Created attachment 929300 [details]
File: dmesg

Looking into this a bit more, it actually looks like the bug happens multiple times per second constantly during operation regardless of whether the machine has an external display plugged in.  However, when the display is plugged in, the warning (or perhaps the ABRT check) causes display lag, which does not happen without the display.

I'm hitting this also, on F21 alpha.  It's quite annoying, even with abrtd stopped, it results in everythng in X 'hanging' for a second, every 4-6 seconds (regular period.  It only happens when my T540p is docked and I connect a mini-DP adapter (external monitor on the other end).   The dock has a DVI connection to a second monitor, all work fine, dispite the kernel warning spam.  My backtrace looks similar also:

kernel: ------------[ cut here ]------------
kernel: WARNING: CPU: 5 PID: 407 at drivers/gpu/drm/i915/intel_dp.c:1306 edp_panel_vdd_off.part.12+0x62/0x70 [i915]()
kernel: eDP VDD not forced on
kernel: Modules linked in:
kernel:  fuse ccm xt_CHECKSUM ipt_MASQUERADE tun nf_conntrack_netbios_ns nf_conntrack_broadcast ip6t_rpfilter ip6t_REJECT xt_conntrack ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6table_nat nf_conntrack_ipv6 nf_defrag_ipv6 nf_nat_ipv6 ip6table_mangle ip6table_security ip6table_raw ip6table_filter ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security iptable_raw bnep arc4 vfat fat iTCO_wdt iTCO_vendor_support uvcvideo videobuf2_vmalloc iwlmvm videobuf2_memops videobuf2_core snd_hda_codec_realtek snd_hda_codec_generic mac80211 snd_hda_codec_hdmi x86_pkg_temp_thermal coretemp v4l2_common snd_hda_intel kvm_intel snd_hda_controller videodev kvm snd_hda_codec snd_hwdep snd_seq snd_seq_device btusb snd_pcm bluetooth
kernel:  iwlwifi media joydev thinkpad_acpi cfg80211 rtsx_pci_ms snd_timer serio_raw snd memstick tpm_tis ie31200_edac mei_me i2c_i801 rfkill edac_core mei lpc_ich soundcore shpchp tpm wmi dm_crypt i915 rtsx_pci_sdmmc mmc_core i2c_algo_bit drm_kms_helper crct10dif_pclmul crc32_pclmul crc32c_intel e1000e drm ghash_clmulni_intel rtsx_pci mfd_core ptp pps_core video
kernel: CPU: 5 PID: 407 Comm: kworker/5:2 Tainted: G        W      3.17.1-303.fc21.x86_64 #1kernel: Hardware name: LENOVO 20BFS0EC00/20BFS0EC00, BIOS GMET62WW (2.10 ) 03/19/2014
kernel: Workqueue: events i915_hotplug_work_func [i915]
kernel:  0000000000000000 00000000910f5887 ffff880403633c90 ffffffff8173dbb1
kernel:  ffff880403633cd8 ffff880403633cc8 ffffffff81096e8d ffff8804042a90e0
kernel:  0000000000000000 ffff8804042b0000 0000000000000011 0000000000000003
kernel: Call Trace:
kernel:  [<ffffffff8173dbb1>] dump_stack+0x45/0x56
kernel:  [<ffffffff81096e8d>] warn_slowpath_common+0x7d/0xa0
kernel:  [<ffffffff81096f0c>] warn_slowpath_fmt+0x5c/0x80
kernel:  [<ffffffffa011ae5b>] ? drm_dp_dpcd_read+0x1b/0x20 [drm_kms_helper]
kernel:  [<ffffffffa01c9ba2>] edp_panel_vdd_off.part.12+0x62/0x70 [i915]
kernel:  [<ffffffffa01ca2d0>] intel_dp_probe_mst.part.15+0xb0/0xf0 [i915]
kernel:  [<ffffffffa01cc889>] intel_dp_detect+0xa9/0x230 [i915]
kernel:  [<ffffffffa01851c8>] i915_hotplug_work_func+0x1f8/0x310 [i915]
kernel:  [<ffffffff810af66d>] process_one_work+0x14d/0x400
kernel:  [<ffffffff810afffb>] worker_thread+0x6b/0x4a0
kernel:  [<ffffffff810aff90>] ? rescuer_thread+0x2a0/0x2a0
kernel:  [<ffffffff810b50da>] kthread+0xea/0x100
kernel:  [<ffffffff810b4ff0>] ? kthread_create_on_node+0x1a0/0x1a0
kernel:  [<ffffffff81744c7c>] ret_from_fork+0x7c/0xb0
kernel:  [<ffffffff810b4ff0>] ? kthread_create_on_node+0x1a0/0x1a0
kernel: ---[ end trace a228f60a8da37091 ]---

Note: It seems I can make the problem stop by opening the laptop lid (it was closed before), then pressing the 'projector mode' button (same as F7).

I noticed after pressing it until I got a projector-mode where all 3 screens are active (instead of just two external monitors).  I have no idea if there's a correlation here, or if I just happened to tickle something at exactly the right time for the kernel to regain some sensibilities :)

I can see the very same problem on Fedora 21 with 3.17.2-300.fc21.x86_64 kernel.

My HW is slightly different:
WARNING: CPU: 1 PID: 3686 at drivers/gpu/drm/i915/intel_dp.c:1306 edp_panel_vdd_off.part.12+0x62/0x70 [i915]()
eDP VDD not forced on
...
CPU: 1 PID: 3686 Comm: kworker/1:3 Tainted: G        W      3.17.2-300.fc21.x86_64 #1
Hardware name: LENOVO 20BFS0W505/20BFS0W505, BIOS GMET66WW (2.14 ) 07/01/2014

In my case is seems that the problem is related to screen blanking in console. The stacktrace pops up every ~ 4 seconds when screen is 'blanked'/backlight is turned off.

Opening/closing lid doesn't change anything (I have disabled sleep mode in logind.conf). Stack trace does not appear anymore after I press any button on keyrboard so backlight is turned on and I can see console again.

Build http://koji.fedoraproject.org/koji/taskinfo?taskID=8094663 fixes the problem for me.

(In reply to Petr Spacek from comment #7)
> Build http://koji.fedoraproject.org/koji/taskinfo?taskID=8094663 fixes the
> problem for me.

I'm having the same problem on F20 - lenovo thinkpad -> docking station -> external dvi monitor

Is there an F20 build that fixes this problem?

It may sound kind of hokey but I've managed to "fix" this problem by shutting-down, re-docking, and powering up my laptop.  Not sure if this implicates the dock/laptop connector are involved somehow or it's just blink-luck WRT to booting w/ the planets aligned.

(In reply to Chris Evich from comment #9)
> It may sound kind of hokey but I've managed to "fix" this problem by
> shutting-down, re-docking, and powering up my laptop.  Not sure if this
> implicates the dock/laptop connector are involved somehow or it's just
> blink-luck WRT to booting w/ the planets aligned.

so
1) shutdown laptop
2) undock
3) dock
4) startup laptop
?

(In reply to Rich Megginson from comment #10)
> (In reply to Chris Evich from comment #9)
> > It may sound kind of hokey but I've managed to "fix" this problem by
> > shutting-down, re-docking, and powering up my laptop.  Not sure if this
> > implicates the dock/laptop connector are involved somehow or it's just
> > blink-luck WRT to booting w/ the planets aligned.
> 
> so
> 1) shutdown laptop
> 2) undock
> 3) dock
> 4) startup laptop
> ?

Yes.  At least once I remember having to do the entire sequence two or three times before it started behaving.  But again, it may be a red-herring if there's some startup-timing related thing in-play, so YMMV.  

Recently, I've been much more "gentle, yet deliberate" when docking, and haven't had this problem re-appear.  Though it could also be a red-herring, since I'm running latest F21 updates-testing.

Such is the challenge w/ reporting from a "in-use" vs "lab" laptop :S

(In reply to Chris Evich from comment #11)
> (In reply to Rich Megginson from comment #10)
> > (In reply to Chris Evich from comment #9)
> > > It may sound kind of hokey but I've managed to "fix" this problem by
> > > shutting-down, re-docking, and powering up my laptop.  Not sure if this
> > > implicates the dock/laptop connector are involved somehow or it's just
> > > blink-luck WRT to booting w/ the planets aligned.
> > 
> > so
> > 1) shutdown laptop
> > 2) undock
> > 3) dock
> > 4) startup laptop
> > ?
> 
> Yes.  At least once I remember having to do the entire sequence two or three
> times before it started behaving.  But again, it may be a red-herring if
> there's some startup-timing related thing in-play, so YMMV.  
> 
> Recently, I've been much more "gentle, yet deliberate" when docking, and
> haven't had this problem re-appear.  Though it could also be a red-herring,
> since I'm running latest F21 updates-testing.
> 
> Such is the challenge w/ reporting from a "in-use" vs "lab" laptop :S

Ok.  I'm hoping that since Petr reported that a kernel update on F21 fixed the problem, a new kernel is forthcoming on F20 that will also fix the problem.

(In reply to Chris Evich from comment #4)
> Note: It seems I can make the problem stop by opening the laptop lid (it was
> closed before), then pressing the 'projector mode' button (same as F7).
> 

I can confirm for 3.17.2-200.fc20.x86_64 that opening the lid causes the crashed to go away, and closing it makes them re-appear, so I will just sit with my lid open (since the display is set to off anyway)

I'm using 2 external monitors, one on DVI and the other on VGA.

*** Bug 1163930 has been marked as a duplicate of this bug. ***

On my machine the problem does not appear anymore with stock package kernel-3.17.3-300.fc21.x86_64.

Oh, the package is in stable now!