Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!think!mit-eddie!genrad!decvax!mcnc!rti-sel!dg_rtp!meissner
From: meissner@dg_rtp.UUCP (Michael Meissner)
Newsgroups: net.unix
Subject: Re: why is fprintf(stderr,) slow?
Message-ID: <549@dg_rtp.UUCP>
Date: Mon, 1-Sep-86 13:04:29 EDT
Article-I.D.: dg_rtp.549
Posted: Mon Sep  1 13:04:29 1986
Date-Received: Wed, 3-Sep-86 20:49:21 EDT
References: <633@hropus.UUCP> <1480@poseidon.UUCP> <7074@utzoo.UUCP> <338@chronon.chronon.UUCP>
Reply-To: meissner@dg_rtp.UUCP (Michael Meissner)
Organization: Data General (Languages @ Research Triangle Park, NC.)
Lines: 19
Summary: Setvbuf on System V.2 and ANSI X3J11

In article <338@chronon.chronon.UUCP> eric@chronon.UUCP (Eric Black) writes:
>
>Of course, the reason that it was made per character is to make sure that
>the error output in fact gets out, and is not lost in stdio buffering somewhere
>when an errant program suddenly gives up the ghost.  On BSD systems,
>at least, you can change this with the setlinebuf(3S) call, to make
>stderr use buffering.  Nowadays, with buffered stdio output less likely
>to get lost (?), perhaps it is too bad that the default is still
>to make a system call per character, rather than per call, but...
>
>On non-BSD systems you're stuck, as far as I know.

On System V.2, V.3, and the ANSI X3J11 draft you can use setvbuf (which allows
to set both the buffersize and a choice between no/line/full buffering).  Also,
I believe a pure System V.2 system buffers per call to *printf instead of per
character (but I'm not sure).

	Michael Meissner, Data General
	...{ decvax, ucbvax }!mcnc!rti-sel!dg_rtp!meissner