From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753420AbbL2O1d (ORCPT ); Tue, 29 Dec 2015 09:27:33 -0500 Received: from szxga01-in.huawei.com ([58.251.152.64]:47258 "EHLO szxga01-in.huawei.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751817AbbL2O13 (ORCPT ); Tue, 29 Dec 2015 09:27:29 -0500 Subject: Re: [PATCH v1 1/3] ARM64 LPC: indirect ISA PORT IO introduced To: Arnd Bergmann , Rongrong Zou References: <1451396032-23708-1-git-send-email-zourongrong@gmail.com> <1451396032-23708-2-git-send-email-zourongrong@gmail.com> <3232651.99Svz5pbDW@wuerfel> CC: , , , , , , , From: Rongrong Zou Message-ID: <56829813.2060803@huawei.com> Date: Tue, 29 Dec 2015 22:26:27 +0800 User-Agent: Mozilla/5.0 (Windows NT 6.1; rv:38.0) Gecko/20100101 Thunderbird/38.3.0 MIME-Version: 1.0 In-Reply-To: <3232651.99Svz5pbDW@wuerfel> Content-Type: text/plain; charset="UTF-8"; format=flowed Content-Transfer-Encoding: 8bit X-Originating-IP: [10.177.30.66] X-CFilter-Loop: Reflected X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A020201.56829822.00F5,ss=1,re=0.000,recu=0.000,reip=0.000,cl=1,cld=1,fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32 X-Mirapoint-Loop-Id: 0442f5c99388746b1cfc955549e50c1a Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Arnd, Thanks for your comment. 在 2015/12/29 21:47, Arnd Bergmann 写道: > On Tuesday 29 December 2015 21:33:50 Rongrong Zou wrote: >> Indirect ISA port I/O accessing introduced, vendors can hook >> their own in/out function to general inb/outb. Drivers can access >> legacy ISA I/O port by inb/outb as it is done in x86 platform. >> >> Signed-off-by: Rongrong Zou > > Looks correct to me, but I have a few style comments > >> --- >> arch/arm64/Kconfig.platforms | 5 ++- >> arch/arm64/include/asm/io.h | 78 ++++++++++++++++++++++++++++++++++++++++++++ >> arch/arm64/kernel/setup.c | 5 +++ >> 3 files changed, 87 insertions(+), 1 deletion(-) >> >> diff --git a/arch/arm64/Kconfig.platforms b/arch/arm64/Kconfig.platforms >> index 4043c35..98ae206 100644 >> --- a/arch/arm64/Kconfig.platforms >> +++ b/arch/arm64/Kconfig.platforms >> @@ -127,5 +127,8 @@ config ARCH_ZYNQMP >> bool "Xilinx ZynqMP Family" >> help >> This enables support for Xilinx ZynqMP Family >> - >> +config ARM64_INDIRECT_PIO >> + bool "ARM64 Indirect port I/O" >> + help >> + This enables support for ARM64 indirect port I/O >> endmenu > > The option should probably go into arch/arm64/Kconfig. Possibly you can make > it a silent option that just gets selected whenever a driver is enabled > that might set the callbacks. I try to bind it with HISILICON platform. > >> diff --git a/arch/arm64/include/asm/io.h b/arch/arm64/include/asm/io.h >> index 44be1e0..0041f3b 100644 >> --- a/arch/arm64/include/asm/io.h >> +++ b/arch/arm64/include/asm/io.h >> @@ -193,6 +193,84 @@ extern void __iomem *ioremap_cache(phys_addr_t phys_addr, size_t size); >> */ >> #define xlate_dev_kmem_ptr(p) p >> >> +#ifdef CONFIG_ARM64_INDIRECT_PIO >> +#define DEF_PCI_HOOK_pio(x) x >> +#else >> +#define DEF_PCI_HOOK_pio(x) NULL >> +#endif > > Maybe just put the entire definition block inside #ifdef and > fall back to the default inb/outb definitions otherwise. > >> +/* >> + * This value is equal to PCIBIOS_MIN_IO >> + */ >> +#define LEGACY_ISA_PORT_MAX 0x1000 > > I would just use PCIBIOS_MIN_IO instead of defining another macro. Because PCIBIOS_MIN_IO is defined in asm/pci.h, and asm/io.h is included by asm/pci.h. so do you mean i define PCIBIOS_MIN_IO here, and in asm/pci.h just use include asm/io.h ? > >> +extern struct arm64_isa_io { >> + u8 (*inb)(unsigned long port); >> + u16 (*inw)(unsigned long port); >> + u32 (*inl)(unsigned long port); >> + void (*outb)(u8 value, unsigned long port); >> + void (*outw)(u16 value, unsigned long port); >> + void (*outl)(u32 value, unsigned long port); >> +} arm64_isa_io; > > Maybe make this a single function pointer like > > void (*arm64_indirect_pio)(unsigned long port, bool write, int size, void *data); > > I'm guessing that this would result in smaller object code at the call sites, > but you'd have to try. OK, i try it. > > Arnd > > _______________________________________________ > linuxarm mailing list > linuxarm@huawei.com > http://rnd-openeuler.huawei.com/mailman/listinfo/linuxarm > > . >