Created attachment 31502 [details] dmesg (from syslog) Trying to play this game: ufoai.ninex.info, as soon as you choose to enter a mission there is a GPU lockup and then the radeon driver hangs. It is impossible to switch to the console, or otherwise kill the game at this point. I can ssh in, but there is no way to get back X into a working state (or killing it either). Only a reboot helps. This happens using r600 classic driver from git mesa. Tried it twice: locked up system both times. The GPU lockup message: Sep 24 18:50:23 deb0 kernel: [32073.983546] ------------[ cut here ]------------ Sep 24 18:50:23 deb0 kernel: [32073.983563] WARNING: at drivers/gpu/drm/radeon/radeon_fence.c:235 radeon_fence_wait+0x247/0x2e5 [radeon]() Sep 24 18:50:23 deb0 kernel: [32073.983565] Hardware name: GA-870A-UD3 Sep 24 18:50:23 deb0 kernel: [32073.983566] GPU lockup (waiting for 0x007483F0 last fence id 0x007483ED) Sep 24 18:50:23 deb0 kernel: [32073.983567] Modules linked in: msr cpuid ip6table_filter ip6_tables ipt_MASQUERADE iptable_nat nf_nat nf_conntrack_ipv4 nf_defrag_ipv4 xt_state nf_conntrack ipt_REJECT xt_tcpudp iptable_filter ip_tables x_tables bridge stp llc powernow_k8 cpufreq_conservative cpufreq_powersave cpufreq_stats cpufreq_userspace cpufreq_ondemand kvm_amd kvm fuse ipv6 radeon ttm drm_kms_helper i2c_algo_bit cfbcopyarea cfbimgblt cfbfillrect drm freq_table mperf it87 hwmon_vid firewire_sbp2 loop snd_hda_codec_atihdmi snd_hda_codec_realtek hid_a4tech usbhid snd_hda_intel snd_hda_codec snd_hwdep snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_oss snd_seq_midi_event snd_seq snd_timer snd_seq_device evdev k10temp hwmon parport_pc snd parport i2c_piix4 xhci_hcd wmi button soundcore snd_page_alloc processor ohci_hcd ehci_hcd sr_mod firewire_ohci firewire_core crc_itu_t cdrom usbcore ssb nls_base thermal thermal_sys [last unloaded: scsi_wait_scan] Sep 24 18:50:23 deb0 kernel: [32073.983598] Pid: 4288, comm: Xorg Not tainted 2.6.36-rc5-phenom #103 Sep 24 18:50:23 deb0 kernel: [32073.983600] Call Trace: Sep 24 18:50:23 deb0 kernel: [32073.983605] [<ffffffff8103be92>] warn_slowpath_common+0x80/0x98 Sep 24 18:50:23 deb0 kernel: [32073.983607] [<ffffffff8103bf3e>] warn_slowpath_fmt+0x41/0x43 Sep 24 18:50:23 deb0 kernel: [32073.983618] [<ffffffffa02b2b9b>] radeon_fence_wait+0x247/0x2e5 [radeon] Sep 24 18:50:23 deb0 kernel: [32073.983620] [<ffffffff81056a4d>] ? autoremove_wake_function+0x0/0x34 Sep 24 18:50:23 deb0 kernel: [32073.983630] [<ffffffffa02b33cf>] radeon_sync_obj_wait+0xc/0xe [radeon] Sep 24 18:50:23 deb0 kernel: [32073.983635] [<ffffffffa027d8cb>] ttm_bo_wait+0xc0/0x15d [ttm] Sep 24 18:50:23 deb0 kernel: [32073.983646] [<ffffffffa02c4eae>] radeon_gem_wait_idle_ioctl+0xa7/0xf7 [radeon] Sep 24 18:50:23 deb0 kernel: [32073.983654] [<ffffffffa023d877>] drm_ioctl+0x26c/0x32e [drm] Sep 24 18:50:23 deb0 kernel: [32073.983665] [<ffffffffa02c4e07>] ? radeon_gem_wait_idle_ioctl+0x0/0xf7 [radeon] Sep 24 18:50:23 deb0 kernel: [32073.983668] [<ffffffff8100abca>] ? save_i387_xstate+0x139/0x1cb Sep 24 18:50:23 deb0 kernel: [32073.983671] [<ffffffff81300300>] ? input_event_to_user+0x4f/0x5f Sep 24 18:50:23 deb0 kernel: [32073.983675] [<ffffffffa0109d29>] ? evdev_read+0x20c/0x22d [evdev] Sep 24 18:50:23 deb0 kernel: [32073.983678] [<ffffffff810f5ec0>] do_vfs_ioctl+0x42d/0x47c Sep 24 18:50:23 deb0 kernel: [32073.983680] [<ffffffff810f5f60>] sys_ioctl+0x51/0x74 Sep 24 18:50:23 deb0 kernel: [32073.983683] [<ffffffff81002a2b>] system_call_fastpath+0x16/0x1b Sep 24 18:50:23 deb0 kernel: [32073.983684] ---[ end trace 037a2fc41d6645c9 ]--- The hung driver message: Sep 24 18:53:50 deb0 kernel: [32280.810244] INFO: task ufo:8518 blocked for more than 120 seconds. Sep 24 18:53:50 deb0 kernel: [32280.810246] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. Sep 24 18:53:50 deb0 kernel: [32280.810248] ufo D 0000000000000000 0 8518 8517 0x00000004 Sep 24 18:53:50 deb0 kernel: [32280.810250] ffff880126fcfc58 0000000000200082 ffff880126fcfbe8 0000000000000000 Sep 24 18:53:50 deb0 kernel: [32280.810253] ffff880126fcffd8 ffff880126fcffd8 ffff88012e00d808 ffff880124e819f0 Sep 24 18:53:50 deb0 kernel: [32280.810255] ffff880124e819e8 ffff880126fce000 ffff880126fce000 ffff880126fce000 Sep 24 18:53:50 deb0 kernel: [32280.810257] Call Trace: Sep 24 18:53:50 deb0 kernel: [32280.810262] [<ffffffff813d86fb>] __mutex_lock_slowpath+0x119/0x18b Sep 24 18:53:50 deb0 kernel: [32280.810278] [<ffffffffa02b2897>] ? radeon_fence_destroy+0x69/0x6e [radeon] Sep 24 18:53:50 deb0 kernel: [32280.810280] [<ffffffff813d8545>] mutex_lock+0x16/0x2d Sep 24 18:53:50 deb0 kernel: [32280.810290] [<ffffffffa02c6de1>] radeon_cs_ioctl+0x39/0x1a4 [radeon] Sep 24 18:53:50 deb0 kernel: [32280.810294] [<ffffffffa027d803>] ? ttm_bo_unreserve+0x115/0x11d [ttm] Sep 24 18:53:50 deb0 kernel: [32280.810304] [<ffffffffa02c500a>] ? radeon_gem_busy_ioctl+0x10c/0x11e [radeon] Sep 24 18:53:50 deb0 kernel: [32280.810312] [<ffffffffa023d877>] drm_ioctl+0x26c/0x32e [drm] Sep 24 18:53:50 deb0 kernel: [32280.810318] [<ffffffff810cbe0b>] ? mmap_region+0x391/0x4bf Sep 24 18:53:50 deb0 kernel: [32280.810328] [<ffffffffa02c6da8>] ? radeon_cs_ioctl+0x0/0x1a4 [radeon] Sep 24 18:53:50 deb0 kernel: [32280.810330] [<ffffffff813dcb03>] ? do_page_fault+0x3ef/0x42c Sep 24 18:53:50 deb0 kernel: [32280.810333] [<ffffffff810f5ec0>] do_vfs_ioctl+0x42d/0x47c Sep 24 18:53:50 deb0 kernel: [32280.810334] [<ffffffff810f5f60>] sys_ioctl+0x51/0x74 Sep 24 18:53:50 deb0 kernel: [32280.810337] [<ffffffff81002a2b>] system_call_fastpath+0x16/0x1b Attached .config and the dmesg (copied from syslog, system rebooted already).
Created attachment 31512 [details] .config
Tracking this with fdo bugzilla https://bugs.freedesktop.org/show_bug.cgi?id=30383