iavf: switch to Page Pool

Browse source

Now that the IAVF driver simply uses dev_alloc_page() + free_page() with
no custom recycling logics, it can easily be switched to using Page
Pool / libeth API instead.
This allows to removing the whole dancing around headroom, HW buffer
size, and page order. All DMA-for-device is now done in the PP core,
for-CPU -- in the libeth helper.
Use skb_mark_for_recycle() to bring back the recycling and restore the
performance. Speaking of performance: on par with the baseline and
faster with the PP optimization series applied. But the memory usage for
1500b MTU is now almost 2x lower (x86_64) thanks to allocating a page
every second descriptor.

Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

...

This commit is contained in:

Alexander Lobakin 2024-04-18 13:36:15 +02:00 • committed by Tony Nguyen

parent 97cadd3d3c

commit 5fa4caff59

5 changed files with 111 additions and 214 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download patch file Download diff file Expand all files Collapse all files

7

drivers/net/ethernet/intel/iavf/iavf_main.c

View file

255

drivers/net/ethernet/intel/iavf/iavf_txrx.c

View file

34

drivers/net/ethernet/intel/iavf/iavf_txrx.h

View file

10

drivers/net/ethernet/intel/iavf/iavf_virtchnl.c

View file

17

include/linux/net/intel/libie/rx.h

View file