Re: [PATCH net-next 0/8] net/packet: better behavior under DDOS

2019-06-14 Thread David Miller
From: Eric Dumazet Date: Wed, 12 Jun 2019 09:52:25 -0700 > Using tcpdump (or other af_packet user) on a busy host can lead to > catastrophic consequences, because suddenly, potentially all cpus > are spinning on a contended spinlock. > > Both packet_rcv() and tpacket_rcv() grab the spinlock > to

Re: [PATCH net-next 0/8] net/packet: better behavior under DDOS

2019-06-12 Thread Willem de Bruijn
On Wed, Jun 12, 2019 at 12:52 PM Eric Dumazet wrote: > > Using tcpdump (or other af_packet user) on a busy host can lead to > catastrophic consequences, because suddenly, potentially all cpus > are spinning on a contended spinlock. > > Both packet_rcv() and tpacket_rcv() grab the spinlock > to eve

[PATCH net-next 0/8] net/packet: better behavior under DDOS

2019-06-12 Thread Eric Dumazet
Using tcpdump (or other af_packet user) on a busy host can lead to catastrophic consequences, because suddenly, potentially all cpus are spinning on a contended spinlock. Both packet_rcv() and tpacket_rcv() grab the spinlock to eventually find there is no room for an additional packet. This patch