>>>>> "Anders" == anders henriksson <[log in to unmask]> writes:

[I am not sure that this is the appropriate list, but I'll answer
 anyway. If you want to continue this thread, maybe comp.lang.ada on
 Usenet would be a better choice.]

Anders> Problem description: We are developing a Gui, using TeleUse
Anders> and SL-GMS, with Ada (Ada83). The Gui is running in one
Anders> Solaris process and it seems that the combination of Ada-delay
Anders> and X.11 isn't any good idea!

Anders> Both X.11 and Ada-delay uses the signal SIGALRM from Solaris.
Anders> After X.11 catched an SIGALRM the SIGALRM is cleared. No
Anders> chance for the AdaRT to catch the signal.

Anders> Is there any solutions to this problem?  It seems that the
Anders> problem is X.11 (as usual)?

There is no "problem" in Ada or X11, the problem is that you are
trying to use an Ada run-time system which is not
SIGALRM-friendly. The fact that your Ada run-time catches SIGALRM can
hardly be seen as a "problem in X11 (as usual)". Using GNAT, I've
built applications that use delays as well as X11 and I have never
encountered such a behaviour.

You should try to compile your project with another compiler and see
if the same thing happens (GNAT is freely available from
ftp://cs.nyu.edu/pub/gnat/ for example).

Samuel Tardieu -- [log in to unmask]