netdev - Re: Ethernet CRC questions

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20091230064324.GJ3486@mea-ext.zmailer.org>
Date:	Wed, 30 Dec 2009 08:43:24 +0200
From:	Matti Aarnio <matti.aarnio@...iler.org>
To:	thomas yang <lampsu@...il.com>
Cc:	netdev@...r.kernel.org
Subject: Re: Ethernet CRC questions

On Wed, Dec 30, 2009 at 01:53:09PM +0800, thomas yang wrote:
> 2009/12/30 Ben Hutchings <bhutchings@...arflare.com>:
> > On Wed, 2009-12-30 at 01:17 +0800, thomas yang wrote:
> >> How does Linux  generate Ethernet CRC ?
> >> by the hardware (Ethernet controller)  or  use crc32_le() /
> >> crc32_be()     ( lib/crc32.c )  ?
> >>
> >> P.S.   my  network card is  rtl8139d  , use  8139too driver
> >
> > The Ethernet frame CRC is normally generated and verified in hardware.
> > These library functions are needed for the few MACs that don't, or where
> > the same CRC function is used for some other purpose (e.g. as a hash
> > function for multicast filtering).
> 
> Does the Ethernet frame CRC (generate and verify in hardware)  make
> the TX / RX  rate  a little smaller , and a longer end-to-end  delay ?
>  Could I disable CRC appending / verifying  in the network card driver
>   on all of my nodes (machines) ?

You would need to similarly disable the checking and generating at your
network switches as well.  It really is not worthwhile.

You are better to use multiple links in between the machines in parallel,
and bond them together as load-sharing link.  Or go to 10G.
(We used to do that kind of things with quad fast-ethernet cards, but
with gigaethers we did not bother anymore.)

> Thanks.

Best Regards,  Matti Aarnio
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html