From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from szxga02-in.huawei.com ([119.145.14.65]:60296 "EHLO szxga02-in.huawei.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755032Ab3DOCfV (ORCPT ); Sun, 14 Apr 2013 22:35:21 -0400 Message-ID: <516B673D.5020201@huawei.com> Date: Mon, 15 Apr 2013 10:34:37 +0800 From: Yijing Wang MIME-Version: 1.0 To: Bjorn Helgaas CC: Tony Luck , "linux-pci@vger.kernel.org" , "linux-kernel@vger.kernel.org" , Hanjun Guo , Jiang Liu , Fenghua Yu , Yinghai Lu , Greg Kroah-Hartman , Thierry Reding , "Rafael J. Wysocki" Subject: Re: [PATCH 2/2] PCI/IA64: fix pci_dev->enable_cnt balance when doing pci hotplug References: <1364805775-12396-1-git-send-email-wangyijing@huawei.com> <1364805775-12396-2-git-send-email-wangyijing@huawei.com> In-Reply-To: Content-Type: text/plain; charset="ISO-8859-1" Sender: linux-pci-owner@vger.kernel.org List-ID: >> @@ -383,6 +383,7 @@ struct pci_bus *pci_acpi_scan_root(struct acpi_pci_root *root) >> } >> >> pci_scan_child_bus(pbus); >> + pci_enable_bridges(pbus); >> return pbus; >> >> out3: > > I think that with this patch, if you hot-add a PCI host bridge, you > will call pci_enable_bridges() twice (once in pci_acpi_scan_root() and > again in acpi_pci_root_add()), so there will be an enable_cnt error in > the opposite direction. > > I'd like to see the pci_enable_bridges() calls pushed up into the > generic code because I don't think there's anything arch-specific > about it. Hi Bjorn, Thanks for your review and comments! This is my fault, I forgot we will enable pci bridges when we hot add host bridge. Push pci_enable_bridges() into the generic code is a good idea, so we don't need to consider enabling bridge in pci arch-specific code. In IA64 we don't assign the unassigned resources like other arch. This is also a weak point. I will update this patch and resend soon. Thanks! Yijing. > > . > -- Thanks! Yijing