On Sat, Sep 25, 2004 at 07:14:14PM -0600, Marc Aurele La France wrote:On Sat, 25 Sep 2004, Phil Dibowitz wrote:I think I might have located the problem. The server is not reporting the correct drawable in a completion event requested by an Xinerama-fied XShmPutImage request. The attached is a possible fix for that.
That works, but I think there's just a typo in the original code that keeps it from working the way it was intended. FOR_NSCREENS should have been FOR_NSCREENS_BACKWARD. The idea was that it counts down and sends the event on the last one (index zero) which had the correct drawable. Alot of functions use the BACKWARD version for this an similar reasons.
OK, thanks.
There are two instances of FOR_NSCREENS in the original code -- which one is supposed to be changed?
The one in ProcPanoramiXShmPutImage().
And are you saying that changing FOR_NSCREENS will fix the problem alone, or that the patch is also needed?
The strace doesn't show enough of each packet read by the server to know for certain if the application requested events in its original XShmPutImage(), so I don't know for certain that this will fix the problem. But the fix is a definite candidate.
I'm really sorry for the huge delay here - work got hellish and I didn't have time to keep testing this.
Anyway, since neither patch worked, the last thing I was asked (and I can't find that email) was to get an strace with it working (no xinerama) and it not working (xinerama). I've done that.
Here they are: http://sol.usc.edu/~phil/x.strace.noxin.bz2 (working) http://sol.usc.edu/~phil/x.strace.xin.bz2 (X freezes)
Again, I'm sorry for the delay... things have calmed down and I should be able to test any patches and get feedback in a normal turn around time again.
I asked you to produce these traces to see if the client expected some reply to its XShmPutImage calls. The non-Xinerama trace shows that it doesn't, so it's possible that the XShmPutImage angle is just a red herring.
Is there something specific I can use to duplicate the problem? You mentionned earlier something about BlackBoard 6.
Marc.
+----------------------------------+-----------------------------------+ | Marc Aurele La France | work: 1-780-492-9310 | | Computing and Network Services | fax: 1-780-492-1729 | | 352 General Services Building | email: tsi@xxxxxxxxxxx | | University of Alberta +-----------------------------------+ | Edmonton, Alberta | | | T6G 2H1 | Standard disclaimers apply | | CANADA | | +----------------------------------+-----------------------------------+ XFree86 developer and VP. ATI driver and X server internals. _______________________________________________ XFree86 mailing list XFree86@xxxxxxxxxxx http://XFree86.Org/mailman/listinfo/xfree86
