Default Tx rates cause very long ISR delays on Tx.
0xff is 510us delay, giving only ~ 2000 interrupts per seconds for
Tx rings cleanup. With these settings udp tx rate was never higher than
~800Mbps on a single stream. Changing min delay to 0xF makes it
way better with ~6Gbps

TCP stream performance is almost unaffected by this change, since LSO
optimizations play important role.

CPU load is affected insignificantly by this change.

Signed-off-by: Pavel Belous <[email protected]>
Signed-off-by: Igor Russkikh <[email protected]>
---
 drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c 
b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c
index 11f7e71..ec68c20 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c
@@ -821,12 +821,12 @@ static int hw_atl_b0_hw_interrupt_moderation_set(struct 
aq_hw_s *self)
                        itr_rx |= rx_max_timer << 0x10U;
                } else {
                        static unsigned int hw_atl_b0_timers_table_tx_[][2] = {
-                               {0xffU, 0xffU}, /* 10Gbit */
-                               {0xffU, 0x1ffU}, /* 5Gbit */
-                               {0xffU, 0x1ffU}, /* 5Gbit 5GS */
-                               {0xffU, 0x1ffU}, /* 2.5Gbit */
-                               {0xffU, 0x1ffU}, /* 1Gbit */
-                               {0xffU, 0x1ffU}, /* 100Mbit */
+                               {0xfU, 0xffU}, /* 10Gbit */
+                               {0xfU, 0x1ffU}, /* 5Gbit */
+                               {0xfU, 0x1ffU}, /* 5Gbit 5GS */
+                               {0xfU, 0x1ffU}, /* 2.5Gbit */
+                               {0xfU, 0x1ffU}, /* 1Gbit */
+                               {0xfU, 0x1ffU}, /* 100Mbit */
                        };
 
                        static unsigned int hw_atl_b0_timers_table_rx_[][2] = {
-- 
2.7.4

Reply via email to