From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yijing Wang Subject: Re: [PATCH 09/28] PCI: Separate pci_host_bridge creation out of pci_create_root_bus() Date: Mon, 19 Jan 2015 10:39:42 +0800 Message-ID: <54BC6E6E.9090200@huawei.com> References: <1421372666-12288-1-git-send-email-wangyijing@huawei.com> <1421372666-12288-10-git-send-email-wangyijing@huawei.com> <11966423.5s0rO0b8cE@wuerfel> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <11966423.5s0rO0b8cE@wuerfel> Sender: linux-m68k-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="us-ascii" To: Arnd Bergmann , linux-arm-kernel@lists.infradead.org Cc: Bjorn Helgaas , Liviu Dudau , Tony Luck , Russell King , Marc Zyngier , linux-pci@vger.kernel.org, x86@kernel.org, linux-kernel@vger.kernel.org, "David S. Miller" , linux-m68k@lists.linux-m68k.org, Geert Uytterhoeven , linux-alpha@vger.kernel.org, linux-ia64@vger.kernel.org, Thomas Gleixner , Guan Xuetao , Yinghai Lu , Jiang Liu On 2015/1/16 17:34, Arnd Bergmann wrote: > On Friday 16 January 2015 09:44:07 Yijing Wang wrote: >> We want to make a generic pci_host_bridge, then we could >> place common PCI infos like domain number in it. Ripping >> out pci_host_bridge creation from pci_create_root_bus() >> make code more better readability. Further more, we could >> use the generic pci_host_bridge to hold host bridge specific >> operations like pcibios_root_bridge_prepare(). >> >> Signed-off-by: Yijing Wang >> > > I assume this patch is doing the right thing, but the description > above doesn't really tell me enough to be sure. > > This is supposed to be transparent to the callers, right? If > so, please mention it explicitly in the text. Sorry, I will refresh the patch log for better description. In this series I have no plan to export the pci_create_host_bridge() to callers, because I think pci_host_bridge_ops could make callers have the ability to setup the platform specific operations. But if this is needed, we could introduce another patches to export it. > > Also you don't explain what the list of host bridges is > used for. Maybe you can split this out into a separate patch > so you have one patch that just moves code from one file > to the other but no functional changes, and a second patch > that exports pci_create_host_bridge and introduces the list, > with an explanation of what it is used for. > pci_host_bridge_list is used to find out whether the pci_host_bridge in domain:bus is already existed, I will split the related code to another patch, thanks. > Arnd > > . > -- Thanks! Yijing