From mboxrd@z Thu Jan 1 00:00:00 1970 From: chetan loke Subject: Re: [RFC PATCH 1/2] af_packet: direct dma for packet ineterface Date: Mon, 6 Mar 2017 13:28:02 -0800 Message-ID: References: <20170127213132.14162.82951.stgit@john-Precision-Tower-5810> <20170127213344.14162.59976.stgit@john-Precision-Tower-5810> <20170130191607.14d964e4@redhat.com> <588FB57B.1060507@gmail.com> <58916D84.7090706@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Cc: Willem de Bruijn , Jesper Dangaard Brouer , bjorn.topel@gmail.com, jasowang@redhat.com, ast@fb.com, alexander.duyck@gmail.com, john.r.fastabend@intel.com, Network Development To: John Fastabend Return-path: Received: from mail-ua0-f194.google.com ([209.85.217.194]:35690 "EHLO mail-ua0-f194.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753906AbdCFViT (ORCPT ); Mon, 6 Mar 2017 16:38:19 -0500 Received: by mail-ua0-f194.google.com with SMTP id u30so7453083uau.2 for ; Mon, 06 Mar 2017 13:37:54 -0800 (PST) In-Reply-To: <58916D84.7090706@gmail.com> Sender: netdev-owner@vger.kernel.org List-ID: On Tue, Jan 31, 2017 at 9:09 PM, John Fastabend wrote: >> If I understand correctly, the difficulty lies in v3 requiring that the >> timer "close" the block when the timer expires. That may not be worth >> implementing, indeed. >> > > Yep that is where I just gave up and decided it wasn't worth it. > Without a support for timeout, when a user-space app has to do its own book-keeping or lets say - shutdown for maintenance/upgrade, then how can they(app) unblock from this operation? Because if the link is idle then the DMA may never happen because there are no frames on the wire. So is there a way to handle this? Chetan