System Environment:
--------------------------
Platform: Haswell
Kernel: drm-intel-next-queued(243ee1643b4c37ef00bb1f3cb3e67cc021690344)


Bug detailed description:
-----------------------------
While doing testdisplay -a with VGA, VGA flashed (i attach a video for more detail　URL:http://youtu.be/-sdoFvv9Skk ).And we can find the Call Trace in dmesg:

Call Trace:
------------[ cut here ]------------
[   37.741923] WARNING: CPU: 2 PID: 16 at drivers/gpu/drm/i915/intel_display.c:6864 hsw_enable_pc8+0x134/0x626 [i915]()
[   37.741924] SPLL enabled
[   37.741925] Modules linked in: dm_mod pcspkr snd_hda_codec_realtek snd_hda_codec_generic i2c_i801 snd_hda_codec_hdmi iTCO_wdt iTCO_vendor_support lpc_ich mfd_core snd_hda_intel snd_hda_codec snd_hwdep snd_pcm snd_timer snd soundcore battery ppdev parport_pc parport ac acpi_cpufreq i915 video button drm_kms_helper drm
[   37.741933] CPU: 2 PID: 16 Comm: kworker/2:0 Tainted: G        W    3.14.0_drm-intel-next-queued_44b65c_20140401+ #1234
[   37.741933] Hardware name: Intel Corporation Shark Bay Client platform/SthiPpvRsvd2, BIOS HSWLPTU1.86C.0134.R00.1310022130 10/02/2013
[   37.741935] Workqueue: pm pm_runtime_work
[   37.741935]  0000000000000000 0000000000000009 ffffffff81717233 ffff88043bdbfc88
[   37.741937]  ffffffff81035052 ffffc90012644024 ffffffffa009ed05 0000000000000000
[   37.741938]  ffff8804369b0000 ffff88009e0d9b38 ffff88009e0d9b40 ffff88043bb68098
[   37.741939] Call Trace:
[   37.741941]  [<ffffffff81717233>] ? dump_stack+0x41/0x51
[   37.741942]  [<ffffffff81035052>] ? warn_slowpath_common+0x73/0x8b
[   37.741947]  [<ffffffffa009ed05>] ? hsw_enable_pc8+0x134/0x626 [i915]
[   37.741948]  [<ffffffff81035102>] ? warn_slowpath_fmt+0x45/0x4a
[   37.741953]  [<ffffffffa008df9c>] ? gen6_read32+0x76/0x81 [i915]
[   37.741958]  [<ffffffffa009ed05>] ? hsw_enable_pc8+0x134/0x626 [i915]
[   37.741962]  [<ffffffffa00600f3>] ? i915_runtime_suspend+0x62/0xb1 [i915]
[   37.741964]  [<ffffffff812f8968>] ? pci_pm_runtime_suspend+0x60/0x119
[   37.741965]  [<ffffffff8138406e>] ? __rpm_callback+0x28/0x4c
[   37.741966]  [<ffffffff813840dd>] ? rpm_callback+0x4b/0x69
[   37.741967]  [<ffffffff81384689>] ? rpm_suspend+0x2a9/0x3f9
[   37.741969]  [<ffffffff8103c75e>] ? internal_add_timer+0xd/0x28
[   37.741970]  [<ffffffff8103cf3f>] ? mod_timer+0x125/0x13d
[   37.741971]  [<ffffffff81385329>] ? pm_runtime_work+0x65/0x7b
[   37.741973]  [<ffffffff81046660>] ? process_one_work+0x1bc/0x2ed
[   37.741974]  [<ffffffff81046bce>] ? worker_thread+0x1c7/0x2bc
[   37.741976]  [<ffffffff81046a07>] ? rescuer_thread+0x251/0x251
[   37.741977]  [<ffffffff8104b53e>] ? kthread+0xc5/0xcd
[   37.741979]  [<ffffffff8104b479>] ? kthread_freezable_should_stop+0x40/0x40
[   37.741981]  [<ffffffff817217fc>] ? ret_from_fork+0x7c/0xb0
[   37.741983]  [<ffffffff8104b479>] ? kthread_freezable_should_stop+0x40/0x40
[   37.741984] ---[ end trace 53debec2e22c7e47 ]---


Reproduce steps:
----------------------------
1. ./testdisplay -a  (using IGT )
2. xinit &  ;  chvt 1  (this method is more effective to reproduce)

first bad commit:
----------------------------
commit a8a8bd547e6323c56295e1c5a03e30e765d42325
Author: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date:   Fri Mar 7 20:08:05 2014 -0300

    drm/i915: make PC8 be part of runtime PM suspend/resume