NetNews Usenet Archive 1992 #31/NN_1992_31.iso/spool/comp/os/vms/19923

Internet Message Format | 1992-12-27 | 2.0 KB

Path: sparky!uunet!cs.utexas.edu!sun-barr!sh.wide!wnoc-tyo-news!sranha!anprda!akira
From: [e-mail] (Akira Takiguchi)
Newsgroups: comp.os.vms
Subject: Re: flushing in VAXC and DECC
Message-ID: <[e-mail]>
Date: 27 Dec 92 11:37:49 GMT
References: <[e-mail]> <[e-mail]> <[e-mail]>
Organization: ATSON, Inc. (a subsidiary of the Asahi Shimbun)
Lines: 30

In article <[e-mail]> [e-mail] writes:
>Glad to know that it's documented on AXP machines.  Now, perhaps you'd clarify
>a point for us:  The AXP documentation doesn't agree with what fsync() acutally
>does under VMS 5.4-2.  In particular, the AXP documentation would have us
>believe that:
>    fsync(fileno(stdout));
>actually causes the results of, e.g., all printf's to be flushed to disk.

     I don't see any reason why anyone is forced to "believe" that standard
I/O and UNIX I/O are interchangable (is there any explicit documentaion in
the VAXC manual?  I guess no).  Yes, fileno() can convert file pointer
to file descriptor.  But that doesn't mean you can mix up standard I/O
functions and UNIX I/O functions.  If you are likely to believe that
fsync(fileno(stdout)) without preceding fflush() will work, wouldn't you
also assume that

     fp=fopen("test.dat","w");fputs("aaa\n",fp);close(fileno(fp));
     fp=fopen("test.dat","w");fputs("aaa\n",fp);fclose(fp);

do the same?  (on VMS, they do - unfortunately.  But NEVER on Unix boxes).

     Fsync takes a file descriptor as argument, which means it's a UNIX I/O
function (fsync has long been on unix boxes - BSD origin, I think).  Anybody
who use UNIX I/O function in his program are supposed to know that you need
special care to use it with standard I/O functions.  I wouldn't blame AXP's
documentation.
--
       Akira Takiguchi  at ATSON, Inc. (a subsidiary of the Asahi Shimbun)
                   WAKO GINZA bldg.  8-10-4 Ginza Chuo-ku Tokyo 104  Japan
   [phone](voice) 7066(fax)  EMAIL TO <[e-mail]>