Searched hist:c377411f2494a931ff7facdbb3a6839b1266bcf6 (Results 1 – 5 of 5) sorted by relevance
/linux/net/ipv6/ |
H A D | udp.c | diff c377411f2494a931ff7facdbb3a6839b1266bcf6 Wed Apr 28 00:13:20 CEST 2010 Eric Dumazet <eric.dumazet@gmail.com> net: sk_add_backlog() take rmem_alloc into account
Current socket backlog limit is not enough to really stop DDOS attacks, because user thread spend many time to process a full backlog each round, and user might crazy spin on socket lock.
We should add backlog size and receive_queue size (aka rmem_alloc) to pace writers, and let user run without being slow down too much.
Introduce a sk_rcvqueues_full() helper, to avoid taking socket lock in stress situations.
Under huge stress from a multiqueue/RPS enabled NIC, a single flow udp receiver can now process ~200.000 pps (instead of ~100 pps before the patch) on a 8 core machine.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
|
/linux/net/sctp/ |
H A D | socket.c | diff c377411f2494a931ff7facdbb3a6839b1266bcf6 Wed Apr 28 00:13:20 CEST 2010 Eric Dumazet <eric.dumazet@gmail.com> net: sk_add_backlog() take rmem_alloc into account
Current socket backlog limit is not enough to really stop DDOS attacks, because user thread spend many time to process a full backlog each round, and user might crazy spin on socket lock.
We should add backlog size and receive_queue size (aka rmem_alloc) to pace writers, and let user run without being slow down too much.
Introduce a sk_rcvqueues_full() helper, to avoid taking socket lock in stress situations.
Under huge stress from a multiqueue/RPS enabled NIC, a single flow udp receiver can now process ~200.000 pps (instead of ~100 pps before the patch) on a 8 core machine.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
|
/linux/net/ipv4/ |
H A D | udp.c | diff c377411f2494a931ff7facdbb3a6839b1266bcf6 Wed Apr 28 00:13:20 CEST 2010 Eric Dumazet <eric.dumazet@gmail.com> net: sk_add_backlog() take rmem_alloc into account
Current socket backlog limit is not enough to really stop DDOS attacks, because user thread spend many time to process a full backlog each round, and user might crazy spin on socket lock.
We should add backlog size and receive_queue size (aka rmem_alloc) to pace writers, and let user run without being slow down too much.
Introduce a sk_rcvqueues_full() helper, to avoid taking socket lock in stress situations.
Under huge stress from a multiqueue/RPS enabled NIC, a single flow udp receiver can now process ~200.000 pps (instead of ~100 pps before the patch) on a 8 core machine.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
|
/linux/net/core/ |
H A D | sock.c | diff c377411f2494a931ff7facdbb3a6839b1266bcf6 Wed Apr 28 00:13:20 CEST 2010 Eric Dumazet <eric.dumazet@gmail.com> net: sk_add_backlog() take rmem_alloc into account
Current socket backlog limit is not enough to really stop DDOS attacks, because user thread spend many time to process a full backlog each round, and user might crazy spin on socket lock.
We should add backlog size and receive_queue size (aka rmem_alloc) to pace writers, and let user run without being slow down too much.
Introduce a sk_rcvqueues_full() helper, to avoid taking socket lock in stress situations.
Under huge stress from a multiqueue/RPS enabled NIC, a single flow udp receiver can now process ~200.000 pps (instead of ~100 pps before the patch) on a 8 core machine.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
|
/linux/include/net/ |
H A D | sock.h | diff c377411f2494a931ff7facdbb3a6839b1266bcf6 Wed Apr 28 00:13:20 CEST 2010 Eric Dumazet <eric.dumazet@gmail.com> net: sk_add_backlog() take rmem_alloc into account
Current socket backlog limit is not enough to really stop DDOS attacks, because user thread spend many time to process a full backlog each round, and user might crazy spin on socket lock.
We should add backlog size and receive_queue size (aka rmem_alloc) to pace writers, and let user run without being slow down too much.
Introduce a sk_rcvqueues_full() helper, to avoid taking socket lock in stress situations.
Under huge stress from a multiqueue/RPS enabled NIC, a single flow udp receiver can now process ~200.000 pps (instead of ~100 pps before the patch) on a 8 core machine.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
|