[Xquartz-dev] 2.2.1_rc1 and Xquartz-1.4.0-apple3

[Jeremy Huddleston]

I've rolled together a new package that puts together all the work  
that has gone into XQuartz over the past week or so.  We've made a lot  
of progress on the stability front and on the migration to the 1.4  
branch.

The 2.2.1 release candidate includes recently released versions of  
some X11 apps and libs (including pixman and cairo), fixes the most  
common crashes in Xquartz, and fixes cmd-tab (which now brings all the  
windows forward).

Release notes for 2.2.1 are here:
http://trac.macosforge.org/projects/xquartz/wiki/X112.2.1

The package is available here:
http://xquartz.macosforge.org/downloads/X11-2.2.1_rc1.pkg

There are no known regressions with 2.2.1_rc1 (so please try to find  
them).

I'd still like to get the server lockup bug fixed for the 2.2.1  
release, but I'm having trouble triggering it.  I need good test cases  
for this.  If you can reliably lockup the server, please let me know:
http://trac.macosforge.org/projects/xquartz/ticket/41 - X11 lockup at  
__xp_async_dequeue

Also, can someone please confirm that fonts work fine on ppc with 2.2.1:
http://trac.macosforge.org/projects/xquartz/ticket/92 - font issues on  
ppc

---

We've also made some progress in getting onto a newer xserver  
codebase.  1.4.0-apple3 is available for your testing here (just  
replace /usr/X11/bin/Xquartz from the 2.2.1_rc1 package with this one):
http://xquartz.macosforge.org/downloads/Xquartz-1.4.0-apple3.bz2

As mentioned before, 1.4 adds in support for xkb and tablets, so if  
you use or want to use anything other than an english keyboard and  
mouse, *please* test it out and notify us of regressions.

It addresses the two reported issues with 1.4.0-apple2:
Dragging windows to another space ---> crash
External monitor ---> bork

There are two known issues:
1) Sometimes when X11 is quit, a crash will occur in the new input  
handling code.  I would appreciate any assistance in narrowing down  
this issue (if there's something you can do that will always trigger  
it, for example).

2) There is sometimes some spew at the beginning about 'Failed to lock  
mieqEnqueue_mutex'... this is because of a race of one thread to use  
the mutex before the other inits it.  I'll have that fixed soon.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3221 bytes
Desc: not available
Url : http://lists.macosforge.org/pipermail/xquartz-dev/attachments/20080422/296932d8/smime.bin