Bug 19500 - syndaemon -k doen't ignore mod4
Summary: syndaemon -k doen't ignore mod4
Alias: None
Product: xkeyboard-config
Classification: Unclassified
Component: General (show other bugs)
Version: unspecified
Hardware: x86 (IA32) Linux (All)
: medium normal
Assignee: xkb
QA Contact:
: 23266 (view as bug list)
Depends on:
Reported: 2009-01-10 09:01 UTC by Mikael Eriksson
Modified: 2009-10-18 14:12 UTC (History)
7 users (show)

See Also:
i915 platform:
i915 features:

Fix Win key as third level chooser (1.18 KB, patch)
2009-10-16 13:11 UTC, Alexey Ten
Details | Splinter Review

Description Mikael Eriksson 2009-01-10 09:01:30 UTC
syndaemon -k doens't ignore mod4, only mod1, control and shift.

xorg-server          1.5.3-4
xf86-input-synaptics 0.99.3-1
xf86-input-evdev     2.1.0-1

shift       Shift_L (0x32),  Shift_R (0x3e)
control     Control_L (0x25),  Control_R (0x69)
mod1        Alt_L (0x40),  Meta_L (0xcd)
mod2        Num_Lock (0x4d)
mod4        Super_L (0xce),  Hyper_L (0xcf)
mod5        ISO_Level3_Shift (0x5c),  Mode_switch (0xcb)
Comment 1 Peter Hutterer 2009-03-01 20:29:08 UTC
not a syndaemon bug actually.

Mod4 is mapped to SUPR, a virtual key. This isn't the same as the physical keys, which are mapped to Super_L and Super_R and don't show up in the core mapping.

setxkbmap -option "altwin:super_win" creates this mapping and it works fine.

Sergey - any reason why this isn't the default?
Comment 2 Mikael Eriksson 2009-03-02 08:28:44 UTC
Thanks! With setxkbmap -option "altwin:super_win" it works.
Comment 3 Sergey V. Udaltsov 2009-03-02 12:55:05 UTC
> Sergey - any reason why this isn't the default?

You mean - why Mod4 is not mapped to LWIN/RWIN in addition to SUPR ? Well, I cannot think of any reason. Should we try that?

Comment 4 Peter Hutterer 2009-03-02 13:03:57 UTC
On Mon, Mar 02, 2009 at 12:55:05PM -0800, bugzilla-daemon@freedesktop.org wrote:
> You mean - why Mod4 is not mapped to LWIN/RWIN in addition to SUPR ? Well, I
> cannot think of any reason. Should we try that?

I think it'd make sense.  Since SUPR is a only virtual key it appears in
xmodmap but isn't actually useable. And RWIN/LWIN are mapped to nothing by
default, which is rather pointless.
Comment 5 Sergey V. Udaltsov 2009-03-04 14:51:50 UTC
Committed! Please check.
Comment 6 Stephane Glondu 2009-06-13 11:17:05 UTC
This change seems to break level3 switch with LWIN/RWIN in GTK+ applications. See Debian bug:


I myself don't really understand why this change would cause this bug, so I guess the problem is in GTK+. Do you have any idea on the subject?
Comment 7 Sergey V. Udaltsov 2009-08-25 16:46:20 UTC
According to #23266, this fix is invalid - it breaks things for people using lv3:lwin_switch.

As the result, I am reverting this patch: http://cgit.freedesktop.org/xkeyboard-config/commit/?id=5de02aa07a8d4bbe1957af3a38212c3507f2436f

I'm afraid, you have to specify altwin:super_win explicitly
Comment 8 Peter Hutterer 2009-08-25 17:58:35 UTC
I can't reproduce this. Tested with 3e6a7a495a4f7ceb8c2dd505003f02c893c18975 added caps:numlock

lv3:lwin_switch, lv3:rwin_switch and lv3:win_switch all work fine and as expected (tested on layout 'de').

what's the magic sauce required to break it?
Comment 9 Sergey V. Udaltsov 2009-08-26 15:32:39 UTC
*** Bug 23266 has been marked as a duplicate of this bug. ***
Comment 10 Sergey V. Udaltsov 2009-08-26 15:33:33 UTC
Reopening, since bug #23266 + Debian's bugzilla say that "fix" is causing troubles
Comment 11 Sergey V. Udaltsov 2009-08-26 15:37:52 UTC
Actually, it works to me as well - but may be it only manifests itself with certain layouts/models...
Comment 12 Alexey Ten 2009-10-16 13:11:38 UTC
Created attachment 30489 [details] [review]
Fix Win key as third level chooser

This patch fixes win-key as third level chooser for me.
Comment 13 Sergey V. Udaltsov 2009-10-18 14:12:28 UTC
Ok, let user explicitly specify altwin:super_win. Let's see how many complains we're going to get...

Thanks for the patch!

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.