Re: [Qemu-devel] [PATCH] Add a memory barrier to DMA functions

[Benjamin Herrenschmidt]

On Tue, 2012-05-22 at 15:03 +0300, Michael S. Tsirkin wrote:
> I have a deja vu. Amos sent perf results when you argued about
> exactly the same issue in guest virtio. Delta was small but
> measureable. At the moment I have no free time or free hardware
> to redo the same work all over again. It's a well known fact that
> actual memory barrier is slow on x86 CPUs. You can't see
> it with network on your laptop? Write a microbenchmark. 

Or just screw it, it's small enough, for emulated devices we can swallow
it, and move on with life, since virtio is unaffected, and we can always
fine tune e1000 if performance of that one is critical.

Ben.