From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on inbox.vuxu.org X-Spam-Level: X-Spam-Status: No, score=-0.1 required=5.0 tests=DKIM_SIGNED,DKIM_VALID, DKIM_VALID_AU autolearn=ham autolearn_force=no version=3.4.4 Received: (qmail 30503 invoked from network); 19 Nov 2022 20:12:00 -0000 Received: from 9front.inri.net (168.235.81.73) by inbox.vuxu.org with ESMTPUTF8; 19 Nov 2022 20:12:00 -0000 Received: from cc-smtpout2.netcologne.de ([89.1.8.212]) by 9front; Sat Nov 19 15:09:15 -0500 2022 Received: from cc-app2.netcologne.de (cc-app2.netcologne.de [89.1.9.191]) by cc-smtpout2.netcologne.de (Postfix) with ESMTP id 6A93D1255C for <9front@9front.org>; Sat, 19 Nov 2022 21:09:09 +0100 (CET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=netcologne.de; s=nc1116a; t=1668888549; bh=OtVQOlQHEfNYiHCGdSAAK+3pXx1Do4/CLIwxHNWMg8k=; h=Date:From:To:Message-ID:In-Reply-To:References:Subject:From; b=RCpptKhx2yPyfmYErCY7eyohs0Ms68VjZPzepybSGjG3BCVHzRXrXiEL3J02nXNTi SbGdSS7PEStPSuA3YMpockjamB1GAM2hzvpfiJnGn3tXggMYGLxB+Y2kCcO8JGvH6Z hOVY8pA44jpcPdvT+1SZJUulnbye3dKpceCOb1scEVG7G5lNeabuTvSlOaDbHEgywj iCJaG6z12gP4NNKiWf9Xqe3JZshFcEbyDkm45oGNM8MD9ZapBsBEKVBrPQKSN8QZTn stB8eOrWTu2YfADElgtHqRAYH01sfF7ikZpFVmAUlo8jlHAHy27MU/QBzMp6QEBNdb FMgORs/feiodg== Received: from cc-app2.netcologne.de (localhost [127.0.0.1]) by cc-app2.netcologne.de (Postfix) with ESMTPA id 3F8E611D96 for <9front@9front.org>; Sat, 19 Nov 2022 21:09:09 +0100 (CET) Date: Sat, 19 Nov 2022 21:09:09 +0100 (CET) From: Arne Meyer To: 9front@9front.org Message-ID: <632533984.161412.1668888549169@comcenter.netcologne.de> In-Reply-To: <1485335506.4596272.1668188142183@comcenter.netcologne.de> References: <1250188960.4593874.1668183193726@comcenter.netcologne.de> <1485335506.4596272.1668188142183@comcenter.netcologne.de> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-Priority: 3 Importance: Normal X-Mailer: Open-Xchange Mailer v7.10.6-Rev30 X-Originating-IP: 89.1.208.52 X-Originating-Client: open-xchange-appsuite X-NetCologne-Spam: L X-Spamd-Bar: - X-Rspamd-Queue-Id: 3F8E611D96 List-ID: <9front.9front.org> List-Help: X-Glyph: ➈ X-Bullshit: proven social interface browser Subject: Re: [9front] [patch] ethervgbe: add rx checksum offloading Reply-To: 9front@9front.org Precedence: bulk Is hw offloading something we want? I ask because I think there is something wrong with ether8169 offloading and I'm working on a patch to fix or axe this. > Arne Meyer hat am 11.11.2022 17:35 GMT geschrieben: > > > To be honest, no. I was reading through the source code and found out that > the ip stack supports RX checksum offloading, so i thought it was fun to implement it. > The vgbe driver is my pet project. > It's been in my local repo and kernel for about a week now and it didn't hurt. Blocks get send up the stack anyway, > the only difference is that the code sets the flags for the blocks. Worst case scenario here is that the driver flags checksums as valid even when they are not. > OpenBSD and FreeBSD both support it, so i think it's stable(ish). But as a fellow network admin, I somewhat agree > with you. > > > hiro <23hiro@gmail.com> hat am 11.11.2022 17:01 GMT geschrieben: > > > > > > did you find concrete use cases where it's worth offloading this? > > some kind of actual bottleneck? > > in practice one of my most frequent network administration related > > tasks is to turn off all kinds of weird (broken) offloading features > > on random ethernet cards. > > > > On 11/11/22, Arne Meyer wrote: > > > Hello, > > > > > > this patch implements rx tcp and udp checksum offloading for ipv4 for the > > > ethervgbe nic driver. > > > > > > Greetings, > > > Arne