Branch: refs/heads/master
Home: https://github.com/Linaro/odp
Commit: fa78c11ad2de706ce07a7691168bdb5b58d060d4
https://github.com/Linaro/odp/commit/fa78c11ad2de706ce07a7691168bdb5b58d060d4
Author: Ola Liljedahl <[email protected]>
Date: 2017-10-12 (Thu, 12 Oct 2017)
Changed paths:
M test/common_plat/performance/odp_pktio_perf.c
Log Message:
-----------
test: disable packet parsing in odp_pktio_perf
Disable packet parsing as it is done in the driver where it affects
scalability of the application. This especially problematic for e.g.
the loopback pktio which has only one RX queue.
Together with the improved sequence number allocation change,
performance improves >2x on Hisilicon D02 (6 and 8 threads) and more of
the scalability of the scheduler itself is exposed.
Signed-off-by: Ola Liljedahl <[email protected]>
Reviewed-by: Brian Brooks <[email protected]>
Reviewed-by: Bill Fischofer <[email protected]>
Reviewed-by: Honnappa Nagarahalli <[email protected]>
Signed-off-by: Maxim Uvarov <[email protected]>
Commit: b529f843814f70068de2090e02df632113afa800
https://github.com/Linaro/odp/commit/b529f843814f70068de2090e02df632113afa800
Author: Ola Liljedahl <[email protected]>
Date: 2017-10-12 (Thu, 12 Oct 2017)
Changed paths:
M test/common_plat/performance/odp_pktio_perf.c
Log Message:
-----------
test: make odp_pktio_perf more scalable
Amortise overhead of atomic increment over multiple packets
so to minimise this scalability bottleneck.
Signed-off-by: Ola Liljedahl <[email protected]>
Reviewed-by: Brian Brooks <[email protected]>
Reviewed-by: Bill Fischofer <[email protected]>
Reviewed-by: Honnappa Nagarahalli <[email protected]>
Signed-off-by: Maxim Uvarov <[email protected]>
Compare: https://github.com/Linaro/odp/compare/5a2376365df9...b529f843814f
