POSIX Signals (reposting)

Steven Suson (suson@tti.TuckerEnergy.com)
Tue, 03 Feb 1998 11:02:13 -0600

I have a couple of suggestions concerning POSIX signals. As far as I
can tell (I don't have an anywhere near recent version of the POSIX
spec.), no behavior is specified for the PID argument to sigqueue.
However, since the description of sigqueue and kill reside together in
O'Reilly's POSIX.4 book (out of date itself, I suppose), seems to imply
that the PID argument would behave similiarly. So, perhaps it would be
possible to add a configuration option (read compile-time option), that
would allow sigqueueinfo to do either a kill_something_info OR a
kill_proc_info. The kill_something_info seems to work just fine in the
2.1.78 kernel, and we beat it to death!

Also, would it be possible to modify the sigaction structure so that
whenever a SA_SIGINFO type handler is installed, it doesn't have to be
cast? Such as (this is from the 2.1.43 Signal patch):

struct sigaction {

union {
__sighandler_t _handler;
void (*_sigaction)(int, siginfo_t *, void *);
} _sa_un;
#define sa_handler _sa_un._handler
#define sa_sigaction _sa_un._sigaction
etc...

Also, I noticed a patch for signal.c going by. I have a problem,
where it seems that a masked (RT) signal (37 to be precise), arrives
anyway in two of my three threads (yea, its a multithreaded process).
The mask is set prior to ANY pthread calls.

And, thanks again.... It looks like we'll make our deadline.

BTW, where did CLONE_PID and signals land? Has a behavior been
decided upon?

Steve Suson
"Keep the faith."