Re: updates?

[christos%zoulas.com@localhost (Christos Zoulas)]

On Jul 23,  8:03pm, charles.cui1984%gmail.com@localhost (Charles Cui) wrote:
-- Subject: Re: updates?

| Hi Christos,
| 
|     I have attached a program to test the delivery order of signals.
| It turns out that both Linux and our current implementation use a FIFO
| policy to deliver signal, (1. only realtime signals and 2. mix of realtime
| signals and standard signals)
| I do not know how FreeBSD handles it.
| Let me know if there are any problems.

Ok good start.  Let's add more things in the queue (or split it into
separate tests) to show that:
	- realtime signals are ordered properly
	- you can queue more than one realtime signal
	- don't use sighold/sigrelse. these are obsolete and release
	  only one signal at a time. use sigprocmask (plus since you are
	  releasing them one at a time, the release order is the order
	  you are going to see them delivered.

Then we can turn it into an ATF test.

Thanks,

christos