summaryrefslogtreecommitdiffstats
path: root/include/asm-arm26/io.h
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@ppc970.osdl.org>2005-04-16 15:20:36 -0700
committerLinus Torvalds <torvalds@ppc970.osdl.org>2005-04-16 15:20:36 -0700
commit1da177e4c3f41524e886b7f1b8a0c1fc7321cac2 (patch)
tree0bba044c4ce775e45a88a51686b5d9f90697ea9d /include/asm-arm26/io.h
downloadblackbird-op-linux-1da177e4c3f41524e886b7f1b8a0c1fc7321cac2.tar.gz
blackbird-op-linux-1da177e4c3f41524e886b7f1b8a0c1fc7321cac2.zip
Linux-2.6.12-rc2v2.6.12-rc2
Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
Diffstat (limited to 'include/asm-arm26/io.h')
-rw-r--r--include/asm-arm26/io.h435
1 files changed, 435 insertions, 0 deletions
diff --git a/include/asm-arm26/io.h b/include/asm-arm26/io.h
new file mode 100644
index 000000000000..02f94d88a124
--- /dev/null
+++ b/include/asm-arm26/io.h
@@ -0,0 +1,435 @@
+/*
+ * linux/include/asm-arm/io.h
+ *
+ * Copyright (C) 1996-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Modifications:
+ * 16-Sep-1996 RMK Inlined the inx/outx functions & optimised for both
+ * constant addresses and variable addresses.
+ * 04-Dec-1997 RMK Moved a lot of this stuff to the new architecture
+ * specific IO header files.
+ * 27-Mar-1999 PJB Second parameter of memcpy_toio is const..
+ * 04-Apr-1999 PJB Added check_signature.
+ * 12-Dec-1999 RMK More cleanups
+ * 18-Jun-2000 RMK Removed virt_to_* and friends definitions
+ */
+#ifndef __ASM_ARM_IO_H
+#define __ASM_ARM_IO_H
+
+#ifdef __KERNEL__
+
+#include <linux/config.h>
+#include <linux/types.h>
+#include <asm/byteorder.h>
+#include <asm/memory.h>
+#include <asm/hardware.h>
+
+/*
+ * Generic IO read/write. These perform native-endian accesses. Note
+ * that some architectures will want to re-define __raw_{read,write}w.
+ */
+extern void __raw_writesb(unsigned int addr, const void *data, int bytelen);
+extern void __raw_writesw(unsigned int addr, const void *data, int wordlen);
+extern void __raw_writesl(unsigned int addr, const void *data, int longlen);
+
+extern void __raw_readsb(unsigned int addr, void *data, int bytelen);
+extern void __raw_readsw(unsigned int addr, void *data, int wordlen);
+extern void __raw_readsl(unsigned int addr, void *data, int longlen);
+
+#define __raw_writeb(v,a) (*(volatile unsigned char *)(a) = (v))
+#define __raw_writew(v,a) (*(volatile unsigned short *)(a) = (v))
+#define __raw_writel(v,a) (*(volatile unsigned int *)(a) = (v))
+
+#define __raw_readb(a) (*(volatile unsigned char *)(a))
+#define __raw_readw(a) (*(volatile unsigned short *)(a))
+#define __raw_readl(a) (*(volatile unsigned int *)(a))
+
+
+/*
+ * Bad read/write accesses...
+ */
+extern void __readwrite_bug(const char *fn);
+
+/*
+ * Now, pick up the machine-defined IO definitions
+ */
+
+#define IO_SPACE_LIMIT 0xffffffff
+
+/*
+ * GCC is totally crap at loading/storing data. We try to persuade it
+ * to do the right thing by using these whereever possible instead of
+ * the above.
+ */
+#define __arch_base_getb(b,o) \
+ ({ \
+ unsigned int v, r = (b); \
+ __asm__ __volatile__( \
+ "ldrb %0, [%1, %2]" \
+ : "=r" (v) \
+ : "r" (r), "Ir" (o)); \
+ v; \
+ })
+
+#define __arch_base_getl(b,o) \
+ ({ \
+ unsigned int v, r = (b); \
+ __asm__ __volatile__( \
+ "ldr %0, [%1, %2]" \
+ : "=r" (v) \
+ : "r" (r), "Ir" (o)); \
+ v; \
+ })
+
+#define __arch_base_putb(v,b,o) \
+ ({ \
+ unsigned int r = (b); \
+ __asm__ __volatile__( \
+ "strb %0, [%1, %2]" \
+ : \
+ : "r" (v), "r" (r), "Ir" (o)); \
+ })
+
+#define __arch_base_putl(v,b,o) \
+ ({ \
+ unsigned int r = (b); \
+ __asm__ __volatile__( \
+ "str %0, [%1, %2]" \
+ : \
+ : "r" (v), "r" (r), "Ir" (o)); \
+ })
+
+/*
+ * We use two different types of addressing - PC style addresses, and ARM
+ * addresses. PC style accesses the PC hardware with the normal PC IO
+ * addresses, eg 0x3f8 for serial#1. ARM addresses are 0x80000000+
+ * and are translated to the start of IO. Note that all addresses are
+ * shifted left!
+ */
+#define __PORT_PCIO(x) (!((x) & 0x80000000))
+
+/*
+ * Dynamic IO functions - let the compiler
+ * optimize the expressions
+ */
+static inline void __outb (unsigned int value, unsigned int port)
+{
+ unsigned long temp;
+ __asm__ __volatile__(
+ "tst %2, #0x80000000\n\t"
+ "mov %0, %4\n\t"
+ "addeq %0, %0, %3\n\t"
+ "strb %1, [%0, %2, lsl #2] @ outb"
+ : "=&r" (temp)
+ : "r" (value), "r" (port), "Ir" (PCIO_BASE - IO_BASE), "Ir" (IO_BASE)
+ : "cc");
+}
+
+static inline void __outw (unsigned int value, unsigned int port)
+{
+ unsigned long temp;
+ __asm__ __volatile__(
+ "tst %2, #0x80000000\n\t"
+ "mov %0, %4\n\t"
+ "addeq %0, %0, %3\n\t"
+ "str %1, [%0, %2, lsl #2] @ outw"
+ : "=&r" (temp)
+ : "r" (value|value<<16), "r" (port), "Ir" (PCIO_BASE - IO_BASE), "Ir" (IO_BASE)
+ : "cc");
+}
+
+static inline void __outl (unsigned int value, unsigned int port)
+{
+ unsigned long temp;
+ __asm__ __volatile__(
+ "tst %2, #0x80000000\n\t"
+ "mov %0, %4\n\t"
+ "addeq %0, %0, %3\n\t"
+ "str %1, [%0, %2, lsl #2] @ outl"
+ : "=&r" (temp)
+ : "r" (value), "r" (port), "Ir" (PCIO_BASE - IO_BASE), "Ir" (IO_BASE)
+ : "cc");
+}
+
+#define DECLARE_DYN_IN(sz,fnsuffix,instr) \
+static inline unsigned sz __in##fnsuffix (unsigned int port) \
+{ \
+ unsigned long temp, value; \
+ __asm__ __volatile__( \
+ "tst %2, #0x80000000\n\t" \
+ "mov %0, %4\n\t" \
+ "addeq %0, %0, %3\n\t" \
+ "ldr" instr " %1, [%0, %2, lsl #2] @ in" #fnsuffix \
+ : "=&r" (temp), "=r" (value) \
+ : "r" (port), "Ir" (PCIO_BASE - IO_BASE), "Ir" (IO_BASE) \
+ : "cc"); \
+ return (unsigned sz)value; \
+}
+
+static inline unsigned int __ioaddr (unsigned int port) \
+{ \
+ if (__PORT_PCIO(port)) \
+ return (unsigned int)(PCIO_BASE + (port << 2)); \
+ else \
+ return (unsigned int)(IO_BASE + (port << 2)); \
+}
+
+#define DECLARE_IO(sz,fnsuffix,instr) \
+ DECLARE_DYN_IN(sz,fnsuffix,instr)
+
+DECLARE_IO(char,b,"b")
+DECLARE_IO(short,w,"")
+DECLARE_IO(int,l,"")
+
+#undef DECLARE_IO
+#undef DECLARE_DYN_IN
+
+/*
+ * Constant address IO functions
+ *
+ * These have to be macros for the 'J' constraint to work -
+ * +/-4096 immediate operand.
+ */
+#define __outbc(value,port) \
+({ \
+ if (__PORT_PCIO((port))) \
+ __asm__ __volatile__( \
+ "strb %0, [%1, %2] @ outbc" \
+ : : "r" (value), "r" (PCIO_BASE), "Jr" ((port) << 2)); \
+ else \
+ __asm__ __volatile__( \
+ "strb %0, [%1, %2] @ outbc" \
+ : : "r" (value), "r" (IO_BASE), "r" ((port) << 2)); \
+})
+
+#define __inbc(port) \
+({ \
+ unsigned char result; \
+ if (__PORT_PCIO((port))) \
+ __asm__ __volatile__( \
+ "ldrb %0, [%1, %2] @ inbc" \
+ : "=r" (result) : "r" (PCIO_BASE), "Jr" ((port) << 2)); \
+ else \
+ __asm__ __volatile__( \
+ "ldrb %0, [%1, %2] @ inbc" \
+ : "=r" (result) : "r" (IO_BASE), "r" ((port) << 2)); \
+ result; \
+})
+
+#define __outwc(value,port) \
+({ \
+ unsigned long v = value; \
+ if (__PORT_PCIO((port))) \
+ __asm__ __volatile__( \
+ "str %0, [%1, %2] @ outwc" \
+ : : "r" (v|v<<16), "r" (PCIO_BASE), "Jr" ((port) << 2)); \
+ else \
+ __asm__ __volatile__( \
+ "str %0, [%1, %2] @ outwc" \
+ : : "r" (v|v<<16), "r" (IO_BASE), "r" ((port) << 2)); \
+})
+
+#define __inwc(port) \
+({ \
+ unsigned short result; \
+ if (__PORT_PCIO((port))) \
+ __asm__ __volatile__( \
+ "ldr %0, [%1, %2] @ inwc" \
+ : "=r" (result) : "r" (PCIO_BASE), "Jr" ((port) << 2)); \
+ else \
+ __asm__ __volatile__( \
+ "ldr %0, [%1, %2] @ inwc" \
+ : "=r" (result) : "r" (IO_BASE), "r" ((port) << 2)); \
+ result & 0xffff; \
+})
+
+#define __outlc(value,port) \
+({ \
+ unsigned long v = value; \
+ if (__PORT_PCIO((port))) \
+ __asm__ __volatile__( \
+ "str %0, [%1, %2] @ outlc" \
+ : : "r" (v), "r" (PCIO_BASE), "Jr" ((port) << 2)); \
+ else \
+ __asm__ __volatile__( \
+ "str %0, [%1, %2] @ outlc" \
+ : : "r" (v), "r" (IO_BASE), "r" ((port) << 2)); \
+})
+
+#define __inlc(port) \
+({ \
+ unsigned long result; \
+ if (__PORT_PCIO((port))) \
+ __asm__ __volatile__( \
+ "ldr %0, [%1, %2] @ inlc" \
+ : "=r" (result) : "r" (PCIO_BASE), "Jr" ((port) << 2)); \
+ else \
+ __asm__ __volatile__( \
+ "ldr %0, [%1, %2] @ inlc" \
+ : "=r" (result) : "r" (IO_BASE), "r" ((port) << 2)); \
+ result; \
+})
+
+#define __ioaddrc(port) \
+({ \
+ unsigned long addr; \
+ if (__PORT_PCIO((port))) \
+ addr = PCIO_BASE + ((port) << 2); \
+ else \
+ addr = IO_BASE + ((port) << 2); \
+ addr; \
+})
+
+#define inb(p) (__builtin_constant_p((p)) ? __inbc(p) : __inb(p))
+#define inw(p) (__builtin_constant_p((p)) ? __inwc(p) : __inw(p))
+#define inl(p) (__builtin_constant_p((p)) ? __inlc(p) : __inl(p))
+#define outb(v,p) (__builtin_constant_p((p)) ? __outbc(v,p) : __outb(v,p))
+#define outw(v,p) (__builtin_constant_p((p)) ? __outwc(v,p) : __outw(v,p))
+#define outl(v,p) (__builtin_constant_p((p)) ? __outlc(v,p) : __outl(v,p))
+#define __ioaddr(p) (__builtin_constant_p((p)) ? __ioaddr(p) : __ioaddrc(p))
+
+/* JMA 18.02.03 added sb,sl from arm/io.h, changing io to ioaddr */
+
+#define outsb(p,d,l) __raw_writesb(__ioaddr(p),d,l)
+#define outsw(p,d,l) __raw_writesw(__ioaddr(p),d,l)
+#define outsl(p,d,l) __raw_writesl(__ioaddr(p),d,l)
+
+#define insb(p,d,l) __raw_readsb(__ioaddr(p),d,l)
+#define insw(p,d,l) __raw_readsw(__ioaddr(p),d,l)
+#define insl(p,d,l) __raw_readsl(__ioaddr(p),d,l)
+
+#define insw(p,d,l) __raw_readsw(__ioaddr(p),d,l)
+#define outsw(p,d,l) __raw_writesw(__ioaddr(p),d,l)
+
+#define readb(c) (__readwrite_bug("readb"),0)
+#define readw(c) (__readwrite_bug("readw"),0)
+#define readl(c) (__readwrite_bug("readl"),0)
+#define readb_relaxed(addr) readb(addr)
+#define readw_relaxed(addr) readw(addr)
+#define readl_relaxed(addr) readl(addr)
+#define writeb(v,c) __readwrite_bug("writeb")
+#define writew(v,c) __readwrite_bug("writew")
+#define writel(v,c) __readwrite_bug("writel")
+
+#define readsw(p,d,l) (__readwrite_bug("readsw"),0)
+#define readsl(p,d,l) (__readwrite_bug("readsl"),0)
+#define writesw(p,d,l) __readwrite_bug("writesw")
+#define writesl(p,d,l) __readwrite_bug("writesl")
+
+#define mmiowb()
+
+/* the following macro is depreciated */
+#define ioaddr(port) __ioaddr((port))
+
+/*
+ * No ioremap support here.
+ */
+#define __arch_ioremap(c,s,f,a) ((void *)(c))
+#define __arch_iounmap(c) do { } while (0)
+
+
+#if defined(__arch_putb) || defined(__arch_putw) || defined(__arch_putl) || \
+ defined(__arch_getb) || defined(__arch_getw) || defined(__arch_getl)
+#warning machine class uses old __arch_putw or __arch_getw
+#endif
+
+/*
+ * IO port access primitives
+ * -------------------------
+ *
+ * The ARM doesn't have special IO access instructions; all IO is memory
+ * mapped. Note that these are defined to perform little endian accesses
+ * only. Their primary purpose is to access PCI and ISA peripherals.
+ *
+ * Note that for a big endian machine, this implies that the following
+ * big endian mode connectivity is in place, as described by numerious
+ * ARM documents:
+ *
+ * PCI: D0-D7 D8-D15 D16-D23 D24-D31
+ * ARM: D24-D31 D16-D23 D8-D15 D0-D7
+ *
+ * The machine specific io.h include defines __io to translate an "IO"
+ * address to a memory address.
+ *
+ * Note that we prevent GCC re-ordering or caching values in expressions
+ * by introducing sequence points into the in*() definitions. Note that
+ * __raw_* do not guarantee this behaviour.
+ */
+/*
+#define outsb(p,d,l) __raw_writesb(__io(p),d,l)
+#define outsw(p,d,l) __raw_writesw(__io(p),d,l)
+
+#define insb(p,d,l) __raw_readsb(__io(p),d,l)
+#define insw(p,d,l) __raw_readsw(__io(p),d,l)
+*/
+#define outb_p(val,port) outb((val),(port))
+#define outw_p(val,port) outw((val),(port))
+#define inb_p(port) inb((port))
+#define inw_p(port) inw((port))
+#define inl_p(port) inl((port))
+
+#define outsb_p(port,from,len) outsb(port,from,len)
+#define outsw_p(port,from,len) outsw(port,from,len)
+#define insb_p(port,to,len) insb(port,to,len)
+#define insw_p(port,to,len) insw(port,to,len)
+
+/*
+ * String version of IO memory access ops:
+ */
+extern void _memcpy_fromio(void *, unsigned long, size_t);
+extern void _memcpy_toio(unsigned long, const void *, size_t);
+extern void _memset_io(unsigned long, int, size_t);
+
+/*
+ * ioremap and friends.
+ *
+ * ioremap takes a PCI memory address, as specified in
+ * Documentation/IO-mapping.txt.
+ */
+extern void * __ioremap(unsigned long, size_t, unsigned long, unsigned long);
+extern void __iounmap(void *addr);
+
+#ifndef __arch_ioremap
+#define ioremap(cookie,size) __ioremap(cookie,size,0,1)
+#define ioremap_nocache(cookie,size) __ioremap(cookie,size,0,1)
+#define iounmap(cookie) __iounmap(cookie)
+#else
+#define ioremap(cookie,size) __arch_ioremap((cookie),(size),0,1)
+#define ioremap_nocache(cookie,size) __arch_ioremap((cookie),(size),0,1)
+#define iounmap(cookie) __arch_iounmap(cookie)
+#endif
+
+/*
+ * DMA-consistent mapping functions. These allocate/free a region of
+ * uncached, unwrite-buffered mapped memory space for use with DMA
+ * devices. This is the "generic" version. The PCI specific version
+ * is in pci.h
+ */
+extern void *consistent_alloc(int gfp, size_t size, dma_addr_t *handle);
+extern void consistent_free(void *vaddr, size_t size, dma_addr_t handle);
+extern void consistent_sync(void *vaddr, size_t size, int rw);
+
+/*
+ * can the hardware map this into one segment or not, given no other
+ * constraints.
+ */
+#define BIOVEC_MERGEABLE(vec1, vec2) \
+ ((bvec_to_phys((vec1)) + (vec1)->bv_len) == bvec_to_phys((vec2)))
+
+/*
+ * 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_ARM_IO_H */
OpenPOWER on IntegriCloud