I've begun getting frequent Xorg crashes (averaging 1 or 2 every 4 hours) while using Eclipse (I think each time when I was using an arrow key to navigate code, but that might be coincidental) starting after a 'dnf update' on December 28. For each incident, I find an entry like the one at the pottom of this post in /var/log/messages (all mentioning miCompositeRects).

I think there was some update to the nouveau_dri.so in the updates applied on December 28 based on the timestamp below.

-rwxr-xr-x. 8 root root 12137000 Dec 28 05:18 /usr/lib64/dri/nouveau_dri.so

I think this library is part of a mesa package but I didn't see an update to a mesa package on December 28 though there was one in the updates applied December 21 per dnf history:

Upgrade mesa-dri-drivers-18.2.6-3.fc29.x86_64 @updates
Upgraded mesa-dri-drivers-18.2.6-2.fc29.x86_64 @@System

The odd thing is that I was doing the same things between December 21 and 28 that I've been doing since and didn't have any issues until after the December 28 update.

Also, odd is the fact that the timestamp on that nouveau_dri.so is 5 hours earlier than the actual 'dnf update' which was at 10 AM local ... but maybe the timestamp is from the library build time?

I'm using the Fedora Cinnamon Spin (not sure if that is relevant to this issue?) and 'uname -a' reports:

Linux mysystem.localdomain 4.19.13-300.fc29.x86_64 #1 SMP Sat Dec 29 22:54:28 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux

I ran an additional 'dnf update' on January 4 (hoping there might be some resolution - but nothing changed as far as the crash frequency).

I'm not sure if there is a simple way to revert the changes to the nouveau_dri.so since I think they happened on December 28 but I don't see any update in that set that seems relevant to the driver. Maybe there is a better way to interpret what actually changed December 28? Or maybe the issue is not really related to the nouveau_dri.so?

I apologize, in advance, if I'm asking questions the wrong way or omitting relevant details. I've not had this type of issue before with Fedora so I'm not sure what information might be most relevant.

Sample stack trace below.

Jan 6 12:48:34 mysystem systemd-coredump[31324]: Process 26526 (Xorg) of user 0 dumped core.#012#012Stack trace of thread 26526:#012#0 0x00007f035bf7a53f raise (libc.so.6)#012#1 0x00007f035bf64895 abort (libc.so.6)#012#2 0x000055c005495d10 OsAbort (Xorg)#012#3 0x000055c00549b8d9 n/a (Xorg)#012#4 0x000055c00549c74d FatalError (Xorg)#012#5 0x000055c005493015 n/a (Xorg)#012#6 0x00007f035c11b030 __restore_rt (libpthread.so.0)#012#7 0x00007f035a8ae2d0 n/a (nouveau_dri.so)#012#8 0x00007f035a9211c5 n/a (nouveau_dri.so)#012#9 0x00007f035a9c52b8 n/a (nouveau_dri.so)#012#10 0x00007f035a921d1f n/a (nouveau_dri.so)#012#11 0x00007f035a922056 n/a (nouveau_dri.so)#012#12 0x00007f035b6f5591 n/a (libglamoregl.so)#012#13 0x00007f035b6f190c n/a (libglamoregl.so)#012#14 0x00007f035b6f1ca8 n/a (libglamoregl.so)#012#15 0x00007f035b6eee72 n/a (libglamoregl.so)#012#16 0x000055c0054143cf n/a (Xorg)#012#17 0x000055c005401811 miCompositeRects (Xorg)#012#18 0x00007f035b6fcf18 n/a (libglamoregl.so)#012#19 0x000055c005408255 n/a (Xorg)#012#20 0x000055c005332d7e n/a (Xorg)#012#21 0x000055c005336dd6 n/a (Xorg)#012#22 0x00007f035bf66413 __libc_start_main (libc.so.6)#012#23 0x000055c0053207be _start (Xorg)#012#012Stack trace of thread 26533:#012#0 0x00007f035c11673c pthread_cond_wait@@GLIBC_2.3.2 (libpthread.so.0)#012#1 0x00007f035aabbc03 n/a (nouveau_dri.so)#012#2 0x00007f035aabb95b n/a (nouveau_dri.so)#012#3 0x00007f035c11058e start_thread (libpthread.so.0)#012#4 0x00007f035c03f6a3 __clone (libc.so.6)#012#012Stack trace of thread 26570:#012#0 0x00007f035c03f9d7 epoll_wait (libc.so.6)#012#1 0x000055c0054939bb n/a (Xorg)#012#2 0x000055c0054911be n/a (Xorg)#012#3 0x00007f035c11058e start_thread (libpthread.so.0)#012#4 0x00007f035c03f6a3 __clone (libc.so.6)