushing refresh requests. Not intel965-specific, but it is a display refresh issue. There's also a couple of April 17 commits that might or might not be about processing of batch requests, one of them i965-specific). But I am really not sure. Anyway: it seems impractical to try and work around such a specific issue in xournal in an unobtrusive way -- xournal is very far removed from the intel driver in the processing chain. If the upgrade to the git version of the intel driver fixes the issue for everyone, I'd advocate contacting the xf86-video-intel maintainer in the relevant distributions to report the problem and request an update of the distribution's intel driver package to include the latest upstream commits. Thanks again for all your detective work! Denis ------------------------------------------------------------------------------ Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay _______________________________________________ Xournal-devel mailing list Xournal-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xournal-devel
