mbox series

[net-next,v2,0/7] ibmvnic rr patchset

Message ID 20240806193706.998148-1-nnac123@linux.ibm.com (mailing list archive)
Headers show
Series ibmvnic rr patchset | expand

Message

Nick Child Aug. 6, 2024, 7:36 p.m. UTC 
Only change from v1 is a commit message change in the last patch.
In response to Jakubs comment, I added a blurb about non-GSO large
packet performance. Thanks Jakub for requesting this, I hadn't thought
to test that prior. Luckily, I think we are safe since xmit_more
is usually true (even in RR tests) due to the segmented packets. My best
case (and feedback is still welcome) to see if we hit any issues with
expensive checksumming was an 8900 byte RR workload with MTU 9000. Noted
those results in the commit message.

Thanks again,
Nick

v1 - https://lore.kernel.org/netdev/20240801212340.132607-1-nnac123@linux.ibm.com/

Nick Child (7):
  ibmvnic: Only replenish rx pool when resources are getting low
  ibmvnic: Use header len helper functions on tx
  ibmvnic: Reduce memcpys in tx descriptor generation
  ibmvnic: Remove duplicate memory barriers in tx
  ibmvnic: Introduce send sub-crq direct
  ibmvnic: Only record tx completed bytes once per handler
  ibmvnic: Perform tx CSO during send scrq direct

 drivers/net/ethernet/ibm/ibmvnic.c | 174 +++++++++++++++++------------
 1 file changed, 102 insertions(+), 72 deletions(-)