X-MimeOLE: Produced By Microsoft Exchange V6.5
Received: by onstor-exch02.onstor.net 
	id <01C7A7AD.E61D693B@onstor-exch02.onstor.net>; Tue, 5 Jun 2007 13:12:53 -0700
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C7A7AD.E61D693B"
Content-class: urn:content-classes:message
Subject: RE: TXRX network observations and questions
Date: Tue, 5 Jun 2007 13:12:53 -0700
Message-ID: <BB375AF679D4A34E9CA8DFA650E2B04E04020537@onstor-exch02.onstor.net>
In-Reply-To: <BB375AF679D4A34E9CA8DFA650E2B04E018A699D@onstor-exch02.onstor.net>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: TXRX network observations and questions
Thread-Index: Acenns+jZfseqkydTlq9Y9CHGG/5PgAAmlAwAADowhAAAU4HYA==
References: <BB375AF679D4A34E9CA8DFA650E2B04E04020463@onstor-exch02.onstor.net> <BB375AF679D4A34E9CA8DFA650E2B04E018A699D@onstor-exch02.onstor.net>
From: "Maxim Kozlovsky" <maxim.kozlovsky@onstor.com>
To: "Jeff Miller" <jeff.miller@onstor.com>,
	"dl-Cougar" <dl-Cougar@onstor.com>

This is a multi-part message in MIME format.

------_=_NextPart_001_01C7A7AD.E61D693B
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

The "various packet sizes" can easily be missed. I don't remember ever
needing to look at this statistic before.=20

_____________________________________________
From: Jeff Miller=20
Sent: Tuesday, June 05, 2007 12:27 PM
To: Maxim Kozlovsky; dl-Cougar
Subject: RE: TXRX network observations and questions

The LSI chip has counters for various packet sizes, six ranges each for
transmit and receive.  The 1480 doesn't have these.  I also think we
should we should use as much error detail as the hardware provides.
Most of the other differences are the types of errors counted and the
exact meaning of some of the counters.
These are the names of the BCM1480 counters:
R_MAC_RMON_TX_BYTES
R_MAC_RMON_COLLISIONS
R_MAC_RMON_LATE_COL
R_MAC_RMON_EX_COL
R_MAC_RMON_FCS_ERROR
R_MAC_RMON_TX_ABORT
R_MAC_RMON_TX_BAD
R_MAC_RMON_TX_GOOD
R_MAC_RMON_TX_RUNT
R_MAC_RMON_TX_OVERSIZE
R_MAC_RMON_RX_BYTES
R_MAC_RMON_RX_MCAST
R_MAC_RMON_RX_BCAST
R_MAC_RMON_RX_BAD
R_MAC_RMON_RX_GOOD
R_MAC_RMON_RX_RUNT
R_MAC_RMON_RX_OVERSIZE
R_MAC_RMON_RX_FCS_ERROR
R_MAC_RMON_RX_LENGTH_ERROR
R_MAC_RMON_RX_CODE_ERROR
R_MAC_RMON_RX_ALIGN_ERROR

_____________________________________________
From: Maxim Kozlovsky=20
Sent: Tuesday, June 05, 2007 11:37 AM
To: Jeff Miller; dl-Cougar
Subject: RE: TXRX network observations and questions

There is no strict requirement for the statistics that I know of.

The following should be supported of course in the new version:

Number of packets send
Number of packets received
Number of bytes send=20
Number of bytes received
Number of packet received with errors, broken down by error kind with as
much detail as the hardware supports.
Number of packet send with errors, broken down by error kind with as
much detail as the hardware supports.

Can you provide more details on what are the missing statistics? We
should decide on them on case by case basis.

The fragmented packets are expected for NFS over UDP only. We are
actively discouraging the customers from using UDP, most current OSs
default to NFS over TCP. I would not try too hard to make the UDP to
work as fast as TCP, software checksums should be fine initially.=20

The IP with options are not seen in the data path, so no optimizations
are necessary for those either.

There are no near plans for IPv6.

Max

_____________________________________________
From: Jeff Miller=20
Sent: Tuesday, June 05, 2007 11:25 AM
To: dl-Cougar
Subject: TXRX network observations and questions

A couple issues that I have uncovered while working on the TXRX network
driver. =20

Cougar uses the BCM1480 MAC instead of the LSI 8104 used on Bobcat and
Cheetah.  The LSI 8104 has several statistics counters that the BCM1480
doesn't have.  Is there any particular requirement for any statistics,
either through SNMP, nfxsh, or TXRX console?  If there is then we may
need to implement in software any missing pieces.

Cougar hardware can not checksum fragmented packets as they are
received.  In Bobcat and Cheetah the LUC would calculate partial
checksums for packet fragments and then software would use the partial
checksums to calculate the full packet checksum.  Cougar (as well as
Bobcat and Cheetah) hardware also does not handle checksums for IP
packets with options.  Does calculating checksums in software for these
cases have any significant performance impact?  Obviously this depends
on how many fragmented packets we see and how fast the software checksum
is.  The data mover DMA could also be used to calculate checksums.

We are not going to see IPv6 packets, right?

Any input on these?
Thanks,
Jeff

------_=_NextPart_001_01C7A7AD.E61D693B
Content-Type: text/html;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Dus-ascii">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
6.5.7652.24">
<TITLE>RE: TXRX network observations and questions</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/rtf format -->

<P DIR=3DLTR><SPAN LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT =
COLOR=3D"#000080" SIZE=3D2 FACE=3D"Arial">The</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"> <FONT COLOR=3D"#000080" =
SIZE=3D2 FACE=3D"Arial">&#8220;</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" =
SIZE=3D2 FACE=3D"Arial">various packet sizes</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" =
SIZE=3D2 FACE=3D"Arial">&#8221;</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" =
SIZE=3D2 FACE=3D"Arial"> can easily be missed</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" =
SIZE=3D2 FACE=3D"Arial">.</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 FACE=3D"Arial"> I =
don</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT =
COLOR=3D"#000080" SIZE=3D2 FACE=3D"Arial">&#8217;</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" =
SIZE=3D2 FACE=3D"Arial">t remember ever</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"> <FONT COLOR=3D"#000080" =
SIZE=3D2 FACE=3D"Arial">needing to</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"> <FONT COLOR=3D"#000080" =
SIZE=3D2 FACE=3D"Arial">look at this statistic =
before.</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"> =
</SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT =
SIZE=3D2 =
FACE=3D"Tahoma">_____________________________________________<BR>
</FONT></SPAN><SPAN LANG=3D"en-us"><B></B></SPAN><SPAN =
LANG=3D"en-us"><B><FONT SIZE=3D2 =
FACE=3D"Tahoma">From:</FONT></B></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Tahoma"> Jeff Miller<BR>
</FONT></SPAN><SPAN LANG=3D"en-us"><B></B></SPAN><SPAN =
LANG=3D"en-us"><B><FONT SIZE=3D2 =
FACE=3D"Tahoma">Sent:</FONT></B></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Tahoma"> Tuesday, June 05, 2007 =
12:27 PM<BR>
</FONT></SPAN><SPAN LANG=3D"en-us"><B></B></SPAN><SPAN =
LANG=3D"en-us"><B><FONT SIZE=3D2 =
FACE=3D"Tahoma">To:</FONT></B></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Tahoma"> Maxim Kozlovsky; =
dl-Cougar<BR>
</FONT></SPAN><SPAN LANG=3D"en-us"><B></B></SPAN><SPAN =
LANG=3D"en-us"><B><FONT SIZE=3D2 =
FACE=3D"Tahoma">Subject:</FONT></B></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT SIZE=3D2 =
FACE=3D"Tahoma"> RE: TXRX network observations and =
questions</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT =
COLOR=3D"#000080" SIZE=3D2 FACE=3D"Arial">The LSI chip has counters for =
various packet sizes, six ranges each for transmit and receive.&nbsp; =
The 1480 doesn&#8217;t have these.&nbsp; I also think we should we =
should use as much error detail as the hardware provides.&nbsp;&nbsp; =
Most of the other differences are the types of errors counted and the =
exact meaning of some of the counters.</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">These are the names of the BCM1480 =
counters:</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_TX_BYTES</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_COLLISIONS</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_LATE_COL</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_EX_COL</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_FCS_ERROR</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_TX_ABORT</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_TX_BAD</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_TX_GOOD</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_TX_RUNT</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_TX_OVERSIZE</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_RX_BYTES</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_RX_MCAST</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_RX_BCAST</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_RX_BAD</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_RX_GOOD</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_RX_RUNT</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_RX_OVERSIZE</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_RX_FCS_ERROR</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_RX_LENGTH_ERROR</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_RX_CODE_ERROR</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">R_MAC_RMON_RX_ALIGN_ERROR</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT =
SIZE=3D2 =
FACE=3D"Tahoma">_____________________________________________<BR>
</FONT></SPAN><SPAN LANG=3D"en-us"><B></B></SPAN><SPAN =
LANG=3D"en-us"><B><FONT SIZE=3D2 =
FACE=3D"Tahoma">From:</FONT></B></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Tahoma"> Maxim Kozlovsky<BR>
</FONT></SPAN><SPAN LANG=3D"en-us"><B></B></SPAN><SPAN =
LANG=3D"en-us"><B><FONT SIZE=3D2 =
FACE=3D"Tahoma">Sent:</FONT></B></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Tahoma"> Tuesday, June 05, 2007 =
11:37 AM<BR>
</FONT></SPAN><SPAN LANG=3D"en-us"><B></B></SPAN><SPAN =
LANG=3D"en-us"><B><FONT SIZE=3D2 =
FACE=3D"Tahoma">To:</FONT></B></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Tahoma"> Jeff Miller; =
dl-Cougar<BR>
</FONT></SPAN><SPAN LANG=3D"en-us"><B></B></SPAN><SPAN =
LANG=3D"en-us"><B><FONT SIZE=3D2 =
FACE=3D"Tahoma">Subject:</FONT></B></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT SIZE=3D2 =
FACE=3D"Tahoma"> RE: TXRX network observations and =
questions</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT =
COLOR=3D"#000080" SIZE=3D2 FACE=3D"Arial">There is no strict requirement =
for the statistics that I know of.</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">The following should be supported of course in the new =
version:</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">Number of packets send</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">Number of packets received</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">Number of bytes send </FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">Number of bytes received</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">Number of packet received with errors, broken down by =
error kind with as much detail as the hardware =
supports.</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">Number of packet send with errors, broken down by error =
kind with as much detail as the hardware supports.</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">Can you provide more details on what are the missing =
statistics? We should decide on them on case by case =
basis.</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">The fragmented packets are expected for NFS over UDP =
only. We are actively discouraging the customers from using UDP, most =
current OSs default to NFS over TCP. I would not try too hard to make =
the UDP to work as fast as TCP, software checksums should be fine =
initially. </FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">The IP with options are not seen in the data path, so no =
optimizations are necessary for those either.</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">There are no near plans for IPv6.</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT COLOR=3D"#000080" SIZE=3D2 =
FACE=3D"Arial">Max</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT =
SIZE=3D2 =
FACE=3D"Tahoma">_____________________________________________<BR>
</FONT></SPAN><SPAN LANG=3D"en-us"><B></B></SPAN><SPAN =
LANG=3D"en-us"><B><FONT SIZE=3D2 =
FACE=3D"Tahoma">From:</FONT></B></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Tahoma"> Jeff Miller<BR>
</FONT></SPAN><SPAN LANG=3D"en-us"><B></B></SPAN><SPAN =
LANG=3D"en-us"><B><FONT SIZE=3D2 =
FACE=3D"Tahoma">Sent:</FONT></B></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Tahoma"> Tuesday, June 05, 2007 =
11:25 AM<BR>
</FONT></SPAN><SPAN LANG=3D"en-us"><B></B></SPAN><SPAN =
LANG=3D"en-us"><B><FONT SIZE=3D2 =
FACE=3D"Tahoma">To:</FONT></B></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Tahoma"> dl-Cougar<BR>
</FONT></SPAN><SPAN LANG=3D"en-us"><B></B></SPAN><SPAN =
LANG=3D"en-us"><B><FONT SIZE=3D2 =
FACE=3D"Tahoma">Subject:</FONT></B></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT SIZE=3D2 =
FACE=3D"Tahoma"> TXRX network observations and =
questions</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT =
SIZE=3D2 FACE=3D"Arial">A couple issues that I have uncovered while =
working on the TXRX network driver.&nbsp; </FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Arial">Cougar =
uses the BCM1480 MAC instead of the LSI 8104 used on Bobcat and =
Cheetah.&nbsp; The LSI 8104 has several statistics counters that the =
BCM1480 doesn't have.&nbsp; Is there any particular requirement for any =
statistics, either through SNMP, nfxsh, or TXRX console?&nbsp; If there =
is then we may need to implement in software any missing =
pieces.</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Arial">Cougar =
hardware can not checksum fragmented packets as they are received.&nbsp; =
In Bobcat and Cheetah the LUC would calculate partial checksums for =
packet fragments and then software would use the partial checksums to =
calculate the full packet checksum.&nbsp; Cougar (as well as Bobcat and =
Cheetah) hardware also does not handle checksums for IP packets with =
options.&nbsp; Does calculating checksums in software for these cases =
have any significant performance impact?&nbsp; Obviously this depends on =
how many fragmented packets we see and how fast the software checksum =
is.&nbsp; The data mover DMA could also be used to calculate =
checksums.</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Arial">We are =
not going to see IPv6 packets, right?</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Arial">Any =
input on these?</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT SIZE=3D2 =
FACE=3D"Arial">Thanks,</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT SIZE=3D2 =
FACE=3D"Arial">Jeff</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"></SPAN></P>

</BODY>
</HTML>
------_=_NextPart_001_01C7A7AD.E61D693B--
