Sample Header Ad - 728x90

Chrome on Gnome randomly fails to go full screen

1 vote
1 answer
1161 views
I have CentOS 7 running the 4.7.1 kernel (elrepo-ml) on an Intel NUC with Intel HD Graphics i915. I'm trying to run Chrome in full screen mode google-chrome --full-screen --kiosk http:192.168.2.20 (just a local server with the default Apache2 page) and it works most of the time, but randomly (every second to fifth time) it doesn't fully display the page and looks like this: screencap You can see that chrome "thinks" it's full screen because of the way that the page is rendered. One interesting behavior is that if I Alt+Tab to another window (hidden in the screen shot), then the page will render full screen under the window just brought to the front. There is nothing interesting in X.org logs – the same lines are repeated in both successful and unsuccessful full-screen attempts. (The same lines are logged in /var/log/gdm/ when gdm is in debug mode).
[352467.250] GetModeLine - scrn: 0 clock: 148500
[352467.250] GetModeLine - hdsp: 1920 hbeg: 2008 hend: 2052 httl: 2200
[352467.250]               vdsp: 1080 vbeg: 1084 vend: 1089 vttl: 1125 flags: 5
I've tried to force a refresh (which may be what Alt+Tab) programmatically with xrefresh, but that doesn't make any difference. What could be causing this behavior and what can I do to consistently get a full screen image? Edited to add more system details: - Hardware: Intel NUC NUC5CPYH Processor - Intel Celeron N3050 with Intel HD Graphics Driver:
$ modinfo i915 | grep version
srcversion:     4F4A9108A6D94B8919E3154
vermagic:       4.7.1-1.el7.elrepo.x86_64 SMP mod_unload modversions
Hardware:
$ dmesg | egrep "i915|[Gg]raphic"
[    0.000000] Reserving Intel graphics stolen memory at 0x7cf00000-0x7eefffff
[    2.356158] [drm] Memory usable by graphics device = 2048M
[    2.498705] [drm] Initialized i915 1.6.0 20160425 for 0000:00:02.0 on minor 0
[    3.284294] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device
[    6.301736] snd_hda_intel 0000:00:1b.0: bound 0000:00:02.0 (ops i915_audio_component_bind_ops [i915])
Asked by MikeV (1408 rep)
Aug 24, 2016, 01:34 AM
Last activity: Aug 24, 2020, 07:53 PM