Every now and then, the display momentarily blinks to black, and the following message appears in the log: [drm:intel_cpu_fifo_underrun_irq_handler] *ERROR* CPU pipe A FIFO underrun Reproduced on Linux 4.2 and 4.3-rc1, on a new Skylake laptop with quad-core i5 6200U. The laptop seems rather beefy so I have trouble believing this is a general RAM bandwidth issue. This is using just the internal laptop display, 1920x1080. Sometimes the problem happens a lot, and sometimes it takes a while. However, I find that I can't complete a grand prix in SuperTuxKart (about 20 mins gameplay) without this happening at least once (usually more times). I can't reproduce on intel-drm-nightly, however SuperTuxKart gameplay is not as smooth there, every few minutes the display hangs for approximately 1 second, with no messages in the log. 00:02.0 VGA compatible controller: Intel Corporation Device 1916 (rev 07) (prog-if 00 [VGA controller]) Subsystem: ASUSTeK Computer Inc. Device 1c3d Flags: bus master, fast devsel, latency 0, IRQ 123 Memory at de000000 (64-bit, non-prefetchable) [size=16M] Memory at c0000000 (64-bit, prefetchable) [size=256M] I/O ports at f000 [size=64] Expansion ROM at <unassigned> [disabled] Capabilities: [40] Vendor Specific Information: Len=0c <?> Capabilities: [70] Express Root Complex Integrated Endpoint, MSI 00 Capabilities: [ac] MSI: Enable+ Count=1/1 Maskable- 64bit- Capabilities: [d0] Power Management version 2 Capabilities: [100] #1b Capabilities: [200] Address Translation Service (ATS) Capabilities: [300] #13 Kernel driver in use: i915
Spoke to soon - just reproduced this on drm-intel-nightly.
Ping - anything I can do to help? We also see this on other Skylake laptops.
Please try Ville's FIFO underrun elimination series [1] on top of drm-intel-nightly, also available at [2], and report back. [1] http://mid.gmane.org/1446146763-31821-1-git-send-email-ville.syrjala@linux.intel.com [2] git://github.com/vsyrjala/linux.git pch_fifo_underrun_fix_4
I can reproduce the problem on that github branch, and also in Linux 4.4. However linus master from today seems fine. Yay!
(In reply to Daniel Drake from comment #4) > I can reproduce the problem on that github branch, and also in Linux 4.4. > However linus master from today seems fine. Yay! Great, thanks for testing.
Closing resolved+fixed based on verification done by Daniel.
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.