From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jakub Kicinski Subject: Re: [PATCH net-next v2 1/5] nfp: make use of the DMA_ATTR_SKIP_CPU_SYNC attribute Date: Fri, 21 Apr 2017 17:36:25 -0700 Message-ID: <20170421173625.30774a5d@cakuba.lan> References: <20170421192310.193875-1-jakub.kicinski@netronome.com> <20170421192310.193875-2-jakub.kicinski@netronome.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: oss-drivers@netronome.com, Jakub Kicinski To: netdev@vger.kernel.org Return-path: Received: from mail-io0-f171.google.com ([209.85.223.171]:35580 "EHLO mail-io0-f171.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1162726AbdDVAga (ORCPT ); Fri, 21 Apr 2017 20:36:30 -0400 Received: by mail-io0-f171.google.com with SMTP id r16so131754662ioi.2 for ; Fri, 21 Apr 2017 17:36:29 -0700 (PDT) In-Reply-To: <20170421192310.193875-2-jakub.kicinski@netronome.com> Sender: netdev-owner@vger.kernel.org List-ID: On Fri, 21 Apr 2017 12:23:06 -0700, Jakub Kicinski wrote: > DMA unmap may destroy changes CPU made to the buffer. To make XDP > run correctly on non-x86 platforms we should use the > DMA_ATTR_SKIP_CPU_SYNC attribute. > > Thanks to using the attribute we can now push the sync operation to the > common code path from XDP handler. > > A little bit of variable name reshuffling is required to bring the > code back to readable state. > > Signed-off-by: Jakub Kicinski I think I need to add the sync for device after XDP DROP as well. Please drop this series, I'll send v3, sorry.