Changes in Kconfig

74d1412a - idpf: use libeth Rx buffer management for payload buffer

Thu, 20 Jun 2024 13:53:47 +0000

idpf: use libeth Rx buffer management for payload bufferidpf uses Page Pool for data buffers with hardcoded buffer lengths of4k for "classic" buffers and 2k for "short" ones. This is not flexibleand does not ensure optimal memory usage. Why would you need 4k bufferswhen the MTU is 1500?Use libeth for the data buffers and don't hardcode any buffer sizes. Letthem be calculated from the MTU for "classics" and then divide thetruesize by 2 for "short" ones. The memory usage is now greatly reducedand 2 buffer queues starts make sense: on frames <= 1024, you'll recycle(and resync) a page only after 4 HW writes rather than two.Signed-off-by: Alexander Lobakin Signed-off-by: Tony Nguyen List of files: /linux-6.15/drivers/net/ethernet/intel/idpf/Kconfig

4309363f - idpf: remove legacy Page Pool Ethtool stats

Thu, 20 Jun 2024 13:53:44 +0000

idpf: remove legacy Page Pool Ethtool statsPage Pool Ethtool stats are deprecated since the Netlink Page Poolinterface introduction.idpf receives big changes in Rx buffer management, including &page_poollayout, so keeping these deprecated stats does only harm, not speakingof that CONFIG_IDPF selects CONFIG_PAGE_POOL_STATS unconditionally,while the latter is often turned off for better performance.Remove all the references to PP stats from the Ethtool code. The statsare still available in their full via the generic Netlink interface.Reviewed-by: Przemek Kitszel Reviewed-by: Jacob Keller Signed-off-by: Alexander Lobakin Signed-off-by: Tony Nguyen List of files: /linux-6.15/drivers/net/ethernet/intel/idpf/Kconfig

1b1b2620 - idpf: reuse libeth's definitions of parsed ptype structures

Thu, 20 Jun 2024 13:53:43 +0000

idpf: reuse libeth's definitions of parsed ptype structuresidpf's in-kernel parsed ptype structure is almost identical to the oneused in the previous Intel drivers, which means it can be converted touse libeth's definitions and even helpers. The only difference is thatit doesn't use a constant table (libie), rather than one obtained fromthe device.Remove the driver counterpart and use libeth's helpers for hashes andchecksums. This slightly optimizes skb fields processing due to fasterchecks. Also don't define big static array of ptypes in &idpf_vport --allocate them dynamically. The pointer to it is anyway cached in&idpf_rx_queue.Reviewed-by: Przemek Kitszel Reviewed-by: Jacob Keller Signed-off-by: Alexander Lobakin Signed-off-by: Tony Nguyen List of files: /linux-6.15/drivers/net/ethernet/intel/idpf/Kconfig

f771314d - idpf: compile singleq code only under default-n CONFIG_IDPF_SINGLEQ

Thu, 20 Jun 2024 13:53:42 +0000

idpf: compile singleq code only under default-n CONFIG_IDPF_SINGLEQCurrently, all HW supporting idpf supports the singleq model, but noneof it advertises it by default, as splitq is supported and preferredfor multiple reasons. Still, this almost dead code often times addshotpath branches and redundant cacheline accesses.While it can't currently be removed, add CONFIG_IDPF_SINGLEQ and buildthe singleq code only when it's enabled manually. This corresponds to-10 Kb of object code size and a good bunch of hotpath checks.idpf_is_queue_model_split() works as a gate and compiles out to `true`when the config option is disabled.Reviewed-by: Przemek Kitszel Signed-off-by: Alexander Lobakin Signed-off-by: Tony Nguyen List of files: /linux-6.15/drivers/net/ethernet/intel/idpf/Kconfig