From mboxrd@z Thu Jan 1 00:00:00 1970 From: Herbert Xu Subject: Re: netlink & rhashtable status Date: Thu, 14 May 2015 10:55:57 +0800 Message-ID: <20150514025557.GB3853@gondor.apana.org.au> References: <20150513062038.GA26944@gondor.apana.org.au> <1431522271.566.132.camel@edumazet-glaptop2.roam.corp.google.com> <1431533884.566.148.camel@edumazet-glaptop2.roam.corp.google.com> <20150513.123520.1301797535605779844.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: eric.dumazet@gmail.com, tgraf@suug.ch, netdev@vger.kernel.org To: David Miller Return-path: Received: from helcar.hengli.com.au ([209.40.204.226]:41380 "EHLO helcar.hengli.com.au" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753170AbbENC4D (ORCPT ); Wed, 13 May 2015 22:56:03 -0400 Content-Disposition: inline In-Reply-To: <20150513.123520.1301797535605779844.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: On Wed, May 13, 2015 at 12:35:20PM -0400, David Miller wrote: > > Alternatively, we could consider reverting the rhashtable conversion > of netlink in the interim. It might be the safest solution for > -stable. The most suspicous change is the dynamic rehashing, i.e., allowing insertions/removals during a rehash. But then Eric says that the same thing happens under 3.17, where every opertion is under a single mutex. So I'm at a loss as to what is wrong and if we can't figure it out then yes I agree that reverting the netlink rhashtable conversion is probably the safest option. Cheers, -- Email: Herbert Xu Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt