!C99Shell v. 2.0 [PHP 7 Update] [25.02.2019]!

Software: nginx/1.23.4. PHP/5.6.40-65+ubuntu20.04.1+deb.sury.org+1 

uname -a: Linux foro-restaurado-2 5.15.0-1040-oracle #46-Ubuntu SMP Fri Jul 14 21:47:21 UTC 2023
aarch64
 

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

Safe-mode: OFF (not secure)

/usr/src/linux-oracle-6.8-headers-6.8.0-1027/include/linux/   drwxr-xr-x
Free 83.29 GB of 96.73 GB (86.11%)
Home    Back    Forward    UPDIR    Refresh    Search    Buffer    Encoder    Tools    Proc.    FTP brute    Sec.    SQL    PHP-code    Update    Feedback    Self remove    Logout    


Viewing file:     percpu-defs.h (17.22 KB)      -rw-r--r--
Select action/file-type:
(+) | (+) | (+) | Code (+) | Session (+) | (+) | SDB (+) | (+) | (+) | (+) | (+) | (+) |
/* SPDX-License-Identifier: GPL-2.0-only */
/*
 * linux/percpu-defs.h - basic definitions for percpu areas
 *
 * DO NOT INCLUDE DIRECTLY OUTSIDE PERCPU IMPLEMENTATION PROPER.
 *
 * This file is separate from linux/percpu.h to avoid cyclic inclusion
 * dependency from arch header files.  Only to be included from
 * asm/percpu.h.
 *
 * This file includes macros necessary to declare percpu sections and
 * variables, and definitions of percpu accessors and operations.  It
 * should provide enough percpu features to arch header files even when
 * they can only include asm/percpu.h to avoid cyclic inclusion dependency.
 */

#ifndef _LINUX_PERCPU_DEFS_H
#define _LINUX_PERCPU_DEFS_H

#ifdef CONFIG_SMP

#ifdef MODULE
#define PER_CPU_SHARED_ALIGNED_SECTION ""
#define PER_CPU_ALIGNED_SECTION ""
#else
#define PER_CPU_SHARED_ALIGNED_SECTION "..shared_aligned"
#define PER_CPU_ALIGNED_SECTION "..shared_aligned"
#endif
#define PER_CPU_FIRST_SECTION "..first"

#else

#define PER_CPU_SHARED_ALIGNED_SECTION ""
#define PER_CPU_ALIGNED_SECTION "..shared_aligned"
#define PER_CPU_FIRST_SECTION ""

#endif

/*
 * Base implementations of per-CPU variable declarations and definitions, where
 * the section in which the variable is to be placed is provided by the
 * 'sec' argument.  This may be used to affect the parameters governing the
 * variable's storage.
 *
 * NOTE!  The sections for the DECLARE and for the DEFINE must match, lest
 * linkage errors occur due the compiler generating the wrong code to access
 * that section.
 */
#define __PCPU_ATTRS(sec)                        \
    __percpu __attribute__((section(PER_CPU_BASE_SECTION sec)))    \
    PER_CPU_ATTRIBUTES

#define __PCPU_DUMMY_ATTRS                        \
    __section(".discard") __attribute__((unused))

/*
 * s390 and alpha modules require percpu variables to be defined as
 * weak to force the compiler to generate GOT based external
 * references for them.  This is necessary because percpu sections
 * will be located outside of the usually addressable area.
 *
 * This definition puts the following two extra restrictions when
 * defining percpu variables.
 *
 * 1. The symbol must be globally unique, even the static ones.
 * 2. Static percpu variables cannot be defined inside a function.
 *
 * Archs which need weak percpu definitions should define
 * ARCH_NEEDS_WEAK_PER_CPU in asm/percpu.h when necessary.
 *
 * To ensure that the generic code observes the above two
 * restrictions, if CONFIG_DEBUG_FORCE_WEAK_PER_CPU is set weak
 * definition is used for all cases.
 */
#if defined(ARCH_NEEDS_WEAK_PER_CPU) || defined(CONFIG_DEBUG_FORCE_WEAK_PER_CPU)
/*
 * __pcpu_scope_* dummy variable is used to enforce scope.  It
 * receives the static modifier when it's used in front of
 * DEFINE_PER_CPU() and will trigger build failure if
 * DECLARE_PER_CPU() is used for the same variable.
 *
 * __pcpu_unique_* dummy variable is used to enforce symbol uniqueness
 * such that hidden weak symbol collision, which will cause unrelated
 * variables to share the same address, can be detected during build.
 */
#define DECLARE_PER_CPU_SECTION(type, name, sec)            \
    extern __PCPU_DUMMY_ATTRS char __pcpu_scope_##name;        \
    extern __PCPU_ATTRS(sec) __typeof__(type) name

#define DEFINE_PER_CPU_SECTION(type, name, sec)                \
    __PCPU_DUMMY_ATTRS char __pcpu_scope_##name;            \
    extern __PCPU_DUMMY_ATTRS char __pcpu_unique_##name;        \
    __PCPU_DUMMY_ATTRS char __pcpu_unique_##name;            \
    extern __PCPU_ATTRS(sec) __typeof__(type) name;            \
    __PCPU_ATTRS(sec) __weak __typeof__(type) name
#else
/*
 * Normal declaration and definition macros.
 */
#define DECLARE_PER_CPU_SECTION(type, name, sec)            \
    extern __PCPU_ATTRS(sec) __typeof__(type) name

#define DEFINE_PER_CPU_SECTION(type, name, sec)                \
    __PCPU_ATTRS(sec) __typeof__(type) name
#endif

/*
 * Variant on the per-CPU variable declaration/definition theme used for
 * ordinary per-CPU variables.
 */
#define DECLARE_PER_CPU(type, name)                    \
    DECLARE_PER_CPU_SECTION(type, name, "")

#define DEFINE_PER_CPU(type, name)                    \
    DEFINE_PER_CPU_SECTION(type, name, "")

/*
 * Declaration/definition used for per-CPU variables that must come first in
 * the set of variables.
 */
#define DECLARE_PER_CPU_FIRST(type, name)                \
    DECLARE_PER_CPU_SECTION(type, name, PER_CPU_FIRST_SECTION)

#define DEFINE_PER_CPU_FIRST(type, name)                \
    DEFINE_PER_CPU_SECTION(type, name, PER_CPU_FIRST_SECTION)

/*
 * Declaration/definition used for per-CPU variables that must be cacheline
 * aligned under SMP conditions so that, whilst a particular instance of the
 * data corresponds to a particular CPU, inefficiencies due to direct access by
 * other CPUs are reduced by preventing the data from unnecessarily spanning
 * cachelines.
 *
 * An example of this would be statistical data, where each CPU's set of data
 * is updated by that CPU alone, but the data from across all CPUs is collated
 * by a CPU processing a read from a proc file.
 */
#define DECLARE_PER_CPU_SHARED_ALIGNED(type, name)            \
    DECLARE_PER_CPU_SECTION(type, name, PER_CPU_SHARED_ALIGNED_SECTION) \
    ____cacheline_aligned_in_smp

#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)            \
    DEFINE_PER_CPU_SECTION(type, name, PER_CPU_SHARED_ALIGNED_SECTION) \
    ____cacheline_aligned_in_smp

#define DECLARE_PER_CPU_ALIGNED(type, name)                \
    DECLARE_PER_CPU_SECTION(type, name, PER_CPU_ALIGNED_SECTION)    \
    ____cacheline_aligned

#define DEFINE_PER_CPU_ALIGNED(type, name)                \
    DEFINE_PER_CPU_SECTION(type, name, PER_CPU_ALIGNED_SECTION)    \
    ____cacheline_aligned

/*
 * Declaration/definition used for per-CPU variables that must be page aligned.
 */
#define DECLARE_PER_CPU_PAGE_ALIGNED(type, name)            \
    DECLARE_PER_CPU_SECTION(type, name, "..page_aligned")        \
    __aligned(PAGE_SIZE)

#define DEFINE_PER_CPU_PAGE_ALIGNED(type, name)                \
    DEFINE_PER_CPU_SECTION(type, name, "..page_aligned")        \
    __aligned(PAGE_SIZE)

/*
 * Declaration/definition used for per-CPU variables that must be read mostly.
 */
#define DECLARE_PER_CPU_READ_MOSTLY(type, name)            \
    DECLARE_PER_CPU_SECTION(type, name, "..read_mostly")

#define DEFINE_PER_CPU_READ_MOSTLY(type, name)                \
    DEFINE_PER_CPU_SECTION(type, name, "..read_mostly")

/*
 * Declaration/definition used for per-CPU variables that should be accessed
 * as decrypted when memory encryption is enabled in the guest.
 */
#ifdef CONFIG_AMD_MEM_ENCRYPT
#define DECLARE_PER_CPU_DECRYPTED(type, name)                \
    DECLARE_PER_CPU_SECTION(type, name, "..decrypted")

#define DEFINE_PER_CPU_DECRYPTED(type, name)                \
    DEFINE_PER_CPU_SECTION(type, name, "..decrypted")
#else
#define DEFINE_PER_CPU_DECRYPTED(type, name)    DEFINE_PER_CPU(type, name)
#endif

/*
 * Intermodule exports for per-CPU variables.  sparse forgets about
 * address space across EXPORT_SYMBOL(), change EXPORT_SYMBOL() to
 * noop if __CHECKER__.
 */
#ifndef __CHECKER__
#define EXPORT_PER_CPU_SYMBOL(var) EXPORT_SYMBOL(var)
#define EXPORT_PER_CPU_SYMBOL_GPL(var) EXPORT_SYMBOL_GPL(var)
#else
#define EXPORT_PER_CPU_SYMBOL(var)
#define EXPORT_PER_CPU_SYMBOL_GPL(var)
#endif

/*
 * Accessors and operations.
 */
#ifndef __ASSEMBLY__

/*
 * __verify_pcpu_ptr() verifies @ptr is a percpu pointer without evaluating
 * @ptr and is invoked once before a percpu area is accessed by all
 * accessors and operations.  This is performed in the generic part of
 * percpu and arch overrides don't need to worry about it; however, if an
 * arch wants to implement an arch-specific percpu accessor or operation,
 * it may use __verify_pcpu_ptr() to verify the parameters.
 *
 * + 0 is required in order to convert the pointer type from a
 * potential array type to a pointer to a single item of the array.
 */
#define __verify_pcpu_ptr(ptr)                        \
do {                                    \
    const void __percpu *__vpp_verify = (typeof((ptr) + 0))NULL;    \
    (void)__vpp_verify;                        \
} while (0)

#ifdef CONFIG_SMP

/*
 * Add an offset to a pointer but keep the pointer as-is.  Use RELOC_HIDE()
 * to prevent the compiler from making incorrect assumptions about the
 * pointer value.  The weird cast keeps both GCC and sparse happy.
 */
#define SHIFT_PERCPU_PTR(__p, __offset)                    \
    RELOC_HIDE((typeof(*(__p)) __kernel __force *)(__p), (__offset))

#define per_cpu_ptr(ptr, cpu)                        \
({                                    \
    __verify_pcpu_ptr(ptr);                        \
    SHIFT_PERCPU_PTR((ptr), per_cpu_offset((cpu)));            \
})

#define raw_cpu_ptr(ptr)                        \
({                                    \
    __verify_pcpu_ptr(ptr);                        \
    arch_raw_cpu_ptr(ptr);                        \
})

#ifdef CONFIG_DEBUG_PREEMPT
#define this_cpu_ptr(ptr)                        \
({                                    \
    __verify_pcpu_ptr(ptr);                        \
    SHIFT_PERCPU_PTR(ptr, my_cpu_offset);                \
})
#else
#define this_cpu_ptr(ptr) raw_cpu_ptr(ptr)
#endif

#else    /* CONFIG_SMP */

#define VERIFY_PERCPU_PTR(__p)                        \
({                                    \
    __verify_pcpu_ptr(__p);                        \
    (typeof(*(__p)) __kernel __force *)(__p);            \
})

#define per_cpu_ptr(ptr, cpu)    ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
#define raw_cpu_ptr(ptr)    per_cpu_ptr(ptr, 0)
#define this_cpu_ptr(ptr)    raw_cpu_ptr(ptr)

#endif    /* CONFIG_SMP */

#define per_cpu(var, cpu)    (*per_cpu_ptr(&(var), cpu))

/*
 * Must be an lvalue. Since @var must be a simple identifier,
 * we force a syntax error here if it isn't.
 */
#define get_cpu_var(var)                        \
(*({                                    \
    preempt_disable();                        \
    this_cpu_ptr(&var);                        \
}))

/*
 * The weird & is necessary because sparse considers (void)(var) to be
 * a direct dereference of percpu variable (var).
 */
#define put_cpu_var(var)                        \
do {                                    \
    (void)&(var);                            \
    preempt_enable();                        \
} while (0)

#define get_cpu_ptr(var)                        \
({                                    \
    preempt_disable();                        \
    this_cpu_ptr(var);                        \
})

#define put_cpu_ptr(var)                        \
do {                                    \
    (void)(var);                            \
    preempt_enable();                        \
} while (0)

/*
 * Branching function to split up a function into a set of functions that
 * are called for different scalar sizes of the objects handled.
 */

extern void __bad_size_call_parameter(void);

#ifdef CONFIG_DEBUG_PREEMPT
extern void __this_cpu_preempt_check(const char *op);
#else
static __always_inline void __this_cpu_preempt_check(const char *op) { }
#endif

#define __pcpu_size_call_return(stem, variable)                \
({                                    \
    typeof(variable) pscr_ret__;                    \
    __verify_pcpu_ptr(&(variable));                    \
    switch(sizeof(variable)) {                    \
    case 1: pscr_ret__ = stem##1(variable); break;            \
    case 2: pscr_ret__ = stem##2(variable); break;            \
    case 4: pscr_ret__ = stem##4(variable); break;            \
    case 8: pscr_ret__ = stem##8(variable); break;            \
    default:                            \
        __bad_size_call_parameter(); break;            \
    }                                \
    pscr_ret__;                            \
})

#define __pcpu_size_call_return2(stem, variable, ...)            \
({                                    \
    typeof(variable) pscr2_ret__;                    \
    __verify_pcpu_ptr(&(variable));                    \
    switch(sizeof(variable)) {                    \
    case 1: pscr2_ret__ = stem##1(variable, __VA_ARGS__); break;    \
    case 2: pscr2_ret__ = stem##2(variable, __VA_ARGS__); break;    \
    case 4: pscr2_ret__ = stem##4(variable, __VA_ARGS__); break;    \
    case 8: pscr2_ret__ = stem##8(variable, __VA_ARGS__); break;    \
    default:                            \
        __bad_size_call_parameter(); break;            \
    }                                \
    pscr2_ret__;                            \
})

#define __pcpu_size_call_return2bool(stem, variable, ...)        \
({                                    \
    bool pscr2_ret__;                        \
    __verify_pcpu_ptr(&(variable));                    \
    switch(sizeof(variable)) {                    \
    case 1: pscr2_ret__ = stem##1(variable, __VA_ARGS__); break;    \
    case 2: pscr2_ret__ = stem##2(variable, __VA_ARGS__); break;    \
    case 4: pscr2_ret__ = stem##4(variable, __VA_ARGS__); break;    \
    case 8: pscr2_ret__ = stem##8(variable, __VA_ARGS__); break;    \
    default:                            \
        __bad_size_call_parameter(); break;            \
    }                                \
    pscr2_ret__;                            \
})

#define __pcpu_size_call(stem, variable, ...)                \
do {                                    \
    __verify_pcpu_ptr(&(variable));                    \
    switch(sizeof(variable)) {                    \
        case 1: stem##1(variable, __VA_ARGS__);break;        \
        case 2: stem##2(variable, __VA_ARGS__);break;        \
        case 4: stem##4(variable, __VA_ARGS__);break;        \
        case 8: stem##8(variable, __VA_ARGS__);break;        \
        default:                         \
            __bad_size_call_parameter();break;        \
    }                                \
} while (0)

/*
 * this_cpu operations (C) 2008-2013 Christoph Lameter <[email protected]>
 *
 * Optimized manipulation for memory allocated through the per cpu
 * allocator or for addresses of per cpu variables.
 *
 * These operation guarantee exclusivity of access for other operations
 * on the *same* processor. The assumption is that per cpu data is only
 * accessed by a single processor instance (the current one).
 *
 * The arch code can provide optimized implementation by defining macros
 * for certain scalar sizes. F.e. provide this_cpu_add_2() to provide per
 * cpu atomic operations for 2 byte sized RMW actions. If arch code does
 * not provide operations for a scalar size then the fallback in the
 * generic code will be used.
 *
 * cmpxchg_double replaces two adjacent scalars at once.  The first two
 * parameters are per cpu variables which have to be of the same size.  A
 * truth value is returned to indicate success or failure (since a double
 * register result is difficult to handle).  There is very limited hardware
 * support for these operations, so only certain sizes may work.
 */

/*
 * Operations for contexts where we do not want to do any checks for
 * preemptions.  Unless strictly necessary, always use [__]this_cpu_*()
 * instead.
 *
 * If there is no other protection through preempt disable and/or disabling
 * interrupts then one of these RMW operations can show unexpected behavior
 * because the execution thread was rescheduled on another processor or an
 * interrupt occurred and the same percpu variable was modified from the
 * interrupt context.
 */
#define raw_cpu_read(pcp)        __pcpu_size_call_return(raw_cpu_read_, pcp)
#define raw_cpu_write(pcp, val)        __pcpu_size_call(raw_cpu_write_, pcp, val)
#define raw_cpu_add(pcp, val)        __pcpu_size_call(raw_cpu_add_, pcp, val)
#define raw_cpu_and(pcp, val)        __pcpu_size_call(raw_cpu_and_, pcp, val)
#define raw_cpu_or(pcp, val)        __pcpu_size_call(raw_cpu_or_, pcp, val)
#define raw_cpu_add_return(pcp, val)    __pcpu_size_call_return2(raw_cpu_add_return_, pcp, val)
#define raw_cpu_xchg(pcp, nval)        __pcpu_size_call_return2(raw_cpu_xchg_, pcp, nval)
#define raw_cpu_cmpxchg(pcp, oval, nval) \
    __pcpu_size_call_return2(raw_cpu_cmpxchg_, pcp, oval, nval)
#define raw_cpu_try_cmpxchg(pcp, ovalp, nval) \
    __pcpu_size_call_return2bool(raw_cpu_try_cmpxchg_, pcp, ovalp, nval)
#define raw_cpu_sub(pcp, val)        raw_cpu_add(pcp, -(val))
#define raw_cpu_inc(pcp)        raw_cpu_add(pcp, 1)
#define raw_cpu_dec(pcp)        raw_cpu_sub(pcp, 1)
#define raw_cpu_sub_return(pcp, val)    raw_cpu_add_return(pcp, -(typeof(pcp))(val))
#define raw_cpu_inc_return(pcp)        raw_cpu_add_return(pcp, 1)
#define raw_cpu_dec_return(pcp)        raw_cpu_add_return(pcp, -1)

/*
 * Operations for contexts that are safe from preemption/interrupts.  These
 * operations verify that preemption is disabled.
 */
#define __this_cpu_read(pcp)                        \
({                                    \
    __this_cpu_preempt_check("read");                \
    raw_cpu_read(pcp);                        \
})

#define __this_cpu_write(pcp, val)                    \
({                                    \
    __this_cpu_preempt_check("write");                \
    raw_cpu_write(pcp, val);                    \
})

#define __this_cpu_add(pcp, val)                    \
({                                    \
    __this_cpu_preempt_check("add");                \
    raw_cpu_add(pcp, val);                        \
})

#define __this_cpu_and(pcp, val)                    \
({                                    \
    __this_cpu_preempt_check("and");                \
    raw_cpu_and(pcp, val);                        \
})

#define __this_cpu_or(pcp, val)                        \
({                                    \
    __this_cpu_preempt_check("or");                    \
    raw_cpu_or(pcp, val);                        \
})

#define __this_cpu_add_return(pcp, val)                    \
({                                    \
    __this_cpu_preempt_check("add_return");                \
    raw_cpu_add_return(pcp, val);                    \
})

#define __this_cpu_xchg(pcp, nval)                    \
({                                    \
    __this_cpu_preempt_check("xchg");                \
    raw_cpu_xchg(pcp, nval);                    \
})

#define __this_cpu_cmpxchg(pcp, oval, nval)                \
({                                    \
    __this_cpu_preempt_check("cmpxchg");                \
    raw_cpu_cmpxchg(pcp, oval, nval);                \
})

#define __this_cpu_sub(pcp, val)    __this_cpu_add(pcp, -(typeof(pcp))(val))
#define __this_cpu_inc(pcp)        __this_cpu_add(pcp, 1)
#define __this_cpu_dec(pcp)        __this_cpu_sub(pcp, 1)
#define __this_cpu_sub_return(pcp, val)    __this_cpu_add_return(pcp, -(typeof(pcp))(val))
#define __this_cpu_inc_return(pcp)    __this_cpu_add_return(pcp, 1)
#define __this_cpu_dec_return(pcp)    __this_cpu_add_return(pcp, -1)

/*
 * Operations with implied preemption/interrupt protection.  These
 * operations can be used without worrying about preemption or interrupt.
 */
#define this_cpu_read(pcp)        __pcpu_size_call_return(this_cpu_read_, pcp)
#define this_cpu_write(pcp, val)    __pcpu_size_call(this_cpu_write_, pcp, val)
#define this_cpu_add(pcp, val)        __pcpu_size_call(this_cpu_add_, pcp, val)
#define this_cpu_and(pcp, val)        __pcpu_size_call(this_cpu_and_, pcp, val)
#define this_cpu_or(pcp, val)        __pcpu_size_call(this_cpu_or_, pcp, val)
#define this_cpu_add_return(pcp, val)    __pcpu_size_call_return2(this_cpu_add_return_, pcp, val)
#define this_cpu_xchg(pcp, nval)    __pcpu_size_call_return2(this_cpu_xchg_, pcp, nval)
#define this_cpu_cmpxchg(pcp, oval, nval) \
    __pcpu_size_call_return2(this_cpu_cmpxchg_, pcp, oval, nval)
#define this_cpu_try_cmpxchg(pcp, ovalp, nval) \
    __pcpu_size_call_return2bool(this_cpu_try_cmpxchg_, pcp, ovalp, nval)
#define this_cpu_sub(pcp, val)        this_cpu_add(pcp, -(typeof(pcp))(val))
#define this_cpu_inc(pcp)        this_cpu_add(pcp, 1)
#define this_cpu_dec(pcp)        this_cpu_sub(pcp, 1)
#define this_cpu_sub_return(pcp, val)    this_cpu_add_return(pcp, -(typeof(pcp))(val))
#define this_cpu_inc_return(pcp)    this_cpu_add_return(pcp, 1)
#define this_cpu_dec_return(pcp)    this_cpu_add_return(pcp, -1)

#endif /* __ASSEMBLY__ */
#endif /* _LINUX_PERCPU_DEFS_H */

:: Command execute ::

Enter:
 
Select:
 

:: Search ::
  - regexp 

:: Upload ::
 
[ Read-Only ]

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

:: Go Dir ::
 
:: Go File ::
 

--[ c99shell v. 2.0 [PHP 7 Update] [25.02.2019] maintained by HackingTool | HackingTool | Generation time: 0.0036 ]--