Testing whether Xlib threading support works in XORG-RELEASE-1branch...

[Jim Gettys]

I guess the answer is that little active testing gets done;
this is not a healthy state. My point is just that Mozilla is
intimidating enough to have inhibited its use in this area.

This may no longer be a viable concern, as mozilla's
release engineering has improved greatly and it can be built
by mere mortals these days. 

When Keith and I were burned by it, it required a Netscape/mozilla
wizard, and our last experience
was frustrating and resulted in just determining that the
hang we were seeing was a Netscape bug, and that no one at
Netscape was interested in helping track that bug down at the time.

We can easily believe there are bugs....
                  - Jim


On Tue, 2004-03-23 at 18:47, Roland Mainz wrote:
> Christopher Blizzard wrote:
> > >While I don't doubt there could be problems, you should
> > >be aware in the past that Mozilla's threading itself was badly
> > >broken (IIRC, at the time wasn't even actually calling Xlib's thread
> > >initialization at all; this was in late Netscape days).  I
> > >remember a very frustrating debugging session on the
> > >phone with Keithp where we ended up having to conclude we
> > >couldn't fix it (it being Netscape, without source available),
> > >and that the fault was 100% Netscape's.
> >
> > Mozilla only does drawing from the main thread, so it's not like we're
> > even pushing that code very hard.  Mozilla's threading doesn't have any
> > serious issues that I know of right now.  I don't know what roland is
> > seeing. 
> 
> See my other reply here... I was never talking about bugs in Mozilla, I
> was interested that the X11 peopple test their Xlib and X extension
> libraries against a real-world application (e.g. mozilla) with Xlib
> threadsafe mode enabled to check whether this works now or not... in the
> past that part of Xlib was simply unuseable since the implementation
> assumed that it is safe to lock a mutex recursively. Xfree86 people did
> some work on that before V4.4.0 was released and my patch for Mozilla
> was mainly ([1]) to test whether the fixes are complete or whether more
> work is required.
> 
> [1]=there is the idea to run plugins in a seperate thread but that
> cannot be implemented in the near future (>=2 years) since the fixes
> have to spread first over all Unix/Linux distributions before we can
> even think to make any use of that.
> 
> ----
> 
> Bye,
> Roland
-- 
Jim Gettys <Jim.Gettys at hp.com>
HP Labs, Cambridge Research Laboratory