From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Adam Langley" Subject: Re: MD5 SG fix Date: Tue, 1 Jul 2008 14:48:31 -0700 Message-ID: <396556a20807011448m534c4c8dvf0b827f4cead4714@mail.gmail.com> References: <396556a20805301217k293e5718h6bbf02bfe0683151@europa> <20080701143943.426c9f24@extreme> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org To: "Stephen Hemminger" Return-path: Received: from rv-out-0506.google.com ([209.85.198.238]:40530 "EHLO rv-out-0506.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751570AbYGAVsd (ORCPT ); Tue, 1 Jul 2008 17:48:33 -0400 Received: by rv-out-0506.google.com with SMTP id k40so94328rvb.1 for ; Tue, 01 Jul 2008 14:48:32 -0700 (PDT) In-Reply-To: <20080701143943.426c9f24@extreme> Content-Disposition: inline Sender: netdev-owner@vger.kernel.org List-ID: On Tue, Jul 1, 2008 at 2:39 PM, Stephen Hemminger wrote: > I would rather see the pool used less and the stack used more. > Global context is more like FORTRAN common blocks. Will do. My only concern was that it was 214 bytes which could easily not be on the stack. > My suggestion would be to have two hash functions, one that takes a block > for the cases of TCP SYN, etc. and another that takes an skb for data packets. You're probably right. If it works I'll break it out like that. > You still need to have some of the places that reset sk_route_caps otherwise > you will get TSO/GSO packets because of the resetting of route_caps after the SYN/ACK. I've copied the hunks from [1] where sk_route_caps were added (not just changed from ~GSO to ~(GSO|SG)). Hopefully those are the right ones. I'll resend in a sec. Cheers, AGL -- Adam Langley agl@imperialviolet.org http://www.imperialviolet.org