From: robherring2@gmail.com (Rob Herring)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 13/15] ARM: make mach/io.h include optional
Date: Mon, 27 Feb 2012 13:31:48 -0600 [thread overview]
Message-ID: <4F4BDA24.1070807@gmail.com> (raw)
In-Reply-To: <201202141716.26930.arnd@arndb.de>
On 02/14/2012 11:16 AM, Arnd Bergmann wrote:
> On Tuesday 14 February 2012, Rob Herring wrote:
>> On 02/14/2012 02:04 AM, Arnd Bergmann wrote:
>>> On Tuesday 14 February 2012, Rob Herring wrote:
>>>
>>>> If we use CONFIG_PCI, then we have to move over all PCI enabled
>>>> platforms at once. With a separate kconfig option, then we can move
>>>> platforms one by one. Some are legacy and we may not want to move. This
>>>> also helped with omap, but Tony has now fixed it.
>>>
>>> We could also generalize the implementation from tegra, which seems
>>> reasonable as a start:
>>>
>>>
>>> #define IO_SPACE_LIMIT 0xffff
>>>
>>> #if defined(CONFIG_ISA) || defined(CONFIG_PCCARD)
>>> #include <mach/io.h>
>>> #elif defined(CONFIG_PCI)
>>> extern void __iomem *pci_io_base;
>>>
>>> static inline void __iomem *__io(unsigned long addr)
>>> {
>>> return pci_io_base + (addr & IO_SPACE_LIMIT);
>>
>> But don't we want a constant pci_io_base? This would certainly be a
>> quicker conversion, but I don't think we want to do it twice.
>
> Yes, at least in the long run. Note that this should make no difference
> at all from a performance point of view, but it does impact code size a bit.
>
>>> }
>>> #else
>>> static inline void __iomem *__io(unsigned long addr)
>>> {
>>> return NULL;
>>> }
>>> #endif
>>> #define __io(a) __io(a)
>>>
>>> Out of the platforms supporting PCI right now, we currently have these three classes:
>>>
>>> 1. portable, but using different virtual addresses:
>>> arch/arm/mach-integrator/include/mach/io.h:#define __io(a) ((void __iomem *)(PCI_IO_VADDR + (a)))
>>> arch/arm/mach-ixp23xx/include/mach/io.h:#define __io(p) ((void __iomem*)((p) + IXP23XX_PCI_IO_VIRT))
>>> arch/arm/mach-ixp2000/include/mach/io.h:#define __io(p) ((void __iomem *)((p)+IXP2000_PCI_IO_VIRT_BASE))
>>> arch/arm/mach-shark/include/mach/io.h:#define __io(a) ((void __iomem *)(0xe0000000 + (a)))
>>> arch/arm/mach-footbridge/include/mach/io.h:#define __io(a) ((void __iomem *)(PCIO_BASE + (a)))
>>> arch/arm/mach-tegra/include/mach/io.h:static inline void __iomem *__io(unsigned long addr)
>>> arch/arm/mach-kirkwood/include/mach/io.h:static inline void __iomem *__io(unsigned long addr)
>>> arch/arm/mach-dove/include/mach/io.h:#define __io(a) ((void __iomem *)(((a) - DOVE_PCIE0_IO_BUS_BASE) + \
>>>
>>> 2. Does not map the I/O space, or does not use it -- I cannot see how
>>> any of these use PIO based PCI devices at all, probably broken already:
>>> arch/arm/mach-cns3xxx/include/mach/io.h:#define __io(a) __typesafe_io(a)
>>> arch/arm/mach-ixp4xx/include/mach/io.h:#define __io(v) __typesafe_io(v)
>>> arch/arm/mach-ks8695/include/mach/io.h:#define __io(a) __typesafe_io(a)
>>> arch/arm/mach-orion5x/include/mach/io.h:#define __io(a) __typesafe_io(a)
>>> arch/arm/mach-sa1100/include/mach/io.h:#define __io(a) __typesafe_io(a)
>>> arch/arm/mach-pxa/include/mach/io.h:#define __io(a) __typesafe_io(a)
>>> arch/arm/mach-versatile/include/mach/io.h:#define __io(a) __typesafe_io(a)
>>>
>>> 3. scary multi-way translation, needs someone to really understand (Nico?, Lennert?)
>>> arch/arm/mach-iop32x/include/mach/io.h:#define __io(p) ((void __iomem *)IOP3XX_PCI_IO_PHYS_TO_VIRT(p))
>>> arch/arm/mach-iop33x/include/mach/io.h:#define __io(p) ((void __iomem *)IOP3XX_PCI_IO_PHYS_TO_VIRT(p))
>>> arch/arm/mach-iop13xx/include/mach/io.h:#define __io(a) __iop13xx_io(a)
>>> arch/arm/mach-mv78xx0/include/mach/io.h:static inline void __iomem *__io(unsigned long addr)
>>>
>>> I think if we can figure out the third category, this should all become
>>> fairly easy for PCI based platforms.
>>>
>>
>> This looks correct AFAICT.
>>
>> We could just call the 3rd category legacy and move on.
>
> I'm not ready to give up that fast. The platforms seem to be implemented
> reasonably well compared to category 2, but they can't be converted in
> a completely mechanical way. The patch below should put iop13xx it into
> category 1, but it really needs testing and review by a person who
> understands that platform. I can also take a look at the others if we
> agree that this is the right approach.
>
It looks fine to me and I'm happy to pull it in, but who can test?
There's almost no commits specific to iop13xx going back to 2009. The
PCI related code goes back to 2007. I've added a couple of Intel folks
that did the original support.
Rob
> When we have all platforms done in that way, we can move the virtual
> mapping base to a fixed location.
>
> 8<--------------
> [PATCH] iop13xx: use more regular PCI I/O space handling
>
> iop13xx confuses I/O port numbers with physical addresses, which breaks
> legacy ISA I/O access behind PCI bridges and makes it unnecessarily hard
> to unify the inb/outb accessors with other platforms. This removes the
> special-casing and just puts all I/O ports into a single 128KB virtually
> mapped I/O port range starting at port zero.
>
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
> arch/arm/mach-iop13xx/include/mach/io.h | 6 +++-
> arch/arm/mach-iop13xx/include/mach/iop13xx.h | 12 ++--------
> arch/arm/mach-iop13xx/io.c | 27 --------------------------
> arch/arm/mach-iop13xx/pci.c | 12 +++++-----
> 4 files changed, 13 insertions(+), 44 deletions(-)
>
> diff --git a/arch/arm/mach-iop13xx/include/mach/io.h b/arch/arm/mach-iop13xx/include/mach/io.h
> index dffb234..90bb7df 100644
> --- a/arch/arm/mach-iop13xx/include/mach/io.h
> +++ b/arch/arm/mach-iop13xx/include/mach/io.h
> @@ -19,9 +19,11 @@
> #ifndef __ASM_ARM_ARCH_IO_H
> #define __ASM_ARM_ARCH_IO_H
>
> -#define IO_SPACE_LIMIT 0xffffffff
> +#include <mach/iop13xx.h>
>
> -#define __io(a) __iop13xx_io(a)
> +#define IO_SPACE_LIMIT (IOP13XX_PCIE_IO_WINDOW_SIZE + IOP13XX_PCIX_IO_WINDOW_SIZE - 1)
> +
> +#define __io(a) (IOP13XX_PCIX_LOWER_IO_VA + ((a) & IO_SPACE_LIMIT))
> #define __mem_pci(a) (a)
> #define __mem_isa(a) (a)
>
> diff --git a/arch/arm/mach-iop13xx/include/mach/iop13xx.h b/arch/arm/mach-iop13xx/include/mach/iop13xx.h
> index 07e9ff7..ee1dfd2 100644
> --- a/arch/arm/mach-iop13xx/include/mach/iop13xx.h
> +++ b/arch/arm/mach-iop13xx/include/mach/iop13xx.h
> @@ -68,17 +68,15 @@ extern unsigned long get_iop_tick_rate(void);
> * 0x8000.0000 + 928M 0x2.8000.0000 (ioremap) PCIE outbound memory window
> *
> * IO MAP
> - * 0x1000 + 64K 0x0.fffb.1000 0xfec6.1000 PCIX outbound i/o window
> + * 0x1000 + 64K 0x0.fffb.1000 0xfed6.1000 PCIX outbound i/o window
> * 0x1000 + 64K 0x0.fffd.1000 0xfed7.1000 PCIE outbound i/o window
> */
> #define IOP13XX_PCIX_IO_WINDOW_SIZE 0x10000UL
> #define IOP13XX_PCIX_LOWER_IO_PA 0xfffb0000UL
> -#define IOP13XX_PCIX_LOWER_IO_VA 0xfec60000UL
> +#define IOP13XX_PCIX_LOWER_IO_VA 0xfed60000UL
> #define IOP13XX_PCIX_LOWER_IO_BA 0x0UL /* OIOTVR */
> #define IOP13XX_PCIX_IO_BUS_OFFSET 0x1000UL
> -#define IOP13XX_PCIX_UPPER_IO_PA (IOP13XX_PCIX_LOWER_IO_PA +\
> - IOP13XX_PCIX_IO_WINDOW_SIZE - 1)
> -#define IOP13XX_PCIX_UPPER_IO_VA (IOP13XX_PCIX_LOWER_IO_VA +\
> +#define IOP13XX_PCIX_UPPER_IO_BA (IOP13XX_PCIX_LOWER_IO_BA +\
> IOP13XX_PCIX_IO_WINDOW_SIZE - 1)
> #define IOP13XX_PCIX_IO_PHYS_TO_VIRT(addr) (u32) ((u32) addr -\
> (IOP13XX_PCIX_LOWER_IO_PA\
> @@ -107,10 +105,6 @@ extern unsigned long get_iop_tick_rate(void);
> #define IOP13XX_PCIE_LOWER_IO_VA 0xfed70000UL
> #define IOP13XX_PCIE_LOWER_IO_BA 0x0UL /* OIOTVR */
> #define IOP13XX_PCIE_IO_BUS_OFFSET 0x1000UL
> -#define IOP13XX_PCIE_UPPER_IO_PA (IOP13XX_PCIE_LOWER_IO_PA +\
> - IOP13XX_PCIE_IO_WINDOW_SIZE - 1)
> -#define IOP13XX_PCIE_UPPER_IO_VA (IOP13XX_PCIE_LOWER_IO_VA +\
> - IOP13XX_PCIE_IO_WINDOW_SIZE - 1)
> #define IOP13XX_PCIE_UPPER_IO_BA (IOP13XX_PCIE_LOWER_IO_BA +\
> IOP13XX_PCIE_IO_WINDOW_SIZE - 1)
> #define IOP13XX_PCIE_IO_PHYS_TO_VIRT(addr) (u32) ((u32) addr -\
> diff --git a/arch/arm/mach-iop13xx/io.c b/arch/arm/mach-iop13xx/io.c
> index 48642e6..7a1ab4d 100644
> --- a/arch/arm/mach-iop13xx/io.c
> +++ b/arch/arm/mach-iop13xx/io.c
> @@ -21,25 +21,6 @@
> #include <linux/io.h>
> #include <mach/hardware.h>
>
> -void * __iomem __iop13xx_io(unsigned long io_addr)
> -{
> - void __iomem * io_virt;
> -
> - switch (io_addr) {
> - case IOP13XX_PCIE_LOWER_IO_PA ... IOP13XX_PCIE_UPPER_IO_PA:
> - io_virt = (void *) IOP13XX_PCIE_IO_PHYS_TO_VIRT(io_addr);
> - break;
> - case IOP13XX_PCIX_LOWER_IO_PA ... IOP13XX_PCIX_UPPER_IO_PA:
> - io_virt = (void *) IOP13XX_PCIX_IO_PHYS_TO_VIRT(io_addr);
> - break;
> - default:
> - BUG();
> - }
> -
> - return io_virt;
> -}
> -EXPORT_SYMBOL(__iop13xx_io);
> -
> void * __iomem __iop13xx_ioremap(unsigned long cookie, size_t size,
> unsigned int mtype)
> {
> @@ -65,12 +46,6 @@ void * __iomem __iop13xx_ioremap(unsigned long cookie, size_t size,
> (cookie - IOP13XX_PBI_LOWER_MEM_RA),
> size, mtype, __builtin_return_address(0));
> break;
> - case IOP13XX_PCIE_LOWER_IO_PA ... IOP13XX_PCIE_UPPER_IO_PA:
> - retval = (void *) IOP13XX_PCIE_IO_PHYS_TO_VIRT(cookie);
> - break;
> - case IOP13XX_PCIX_LOWER_IO_PA ... IOP13XX_PCIX_UPPER_IO_PA:
> - retval = (void *) IOP13XX_PCIX_IO_PHYS_TO_VIRT(cookie);
> - break;
> case IOP13XX_PMMR_PHYS_MEM_BASE ... IOP13XX_PMMR_UPPER_MEM_PA:
> retval = (void *) IOP13XX_PMMR_PHYS_TO_VIRT(cookie);
> break;
> @@ -100,8 +75,6 @@ void __iop13xx_iounmap(void __iomem *addr)
> goto skip;
>
> switch ((u32) addr) {
> - case IOP13XX_PCIE_LOWER_IO_VA ... IOP13XX_PCIE_UPPER_IO_VA:
> - case IOP13XX_PCIX_LOWER_IO_VA ... IOP13XX_PCIX_UPPER_IO_VA:
> case IOP13XX_PMMR_VIRT_MEM_BASE ... IOP13XX_PMMR_UPPER_MEM_VA:
> goto skip;
> }
> diff --git a/arch/arm/mach-iop13xx/pci.c b/arch/arm/mach-iop13xx/pci.c
> index b8f5a87..d72eddf 100644
> --- a/arch/arm/mach-iop13xx/pci.c
> +++ b/arch/arm/mach-iop13xx/pci.c
> @@ -1042,8 +1042,8 @@ int iop13xx_pci_setup(int nr, struct pci_sys_data *sys)
> << IOP13XX_ATUX_PCIXSR_FUNC_NUM;
> __raw_writel(pcixsr, IOP13XX_ATUX_PCIXSR);
>
> - res[0].start = IOP13XX_PCIX_LOWER_IO_PA + IOP13XX_PCIX_IO_BUS_OFFSET;
> - res[0].end = IOP13XX_PCIX_UPPER_IO_PA;
> + res[0].start = IOP13XX_PCIX_LOWER_IO_BA + IOP13XX_PCIX_IO_BUS_OFFSET;
> + res[0].end = IOP13XX_PCIX_UPPER_IO_BA;
> res[0].name = "IQ81340 ATUX PCI I/O Space";
> res[0].flags = IORESOURCE_IO;
>
> @@ -1052,7 +1052,7 @@ int iop13xx_pci_setup(int nr, struct pci_sys_data *sys)
> res[1].name = "IQ81340 ATUX PCI Memory Space";
> res[1].flags = IORESOURCE_MEM;
> sys->mem_offset = IOP13XX_PCIX_MEM_OFFSET;
> - sys->io_offset = IOP13XX_PCIX_LOWER_IO_PA;
> + sys->io_offset = IOP13XX_PCIX_LOWER_IO_BA;
> break;
> case IOP13XX_INIT_ATU_ATUE:
> /* Note: the function number field in the PCSR is ro */
> @@ -1063,8 +1063,8 @@ int iop13xx_pci_setup(int nr, struct pci_sys_data *sys)
>
> __raw_writel(pcsr, IOP13XX_ATUE_PCSR);
>
> - res[0].start = IOP13XX_PCIE_LOWER_IO_PA + IOP13XX_PCIE_IO_BUS_OFFSET;
> - res[0].end = IOP13XX_PCIE_UPPER_IO_PA;
> + res[0].start = IOP13XX_PCIE_LOWER_IO_BA + IOP13XX_PCIE_IO_BUS_OFFSET;
> + res[0].end = IOP13XX_PCIE_UPPER_IO_BA;
> res[0].name = "IQ81340 ATUE PCI I/O Space";
> res[0].flags = IORESOURCE_IO;
>
> @@ -1073,7 +1073,7 @@ int iop13xx_pci_setup(int nr, struct pci_sys_data *sys)
> res[1].name = "IQ81340 ATUE PCI Memory Space";
> res[1].flags = IORESOURCE_MEM;
> sys->mem_offset = IOP13XX_PCIE_MEM_OFFSET;
> - sys->io_offset = IOP13XX_PCIE_LOWER_IO_PA;
> + sys->io_offset = IOP13XX_PCIE_LOWER_IO_BA;
> sys->map_irq = iop13xx_pcie_map_irq;
> break;
> default:
next prev parent reply other threads:[~2012-02-27 19:31 UTC|newest]
Thread overview: 74+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-02-13 21:43 [PATCH 00/15] mach/io.h cleanup and removal Rob Herring
2012-02-13 21:43 ` [PATCH 01/15] usb: ohci-pxa27x: add explicit include of hardware.h Rob Herring
2012-02-13 21:43 ` [PATCH 02/15] ARM: add explicit include of system.h to processor.h Rob Herring
2012-02-13 22:14 ` H Hartley Sweeten
2012-02-13 21:43 ` [PATCH 03/15] ARM: provide runtime hook for ioremap Rob Herring
2012-02-13 22:13 ` H Hartley Sweeten
2012-02-13 22:30 ` Russell King - ARM Linux
2012-02-13 22:48 ` Rob Herring
2012-02-13 21:43 ` [PATCH 04/15] ARM: imx: convert to common runtime ioremap hook Rob Herring
2012-02-16 0:17 ` Shawn Guo
2012-02-13 21:43 ` [PATCH 05/15] ARM: msm: use " Rob Herring
2012-02-13 23:05 ` David Brown
2012-02-13 21:43 ` [PATCH 06/15] ARM: msm: clean-up mach/io.h Rob Herring
2012-02-13 21:43 ` [PATCH 07/15] ARM: at91: " Rob Herring
2012-02-14 9:21 ` Nicolas Ferre
2012-02-14 13:24 ` Rob Herring
2012-02-16 7:43 ` Jean-Christophe PLAGNIOL-VILLARD
2012-02-16 14:08 ` Rob Herring
2012-02-16 14:23 ` Jean-Christophe PLAGNIOL-VILLARD
2012-02-23 17:26 ` Nicolas Ferre
2012-02-27 16:55 ` Rob Herring
2012-02-27 17:27 ` Jean-Christophe PLAGNIOL-VILLARD
2012-02-13 21:43 ` [PATCH 08/15] ARM: davinci: remove unneeded mach/io.h include Rob Herring
2012-02-13 21:43 ` [PATCH 09/15] ARM: orion5x: clean-up mach/io.h Rob Herring
2012-02-13 21:43 ` [PATCH 10/15] ARM: tegra: " Rob Herring
2012-02-13 21:43 ` [PATCH 11/15] ARM: ep93xx: " Rob Herring
2012-02-13 21:52 ` Ryan Mallon
2012-02-13 22:15 ` Rob Herring
2012-02-13 22:16 ` H Hartley Sweeten
2012-02-27 15:17 ` [PATCH] " Rob Herring
2012-02-13 21:43 ` [PATCH 12/15] ARM: clps711x: remove unneeded include of mach/io.h Rob Herring
2012-02-13 21:43 ` [PATCH 13/15] ARM: make mach/io.h include optional Rob Herring
2012-02-13 22:14 ` H Hartley Sweeten
2012-02-13 22:36 ` Russell King - ARM Linux
2012-02-13 22:55 ` Rob Herring
2012-02-14 2:03 ` Arnd Bergmann
2012-02-14 2:54 ` Rob Herring
2012-02-14 8:04 ` Arnd Bergmann
2012-02-14 14:36 ` Rob Herring
2012-02-14 17:16 ` Arnd Bergmann
2012-02-14 17:40 ` Russell King - ARM Linux
2012-02-14 18:12 ` Arnd Bergmann
2012-02-14 23:09 ` Rob Herring
2012-02-14 23:43 ` Russell King - ARM Linux
2012-02-15 0:25 ` Nicolas Pitre
2012-02-15 14:14 ` Rob Herring
2012-02-15 0:57 ` Arnd Bergmann
2012-02-27 19:31 ` Rob Herring [this message]
2012-02-28 16:10 ` Arnd Bergmann
2012-02-27 22:31 ` Rob Herring
2012-02-28 16:32 ` Arnd Bergmann
2012-02-13 23:15 ` H Hartley Sweeten
2012-02-14 1:06 ` Arnd Bergmann
2012-02-14 17:38 ` H Hartley Sweeten
2012-02-14 18:20 ` Arnd Bergmann
2012-02-13 21:43 ` [PATCH 14/15] ARM: remove bunch of now unused mach/io.h files Rob Herring
2012-02-13 22:16 ` H Hartley Sweeten
2012-02-16 0:19 ` Shawn Guo
2012-02-16 18:57 ` Linus Walleij
2012-02-13 21:43 ` [PATCH 15/15] ARM: kill off __mem_pci Rob Herring
2012-02-13 22:22 ` [PATCH 00/15] mach/io.h cleanup and removal Tony Lindgren
2012-02-13 23:56 ` Tony Lindgren
2012-02-14 3:09 ` Rob Herring
2012-02-13 23:41 ` Tony Lindgren
2012-02-14 3:20 ` Rob Herring
2012-02-14 17:24 ` Tony Lindgren
2012-02-14 17:57 ` Arnd Bergmann
2012-02-14 18:28 ` Nicolas Pitre
2012-02-14 19:41 ` Rob Herring
2012-02-14 20:43 ` Tony Lindgren
2012-02-14 21:26 ` Arnd Bergmann
2012-02-14 21:54 ` Rob Herring
2012-02-14 22:38 ` Arnd Bergmann
2012-02-21 22:47 ` Stephen Warren
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4F4BDA24.1070807@gmail.com \
--to=robherring2@gmail.com \
--cc=linux-arm-kernel@lists.infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).