Revision 4ceb5db9757aaeadcf8fbbf97d76bd42aa4df0d6 authored by Linus Torvalds on 01 August 2005, 18:14:49 UTC, committed by Linus Torvalds on 01 August 2005, 18:14:49 UTC
There's no real guarantee that handle_mm_fault() will always be able to break a COW situation - if an update from another thread ends up modifying the page table some way, handle_mm_fault() may end up requiring us to re-try the operation. That's normally fine, but get_user_pages() ended up re-trying it as a read, and thus a write access could in theory end up losing the dirty bit or be done on a page that had not been properly COW'ed. This makes get_user_pages() always retry write accesses as write accesses by making "follow_page()" require that a writable follow has the dirty bit set. That simplifies the code and solves the race: if the COW break fails for some reason, we'll just loop around and try again. Signed-off-by: Linus Torvalds <torvalds@osdl.org>
1 parent 8d894c4
io.h
/* io.h: FRV I/O operations
*
* Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
* Written by David Howells (dhowells@redhat.com)
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version
* 2 of the License, or (at your option) any later version.
*
* This gets interesting when talking to the PCI bus - the CPU is in big endian
* mode, the PCI bus is little endian and the hardware in the middle can do
* byte swapping
*/
#ifndef _ASM_IO_H
#define _ASM_IO_H
#ifdef __KERNEL__
#include <linux/config.h>
#include <asm/virtconvert.h>
#include <asm/string.h>
#include <asm/mb-regs.h>
#include <linux/delay.h>
/*
* swap functions are sometimes needed to interface little-endian hardware
*/
static inline unsigned short _swapw(unsigned short v)
{
return ((v << 8) | (v >> 8));
}
static inline unsigned long _swapl(unsigned long v)
{
return ((v << 24) | ((v & 0xff00) << 8) | ((v & 0xff0000) >> 8) | (v >> 24));
}
//#define __iormb() asm volatile("membar")
//#define __iowmb() asm volatile("membar")
#define __raw_readb(addr) __builtin_read8((void *) (addr))
#define __raw_readw(addr) __builtin_read16((void *) (addr))
#define __raw_readl(addr) __builtin_read32((void *) (addr))
#define __raw_writeb(datum, addr) __builtin_write8((void *) (addr), datum)
#define __raw_writew(datum, addr) __builtin_write16((void *) (addr), datum)
#define __raw_writel(datum, addr) __builtin_write32((void *) (addr), datum)
static inline void io_outsb(unsigned int addr, const void *buf, int len)
{
unsigned long __ioaddr = (unsigned long) addr;
const uint8_t *bp = buf;
while (len--)
__builtin_write8((volatile void __iomem *) __ioaddr, *bp++);
}
static inline void io_outsw(unsigned int addr, const void *buf, int len)
{
unsigned long __ioaddr = (unsigned long) addr;
const uint16_t *bp = buf;
while (len--)
__builtin_write16((volatile void __iomem *) __ioaddr, (*bp++));
}
extern void __outsl_ns(unsigned int addr, const void *buf, int len);
extern void __outsl_sw(unsigned int addr, const void *buf, int len);
static inline void __outsl(unsigned int addr, const void *buf, int len, int swap)
{
unsigned long __ioaddr = (unsigned long) addr;
if (!swap)
__outsl_ns(__ioaddr, buf, len);
else
__outsl_sw(__ioaddr, buf, len);
}
static inline void io_insb(unsigned long addr, void *buf, int len)
{
uint8_t *bp = buf;
while (len--)
*bp++ = __builtin_read8((volatile void __iomem *) addr);
}
static inline void io_insw(unsigned long addr, void *buf, int len)
{
uint16_t *bp = buf;
while (len--)
*bp++ = __builtin_read16((volatile void __iomem *) addr);
}
extern void __insl_ns(unsigned long addr, void *buf, int len);
extern void __insl_sw(unsigned long addr, void *buf, int len);
static inline void __insl(unsigned long addr, void *buf, int len, int swap)
{
if (!swap)
__insl_ns(addr, buf, len);
else
__insl_sw(addr, buf, len);
}
/*
* make the short names macros so specific devices
* can override them as required
*/
static inline void memset_io(volatile void __iomem *addr, unsigned char val, int count)
{
memset((void __force *) addr, val, count);
}
static inline void memcpy_fromio(void *dst, volatile void __iomem *src, int count)
{
memcpy(dst, (void __force *) src, count);
}
static inline void memcpy_toio(volatile void __iomem *dst, const void *src, int count)
{
memcpy((void __force *) dst, src, count);
}
static inline uint8_t inb(unsigned long addr)
{
return __builtin_read8((void *)addr);
}
static inline uint16_t inw(unsigned long addr)
{
uint16_t ret = __builtin_read16((void *)addr);
if (__is_PCI_IO(addr))
ret = _swapw(ret);
return ret;
}
static inline uint32_t inl(unsigned long addr)
{
uint32_t ret = __builtin_read32((void *)addr);
if (__is_PCI_IO(addr))
ret = _swapl(ret);
return ret;
}
static inline void outb(uint8_t datum, unsigned long addr)
{
__builtin_write8((void *)addr, datum);
}
static inline void outw(uint16_t datum, unsigned long addr)
{
if (__is_PCI_IO(addr))
datum = _swapw(datum);
__builtin_write16((void *)addr, datum);
}
static inline void outl(uint32_t datum, unsigned long addr)
{
if (__is_PCI_IO(addr))
datum = _swapl(datum);
__builtin_write32((void *)addr, datum);
}
#define inb_p(addr) inb(addr)
#define inw_p(addr) inw(addr)
#define inl_p(addr) inl(addr)
#define outb_p(x,addr) outb(x,addr)
#define outw_p(x,addr) outw(x,addr)
#define outl_p(x,addr) outl(x,addr)
#define outsb(a,b,l) io_outsb(a,b,l)
#define outsw(a,b,l) io_outsw(a,b,l)
#define outsl(a,b,l) __outsl(a,b,l,0)
#define insb(a,b,l) io_insb(a,b,l)
#define insw(a,b,l) io_insw(a,b,l)
#define insl(a,b,l) __insl(a,b,l,0)
#define IO_SPACE_LIMIT 0xffffffff
static inline uint8_t readb(const volatile void __iomem *addr)
{
return __builtin_read8((volatile uint8_t __force *) addr);
}
static inline uint16_t readw(const volatile void __iomem *addr)
{
uint16_t ret = __builtin_read16((volatile uint16_t __force *)addr);
if (__is_PCI_MEM(addr))
ret = _swapw(ret);
return ret;
}
static inline uint32_t readl(const volatile void __iomem *addr)
{
uint32_t ret = __builtin_read32((volatile uint32_t __force *)addr);
if (__is_PCI_MEM(addr))
ret = _swapl(ret);
return ret;
}
static inline void writeb(uint8_t datum, volatile void __iomem *addr)
{
__builtin_write8((volatile uint8_t __force *) addr, datum);
if (__is_PCI_MEM(addr))
__flush_PCI_writes();
}
static inline void writew(uint16_t datum, volatile void __iomem *addr)
{
if (__is_PCI_MEM(addr))
datum = _swapw(datum);
__builtin_write16((volatile uint16_t __force *) addr, datum);
if (__is_PCI_MEM(addr))
__flush_PCI_writes();
}
static inline void writel(uint32_t datum, volatile void __iomem *addr)
{
if (__is_PCI_MEM(addr))
datum = _swapl(datum);
__builtin_write32((volatile uint32_t __force *) addr, datum);
if (__is_PCI_MEM(addr))
__flush_PCI_writes();
}
/* Values for nocacheflag and cmode */
#define IOMAP_FULL_CACHING 0
#define IOMAP_NOCACHE_SER 1
#define IOMAP_NOCACHE_NONSER 2
#define IOMAP_WRITETHROUGH 3
extern void __iomem *__ioremap(unsigned long physaddr, unsigned long size, int cacheflag);
extern void __iounmap(void __iomem *addr, unsigned long size);
static inline void __iomem *ioremap(unsigned long physaddr, unsigned long size)
{
return __ioremap(physaddr, size, IOMAP_NOCACHE_SER);
}
static inline void __iomem *ioremap_nocache(unsigned long physaddr, unsigned long size)
{
return __ioremap(physaddr, size, IOMAP_NOCACHE_SER);
}
static inline void __iomem *ioremap_writethrough(unsigned long physaddr, unsigned long size)
{
return __ioremap(physaddr, size, IOMAP_WRITETHROUGH);
}
static inline void __iomem *ioremap_fullcache(unsigned long physaddr, unsigned long size)
{
return __ioremap(physaddr, size, IOMAP_FULL_CACHING);
}
extern void iounmap(void __iomem *addr);
static inline void flush_write_buffers(void)
{
__asm__ __volatile__ ("membar" : : :"memory");
}
/*
* Convert a physical pointer to a virtual kernel pointer for /dev/mem
* access
*/
#define xlate_dev_mem_ptr(p) __va(p)
/*
* Convert a virtual cached pointer to an uncached pointer
*/
#define xlate_dev_kmem_ptr(p) p
#endif /* __KERNEL__ */
#endif /* _ASM_IO_H */
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...