On Sun, Dec 08, 2002 at 04:31:30PM +0100, Thomas Finneid wrote:

 > Ok, so this is why the ip processing is consuming so many clock cycles. I
 > have'nt gotten to the details of the ip part just yet. I thought the
 > checksum might have be done in the socket layer when copying the data from
 > the user process, but I could'nt find it there either. 

Right.  And, to be accurate, it's not like this makes IP processing "more
expensive", because the deferred checksum only happens if a higher-level
protocol set it up, and that higher-level protocol would have had to pay
the cost of the checksum anyway.  We are just deferring it until we know
whether or not it will be needed.

 > The page loan stuff(zero-copy path ?) in the socket layer, do I understand
 > it right in that it transfers ownership, temporarily, of the page from the
 > process to the kernel so as to avoid the data copying at that level? So the
 > data is only copied when transferred from memory to the NIC, 

Correct.  Note, in most cases, the "copy" that happens from memory->NIC
is DMA by the NIC itself, so the CPU is not directly involved in the data
movement.

-- 
        -- Jason R. Thorpe <thorpej@wasabisystems.com>