Frank Van Der Linden wrote:
> Roland Mainz wrote:
> > Ok... what about the following change:
> > Signal bits 0-63 are used or reserved for "normal" signals and signals
> > from 64-95 are used by realtime signals. Anything signal numbers from
> > 96-128 are reserved.
> > This would allow 32 realtime signals (like in Linux) and avoid that they
> > have to move each time a signal gets added...
> > ... would that work ?
> >
> Bumping the number beyond 64 is more work than I'd like to do..

AFAIK this is unavoidable assuming we want to bump the number of
realtime signals to 32.

>From "/usr/include/sys/iso/signal_iso.h":
-- snip --
#define SIGLOST 37      /* resource lost (eg, record-lock lost) */
#define SIGXRES 38      /* resource control exceeded */
#define SIGJVM1 39      /* reserved signal for Java Virtual Machine */
#define SIGJVM2 40      /* reserved signal for Java Virtual Machine */

/* insert new signals here, and move _SIGRTM* appropriately */
#define _SIGRTMIN 41    /* first (highest-priority) realtime signal */
#define _SIGRTMAX 48    /* last (lowest-priority) realtime signal *
-- snip --

SIGJVM2+32 = 72

> also, as
> Casper noted, the realtime signals are the ones that are actually suited
> for moving, since SIGRTMIN and SIGRTMAX are defined as sysconf(2) calls.

Right... my idea was to avoid trouble for the kernel people, not the
userland consumers...

> The real issue here will be the value of NSIG, the number of signals.
> This value is not in any standard, and its use is discouraged. But, it
> is being used anyway, so its value will be compiled in to applications
> (though not many), and it will have to change.

I didn a quick look at the list
(http://src.opensolaris.org/source/search?n=25&start=0&refs=NSIG) ...
most of the entries "look" harmless but
http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/ucblib/libucb/port/gen/siglist.c#52
worries me a little bit (or maybe I am worrying too much - the table
doesn't care for realtime signals or SIGJVM&co. ... I have to dig a
little bit in this case...) ... ;-(

----

Bye,
Roland

-- 
  __ .  . __
 (o.\ \/ /.o) [EMAIL PROTECTED]
  \__\/\/__/  MPEG specialist, C&&JAVA&&Sun&&Unix programmer
  /O /==\ O\  TEL +49 641 7950090
 (;O/ \/ \O;)
_______________________________________________
opensolaris-code mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/opensolaris-code

Reply via email to