Hi!
Can anyone explain this? He's got 2.4.2 with libc6 2.2.x. df -k returns the
correct results for his partition. Both statfs() and statvfs() returns bogus
values. From his last email it looks like it's actually returning stats for
/dev/hda8 instead of /dev/hda9.
Is this a (known) bug?
Here is the source for the program mentioned in the email:
------------< snip <------< snip <------< snip <------------
root@crystal:~# cat /tmp/t.c
#include <stdio.h>
#include <sys/vfs.h>
#include <sys/statvfs.h>
static const char part[] = "/dev/hda9";
int main ()
{
struct statfs stat;
struct statvfs vfs;
if (statfs (part,&stat) < 0) {
perror ("statfs");
return(1);
}
printf("f_blocks: %ld, f_bsize: %d\n",stat.f_blocks,stat.f_bsize);
if (statvfs (part,&vfs) < 0) {
perror ("statvfs");
return(1);
}
printf("f_bsize: %lu, f_blocks: %lu\n",vfs.f_bsize,vfs.f_blocks);
return(0);
}
------------< snip <------< snip <------< snip <------------
--Regards Abraham
Avoid Quiet and Placid persons unless you are in Need of Sleep. -- National Lampoon, "Deteriorata"
__________________________________________________________ Abraham vd Merwe - 2d3D, Inc.
Device Driver Development, Outsourcing, Embedded Systems
Cell: +27 82 565 4451 Snailmail: Tel: +27 21 761 7549 Block C, Aintree Park Fax: +27 21 761 7648 Doncaster Road Email: abraham@2d3d.co.za Kenilworth, 7700 Http: http://www.2d3d.com South Africa
attached mail follows:
> allen@eagle:~$ uname -a > Linux eagle.spi.co.za 2.4.2-2 #1 Sun Apr 8 20:41:30 EDT 2001 i686 unknown
I just had a look at the source for the df from fileutils 4.0.37 and it can get the filesystem stats in a number of ways
1. statfs() 2. reading a filsys structure directly from the superblock of the filesystem. 3. statvfs()
I've modified my test program to display both. See if you get different output. Also try compiling the program with -D_LARGEFILE64_SOURCE and see if that makes a difference (shouldn't since your partition is quite small).
< < I have also played with options 1 and 3, using statfs() and statvfs() which < both display the same information. < Compiling t.c with -D_LARGEFILE64_SOURCE doesn't make a differance < as you thought. I was hoping not to get involved with option 2 for now, < since time does not permit and I'm on the verge of execing("df -k").
(Which reminds me. How big is your partition really? I take it df gave the correct output which makes it a 2.8G partition?)
< Yep, its quite small. I have to re-install this machine this weekend and that's < one of the things I need to fix. :)
Also, what filesystem do you have on there?
< ext2fs is all I use on this machine.
If this is the same I give up. In that case I suggest you grab a copy of the latest fileutils, compile it with debugging and run df through the debugger. (break on get_fs_usage and step through the function to see what method it uses on your system).
< yes it is the same. I will check out fileutils. < yet I saw something very interesting...hmmm.... < don't understand it, but all the programs (we) wrote < did return valid values, but for / (root) partition, which < is on /dev/hda8 (even tho I explicitly tested with /dev/hda9) < see this:
gcc t.c (your latest t.c) allen@eagle:~$ ./a.out f_blocks: 350021, f_bsize: 1024 f_bsize: 1024, f_blocks: 350021
allen@eagle:~$ ./fs /dev/hda /dev/hda: total blocks 350021 free blocks 217464
allen@eagle:~$ ./fs /dev/hda9 /dev/hda9: total blocks 350021 free blocks 217458
allen@eagle:~$ df -k Filesystem 1k-blocks Used Available Use% Mounted on /dev/hda8 350021 132578 199372 40% / /dev/hda9 3028080 2575624 298636 90% /usr ...
What can you make of it?
-Allen
-- Clug-tech mailing list Clug-tech@clug.org.za To (un)subscribe: http://www2.clug.org.za/mailman/listinfo/clug-tech
This archive was generated by hypermail 2b29 : Tue Jul 23 2002 - 22:00:42 EST