21143 updates, and Re: large number of TX packets on ftp

[Peter Stein]

>> (it appears /dev/null is not diskless).

>What is that supposed to mean?  Of course accessing /dev/null does not
>cause disk activity.

Oh yes it does. This anomaly is even pointed out in some of the linux
documentation. But I'm not trying to be argumentative. Please read what
follows.

>In fact here are the exact functions used to implement the read and write 
>syscalls for /dev/null (in drivers/char/mem.c)

>static ssize_t read_null(struct file * file, char * buf,
>                         size_t count, loff_t *ppos)
>{
>        return 0;
>}
>
>static ssize_t write_null(struct file * file, const char * buf,
>                          size_t count, loff_t *ppos)
>{
>        return count;
>}

Guess what? My mem.c contains different code! Here are my mem.c functions:

static int read_null(struct inode * node, struct file * file,
                     char * buf, int count)
{
        return 0;
}

static int write_null(struct inode * inode, struct file * file,
                      const char * buf, int count)
{
        return count;
}

Clearly a hint that things are different on our systems. Your versions
look much newer and I suspect in your distribution this problem was
fixed. My code base is RedHat 4.2 upgraded from 4.1 and I assumed all
the code is fairly current, but apparently this is not the case. Which
distribution did your mem.c come from?

Peter Stein
nbi@xnet.com