All of lore.kernel.org
 help / color / mirror / Atom feed
From: Kurt Garloff <garloff@suse.de>
To: Andrea Arcangeli <andrea@suse.de>,
	Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: Linux kernel list <linux-kernel@vger.kernel.org>
Subject: Re: nfs MAP_SHARED corruption fix
Date: Wed, 9 May 2001 12:55:51 +0200	[thread overview]
Message-ID: <20010509125551.C420@garloff.suse.de> (raw)


[-- Attachment #1.1: Type: text/plain, Size: 544 bytes --]

Hi Andrea, Trond,

the demo for the NFS SHARED_MAP corruption:

garloff@daubechies:~/C $ uname -sr
Linux 2.4.4
garloff@daubechies:~/C $ ./test_nfs_shared_map ; head -1 ./testfile; sleep 10; head -1 ./testfile
Linux NFS rocks.
Linux NFS sucks.

Sources attached. I still have to test your fix, Trond.

Regards,
-- 
Kurt Garloff  <garloff@suse.de>                          Eindhoven, NL
GPG key: See mail header, key servers         Linux kernel development
SuSE GmbH, Nuernberg, FRG                               SCSI, Security

[-- Attachment #1.2: test_nfs_shared_map.c --]
[-- Type: text/plain, Size: 1197 bytes --]

/** test_nfs_shared_map.c
 *
 * Creates a file, expands it by ftruncate, mmaps it, writes
 * to the mapped memory, unmaps it and closes the file again.
 * 
 * This triggers a bug in 2.4.4 NFS client code: The file won't
 * contain the data written to the mapped memory.
 * 
 * (c) Kurt Garloff <garloff@suse.de>, 2001-05-09, GNU GPL
 */

#include <stdio.h>
#include <string.h>
#include <unistd.h>
#include <fcntl.h>
#include <sys/mman.h>
#include <errno.h>


const char * const bad  = "Linux NFS sucks.\n";
const char * const good = "Linux NFS rocks.\n";
const char * const name = "testfile";

#define LEN 4096

int die (const char* const txt)
{
	perror (txt);
	exit (errno);
}

int main ()
{
	char* adr; int err;
	int fd = open (name, O_RDWR | O_CREAT | O_TRUNC, 0644);
	if (fd <= 0) die ("create testfile");
	err = write (fd, bad, strlen (bad));
	close (fd); 
	truncate (name, LEN);
	sync ();
	fd = open (name, O_RDWR);
	if (fd <= 0) die ("open testfile");
	adr = (char*) mmap (0, LEN, PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0);
	if (!adr) die ("mmap failed");
	strcpy (adr, good);
	strcpy (adr+32, good);
#ifdef NEED_MSYNC
	msync (adr, LEN, MS_SYNC);
#endif
	munmap (adr, LEN);
	close (fd);
}

[-- Attachment #2: Type: application/pgp-signature, Size: 232 bytes --]

             reply	other threads:[~2001-05-09 10:59 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-05-09 10:55 Kurt Garloff [this message]
  -- strict thread matches above, loose matches on Subject: below --
2001-05-08 14:00 nfs MAP_SHARED corruption fix Andrea Arcangeli
2001-05-08 15:21 ` Trond Myklebust
2001-05-08 15:38   ` Kurt Garloff
2001-05-09  7:30     ` Trond Myklebust
2001-05-09 13:25       ` Andrea Arcangeli
2001-05-09 22:02       ` Marcelo Tosatti
2001-05-10  0:08         ` Andrea Arcangeli
2001-05-09 22:38           ` Marcelo Tosatti
2001-05-10  1:16             ` Andrea Arcangeli
2001-05-10  0:00               ` Marcelo Tosatti
2001-05-10 10:11                 ` Trond Myklebust
2001-05-10 10:14                 ` Trond Myklebust
2001-05-09  2:48   ` Andrea Arcangeli
2001-05-09  7:00     ` Trond Myklebust

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20010509125551.C420@garloff.suse.de \
    --to=garloff@suse.de \
    --cc=andrea@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=trond.myklebust@fys.uio.no \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.