Re: Performance regression with virtio_net

[Euan Kemp]

On Mon, Jul 31, 2017 at 04:26:08PM +0300, Michael S. Tsirkin wrote:
> I kept this around unchanged from
> ab7db91705e95ed1bba1304388936fccfa58c992.  That commit had an internal
> reason not to account for that space: not enough bits to do it.  No
> longer true so let's account for length exactly.  I'll send a proper
> patch after a bit of testing, would appreciate reports reports of
> whether this helps very much.

I can confirm your patch fixes the issue for me.

- Euan

Re: Performance regression with virtio_net

[Euan Kemp]

I've also observed this performance regression.

The minimal fix for me is removing the two
> if (unlikely(len > (unsigned long)ctx))
checks added in 680557c.

After digging a little more, the reason that check can fail appears to
be that add_recvbuf_mergeable sometimes includes a hole at the end,
which is included in len but not ctx.

I'd send a patch removing those conditions, but I'm not certain
whether "truesize" in receive_mergeable should also be changed back to
be the max of len/ctx, or should remain as-is.

- Euan