Re: more NET_LOCK in pppx

2018-03-27 Thread David Gwynne
ok. > On 28 Mar 2018, at 10:07, Jonathan Matthew wrote: > > Reading through if_pppx.c I spotted a couple of places we need NET_LOCK(). > > ok? > > Index: if_pppx.c > === > RCS file: /cvs/src/sys/net/if_pppx.c,v

Re: More NET_LOCK()

2016-12-20 Thread Martin Pieuchot
On 20/12/16(Tue) 19:51, Alexander Bluhm wrote: > [...] > The sosplice thread is only used for TCP. UDP runs in the softnet > thread. As we are reading 64 bit values, we need some sort of > locking. Now everything is in the kernel lock, but when we start > working to unlock the socket layer, be

Re: More NET_LOCK()

2016-12-20 Thread Alexander Bluhm
On Tue, Dec 20, 2016 at 01:55:03PM +0100, Martin Pieuchot wrote: > rzalamena@ found that the IPv4 multicast code have two code paths ending > in ip_output() that are not yet taking the NET_LOCK: sosetopt() and > pffasttimo(). I see this also on my regress test machine. splassert: ip_output: want