http://bugzilla.opensuse.org/show_bug.cgi?id=1139889 http://bugzilla.opensuse.org/show_bug.cgi?id=1139889#c3 Felix Miata changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |mrmazda@earthlink.net --- Comment #3 from Felix Miata --- I saw this less than a week ago and ignored it. I thought it was host kt880 using NV gt218 with modesetting DDX on 32bit 20190623 @1920x1200 on HDMI, but it's not doing it today, nor is mcp61 using NV C61/GeForce 6150SE on 64bit 20190621 @1680x1050 on VGA. Host kt880 has onboard i915G, so it's possible I saw pointer corruption before installing the gt218 and that's why I ignored it. It could have been on KDE3, TDE or IceWM. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1139889 http://bugzilla.opensuse.org/show_bug.cgi?id=1139889#c15 Felix Miata changed: What |Removed |Added ---------------------------------------------------------------------------- Component|X.Org |Installation Assignee|xorg-maintainer-bugs@forge. |yast2-maintainers@suse.de |provo.novell.com | QA Contact|xorg-maintainer-bugs@forge. |jsrain@suse.com |provo.novell.com | --- Comment #15 from Felix Miata --- Suse User, nomodeset is primarily a troubleshooting/safe mode parameter that should not be required or retained in the installed system's bootloader configuration. If it remains in your /etc/default/grub and /boot/grub2/grub.cfg it needs to be removed, so that you can enjoy a normally operating system. If you have any doubts about nomodeset, read https://en.opensuse.org/SDB:Nomodeset:_Work_Around_Graphic_Upgrade_&_Installation_Obstacles and bug 1110041 -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1139889 http://bugzilla.opensuse.org/show_bug.cgi?id=1139889#c18 Stefan Dirsch changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |REOPENED Component|Installation |X.Org Resolution|WONTFIX |--- Assignee|yast2-maintainers@suse.de |xorg-maintainer-bugs@forge. | |provo.novell.com QA Contact|jsrain@suse.com |xorg-maintainer-bugs@forge. | |provo.novell.com --- Comment #18 from Stefan Dirsch --- I suggest to add nomodeset permanently to /etc/default/grub to line GRUB_CMDLINE_LINUX_DEFAULT. Then run grub2-mkconfig -o /boot/grub2/grub.cfg as also mentioned in this file, so it no longer needs to be specified any longer during boot time. Then from time to time remove it in grub boot menu during boot in order to see whether things have improved. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1139889 http://bugzilla.opensuse.org/show_bug.cgi?id=1139889#c19 Felix Miata changed: What |Removed |Added ---------------------------------------------------------------------------- Summary|915GM: Mouse pointer |915G[M]: Mouse pointer |corrupt in graphics |corrupt in graphics |installation |installation --- Comment #19 from Felix Miata --- (In reply to Stefan Dirsch from comment #18)

I suggest to add

nomodeset

permanently to /etc/default/grub to line GRUB_CMDLINE_LINUX_DEFAULT.

Why? The corruption is only during installation! I just reproduced using TW20190705 NET on 915G (not M). The installed TW20190708 system has no such problem running without nomodeset with xf86-video-intel, as appears implied by comment 16, and as noted in comment 14. # inxi -GxxSa System: Host: gx280 Kernel: 5.1.16-1-default i686 bits: 32 compiler: gcc v: 9.1.1 parameters: root=LABEL=17ostw ipv6.disable=1 net.ifnames=0 noresume mitigations=auto consoleblank=0 vga=791 video=1440x900@60 5 Desktop: KDE 3.5.10 tk: Qt 3.3.8c wm: kwin dm: N/A Distro: openSUSE Tumbleweed 20190708 Graphics: Device-1: Intel 82915G/GV/910GL Integrated Graphics vendor: Dell driver: i915 v: kernel bus ID: 00:02.0 chip ID: 8086:2582 Display: x11 server: X.Org 1.20.5 driver: intel unloaded: fbdev,modesetting,vesa tty: N/A OpenGL: renderer: Mesa DRI Intel 915G x86/MMX/SSE2 v: 1.4 Mesa 19.1.1 direct render: Yes # xrandr | head -n5 Screen 0: minimum 8 x 8, current 1440 x 900, maximum 32767 x 32767 VGA1 connected primary 1440x900+0+0 (normal left inverted right x axis y axis) 470mm x 300mm 1680x1050 59.93 + 1280x1024 75.02 60.02 1440x900 74.98 59.89* The installer GUI crashed when I hit the PrtScr key to try to capture the corrupt pointer that looked just like the comment 0 attachment. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1139889 http://bugzilla.opensuse.org/show_bug.cgi?id=1139889#c21 --- Comment #21 from Felix Miata --- Created attachment 810597 --> http://bugzilla.opensuse.org/attachment.cgi?id=810597&action=edit y2logs from initialized installation with corrupt mouse pointer I began a TW20190708 NET installation, clicked OK on first screen, let next screen paint complete with corrupt mouse pointer, then saved logs. Maybe there's a clue to be had in them. (In reply to Stefan Dirsch from comment #20)

We may still be able to figure this out by experimenting with

/etc/X11/xorg.conf.d/50-device.conf

Section "Device" Identifier "Default Device" Driver "intel" #Driver "modeset" #Driver "fbdev" EndSection

and see if can get the corrupted cursor also after installation.

I tried. No such luck: # kmsINTELok # grep -i 'l command lin' /var/log/Xorg.0.log [ 144.883] Kernel command line: root=/dev/sda17 noresume mitigations=auto consoleblank=0 3 # grep river /etc/X11/xorg.conf.d/50-device.conf Driver "intel" # Driver "modeset" # Driver "fbdev" # xrandr | head -n4 Screen 0: minimum 8 x 8, current 1680 x 1050, maximum 32767 x 32767 VGA1 connected primary 1680x1050+0+0 (normal left inverted right x axis y axis) 470mm x 300mm 1680x1050 59.93*+ 1280x1024 75.02 60.02 EOF # kmsFBDEVok # grep -i 'l command lin' /var/log/Xorg.0.log [ 144.883] Kernel command line: root=/dev/sda17 noresume mitigations=auto consoleblank=0 3 # grep river /etc/X11/xorg.conf.d/50-device.conf # Driver "intel" # Driver "modeset" Driver "fbdev" # xrandr | head -n4 # xrandr xrandr: Failed to get size of gamma for output default Screen 0: minimum 1680 x 1050, current 1680 x 1050, maximum 1680 x 1050 default connected 1680x1050+0+0 0mm x 0mm 1680x1050 0.00* # grep V\(0\) /var/log/Xorg.0.log | head -n5 [ 837.117] (**) FBDEV(0): claimed PCI slot 0@0:2:0 [ 837.117] (II) FBDEV(0): using default device [ 837.117] (II) FBDEV(0): Creating default Display subsection in Screen section [ 837.117] (==) FBDEV(0): Depth 24, (==) framebuffer bpp 32 [ 837.117] (==) FBDEV(0): RGB weight 888 EOF # kmsMODESETTING0glamor # grep -i 'l command lin' /var/log/Xorg.0.log [ 144.883] Kernel command line: root=/dev/sda17 noresume mitigations=auto consoleblank=0 3 # grep river /etc/X11/xorg.conf.d/50-device.conf # Driver "intel" Driver "modeset" # Driver "fbdev" [ 646.482] (EE) Failed to load module "modeset" (module does not exist, 0) [ 646.482] (EE) No drivers available. [ 1522.609] (EE) modeset(0): Failed to initialize glamor at ScreenInit() time. [ 1522.609] (EE) Fatal server error: [ 1522.610] (EE) AddScreen/ScreenInit failed for driver 0 EOF # kmsMODESET0exist # grep -i 'l command lin' /var/log/Xorg.0.log [ 144.883] Kernel command line: root=/dev/sda17 noresume mitigations=auto consoleblank=0 3 # grep river /etc/X11/xorg.conf.d/50-device.conf # Driver "intel" Driver "modeset" # Driver "fbdev" # Driver "modesetting" [ 1998.979] (WW) Warning, couldn't open module modeset [ 1998.979] (EE) Failed to load module "modeset" (module does not exist, 0) [ 1998.979] (EE) No drivers available. EOF # nomodesetINTEL0devices [ 311.026] Kernel command line: root=/dev/sda17 noresume mitigations=auto nomodeset 3 # grep river /etc/X11/xorg.conf.d/50-device.conf Driver "intel" # Driver "modeset" # Driver "fbdev" [ 313.892] (EE) No devices detected. EOF # nomodesetOKvesa # grep \(0\) /var/log/Xorg.0.log | grep 'irtual size is' [ 647.533] (II) VESA(0): Virtual size is 1600x1200 (pitch 1600) # inxi -GxxSa System: Host: gx280 Kernel: 5.1.16-1-default i686 bits: 32 compiler: gcc v: 9.1.1 parameters: root=LABEL=17ostw ipv6.disable=1 net.ifnames=0 noresume mitigations=auto consoleblank=0 nomodeset 3 Desktop: KDE 3.5.10 tk: Qt 3.3.8c wm: kwin dm: startx Distro: openSUSE Tumbleweed 20190708 Graphics: Device-1: Intel 82915G/GV/910GL Integrated Graphics vendor: Dell driver: N/A bus ID: 00:02.0 chip ID: 8086:2582 Display: tty server: X.Org 1.20.5 driver: intel,vesa unloaded: fbdev,modesetting resolution: 1600x1200~N/A OpenGL: renderer: llvmpipe (LLVM 8.0 128 bits) v: 3.3 Mesa 19.1.1 compat-v: 3.1 direct render: Yes EOF # nomodesetMODESETTING0screensfound [ 1105.572] Kernel command line: root=/dev/sda17 noresume mitigations=auto nomodeset 3 # grep river /etc/X11/xorg.conf.d/50-device.conf # Driver "intel" # Driver "modeset" # Driver "fbdev" Driver "modesetting" [ 215.623] (II) modesetting: Driver for Modesetting Kernel Drivers: kms [ 215.624] (EE) open /dev/dri/card0: No such file or directory [ 215.624] (WW) Falling back to old probe method for modesetting [ 215.624] (EE) open /dev/dri/card0: No such file or directory [ 215.624] (EE) Screen 0 deleted because of no matching config section. [ 215.624] (II) UnloadModule: "modesetting" [ 215.624] (EE) Device(s) detected, but none match those in the config file. [ 215.624] (EE) Fatal server error: [ 215.624] (EE) no screens found(EE) EOF # nomodesetFBDEVnoscreensfound [ 1105.572] Kernel command line: root=/dev/sda17 noresume mitigations=auto nomodeset 3 # grep river /etc/X11/xorg.conf.d/50-device.conf # Driver "intel" # Driver "modeset" Driver "fbdev" # Driver "modesetting" [ 1105.595] (EE) Unable to find a valid framebuffer device EOF # nomodesetMODESETTINGnoscreensfound [ 1105.572] Kernel command line: root=/dev/sda17 noresume mitigations=auto nomodeset 3 # grep river /etc/X11/xorg.conf.d/50-device.conf # Driver "intel" Driver "modeset" # Driver "fbdev" # Driver "modesetting" [ 1449.178] (EE) Failed to load module "modeset" (module does not exist, 0) [ 1449.178] (EE) No drivers available. EOF -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1139889 http://bugzilla.opensuse.org/show_bug.cgi?id=1139889#c23 --- Comment #23 from Felix Miata --- With and without nomodeset at boot, the three you typed, plus modesetting make 8 total permutations I tried in TW20190708 without being able to recreate, which is why I collected y2logs from initializing an installation without nomodeset on installer cmdline, where the Xorg.0.log shows modesetting loaded and modeset(0) on all ModeLine lines. Because of what you typed I thought maybe YaST has a customized DDX named modeset. I know modesetting is upstream's name for what used to be provided by xf86-video-modesetting until it was moved into the server package 4 years ago. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1139889 http://bugzilla.opensuse.org/show_bug.cgi?id=1139889#c25 --- Comment #25 from Felix Miata --- Created attachment 810657 --> http://bugzilla.opensuse.org/attachment.cgi?id=810657&action=edit Xorg.0.log from attempting to use modesetting DDX on 8086:2582 i915G Highlight: modeset(0): Failed to initialize glamor at ScreenInit() time.... (In reply to Stefan Dirsch from comment #24)

I'm only interested in results without 'nomodeset'.

I was trying every permutation I could think of to reproduce. The only way I could come up with was using the modesetting DDX during TW installation, which apparently needs no help to use.

"modesetting" and "Intel" driver cannot workout without KMS.

I have understood this for many many moons. The y2logs attachment includes the installer's Xorg.0.log showing boot was without nomodeset and used the modesetting DDX to produce the subject corruption. However, my understanding has been that Grantsdale GMA 900 i915G GPU as 3rd gen Intel Graphics is too old for modesetting support, which began with 4th gen GMA 3000 Lakeport/Broadwater 946/963/965. I don't recall ever even trying to get modesetting DDX to work on 915G. Seeing it successfully used for TW installation was very surprising to me. -- You are receiving this mail because: You are on the CC list for the bug.