Re: XDP redirect measurements, gotchas and tracepoints

On Tue, Aug 22, 2017 at 11:59 PM, Michael Chan
[off-list ref] wrote:

On Tue, Aug 22, 2017 at 6:06 PM, Alexander Duyck
[off-list ref] wrote:

quoted

On Tue, Aug 22, 2017 at 1:04 PM, Michael Chan [off-list ref] wrote:

quoted

Right, but it's conceivable to add an API to "return" the buffer to
the input device, right?

You could, it is just added complexity. "just free the buffer" in
ixgbe usually just amounts to one atomic operation to decrement the
total page count since page recycling is already implemented in the
driver. You still would have to unmap the buffer regardless of if you
were recycling it or not so all you would save is 1.000015259 atomic
operations per packet. The fraction is because once every 64K uses we
have to bulk update the count on the page.

If the buffer is returned to the input device, the input device can
keep the DMA mapping.  All it needs to do is to dma_sync it back to
the input device when the buffer is returned.

That is what ixgbe is already doing. The Rx path doesn't free the
page, it just treats it as a bounce buffer and uses the page count to
make certain we don't use a section of the buffer that is already in
use.

- Alex