From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-2.2 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED,USER_AGENT_MUTT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 31B7DC31E49 for ; Thu, 13 Jun 2019 15:55:10 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 0B396208CA for ; Thu, 13 Jun 2019 15:55:10 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1731581AbfFMPyS (ORCPT ); Thu, 13 Jun 2019 11:54:18 -0400 Received: from verein.lst.de ([213.95.11.211]:36725 "EHLO newverein.lst.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1731572AbfFMIyc (ORCPT ); Thu, 13 Jun 2019 04:54:32 -0400 Received: by newverein.lst.de (Postfix, from userid 2407) id 2E87468B02; Thu, 13 Jun 2019 10:54:03 +0200 (CEST) Date: Thu, 13 Jun 2019 10:54:02 +0200 From: Christoph Hellwig To: Keith Busch Cc: Daniel Drake , Bjorn Helgaas , Jens Axboe , Keith Busch , Christoph Hellwig , Linux PCI , linux-ide@vger.kernel.org, Sagi Grimberg , linux-nvme , Linux Upstreaming Team Subject: Re: [PATCH] PCI: Add Intel remapped NVMe device support Message-ID: <20190613085402.GC13442@lst.de> References: <20190610074456.2761-1-drake@endlessm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.17 (2007-11-01) Sender: linux-pci-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-pci@vger.kernel.org On Wed, Jun 12, 2019 at 08:32:46AM -0600, Keith Busch wrote: > On Mon, Jun 10, 2019 at 8:46 PM Daniel Drake wrote: > > What's the specific problem that you see here? > > Performance. Have you had a chance to benchmark these storage devices > comparing legacy vs MSI interrupts? I don't think anyone would chose > the slower one on purpose. These platforms have an option to disable > raid mode, so the kernel's current recommendation should be the user's > best option. And it isn't just performance. I really don't understand how a) quirks on the PCI ID b) reset handling, including the PCI device removal as the last escalation step c) SR-IOV VFs and their management d) power management and probably various other bits I didn't even think of are going to work. So until we get very clear and good documentation from Intel on that I don't think any form of upstream support will fly. And given that Dan who submitted the original patch can't even talk about this thing any more and apparently got a gag order doesn't really give me confidence any of this will ever work.