[PATCH v8 RFC 0/3] Generic IOMMU pooled allocator

Sowmini Varadhan sowmini.varadhan at oracle.com
Thu Apr 2 23:51:52 AEDT 2015


On (03/31/15 23:12), David Miller wrote:
> 
> It's much more amortized with smart buffering strategies, which are
> common on current generation networking cards.
> 
> There you only eat one map/unmap per "PAGE_SIZE / rx_pkt_size".
> 
> Maybe the infiniband stuff is doing things very suboptimally, and
> actually with that subsystem and drivers absolutely nothing would
> surprise me.

yeh, we are trying to get more info from them about what their
bottle-necks are. Until then, lets just stick with the spin_lock.

do I need to resubmit this without the RFC tag? Perhaps I should
have dropped that some time ago.

--Sowmini


More information about the Linuxppc-dev mailing list