On 11/01/2013 10:13 AM, Rusty Russell wrote:

Jason Wang [off-list ref] writes:

quoted

Commit 2613af0ed18a11d5c566a81f9a6510b73180660a (virtio_net: migrate mergeable
rx buffers to page frag allocators) try to increase the payload/truesize for
MTU-sized traffic. But this will introduce the extra overhead for GSO packets
received because of the frag list. This commit tries to reduce this issue by
coalesce the possible rx frags when possible during rx. Test result shows the
about 15% improvement on full size GSO packet receiving (and even better than
commit 2613af0ed18a11d5c566a81f9a6510b73180660a).

I don't know about the other users of skb_add_rx_frag, but should
this coalesce-if-possible code be built into that?

There're several other users without the possibility check. And Eric
point out in do_tcp_sendpages() skb_add_rx_frag() and skb_can_coalesce()
were used separatedly which can allows us to do some other things in the
middle. So it was ok here.

Thanks,
Rusty.