Re: XOrg-7.2 - libXcb

Matthew Burgess wrote:
> 1) The book instructions mention that XCB is "optional", whereas libX11 pretty 
> clearly wants/expects it to be there.  Therefore, should it be upgraded 
> to "recommended"? (I have noted the book mentions the --without-xcb configure 
> option, btw).

This should be added to the ./configure line

> 3) Is there a reason XCB was not added to the book?  (It needs the 
> libpthread-stubs package and xsltproc from a quick look at the dependencies).

Yes - Java and internal Xorg problems. E.g., with libXi 1.0.2 (that comes 
with stock Xorg-7.2 without updates) any program using the XINPUT extension 
(e.g., Xdmx) would print a message about a failed assertion if you attempt 
to run it against an X server that doesn't support this extension.

XCB merely exposes locking bugs in other applications. But, since there are 
such bugs even in Xorg itself, and there is not enough time and developers 
to collect/confirm the fixes and make sure that we collected all of them, 
the decision has been made to discourage the use of XCB for Xorg-7.2.

-- 
Alexander E. Patrakov
--

-- 
http://linuxfromscratch.org/mailman/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Re: XOrg-7.2 - libXcb

On 6/30/07, Alexander E. Patrakov <patrakov <at> ums.usu.ru> wrote:
> Matthew Burgess wrote:
>
> > 3) Is there a reason XCB was not added to the book?  (It needs the
> > libpthread-stubs package and xsltproc from a quick look at the dependencies).
>
> Yes - Java and internal Xorg problems. E.g., with libXi 1.0.2 (that comes
> with stock Xorg-7.2 without updates) any program using the XINPUT extension
> (e.g., Xdmx) would print a message about a failed assertion if you attempt
> to run it against an X server that doesn't support this extension.
>
> XCB merely exposes locking bugs in other applications. But, since there are
> such bugs even in Xorg itself, and there is not enough time and developers
> to collect/confirm the fixes and make sure that we collected all of them,
> the decision has been made to discourage the use of XCB for Xorg-7.2.

I think we should just push in XCB, but disable the assertions since I
don't think we can track down all these issues yet. Attached patch is
what I'm sitting on + CPPFLAGS=-DNDEBUG for libxcb. What do you guys
think?

I'll also start making the Xorg-7.2 updates soon. Just have been busy
at work and travelling.

--
Dan

--

-- 

Re: XOrg-7.2 - libXcb

Dan Nicholson wrote:
> I think we should just push in XCB, but disable the assertions since I
> don't think we can track down all these issues yet. Attached patch is
> what I'm sitting on + CPPFLAGS=-DNDEBUG for libxcb. What do you guys
> think?

You'll get bashed by Xorg developers if you do this. Without XCB, we can at 
least pretend that we are just lazy.

>  848b80f77b20ae1fa5c882bbfa531ebc  libX11-1.1.1.tar.bz2

This contains a known idle-spin bug when compiled against XCB: 
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=415541 - and note that this 
is not a locking bug that can be worked around with -DNDEBUG.

Also you have libXi-1.0.2. While this will work with -DNDEBUG, I don't 
recommend this version.

-- 
Alexander E. Patrakov
--

-- 
http://linuxfromscratch.org/mailman/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Re: XOrg-7.2 - libXcb

On 7/2/07, Alexander E. Patrakov <patrakov <at> ums.usu.ru> wrote:
> Dan Nicholson wrote:
> > I think we should just push in XCB, but disable the assertions since I
> > don't think we can track down all these issues yet. Attached patch is
> > what I'm sitting on + CPPFLAGS=-DNDEBUG for libxcb. What do you guys
> > think?
>
> You'll get bashed by Xorg developers if you do this. Without XCB, we can at
> least pretend that we are just lazy.

I think it was the XCB developers to be exact, but same difference. I
understand their argument, but the alternative isn't much better. I
just read that Skype has the same locking issues. What would I do in
that case? Even if I could convince the Skype people there was a bug,
I'd have to wait until they made another release. And who knows when
that would be.

> >  848b80f77b20ae1fa5c882bbfa531ebc  libX11-1.1.1.tar.bz2
>
> This contains a known idle-spin bug when compiled against XCB:
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=415541 - and note that this
> is not a locking bug that can be worked around with -DNDEBUG.
>
> Also you have libXi-1.0.2. While this will work with -DNDEBUG, I don't
> recommend this version.

Well, why don't I work on the 7.2 updates first? libX11-1.0.2 needs a
patch, right? Looks like you've got it on the livecd.

http://wiki.linuxfromscratch.org/livecd/browser/trunk/packages/Xorg-lib/libX11-1.1.2-bug11222-1.patch

Re: XOrg-7.2 - libXcb

Dan Nicholson wrote:

> I think it was the XCB developers to be exact, but same difference. I
> understand their argument, but the alternative isn't much better. I
> just read that Skype has the same locking issues. What would I do in
> that case? Even if I could convince the Skype people there was a bug,
> I'd have to wait until they made another release. And who knows when
> that would be.

Then, providing a patch that skips assertions conditionally, if an 
environment variable is set, would be better - at least, it won't exclude 
LFS users from potential bug reporters. The patch is (however, badly 
damaged) at http://lists4.opensuse.org/opensuse-commit/2007-03/msg00176.html

Also please make it clear that this is a workaround and that the bugs are 
not in libxcb, but in applications.

> Well, why don't I work on the 7.2 updates first? libX11-1.0.2 needs a
> patch, right?

Yes. AFAIK, this and

http://gitweb.freedesktop.org/?p=xorg/lib/libX11.git;a=commitdiff;h=c2f88cdf5cd9c94b77e5bfdac572b5ac06ab4aa8 
are the only known brown paper bag bugs in this version.

-- 
Alexander E. Patrakov
--

-- 
http://linuxfromscratch.org/mailman/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html

Re: XOrg-7.2 - libXcb

Alexander E. Patrakov wrote these words on 07/02/07 09:46 CST:
> Dan Nicholson wrote:
>
> Also please make it clear that this is a workaround and that the bugs are 
> not in libxcb, but in applications.
> 
>> Well, why don't I work on the 7.2 updates first? libX11-1.0.2 needs a
>> patch, right?
> 
> Yes. AFAIK, this and 
>
http://gitweb.freedesktop.org/?p=xorg/lib/libX11.git;a=commitdiff;h=c2f88cdf5cd9c94b77e5bfdac572b5ac06ab4aa8 
> are the only known brown paper bag bugs in this version.

I'm trying to follow this thread, but I'm a bit confused as to what
exactly is broken. Is the bottom line that the BLFS instructions for
installing Xorg-7.2 are broken and shouldn't be used?

Or is eliminating libxcb a solution that works and has no affect on
other packages down the road?

I'm asking because on the Xorg site, it says that libxcb is now an
official part of Xorg-7.2, but we aren't installing it.

Can anyone install applicable patches to
http://www.linuxfromscratch.org/patches/downloads/xorg/ that should
be used in a BLFS installation with libxcb and one to fix the other
bug Alex is speaking of? Is there such patches available already?

Sorry for my ignorance about this, but I would like to get a 7.2

Re: XOrg-7.2 - libXcb

On 7/3/07, Randy McMurchy <randy <at> linuxfromscratch.org> wrote:
> Alexander E. Patrakov wrote these words on 07/02/07 09:46 CST:
> > Dan Nicholson wrote:
> >
> > Also please make it clear that this is a workaround and that the bugs are
> > not in libxcb, but in applications.
> >
> >> Well, why don't I work on the 7.2 updates first? libX11-1.0.2 needs a
> >> patch, right?
> >
> > Yes. AFAIK, this and
> > http://gitweb.freedesktop.org/?p=xorg/lib/libX11.git;a=commitdiff;h=c2f88cdf5cd9c94b77e5bfdac572b5ac06ab4aa8
> > are the only known brown paper bag bugs in this version.
>
> I'm trying to follow this thread, but I'm a bit confused as to what
> exactly is broken. Is the bottom line that the BLFS instructions for
> installing Xorg-7.2 are broken and shouldn't be used?

Sure, I'll be glad to have another opinion here.

Our instructions are fine (besides a couple versions Alexander pointed
out that have bugs and should be updated), they're just not using the
full official way of libX11 on libxcb.

> Or is eliminating libxcb a solution that works and has no affect on
> other packages down the road?
>
> I'm asking because on the Xorg site, it says that libxcb is now an
> official part of Xorg-7.2, but we aren't installing it.

Re: XOrg-7.2 - libXcb

Dan Nicholson wrote:
> Hope that sums things up.

Yes, thats a good summary.

> You have an opinion on the matter? Keep in
> mind that those last patches that Alexander pointed to are a separate
> issue that will be addressed soon. Using the libX11 in the book right
> now should be fine.

They are not exactly a separate issue. My point was that neither myself nor 
Dan made a proper job on eliminating XCB-related assertions from Xorg 
libraries (I am not talking about external applications). My own personal 
opinion is that the Xorg team suggestion to use XCB with Xorg-7.2 is a bit 
irresponsible in this situation.

If the only problem were external applications, I would not oppose XCB 
inclusion that much.

-- 
Alexander E. Patrakov
--

-- 
http://linuxfromscratch.org/mailman/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Re: XOrg-7.2 - libXcb

Randy McMurchy wrote:

> I'm trying to follow this thread, but I'm a bit confused as to what
> exactly is broken. Is the bottom line that the BLFS instructions for
> installing Xorg-7.2 are broken and shouldn't be used?
> 
> Or is eliminating libxcb a solution that works and has no affect on
> other packages down the road?

Eliminating libxcb just means staying with the old and tried code.

However, the existing Xorg-7.2 instructions should not be used, for reasons 
unrelated to XCB. E.g., accelerated indirect rendering (AIGLX, needed for 
beryl and compiz) won't work, because Mesa and Xorg disagree on the location 
of DRI drivers. There may be other directory mismatches.

The LiveCD deviates a lot from the book and should not be treated as a proof 
that it is correct. Moreover, it has its own bugs, probably caused by such 
deviations (e.g., Xnest doesn't regenerate properly after the last client 
exits).

-- 
Alexander E. Patrakov
--

-- 
http://linuxfromscratch.org/mailman/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Re: XOrg-7.2 - libXcb

Alexander E. Patrakov wrote:
> ...the existing Xorg-7.2 instructions should not be used, for reasons 
> unrelated to XCB. E.g., accelerated indirect rendering (AIGLX, needed for 
> beryl and compiz) won't work, because Mesa and Xorg disagree on the location 
> of DRI drivers...
>
>   
Apologies for deviating from the thread-topic, Alexander, but as someone 
who's interested in installing Beryl, the above passage has me a bit 
confused. The instructions in the development book for MesaLib-6.5.2 & 
Xorg-Server-1.2.0 appear to use configuration switches to correlate the 
location of the drivers between the two. If that is incorrect or 
insufficient to get a working AIGLX-capable server, are there other or 
alternate steps that need to be taken?

Thanks in advance, taipan

--

-- 
http://linuxfromscratch.org/mailman/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page