Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756286AbdCWWdz (ORCPT ); Thu, 23 Mar 2017 18:33:55 -0400 Received: from mail-io0-f169.google.com ([209.85.223.169]:33864 "EHLO mail-io0-f169.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756140AbdCWWdx (ORCPT ); Thu, 23 Mar 2017 18:33:53 -0400 Date: Thu, 23 Mar 2017 15:33:49 -0700 From: Brian Norris To: Bjorn Helgaas Cc: Bjorn Helgaas , linux-kernel@vger.kernel.org, Shawn Lin , Jeffy Chen , Wenrui Li , linux-pci@vger.kernel.org, linux-rockchip@lists.infradead.org Subject: Re: [PATCH v2 1/5] PCI: rockchip: fix sign issues for current limits Message-ID: <20170323223348.GA102312@google.com> References: <20170310024617.67303-1-briannorris@chromium.org> <20170323222717.GD23612@bhelgaas-glaptop.roam.corp.google.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170323222717.GD23612@bhelgaas-glaptop.roam.corp.google.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1505 Lines: 40 On Thu, Mar 23, 2017 at 05:27:17PM -0500, Bjorn Helgaas wrote: > On Thu, Mar 09, 2017 at 06:46:13PM -0800, Brian Norris wrote: > > The regulator framework can return negative error codes via > > regulator_get_current_limit() for regulators that don't provide current > > information. The subsequent check for postive values isn't very useful, > > if the variable is unsigned. > > > > Let's just match the signedness of the return value. > > > > Prevents error messages like this, seen on Samsung Chromebook Plus: > > > > [ 1.069372] rockchip-pcie f8000000.pcie: invalid power supply > > > > Fixes: 4816c4c7b82b ("PCI: rockchip: Provide captured slot power limit and scale") > > Signed-off-by: Brian Norris > > Acked-by: Shawn Lin > > I applied the first two patches (this already has Shawn's ack and the > second is trivially obvious) to pci/host-rockchip. Thanks! > I'm not sure what the > current state of the others is. Patch 4 seems like it should be fine (it was discussed previously, but never done). Apart from existing leaks in the PCI framework (which Jeffy and Shawn are trying to patch [1]), I don't think there are any known issues with 3 and 5. It's certainly better than having 100% broken unbind at least, IMO. I suppose it's worth getting an ack/nack from Shawn though. Brian [1] https://patchwork.kernel.org/patch/9638353/ https://patchwork.kernel.org/patch/9640545/ https://patchwork.kernel.org/patch/9640549/