/* * Copyright (c) 2011-2014, Wind River Systems, Inc. * * SPDX-License-Identifier: Apache-2.0 */ /** * @file * @brief Misc utilities * * Misc utilities usable by the kernel and application code. */ #ifndef ZEPHYR_INCLUDE_SYS_UTIL_H_ #define ZEPHYR_INCLUDE_SYS_UTIL_H_ #include #include /* needs to be outside _ASMLANGUAGE so 'true' and 'false' can turn * into '1' and '0' for asm or linker scripts */ #include #ifndef _ASMLANGUAGE #include #include #include #include /** @brief Number of bits that make up a type */ #define NUM_BITS(t) (sizeof(t) * 8) #ifdef __cplusplus extern "C" { #endif /** * @defgroup sys-util Utility Functions * @since 2.4 * @version 0.1.0 * @ingroup utilities * @{ */ /** @brief Cast @p x, a pointer, to an unsigned integer. */ #define POINTER_TO_UINT(x) ((uintptr_t) (x)) /** @brief Cast @p x, an unsigned integer, to a void*. */ #define UINT_TO_POINTER(x) ((void *) (uintptr_t) (x)) /** @brief Cast @p x, a pointer, to a signed integer. */ #define POINTER_TO_INT(x) ((intptr_t) (x)) /** @brief Cast @p x, a signed integer, to a void*. */ #define INT_TO_POINTER(x) ((void *) (intptr_t) (x)) #if !(defined(__CHAR_BIT__) && defined(__SIZEOF_LONG__) && defined(__SIZEOF_LONG_LONG__)) # error Missing required predefined macros for BITS_PER_LONG calculation #endif /** Number of bits in a long int. */ #define BITS_PER_LONG (__CHAR_BIT__ * __SIZEOF_LONG__) /** Number of bits in a long long int. */ #define BITS_PER_LONG_LONG (__CHAR_BIT__ * __SIZEOF_LONG_LONG__) /** * @brief Create a contiguous bitmask starting at bit position @p l * and ending at position @p h. */ #define GENMASK(h, l) \ (((~0UL) - (1UL << (l)) + 1) & (~0UL >> (BITS_PER_LONG - 1 - (h)))) /** * @brief Create a contiguous 64-bit bitmask starting at bit position @p l * and ending at position @p h. */ #define GENMASK64(h, l) \ (((~0ULL) - (1ULL << (l)) + 1) & (~0ULL >> (BITS_PER_LONG_LONG - 1 - (h)))) /** @brief Extract the Least Significant Bit from @p value. */ #define LSB_GET(value) ((value) & -(value)) /** * @brief Extract a bitfield element from @p value corresponding to * the field mask @p mask. */ #define FIELD_GET(mask, value) (((value) & (mask)) / LSB_GET(mask)) /** * @brief Prepare a bitfield element using @p value with @p mask representing * its field position and width. The result should be combined * with other fields using a logical OR. */ #define FIELD_PREP(mask, value) (((value) * LSB_GET(mask)) & (mask)) /** @brief 0 if @p cond is true-ish; causes a compile error otherwise. */ #define ZERO_OR_COMPILE_ERROR(cond) ((int) sizeof(char[1 - 2 * !(cond)]) - 1) #if defined(__cplusplus) /* The built-in function used below for type checking in C is not * supported by GNU C++. */ #define ARRAY_SIZE(array) (sizeof(array) / sizeof((array)[0])) #else /* __cplusplus */ /** * @brief Zero if @p array has an array type, a compile error otherwise * * This macro is available only from C, not C++. */ #define IS_ARRAY(array) \ ZERO_OR_COMPILE_ERROR( \ !__builtin_types_compatible_p(__typeof__(array), \ __typeof__(&(array)[0]))) /** * @brief Number of elements in the given @p array * * In C++, due to language limitations, this will accept as @p array * any type that implements operator[]. The results may not be * particularly meaningful in this case. * * In C, passing a pointer as @p array causes a compile error. */ #define ARRAY_SIZE(array) \ ((size_t) (IS_ARRAY(array) + (sizeof(array) / sizeof((array)[0])))) #endif /* __cplusplus */ /** * @brief Whether @p ptr is an element of @p array * * This macro can be seen as a slightly stricter version of @ref PART_OF_ARRAY * in that it also ensures that @p ptr is aligned to an array-element boundary * of @p array. * * In C, passing a pointer as @p array causes a compile error. * * @param array the array in question * @param ptr the pointer to check * * @return 1 if @p ptr is part of @p array, 0 otherwise */ #define IS_ARRAY_ELEMENT(array, ptr) \ ((ptr) && POINTER_TO_UINT(array) <= POINTER_TO_UINT(ptr) && \ POINTER_TO_UINT(ptr) < POINTER_TO_UINT(&(array)[ARRAY_SIZE(array)]) && \ (POINTER_TO_UINT(ptr) - POINTER_TO_UINT(array)) % sizeof((array)[0]) == 0) /** * @brief Index of @p ptr within @p array * * With `CONFIG_ASSERT=y`, this macro will trigger a runtime assertion * when @p ptr does not fall into the range of @p array or when @p ptr * is not aligned to an array-element boundary of @p array. * * In C, passing a pointer as @p array causes a compile error. * * @param array the array in question * @param ptr pointer to an element of @p array * * @return the array index of @p ptr within @p array, on success */ #define ARRAY_INDEX(array, ptr) \ ({ \ __ASSERT_NO_MSG(IS_ARRAY_ELEMENT(array, ptr)); \ (__typeof__((array)[0]) *)(ptr) - (array); \ }) /** * @brief Check if a pointer @p ptr lies within @p array. * * In C but not C++, this causes a compile error if @p array is not an array * (e.g. if @p ptr and @p array are mixed up). * * @param array an array * @param ptr a pointer * @return 1 if @p ptr is part of @p array, 0 otherwise */ #define PART_OF_ARRAY(array, ptr) \ ((ptr) && POINTER_TO_UINT(array) <= POINTER_TO_UINT(ptr) && \ POINTER_TO_UINT(ptr) < POINTER_TO_UINT(&(array)[ARRAY_SIZE(array)])) /** * @brief Array-index of @p ptr within @p array, rounded down * * This macro behaves much like @ref ARRAY_INDEX with the notable * difference that it accepts any @p ptr in the range of @p array rather than * exclusively a @p ptr aligned to an array-element boundary of @p array. * * With `CONFIG_ASSERT=y`, this macro will trigger a runtime assertion * when @p ptr does not fall into the range of @p array. * * In C, passing a pointer as @p array causes a compile error. * * @param array the array in question * @param ptr pointer to an element of @p array * * @return the array index of @p ptr within @p array, on success */ #define ARRAY_INDEX_FLOOR(array, ptr) \ ({ \ __ASSERT_NO_MSG(PART_OF_ARRAY(array, ptr)); \ (POINTER_TO_UINT(ptr) - POINTER_TO_UINT(array)) / sizeof((array)[0]); \ }) /** * @brief Iterate over members of an array using an index variable * * @param array the array in question * @param idx name of array index variable */ #define ARRAY_FOR_EACH(array, idx) for (size_t idx = 0; (idx) < ARRAY_SIZE(array); ++(idx)) /** * @brief Iterate over members of an array using a pointer * * @param array the array in question * @param ptr pointer to an element of @p array */ #define ARRAY_FOR_EACH_PTR(array, ptr) \ for (__typeof__(*(array)) *ptr = (array); (size_t)((ptr) - (array)) < ARRAY_SIZE(array); \ ++(ptr)) /** * @brief Validate if two entities have a compatible type * * @param a the first entity to be compared * @param b the second entity to be compared * @return 1 if the two elements are compatible, 0 if they are not */ #define SAME_TYPE(a, b) __builtin_types_compatible_p(__typeof__(a), __typeof__(b)) /** * @brief Validate CONTAINER_OF parameters, only applies to C mode. */ #ifndef __cplusplus #define CONTAINER_OF_VALIDATE(ptr, type, field) \ BUILD_ASSERT(SAME_TYPE(*(ptr), ((type *)0)->field) || \ SAME_TYPE(*(ptr), void), \ "pointer type mismatch in CONTAINER_OF"); #else #define CONTAINER_OF_VALIDATE(ptr, type, field) #endif /** * @brief Get a pointer to a structure containing the element * * Example: * * struct foo { * int bar; * }; * * struct foo my_foo; * int *ptr = &my_foo.bar; * * struct foo *container = CONTAINER_OF(ptr, struct foo, bar); * * Above, @p container points at @p my_foo. * * @param ptr pointer to a structure element * @param type name of the type that @p ptr is an element of * @param field the name of the field within the struct @p ptr points to * @return a pointer to the structure that contains @p ptr */ #define CONTAINER_OF(ptr, type, field) \ ({ \ CONTAINER_OF_VALIDATE(ptr, type, field) \ ((type *)(((char *)(ptr)) - offsetof(type, field))); \ }) /** * @brief Report the size of a struct field in bytes. * * @param type The structure containing the field of interest. * @param member The field to return the size of. * * @return The field size. */ #define SIZEOF_FIELD(type, member) sizeof((((type *)0)->member)) /** * @brief Concatenate input arguments * * Concatenate provided tokens into a combined token during the preprocessor pass. * This can be used to, for ex., build an identifier out of multiple parts, * where one of those parts may be, for ex, a number, another macro, or a macro argument. * * @param ... Tokens to concatencate * * @return Concatenated token. */ #define CONCAT(...) \ UTIL_CAT(_CONCAT_, NUM_VA_ARGS_LESS_1(__VA_ARGS__))(__VA_ARGS__) /** * @brief Check if @p ptr is aligned to @p align alignment */ #define IS_ALIGNED(ptr, align) (((uintptr_t)(ptr)) % (align) == 0) /** * @brief Value of @p x rounded up to the next multiple of @p align. */ #define ROUND_UP(x, align) \ ((((unsigned long)(x) + ((unsigned long)(align) - 1)) / \ (unsigned long)(align)) * (unsigned long)(align)) /** * @brief Value of @p x rounded down to the previous multiple of @p align. */ #define ROUND_DOWN(x, align) \ (((unsigned long)(x) / (unsigned long)(align)) * (unsigned long)(align)) /** @brief Value of @p x rounded up to the next word boundary. */ #define WB_UP(x) ROUND_UP(x, sizeof(void *)) /** @brief Value of @p x rounded down to the previous word boundary. */ #define WB_DN(x) ROUND_DOWN(x, sizeof(void *)) /** * @brief Divide and round up. * * Example: * @code{.c} * DIV_ROUND_UP(1, 2); // 1 * DIV_ROUND_UP(3, 2); // 2 * @endcode * * @param n Numerator. * @param d Denominator. * * @return The result of @p n / @p d, rounded up. */ #define DIV_ROUND_UP(n, d) (((n) + (d) - 1) / (d)) /** * @brief Divide and round to the nearest integer. * * Example: * @code{.c} * DIV_ROUND_CLOSEST(5, 2); // 3 * DIV_ROUND_CLOSEST(5, -2); // -3 * DIV_ROUND_CLOSEST(5, 3); // 2 * @endcode * * @param n Numerator. * @param d Denominator. * * @return The result of @p n / @p d, rounded to the nearest integer. */ #define DIV_ROUND_CLOSEST(n, d) \ ((((n) < 0) ^ ((d) < 0)) ? ((n) - ((d) / 2)) / (d) : \ ((n) + ((d) / 2)) / (d)) /** * @brief Ceiling function applied to @p numerator / @p divider as a fraction. * @deprecated Use DIV_ROUND_UP() instead. */ #define ceiling_fraction(numerator, divider) __DEPRECATED_MACRO \ DIV_ROUND_UP(numerator, divider) #ifndef MAX /** * @brief Obtain the maximum of two values. * * @note Arguments are evaluated twice. Use Z_MAX for a GCC-only, single * evaluation version * * @param a First value. * @param b Second value. * * @returns Maximum value of @p a and @p b. */ #define MAX(a, b) (((a) > (b)) ? (a) : (b)) #endif #ifndef MIN /** * @brief Obtain the minimum of two values. * * @note Arguments are evaluated twice. Use Z_MIN for a GCC-only, single * evaluation version * * @param a First value. * @param b Second value. * * @returns Minimum value of @p a and @p b. */ #define MIN(a, b) (((a) < (b)) ? (a) : (b)) #endif #ifndef CLAMP /** * @brief Clamp a value to a given range. * * @note Arguments are evaluated multiple times. Use Z_CLAMP for a GCC-only, * single evaluation version. * * @param val Value to be clamped. * @param low Lowest allowed value (inclusive). * @param high Highest allowed value (inclusive). * * @returns Clamped value. */ #define CLAMP(val, low, high) (((val) <= (low)) ? (low) : MIN(val, high)) #endif /** * @brief Checks if a value is within range. * * @note @p val is evaluated twice. * * @param val Value to be checked. * @param min Lower bound (inclusive). * @param max Upper bound (inclusive). * * @retval true If value is within range * @retval false If the value is not within range */ #define IN_RANGE(val, min, max) ((val) >= (min) && (val) <= (max)) /** * @brief Is @p x a power of two? * @param x value to check * @return true if @p x is a power of two, false otherwise */ static inline bool is_power_of_two(unsigned int x) { return IS_POWER_OF_TWO(x); } /** * @brief Is @p p equal to ``NULL``? * * Some macros may need to check their arguments against NULL to support * multiple use-cases, but NULL checks can generate warnings if such a macro * is used in contexts where that particular argument can never be NULL. * * The warnings can be triggered if: * a) all macros are expanded (e.g. when using CONFIG_COMPILER_SAVE_TEMPS=y) * or * b) tracking of macro expansions are turned off (-ftrack-macro-expansion=0) * * The warnings can be circumvented by using this inline function for doing * the NULL check within the macro. The compiler is still able to optimize the * NULL check out at a later stage. * * @param p Pointer to check * @return true if @p p is equal to ``NULL``, false otherwise */ static ALWAYS_INLINE bool is_null_no_warn(void *p) { return p == NULL; } /** * @brief Arithmetic shift right * @param value value to shift * @param shift number of bits to shift * @return @p value shifted right by @p shift; opened bit positions are * filled with the sign bit */ static inline int64_t arithmetic_shift_right(int64_t value, uint8_t shift) { int64_t sign_ext; if (shift == 0U) { return value; } /* extract sign bit */ sign_ext = (value >> 63) & 1; /* make all bits of sign_ext be the same as the value's sign bit */ sign_ext = -sign_ext; /* shift value and fill opened bit positions with sign bit */ return (value >> shift) | (sign_ext << (64 - shift)); } /** * @brief byte by byte memcpy. * * Copy `size` bytes of `src` into `dest`. This is guaranteed to be done byte by byte. * * @param dst Pointer to the destination memory. * @param src Pointer to the source of the data. * @param size The number of bytes to copy. */ static inline void bytecpy(void *dst, const void *src, size_t size) { size_t i; for (i = 0; i < size; ++i) { ((volatile uint8_t *)dst)[i] = ((volatile const uint8_t *)src)[i]; } } /** * @brief byte by byte swap. * * Swap @a size bytes between memory regions @a a and @a b. This is * guaranteed to be done byte by byte. * * @param a Pointer to the first memory region. * @param b Pointer to the second memory region. * @param size The number of bytes to swap. */ static inline void byteswp(void *a, void *b, size_t size) { uint8_t t; uint8_t *aa = (uint8_t *)a; uint8_t *bb = (uint8_t *)b; for (; size > 0; --size) { t = *aa; *aa++ = *bb; *bb++ = t; } } /** * @brief Convert a single character into a hexadecimal nibble. * * @param c The character to convert * @param x The address of storage for the converted number. * * @return Zero on success or (negative) error code otherwise. */ int char2hex(char c, uint8_t *x); /** * @brief Convert a single hexadecimal nibble into a character. * * @param c The number to convert * @param x The address of storage for the converted character. * * @return Zero on success or (negative) error code otherwise. */ int hex2char(uint8_t x, char *c); /** * @brief Convert a binary array into string representation. * * @param buf The binary array to convert * @param buflen The length of the binary array to convert * @param hex Address of where to store the string representation. * @param hexlen Size of the storage area for string representation. * * @return The length of the converted string, or 0 if an error occurred. */ size_t bin2hex(const uint8_t *buf, size_t buflen, char *hex, size_t hexlen); /** * @brief Convert a hexadecimal string into a binary array. * * @param hex The hexadecimal string to convert * @param hexlen The length of the hexadecimal string to convert. * @param buf Address of where to store the binary data * @param buflen Size of the storage area for binary data * * @return The length of the binary array, or 0 if an error occurred. */ size_t hex2bin(const char *hex, size_t hexlen, uint8_t *buf, size_t buflen); /** * @brief Convert a binary coded decimal (BCD 8421) value to binary. * * @param bcd BCD 8421 value to convert. * * @return Binary representation of input value. */ static inline uint8_t bcd2bin(uint8_t bcd) { return ((10 * (bcd >> 4)) + (bcd & 0x0F)); } /** * @brief Convert a binary value to binary coded decimal (BCD 8421). * * @param bin Binary value to convert. * * @return BCD 8421 representation of input value. */ static inline uint8_t bin2bcd(uint8_t bin) { return (((bin / 10) << 4) | (bin % 10)); } /** * @brief Convert a uint8_t into a decimal string representation. * * Convert a uint8_t value into its ASCII decimal string representation. * The string is terminated if there is enough space in buf. * * @param buf Address of where to store the string representation. * @param buflen Size of the storage area for string representation. * @param value The value to convert to decimal string * * @return The length of the converted string (excluding terminator if * any), or 0 if an error occurred. */ uint8_t u8_to_dec(char *buf, uint8_t buflen, uint8_t value); /** * @brief Sign extend an 8, 16 or 32 bit value using the index bit as sign bit. * * @param value The value to sign expand. * @param index 0 based bit index to sign bit (0 to 31) */ static inline int32_t sign_extend(uint32_t value, uint8_t index) { __ASSERT_NO_MSG(index <= 31); uint8_t shift = 31 - index; return (int32_t)(value << shift) >> shift; } /** * @brief Sign extend a 64 bit value using the index bit as sign bit. * * @param value The value to sign expand. * @param index 0 based bit index to sign bit (0 to 63) */ static inline int64_t sign_extend_64(uint64_t value, uint8_t index) { __ASSERT_NO_MSG(index <= 63); uint8_t shift = 63 - index; return (int64_t)(value << shift) >> shift; } /** * @brief Properly truncate a NULL-terminated UTF-8 string * * Take a NULL-terminated UTF-8 string and ensure that if the string has been * truncated (by setting the NULL terminator) earlier by other means, that * the string ends with a properly formatted UTF-8 character (1-4 bytes). * * @htmlonly * Example: * char test_str[] = "€€€"; * char trunc_utf8[8]; * * printf("Original : %s\n", test_str); // €€€ * strncpy(trunc_utf8, test_str, sizeof(trunc_utf8)); * trunc_utf8[sizeof(trunc_utf8) - 1] = '\0'; * printf("Bad : %s\n", trunc_utf8); // €€� * utf8_trunc(trunc_utf8); * printf("Truncated: %s\n", trunc_utf8); // €€ * @endhtmlonly * * @param utf8_str NULL-terminated string * * @return Pointer to the @p utf8_str */ char *utf8_trunc(char *utf8_str); /** * @brief Copies a UTF-8 encoded string from @p src to @p dst * * The resulting @p dst will always be NULL terminated if @p n is larger than 0, * and the @p dst string will always be properly UTF-8 truncated. * * @param dst The destination of the UTF-8 string. * @param src The source string * @param n The size of the @p dst buffer. Maximum number of characters copied * is @p n - 1. If 0 nothing will be done, and the @p dst will not be * NULL terminated. * * @return Pointer to the @p dst */ char *utf8_lcpy(char *dst, const char *src, size_t n); #define __z_log2d(x) (32 - __builtin_clz(x) - 1) #define __z_log2q(x) (64 - __builtin_clzll(x) - 1) #define __z_log2(x) (sizeof(__typeof__(x)) > 4 ? __z_log2q(x) : __z_log2d(x)) /** * @brief Compute log2(x) * * @note This macro expands its argument multiple times (to permit use * in constant expressions), which must not have side effects. * * @param x An unsigned integral value to compute logarithm of (positive only) * * @return log2(x) when 1 <= x <= max(x), -1 when x < 1 */ #define LOG2(x) ((x) < 1 ? -1 : __z_log2(x)) /** * @brief Compute ceil(log2(x)) * * @note This macro expands its argument multiple times (to permit use * in constant expressions), which must not have side effects. * * @param x An unsigned integral value * * @return ceil(log2(x)) when 1 <= x <= max(type(x)), 0 when x < 1 */ #define LOG2CEIL(x) ((x) < 1 ? 0 : __z_log2((x)-1) + 1) /** * @brief Compute next highest power of two * * Equivalent to 2^ceil(log2(x)) * * @note This macro expands its argument multiple times (to permit use * in constant expressions), which must not have side effects. * * @param x An unsigned integral value * * @return 2^ceil(log2(x)) or 0 if 2^ceil(log2(x)) would saturate 64-bits */ #define NHPOT(x) ((x) < 1 ? 1 : ((x) > (1ULL<<63) ? 0 : 1ULL << LOG2CEIL(x))) /** * @brief Determine if a buffer exceeds highest address * * This macro determines if a buffer identified by a starting address @a addr * and length @a buflen spans a region of memory that goes beyond the highest * possible address (thereby resulting in a pointer overflow). * * @param addr Buffer starting address * @param buflen Length of the buffer * * @return true if pointer overflow detected, false otherwise */ #define Z_DETECT_POINTER_OVERFLOW(addr, buflen) \ (((buflen) != 0) && \ ((UINTPTR_MAX - (uintptr_t)(addr)) <= ((uintptr_t)((buflen) - 1)))) /** * @brief XOR n bytes * * @param dst Destination of where to store result. Shall be @p len bytes. * @param src1 First source. Shall be @p len bytes. * @param src2 Second source. Shall be @p len bytes. * @param len Number of bytes to XOR. */ static inline void mem_xor_n(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, size_t len) { while (len--) { *dst++ = *src1++ ^ *src2++; } } /** * @brief XOR 32 bits * * @param dst Destination of where to store result. Shall be 32 bits. * @param src1 First source. Shall be 32 bits. * @param src2 Second source. Shall be 32 bits. */ static inline void mem_xor_32(uint8_t dst[4], const uint8_t src1[4], const uint8_t src2[4]) { mem_xor_n(dst, src1, src2, 4U); } /** * @brief XOR 128 bits * * @param dst Destination of where to store result. Shall be 128 bits. * @param src1 First source. Shall be 128 bits. * @param src2 Second source. Shall be 128 bits. */ static inline void mem_xor_128(uint8_t dst[16], const uint8_t src1[16], const uint8_t src2[16]) { mem_xor_n(dst, src1, src2, 16); } #ifdef __cplusplus } #endif /* This file must be included at the end of the !_ASMLANGUAGE guard. * It depends on macros defined in this file above which cannot be forward declared. */ #include #endif /* !_ASMLANGUAGE */ /** @brief Number of bytes in @p x kibibytes */ #ifdef _LINKER /* This is used in linker scripts so need to avoid type casting there */ #define KB(x) ((x) << 10) #else #define KB(x) (((size_t)(x)) << 10) #endif /** @brief Number of bytes in @p x mebibytes */ #define MB(x) (KB(x) << 10) /** @brief Number of bytes in @p x gibibytes */ #define GB(x) (MB(x) << 10) /** @brief Number of Hz in @p x kHz */ #define KHZ(x) ((x) * 1000) /** @brief Number of Hz in @p x MHz */ #define MHZ(x) (KHZ(x) * 1000) /** * @brief For the POSIX architecture add a minimal delay in a busy wait loop. * For other architectures this is a no-op. * * In the POSIX ARCH, code takes zero simulated time to execute, * so busy wait loops become infinite loops, unless we * force the loop to take a bit of time. * Include this macro in all busy wait/spin loops * so they will also work when building for the POSIX architecture. * * @param t Time in microseconds we will busy wait */ #if defined(CONFIG_ARCH_POSIX) #define Z_SPIN_DELAY(t) k_busy_wait(t) #else #define Z_SPIN_DELAY(t) #endif /** * @brief Wait for an expression to return true with a timeout * * Spin on an expression with a timeout and optional delay between iterations * * Commonly needed when waiting on hardware to complete an asynchronous * request to read/write/initialize/reset, but useful for any expression. * * @param expr Truth expression upon which to poll, e.g.: XYZREG & XYZREG_EN * @param timeout Timeout to wait for in microseconds, e.g.: 1000 (1ms) * @param delay_stmt Delay statement to perform each poll iteration * e.g.: NULL, k_yield(), k_msleep(1) or k_busy_wait(1) * * @retval expr As a boolean return, if false then it has timed out. */ #define WAIT_FOR(expr, timeout, delay_stmt) \ ({ \ uint32_t _wf_cycle_count = k_us_to_cyc_ceil32(timeout); \ uint32_t _wf_start = k_cycle_get_32(); \ while (!(expr) && (_wf_cycle_count > (k_cycle_get_32() - _wf_start))) { \ delay_stmt; \ Z_SPIN_DELAY(10); \ } \ (expr); \ }) /** * @} */ #endif /* ZEPHYR_INCLUDE_SYS_UTIL_H_ */