Re: [PATCH net-next] udp_diag: fix udp_dump_one() vs SO_REUSEPORT

2016-01-05 Thread Eric Dumazet
On Tue, 2016-01-05 at 17:26 -0500, Craig Gallek wrote: > Thanks Eric! Still reading through this, but I noticed a couple other > paths as well. Anything that comes through the exported > udp4_lib_lookup. Currently xt_TPROXY.c and xt_socket.c. Though, > these could probably be changed to pass th

Re: [PATCH net-next] udp_diag: fix udp_dump_one() vs SO_REUSEPORT

2016-01-05 Thread Craig Gallek
On Tue, Jan 5, 2016 at 5:22 PM, Eric Dumazet wrote: > On Tue, 2016-01-05 at 14:08 -0800, Eric Dumazet wrote: >> From: Eric Dumazet >> >> udp_dump_one() uses __udp4_lib_lookup() & __udp6_lib_lookup() >> which cannot properly handle the provided cookie when SO_REUSEPORT >> is used, as many sockets

Re: [PATCH net-next] udp_diag: fix udp_dump_one() vs SO_REUSEPORT

2016-01-05 Thread Eric Dumazet
On Tue, 2016-01-05 at 14:08 -0800, Eric Dumazet wrote: > From: Eric Dumazet > > udp_dump_one() uses __udp4_lib_lookup() & __udp6_lib_lookup() > which cannot properly handle the provided cookie when SO_REUSEPORT > is used, as many sockets share the same 4-tuple > > Instead, let's use the provided

[PATCH net-next] udp_diag: fix udp_dump_one() vs SO_REUSEPORT

2016-01-05 Thread Eric Dumazet
From: Eric Dumazet udp_dump_one() uses __udp4_lib_lookup() & __udp6_lib_lookup() which cannot properly handle the provided cookie when SO_REUSEPORT is used, as many sockets share the same 4-tuple Instead, let's use the provided 64bit cookie to uniquely identify the socket. This will allow us to