Created attachment 6186 switching between workspace 1 and 2 When maximizing/switching to a browser or changing a workspace, browser address and tabs bars (sometimes also 1/4 upper part of a screen) are not refreshed until clicking refresh website button or going with mouse pointer over all tabs or activating some other app on second monitor. See attached picture (red color - Chrome on workspace 1, blue color - Opera on workspace 2). Checked only with Opera and Chrome.
That really looks weird. - Do you have compositing enabled? - do you use xfwm4 compositor or some other separate compositor (compton, xcompmgr, etc.) - What if you disable compositing? - What Xorg driver/version do you use?
1. Yup, compositing is enabled in Window Manager Tweaks. 2. xfwm4 (I also have compiz but disabled). 3. No issue anymore - after disabling compositing 1st workspace switch is erroneous, next switches work as expected. 4. 1.15.1
(In reply to Kubsztal from comment #2) > 1. Yup, compositing is enabled in Window Manager Tweaks. > 2. xfwm4 (I also have compiz but disabled). I cannot reproduce that. I tried with google-chrome and Mozilla (I cannot install Opera because all they seem to offer is a deb package and I am not using a Debian based distribution) - And to be honest, I am a little wary with closed source proprietary software/drivers. > 3. No issue anymore - after disabling compositing 1st workspace switch is > erroneous, next switches work as expected. That does not sound right, you still get an error at first workspace switch even after disabling compositing? Can you reproduce without Opera? > 4. 1.15.1 You didn't mention what driver.
(In reply to Olivier Fourdan from comment #3) > I cannot reproduce that. I tried with google-chrome and Mozilla. I've checked also with Firefox - works fine. It seems there is a problem only with Chrome and Opera. But I catched another side effect - after disabling compositing, dropbox icon vanishes from system tray. Physically it's still there as RMB click opens dropbox menu, just icon itself is not visible. > That does not sound right, you still get an error at first workspace switch > even after disabling compositing? The same in other direction, after enabling compositing 1st workspace switch is OK. So it seems that toggling compositing doesn't fully refresh Opera/Chrome off-screen buffers. > Can you reproduce without Opera? Yup, still NOK with Chrome. > You didn't mention what driver. Sorry, forgot about that. [ 3.158] (II) LoadModule: "intel" [ 3.158] (II) Loading /usr/lib/xorg/modules/drivers/intel_drv.so [ 3.161] (II) Module intel: vendor="X.Org Foundation" [ 3.161] compiled for 1.15.1, module version = 2.99.910 [ 3.161] Module class: X.Org Video Driver [ 3.161] ABI class: X.Org Video Driver, version 15.0
Created attachment 6208 filenames say everything
Created attachment 6209 comp_enabled_after_minimising-maximising-1 This was intended to be a separate bug report: I don't think that the issue has anything to do with running Opera, I never installed it and yet I have problems. It also probably has little to do with Chromium as I was running it finely on Compiz (before switching to xfwm4). With compostor enabled, each time I minimise and then maximise Chromium window or move to secondary workspace and than back to primary, entire Chromium window contents or some elements fail to draw until I cause some change in it by scrolling or openig a new tab. I am using xfce with Linux Mint 17.1. Driver is Intel You should be able to reproduce behavior by repeating steps mentioned above. There are screenshots of what is happening with compostor enabled and you will see that everything worked well after I disabled it. $ dmesg | grep "Intel" does not return anything about Intel graphic driver. This is what I get: $ dmesg | grep "Intel" [ 0.000000] Intel GenuineIntel [ 0.000000] Reserving Intel graphics stolen memory at 0xaba00000-0xaf9fffff [ 0.068701] smpboot: CPU0: Intel(R) Pentium(R) CPU B950 @ 2.10GHz (fam: 06, model: 2a, stepping: 07) [ 0.068737] Performance Events: PEBS fmt1+, 16-deep LBR, SandyBridge events, full-width counters, Intel PMU driver. [ 0.778250] Intel P-state driver initializing. [ 0.778294] Intel pstate controlling: cpu 0 [ 0.778317] Intel pstate controlling: cpu 1 [ 9.172844] Intel(R) Wireless WiFi driver for Linux, in-tree: [ 9.172848] Copyright(c) 2003- 2014 Intel Corporation [ 9.818641] iwlwifi 0000:02:00.0: Detected Intel(R) Centrino(R) Wireless-N 130 BGN, REV=0xB0 [ 10.058744] input: HDA Intel PCH Mic as /devices/pci0000:00/0000:00:1b.0/sound/card0/input10 [ 10.058818] input: HDA Intel PCH Headphone as /devices/pci0000:00/0000:00:1b.0/sound/card0/input11 [ 10.058884] input: HDA Intel PCH HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:1b.0/sound/card0/input12 [ 804.051031] Intel pstate controlling: cpu 1 [ 8560.530330] Intel pstate controlling: cpu 1 [14096.204624] Intel pstate controlling: cpu 1 [15455.021389] Intel pstate controlling: cpu 1 Maybe this has some relevance: $ dpkg -s xserver-xorg-video-intel Package: xserver-xorg-video-intel Status: install ok installed Priority: optional Section: x11 Installed-Size: 2774 Maintainer: Ubuntu Developers <ubuntu-devel-discuss@lists.ubuntu.com> Architecture: amd64 Version: 2:2.99.910-0ubuntu1.4 Provides: xorg-driver-video Depends: libc6 (>= 2.17), libdrm-intel1 (>= 2.4.38), libdrm2 (>= 2.4.25), libpciaccess0 (>= 0.8.0+git20071002), libpixman-1-0 (>= 0.30.0), libudev1 (>= 183), libx11-6, libx11-xcb1, libxcb-dri2-0, libxcb-util0 (>= 0.3.8), libxcb1, libxv1, libxvmc1, xorg-video-abi-15, xserver-xorg-core (>= 2:1.14.99.902) Description: X.Org X server -- Intel i8xx, i9xx display driver This package provides the driver for the Intel i8xx and i9xx family of chipsets, including i810, i815, i830, i845, i855, i865, i915, i945 and i965 series chips. . This package also provides XvMC (XVideo Motion Compensation) drivers for i810/i815 and i9xx and newer chipsets. . This package is built from the X.org xf86-video-intel driver module. Homepage: http://www.x.org/ Original-Maintainer: Debian X Strike Force <debian-x@lists.debian.org>
Created attachment 6210 comp_enabled_after_minimising-maximising-2
Created attachment 6211 comp_enabled_after_switching-workspace
Created attachment 6212 comp_disabled_after_minimising-maximising
Created attachment 6213 comp_disabled_after_switching-workspace
Created attachment 6214 before_everything forgive my silly minimiSe maximiSe mistake :-)
And now, after disabling compostor, installing and running Compton everything works just fine.
I cannot reproduce yet I use intel drivers on most of my laptops. But I use more recent versions of Xorg/intel driver... Any way you can try to reproduce using an up-to-date software stack?
(In reply to Olivier Fourdan from comment #13) > Any way you can try > to reproduce using an up-to-date software stack? After updating to the intel driver version 2.99.917 from xorg-edgers ppa, dropbox icon in system tray became visible with disabled compositing. After enabling compositing problem with screen drawing while switching between opera/chrome remains.
(In reply to Kubsztal from comment #14) What about updating Xorg?
(In reply to Olivier Fourdan from comment #15) > What about updating Xorg? Found xorg-server-lts-vivid - 2:1.17.1-0ubuntu3~trusty1 here: https://launchpad.net/~canonical-x/+archive/ubuntu/x-staging/+packages but still not available due to build errors. Any other ppa? Don't have a time for compiling and installing from scratch. Xorg-edgers has only 1.15.1 for trusty.
(In reply to Kubsztal from comment #16) > Found xorg-server-lts-vivid - 2:1.17.1-0ubuntu3~trusty1 here: > https://launchpad.net/~canonical-x/+archive/ubuntu/x-staging/+packages > but still not available due to build errors. Any other ppa? Don't have a > time for compiling and installing from scratch. Xorg-edgers has only 1.15.1 > for trusty. Sorry, I don use Debian/Ubuntu so I can't help here. But you don't necessarily have to update your laptop, I was thinking more about trying a live-cd with a more recent software stack.
Something like what was done for bug 11473 for example
"" The problem seems to be in the graphics driver. For those having Intel® HD Graphics 4000, check whether this xorg.conf solves the problem: $ cat /etc/X11/xorg.conf.d/20-intel.conf Section "Device" Identifier "Intel Graphics" Driver "intel" Option "AccelMethod" "UXA" EndSection "" I have Intel HD Graphics 2000, and this worked on my Mint 17.1. You can also disable hardware acceleration, but I didn't like that 'solution'. Comment #25: https://bugs.launchpad.net/ubuntu/+source/chromium-browser/+bug/1309801
(In reply to Olivier Fourdan from comment #18) > Something like what was done for bug 11473 for example I also tried to reproduce this bug on Xubuntu 15.04 and OpenSUSE 13.2 (more recent graphic drivers) and I couldn't. I guess that invalidates this bug.
Driver issue Then.
(In reply to Goran Janković from comment #19) > $ cat /etc/X11/xorg.conf.d/20-intel.conf I don't have xorg.conf.d in /etc/X11/ folder on Mint 17.1...
(In reply to Kubsztal from comment #22) > (In reply to Goran Janković from comment #19) > > $ cat /etc/X11/xorg.conf.d/20-intel.conf > > I don't have xorg.conf.d in /etc/X11/ folder on Mint 17.1... /xorg.conf.d/20-intel.conf Create folder and file, I had to create them too.
(In reply to Goran Janković from comment #23) > /xorg.conf.d/20-intel.conf > > Create folder and file, I had to create them too. Works great, thanks Goran! :)