Re: Question about ip_defrag

2017-08-30 Thread Jesper Dangaard Brouer
On Wed, 30 Aug 2017 13:58:20 +0200 Florian Westphal wrote: > Jesper Dangaard Brouer wrote: > > > I take 2) back. Its wrong to do this, for large NR_CPU values it > > > would even overflow. > > > > Alternatively solution 3: > > Why do we want to maintain a

Re: Question about ip_defrag

2017-08-30 Thread Florian Westphal
Jesper Dangaard Brouer wrote: > > I take 2) back. Its wrong to do this, for large NR_CPU values it > > would even overflow. > > Alternatively solution 3: > Why do we want to maintain a (4MBytes) memory limit, across all CPUs? > Couldn't we just allow each CPU to have a memory

Re: Question about ip_defrag

2017-08-30 Thread Jesper Dangaard Brouer
(trimmed CC list a bit) On Tue, 29 Aug 2017 09:53:15 +0200 Florian Westphal wrote: > Jesper Dangaard Brouer wrote: > > On Mon, 28 Aug 2017 16:00:32 +0200 > > Florian Westphal wrote: > > > > > liujian (CE) wrote: >

RE: Question about ip_defrag

2017-08-29 Thread liujian (CE)
ev@vger.kernel.org; > Wangkefeng (Kevin); weiyongjun (A) > Subject: Re: Question about ip_defrag > > liujian (CE) <liujia...@huawei.com> wrote: > > [ trimming cc list ] > > > Now, I have not the real environment. > > I use iperf generate fragment packets; and I

Re: Question about ip_defrag

2017-08-29 Thread Florian Westphal
liujian (CE) wrote: [ trimming cc list ] > Now, I have not the real environment. > I use iperf generate fragment packets; > and I always change NIC rx irq's affinity cpu, to make sure frag_mem_limit > reach to thresh. > my test machine, CPU num is 384. Oh well, that

RE: Question about ip_defrag

2017-08-29 Thread liujian (CE)
duma...@google.com; > netdev@vger.kernel.org; Wangkefeng (Kevin); weiyongjun (A) > Subject: RE: Question about ip_defrag > > > > > -Original Message- > > From: netdev-ow...@vger.kernel.org > > [mailto:netdev-ow...@vger.kernel.org] > > On Behalf Of Fl

Re: Question about ip_defrag

2017-08-29 Thread Florian Westphal
Jesper Dangaard Brouer wrote: > On Mon, 28 Aug 2017 16:00:32 +0200 > Florian Westphal wrote: > > > liujian (CE) wrote: > > > Hi > > > > > > I checked our 3.10 kernel, we had backported all percpu_counter bug fix > > > in

RE: Question about ip_defrag

2017-08-29 Thread liujian (CE)
ogle.com; > netdev@vger.kernel.org; Wangkefeng (Kevin); weiyongjun (A); > bro...@redhat.com > Subject: Re: Question about ip_defrag > > On Mon, 28 Aug 2017 16:00:32 +0200 > Florian Westphal <f...@strlen.de> wrote: > > > liujian (CE) <liujia...@huawei.com>

RE: Question about ip_defrag

2017-08-29 Thread liujian (CE)
shf...@linux-ipv6.org; elena.reshet...@intel.com; eduma...@google.com; > netdev@vger.kernel.org; Wangkefeng (Kevin); weiyongjun (A) > Subject: Re: Question about ip_defrag > > liujian (CE) <liujia...@huawei.com> wrote: > > Hi > > > > I checked our 3.10 kern

Re: Question about ip_defrag

2017-08-29 Thread Jesper Dangaard Brouer
On Mon, 28 Aug 2017 16:00:32 +0200 Florian Westphal wrote: > liujian (CE) wrote: > > Hi > > > > I checked our 3.10 kernel, we had backported all percpu_counter bug fix in > > lib/percpu_counter.c and include/linux/percpu_counter.h. > > And I check

Re: Question about ip_defrag

2017-08-28 Thread Florian Westphal
liujian (CE) wrote: > Hi > > I checked our 3.10 kernel, we had backported all percpu_counter bug fix in > lib/percpu_counter.c and include/linux/percpu_counter.h. > And I check 4.13-rc6, also has the issue if NIC's rx cpu num big enough. > > > > > > the issue: > > > > >

RE: Question about ip_defrag

2017-08-28 Thread liujian (CE)
7 9:33 AM > To: 'Jesper Dangaard Brouer' > Cc: da...@davemloft.net; kuz...@ms2.inr.ac.ru; yoshf...@linux-ipv6.org; > elena.reshet...@intel.com; eduma...@google.com; netdev@vger.kernel.org; > Wangkefeng (Kevin); weiyongjun (A) > Subject: RE: Question about ip_defrag > > > >

RE: Question about ip_defrag

2017-08-24 Thread liujian (CE)
ev@vger.kernel.org; > bro...@redhat.com > Subject: Re: Question about ip_defrag > > > On Thu, 24 Aug 2017 16:04:41 + "liujian (CE)" <liujia...@huawei.com> > wrote: > > > >What kernel version have you seen this issue with? > > > > 3.10，wi

Re: Question about ip_defrag

2017-08-24 Thread Jesper Dangaard Brouer
g<mailto:yoshf...@linux-ipv6.org>;elena.reshet...@intel.com<mailto:elena.reshet...@intel.com>;eduma...@google.com<mailto:eduma...@google.com>;netdev@vger.kernel.org<mailto:netdev@vger.kernel.org>;bro...@redhat.com<mailto:bro...@redhat.com> > 主题： Re: Question about ip_

Re: Question about ip_defrag

2017-08-24 Thread Jesper Dangaard Brouer
On Thu, 24 Aug 2017 13:15:33 + "liujian (CE)" wrote: > Hello, > > With below patch we met one issue. > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?h=v4.13-rc6=6d7b857d541e > > the issue: > Ip_defrag fail caused by frag_mem_limit reached