Comment # 5 on bug 966179 from Takashi Iwai

(In reply to Jan Engelhardt from comment #4)
> "framebuffer console remains black"... it is vague, but I have experienced
> something similar - maybe related. I observe that there is a regression
> between 4.3.x and 4.4 (both with a *13.2* not Leap not TW userspace),
> whereby, after resuming from suspend-to-ram for the Nth time (or after N
> days - this kind of makes it hard and long-winded to reproduce so far),
> certain small rectangular areas of the X framebuffer are black, and
> switching to text console produces a no-text console, but getty is
> definitely running and one can log in blind. The X fb is repaired when
> switching back to tty7, but textcon stays dark forever.
> 
> 00:02.0 VGA compatible controller: Intel Corporation Haswell-ULT Integrated
> Graphics Controller (rev 0b)
> 00:02.0 0300: 8086:0a16 (rev 0b)

The S3 resume problem might be a different issue.  I myself also found this
regression, already reported to upstream.  Through the discussion, the patch
was suggested, which had landed in 4.5-rc with the commit:

0c82312f3f15538f4e6ceda2a82caee8fbac4501
    drm/i915: Pin the ifbdev for the info->system_base GGTT mmapping

I haven't seen the issue after applying it.  Alas, the S3 resume problem isn't
reliably reproducible, so I can't say exactly whether it is cured by that.