On Mon, Sep 28, 2015 at 4:00 PM, Florian Westphal wrote:
> Tom Herbert wrote:
>> RFC6296 doesn't work because it allows an invalid checksum to be sent
>> on wire relative to the addresses used on the wire. That means we
>> would lose CHECKSUM_UNNECESSARY for ILA which is way too big of a
>> perfo
Tom Herbert wrote:
> RFC6296 doesn't work because it allows an invalid checksum to be sent
> on wire relative to the addresses used on the wire. That means we
> would lose CHECKSUM_UNNECESSARY for ILA which is way too big of a
> performance hit.
Not following. I did not say you should use NPT in
On Sun, Sep 27, 2015 at 1:10 AM, Florian Westphal wrote:
> David Miller wrote:
>> > This patch set addresses the issue for ILA by adding a fast locator
>> > lookup that occurs before early demux. This is done by using a hook
>> > at NF_INET_PRE_ROUTING. For the backend we implement an rhashtable
David Miller wrote:
> > This patch set addresses the issue for ILA by adding a fast locator
> > lookup that occurs before early demux. This is done by using a hook
> > at NF_INET_PRE_ROUTING. For the backend we implement an rhashtable
> > that contains identifier to locator to mappings. The table
From: Tom Herbert
Date: Thu, 24 Sep 2015 09:30:20 -0700
> This patch set addresses the issue for ILA by adding a fast locator
> lookup that occurs before early demux. This is done by using a hook
> at NF_INET_PRE_ROUTING. For the backend we implement an rhashtable
> that contains identifier to lo
In the current implementation of ILA, LWT is used to perform
translation on both the input and output paths. This is functional,
however there is a big performance hit in the receive path. Early
demux occurs before the routing lookup (a hit actually obviates the
route lookup). Therefore the stack c
