Simon Thum authored Jul 23, 2008 and Peter Hutterer committed Jul 28, 2008

Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>

Apart from the obvious reference counting issue, this fixes
http://bugs.freedesktop.org/show_bug.cgi?id=16867 .

Could have crashed otherwise if the num(Back)ClipRects variables referenced by
the caller weren't pre-initialized to 0.

Looks like an artifact from early modularization.

Only strings are parsed by the server, all others are ignored. Doesn't matter,
specifying int options as strings works fine anyway.

Really not necessary, we can just walk the list and spare us the special
treatment of the VCK.

Put out a warning if xorg.conf has InputDevice sections, but these aren't
referenced in the used ServerLayout. This is only performed if AllowEmptyInput
is enabled.
The reason behind this is that the server used to auto-add the first
mouse/keyboard sections if none where referenced. Now, with HAL and AEI
enabled by default, setups that relied on this auto-adding break and are left
without input devices. The least we can do is warn them.

Core events don't happen until later in the DIX, so pump device events down
instead. This makes modifiers work again when SlowKeys is enabled.

Two corrections
1. the "detail" field has NotifyVirtual, etc., not the "mode" field. This was
a clear bug.
2. don't set/unset the flags for NotifyGrab or NotifyUngrab. Clients are
expected to deal with multiple enter/leave events per window if the mode is
not NotifyNormal.

Testable with TCL menu boxes (such as used in gitk):
tk_optionMenu .menu globVar Val1 Val2 Val3 ValJunk
pack .menu

Thanks to Michel Dänzer for pointing this out.

Daniel Drake authored Jul 24, 2008 and Daniel Drake committed Jul 24, 2008

GLX is enabled by default, but the current swrast behaviour causes X
to abort with fatal error if the swrast dri library dlopen fails.

Handle the case where the swrast library is not present, and do not
register the GLX extension unless at least one screen has a usable
GL provider.

These are AT scancode specific, should really be part of xf86-input-keyboard.
Remove from server, move to xf86-input-keyboard.

Us shipping a GUI configuration utility (especially as part of the
server!) was pretty pointless.  There was pretty much nothing it could
configure which wasn't already runtime adjustable: if you could get a
server up with functioning input and output, there wasn't much xorgcfg
could do for you.

Au revoir.

I don't know why we would possibly be shipping this.

We only built this on BSD and Solaris, and if such a tool is generally
useful, ship it with the OS.

Reorder to minimize comparisons

gcc 3.0 was seven years ago.  Get with the times.

Again, hasn't worked since at least 7.0.

Using strncasecmp(3) with the lenght of the user-supplied colour name
will result in a false positive when the db key starts out with the
same string.

Eg, blue will also match BlueViolet (aka blue violet).

Since the shorter strings occur first in the database, avoid such
errors by treating a 0 result from strncasecmp(3) as a positive result
when the key’s length is longer than the supplied string’s.