You are not logged in.
I'm having a rather strange issue where the cursor will appear around 50px higher than it actually is when the screen is rotated 90* clockwise.
The issue only occurs when the screen is clockwise - it works fine anti-clockwise and in landscape modes.
The issue wasn't present in Gnome 3.16, and happens whether the screen is primary, secondary, or the only screen attached.
Very confused, and not really sure how to go about diagnosing it or fixing it, so any help with be appreciated.
Last edited by James McMinn (2015-10-09 20:34:36)
Offline
Yes! I'm glad I'm not the only one. I wasn't having this issue yesterday (or at least I'm pretty sure I would've noticed) and it's definitely an annoying one.
Things I've verified:
I'm experiencing it on both i3 and gnome (I usually use i3)
Creating a new layout with arandr doesn't help.
Rebuilding and downgrading xf86-video-intel (updated yesterday) to the previously pacman version (pulled from /var/log/pacman.log) does not help.
Nothing obvious is showing up in journalctl.
Additional symptoms:
For me the cursor isn't higher, but to the left of where I'm seeing the UI react.
If I place the mouse on the boundary between my middle screen and the rotated screen (left of the main), I see two cursors.
If I overlap the two screens with arandr, I see the cursor in its correct position on my main screen and offset to the left on the rotated screen.
It does appear that rotating anti-clockwise is a workaround, but it's a bit of an unfortunate workaround as my monitor stands only allow me to rotate clockwise.
Last edited by andrewstuart (2015-10-09 20:06:05)
Offline
I just enabled the testing repository in /etc/pacman.conf which appears to have fixed the issue.
There were a few upgrades, but it seems most likely that the fix was a more recent version of xf86-video-intel (upgraded xf86-video-intel (1:2.99.917+472+gf0fd4d5-1 -> 1:2.99.917+478+gdf72bc5-1)).
Offline
Hmm. Was there anything else you did?
I tried updating xf86-video-intel first to the version you specified, with no luck still (in i3 or in gnome), and then I tried just updating everything to the versions in the testing repo, and still no luck there.
Offline
All I did was update everything from testing and then reboot.
Given that I had a vertical offset while you have a horizontal offset, I suppose it's possible that these are similar but separate issues?
Relevant section from pacman.log is below.
[2015-10-09 18:49] [PACMAN] starting full system upgrade # Testing wasn't enable for the first upgrade
[2015-10-09 18:49] [ALPM] transaction started
[2015-10-09 18:49] [ALPM] upgraded baobab (3.18.0-1 -> 3.18.1-1)
[2015-10-09 18:49] [ALPM] upgraded ghostscript (9.16-1 -> 9.18-1)
[2015-10-09 18:49] [ALPM] upgraded groff (1.22.3-3 -> 1.22.3-4)
[2015-10-09 18:49] [ALPM] upgraded xdg-utils (1.1.0.git20150323-1 -> 1.1.1-1)
[2015-10-09 18:49] [ALPM] transaction completed
[2015-10-09 18:50] [PACMAN] Running 'pacman -Syu' # I enabled testing here
[2015-10-09 18:50] [PACMAN] synchronizing package lists
[2015-10-09 18:50] [PACMAN] starting full system upgrade
[2015-10-09 18:50] [ALPM] transaction started
[2015-10-09 18:50] [ALPM] upgraded xz (5.2.1-1 -> 5.2.2-1)
[2015-10-09 18:50] [ALPM] upgraded libsystemd (226-3 -> 227-1)
[2015-10-09 18:50] [ALPM] upgraded libutil-linux (2.27-5 -> 2.27-6)
[2015-10-09 18:50] [ALPM] upgraded util-linux (2.27-5 -> 2.27-6)
[2015-10-09 18:50] [ALPM] upgraded systemd (226-3 -> 227-1)
[2015-10-09 18:50] [ALPM] upgraded device-mapper (2.02.125-1 -> 2.02.132-1)
[2015-10-09 18:50] [ALPM] upgraded cryptsetup (1.6.7-1 -> 1.6.8-1)
[2015-10-09 18:50] [ALPM] upgraded curl (7.44.0-1 -> 7.45.0-1)
[2015-10-09 18:50] [ALPM] upgraded x265 (1.7-2 -> 1.8-1)
[2015-10-09 18:50] [ALPM] upgraded ffmpeg (1:2.8-1 -> 1:2.8-2)
[2015-10-09 18:50] [ALPM] upgraded gpgme (1.6.0-1 -> 1.6.0-2)
[2015-10-09 18:50] [ALPM] upgraded gst-plugins-bad (1.6.0-1 -> 1.6.0-2)
[2015-10-09 18:50] [ALPM] upgraded iputils (20140519.fad11dc-1 -> 20150815.1b5d03a-2)
[2015-10-09 18:50] [ALPM] upgraded libcap-ng (0.7.4-1 -> 0.7.7-1)
[2015-10-09 18:50] [ALPM] upgraded linux (4.2.2-1 -> 4.2.3-1)
[2015-10-09 18:50] [ALPM-SCRIPTLET] >>> Updating module dependencies. Please wait ...
[2015-10-09 18:50] [ALPM-SCRIPTLET] >>> Generating initial ramdisk, using mkinitcpio. Please wait...
[2015-10-09 18:50] [ALPM-SCRIPTLET] ==> Building image from preset: /etc/mkinitcpio.d/linux.preset: 'default'
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> -k /boot/vmlinuz-linux -c /etc/mkinitcpio.conf -g /boot/initramfs-linux.img
[2015-10-09 18:50] [ALPM-SCRIPTLET] ==> Starting build: 4.2.3-1-ARCH
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> Running build hook: [base]
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> Running build hook: [udev]
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> Running build hook: [autodetect]
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> Running build hook: [modconf]
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> Running build hook: [block]
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> Running build hook: [filesystems]
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> Running build hook: [keyboard]
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> Running build hook: [fsck]
[2015-10-09 18:50] [ALPM-SCRIPTLET] ==> Generating module dependencies
[2015-10-09 18:50] [ALPM-SCRIPTLET] ==> Creating gzip-compressed initcpio image: /boot/initramfs-linux.img
[2015-10-09 18:50] [ALPM-SCRIPTLET] ==> Image generation successful
[2015-10-09 18:50] [ALPM-SCRIPTLET] ==> Building image from preset: /etc/mkinitcpio.d/linux.preset: 'fallback'
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> -k /boot/vmlinuz-linux -c /etc/mkinitcpio.conf -g /boot/initramfs-linux-fallback.img -S autodetect
[2015-10-09 18:50] [ALPM-SCRIPTLET] ==> Starting build: 4.2.3-1-ARCH
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> Running build hook: [base]
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> Running build hook: [udev]
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> Running build hook: [modconf]
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> Running build hook: [block]
[2015-10-09 18:50] [ALPM-SCRIPTLET] ==> WARNING: Possibly missing firmware for module: wd719x
[2015-10-09 18:50] [ALPM-SCRIPTLET] ==> WARNING: Possibly missing firmware for module: aic94xx
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> Running build hook: [filesystems]
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> Running build hook: [keyboard]
[2015-10-09 18:50] [ALPM-SCRIPTLET] -> Running build hook: [fsck]
[2015-10-09 18:50] [ALPM-SCRIPTLET] ==> Generating module dependencies
[2015-10-09 18:50] [ALPM-SCRIPTLET] ==> Creating gzip-compressed initcpio image: /boot/initramfs-linux-fallback.img
[2015-10-09 18:51] [ALPM-SCRIPTLET] ==> Image generation successful
[2015-10-09 18:51] [ALPM] upgraded thin-provisioning-tools (0.4.1-1 -> 0.5.3-1)
[2015-10-09 18:51] [ALPM] upgraded lvm2 (2.02.125-1 -> 2.02.132-1)
[2015-10-09 18:51] [ALPM] installed python-urllib3 (1.12-2)
[2015-10-09 18:51] [ALPM] installed python-chardet (2.3.0-2)
[2015-10-09 18:51] [ALPM] upgraded python-requests (2.7.0-2 -> 2.8.0-1)
[2015-10-09 18:51] [ALPM] upgraded systemd-sysvcompat (226-3 -> 227-1)
[2015-10-09 18:51] [ALPM] upgraded vlc (2.2.1-7 -> 2.2.1-8)
[2015-10-09 18:51] [ALPM] upgraded xf86-video-intel (1:2.99.917+472+gf0fd4d5-1 -> 1:2.99.917+478+gdf72bc5-1)
[2015-10-09 18:51] [ALPM] transaction completedOffline
A quick look at the recent commits to xf86-video-intel suggests that a fix might have been commited 9 hours ago: http://cgit.freedesktop.org/xorg/driver … 554d589160
It looks like there have been a bunch of commits recently related to cursor position, so it seems there's a number of little things broken at the moment: http://cgit.freedesktop.org/xorg/driver … intel/log/
Offline
It seems like you just had better luck with the `testing` repo than I did. Once I pulled the latest source with ABS and built it myself, everything is hunky dory again.
Thanks for pointing me in the right direction!
Last edited by andrewstuart (2015-10-10 02:08:43)
Offline
I am affected too, good to hear that a cure is in testing.
Usually I had Chrome on my second rotated display and the bug made me not able to reach the close button. Furthermore, every click was offset to the left.
Thanks for the links, I will wait for the driver to leave testing.
Offline
In case anybody else is interested in the steps I had to take, since they're a bit more involved:
(Update: this is apparently not the preferred way. Use makepkg so that pacman can manage the install, see below.)
git clone git://anongit.freedesktop.org/xorg/driver/xf86-video-intel
cd xf86-video-intel
./autogen.sh
make && sudo make installLast edited by andrewstuart (2015-10-10 19:15:17)
Offline
Don't do that. Installing things outside of pacman is a Bad Idea™. Use the xf86-video-intel-git package from the AUR.
Sakura:-
Mobo: MSI MAG X570S TORPEDO MAX // Processor: AMD Ryzen 9 5950X @4.9GHz // GFX: AMD Radeon RX 5700 XT // RAM: 32GB (4x 8GB) Corsair DDR4 (@ 3000MHz) // Storage: 1x 3TB HDD, 6x 1TB SSD, 2x 120GB SSD, 1x 275GB M2 SSD
Making lemonade from lemons since 2015.
Offline
Alright, if you want to use AUR and make sure Pacman knows what's going on (even though the bugger got me into this mess :-P ) then here are some alternate steps:
yaourt -G xf86-video-intel-git
#if yaourt not installed:
# curl https://aur.archlinux.org/cgit/aur.git/snapshot/xf86-video-intel-git.tar.gz | tar -xz
cd xf86-video-intel-git
# I like to add -j<numCores> after all the instances of `make` in the PKGBUILD now, but not necessary.
makepkg -csi # clean, get dependencies, and install when builtOffline
xf86-video-intel 1:2.99.917+478+gdf72bc5-2 moved into extra.
... and now is fixed for everyone, no need to enable testing or install from AUR.
Offline