Since upgrading from xf86-video-intel-2.21.15 to 2.99.907-12.99.907, I get frequent crashes when moving or resizing windows. Such a crash leaves the graphics card completely disfunctional so that I even cannot switch to a console anymore. The dmesg output is like: [ 1762.336549] [drm] stuck on render ring [ 1762.336562] [drm] capturing error event; look for more information in /sys/class/drm/card0/error [ 1762.341549] [drm:i915_set_reset_status] *ERROR* render ring hung inside bo (0xd73f000 ctx 0) at 0xd73f12c [ 1762.409886] [drm] GMBUS [i915 gmbus vga] timed out, falling back to bit banging on pin 2 [ 1762.846568] [drm:i915_reset] *ERROR* Failed to reset chip. [ 1763.430449] ------------[ cut here ]------------ [ 1763.430472] WARNING: CPU: 0 PID: 303 at drivers/gpu/drm/i915/intel_display.c:929 assert_pll+0x64/0x70 [i915]() [ 1763.430474] PLL state assertion failure (expected on, current off) [ 1763.430475] Modules linked in: fuse iTCO_wdt iTCO_vendor_support joydev snd_hda_codec_analog coretemp kvm_intel kvm pcmcia btusb bluetooth microcode lpc_ich psmouse pcspkr i915 r592 serio_raw snd_hda_intel memstick snd_hda_codec i2c_i801 arc4 evdev yenta_socket snd_hwdep pcmcia_rsrc pcmcia_core drm_kms_helper snd_pcm snd_page_alloc snd_timer thinkpad_acpi e1000e drm nvram snd iwl4965 iwlegacy mac80211 cfg80211 soundcore i2c_algo_bit shpchp intel_agp mei_me rfkill i2c_core ac intel_gtt mei wmi thermal tpm_tis tpm tpm_bios video battery ptp pps_core button acpi_cpufreq processor ext4 crc16 mbcache jbd2 hid_generic usbhid hid sr_mod cdrom sd_mod ata_generic ahci libahci ata_piix sdhci_pci sdhci mmc_core firewire_ohci firewire_core crc_itu_t pata_acpi ehci_pci uhci_hcd ehci_hcd libata usbcore usb_common [ 1763.430534] scsi_mod [ 1763.430537] CPU: 0 PID: 303 Comm: Xorg Not tainted 3.12.7-2-ARCH #1 [ 1763.430539] Hardware name: LENOVO 64669YG/64669YG, BIOS 7LETC5WW (2.25 ) 11/14/2008 [ 1763.430541] 0000000000000009 ffff8800bafed9f8 ffffffff814ec113 ffff8800bafeda40 [ 1763.430544] ffff8800bafeda30 ffffffff81061b8d 0000000000000001 ffff8800ba7c4000 [ 1763.430547] ffff880133ec2000 0000000000000000 0000000000000000 ffff8800bafeda90 [ 1763.430550] Call Trace: [ 1763.430557] [<ffffffff814ec113>] dump_stack+0x54/0x8d [ 1763.430562] [<ffffffff81061b8d>] warn_slowpath_common+0x7d/0xa0 [ 1763.430565] [<ffffffff81061bfc>] warn_slowpath_fmt+0x4c/0x50 [ 1763.430574] [<ffffffffa0617c40>] ? i915_write32+0xb0/0x1a0 [i915] [ 1763.430584] [<ffffffffa05ddcc4>] assert_pll+0x64/0x70 [i915] [ 1763.430592] [<ffffffffa05dde2c>] intel_crtc_load_lut+0x15c/0x170 [i915] [ 1763.430601] [<ffffffffa05deb3e>] i9xx_crtc_enable+0x20e/0x420 [i915] [ 1763.430610] [<ffffffffa05e5317>] __intel_set_mode+0x857/0x13d0 [i915] [ 1763.430621] [<ffffffffa05e8336>] intel_set_mode+0x16/0x30 [i915] [ 1763.430630] [<ffffffffa05e8c33>] intel_crtc_set_config+0x7e3/0x9d0 [i915] [ 1763.430639] [<ffffffffa04b550d>] drm_mode_set_config_internal+0x5d/0xe0 [drm] [ 1763.430646] [<ffffffffa04b8197>] drm_mode_setcrtc+0xf7/0x650 [drm] [ 1763.430652] [<ffffffffa04a8c62>] drm_ioctl+0x502/0x630 [drm] [ 1763.430656] [<ffffffff810715a2>] ? __set_task_blocked+0x32/0x70 [ 1763.430661] [<ffffffff811b6435>] do_vfs_ioctl+0x2e5/0x4d0 [ 1763.430663] [<ffffffff811b66a1>] SyS_ioctl+0x81/0xa0 [ 1763.430667] [<ffffffff814fac2d>] system_call_fastpath+0x1a/0x1f [ 1763.430669] ---[ end trace 75438fb0733a4063 ]--- [ 1763.430697] ------------[ cut here ]------------ I'll attach two complete dmesg logs (one with --enable-glamor, one with --disable-glamor). I'm using Arch Linux where I reported the bug first: https://bugs.archlinux.org/task/38525
Created attachment 92143 [details] Crash when configured with --enable-glamor Since this version, the Arch Linux builds of the driver use --enable-glamor.
Created attachment 92145 [details] Crash when configured with --disable-glamor I've though the problem could be triggered by compiling with glamor support (which was introduced with this version in Arch Linux). So I recompiled the drivers with --disable-glamor, but then I got the crash whose dmesg output is attached here.
For the time being, I've switched back to xf86-video-intel-2.21.15 which works flawlessly for me.
*** This bug has been marked as a duplicate of bug 73644 ***
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.