From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vlad Buslov Subject: Re: [PATCH net-next v4 00/11] Modify action API for implementing lockless actions Date: Fri, 01 Jun 2018 18:19:27 +0300 Message-ID: References: <1527753499-32124-1-git-send-email-vladbu@mellanox.com> <262fbd11-401e-90cf-4226-39b1604eb16d@mojatatu.com> Mime-Version: 1.0 Content-Type: text/plain Cc: netdev@vger.kernel.org, davem@davemloft.net, xiyou.wangcong@gmail.com, jiri@resnulli.us, pablo@netfilter.org, kadlec@blackhole.kfki.hu, fw@strlen.de, ast@kernel.org, daniel@iogearbox.net, edumazet@google.com, keescook@chromium.org, marcelo.leitner@gmail.com, kliteyn@mellanox.com To: Jamal Hadi Salim Return-path: Received: from mail-eopbgr00054.outbound.protection.outlook.com ([40.107.0.54]:58336 "EHLO EUR02-AM5-obe.outbound.protection.outlook.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1752794AbeFAPTk (ORCPT ); Fri, 1 Jun 2018 11:19:40 -0400 In-reply-to: Sender: netdev-owner@vger.kernel.org List-ID: On Fri 01 Jun 2018 at 12:24, Jamal Hadi Salim wrote: > On 31/05/18 08:38 AM, Vlad Buslov wrote: > >> Hi Jamal, >> >> On current net-next I still have action with single reference after last >> step: >> ~$ sudo $TC -s actions ls action skbedit >> total acts 1 >> >> action order 0: skbedit mark 1 pipe >> index 1 ref 2 bind 1 installed 47 sec used 47 sec >> Action statistics: >> Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) >> backlog 0b 0p requeues 0 >> ~$ sudo $TC filter del dev lo parent ffff: protocol ip prio 1 \ >>> u32 match ip dst 127.0.0.1/32 flowid 1:1 >> ~$ sudo $TC -s actions ls action skbedit >> total acts 1 >> >> action order 0: skbedit mark 1 pipe >> index 1 ref 1 bind 0 installed 80 sec used 80 sec >> Action statistics: >> Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) >> backlog 0b 0p requeues 0 >> >> Which branch are you testing on? > > You are correct - this is how it works now (I was thinking of a > very old version before Cong made some changes a while back). > Just vet this continues to work as above. > > cheers, > jamal Indeed, there was a problem that changed a behavior in this case. I fixed it, re-run the test suite, manually checked with test you proposed in this thread, and sent V5. Thanks, Vlad