Description of problem: Boot computer and watch boot messages flash by (alternate, watch fedora infinity turn from white background to full-fledged)... and then the computer just hangs! Version-Release number of selected component (if applicable): both kernels, 4.5.3 and 4.5.4 How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info: When I booted with nomodeset and without rhgb and without quiet, I was able to watch the text and one of the very last lines was something about GDM, when, suddenly, the display flashed and entirely new text appeared, no longer in colour, but white on black, which appears to be a kernel crash output. The computer does boot with the nomodeset option in the linux16 line, but the display is wonky, constantly switching modes and leaving coloured rectangles on the screen, before switching anew. With this method of starting, KDE Plasma will not work, only a black screen and an active mouse pointer (a monkey :-) told me that there is already such a bug open), but Gnome does work (but I'm not sure that the stability, smoothness and speed of the display isn't affected). Anyway, the computer worked fine under f23 and I know there's nothing wrong with it. Something is preventing GDM from starting and I think it is causing the kernel to crash... and that's a bug!
We need to see the backtrace. Can you take a picture of it or get it from your logs?
OK, I will try to get it in a couple of minutes (boot computer, try to get the journal output onto a memory stick, transfer to this computer, etc). I also filed the same bug against xorg/Intel driver on freedesktop, since Adam said I should (but I had already filed here first). Give me a moment.
https://bugs.freedesktop.org/show_bug.cgi?id=95518
Created attachment 1160141 [details] journal lines for all boots today Sorry it took so long :-( I read the manual and googled, but I was unable to learn how to output selected lines from journalctl, so I -o the entire thing to a text file and deleted everything from yesterday.
I think I'm seeing the same thing in 4.5.4-300.fc24.i686+PAE as here, with 00:02.0 VGA compatible controller: Intel Corporation 82915G/GV/910GL Integrated Graphics Controller (rev 04) which looks like was previously reported in bug 1331593 and/or bug 1302071 and/or bug 1283348 and https://lists.fedoraproject.org/archives/list/test@lists.fedoraproject.org/message/FITMTLVW5OBE7ZT4S7YRCD3RLGTHN2KW/ . I can remote login to reboot, and collect dmesg. This is from its tail: [ 81.777762] ------------[ cut here ]------------ [ 81.777796] kernel BUG at include/linux/page-flags.h:272! [ 81.777826] invalid opcode: 0000 [#1] SMP [ 81.777854] Modules linked in: rpcsec_gss_krb5 nfsv4 dns_resolver nfs fscache fuse msdos fat snd_intel8x0 i915 snd_ac97_codec ac97_bus snd_seq snd_seq_device snd_pcm snd_timer video tg3 i2c_algo_bit drm_kms_helper sata_sil iTCO_wdt gpio_ich iTCO_vendor_support snd ppdev dcdbas drm soundcore i2c_i801 lpc_ich ptp pps_core parport_pc fjes parport acpi_cpufreq tpm_tis tpm nfsd auth_rpcgss nfs_acl lockd grace sunrpc serio_raw ata_generic pata_acpi [ 81.778201] CPU: 0 PID: 1114 Comm: Xorg Not tainted 4.5.4-300.fc24.i686+PAE #1 [ 81.778239] Hardware name: Dell Inc. OptiPlex GX280 /0G5611, BIOS A08 03/03/2006 [ 81.778289] task: f67ab180 ti: f3656000 task.ti: f3656000 [ 81.778319] EIP: 0060:[<f8597e42>] EFLAGS: 00013286 CPU: 0 [ 81.778377] EIP is at drm_pci_alloc+0xc2/0x1b0 [drm] [ 81.778405] EAX: 00000000 EBX: 00004000 ECX: f775d848 EDX: 00000007 [ 81.778438] ESI: f58ba490 EDI: c0419f90 EBP: f3657b90 ESP: f3657b6c [ 81.778472] DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 [ 81.778499] CR0: 80050033 CR2: b68bd000 CR3: 35a65340 CR4: 000006f0 [ 81.778499] Stack: [ 81.778499] 024040c0 00000000 f701a864 f88b9688 024040c0 80612224 00000000 f5a62c00 [ 81.778499] f5a62c78 f3657ba8 f882e8ef 00000100 f647c000 f37380c0 f3738000 f3657bdc [ 81.778499] f8875b4b f35e0000 f5b9ec00 f5f33800 f885d701 f3653400 f66d4800 00000000 [ 81.778499] Call Trace: [ 81.778499] [<f882e8ef>] i915_gem_object_attach_phys+0xef/0x190 [i915] [ 81.778499] [<f8875b4b>] intel_prepare_plane_fb+0x18b/0x300 [i915] [ 81.778499] [<f885d701>] ? intel_crtc_destroy+0x41/0x60 [i915] [ 81.778499] [<f85de385>] drm_atomic_helper_prepare_planes+0x45/0xb0 [drm_kms_helper] [ 81.778499] [<f886d1c0>] intel_atomic_commit+0x270/0x1770 [i915] [ 81.778499] [<f886fb26>] ? intel_atomic_check+0x886/0x10a0 [i915] [ 81.778499] [<f886f2a0>] ? intel_link_compute_m_n+0x50/0x50 [i915] [ 81.778499] [<f85ace9f>] ? drm_atomic_check_only+0x19f/0x670 [drm] [ 81.778499] [<f85ac664>] ? drm_atomic_get_crtc_state+0x54/0xc0 [drm] [ 81.778499] [<c058cb4a>] ? kmemdup+0x2a/0x40 [ 81.778499] [<f886cf50>] ? modeset_get_crtc_power_domains+0x140/0x140 [i915] [ 81.778499] [<f85ad3a4>] drm_atomic_commit+0x34/0x60 [drm] [ 81.778499] [<f85ded5c>] drm_atomic_helper_update_plane+0xbc/0x100 [drm_kms_helper] [ 81.778499] [<f85deca0>] ? drm_atomic_helper_wait_for_vblanks+0x210/0x210 [drm_kms_helper] [ 81.778499] [<f859ceeb>] __setplane_internal+0x1eb/0x230 [drm] [ 81.778499] [<f859d0ae>] drm_mode_cursor_common+0x17e/0x3a0 [drm] [ 81.778499] [<f85a13b0>] ? drm_mode_setcrtc+0x570/0x570 [drm] [ 81.778499] [<f85a1407>] drm_mode_cursor_ioctl+0x57/0x70 [drm] [ 81.778499] [<f85923f9>] drm_ioctl+0x149/0x4f0 [drm] [ 81.778499] [<f85a13b0>] ? drm_mode_setcrtc+0x570/0x570 [drm] [ 81.778499] [<c059bed7>] ? __do_fault+0x67/0x180 [ 81.778499] [<c059ff64>] ? handle_mm_fault+0x204/0x18a0 [ 81.778499] [<f85922b0>] ? drm_getmap+0xc0/0xc0 [drm] [ 81.778499] [<c05ec931>] do_vfs_ioctl+0x91/0x6f0 [ 81.778499] [<c05a703a>] ? do_mmap+0x24a/0x320 [ 81.778499] [<c05ecff8>] SyS_ioctl+0x68/0x80 [ 81.778499] [<c040388f>] do_fast_syscall_32+0x8f/0x140 [ 81.778499] [<c0b2f19b>] sysenter_past_esp+0x40/0x61 [ 81.778499] Code: 70 8d 82 00 00 00 40 c1 e8 0c 8d 0c 80 a1 84 e3 04 c1 8d 04 c8 8b 08 80 e5 40 74 7d 90 8d 74 26 00 ba 18 e5 5b f8 e8 2e 1a 00 c8 <0f> 0b 8d 74 26 00 8d 55 ec 8d 45 e4 e8 dd 1f e8 c7 84 c0 0f 84 [ 81.778499] EIP: [<f8597e42>] drm_pci_alloc+0xc2/0x1b0 [drm] SS:ESP 0068:f3657b6c [ 81.824939] ---[ end trace 0947e3cb36165af0 ]---
BIOS is 07/24/2009 F.16 uname -p reports i686 I can't type it all out by hand, but lspci|grep VGA reports: VGA compatible Intel mobile 915GM/GMS/910GML Express, rev 03
cat /proc/cpuinfo: processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 13 model name : Intel(R) Pentium(R) M processor 1.73GHz stepping : 8 microcode : 0x20 cpu MHz : 1067.000 cache size : 2048 KB physical id : 0 siblings : 1 core id : 0 cpu cores : 1 apicid : 0 initial apicid : 0 fdiv_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 2 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov clflush dts acpi mmx fxsr sse sse2 ss tm pbe nx bts est tm2 bugs : bogomips : 2129.40 clflush size : 64 cache_alignment : 64 address sizes : 32 bits physical, 32 bits virtual power management: It does not show lm or anything I have been told to indicate x86_64. man uname says -p option gives processor capability. This output indicates pae. I used to run the pae kernel, until I read that pae has something to do with memory paging, I believe, and 4GB are required as a minimum. My PC has only 1GB, hence I have gone back to the straight kernel.
As I had indicated, I had also followed Adam Williamson's suggestion (after I had already opened this bug) and reported the same bug on freedesktop as an xorg/Intel driver bug. This afternoon, Chris Wilson closed the bug with the resolution 'NOTOURBUG' and the comment "nomodeset disables the i915.ko module. Don't set it." That means that nomodeset should not be in the grub kernel boot line and, as we have seen above, removing it from the boot line causes a kernel crash.
*** This bug has been marked as a duplicate of bug 1335173 ***
(In reply to Peter Gückel from comment #8) > As I had indicated, I had also followed Adam Williamson's suggestion (after > I had already opened this bug) and reported the same bug on freedesktop as > an xorg/Intel driver bug. This afternoon, Chris Wilson closed the bug with > the resolution 'NOTOURBUG' and the comment "nomodeset disables the i915.ko > module. Don't set it." > > That means that nomodeset should not be in the grub kernel boot line and, as > we have seen above, removing it from the boot line causes a kernel crash. I think there must have been a misunderstanding in that particular report, because Chris' message implies that he thought you were reporting a problem *with 'nomodeset' enabled*, which is not what you are reporting AIUI.
(In reply to Adam Williamson from comment #10) > I think there must have been a misunderstanding in that particular report, > because Chris' message implies that he thought you were reporting a problem > *with 'nomodeset' enabled*, which is not what you are reporting AIUI. okayyy... :-) So, this is still a duplicate of bug 1335173, as Josh indicated in Comment 9, right?
yes, we can deal with it there. I was just making a comment on what I think happened with the upstream report.
(In reply to Adam Williamson from comment #12) > yes, we can deal with it there. super. glad it can be dealt with here. feels more like home :-) > I was just making a comment on what I think > happened with the upstream report. they did misunderstand and i misunderstood their misunderstanding, etc ;-) you put the parameter in to have no modesetting and don't put it in to have modesetting. it's hard to explain what you're doing or supposed to do :-))