| Message ID | 20230216123419.461016-13-bhe@redhat.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | None | expand |
On Thu, Feb 16, 2023 at 08:34:15PM +0800, Baoquan He wrote: > Meanwhile, add macro ARCH_HAS_IOREMAP_WC since the added ioremap_wc() > will conflict with the one in include/asm-generic/iomap.h, then an > compiling error is seen: Huh? ARCH_HAS_IOREMAP_WC comes up nowhere else in this patchset, and the current definition of ioremap_wc() is guarded by an ifndef ioremap_wc > +#define ioremap_wc(addr, size) \ > + ioremap_prot((addr), (size), _PAGE_IOREMAP) This should be enough all by itself.
On 02/16/23 at 01:50pm, Matthew Wilcox wrote: > On Thu, Feb 16, 2023 at 08:34:15PM +0800, Baoquan He wrote: > > Meanwhile, add macro ARCH_HAS_IOREMAP_WC since the added ioremap_wc() > > will conflict with the one in include/asm-generic/iomap.h, then an > > compiling error is seen: > Thanks for reviewing. > Huh? ARCH_HAS_IOREMAP_WC comes up nowhere else in this patchset, and > the current definition of ioremap_wc() is guarded by an ifndef ioremap_wc Because another patch of powerpc has got ARCH_HAS_IOREMAP_WC in the existed code. > > > +#define ioremap_wc(addr, size) \ > > + ioremap_prot((addr), (size), _PAGE_IOREMAP) > > This should be enough all by itself. It's not if including asm-generic/iomap.h. The ARCH_HAS_IOREMAP_xx is to avoid redefinition there. include/asm-generic/iomap.h: ---- #ifndef ARCH_HAS_IOREMAP_WC #define ioremap_wc ioremap #endif >
On Thu, Feb 16, 2023, at 16:02, Baoquan He wrote: > On 02/16/23 at 01:50pm, Matthew Wilcox wrote: > It's not if including asm-generic/iomap.h. The ARCH_HAS_IOREMAP_xx is to > avoid redefinition there. > > include/asm-generic/iomap.h: > ---- > #ifndef ARCH_HAS_IOREMAP_WC > #define ioremap_wc ioremap > #endif I'd change that to the usual '#ifndef ioremap_wc' in that case. Arnd
On 02/16/23 at 04:18pm, Arnd Bergmann wrote: > On Thu, Feb 16, 2023, at 16:02, Baoquan He wrote: > > On 02/16/23 at 01:50pm, Matthew Wilcox wrote: > > It's not if including asm-generic/iomap.h. The ARCH_HAS_IOREMAP_xx is to > > avoid redefinition there. > > > > include/asm-generic/iomap.h: > > ---- > > #ifndef ARCH_HAS_IOREMAP_WC > > #define ioremap_wc ioremap > > #endif > > I'd change that to the usual '#ifndef ioremap_wc' in that case. Not sure if I got you. Kill all ARCH_HAS_IOREMAP_xxx in kernel? If yes, sounds like a good idea.
Le 17/02/2023 à 14:31, Baoquan He a écrit : > On 02/16/23 at 04:18pm, Arnd Bergmann wrote: >> On Thu, Feb 16, 2023, at 16:02, Baoquan He wrote: >>> On 02/16/23 at 01:50pm, Matthew Wilcox wrote: >>> It's not if including asm-generic/iomap.h. The ARCH_HAS_IOREMAP_xx is to >>> avoid redefinition there. >>> >>> include/asm-generic/iomap.h: >>> ---- >>> #ifndef ARCH_HAS_IOREMAP_WC >>> #define ioremap_wc ioremap >>> #endif >> >> I'd change that to the usual '#ifndef ioremap_wc' in that case. > > Not sure if I got you. Kill all ARCH_HAS_IOREMAP_xxx in kernel? If yes, > sounds like a good idea. > At least kill that one at the first place in your series, and then the other ones in a follow-up series maybe.
On 02/17/23 at 01:46pm, Christophe Leroy wrote: > > > Le 17/02/2023 à 14:31, Baoquan He a écrit : > > On 02/16/23 at 04:18pm, Arnd Bergmann wrote: > >> On Thu, Feb 16, 2023, at 16:02, Baoquan He wrote: > >>> On 02/16/23 at 01:50pm, Matthew Wilcox wrote: > >>> It's not if including asm-generic/iomap.h. The ARCH_HAS_IOREMAP_xx is to > >>> avoid redefinition there. > >>> > >>> include/asm-generic/iomap.h: > >>> ---- > >>> #ifndef ARCH_HAS_IOREMAP_WC > >>> #define ioremap_wc ioremap > >>> #endif > >> > >> I'd change that to the usual '#ifndef ioremap_wc' in that case. > > > > Not sure if I got you. Kill all ARCH_HAS_IOREMAP_xxx in kernel? If yes, > > sounds like a good idea. > > > > At least kill that one at the first place in your series, and then the > other ones in a follow-up series maybe. Then we can make a preparation patch to change that in iomap.h, then remove all ARCH_HAS_IOREMAP_WC definition in arch. I thought to let this patch as is, then get rid of all ARCH_HAS_IOREMAP_xxx in a follow-up series. While the former is also fine to me. Thanks, Christophe. diff --git a/include/asm-generic/iomap.h b/include/asm-generic/iomap.h index 08237ae8b840..5fa1e9ca951c 100644 --- a/include/asm-generic/iomap.h +++ b/include/asm-generic/iomap.h @@ -93,7 +93,7 @@ extern void __iomem *ioport_map(unsigned long port, unsigned int nr); extern void ioport_unmap(void __iomem *); #endif -#ifndef ARCH_HAS_IOREMAP_WC +#ifndef ioremap_wc #define ioremap_wc ioremap #endif [~]$ git grep ARCH_HAS_IOREMAP_WC arch/loongarch/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC arch/mips/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC arch/powerpc/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC arch/x86/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC drivers/net/ethernet/sfc/io.h:#ifdef ARCH_HAS_IOREMAP_WC drivers/net/ethernet/sfc/siena/io.h:#ifdef ARCH_HAS_IOREMAP_WC include/asm-generic/iomap.h:#ifndef ARCH_HAS_IOREMAP_WC
Le 17/02/2023 à 15:21, Baoquan He a écrit : > On 02/17/23 at 01:46pm, Christophe Leroy wrote: >> >> >> Le 17/02/2023 à 14:31, Baoquan He a écrit : >>> On 02/16/23 at 04:18pm, Arnd Bergmann wrote: >>>> On Thu, Feb 16, 2023, at 16:02, Baoquan He wrote: >>>>> On 02/16/23 at 01:50pm, Matthew Wilcox wrote: >>>>> It's not if including asm-generic/iomap.h. The ARCH_HAS_IOREMAP_xx is to >>>>> avoid redefinition there. >>>>> >>>>> include/asm-generic/iomap.h: >>>>> ---- >>>>> #ifndef ARCH_HAS_IOREMAP_WC >>>>> #define ioremap_wc ioremap >>>>> #endif >>>> >>>> I'd change that to the usual '#ifndef ioremap_wc' in that case. >>> >>> Not sure if I got you. Kill all ARCH_HAS_IOREMAP_xxx in kernel? If yes, >>> sounds like a good idea. >>> >> >> At least kill that one at the first place in your series, and then the >> other ones in a follow-up series maybe. > > Then we can make a preparation patch to change that in iomap.h, then > remove all ARCH_HAS_IOREMAP_WC definition in arch. I thought to let this > patch as is, then get rid of all ARCH_HAS_IOREMAP_xxx in a follow-up > series. While the former is also fine to me. Thanks, Christophe. > > diff --git a/include/asm-generic/iomap.h b/include/asm-generic/iomap.h > index 08237ae8b840..5fa1e9ca951c 100644 > --- a/include/asm-generic/iomap.h > +++ b/include/asm-generic/iomap.h > @@ -93,7 +93,7 @@ extern void __iomem *ioport_map(unsigned long port, unsigned int nr); > extern void ioport_unmap(void __iomem *); > #endif > > -#ifndef ARCH_HAS_IOREMAP_WC > +#ifndef ioremap_wc > #define ioremap_wc ioremap > #endif > > [~]$ git grep ARCH_HAS_IOREMAP_WC > arch/loongarch/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > arch/mips/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > arch/powerpc/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > arch/x86/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > drivers/net/ethernet/sfc/io.h:#ifdef ARCH_HAS_IOREMAP_WC > drivers/net/ethernet/sfc/siena/io.h:#ifdef ARCH_HAS_IOREMAP_WC > include/asm-generic/iomap.h:#ifndef ARCH_HAS_IOREMAP_WC > > Not so many: $ git grep ARCH_HAS_IOREMAP_WC | grep define arch/loongarch/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC arch/mips/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC arch/powerpc/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC arch/x86/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC And also make sure they define ioremap_wc : $ git grep "define ioremap_wc" `git grep -l "define ARCH_HAS_IOREMAP_WC"` arch/loongarch/include/asm/io.h:#define ioremap_wc(offset, size) \ arch/mips/include/asm/io.h:#define ioremap_wc(offset, size) \ arch/powerpc/include/asm/io.h:#define ioremap_wc ioremap_wc arch/x86/include/asm/io.h:#define ioremap_wc ioremap_wc Christophe
Le 17/02/2023 à 15:33, Christophe Leroy a écrit : > > > Le 17/02/2023 à 15:21, Baoquan He a écrit : >> On 02/17/23 at 01:46pm, Christophe Leroy wrote: >>> >>> >>> Le 17/02/2023 à 14:31, Baoquan He a écrit : >>>> On 02/16/23 at 04:18pm, Arnd Bergmann wrote: >>>>> On Thu, Feb 16, 2023, at 16:02, Baoquan He wrote: >>>>>> On 02/16/23 at 01:50pm, Matthew Wilcox wrote: >>>>>> It's not if including asm-generic/iomap.h. The ARCH_HAS_IOREMAP_xx >>>>>> is to >>>>>> avoid redefinition there. >>>>>> >>>>>> include/asm-generic/iomap.h: >>>>>> ---- >>>>>> #ifndef ARCH_HAS_IOREMAP_WC >>>>>> #define ioremap_wc ioremap >>>>>> #endif >>>>> >>>>> I'd change that to the usual '#ifndef ioremap_wc' in that case. >>>> >>>> Not sure if I got you. Kill all ARCH_HAS_IOREMAP_xxx in kernel? If yes, >>>> sounds like a good idea. >>>> >>> >>> At least kill that one at the first place in your series, and then the >>> other ones in a follow-up series maybe. >> >> Then we can make a preparation patch to change that in iomap.h, then >> remove all ARCH_HAS_IOREMAP_WC definition in arch. I thought to let this >> patch as is, then get rid of all ARCH_HAS_IOREMAP_xxx in a follow-up >> series. While the former is also fine to me. Thanks, Christophe. >> >> diff --git a/include/asm-generic/iomap.h b/include/asm-generic/iomap.h >> index 08237ae8b840..5fa1e9ca951c 100644 >> --- a/include/asm-generic/iomap.h >> +++ b/include/asm-generic/iomap.h >> @@ -93,7 +93,7 @@ extern void __iomem *ioport_map(unsigned long port, >> unsigned int nr); >> extern void ioport_unmap(void __iomem *); >> #endif >> -#ifndef ARCH_HAS_IOREMAP_WC >> +#ifndef ioremap_wc >> #define ioremap_wc ioremap >> #endif >> >> [~]$ git grep ARCH_HAS_IOREMAP_WC >> arch/loongarch/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC >> arch/mips/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC >> arch/powerpc/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC >> arch/x86/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC >> drivers/net/ethernet/sfc/io.h:#ifdef ARCH_HAS_IOREMAP_WC >> drivers/net/ethernet/sfc/siena/io.h:#ifdef ARCH_HAS_IOREMAP_WC >> include/asm-generic/iomap.h:#ifndef ARCH_HAS_IOREMAP_WC >> >> > > Not so many: > > $ git grep ARCH_HAS_IOREMAP_WC | grep define > arch/loongarch/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > arch/mips/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > arch/powerpc/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > arch/x86/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > > And also make sure they define ioremap_wc : > > $ git grep "define ioremap_wc" `git grep -l "define ARCH_HAS_IOREMAP_WC"` > arch/loongarch/include/asm/io.h:#define ioremap_wc(offset, size) \ > arch/mips/include/asm/io.h:#define ioremap_wc(offset, size) > \ > arch/powerpc/include/asm/io.h:#define ioremap_wc ioremap_wc > arch/x86/include/asm/io.h:#define ioremap_wc ioremap_wc > By the way there are so few, you can make all at once: $ git grep ARCH_HAS_IOREMAP_ | grep define arch/loongarch/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC arch/m68k/include/asm/kmap.h:#define ARCH_HAS_IOREMAP_WT arch/mips/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC arch/powerpc/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC arch/powerpc/include/asm/io.h:#define ARCH_HAS_IOREMAP_WT arch/x86/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC arch/x86/include/asm/io.h:#define ARCH_HAS_IOREMAP_WT Christophe
On 02/17/23 at 02:35pm, Christophe Leroy wrote: ...... > > Not so many: > > > > $ git grep ARCH_HAS_IOREMAP_WC | grep define > > arch/loongarch/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > > arch/mips/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > > arch/powerpc/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > > arch/x86/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > > > > And also make sure they define ioremap_wc : > > > > $ git grep "define ioremap_wc" `git grep -l "define ARCH_HAS_IOREMAP_WC"` > > arch/loongarch/include/asm/io.h:#define ioremap_wc(offset, size) \ > > arch/mips/include/asm/io.h:#define ioremap_wc(offset, size) > > \ > > arch/powerpc/include/asm/io.h:#define ioremap_wc ioremap_wc > > arch/x86/include/asm/io.h:#define ioremap_wc ioremap_wc > > > > > By the way there are so few, you can make all at once: OK, I will try to pack this into one preparation patch. Thanks. > > $ git grep ARCH_HAS_IOREMAP_ | grep define > arch/loongarch/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > arch/m68k/include/asm/kmap.h:#define ARCH_HAS_IOREMAP_WT > arch/mips/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > arch/powerpc/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > arch/powerpc/include/asm/io.h:#define ARCH_HAS_IOREMAP_WT > arch/x86/include/asm/io.h:#define ARCH_HAS_IOREMAP_WC > arch/x86/include/asm/io.h:#define ARCH_HAS_IOREMAP_WT > > Christophe
Hi Christophe, Arnd, On 02/17/23 at 01:46pm, Christophe Leroy wrote: > > > Le 17/02/2023 à 14:31, Baoquan He a écrit : > > On 02/16/23 at 04:18pm, Arnd Bergmann wrote: > >> On Thu, Feb 16, 2023, at 16:02, Baoquan He wrote: > >>> On 02/16/23 at 01:50pm, Matthew Wilcox wrote: > >>> It's not if including asm-generic/iomap.h. The ARCH_HAS_IOREMAP_xx is to > >>> avoid redefinition there. > >>> > >>> include/asm-generic/iomap.h: > >>> ---- > >>> #ifndef ARCH_HAS_IOREMAP_WC > >>> #define ioremap_wc ioremap > >>> #endif > >> > >> I'd change that to the usual '#ifndef ioremap_wc' in that case. > > > > Not sure if I got you. Kill all ARCH_HAS_IOREMAP_xxx in kernel? If yes, > > sounds like a good idea. > > > > At least kill that one at the first place in your series, and then the > other ones in a follow-up series maybe. I made a patch to remove all ARCH_HAS_IOREMAP_xx macros in architectures and the ifdeffery of ARCH_HAS_IOREMAP_xx in asm-generic/iomap.h. But the change will cause building error as below. Becuase we usually have '#include <asm-generic/iomap.h>' at the beginning of arch/xx/include/asm/io.h, and have '#include <asm-generic/io.h>' at the end of arch/xx/include/asm/io.h. For architecutres which has ARCH_HAS_IOREMAP_xx defining, we need move ''#include <asm-generic/iomap.h> dowe to below '#include <asm-generic/io.h>'. Please help check if it's still worth doing. ***move '#include <asm-generic/iomap.h>' below '#include <asm-generic/io.h>' *** diff --git a/arch/x86/include/asm/io.h b/arch/x86/include/asm/io.h index 8ab68cde1f13..a8d55fc62959 100644 --- a/arch/x86/include/asm/io.h +++ b/arch/x86/include/asm/io.h @@ -209,8 +209,6 @@ void memset_io(volatile void __iomem *, int, size_t); #define memcpy_toio memcpy_toio #define memset_io memset_io -#include <asm-generic/iomap.h> - /* * ISA space is 'always mapped' on a typical x86 system, no need to * explicitly ioremap() it. The fact that the ISA IO space is mapped @@ -329,6 +327,8 @@ extern bool is_early_ioremap_ptep(pte_t *ptep); #include <asm-generic/io.h> #undef PCI_IOBASE +#include <asm-generic/iomap.h> + #ifdef CONFIG_MTRR extern int __must_check arch_phys_wc_index(int handle); #define arch_phys_wc_index arch_phys_wc_index ***Building error after removing ARCH_HAS_IOREMAP_xx *** In file included from ./include/linux/io.h:13, from ./include/linux/irq.h:20, from ./include/xen/events.h:6, from arch/x86/entry/common.c:25: ./arch/x86/include/asm/io.h:321: warning: "ioremap_wc" redefined 321 | #define ioremap_wc ioremap_wc | In file included from ./arch/x86/include/asm/io.h:212: ./include/asm-generic/iomap.h:97: note: this is the location of the previous definition 97 | #define ioremap_wc ioremap | ./arch/x86/include/asm/io.h:323: warning: "ioremap_wt" redefined 323 | #define ioremap_wt ioremap_wt | ./include/asm-generic/iomap.h:101: note: this is the location of the previous definition 101 | #define ioremap_wt ioremap
diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig index a98940e64243..0ed18e673aba 100644 --- a/arch/parisc/Kconfig +++ b/arch/parisc/Kconfig @@ -36,6 +36,7 @@ config PARISC select GENERIC_ATOMIC64 if !64BIT select GENERIC_IRQ_PROBE select GENERIC_PCI_IOMAP + select GENERIC_IOREMAP select ARCH_HAVE_NMI_SAFE_CMPXCHG select GENERIC_SMP_IDLE_THREAD select GENERIC_ARCH_TOPOLOGY if SMP diff --git a/arch/parisc/include/asm/io.h b/arch/parisc/include/asm/io.h index c05e781be2f5..4c386b0cee08 100644 --- a/arch/parisc/include/asm/io.h +++ b/arch/parisc/include/asm/io.h @@ -2,6 +2,8 @@ #ifndef _ASM_IO_H #define _ASM_IO_H +#define ARCH_HAS_IOREMAP_WC + #include <linux/types.h> #include <linux/pgtable.h> @@ -125,12 +127,17 @@ static inline void gsc_writeq(unsigned long long val, unsigned long addr) /* * The standard PCI ioremap interfaces */ -void __iomem *ioremap(unsigned long offset, unsigned long size); -#define ioremap_wc ioremap -#define ioremap_uc ioremap -#define pci_iounmap pci_iounmap +#define ioremap_prot ioremap_prot + +#define _PAGE_IOREMAP (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | \ + _PAGE_ACCESSED | _PAGE_NO_CACHE) -extern void iounmap(const volatile void __iomem *addr); +#define ioremap_wc(addr, size) \ + ioremap_prot((addr), (size), _PAGE_IOREMAP) +#define ioremap_uc(addr, size) \ + ioremap_prot((addr), (size), _PAGE_IOREMAP) + +#define pci_iounmap pci_iounmap void memset_io(volatile void __iomem *addr, unsigned char val, int count); void memcpy_fromio(void *dst, const volatile void __iomem *src, int count); diff --git a/arch/parisc/mm/ioremap.c b/arch/parisc/mm/ioremap.c index 345ff0b66499..fd996472dfe7 100644 --- a/arch/parisc/mm/ioremap.c +++ b/arch/parisc/mm/ioremap.c @@ -13,25 +13,9 @@ #include <linux/io.h> #include <linux/mm.h> -/* - * Generic mapping function (not visible outside): - */ - -/* - * Remap an arbitrary physical address space into the kernel virtual - * address space. - * - * NOTE! We need to allow non-page-aligned mappings too: we will obviously - * have to convert them into an offset in a page-aligned mapping, but the - * caller shouldn't need to know that small detail. - */ -void __iomem *ioremap(unsigned long phys_addr, unsigned long size) +void __iomem *ioremap_prot(phys_addr_t phys_addr, size_t size, + unsigned long prot) { - void __iomem *addr; - struct vm_struct *area; - unsigned long offset, last_addr; - pgprot_t pgprot; - #ifdef CONFIG_EISA unsigned long end = phys_addr + size - 1; /* Support EISA addresses */ @@ -40,11 +24,6 @@ void __iomem *ioremap(unsigned long phys_addr, unsigned long size) phys_addr |= F_EXTEND(0xfc000000); #endif - /* Don't allow wraparound or zero size */ - last_addr = phys_addr + size - 1; - if (!size || last_addr < phys_addr) - return NULL; - /* * Don't allow anybody to remap normal RAM that we're using.. */ @@ -62,39 +41,6 @@ void __iomem *ioremap(unsigned long phys_addr, unsigned long size) } } - pgprot = __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | - _PAGE_ACCESSED | _PAGE_NO_CACHE); - - /* - * Mappings have to be page-aligned - */ - offset = phys_addr & ~PAGE_MASK; - phys_addr &= PAGE_MASK; - size = PAGE_ALIGN(last_addr + 1) - phys_addr; - - /* - * Ok, go for it.. - */ - area = get_vm_area(size, VM_IOREMAP); - if (!area) - return NULL; - - addr = (void __iomem *) area->addr; - if (ioremap_page_range((unsigned long)addr, (unsigned long)addr + size, - phys_addr, pgprot)) { - vunmap(addr); - return NULL; - } - - return (void __iomem *) (offset + (char __iomem *)addr); -} -EXPORT_SYMBOL(ioremap); - -void iounmap(const volatile void __iomem *io_addr) -{ - unsigned long addr = (unsigned long)io_addr & PAGE_MASK; - - if (is_vmalloc_addr((void *)addr)) - vunmap((void *)addr); + return generic_ioremap_prot(phys_addr, size, __pgprot(prot)); } -EXPORT_SYMBOL(iounmap); +EXPORT_SYMBOL(ioremap_prot);
By taking GENERIC_IOREMAP method, the generic generic_ioremap_prot(), generic_iounmap(), and their generic wrapper ioremap_prot(), ioremap() and iounmap() are all visible and available to arch. Arch needs to provide wrapper functions to override the generic versions if there's arch specific handling in its ioremap_prot(), ioremap() or iounmap(). This change will simplify implementation by removing duplicated codes with generic_ioremap_prot() and generic_iounmap(), and has the equivalent functioality as before. Here, add wrapper function ioremap_prot() for parisc's special operation when iounmap(). Meanwhile, add macro ARCH_HAS_IOREMAP_WC since the added ioremap_wc() will conflict with the one in include/asm-generic/iomap.h, then an compiling error is seen: ./include/asm-generic/iomap.h:97: warning: "ioremap_wc" redefined 97 | #define ioremap_wc ioremap And benefit from the commit 437b6b35362b ("parisc: Use the generic IO helpers"), those macros don't need be added any more. Signed-off-by: Baoquan He <bhe@redhat.com> Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com> Cc: Helge Deller <deller@gmx.de> Cc: linux-parisc@vger.kernel.org --- arch/parisc/Kconfig | 1 + arch/parisc/include/asm/io.h | 17 +++++++--- arch/parisc/mm/ioremap.c | 62 +++--------------------------------- 3 files changed, 17 insertions(+), 63 deletions(-)