: net: Do software flow seperation on receive.
Subsystems (2)
David Miller
maintainerPATCH 2/2 : net: Do software flow seperation on receive.
net: Do software flow seperation on receive.
Push netif_receive_skb() work to remote cpus via flow hashing and remove softirqs.
--- include/linux/interrupt.h | 1 + include/linux/netdevice.h | 2 - include/linux/skbuff.h | 3 + net/core/dev.c | 273 +++++++++++++++++++++++++-------------------- 4 files changed, 157 insertions(+), 122 deletions(-)
include/linux/interrupt.h +1 -0
@@ -247,6 +247,7 @@ enumTIMER_SOFTIRQ ,NET_TX_SOFTIRQ ,NET_RX_SOFTIRQ , +NET_RECEIVE_SOFTIRQ ,BLOCK_SOFTIRQ ,TASKLET_SOFTIRQ , SCHED_SOFTIRQ,
include/linux/netdevice.h +0 -2
@@ -965,11 +965,9 @@ static inline int unregister_gifconf(unsigned int family)struct softnet_data {struct Qdisc *output_queue; -struct sk_buff_head input_pkt_queue;struct list_head poll_list;struct sk_buff *completion_queue; -struct napi_struct backlog;#ifdef CONFIG_NET_DMA struct dma_chan *net_dma;#endif
include/linux/skbuff.h +3 -0
@@ -18,6 +18,7 @@#include <linux/compiler.h> #include <linux/time.h> #include <linux/cache.h> +#include <linux/smp.h> #include <asm/atomic.h> #include <asm/types.h>
@@ -255,6 +256,8 @@ struct sk_buff {struct sk_buff *next;struct sk_buff *prev; +struct call_single_data csd; +struct sock *sk;ktime_t tstamp;struct net_device *dev;
net/core/dev.c +153 -120
370 lines not on this page · raw message
Signature · 1 line
1.5.6.5
Hello,
Quoted 7 lines · David Miller
net: Do software flow seperation on receive.
Push
netif_receive_skb()work to remote cpus via flow hashing and remove softirqs.Signed-off-by David S. Miller <[email protected]>
<CUT>
What is the future of this patch? Was it only a proof of concept or is it going to be included in the mainline one day?
Best regards,
Krzysztof Olędzki
David Miller
maintainerFrom: Krzysztof Oledzki <[email protected]>
Date: Sat, 27 Jun 2009 01:40:35 +0200 (CEST)
Quoted 2 lines
What is the future of this patch? Was it only a proof of concept or is it going to be included in the mainline one day?
Lots of work and thinking is necessary to do this right.
Maybe something will be integrated into my tree in the next few months or so.