| From: |
| David Miller <davem-AT-davemloft.net> |
| To: |
| edumazet-AT-google.com |
| Subject: |
| Re: [PATCH net-next 0/5] tcp: add zero copy receive |
| Date: |
| Mon, 16 Apr 2018 18:48:37 -0400 (EDT) |
| Message-ID: |
| <20180416.184837.997129432524220807.davem@davemloft.net> |
| Cc: |
| netdev-AT-vger.kernel.org, ncardwell-AT-google.com, ycheng-AT-google.com, soheil-AT-google.com, eric.dumazet-AT-gmail.com |
From: Eric Dumazet <edumazet@google.com>
Date: Mon, 16 Apr 2018 10:33:34 -0700
> This patch series add mmap() support to TCP sockets for RX zero copy.
>
> While tcp_mmap() patch itself is quite small (~100 LOC), optimal support
> for asynchronous mmap() required better SO_RCVLOWAT behavior, and a
> test program to demonstrate how mmap() on TCP sockets can be used.
>
> Note that mmap() (and associated munmap()) calls are adding more
> pressure on per-process VM semaphore, so might not show benefit
> for processus with high number of threads.
Great work. I can see how it is less effective without the rcvlowat
fixes/tweaks.
Series applied, thanks Eric!
