From mboxrd@z Thu Jan 1 00:00:00 1970 From: Masahiro Yamada Subject: Re: [PATCH 4/8] PCI: consolidate PCI config entry in drivers/pci Date: Fri, 19 Oct 2018 14:07:04 +0900 Message-ID: References: <20181017080201.10866-1-hch@lst.de> <20181017080201.10866-5-hch@lst.de> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Return-path: In-Reply-To: <20181017080201.10866-5-hch@lst.de> Sender: linux-kernel-owner@vger.kernel.org To: Christoph Hellwig , linux-pci@vger.kernel.org Cc: mporter@kernel.crashing.org, Alex Bounine , Dominik Brodowski , Linux Kbuild mailing list , linux-scsi , linux-arch , Linux Kernel Mailing List , linuxppc-dev List-Id: linux-arch.vger.kernel.org On Wed, Oct 17, 2018 at 5:04 PM Christoph Hellwig wrote: > > There is no good reason to duplicate the PCI menu in every architecture. > Instead provide a selectable HAS_PCI symbol that indicates availability HAS_PCI -> HAVE_PCI > of PCI support and the handle the rest in drivers/pci. > > Note that for powerpc we now select HAVE_PCI globally instead of the > convoluted mess of conditional or or non-conditional support per board, > similar to what we do e.g. on x86. For alpha PCI is selected for the > non-jensen configs as it was the default before, and a lot of code does > not compile without PCI enabled. On other architectures with limited > PCI support that wasn't as complicated I've left the selection as-is. > > Signed-off-by: Christoph Hellwig > Acked-by: Thomas Gleixner > Acked-by: Bjorn Helgaas Just in case, could you double-check these? PCI_ENDPOINT PCI_ENDPOINT_CONFIGFS PCI_EPF_TEST Previously, architecture without "source drivers/pci/Kconfig" could not enable PCI_ENDPOINT. Now, any architecture can enable it regardless of its actual PCI availability because PCI_ENDPOINT is only guarded by HAS_DMA. We could add 'depends on HAVE_PCI' or something to guard it to avoid changing the logic. config PCI_ENDPOINT bool "PCI Endpoint Support" depends on HAVE_PCI # Is this correct ?? depends on HAS_DMA or better to have 'depends on PCI' ? PCI ML is also CC'ed, so comments are appreciated. -- Best Regards Masahiro Yamada From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from conssluserg-03.nifty.com ([210.131.2.82]:61395 "EHLO conssluserg-03.nifty.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726942AbeJSNML (ORCPT ); Fri, 19 Oct 2018 09:12:11 -0400 MIME-Version: 1.0 References: <20181017080201.10866-1-hch@lst.de> <20181017080201.10866-5-hch@lst.de> In-Reply-To: <20181017080201.10866-5-hch@lst.de> From: Masahiro Yamada Date: Fri, 19 Oct 2018 14:07:04 +0900 Message-ID: Subject: Re: [PATCH 4/8] PCI: consolidate PCI config entry in drivers/pci Content-Type: text/plain; charset="UTF-8" Sender: linux-arch-owner@vger.kernel.org List-ID: To: Christoph Hellwig , linux-pci@vger.kernel.org Cc: mporter@kernel.crashing.org, Alex Bounine , Dominik Brodowski , Linux Kbuild mailing list , linux-scsi , linux-arch , Linux Kernel Mailing List , linuxppc-dev Message-ID: <20181019050704.VjlOapMeRBw8gZAAoBImS-Y9D1fO0GwbWDIZk1bwjkg@z> On Wed, Oct 17, 2018 at 5:04 PM Christoph Hellwig wrote: > > There is no good reason to duplicate the PCI menu in every architecture. > Instead provide a selectable HAS_PCI symbol that indicates availability HAS_PCI -> HAVE_PCI > of PCI support and the handle the rest in drivers/pci. > > Note that for powerpc we now select HAVE_PCI globally instead of the > convoluted mess of conditional or or non-conditional support per board, > similar to what we do e.g. on x86. For alpha PCI is selected for the > non-jensen configs as it was the default before, and a lot of code does > not compile without PCI enabled. On other architectures with limited > PCI support that wasn't as complicated I've left the selection as-is. > > Signed-off-by: Christoph Hellwig > Acked-by: Thomas Gleixner > Acked-by: Bjorn Helgaas Just in case, could you double-check these? PCI_ENDPOINT PCI_ENDPOINT_CONFIGFS PCI_EPF_TEST Previously, architecture without "source drivers/pci/Kconfig" could not enable PCI_ENDPOINT. Now, any architecture can enable it regardless of its actual PCI availability because PCI_ENDPOINT is only guarded by HAS_DMA. We could add 'depends on HAVE_PCI' or something to guard it to avoid changing the logic. config PCI_ENDPOINT bool "PCI Endpoint Support" depends on HAVE_PCI # Is this correct ?? depends on HAS_DMA or better to have 'depends on PCI' ? PCI ML is also CC'ed, so comments are appreciated. -- Best Regards Masahiro Yamada