Subject: Re: new faster in_cksum.c
To: Chris Gilbert <cg110@york.ac.uk>
From: Ale Terlevich <A.I.Terlevich@durham.ac.uk>
List: port-arm32
Date: 04/30/1997 12:05:09
On Tue, 29 Apr 1997, Chris Gilbert wrote:

> 
> hi,
> 
> for those of you that build your own kernels and have a lot of inet stuff
> going on I've made a arm32 version of this.  It's based on the sparc
> version (in fact it's identical but for the asm parts)
> 
> If people who've got machines using inet a lot could test it out see if it
> is faster, and let me know if it works.
> 

  I'm afraid it seems to not work.

I compiled it as instructed (i.e. put it into arch/arm32/arm32 and 
changed files.arm32, nad when I rebooted the kernel, only ping worked.

A netstat -p tcp reveals that it is rejecting all incoming packest due to 
bad checksums....

  I also compiled the new ntohl.S and htonl.S in the same kernel, but as 
ping works, I assume these aren't the problem.

Ale.

tcp:
        7 packets sent
                0 data packets (0 bytes)
                0 data packets (0 bytes) retransmitted
                0 ack-only packets (0 delayed)
                0 URG only packets
                0 window probe packets
                0 window update packets
                7 control packets
        8 packets received
                0 acks (for 0 bytes)
                0 duplicate acks
                0 acks for unsent data
                0 packets (0 bytes) received in-sequence
                0 completely duplicate packets (0 bytes)
                0 old duplicate packets
                0 packets with some dup. data (0 bytes duped)
                0 out-of-order packets (0 bytes)
                0 packets (0 bytes) of data after window
                0 window probes
                0 window update packets
                0 packets received after close
                8 discarded for bad checksums
                0 discarded for bad header offset fields
                0 discarded because packet too short
....