Re: [PATCH RFC net-next] arp: Really delete an arp entry on "arp -d"

2017-05-31 Thread Sowmini Varadhan
On (05/30/17 18:42), David Ahern wrote: > It has been noticed. I have not sent a patch since adjusting gc > parameters will reclaim FAILED entries at whatever rate the user wants. Overly aggressive garbage collection will delete other (non-FAILED) entries as well, and can trigger needless re-ARP

Re: [PATCH RFC net-next] arp: Really delete an arp entry on "arp -d"

2017-05-30 Thread David Ahern
On 5/30/17 5:32 PM, Sowmini Varadhan wrote: > On (05/30/17 16:20), Stephen Hemminger wrote: >> >> Please don't copy/paste chunks of code. Instead refactor and make this >> into a helper function. > > sure, I have no problems with that, and as I pointed out, I've not > tested ipv6 for this yet

Re: [PATCH RFC net-next] arp: Really delete an arp entry on "arp -d"

2017-05-30 Thread Sowmini Varadhan
On (05/30/17 16:20), Stephen Hemminger wrote: > > Please don't copy/paste chunks of code. Instead refactor and make this > into a helper function. sure, I have no problems with that, and as I pointed out, I've not tested ipv6 for this yet either. I'll do all of this after getting some feedback

Re: [PATCH RFC net-next] arp: Really delete an arp entry on "arp -d"

2017-05-30 Thread Stephen Hemminger
On Mon, 29 May 2017 20:48:16 -0700 Sowmini Varadhan wrote: > + np = >hash_buckets[hash_val]; > + while ((n = rcu_dereference_protected(*np, > + lockdep_is_held(>lock))) != NULL) { > + write_lock(>lock); > +

[PATCH RFC net-next] arp: Really delete an arp entry on "arp -d"

2017-05-30 Thread Sowmini Varadhan
Noticed during some testing: the command # arp -s 62.2.0.1 a:b:c:d:e:f dev eth2 adds an entry like the following (listed by "arp -an") ? (62.2.0.1) at 0a:0b:0c:0d:0e:0f [ether] PERM on eth2 but the symmetric deletion command # arp -i eth2 -d 62.2.0.1 does not remove the PERM entry from the