From mboxrd@z Thu Jan 1 00:00:00 1970 From: Logan Gunthorpe Subject: Re: [PATCH v2 04/10] PCI/P2PDMA: Clear ACS P2P flags for all devices behind switches Date: Thu, 1 Mar 2018 16:59:27 -0700 Message-ID: References: <20180228234006.21093-1-logang@deltatee.com> <20180228234006.21093-5-logang@deltatee.com> <20180301180257.GH13722@bhelgaas-glaptop.roam.corp.google.com> <0D05579B-789C-4A19-B3A2-C1A630BE31C0@raithlin.com> <20180301231536.GB74737@bhelgaas-glaptop.roam.corp.google.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20180301231536.GB74737-1RhO1Y9PlrlHTL0Zs8A6p5iNqAH0jzoTYJqu5kTmcBRl57MIdRCFDg@public.gmane.org> Content-Language: en-US List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: linux-nvdimm-bounces-hn68Rpc1hR1g9hUCZPvPmw@public.gmane.org Sender: "Linux-nvdimm" To: Bjorn Helgaas , Stephen Bates Cc: Jens Axboe , Keith Busch , Alex Williamson , "linux-nvdimm-hn68Rpc1hR1g9hUCZPvPmw@public.gmane.org" , "linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , "linux-pci-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , "linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , "linux-nvme-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org" , "linux-block-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , =?UTF-8?B?SsOpcsO0bWUgR2xpc3Nl?= , Jason Gunthorpe , Benjamin Herrenschmidt , Bjorn Helgaas , Max Gurtovoy , Christoph Hellwig List-Id: linux-rdma@vger.kernel.org On 01/03/18 04:15 PM, Bjorn Helgaas wrote: > The question is what the relevant switch is. We call pci_enable_acs() > on every PCI device, including Root Ports. It looks like this relies > on get_upstream_bridge_port() to filter out some things. I don't > think get_upstream_bridge_port() is doing the right thing, so we need > to sort that out first, I guess. Yes, well if a port has an upstream bridge port and that port has an upstream bridge port we are almost certainly looking at a port in a switch. If you have a better suggestion to only disable ACS bits on downstream switch ports I'd be happy to consider it. > This might be a good start, but whatever the implications are, they > need to be distilled and made clear directly in the changelog and the > Kconfig help text. Well, I tried. I'll expand on it for v3. Logan