08:02hrw: http://pastebin.com/XUNaDE4Z shows strange thing:
08:02hrw: [ 12.251909] WARNING: CPU: 3 PID: 341 at lib/dma-debug.c:1087 check_unmap+0x8bc/0x9e0()
08:02hrw: [ 12.259800] nouveau 0000:01:00.0: DMA-API: device driver tries to free an invalid DMA memory address
08:10mlankhorst: you're probably trying to free something not mapped by the dma api
08:10mlankhorst: oh right the -14 looks familiar
08:11mlankhorst: hrw: iirc it's the glob->dummy_read_page that fails..
08:12hrw: mlankhorst: http://pastebin.com/EKMUSTVu has more data
08:14mlankhorst: hrw: probably safe to ignore :P just fixup dummy_read_page
08:15hrw: have to grab better card than gts250 first. it is too loud
08:15mlankhorst: hrw: iirc there is no GFP_DMA32 on arm64 but only GFP_DMA, which is why that failed
08:16mlankhorst: just fix it up and you'll get further
21:31xuhdev: Hi, I have a display connected by a displayport to HDMI cable. When I unplugged the cable and replugged in the cable, the system hangs (not only the display).
21:31xuhdev: dmesg is here http://pastebin.com/yK1Pv65P
21:31xuhdev: The last two lines seem indicating this is a nouveau problem
21:32imirkin: ouch, that seems bad... were there any more lines?
21:32xuhdev: The full dmesg is pasted in the link above
21:34xuhdev: This is the only display this machine is connected to
21:40imirkin: xuhdev: can you see if you can get the full BUG output, maybe netconsole?
21:41xuhdev: What do you mean, I'm not familiar with driver development...
21:42imirkin: search around for netconsole... basically a way to send messages to a second machine
21:45xuhdev: ok, hold on a moment
22:00xuhdev: imirkin: http://pastebin.com/xWTCDrQG
22:01imirkin: that's really useful, thanks!
22:01imirkin: skeggsb_: --^
22:01imirkin: [ 1951.120229] RIP: 0010:[<ffffffffa00ec840>] [<ffffffffa00ec840>] nouveau_i2c_try_acquire_pad+0x30/0x70 [nouveau]
22:01xuhdev: no problem!
22:01imirkin: this seems somehow familiar
22:15imirkin: xuhdev: looks like the same issue as https://bugs.freedesktop.org/show_bug.cgi?id=84101
22:15imirkin: xuhdev: you should add yourself on that bug
22:16imirkin: and there's another instance of it at https://bugzilla.redhat.com/show_bug.cgi?id=1179647
22:17xuhdev: Ah, thanks!
22:19imirkin: your situation is a tad different since you get it on a plug unplug, so perhaps mention it in the bug. i suspect the underlying cause is the same. perhaps also attach your vbios
22:23xuhdev: I also noticed that the original bug reporter says he can still log in through ssh, but in my situation it's completely crashed
22:23imirkin: prolly coz the workqueue gets stuck