[libvirt] problem of escaped scancodes

[KAMEZAWA Hiroyuki]

On Thu, 25 Aug 2011 17:51:49 +0100
"Daniel P. Berrange" <berrange at redhat.com> wrote:

> On Thu, Aug 25, 2011 at 10:25:57AM +0900, KAMEZAWA Hiroyuki wrote:
> > On Tue, 23 Aug 2011 14:54:11 +0100
> > "Daniel P. Berrange" <berrange at redhat.com> wrote:
> >  
> > > No, we just need to switch to use the correct scancode set.
> > > 
> > > The original keymap-gen.pl script in GTK-VNC has code for auto-generating
> > > the RFB scancode set, from the data for the XT KBD scancode set. It is a
> > > straighforward re-encoding for the extended keys:
> > > 
> > >     # RFB keycodes are XT kbd keycodes with a slightly
> > >     # different encoding of 0xe0 scan codes. RFB uses
> > >     # the high bit of the first byte, instead of the low
> > >     # bit of the second byte.
> > >     rfbkey = (xtkbdkey & 0x100) >> 1 | (ktkbdkey & 0x7f)
> > > 
> > like this ?
> > 
> > ==
> > From 7ea5214b7832acbb7bae81d7d8ceeaa19890e32b Mon Sep 17 00:00:00 2001
> > From: KAMEZAWA Hiroyuki <kamezawa.hiroyu at jp.fujitsu.com>
> > Date: Thu, 25 Aug 2011 10:29:57 +0900
> > Subject: [PATCH] fix to adjust xt_kbd scancode
> > 
> > The keycode in http://git.gnome.org/browse/gtk-vnc/plain/src/keymaps.csv
> > handles escaped scancode in different way from qemu's one.
> > 
> > all escaped scancode "e0 ??" are encoded as "0x100 | ??" in this csv.
> > But qemu handles escaped scancode as "0x80 | ??" and this scancode
> > is more common.
> > ---
> >  src/util/virkeycode-mapgen.py |   14 +++++++++++++-
> >  1 files changed, 13 insertions(+), 1 deletions(-)
> > 
> > diff --git a/src/util/virkeycode-mapgen.py b/src/util/virkeycode-mapgen.py
> > index acf7364..5d83a39 100755
> > --- a/src/util/virkeycode-mapgen.py
> > +++ b/src/util/virkeycode-mapgen.py
> > @@ -14,6 +14,7 @@ import sys
> >  import re
> >  
> >  namecolums = (0,2,10)
> > +xtkbdkey_index = 8
> >  
> >  def quotestring(str):
> >      if str[0] != '"':
> > @@ -38,7 +39,18 @@ for line in sys.stdin.xreadlines():
> >      for i in namecolums:
> >          b = b + (a[i] and quotestring(a[i]) or 'NULL') + ','
> >      for i in [ x for x in range(12) if not x in namecolums ]:
> > -        b = b + (a[i] or '0') + ','
> > +        if i != xtkbdkey_index :
> > +            b = b + (a[i] or '0') + ','
> > +        elif a[i] == '' :
> > +            b = b + '0' + ','
> > +        else :
> > +            # RFB keycodes are XT kbd keycodes with a slightly
> > +            # different encoding of 0xe0 scan codes. RFB uses
> > +            # the high bit of the first byte, instead of the low
> > +            # bit of the second byte.
> > +            code = int(a[i])
> > +            code = (code & 0x100) >> 1 | (code & 0x7f)
> > +            b = b + str(code) + ','
> >      print "    { " + b + "},"
> >  
> >  print '};'
> 
> This is not quite right. We don't want to change the XT KBD codeset
> since that's defined that way by the Linux driver. We instead want
> to add a new RFB codeset and make the QEMU driver use that. Check
> out this patch
> 
> https://www.redhat.com/archives/libvir-list/2011-August/msg01256.html
> 
> In my tests it made  'virsh send-key f16x86_64 KEY_RIGHTCTRL KEY_C'
> work correctly
> 

Okay, thank you.

-Kame