C语言嵌入式处理器扩展：关键特性与专利通知

需积分: 10 46 浏览量更新于2024-07-21 收藏 365KB PDF 举报

本篇文章主要探讨了C语言在支持嵌入式处理器方面的重要扩展，由ISO/IEC JTC1 SC22 WG14于2006年4月4日发布为ISO/IEC TR 18037草案。这份技术报告并不是正式的国际标准，而是供审查和讨论的草稿，可能会随时进行修改，因此不应用于任何官方引用。 C语言作为一种广泛使用的编程语言，在嵌入式系统领域因其灵活性和性能而受到青睐。随着技术的发展，针对嵌入式处理器特性的C语言扩展变得至关重要。这些扩展旨在优化代码效率、降低内存占用和增强与硬件的交互能力，以适应资源受限的环境。文中提到的ISO/IEC TR 18037详细列出了针对嵌入式处理器设计的特定C语言特性，可能包括但不限于： 1. **低级硬件访问**：提供了对寄存器操作、中断处理和硬件控制寄存器的直接访问，允许程序员更精细地控制硬件资源。 2. **内存管理优化**：考虑到嵌入式系统的内存限制，可能包含对内存分配和释放的特殊处理，或者对数据结构和算法的优化，以减少内存消耗。 3. **实时性增强**：对于实时操作系统(RTOS)的支持，可能包括时间片管理、同步和互斥机制，以确保程序的可靠性和响应速度。 4. **能耗管理**：在电池供电的设备上，C语言扩展可能包含节能模式或低功耗模式的实现，如电源管理库和编译时优化。 5. **硬件抽象层(HAL)**：提供了一种标准化的方式来编写与不同嵌入式处理器兼容的代码，降低了移植成本。 6. **错误处理和调试**：针对小型系统，可能增加了对异常处理和故障检测的简化接口，以及针对嵌入式环境的调试工具支持。 7. **硬件描述语言(HDL)集成**：允许程序员将硬件描述与C代码无缝结合，提高了硬件和软件协同开发的效率。 8. **硬件驱动程序和外设支持**：为了连接各种传感器、接口和外围设备，C语言可能提供了一套统一的驱动编程框架。由于这是一份草案，它还可能包含了尚未标准化的新功能和建议，供业界专家和开发者评估其潜在价值和适用性。收到此草案的读者被邀请提交他们的意见，包括专利声明和相关文档，以帮助完善这个领域内的技术标准。 C语言的嵌入式处理器扩展是C语言发展的一个重要方向，它反映了编程语言如何随着技术进步不断适应新的应用场景和挑战。通过了解和掌握这些扩展，开发人员可以更高效地在资源受限的嵌入式环境中构建和维护高质量的应用。

ISO/IEC TR 18037:2006(E)

closest larger value representable by the fixed point result type may be returned as the result, even

if the result type can represent the value -1 exactly. The circumstances in which a 1 or -1 result

might be replaced in this manner are implementation-defined.

Note that the special treatment of the values 1 and -1 as result of a * operation, as indicated in this

paragraph, is only included to ensure that certain specific implementations, that otherwise would not

be conformant, can conform to this Technical Report; it is the intention to deprecate this treatment in

future revisions of this specification. For more discussion, see Annex A.7.

If the value of the second operand of the / operator is zero, the behavior is undefined.

According to the rules above, the result type of an arithmetic operation where one operand has a

fixed-point type and the other operand has an integer or a fixed-point type is always a fixed-point

type. Other combinations of operand and result types for arithmetic operations are supported

through special functions. The names for these functions are mulifx, divifx, fxdivi and

idivfx, where fx stands for one of r, lr, k, lk, ur, ulr, uk and ulk. The muli functions

multiply an integer operand by a fixed-point operand and return an integer value; the divi

functions divide the first (integer) operand by a fixed-point operand (divifx) yielding an integer

type result, or divide two integer operands (fxdivi) yielding a fixed-point type result; the idivfx

functions divide the two fixed-point type operands (with the same type) and return an integer result.

If an integer result of one of these functions overflows, the behavior is undefined.

4.1.6.2.2 Bitwise shift operators

Shifts of fixed-point values using the standard << and >> operators are defined to be equivalent to

multiplication or division by a power of two (including the resulting rounding and overflow behavior).

The right operand must have integer type and must be nonnegative and less than the total number

of (nonpadding) bits of the fixed-point operand (the left operand). The result type is the same as

that of the fixed-point operand. An exact result is calculated and then converted to the result type in

the same way as the other fixed-point arithmetic operators.

4.1.6.2.3 Relational operators, equality operators

The standard relational operators (<, <=, >=, and >) and equality operators (==, and !=) accept

fixed-point operands. When comparing fixed-point values with fixed-point values or integer values,

the values are compared directly; the values of the operands are not converted before the

comparison is made. Otherwise, the usual arithmetic conversions are applied before the

comparison is made.

4.1.6.3 Assignment operators

The standard assignment operators +=, -=, *=, and /= are defined in the usual way when either

operand is fixed-point.

The standard assignment operators <<= and >>= are defined in the usual way when the left

operand is fixed-point.

ISO/IEC TR 18037:2006(E)

4.1.7.3 The fixed-point bit countls functions

The bit count functions countlsfx, where fx stands for one of hr, r, lr, hk, k, lk, uhr, ur,

ulr, uhk, uk or ulk, take one fixed-point type argument (corresponding to fx); the result type is

int.

The integer return value of the above functions is defined as follows:

- if the value of the fixed-point argument is non-zero, the return value is the largest integer k for

which the expression a<<k does not overflow;

- if the value of the fixed-point argument is zero, an integer value is returned that is at least as

large as N, where N is the total number of value bits of the fixed-point type of the argument.

4.1.7.4 The bitwise fixed-point to integer conversion functions

The bitwise fixed-point to integer conversion functions bitsfx, where fx stands for one of hr, r,

lr, hk, k, lk, uhr, ur, ulr, uhk, uk or ulk, take one fixed-point type argument (corresponding

to fx); the type of the function is an implementation-defined integer type int_fx_t (for signed

fixed-point types) or uint_fx_t (for unsigned fixed-point types), defined in the <stdfix.h>

headerfile, that is large enough to hold all the bits in the fixed-point type.

The bitwise fixed-point to integer conversion functions return an integer value equal to the fixed-

point value of the argument multiplied by 2

, where F is the number of fractional bits in the fixed-

point type. The result type is an integer type big enough to hold all valid result values for the given

fixed-point argument type. For example, if the fract type has 15 fractional bits, then after the

declaration

fract a = 0.5;

the value of bitsr(a) is 0.5 * 2^15 = 0x4000.

4.1.7.5 The bitwise integer to fixed-point conversion functions

The bitwise integer to fixed-point conversion functions fxbits, where fx stands for one of hr, r,

lr, hk, k, lk, uhr, ur, ulr, uhk, uk or ulk, take one argument with type int_fx_t or

uint_fx_t, the result type is a fixed-point type (corresponding to fx).

The bitwise fixed-point to integer conversion functions return an fixed-point value equal to the

integer value of the argument divided by 2

, where F is the number of fractional bits in the fixed-

point result type of the function. For example, if fract has 15 fractional bits, then the value of

rbits(0x2000) is 0.25.

ISO/IEC TR 18037:2006(E)

4.1.7.6 Type-generic fixed-point functions

The header <stdfix.h> defines the following fixed-point type-generic macros. For each of the

fixed-point absolute value functions in 4.1.7.1, the fixed-point round functions in 4.1.7.2 and the

fixed-point countls functions in 4.1.7.3, a type-generic macro is defined as follows:

type-generic macro

the fixed-point absolute value functions

absfx

the fixed-point rounding functions

roundfx

the fixed-point countls functions

countlsfx

4.1.7.7 Fixed-point numeric conversion functions

The fixed-point numeric conversion functions strtofxfx, where fx stands for one of hr, r, lr,

hk, k, lk, uhr, ur, ulr, uhk, uk or ulk, take two arguments: the first argument has

const char * restrict type, the second argument has char ** restrict type; the

result type is a fixed-point type (corresponding to fx).

Similar to the strtod function, the strtofxfx functions convert a portion of the string pointed to

by the first argument to a fixed-point representation, with a type corresponding to fx, and return that

fixed-point type value.

4.1.8 Fixed-point definitions <stdfix.h>

The header <stdfix.h> defines macros that specify the precision of the fixed-point types and

declares functions that support fixed-point arithmetic.

4.1.9 Formatted I/O functions for fixed-point arguments

Additional conversion specifiers for fixed-point arguments are defined as follows:

r for (signed) fract types

R for unsigned fract types

k for (signed) accum types

K for unsigned accum types.

Together with the standard length modifiers h (for short fixed-point arguments) and l (for long fixed-

point arguments) all fixed-point types can be converted in the normal manner. Conversions to and

from infinity and NaN representations are not supported.

The fprintf function and its derived functions with the r, R, k and K conversion formats convert

the argument with a fixed-point type representing a fixed-point number to decimal notation in the

style [-]ddd.ddd, where the number of digits after the decimal point is equal to the precision

specification (i.e., it corresponds to the output format of the f (F) conversion specifier).

ISO/IEC TR 18037:2006(E)

The fscanf function and its derived functions match an optionally signed fixed-point number

whose format is the same as expected for the subject sequence of the corresponding strtofxfx

function. The corresponding argument of fscanf shall be a pointer to a fixed-point type variable with

a type corresponding to fx.

4.2 Detailed changes to ISO/IEC 9899:1999

This clause details the modifications to ISO/IEC 9899:1999 needed to incorporate the fixed-point

functionality as described in clause 4.1 of this Technical Report. The changes listed in this clause

are limited to syntax and semantics; examples, (forward) references and other descriptive

information are omitted. The modifications are ordered according to the clauses of

ISO/IEC 9899:1999 to which they refer. If a clause of ISO/IEC 9899:1999 is not mentioned, no

changes to that clause are needed. New clauses are indicated with (NEW CLAUSE), however

resulting changes in the existing numbering are not indicated; the clause number mm.nna of new

clause indicates that this clause follows immediately clause mm.nn at the same level.

Clause 5.2.4.2.3 - Characteristics of fixed-point types (NEW CLAUSE)

The characteristics of fixed-point data types are defined in terms of a model that describes a

representation of fixed-point numbers and values that provide information about an implementation's

fixed-point arithmetic. (The fixed-point model is intended to clarify the description of each fixed-point

characteristic and does not require the fixed-point arithmetic of the implementation to be identical.)

Analogous to the Scaled data type, as defined in ISO/IEC 11404:1996 - Language-Independent

Datatypes (LID), a fixed-point number ( x ) is defined by the following model:

x = s * n * (b

)

with the following parameters:

s sign (±1)

b base or radix of nominator representation (an integer > 1)

p precision (the number of base-b digits in the nominator)

n nominator (nonnegative integer less than b raised to the power p)

f factor (an integer value).

For the purpose of this Technical Report, the following restrictions to the above general model

apply:

- b equals 2: only binary fixed-point is considered;

- (-p) <= f < 0: integer values (f >= 0) are not considered to form part of the fixed-point values, and

the radix dot is assumed to be somewhere between the most significant digit and the least

significant digit in the nominator, or immediately to the left of the most significant digit in the

nominator.

Fixed-point infinities or NaNs are not supported.

剩余104页未读，继续阅读

npu_wengjun

粉丝: 0
资源: 1

C语言嵌入式处理器扩展：关键特性与专利通知

architecture-instruction-set-extensions-programming-reference

file_system_util.rar_IEC

Intel Architecture - Instruction Set Extensions and Future Features - Programming Reference - March 2020 (319433-038)-计算机科学

ieee-standard-for-extensions-to-standard-test-interface-language.pdf

stable-diffusion-webui-extensions 扩展

05-27-2015_Intro-to-App-Extensions-Using-Swift:Will Zink 使用 Swift 介绍 Apple Watch 和应用程序扩展

Novosec - Bouncy Castle - Extensions-开源

Win32 Perl Programming-The Standard Extensions

Intel Architecture Instruction Set Extensions Programming Reference-2013.7

Git-2.37.3-64-bit 和 GitExtensions-3.5.4 安装包

最新资源