Re: zero-copy TCP fileserving

Richard B. Johnson (root@chaos.analogic.com)
Thu, 3 Jun 1999 18:36:35 -0400 (EDT)

On Thu, 3 Jun 1999, Bjorn Wesen wrote:
[SNIPPED]
>
> Speaking of checksumming outgoing fragments in HW - it is trivial to make
> the HW calculate the checksum itself, but I see a problem with having it
> inserting it in the stream - mainly because the checksum field passes
> through the HW _before_ the data it is supposed to checksum. How is this
> solved in the HW that can do outgoing checksumming ? Does it have a FIFO
> large enough to keep an MTU (and manipulate the header) ?
>

I don't know. I do know that the normal CRC is not just appended in the
NE* SNIC. You can 'remote DMA' data both ways and when you get it back
it doesn't have the CRC. Several years ago I tried for a long time to
be able to read the CRC of the packet I was about to send. We use these
things are RF modulators in a UWB communications system.

It looks as though the normal CRC is appended during the actual transmit.
If the offset of the checksum was known to a controller, it could fill in
the checksum using the same mechanism.

Cheers,
Dick Johnson
***** FILE SYSTEM WAS MODIFIED *****
Penguin : Linux version 2.2.6 on an i686 machine (400.59 BogoMips).
Warning : It's hard to remain at the trailing edge of technology.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/