Subject: Re: [need review] pax(1) change to fix PR #8868
To: None <tech-userlevel@netbsd.org>
From: Todd Whitesel <toddpw@best.com>
List: tech-userlevel
Date: 02/10/2000 23:55:11
> The cause of the problem is that if the SIGINFO is caught
> in the middle of write(2) system call, write() does not
> continue the operation after the signal and returns with
> partial operation.

Is there some reason why we cannot enable "SA_RESTART" on the SIGINFO
signal, so the kernel takes care of this automatically for read() and
write() ?

Todd Whitesel
toddpw @ best.com