Re: [PATCH v2 net-next 0/7] net: speedup netns create/delete time

2017-12-13 Thread Dmitry Torokhov
On Wed, Dec 13, 2017 at 1:52 PM, Eric Dumazet wrote: > On Wed, 2017-12-13 at 13:43 -0800, Dmitry Torokhov wrote: >> Hi Eric, >> >> On Thu, Oct 19, 2017 at 7:11 AM, Eric Dumazet wrote: >> > >> > On Thu, Oct 19, 2017 at 4:48 AM, Tariq Toukan

Re: [PATCH v2 net-next 0/7] net: speedup netns create/delete time

2017-12-13 Thread Eric Dumazet
On Wed, 2017-12-13 at 13:43 -0800, Dmitry Torokhov wrote: > Hi Eric, > > On Thu, Oct 19, 2017 at 7:11 AM, Eric Dumazet wrote: > > > > On Thu, Oct 19, 2017 at 4:48 AM, Tariq Toukan wrote: > > > > > > Hi Eric, > > > > > > I just wanted to check if this

Re: [PATCH v2 net-next 0/7] net: speedup netns create/delete time

2017-12-13 Thread Dmitry Torokhov
Hi Eric, On Thu, Oct 19, 2017 at 7:11 AM, Eric Dumazet wrote: > > On Thu, Oct 19, 2017 at 4:48 AM, Tariq Toukan wrote: > > > > Hi Eric, > > > > I just wanted to check if this is solved already, as I don't want to keep an > > unnecessary revert patch in

Re: [PATCH v2 net-next 0/7] net: speedup netns create/delete time

2017-10-19 Thread Eric Dumazet
On Thu, Oct 19, 2017 at 4:48 AM, Tariq Toukan wrote: > > Hi Eric, > > I just wanted to check if this is solved already, as I don't want to keep an > unnecessary revert patch in our internal branches. > According to my check bug still exists. > I will handle this today, thanks

Re: [PATCH v2 net-next 0/7] net: speedup netns create/delete time

2017-10-19 Thread Tariq Toukan
On 26/09/2017 6:30 PM, Eric Dumazet wrote: On Tue, Sep 26, 2017 at 8:22 AM, Dmitry Torokhov wrote: It is in Greg's tree where all kobject patches should go through as far as I know. Yes, I will fix this, adding a second memmove() Hi Eric, I just wanted to

Re: [PATCH v2 net-next 0/7] net: speedup netns create/delete time

2017-09-26 Thread Eric Dumazet
On Tue, Sep 26, 2017 at 8:22 AM, Dmitry Torokhov wrote: > It is in Greg's tree where all kobject patches should go through as far as I > know. Yes, I will fix this, adding a second memmove()

Re: [PATCH v2 net-next 0/7] net: speedup netns create/delete time

2017-09-26 Thread Tariq Toukan
On 26/09/2017 6:13 PM, Eric Dumazet wrote: On Tue, Sep 26, 2017 at 8:04 AM, Tariq Toukan wrote: On 26/09/2017 3:51 PM, Eric Dumazet wrote: On Tue, Sep 26, 2017 at 4:21 AM, Tariq Toukan wrote: Hi Eric, We see a regression introduced in this

Re: [PATCH v2 net-next 0/7] net: speedup netns create/delete time

2017-09-26 Thread Dmitry Torokhov
On September 26, 2017 8:13:21 AM PDT, Eric Dumazet wrote: >On Tue, Sep 26, 2017 at 8:04 AM, Tariq Toukan >wrote: >> >> >> On 26/09/2017 3:51 PM, Eric Dumazet wrote: >>> >>> On Tue, Sep 26, 2017 at 4:21 AM, Tariq Toukan >wrote:

Re: [PATCH v2 net-next 0/7] net: speedup netns create/delete time

2017-09-26 Thread Eric Dumazet
On Tue, Sep 26, 2017 at 8:04 AM, Tariq Toukan wrote: > > > On 26/09/2017 3:51 PM, Eric Dumazet wrote: >> >> On Tue, Sep 26, 2017 at 4:21 AM, Tariq Toukan wrote: >>> >>> >>> Hi Eric, >>> >>> We see a regression introduced in this series, specifically in

Re: [PATCH v2 net-next 0/7] net: speedup netns create/delete time

2017-09-26 Thread Tariq Toukan
On 26/09/2017 3:51 PM, Eric Dumazet wrote: On Tue, Sep 26, 2017 at 4:21 AM, Tariq Toukan wrote: Hi Eric, We see a regression introduced in this series, specifically in the patches touching lib/kobject_uevent.c. We tried to figure out what is wrong there, but couldn't

Re: [PATCH v2 net-next 0/7] net: speedup netns create/delete time

2017-09-26 Thread Eric Dumazet
On Tue, Sep 26, 2017 at 4:21 AM, Tariq Toukan wrote: > > > Hi Eric, > > We see a regression introduced in this series, specifically in the patches > touching lib/kobject_uevent.c. > We tried to figure out what is wrong there, but couldn't point it out. > > Bug is that mlx4

Re: [PATCH v2 net-next 0/7] net: speedup netns create/delete time

2017-09-26 Thread Tariq Toukan
On 20/09/2017 2:27 AM, Eric Dumazet wrote: When rate of netns creation/deletion is high enough, we observe softlockups in cleanup_net() caused by huge list of netns and way too many rcu_barrier() calls. This patch series does some optimizations in kobject, and add batching to tunnels so that

Re: [PATCH v2 net-next 0/7] net: speedup netns create/delete time

2017-09-19 Thread David Miller
From: Eric Dumazet Date: Tue, 19 Sep 2017 16:27:02 -0700 > When rate of netns creation/deletion is high enough, > we observe softlockups in cleanup_net() caused by huge list > of netns and way too many rcu_barrier() calls. > > This patch series does some optimizations in

[PATCH v2 net-next 0/7] net: speedup netns create/delete time

2017-09-19 Thread Eric Dumazet
When rate of netns creation/deletion is high enough, we observe softlockups in cleanup_net() caused by huge list of netns and way too many rcu_barrier() calls. This patch series does some optimizations in kobject, and add batching to tunnels so that netns dismantles are less costly. IPv6