!C99Shell v.2.1 [PHP 7 Update] [1.12.2019]!

Software: Apache/2.2.8 (Ubuntu) PHP/5.2.4-2ubuntu5.12 with Suhosin-Patch mod_ssl/2.2.8 OpenSSL/0.9.8g. PHP/5.2.4-2ubuntu5.12 

uname -a: Linux forum.circlefusion.com 2.6.24-19-server #1 SMP Wed Jun 18 15:18:00 UTC 2008 i686 

uid=33(www-data) gid=33(www-data) groups=33(www-data) 

Safe-mode: OFF (not secure)

/usr/src/linux-headers-2.6.24-28/include/asm-alpha/   drwxr-xr-x
Free 1.97 GB of 97.11 GB (2.03%)
Home    Back    Forward    UPDIR    Refresh    Search    Buffer    Encoder    Tools    Proc.    FTP brute    Sec.    SQL    PHP-code    Update    Feedback    Self remove    Logout    


Viewing file:     uaccess.h (14.27 KB)      -rw-r--r--
Select action/file-type:
(+) | (+) | (+) | Code (+) | Session (+) | (+) | SDB (+) | (+) | (+) | (+) | (+) | (+) |
#ifndef __ALPHA_UACCESS_H
#define __ALPHA_UACCESS_H

#include <linux/errno.h>
#include <linux/sched.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.
 *
 * Or at least it did once upon a time.  Nowadays it is a mask that
 * defines which bits of the address space are off limits.  This is a
 * wee bit faster than the above.
 *
 * For historical reasons, these macros are grossly misnamed.
 */

#define KERNEL_DS    ((mm_segment_t) { 0UL })
#define USER_DS        ((mm_segment_t) { -0x40000000000UL })

#define VERIFY_READ    0
#define VERIFY_WRITE    1

#define get_fs()  (current_thread_info()->addr_limit)
#define get_ds()  (KERNEL_DS)
#define set_fs(x) (current_thread_info()->addr_limit = (x))

#define segment_eq(a,b)    ((a).seg == (b).seg)

/*
 * Is a address valid? This does a straightforward calculation rather
 * than tests.
 *
 * Address valid if:
 *  - "addr" doesn't have any high-bits set
 *  - AND "size" doesn't have any high-bits set
 *  - AND "addr+size" doesn't have any high-bits set
 *  - OR we are in kernel mode.
 */
#define __access_ok(addr,size,segment) \
    (((segment).seg & (addr | size | (addr+size))) == 0)

#define access_ok(type,addr,size)                \
({                                \
    __chk_user_ptr(addr);                    \
    __access_ok(((unsigned long)(addr)),(size),get_fs());    \
})

/*
 * These are the main single-value transfer routines.  They automatically
 * use the right size if we just have the right pointer type.
 *
 * As the alpha uses the same address space for kernel and user
 * data, we can just do these as direct assignments.  (Of course, the
 * exception handling means that it's no longer "just"...)
 *
 * Careful to not
 * (a) re-use the arguments for side effects (sizeof/typeof is ok)
 * (b) require any knowledge of processes at this stage
 */
#define put_user(x,ptr) \
  __put_user_check((__typeof__(*(ptr)))(x),(ptr),sizeof(*(ptr)),get_fs())
#define get_user(x,ptr) \
  __get_user_check((x),(ptr),sizeof(*(ptr)),get_fs())

/*
 * The "__xxx" versions do not do address space checking, useful when
 * doing multiple accesses to the same area (the programmer has to do the
 * checks by hand with "access_ok()")
 */
#define __put_user(x,ptr) \
  __put_user_nocheck((__typeof__(*(ptr)))(x),(ptr),sizeof(*(ptr)))
#define __get_user(x,ptr) \
  __get_user_nocheck((x),(ptr),sizeof(*(ptr)))
  
/*
 * The "lda %1, 2b-1b(%0)" bits are magic to get the assembler to
 * encode the bits we need for resolving the exception.  See the
 * more extensive comments with fixup_inline_exception below for
 * more information.
 */

extern void __get_user_unknown(void);

#define __get_user_nocheck(x,ptr,size)                \
({                                \
    long __gu_err = 0;                    \
    unsigned long __gu_val;                    \
    __chk_user_ptr(ptr);                    \
    switch (size) {                        \
      case 1: __get_user_8(ptr); break;            \
      case 2: __get_user_16(ptr); break;            \
      case 4: __get_user_32(ptr); break;            \
      case 8: __get_user_64(ptr); break;            \
      default: __get_user_unknown(); break;            \
    }                            \
    (x) = (__typeof__(*(ptr))) __gu_val;            \
    __gu_err;                        \
})

#define __get_user_check(x,ptr,size,segment)                \
({                                    \
    long __gu_err = -EFAULT;                    \
    unsigned long __gu_val = 0;                    \
    const __typeof__(*(ptr)) __user *__gu_addr = (ptr);        \
    if (__access_ok((unsigned long)__gu_addr,size,segment)) {    \
        __gu_err = 0;                        \
        switch (size) {                        \
          case 1: __get_user_8(__gu_addr); break;        \
          case 2: __get_user_16(__gu_addr); break;        \
          case 4: __get_user_32(__gu_addr); break;        \
          case 8: __get_user_64(__gu_addr); break;        \
          default: __get_user_unknown(); break;            \
        }                            \
    }                                \
    (x) = (__typeof__(*(ptr))) __gu_val;                \
    __gu_err;                            \
})

struct __large_struct { unsigned long buf[100]; };
#define __m(x) (*(struct __large_struct __user *)(x))

#define __get_user_64(addr)                \
    __asm__("1: ldq %0,%2\n"            \
    "2:\n"                        \
    ".section __ex_table,\"a\"\n"            \
    "    .long 1b - .\n"                \
    "    lda %0, 2b-1b(%1)\n"            \
    ".previous"                    \
        : "=r"(__gu_val), "=r"(__gu_err)    \
        : "m"(__m(addr)), "1"(__gu_err))

#define __get_user_32(addr)                \
    __asm__("1: ldl %0,%2\n"            \
    "2:\n"                        \
    ".section __ex_table,\"a\"\n"            \
    "    .long 1b - .\n"                \
    "    lda %0, 2b-1b(%1)\n"            \
    ".previous"                    \
        : "=r"(__gu_val), "=r"(__gu_err)    \
        : "m"(__m(addr)), "1"(__gu_err))

#ifdef __alpha_bwx__
/* Those lucky bastards with ev56 and later CPUs can do byte/word moves.  */

#define __get_user_16(addr)                \
    __asm__("1: ldwu %0,%2\n"            \
    "2:\n"                        \
    ".section __ex_table,\"a\"\n"            \
    "    .long 1b - .\n"                \
    "    lda %0, 2b-1b(%1)\n"            \
    ".previous"                    \
        : "=r"(__gu_val), "=r"(__gu_err)    \
        : "m"(__m(addr)), "1"(__gu_err))

#define __get_user_8(addr)                \
    __asm__("1: ldbu %0,%2\n"            \
    "2:\n"                        \
    ".section __ex_table,\"a\"\n"            \
    "    .long 1b - .\n"                \
    "    lda %0, 2b-1b(%1)\n"            \
    ".previous"                    \
        : "=r"(__gu_val), "=r"(__gu_err)    \
        : "m"(__m(addr)), "1"(__gu_err))
#else
/* Unfortunately, we can't get an unaligned access trap for the sub-word
   load, so we have to do a general unaligned operation.  */

#define __get_user_16(addr)                        \
{                                    \
    long __gu_tmp;                            \
    __asm__("1: ldq_u %0,0(%3)\n"                    \
    "2:    ldq_u %1,1(%3)\n"                    \
    "    extwl %0,%3,%0\n"                    \
    "    extwh %1,%3,%1\n"                    \
    "    or %0,%1,%0\n"                        \
    "3:\n"                                \
    ".section __ex_table,\"a\"\n"                    \
    "    .long 1b - .\n"                        \
    "    lda %0, 3b-1b(%2)\n"                    \
    "    .long 2b - .\n"                        \
    "    lda %0, 3b-2b(%2)\n"                    \
    ".previous"                            \
        : "=&r"(__gu_val), "=&r"(__gu_tmp), "=r"(__gu_err)    \
        : "r"(addr), "2"(__gu_err));                \
}

#define __get_user_8(addr)                        \
    __asm__("1: ldq_u %0,0(%2)\n"                    \
    "    extbl %0,%2,%0\n"                    \
    "2:\n"                                \
    ".section __ex_table,\"a\"\n"                    \
    "    .long 1b - .\n"                        \
    "    lda %0, 2b-1b(%1)\n"                    \
    ".previous"                            \
        : "=&r"(__gu_val), "=r"(__gu_err)            \
        : "r"(addr), "1"(__gu_err))
#endif

extern void __put_user_unknown(void);

#define __put_user_nocheck(x,ptr,size)                \
({                                \
    long __pu_err = 0;                    \
    __chk_user_ptr(ptr);                    \
    switch (size) {                        \
      case 1: __put_user_8(x,ptr); break;            \
      case 2: __put_user_16(x,ptr); break;            \
      case 4: __put_user_32(x,ptr); break;            \
      case 8: __put_user_64(x,ptr); break;            \
      default: __put_user_unknown(); break;            \
    }                            \
    __pu_err;                        \
})

#define __put_user_check(x,ptr,size,segment)                \
({                                    \
    long __pu_err = -EFAULT;                    \
    __typeof__(*(ptr)) __user *__pu_addr = (ptr);            \
    if (__access_ok((unsigned long)__pu_addr,size,segment)) {    \
        __pu_err = 0;                        \
        switch (size) {                        \
          case 1: __put_user_8(x,__pu_addr); break;        \
          case 2: __put_user_16(x,__pu_addr); break;        \
          case 4: __put_user_32(x,__pu_addr); break;        \
          case 8: __put_user_64(x,__pu_addr); break;        \
          default: __put_user_unknown(); break;            \
        }                            \
    }                                \
    __pu_err;                            \
})

/*
 * The "__put_user_xx()" macros tell gcc they read from memory
 * instead of writing: this is because they do not write to
 * any memory gcc knows about, so there are no aliasing issues
 */
#define __put_user_64(x,addr)                    \
__asm__ __volatile__("1: stq %r2,%1\n"                \
    "2:\n"                            \
    ".section __ex_table,\"a\"\n"                \
    "    .long 1b - .\n"                    \
    "    lda $31,2b-1b(%0)\n"                \
    ".previous"                        \
        : "=r"(__pu_err)                \
        : "m" (__m(addr)), "rJ" (x), "0"(__pu_err))

#define __put_user_32(x,addr)                    \
__asm__ __volatile__("1: stl %r2,%1\n"                \
    "2:\n"                            \
    ".section __ex_table,\"a\"\n"                \
    "    .long 1b - .\n"                    \
    "    lda $31,2b-1b(%0)\n"                \
    ".previous"                        \
        : "=r"(__pu_err)                \
        : "m"(__m(addr)), "rJ"(x), "0"(__pu_err))

#ifdef __alpha_bwx__
/* Those lucky bastards with ev56 and later CPUs can do byte/word moves.  */

#define __put_user_16(x,addr)                    \
__asm__ __volatile__("1: stw %r2,%1\n"                \
    "2:\n"                            \
    ".section __ex_table,\"a\"\n"                \
    "    .long 1b - .\n"                    \
    "    lda $31,2b-1b(%0)\n"                \
    ".previous"                        \
        : "=r"(__pu_err)                \
        : "m"(__m(addr)), "rJ"(x), "0"(__pu_err))

#define __put_user_8(x,addr)                    \
__asm__ __volatile__("1: stb %r2,%1\n"                \
    "2:\n"                            \
    ".section __ex_table,\"a\"\n"                \
    "    .long 1b - .\n"                    \
    "    lda $31,2b-1b(%0)\n"                \
    ".previous"                        \
        : "=r"(__pu_err)                \
        : "m"(__m(addr)), "rJ"(x), "0"(__pu_err))
#else
/* Unfortunately, we can't get an unaligned access trap for the sub-word
   write, so we have to do a general unaligned operation.  */

#define __put_user_16(x,addr)                    \
{                                \
    long __pu_tmp1, __pu_tmp2, __pu_tmp3, __pu_tmp4;    \
    __asm__ __volatile__(                    \
    "1:    ldq_u %2,1(%5)\n"                \
    "2:    ldq_u %1,0(%5)\n"                \
    "    inswh %6,%5,%4\n"                \
    "    inswl %6,%5,%3\n"                \
    "    mskwh %2,%5,%2\n"                \
    "    mskwl %1,%5,%1\n"                \
    "    or %2,%4,%2\n"                    \
    "    or %1,%3,%1\n"                    \
    "3:    stq_u %2,1(%5)\n"                \
    "4:    stq_u %1,0(%5)\n"                \
    "5:\n"                            \
    ".section __ex_table,\"a\"\n"                \
    "    .long 1b - .\n"                    \
    "    lda $31, 5b-1b(%0)\n"                \
    "    .long 2b - .\n"                    \
    "    lda $31, 5b-2b(%0)\n"                \
    "    .long 3b - .\n"                    \
    "    lda $31, 5b-3b(%0)\n"                \
    "    .long 4b - .\n"                    \
    "    lda $31, 5b-4b(%0)\n"                \
    ".previous"                        \
        : "=r"(__pu_err), "=&r"(__pu_tmp1),        \
          "=&r"(__pu_tmp2), "=&r"(__pu_tmp3),        \
          "=&r"(__pu_tmp4)                \
        : "r"(addr), "r"((unsigned long)(x)), "0"(__pu_err)); \
}

#define __put_user_8(x,addr)                    \
{                                \
    long __pu_tmp1, __pu_tmp2;                \
    __asm__ __volatile__(                    \
    "1:    ldq_u %1,0(%4)\n"                \
    "    insbl %3,%4,%2\n"                \
    "    mskbl %1,%4,%1\n"                \
    "    or %1,%2,%1\n"                    \
    "2:    stq_u %1,0(%4)\n"                \
    "3:\n"                            \
    ".section __ex_table,\"a\"\n"                \
    "    .long 1b - .\n"                    \
    "    lda $31, 3b-1b(%0)\n"                \
    "    .long 2b - .\n"                    \
    "    lda $31, 3b-2b(%0)\n"                \
    ".previous"                        \
        : "=r"(__pu_err),                \
            "=&r"(__pu_tmp1), "=&r"(__pu_tmp2)        \
        : "r"((unsigned long)(x)), "r"(addr), "0"(__pu_err)); \
}
#endif


/*
 * Complex access routines
 */

/* This little bit of silliness is to get the GP loaded for a function
   that ordinarily wouldn't.  Otherwise we could have it done by the macro
   directly, which can be optimized the linker.  */
#ifdef MODULE
#define __module_address(sym)        "r"(sym),
#define __module_call(ra, arg, sym)    "jsr $" #ra ",(%" #arg ")," #sym
#else
#define __module_address(sym)
#define __module_call(ra, arg, sym)    "bsr $" #ra "," #sym " !samegp"
#endif

extern void __copy_user(void);

extern inline long
__copy_tofrom_user_nocheck(void *to, const void *from, long len)
{
    register void * __cu_to __asm__("$6") = to;
    register const void * __cu_from __asm__("$7") = from;
    register long __cu_len __asm__("$0") = len;

    __asm__ __volatile__(
        __module_call(28, 3, __copy_user)
        : "=r" (__cu_len), "=r" (__cu_from), "=r" (__cu_to)
        : __module_address(__copy_user)
          "0" (__cu_len), "1" (__cu_from), "2" (__cu_to)
        : "$1","$2","$3","$4","$5","$28","memory");

    return __cu_len;
}

extern inline long
__copy_tofrom_user(void *to, const void *from, long len, const void __user *validate)
{
    if (__access_ok((unsigned long)validate, len, get_fs()))
        len = __copy_tofrom_user_nocheck(to, from, len);
    return len;
}

#define __copy_to_user(to,from,n)                    \
({                                    \
    __chk_user_ptr(to);                        \
    __copy_tofrom_user_nocheck((__force void *)(to),(from),(n));    \
})
#define __copy_from_user(to,from,n)                    \
({                                    \
    __chk_user_ptr(from);                        \
    __copy_tofrom_user_nocheck((to),(__force void *)(from),(n));    \
})

#define __copy_to_user_inatomic __copy_to_user
#define __copy_from_user_inatomic __copy_from_user


extern inline long
copy_to_user(void __user *to, const void *from, long n)
{
    return __copy_tofrom_user((__force void *)to, from, n, to);
}

extern inline long
copy_from_user(void *to, const void __user *from, long n)
{
    return __copy_tofrom_user(to, (__force void *)from, n, from);
}

extern void __do_clear_user(void);

extern inline long
__clear_user(void __user *to, long len)
{
    register void __user * __cl_to __asm__("$6") = to;
    register long __cl_len __asm__("$0") = len;
    __asm__ __volatile__(
        __module_call(28, 2, __do_clear_user)
        : "=r"(__cl_len), "=r"(__cl_to)
        : __module_address(__do_clear_user)
          "0"(__cl_len), "1"(__cl_to)
        : "$1","$2","$3","$4","$5","$28","memory");
    return __cl_len;
}

extern inline long
clear_user(void __user *to, long len)
{
    if (__access_ok((unsigned long)to, len, get_fs()))
        len = __clear_user(to, len);
    return len;
}

#undef __module_address
#undef __module_call

/* Returns: -EFAULT if exception before terminator, N if the entire
   buffer filled, else strlen.  */

extern long __strncpy_from_user(char *__to, const char __user *__from, long __to_len);

extern inline long
strncpy_from_user(char *to, const char __user *from, long n)
{
    long ret = -EFAULT;
    if (__access_ok((unsigned long)from, 0, get_fs()))
        ret = __strncpy_from_user(to, from, n);
    return ret;
}

/* Returns: 0 if bad, string length+1 (memory size) of string if ok */
extern long __strlen_user(const char __user *);

extern inline long strlen_user(const char __user *str)
{
    return access_ok(VERIFY_READ,str,0) ? __strlen_user(str) : 0;
}

/* Returns: 0 if exception before NUL or reaching the supplied limit (N),
 * a value greater than N if the limit would be exceeded, else strlen.  */
extern long __strnlen_user(const char __user *, long);

extern inline long strnlen_user(const char __user *str, long n)
{
    return access_ok(VERIFY_READ,str,0) ? __strnlen_user(str, n) : 0;
}

/*
 * About the exception table:
 *
 * - insn is a 32-bit pc-relative offset from the faulting insn.
 * - nextinsn is a 16-bit offset off of the faulting instruction
 *   (not off of the *next* instruction as branches are).
 * - errreg is the register in which to place -EFAULT.
 * - valreg is the final target register for the load sequence
 *   and will be zeroed.
 *
 * Either errreg or valreg may be $31, in which case nothing happens.
 *
 * The exception fixup information "just so happens" to be arranged
 * as in a MEM format instruction.  This lets us emit our three
 * values like so:
 *
 *      lda valreg, nextinsn(errreg)
 *
 */

struct exception_table_entry
{
    signed int insn;
    union exception_fixup {
        unsigned unit;
        struct {
            signed int nextinsn : 16;
            unsigned int errreg : 5;
            unsigned int valreg : 5;
        } bits;
    } fixup;
};

/* Returns the new pc */
#define fixup_exception(map_reg, fixup, pc)            \
({                                \
    if ((fixup)->fixup.bits.valreg != 31)            \
        map_reg((fixup)->fixup.bits.valreg) = 0;    \
    if ((fixup)->fixup.bits.errreg != 31)            \
        map_reg((fixup)->fixup.bits.errreg) = -EFAULT;    \
    (pc) + (fixup)->fixup.bits.nextinsn;            \
})


#endif /* __ALPHA_UACCESS_H */

:: Command execute ::

Enter:
 
Select:
 

:: Search ::
  - regexp 

:: Upload ::
 
[ Read-Only ]

:: Make Dir ::
 
[ Read-Only ]
:: Make File ::
 
[ Read-Only ]

:: Go Dir ::
 
:: Go File ::
 

--[ c99shell v.2.1 [PHP 7 Update] [1.12.2019] maintained by KaizenLouie and updated by cermmik | C99Shell Github (MySQL update) | Generation time: 0.0147 ]--