Return-path: Received: from mail-wi0-f172.google.com ([209.85.212.172]:37495 "EHLO mail-wi0-f172.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750824AbbCBJXy convert rfc822-to-8bit (ORCPT ); Mon, 2 Mar 2015 04:23:54 -0500 Received: by wivr20 with SMTP id r20so13225552wiv.2 for ; Mon, 02 Mar 2015 01:23:53 -0800 (PST) MIME-Version: 1.0 In-Reply-To: <1425287785.1906.0.camel@sipsolutions.net> References: <1425287344-30594-1-git-send-email-michal.kazior@tieto.com> <1425287344-30594-2-git-send-email-michal.kazior@tieto.com> <1425287785.1906.0.camel@sipsolutions.net> Date: Mon, 2 Mar 2015 10:23:53 +0100 Message-ID: (sfid-20150302_102359_087675_0642F7A9) Subject: Re: [PATCH 2/3] ath10k: always save/restore pci config space From: Michal Kazior To: Johannes Berg Cc: "ath10k@lists.infradead.org" , linux-wireless Content-Type: text/plain; charset=UTF-8 Sender: linux-wireless-owner@vger.kernel.org List-ID: On 2 March 2015 at 10:16, Johannes Berg wrote: > On Mon, 2015-03-02 at 10:09 +0100, Michal Kazior wrote: >> The check isn't really necessary and couldn't even >> work because becase pci_restore_state() restores >> only first 64 bytes of PCI configuration space. >> >> This is necessary for future WoWLAN support. > >> + pci_save_state(pdev); >> + pci_disable_device(pdev); > > I believe the whole thing isn't necessary at all since the kernel's PCIe > layer will take care of it. You mean only the pci_save_state() and pci_disable_device() or, the resume counterpart as well? Hmm. Earlier the resume counterpart didn't do any restore stuff and the device ended up pretty confused.. Michał