[PATCH] xtensa: Architecture support for Tensilica Xtensa Part 6
The attached patches provides part 6 of an architecture implementation for the
Tensilica Xtensa CPU series.
Signed-off-by: Chris Zankel <chris@zankel.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
diff --git a/include/asm-xtensa/uaccess.h b/include/asm-xtensa/uaccess.h
new file mode 100644
index 0000000..35576b2
--- /dev/null
+++ b/include/asm-xtensa/uaccess.h
@@ -0,0 +1,532 @@
+/*
+ * include/asm-xtensa/uaccess.h
+ *
+ * User space memory access functions
+ *
+ * These routines provide basic accessing functions to the user memory
+ * space for the kernel. This header file provides fuctions such as:
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_UACCESS_H
+#define _XTENSA_UACCESS_H
+
+#include <linux/errno.h>
+
+#define VERIFY_READ 0
+#define VERIFY_WRITE 1
+
+#ifdef __ASSEMBLY__
+
+#define _ASMLANGUAGE
+#include <asm/current.h>
+#include <asm/offsets.h>
+#include <asm/processor.h>
+
+/*
+ * These assembly macros mirror the C macros that follow below. They
+ * should always have identical functionality. See
+ * arch/xtensa/kernel/sys.S for usage.
+ */
+
+#define KERNEL_DS 0
+#define USER_DS 1
+
+#define get_ds (KERNEL_DS)
+
+/*
+ * get_fs reads current->thread.current_ds into a register.
+ * On Entry:
+ * <ad> anything
+ * <sp> stack
+ * On Exit:
+ * <ad> contains current->thread.current_ds
+ */
+ .macro get_fs ad, sp
+ GET_CURRENT(\ad,\sp)
+ l32i \ad, \ad, THREAD_CURRENT_DS
+ .endm
+
+/*
+ * set_fs sets current->thread.current_ds to some value.
+ * On Entry:
+ * <at> anything (temp register)
+ * <av> value to write
+ * <sp> stack
+ * On Exit:
+ * <at> destroyed (actually, current)
+ * <av> preserved, value to write
+ */
+ .macro set_fs at, av, sp
+ GET_CURRENT(\at,\sp)
+ s32i \av, \at, THREAD_CURRENT_DS
+ .endm
+
+/*
+ * kernel_ok determines whether we should bypass addr/size checking.
+ * See the equivalent C-macro version below for clarity.
+ * On success, kernel_ok branches to a label indicated by parameter
+ * <success>. This implies that the macro falls through to the next
+ * insruction on an error.
+ *
+ * Note that while this macro can be used independently, we designed
+ * in for optimal use in the access_ok macro below (i.e., we fall
+ * through on error).
+ *
+ * On Entry:
+ * <at> anything (temp register)
+ * <success> label to branch to on success; implies
+ * fall-through macro on error
+ * <sp> stack pointer
+ * On Exit:
+ * <at> destroyed (actually, current->thread.current_ds)
+ */
+
+#if ((KERNEL_DS != 0) || (USER_DS == 0))
+# error Assembly macro kernel_ok fails
+#endif
+ .macro kernel_ok at, sp, success
+ get_fs \at, \sp
+ beqz \at, \success
+ .endm
+
+/*
+ * user_ok determines whether the access to user-space memory is allowed.
+ * See the equivalent C-macro version below for clarity.
+ *
+ * On error, user_ok branches to a label indicated by parameter
+ * <error>. This implies that the macro falls through to the next
+ * instruction on success.
+ *
+ * Note that while this macro can be used independently, we designed
+ * in for optimal use in the access_ok macro below (i.e., we fall
+ * through on success).
+ *
+ * On Entry:
+ * <aa> register containing memory address
+ * <as> register containing memory size
+ * <at> temp register
+ * <error> label to branch to on error; implies fall-through
+ * macro on success
+ * On Exit:
+ * <aa> preserved
+ * <as> preserved
+ * <at> destroyed (actually, (TASK_SIZE + 1 - size))
+ */
+ .macro user_ok aa, as, at, error
+ movi \at, (TASK_SIZE+1)
+ bgeu \as, \at, \error
+ sub \at, \at, \as
+ bgeu \aa, \at, \error
+ .endm
+
+/*
+ * access_ok determines whether a memory access is allowed. See the
+ * equivalent C-macro version below for clarity.
+ *
+ * On error, access_ok branches to a label indicated by parameter
+ * <error>. This implies that the macro falls through to the next
+ * instruction on success.
+ *
+ * Note that we assume success is the common case, and we optimize the
+ * branch fall-through case on success.
+ *
+ * On Entry:
+ * <aa> register containing memory address
+ * <as> register containing memory size
+ * <at> temp register
+ * <sp>
+ * <error> label to branch to on error; implies fall-through
+ * macro on success
+ * On Exit:
+ * <aa> preserved
+ * <as> preserved
+ * <at> destroyed
+ */
+ .macro access_ok aa, as, at, sp, error
+ kernel_ok \at, \sp, .Laccess_ok_\@
+ user_ok \aa, \as, \at, \error
+.Laccess_ok_\@:
+ .endm
+
+/*
+ * verify_area determines whether a memory access is allowed. It's
+ * mostly an unnecessary wrapper for access_ok, but we provide it as a
+ * duplicate of the verify_area() C inline function below. See the
+ * equivalent C version below for clarity.
+ *
+ * On error, verify_area branches to a label indicated by parameter
+ * <error>. This implies that the macro falls through to the next
+ * instruction on success.
+ *
+ * Note that we assume success is the common case, and we optimize the
+ * branch fall-through case on success.
+ *
+ * On Entry:
+ * <aa> register containing memory address
+ * <as> register containing memory size
+ * <at> temp register
+ * <error> label to branch to on error; implies fall-through
+ * macro on success
+ * On Exit:
+ * <aa> preserved
+ * <as> preserved
+ * <at> destroyed
+ */
+ .macro verify_area aa, as, at, sp, error
+ access_ok \at, \aa, \as, \sp, \error
+ .endm
+
+
+#else /* __ASSEMBLY__ not defined */
+
+#include <linux/sched.h>
+#include <asm/types.h>
+
+/*
+ * The fs value determines whether argument validity checking should
+ * be performed or not. If get_fs() == USER_DS, checking is
+ * performed, with get_fs() == KERNEL_DS, checking is bypassed.
+ *
+ * For historical reasons (Data Segment Register?), these macros are
+ * grossly misnamed.
+ */
+
+#define KERNEL_DS ((mm_segment_t) { 0 })
+#define USER_DS ((mm_segment_t) { 1 })
+
+#define get_ds() (KERNEL_DS)
+#define get_fs() (current->thread.current_ds)
+#define set_fs(val) (current->thread.current_ds = (val))
+
+#define segment_eq(a,b) ((a).seg == (b).seg)
+
+#define __kernel_ok (segment_eq(get_fs(), KERNEL_DS))
+#define __user_ok(addr,size) (((size) <= TASK_SIZE)&&((addr) <= TASK_SIZE-(size)))
+#define __access_ok(addr,size) (__kernel_ok || __user_ok((addr),(size)))
+#define access_ok(type,addr,size) __access_ok((unsigned long)(addr),(size))
+
+extern inline int verify_area(int type, const void * addr, unsigned long size)
+{
+ return access_ok(type,addr,size) ? 0 : -EFAULT;
+}
+
+/*
+ * These are the main single-value transfer routines. They
+ * automatically use the right size if we just have the right pointer
+ * type.
+ *
+ * This gets kind of ugly. We want to return _two_ values in
+ * "get_user()" and yet we don't want to do any pointers, because that
+ * is too much of a performance impact. Thus we have a few rather ugly
+ * macros here, and hide all the uglyness from the user.
+ *
+ * Careful to not
+ * (a) re-use the arguments for side effects (sizeof is ok)
+ * (b) require any knowledge of processes at this stage
+ */
+#define put_user(x,ptr) __put_user_check((x),(ptr),sizeof(*(ptr)))
+#define get_user(x,ptr) __get_user_check((x),(ptr),sizeof(*(ptr)))
+
+/*
+ * The "__xxx" versions of the user access functions are versions that
+ * do not verify the address space, that must have been done previously
+ * with a separate "access_ok()" call (this is used when we do multiple
+ * accesses to the same area of user memory).
+ */
+#define __put_user(x,ptr) __put_user_nocheck((x),(ptr),sizeof(*(ptr)))
+#define __get_user(x,ptr) __get_user_nocheck((x),(ptr),sizeof(*(ptr)))
+
+
+extern long __put_user_bad(void);
+
+#define __put_user_nocheck(x,ptr,size) \
+({ \
+ long __pu_err; \
+ __put_user_size((x),(ptr),(size),__pu_err); \
+ __pu_err; \
+})
+
+#define __put_user_check(x,ptr,size) \
+({ \
+ long __pu_err = -EFAULT; \
+ __typeof__(*(ptr)) *__pu_addr = (ptr); \
+ if (access_ok(VERIFY_WRITE,__pu_addr,size)) \
+ __put_user_size((x),__pu_addr,(size),__pu_err); \
+ __pu_err; \
+})
+
+#define __put_user_size(x,ptr,size,retval) \
+do { \
+ retval = 0; \
+ switch (size) { \
+ case 1: __put_user_asm(x,ptr,retval,1,"s8i"); break; \
+ case 2: __put_user_asm(x,ptr,retval,2,"s16i"); break; \
+ case 4: __put_user_asm(x,ptr,retval,4,"s32i"); break; \
+ case 8: { \
+ __typeof__(*ptr) __v64 = x; \
+ retval = __copy_to_user(ptr,&__v64,8); \
+ break; \
+ } \
+ default: __put_user_bad(); \
+ } \
+} while (0)
+
+
+/*
+ * Consider a case of a user single load/store would cause both an
+ * unaligned exception and an MMU-related exception (unaligned
+ * exceptions happen first):
+ *
+ * User code passes a bad variable ptr to a system call.
+ * Kernel tries to access the variable.
+ * Unaligned exception occurs.
+ * Unaligned exception handler tries to make aligned accesses.
+ * Double exception occurs for MMU-related cause (e.g., page not mapped).
+ * do_page_fault() thinks the fault address belongs to the kernel, not the
+ * user, and panics.
+ *
+ * The kernel currently prohibits user unaligned accesses. We use the
+ * __check_align_* macros to check for unaligned addresses before
+ * accessing user space so we don't crash the kernel. Both
+ * __put_user_asm and __get_user_asm use these alignment macros, so
+ * macro-specific labels such as 0f, 1f, %0, %2, and %3 must stay in
+ * sync.
+ */
+
+#define __check_align_1 ""
+
+#define __check_align_2 \
+ " _bbci.l %2, 0, 1f \n" \
+ " movi %0, %3 \n" \
+ " _j 2f \n"
+
+#define __check_align_4 \
+ " _bbsi.l %2, 0, 0f \n" \
+ " _bbci.l %2, 1, 1f \n" \
+ "0: movi %0, %3 \n" \
+ " _j 2f \n"
+
+
+/*
+ * We don't tell gcc that we are accessing memory, but this is OK
+ * because we do not write to any memory gcc knows about, so there
+ * are no aliasing issues.
+ *
+ * WARNING: If you modify this macro at all, verify that the
+ * __check_align_* macros still work.
+ */
+#define __put_user_asm(x, addr, err, align, insn) \
+ __asm__ __volatile__( \
+ __check_align_##align \
+ "1: "insn" %1, %2, 0 \n" \
+ "2: \n" \
+ " .section .fixup,\"ax\" \n" \
+ " .align 4 \n" \
+ "4: \n" \
+ " .long 2b \n" \
+ "5: \n" \
+ " l32r %2, 4b \n" \
+ " movi %0, %3 \n" \
+ " jx %2 \n" \
+ " .previous \n" \
+ " .section __ex_table,\"a\" \n" \
+ " .long 1b, 5b \n" \
+ " .previous" \
+ :"=r" (err) \
+ :"r" ((int)(x)), "r" (addr), "i" (-EFAULT), "0" (err))
+
+#define __get_user_nocheck(x,ptr,size) \
+({ \
+ long __gu_err, __gu_val; \
+ __get_user_size(__gu_val,(ptr),(size),__gu_err); \
+ (x) = (__typeof__(*(ptr)))__gu_val; \
+ __gu_err; \
+})
+
+#define __get_user_check(x,ptr,size) \
+({ \
+ long __gu_err = -EFAULT, __gu_val = 0; \
+ const __typeof__(*(ptr)) *__gu_addr = (ptr); \
+ if (access_ok(VERIFY_READ,__gu_addr,size)) \
+ __get_user_size(__gu_val,__gu_addr,(size),__gu_err); \
+ (x) = (__typeof__(*(ptr)))__gu_val; \
+ __gu_err; \
+})
+
+extern long __get_user_bad(void);
+
+#define __get_user_size(x,ptr,size,retval) \
+do { \
+ retval = 0; \
+ switch (size) { \
+ case 1: __get_user_asm(x,ptr,retval,1,"l8ui"); break; \
+ case 2: __get_user_asm(x,ptr,retval,2,"l16ui"); break; \
+ case 4: __get_user_asm(x,ptr,retval,4,"l32i"); break; \
+ case 8: retval = __copy_from_user(&x,ptr,8); break; \
+ default: (x) = __get_user_bad(); \
+ } \
+} while (0)
+
+
+/*
+ * WARNING: If you modify this macro at all, verify that the
+ * __check_align_* macros still work.
+ */
+#define __get_user_asm(x, addr, err, align, insn) \
+ __asm__ __volatile__( \
+ __check_align_##align \
+ "1: "insn" %1, %2, 0 \n" \
+ "2: \n" \
+ " .section .fixup,\"ax\" \n" \
+ " .align 4 \n" \
+ "4: \n" \
+ " .long 2b \n" \
+ "5: \n" \
+ " l32r %2, 4b \n" \
+ " movi %1, 0 \n" \
+ " movi %0, %3 \n" \
+ " jx %2 \n" \
+ " .previous \n" \
+ " .section __ex_table,\"a\" \n" \
+ " .long 1b, 5b \n" \
+ " .previous" \
+ :"=r" (err), "=r" (x) \
+ :"r" (addr), "i" (-EFAULT), "0" (err))
+
+
+/*
+ * Copy to/from user space
+ */
+
+/*
+ * We use a generic, arbitrary-sized copy subroutine. The Xtensa
+ * architecture would cause heavy code bloat if we tried to inline
+ * these functions and provide __constant_copy_* equivalents like the
+ * i386 versions. __xtensa_copy_user is quite efficient. See the
+ * .fixup section of __xtensa_copy_user for a discussion on the
+ * X_zeroing equivalents for Xtensa.
+ */
+
+extern unsigned __xtensa_copy_user(void *to, const void *from, unsigned n);
+#define __copy_user(to,from,size) __xtensa_copy_user(to,from,size)
+
+
+static inline unsigned long
+__generic_copy_from_user_nocheck(void *to, const void *from, unsigned long n)
+{
+ return __copy_user(to,from,n);
+}
+
+static inline unsigned long
+__generic_copy_to_user_nocheck(void *to, const void *from, unsigned long n)
+{
+ return __copy_user(to,from,n);
+}
+
+static inline unsigned long
+__generic_copy_to_user(void *to, const void *from, unsigned long n)
+{
+ prefetch(from);
+ if (access_ok(VERIFY_WRITE, to, n))
+ return __copy_user(to,from,n);
+ return n;
+}
+
+static inline unsigned long
+__generic_copy_from_user(void *to, const void *from, unsigned long n)
+{
+ prefetchw(to);
+ if (access_ok(VERIFY_READ, from, n))
+ return __copy_user(to,from,n);
+ else
+ memset(to, 0, n);
+ return n;
+}
+
+#define copy_to_user(to,from,n) __generic_copy_to_user((to),(from),(n))
+#define copy_from_user(to,from,n) __generic_copy_from_user((to),(from),(n))
+#define __copy_to_user(to,from,n) __generic_copy_to_user_nocheck((to),(from),(n))
+#define __copy_from_user(to,from,n) __generic_copy_from_user_nocheck((to),(from),(n))
+#define __copy_to_user_inatomic __copy_to_user
+#define __copy_from_user_inatomic __copy_from_user
+
+
+/*
+ * We need to return the number of bytes not cleared. Our memset()
+ * returns zero if a problem occurs while accessing user-space memory.
+ * In that event, return no memory cleared. Otherwise, zero for
+ * success.
+ */
+
+extern inline unsigned long
+__xtensa_clear_user(void *addr, unsigned long size)
+{
+ if ( ! memset(addr, 0, size) )
+ return size;
+ return 0;
+}
+
+extern inline unsigned long
+clear_user(void *addr, unsigned long size)
+{
+ if (access_ok(VERIFY_WRITE, addr, size))
+ return __xtensa_clear_user(addr, size);
+ return size ? -EFAULT : 0;
+}
+
+#define __clear_user __xtensa_clear_user
+
+
+extern long __strncpy_user(char *, const char *, long);
+#define __strncpy_from_user __strncpy_user
+
+extern inline long
+strncpy_from_user(char *dst, const char *src, long count)
+{
+ if (access_ok(VERIFY_READ, src, 1))
+ return __strncpy_from_user(dst, src, count);
+ return -EFAULT;
+}
+
+
+#define strlen_user(str) strnlen_user((str), TASK_SIZE - 1)
+
+/*
+ * Return the size of a string (including the ending 0!)
+ */
+extern long __strnlen_user(const char *, long);
+
+extern inline long strnlen_user(const char *str, long len)
+{
+ unsigned long top = __kernel_ok ? ~0UL : TASK_SIZE - 1;
+
+ if ((unsigned long)str > top)
+ return 0;
+ return __strnlen_user(str, len);
+}
+
+
+struct exception_table_entry
+{
+ unsigned long insn, fixup;
+};
+
+/* Returns 0 if exception not found and fixup.unit otherwise. */
+
+extern unsigned long search_exception_table(unsigned long addr);
+extern void sort_exception_table(void);
+
+/* Returns the new pc */
+#define fixup_exception(map_reg, fixup_unit, pc) \
+({ \
+ fixup_unit; \
+})
+
+#endif /* __ASSEMBLY__ */
+#endif /* _XTENSA_UACCESS_H */