On Sun, Dec 04, 2011 at 05:16:59PM +0200, Avi Kivity wrote:

On 12/04/2011 05:11 PM, Michael S. Tsirkin wrote:

quoted

quoted

There's also the used ring, but that's a
mistake if you have out of order completion.  We should have used copying.

Seems unrelated... unless you want used to be written into
descriptor ring itself?

The avail/used rings are in addition to the regular ring, no?

Yes. A couple of extra pages, if we reduce alignment we could pack this
in a single extra page.

 If you
copy descriptors, then it goes away.

The avail ring could go away. used could if we make descriptors
writeable. IIUC it was made RO in the hope that will make it
easier for xen to adopt. Still relevant?

quoted

But, I don't really know why does virtio ring insist on
making the 3 buffers (avail/used/descriptor)
physically contigious. Rusty?

Let's drop them instead.

quoted

quoted

16kB worth of descriptors is 1024 entries.  With 4kB buffers, that's 4MB
worth of data, or 4 ms at 10GbE line speed.  With 1500 byte buffers it's
just 1.5 ms.  In any case I think it's sufficient.

Right. So I think that without indirect, we waste about 3 entries
per packet for virtio header and transport etc headers.

That does suck.  Are there issues in increasing the ring size?  Or
making it discontiguous?

discontiguous ring is what indirect is, basically.

Can you take a peek at how Xen manages its rings?  They have the same
problems we do.

-- 
error compiling committee.c: too many arguments to function