[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180913163848.ni5xc4gc4d6uusdn@breakpoint.cc>
Date: Thu, 13 Sep 2018 18:38:48 +0200
From: Florian Westphal <fw@...len.de>
To: Wolfgang Walter <linux@...m.de>
Cc: Florian Westphal <fw@...len.de>, netdev@...r.kernel.org,
Steffen Klassert <steffen.klassert@...unet.com>,
David Miller <davem@...hat.com>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
Linus Torvalds <torvalds@...ux-foundation.org>
Subject: Re: Regression: kernel 4.14 an later very slow with many ipsec
tunnels
Wolfgang Walter <linux@...m.de> wrote:
> I'll try that but this isn't that easy as the router image does not contain
> perf. I also have to do that on our production router. I try to do that
> tomorrow evening.
No need if its too difficult.
> What I can say is that it depends mainly on number of policy rules and SA.
Thats already a good hint, I guess we're hitting long hash chains in
xfrm_policy_lookup_bytype().
Powered by blists - more mailing lists