asgardius/android_kernel_motorola_sm6225
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

net: increase receive packet quantum

Browse source 
This patch gets about 1.25% back on tbench regression.

My change to NAPI for multiqueue support changed the time limit on
network receive processing.  Under sustained loads like tbench, this
can cause the receiver to reschedule prematurely. 

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
Stephen Hemminger 2008-11-03 17:14:38 -08:00 committed by David S. Miller
parent b9ac99855d
commit 24f8b2385e
1 changed files with 4 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

11
net/core/dev.c
View file

@ -2373,7 +2373,7 @@ EXPORT_SYMBOL(__napi_schedule);
static void net_rx_action(struct softirq_action *h) static void net_rx_action(struct softirq_action *h)
{ {
struct list_head *list = &__get_cpu_var(softnet_data).poll_list; struct list_head *list = &__get_cpu_var(softnet_data).poll_list;
unsigned long start_time = jiffies; unsigned long time_limit = jiffies + 2;
int budget = netdev_budget; int budget = netdev_budget;
void *have; void *have;
@ -2384,13 +2384,10 @@ static void net_rx_action(struct softirq_action *h)
int work, weight; int work, weight;
/* If softirq window is exhuasted then punt. /* If softirq window is exhuasted then punt.
* * Allow this to run for 2 jiffies since which will allow
* Note that this is a slight policy change from the * an average latency of 1.5/HZ.
* previous NAPI code, which would allow up to 2
* jiffies to pass before breaking out. The test
* used to be "jiffies - start_time > 1".
*/ */
if (unlikely(budget <= 0 || jiffies != start_time)) if (unlikely(budget <= 0 || time_after(jiffies, time_limit)))
goto softnet_break; goto softnet_break;
local_irq_enable(); local_irq_enable();