On 7/17/24 12:16 AM, Masaru Nomiya wrote:

In case you are wondering, sddm executes xprofile natively, just like xinitrc, so it is configured as follows.

$ cat /usr/share/sddm/scripts/Xsession | grep profile [...] $SHELL -c "if (-f /etc/csh.login) source /etc/csh.login; if (-f ~/.login) source ~/.login; /bin/sh -c 'export -p' >! $xsess_tmp" . $xsess_tmp rm -f $xsess_tmp ;; */fish) [ -f /etc/profile ] && . /etc/profile [ -f $HOME/.profile ] && . $HOME/.profile -- $SHELL --login -c "/bin/sh -c 'export -p' > $xsess_tmp" . $xsess_tmp rm -f $xsess_tmp ;; *) # Plain sh, ksh, and anything we do not know. [ -f /etc/profile ] && . /etc/profile [ -f $HOME/.profile ] && . $HOME/.profile ;; esac

[ -f /etc/xprofile ] && . /etc/xprofile [ -f /usr/local/etc/xprofile ] && . /usr/local/etc/xprofile [ -f $HOME/.xprofile ] && . $HOME/.xprofile

That is, profile and xprofile are treated equally.

Thank you Felix and Masaru, I think I see the problem. There were changes to the X init between 15.4 and TW moving much to the individual display manager Xsession files. But since I use KDE3/kdm those changes were not made there leaving kdm not sourcing xprofile. So I'll try both ways and see if I can make them work. The short script in xinit.d seems the easiest. But long term for KDE3/kdm do we need to create a kdm/Xsessions to handle this and submit a patch? -- David C. Rankin, J.D.,P.E.

On 7/17/24 2:22 AM, Masaru Nomiya wrote:

What is kdm?

Are you making a mistake?

Well... I might be, but kdm is kdebase3-kdm-3.5.10.1-380.14.x86_64 from the KDE3 repository It is sddm for KDE3 basically. Looking at it's content and how changes have been made to not provide /etc/xdm/Xsession any more in favor of moving the Xsession file to, e.g. /usr/share/sddm/scripts/Xsession, there is now no equivalent Xsession that is read by kdm in order to source xprofile. To my understanding that is why my ~/.xprofile is not sourced when using kdm as the desktop manager. So my thought was -- if this is the new way that Xsession is handled, then I can work on a patch that does that for kdm. Also Felix, my TW install does not have /etc/X11/xinit/xinit.d My 15.4 install does, but my TW install simply has: $ ls -al /etc/X11/xinit/ total 8 drwxr-xr-x 2 root root 4096 Jul 17 00:58 . drwxr-xr-x 5 root root 4096 Jul 7 17:54 .. lrwxrwxrwx 1 root root 41 Jul 4 07:02 xinitrc.common -> ../../../usr/etc/X11/xinit/xinitrc.common Basically a fresh install of TW leaves you with an empty /etc/X11 compared to earlier releases, e.g. $ tree /etc/X11 /etc/X11 ├── kstylerc ├── qt_plugins_3.3rc ├── qtrc ├── xdm │   ├── Xservers │   └── xdm-config ├── xinit │   └── xinitrc.common -> ../../../usr/etc/X11/xinit/xinitrc.common ├── xorg.conf.d │   └── 00-keyboard.conf └── xorg.conf.install Where 15.4 has all the expected files: $ tree /mnt/wz/etc/X11 /mnt/wz/etc/X11 ├── Xmodmap ├── Xmodmap.remote ├── Xresources ├── lbxproxy │   └── AtomControl ├── proxymngr │   └── pmconfig ├── qt_plugins_3.3rc ├── qtrc ├── xdm │   ├── GiveDevices │   ├── Keyboard.map │   ├── README.SUSE │   ├── README.security │   ├── RunChooser │   ├── SUSEconfig.xdm │   ├── SuSEconfig.xdm -> SUSEconfig.xdm │   ├── TakeDevices │   ├── Xaccess │   ├── Xreset │   ├── Xresources │   ├── Xservers │   ├── Xservers.fs │   ├── Xsession │   ├── Xsetup │   ├── Xstartup │   ├── Xwilling │   ├── authdir -> /var/lib/xdm/authdir │   ├── keytable │   ├── scripts │   │   ├── 09-ssh-vars │   │   ├── 10-gpg-agent │   │   ├── 11-ssh-agent │   │   ├── 20-dbus │   │   ├── 30-console-kit │   │   └── session-scriptlets.HOWTO │   ├── sys.xsession │   └── xdm-config ├── xim ├── xim.d │   └── none ├── xinit │   ├── xinitrc │   ├── xinitrc.common │   ├── xinitrc.d │   │   ├── 50-systemd-user.sh │   │   └── libcanberra-gtk-module.sh │   └── xserverrc ├── xorg.conf.d │   ├── 00-keyboard.conf │   ├── 10-evdev.conf │   ├── 10-quirks.conf │   ├── 11-evdev.conf │   ├── 40-libinput.conf │   ├── 50-device.conf │   ├── 50-elotouch.conf │   ├── 50-extensions.conf │   ├── 50-monitor.conf │   ├── 50-screen.conf │   ├── 70-synaptics-dcr.conf │   └── 70-wacom.conf ├── xorg.conf.install └── xorg_pci_ids └── modesetting.ids This is where the issue is coming from. The only xinit.d in TW is /usr/etc/X11/xinit/xinit.d/ -- David C. Rankin, J.D.,P.E.

On 7/17/24 3:36 AM, Masaru Nomiya wrote:

MN> > Are you making a mistake?

DCR> Well... I might be, but kdm is

DCR> kdebase3-kdm-3.5.10.1-380.14.x86_64 from the KDE3 repository

DCR> It is sddm for KDE3 basically. Looking at it's content and how DCR> changes have been made to not provide /etc/xdm/Xsession any more DCR> in favor of moving the Xsession file to, DCR> e.g. /usr/share/sddm/scripts/Xsession, there is now no DCR> equivalent Xsession that is read by kdm in order to source DCR> xprofile. To my understanding that is why my ~/.xprofile is not DCR> sourced when using kdm as the desktop manager. [...]

I'm using sddm with /etc/sddm.conf as follows;

[General]

Thanks Masaru, I was able to restore sourcing of both ~/.xprofile and /etc/profile by adding a short script to /usr/etc/X11/xinit/xinit.d as follows: $ cat /usr/etc/X11/xinit/xinitrc.d/60-my-xinit.sh #!/bin/sh [ -f /etc/xprofile ] && . /etc/xprofile [ -f ~/.xprofile ] && . ~/.xprofile This works fine and will be generic for all desktops. The xprofile I use is innocuous, so if it is source multiple times it makes no difference, e.g. $ cat ~/.xprofile #!/bin/sh ## kill the bloody pc speaker xset -b ## set the keyboard delay and repeat rate xset r rate 276 50 ======================================================================== Now the next big issue is why the middle mouse paste of the select buffer is only working ever 3rd time or so after a long press with the USB-mouse (yes I've tried multiple good mice). Ironically it works fine from the touchpad buttons. So it looks like there is some conflict in TW between the USB middle-mouse and the touchpad middle-mouse. The middle-mouse-wheel does not work in Firefox/Thunderbird unless they are top-level BUT the middle-mouse-wheel scroll from the touchpad does. It's almost like the Microsoft 3-Button generic mouse driver is broken. The xinput output is: $ xinput ⎡ Virtual core pointer id=2 [master pointer (3)] ⎜ ↳ Virtual core XTEST pointer id=4 [slave pointer (2)] ⎜ ↳ Microsoft Microsoft 3-Button Mouse with IntelliEye(TM) id=9 [slave pointer (2)] ⎜ ↳ PS/2 Generic Mouse id=11 [slave pointer (2)] ⎜ ↳ SynPS/2 Synaptics TouchPad id=12 [slave pointer (2)] ⎣ Virtual core keyboard id=3 [master keyboard (2)] ↳ Virtual core XTEST keyboard id=5 [slave keyboard (3)] ↳ Power Button id=6 [slave keyboard (3)] ↳ Video Bus id=7 [slave keyboard (3)] ↳ Sleep Button id=8 [slave keyboard (3)] ↳ AT Translated Set 2 keyboard id=10 [slave keyboard (3)] ↳ HP WMI hotkeys id=13 [slave keyboard (3)] The touchpad (id=12) works perfect. The micrsoft mouse (id=9) is really flaky under TW. -- David C. Rankin, J.D.,P.E.

On 7/17/24 11:58 PM, Masaru Nomiya wrote:

DCR> The xinput output is:

DCR> $ xinput DCR> ⎡ Virtual core pointer id=2 [master pointer (3)] DCR> ⎜ ↳ Virtual core XTEST pointer id=4 [slave pointer (2)] DCR> ⎜ ↳ Microsoft Microsoft 3-Button Mouse with IntelliEye(TM) id=9 [slave DCR> pointer (2)] DCR> ⎜ ↳ PS/2 Generic Mouse id=11 [slave pointer (2)] DCR> ⎜ ↳ SynPS/2 Synaptics TouchPad id=12 [slave pointer (2)] DCR> ⎣ Virtual core keyboard id=3 [master keyboard (2)] DCR> ↳ Virtual core XTEST keyboard id=5 [slave keyboard (3)] DCR> ↳ Power Button id=6 [slave keyboard (3)] DCR> ↳ Video Bus id=7 [slave keyboard (3)] DCR> ↳ Sleep Button id=8 [slave keyboard (3)] DCR> ↳ AT Translated Set 2 keyboard id=10 [slave keyboard (3)] DCR> ↳ HP WMI hotkeys id=13 [slave keyboard (3)]

DCR> The touchpad (id=12) works perfect. The micrsoft mouse (id=9) DCR> is really flaky under TW.

1. What is th bloody speaker?

Chuckling.... Alright, is this a trick question? What speaker are you looking at? (the broken kmix is another issue :)

2. What's this?

PS/2 Generic Mouse id=11 [slave pointer (2)]

This is the stupid pencil eraser size nub between the 'G' and 'H' keys that you can also use to move the mouse -- quite worthless actually and a PITA to get used to having that thing in the middle of the keyboard. I think they first appeared on IBM laptops years ago, and then HP picked them up too.

Wouldn't these give you hints?

I know what they are, but the detailed info doesn't tell me much, e.g.

1. $ xinput --list-props "Microsoft Microsoft 3-Button Mouse with IntelliEye(TM)"

$ xinput --list-props "Microsoft Microsoft 3-Button Mouse with IntelliEye(TM)" Device 'Microsoft Microsoft 3-Button Mouse with IntelliEye(TM)': Device Enabled (153): 1 Coordinate Transformation Matrix (155): 1.000000, 0.000000, 0.000000, 0.000000, 1.000000, 0.000000, 0.000000, 0.000000, 1.000000 libinput Natural Scrolling Enabled (283): 0 libinput Natural Scrolling Enabled Default (284): 0 libinput Scroll Methods Available (285): 0, 0, 1 libinput Scroll Method Enabled (286): 0, 0, 0 libinput Scroll Method Enabled Default (287): 0, 0, 0 libinput Button Scrolling Button (288): 2 libinput Button Scrolling Button Default (289): 2 libinput Button Scrolling Button Lock Enabled (290): 0 libinput Button Scrolling Button Lock Enabled Default (291): 0 libinput Middle Emulation Enabled (292): 0 libinput Middle Emulation Enabled Default (293): 0 libinput Rotation Angle (266): 0.000000 libinput Rotation Angle Default (267): 0.000000 libinput Accel Speed (294): 0.000000 libinput Accel Speed Default (295): 0.000000 libinput Accel Profiles Available (296): 1, 1, 1 libinput Accel Profile Enabled (297): 1, 0, 0 libinput Accel Profile Enabled Default (298): 1, 0, 0 libinput Accel Custom Fallback Points (299): <no items> libinput Accel Custom Fallback Step (300): 0.000000 libinput Accel Custom Motion Points (301): <no items> libinput Accel Custom Motion Step (302): 0.000000 libinput Accel Custom Scroll Points (303): <no items> libinput Accel Custom Scroll Step (304): 0.000000 libinput Left Handed Enabled (305): 0 libinput Left Handed Enabled Default (306): 0 libinput Send Events Modes Available (268): 1, 0 libinput Send Events Mode Enabled (269): 0, 0 libinput Send Events Mode Enabled Default (270): 0, 0 Device Node (271): "/dev/input/event4" Device Product ID (272): 1118, 64 libinput Drag Lock Buttons (307): <no items> libinput Horizontal Scroll Enabled (308): 1 libinput Scrolling Pixel Distance (309): 15 libinput Scrolling Pixel Distance Default (310): 15 libinput High Resolution Wheel Scroll Enabled (311): 1

2. $ xinput --list-props "PS/2 Generic Mouse"

$ xinput --list-props "PS/2 Generic Mouse" Device 'PS/2 Generic Mouse': Device Enabled (153): 1 Coordinate Transformation Matrix (155): 1.000000, 0.000000, 0.000000, 0.000000, 1.000000, 0.000000, 0.000000, 0.000000, 1.000000 libinput Natural Scrolling Enabled (283): 0 libinput Natural Scrolling Enabled Default (284): 0 libinput Scroll Methods Available (285): 0, 0, 1 libinput Scroll Method Enabled (286): 0, 0, 1 libinput Scroll Method Enabled Default (287): 0, 0, 1 libinput Button Scrolling Button (288): 2 libinput Button Scrolling Button Default (289): 2 libinput Button Scrolling Button Lock Enabled (290): 0 libinput Button Scrolling Button Lock Enabled Default (291): 0 libinput Middle Emulation Enabled (292): 0 libinput Middle Emulation Enabled Default (293): 0 libinput Rotation Angle (266): 0.000000 libinput Rotation Angle Default (267): 0.000000 libinput Accel Speed (294): 0.000000 libinput Accel Speed Default (295): 0.000000 libinput Accel Profiles Available (296): 1, 1, 1 libinput Accel Profile Enabled (297): 1, 0, 0 libinput Accel Profile Enabled Default (298): 1, 0, 0 libinput Accel Custom Fallback Points (299): <no items> libinput Accel Custom Fallback Step (300): 0.000000 libinput Accel Custom Motion Points (301): <no items> libinput Accel Custom Motion Step (302): 0.000000 libinput Accel Custom Scroll Points (303): <no items> libinput Accel Custom Scroll Step (304): 0.000000 libinput Left Handed Enabled (305): 0 libinput Left Handed Enabled Default (306): 0 libinput Send Events Modes Available (268): 1, 0 libinput Send Events Mode Enabled (269): 0, 0 libinput Send Events Mode Enabled Default (270): 0, 0 Device Node (271): "/dev/input/event2" Device Product ID (272): 2, 1 libinput Drag Lock Buttons (307): <no items> libinput Horizontal Scroll Enabled (308): 1 libinput Scrolling Pixel Distance (309): 15 libinput Scrolling Pixel Distance Default (310): 15 libinput High Resolution Wheel Scroll Enabled (311): 1 But the important touchpad that DOES work to scroll FF without the GDK_CORE_DEVICE_EVENTS=1 info may be helpful (2-finger scroll): $ xinput --list-props "SynPS/2 Synaptics TouchPad" Device 'SynPS/2 Synaptics TouchPad': Device Enabled (153): 1 Coordinate Transformation Matrix (155): 1.000000, 0.000000, 0.000000, 0.000000, 1.000000, 0.000000, 0.000000, 0.000000, 1.000000 Device Accel Profile (279): 1 Device Accel Constant Deceleration (280): 2.500000 Device Accel Adaptive Deceleration (281): 1.000000 Device Accel Velocity Scaling (282): 12.500000 Synaptics Edges (312): 1767, 5397, 1649, 4615 Synaptics Finger (313): 25, 30, 0 Synaptics Tap Time (314): 180 Synaptics Tap Move (315): 239 Synaptics Tap Durations (316): 180, 180, 100 Synaptics ClickPad (317): 0 Synaptics Middle Button Timeout (318): 75 Synaptics Two-Finger Pressure (319): 282 Synaptics Two-Finger Width (320): 7 Synaptics Scrolling Distance (321): 108, 108 Synaptics Edge Scrolling (322): 0, 0, 0 Synaptics Two-Finger Scrolling (323): 1, 0 Synaptics Move Speed (324): 1.000000, 1.750000, 0.036704, 0.000000 Synaptics Off (325): 1 Synaptics Locked Drags (326): 0 Synaptics Locked Drags Timeout (327): 5000 Synaptics Tap Action (328): 0, 0, 0, 0, 1, 3, 2 Synaptics Click Action (329): 1, 1, 1 Synaptics Circular Scrolling (330): 0 Synaptics Circular Scrolling Distance (331): 0.100000 Synaptics Circular Scrolling Trigger (332): 0 Synaptics Circular Pad (333): 0 Synaptics Palm Detection (334): 0 Synaptics Palm Dimensions (335): 10, 200 Synaptics Coasting Speed (336): 20.000000, 50.000000 Synaptics Pressure Motion (337): 30, 160 Synaptics Pressure Motion Factor (338): 1.000000, 1.000000 Synaptics Grab Event Device (339): 0 Synaptics Gestures (340): 1 Synaptics Capabilities (341): 1, 1, 1, 1, 1, 1, 1 Synaptics Pad Resolution (342): 81, 40 Synaptics Area (343): 0, 0, 0, 0 Synaptics Noise Cancellation (344): 27, 27 Device Product ID (272): 2, 7 Device Node (271): "/dev/input/event3" I don't see anything that immediately says -- "this /dev/input/eventX is hosed". I'm no expert in the finer points of mouse-hardware, but they seem fine. What is really "bloody" odd about the whole situation is I attached xev to the firefox window and captured the events between the mouse (with and without GDK_CORE_DEVICE_EVENTS=1) and touchpad and the X events captured are for all practical purposes identical. I've literally pulled my hair out trying to figure out how to tell what the difference is between what the FF window sees from the mouse and what it sees from the touchpad. Moreover, why it scrolls with the touchpad, but not with the mouse unless GDK_CORE_DEVICE_EVENTS=1. And further why the mouse scrolls FF fine in fluxbox without it. Do you see anything in the information? Let me know if you want to see the xev event traces and I'll post them on paste.opensuse.org. I basically just created short wrappers to launch FF and TB with the environment variable set, e.g.: $ cat ~/bin/firefox-esr-fix.sh #!/bin/sh GDK_CORE_DEVICE_EVENTS=1 firefox-esr & I'm pretty much out of ideas. This is just bizarre, but without mousewheel scroll after 20 years of using focus-follows-mouse, it will drive you nuts :) -- David C. Rankin, J.D.,P.E.

On 7/18/24 4:58 AM, Masaru Nomiya wrote:

MN> > 1. What is th bloody speaker?

DCR> Chuckling.... Alright, is this a trick question? What speaker DCR> are you looking at? (the broken kmix is another issue :)

I think alsamixer is not available either?

It is, and it works fine -- that's what I'm using for volume. I can also use "kamix", but the icon is HUGE in the systray. "kmix" simply cannot find a mixer. You open kmix config and the select a mixer combo-box drop down is empty. Sound works just fine - it's just kmix has lost its bloody mind :)

Sorry, but I can only say, "How about installing imwheel?"

Wasn't needed on 15.4, but worth a shot. One thing I did note about the xinput --info outputs between the MS-Mouse and the Generic PS/2 Mouse is it looks like the MS-Mouse is being treated the same as the Generic PS/2 Mouse (the pencil-eraser nub in the middle of the keyboard) I wouldn't expect recognition of the iconic MS 3-Button Wheel Mouse to change in firmware or in how it's detected over USB, but looking at the Mouse and Generic info output, they looked quite similar, and the Generic PS/2 eraser-nub has no button or wheel. Worse-case scenario, I'll just continue using the wrappers that sets GDK_CORE_DEVICE_EVENTS=1 for firefox-esr and thunderbird in Tumbleweed. That works fine, and so for the side-effects seem limited to just killing smooth-scrolling in those two apps (that doesn't matter to me, I never noticed much usable difference anyway) Really strange issue. But this issue has appeared on/off since mozilla went to Gtk+3 and now Gtk4, so there is likely a regression somewhere, or something kdm/kwin isn't setting that used to be set in the normal X11 setup with the full Xsession, etc.... setup and possibly the sysconfig. Since the issue isn't present in fluxbox, this points to a kwin (or how it is launched) issue. Thank you for all your ideas Masaru! -- David C. Rankin, J.D.,P.E.

David C. Rankin composed on 2024-07-17 02:50 (UTC-0500):

Masaru Nomiya wrote:

...

What is kdm?

...

Are you making a mistake?

Well... I might be, but kdm is

kdebase3-kdm-3.5.10.1-380.14.x86_64 from the KDE3 repository

It is sddm for KDE3 basically.

my TW install does not have /etc/X11/xinit/xinit.d My 15.4 install does ... ├── xinit │   ├── xinitrc │   ├── xinitrc.common │   ├── xinitrc.d │   │   ├── 50-systemd-user.sh │   │   └── libcanberra-gtk-module.sh │   └── xserverrc ├── xorg.conf.d ... /etc/X11/xinit/xinitrc.d/, like /etc/x11/xorg.conf.d/, is an optional directory, which new TW installations don't find necessary to create by default. Those who need them for overriding defaults can create them. Not creating optional

Hardly. Sddm is upstream's pathetic replacement for KDM, missing a wealth of optional functionality, and some basic functionality. Of all places, Fedora never dropped KDM from its standard repos, even though its X11 is stuck a whole bunch of releases behind upstream & TW. directories has been a basic characteristic of /etc/ since /usr/etc/ was created.

This is where the issue is coming from. The only xinit.d in TW is /usr/etc/X11/xinit/xinit.d/

# ls -ld /etc/X11/xinit/xinitrc.d/ drwxr-xr-x 2 root root 4096 Aug 17 2023 /etc/X11/xinit/xinitrc.d/ # -- Evolution as taught in public schools is, like religion, based on faith, not based on science. Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata