On Sat, 2005-06-04 at 17:15 -0700, John Davidorff Pell wrote:
> Aside from modularisation, what changes will be made in the next
> release? I know there are a lot of bug fùÇ . +› .. úÇ
Repository wÇ ÔEntries 9Ç Ä Entries.Log wÇ °Entries.Backup /cvs/xorg/xc/programs/Xserver/hw/xfree86/etc/bindist/Cygwin
/Imakefile/1.3/=//
D
reedesktop.org>; Sat, 4 Jun 2005 21:57:17 -0700 (PDT)
Received: from [2002:83fc:f325:1:212:f0ff:fe24:d97c]
(helo=[IPv6:2002:83fc:f325:1:212:f0ff:fe24:d97c])
by dor.minilop.net with esmtpsa (TLS-1.0:RSA_ARCFOUR_MD5:16)
(Exim 4.50) id 1DemFm-0006RP-AN
for xorg at lists.freedesktop.org; Sat, 04 Jun 2005 20:56:50 -0700
From: Jamey Sharp <jamey at minilop.net>
To: xorg at lists.freedesktop.org
In-Reply-To: <200506042358.04302.ajax at nwnk.net>
References: <0F693E89-AA54-434E-8CA5-598406E22406 at mac.com>
<200506042132.17062.ajax at nwnk.net>
<1117938485.6765.99.camel at id.minilop.net>
<200506042358.04302.ajax at nwnk.net>
Content-Type: text/plain
Date: Sat, 04 Jun 2005 21:57:14 -0700
Message-Id: <1117947434.6765.146.camel at id.minilop.net>
Mime-Version: 1.0
X-Mailer: Evolution 2.2.2
Content-Transfer-Encoding: 7bit
Subject: Re: X11R7
X-BeenThere: xorg at lists.freedesktop.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discuss issues related to the xorg tree <xorg.lists.freedesktop.org>
List-Unsubscribe: <http://lists.freedesktop.org/mailman/listinfo/xorg>,
<mailto:xorg-request at lists.freedesktop.org?subject=unsubscribe>
List-Archive: <http://lists.freedesktop.org/archives/xorg>
List-Post: <mailto:xorg at lists.freedesktop.org>
List-Help: <mailto:xorg-request at lists.freedesktop.org?subject=help>
List-Subscribe: <http://lists.freedesktop.org/mailman/listinfo/xorg>,
<mailto:xorg-request at lists.freedesktop.org?subject=subscribe>
X-List-Received-Date: Sun, 05 Jun 2005 04:57:18 -0000
On Sat, 2005-06-04 at 23:58 -0400, Adam Jackson wrote:
> On Saturday 04 June 2005 22:28, Jamey Sharp wrote:
> > On Sat, 2005-06-04 at 21:32 -0400, Adam Jackson wrote:
> > > XCB will probably not be included, it's not baked enough to be considered
> > > stable
> >
> > Hey now, putting it that strongly isn't very nice. ;-) I don't know of
> > any bugs in XCB's implementation; it's just the API design and Xlib
> > hacks that need a little work.
>
> Well, and the extensions that aren't quite done yet.
After quick inspection, I can't find any extensions that are supported
by XCB at all but have some legal request that XCB can't be used to
produce. There are certainly pending API changes that will make it
*easier* to produce some requests... I may, of course, have missed
something.
Anyway, Xlib compiled --with-xcb doesn't use XCB's extension stubs, or
even most of its core protocol stubs. All extensions currently supported
by Xlib, including those supported by closed-source binary libraries,
continue to be supported by Xlib/XCB.
> But failing the testsuite on Xlib is pretty much a showstopper, yes?
> Conversely, passing the testsuite is the gold standard for "awesome,
> enabled by default".
Showstopper for Xlib --with-xcb, sure. I believe all the test suite
failures are due to bugs in that version of Xlib, not in XCB. So the
twenty pure-XCB apps and libraries in the world are fine. ;-)
Actually, in practice the test suite failures don't have much effect.
Obviously they need to be fixed but they're generally small issues: a
complete desktop has been quite usable on Xlib/XCB for months now.
I'm not arguing XCB *should* be in 6.9/7.0, just that it's a little
better than "not baked enough to be considered stable". :-)
> It's your code. The "same code, different build systems" dictum of the 7.0
> split makes me want to reduce the amount of new big subsystems added to 6.9,
> just because it's work that'll get tossed after one release.
Heh, hard to argue with that.
> If XCB gets to
> a point where it's rock-solid and conformant on every platform we can find,
> then by all means let's add it.
>
> ... my
> perspective is that XCB doesn't look like it's getting quite enough love to
> be shippable yet. I'm happy to be convinced otherwise. ;)
I'd be happy to convince you otherwise too, but in all honesty I can't
just yet. :-)
So hey, if there are folks out there looking for some way to contribute
to the future of the X Window System, let me encourage you to check out
http://xcb.freedesktop.org and test out Xlib/XCB on whatever platforms
you can find, or dive into the code and start hacking. Let's get some
love over here and convince ajax. :-)
--Jamey