Re: Multi-thread udp 4.7 regression, bisected to 71d8c47fc653

2016-07-12 Thread Pablo Neira Ayuso
On Mon, Jul 11, 2016 at 06:17:39PM -0300, Marc Dionne wrote: > Hi Pablo, > > Testing out your patch: > > 1) With no NAT in place, the clash resolution happens, with no side > effects. No EPERM errors are seen. > > 2) With ip(6)table_nat loaded, the clash resolution fails and I get > some EPERM

Re: Multi-thread udp 4.7 regression, bisected to 71d8c47fc653

2016-07-11 Thread Marc Dionne
On Mon, Jul 11, 2016 at 1:26 PM, Pablo Neira Ayuso wrote: > On Sun, Jul 10, 2016 at 04:48:26PM -0300, Marc Dionne wrote: >> An update here since I've had some interactions with Pablo off list. >> >> Further testing shows that the underlying cause of the different test >> results is a udp packet th

Re: Multi-thread udp 4.7 regression, bisected to 71d8c47fc653

2016-07-11 Thread Pablo Neira Ayuso
On Sun, Jul 10, 2016 at 04:48:26PM -0300, Marc Dionne wrote: > An update here since I've had some interactions with Pablo off list. > > Further testing shows that the underlying cause of the different test > results is a udp packet that has a bogus source port number. In the > test the server pro

Re: Multi-thread udp 4.7 regression, bisected to 71d8c47fc653

2016-07-10 Thread Marc Dionne
On Tue, Jul 5, 2016 at 9:28 AM, Pablo Neira Ayuso wrote: > Hi, > > On Mon, Jul 04, 2016 at 09:35:28AM -0300, Marc Dionne wrote: >> If there is no quick fix, seems like a revert should be considered: >> - Looks to me like the commit attempts to fix a long standing bug >> (exists at least as far bac

Re: Multi-thread udp 4.7 regression, bisected to 71d8c47fc653

2016-07-05 Thread Pablo Neira Ayuso
Hi, On Mon, Jul 04, 2016 at 09:35:28AM -0300, Marc Dionne wrote: > If there is no quick fix, seems like a revert should be considered: > - Looks to me like the commit attempts to fix a long standing bug > (exists at least as far back as 3.5, > https://bugzilla.kernel.org/show_bug.cgi?id=52991) > -

Re: Multi-thread udp 4.7 regression, bisected to 71d8c47fc653

2016-07-04 Thread Marc Dionne
On Mon, Jun 27, 2016 at 2:21 PM, Marc Dionne wrote: > On Mon, Jun 27, 2016 at 12:38 PM, Florian Westphal wrote: >> Marc Dionne wrote: >>> On Mon, Jun 27, 2016 at 11:22 AM, Florian Westphal wrote: >>> > Marc Dionne wrote: >>> >> Hi, >> >>> > hlist_nulls_for_each_entry(h, n, &nf_conntrac

Re: Multi-thread udp 4.7 regression, bisected to 71d8c47fc653

2016-06-27 Thread Marc Dionne
On Mon, Jun 27, 2016 at 12:38 PM, Florian Westphal wrote: > Marc Dionne wrote: >> On Mon, Jun 27, 2016 at 11:22 AM, Florian Westphal wrote: >> > Marc Dionne wrote: >> >> Hi, > >> > hlist_nulls_for_each_entry(h, n, &nf_conntrack_hash[hash], hnnode) >> > if (nf_ct_key_equa

Re: Multi-thread udp 4.7 regression, bisected to 71d8c47fc653

2016-06-27 Thread Florian Westphal
Marc Dionne wrote: > On Mon, Jun 27, 2016 at 11:22 AM, Florian Westphal wrote: > > Marc Dionne wrote: > >> Hi, > > hlist_nulls_for_each_entry(h, n, &nf_conntrack_hash[hash], hnnode) > > if (nf_ct_key_equal(h, > > &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, > > -

Re: Multi-thread udp 4.7 regression, bisected to 71d8c47fc653

2016-06-27 Thread Marc Dionne
On Mon, Jun 27, 2016 at 11:22 AM, Florian Westphal wrote: > Marc Dionne wrote: >> Hi, >> >> I've been seeing issues with 4.7-rc kernels with some of our >> multi-thread test cases. I've bisected it down to this commit: >> >> commit 71d8c47fc653711c41bc3282e5b0e605b3727956 >> Author: Pablo Ne

Re: Multi-thread udp 4.7 regression, bisected to 71d8c47fc653

2016-06-27 Thread Florian Westphal
Marc Dionne wrote: > Hi, > > I've been seeing issues with 4.7-rc kernels with some of our > multi-thread test cases. I've bisected it down to this commit: > > commit 71d8c47fc653711c41bc3282e5b0e605b3727956 > Author: Pablo Neira Ayuso > Date: Sun May 1 00:28:40 2016 +0200 > > ne