Messages in this thread | | | Date | Fri, 7 Aug 1998 02:16:02 -0400 (EDT) | From | <> | Subject | Re: Idea to make a kickass TCP read! |
| |
Does *ANY* company make a 100mbit card that can do it?
(I assume do it involves sending a message to the card like send(pointertoheader,size,pointertodata,size) and having it to all the work like checksumming and gathering up the fragments out of memory)
On Fri, 7 Aug 1998, Alan Cox wrote:
> > What if we created a read call that was identical in every way, except > > you passed it a SHM handle. It would directly write the data to the > > SHM segment rather than doing a NIC -> kernel -> user operation? > > > > This would make a one copy high performance, low-latency TCP read for > > the Cluster nuts. > > No it wouldnt. Zero copy TCP is hard and on current PC type cards basically > impossible for most cases > > > > - > 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.altern.org/andrebalsa/doc/lkml-faq.html >
- 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.altern.org/andrebalsa/doc/lkml-faq.html
| |