Re: [PATCH] ipv4: only produce one record in fib_seq_show

From: Arnaldo Carvalho de Melo (acme@conectiva.com.br)
Date: Sun Oct 20 2002 - 00:08:49 EST


Em Sat, Oct 19, 2002 at 09:13:07PM -0700, David S. Miller escreveu:
> From: Arnaldo Carvalho de Melo <acme@conectiva.com.br>
> Date: Sat, 19 Oct 2002 22:03:31 -0300
    
> We'll, if everybody stopped using net-tools and started using iproute2 the
> world would be a better place

> 'iproute2' would need to be able to obtain things, such as
> the snmp statistics, some other way. Currently /proc/net/snmp
> is the only place these values are provided.

> Ditto for UDP socket listings et al. Only tcp_diag has moved
> the TCP socket information into the non-proc realm of netlink.

> If these facilities existed already, I'd agree with you.
> :-)

Hey, but those can be converted, can't they? I for one would be willing to
spend the time and study this to implement it. Indeed I'll do that, unless you
object, of course.

And take a look at this:

CONFIG_PROC_FS=y
[acme@oops net-2.5]$ l net/ipv4/built-in.o
-rw-rw-r-- 1 acme acme 328783 Out 20 01:44 net/ipv4/built-in.o

CONFIG_PROC_FS=n
[acme@oops net-2.5]$ l net/ipv4/built-in.o
-rw-rw-r-- 1 acme acme 320708 Out 20 02:03 net/ipv4/built-in.o

Both with CONFIG_SMP, CONFIG_NR_CPUS=2, so almos whooping 2 pages! Almost
one third of what CONFIG_SECURITY would add! ia32! Imagine on Sparc64! 8-P

- Arnaldo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Wed Oct 23 2002 - 22:00:49 EST