Date: Sat, 11 Jan 2003 06:02:12 +1100 (EST) From: Bruce Evans <bde@zeta.org.au> To: Tony Finch <dot@dotat.at> Cc: freebsd-net@FreeBSD.ORG Subject: Re: Docco patch for review Message-ID: <20030111053338.K12262-100000@gamplex.bde.org> In-Reply-To: <20030110160714.D32571@chiark.greenend.org.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 10 Jan 2003, Tony Finch wrote:
> There seems to be a slight lacuna in the signal documentation:
>
> --- src/lib/libc/gen/signal.3 19 Dec 2002 09:40:21 -0000 1.32
> +++ src/lib/libc/gen/signal.3 10 Jan 2003 16:03:16 -0000
> @@ -187,7 +187,9 @@
> and during a
> .Xr ioctl 2
> or
> -.Xr wait 2 .
> +.Xr wait 2
> +or any other system call whose documentation says it may return
> +.Er EINTR .
This is not correct, since the relevant system calls are those that
are affected by SA_RESTART, not the ones that may return EINTR.
select(2) is the canonical example of a syscall that returns EINTR
but is not affected by SA_RESTART.
Similarly for sigaction.2.
The bug is more that the effect of SA_RESTART isn't documented in the
individual man pages for important syscalls like read(2). It is only
documented in:
fcntl.2.gz:.Dv SA_RESTART
nanosleep.2.gz:.Dv SA_RESTART
sigaction.2.gz:.It Dv SA_RESTART
sigaction.2.gz:.Dv SA_RESTART
sigaction.2.gz:.Dv SA_RESTART
wait.2.gz:.Dv SA_RESTART
wait3.2.gz:.Dv SA_RESTART
wait4.2.gz:.Dv SA_RESTART
waitpid.2.gz:.Dv SA_RESTART
SA_RESTART actually applies to almost all syscalls (although it can only
affect ones that return EINTR), so it would be much easier to document
the ones that it doesn't affect. I could only find select(), poll(),
aio_waitcoomplete(), nanosleep(), sigsuspend(), connect(), kevent() and
open().
Bruce
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-net" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030111053338.K12262-100000>