LCLint User's Guide

Version 2.4
April 1998

David Evans
University of Virginia, Computer Science Group
MIT Laboratory for Computer Science

LCLint is a tool for statically checking C programs. With minimal effort, LCLint can be used as a better lint.[1] If additional effort is invested adding annotations to programs, LCLint can perform stronger checks than can be done by any standard lint.

Some problems detected by LCLint include:

LCLint checking can be customized to select what classes of errors are reported using command line flags and stylized comments in the code.

This document is a guide to using LCLint. Section 1 is a brief overview of the design goals of LCLint. Section 2 explains how to run LCLint, interpret messages produce, and control checking. Sections 3-10 describe particular checks done by LCLint.

Contents

  1. Overview
  2. Operation
  3. Abstract Types
  4. Function Interfaces
  5. Memory Management
  6. Sharing
  7. Value Constraints
  8. Macros
  9. Naming Conventions
  10. Other Checks

Appendices

Appendix A. Availability
Appendix B. Communication
Appendix C. Flags
Appendix D. Annotations
Appendix E. Control Comments
Appendix F. Libraries
Appendix G. Specifications
Appendix H. Emacs

Figures

References
Acknowledgements

1. Overview

The main goals for LCLint are to:

LCLint does many of the traditional lint checks including unused declarations, type inconsistencies, use-before-definition, unreachable code, ignored return values, execution paths with no return, likely infinite loops, and fall-through cases. This document focuses on more powerful checks that are made possible by additional information given in source code annotations. [2] Annotations are stylized comments that document certain assumptions about functions, variables, parameters, and types. They may be used to indicate where the representation of a user-defined type is hidden, to limit where a global variable may be used or modified, to constrain what a function implementation may do to its parameters, and to express checked assumptions about variables, types, structure fields, function parameters, and function results. In addition to the checks specifically enabled by annotations, many of the traditional lint checks are improved by exploiting this additional information.

2. Operation

LCLint is invoked by listing files to be checked. Initialization files, command line flags, and stylized comments may be used to customize checking globally and locally.

The best way to learn to use LCLint, of course, is to actually use it (if you don't already have LCLint installed on your system, download it). Before you read much further in this document, I recommend finding a small C program. Then, try running:

lclint *.c

For the most C programs, this will produce a large number of messages. To turn off reporting for some of the messages, try:

lclint -weak *.c

The -weak flag is a mode flag that sets many checking parameters to select weaker checking than is done in the default mode. Other LCLint flags will be introduced in the following sections; a complete list is given in Appendix C.

2.1 Messages

The format and content of messages printed by LCLint can be customized by the user. A typical message is:

sample.c: (in function faucet)
sample.c:11:12: Fresh storage x not released before return
  A memory leak has been detected.  Newly-allocated or only-qualified storage
  is not released before the last reference to is it lost.  Use -mustfree to
  suppress message.
   sample.c:5:47: Fresh storage x allocated
The first line gives the name of the function in which the error is found. This is printed before the first message reported for a function. (The function context is not printed if -showfunc is used.)

The second line is the text of the message. This message reports a memory leak - storage allocated in a function is not deallocated before the function returns. The text is preceded by the file name, line and column number where the error is located. The column numbers are used by the emacs mode (see Appendix H) to jump to the appropriate line and column location. (Column numbers are not printed if -showcolumn is used.)

The next line is a hint giving more information about the suspected error. Most hints also include information on how the message may be suppressed. For this message, setting the -mustfree flag would prevent the message from being reported. Hints may be turned off by using -hints. Normally, a hint is given only the first time a class of error is reported. To have LCLint print a hint for every message regardless, use +forcehints.

The final line of the message gives additional location information. For this message, it tells where the leaking storage is allocated.

The generic message format is (parts enclosed in square brackets are optional):

  [<file>:<line> (in <context>)]
  <file>:<line>[,<column>]: message
     [hint]
      <file>:<line>,<column>: extra location information, if appopriate

The text of messages and hints may be longer than one line. They are split into lines of length less than the value set using -linelen <number>. The default line length is 80 characters. LCLint attempts to split lines in a sensible place as near to the line length limit as possible.

The +parenfileformat flag can be used to generate file locations in the format recognized by Microsoft Developer Studio. If +parenfileformat is set, the line number follows the file name in parentheses (e.g., sample.c(11).).

2.2 Flags

So that many programming styles can be supported, LCLint provides over 300 flags for controlling checking and message reporting. Some of the flags are introduced in the body of this document. Apppendix C describes every flag. Modes and shortcut flags are provided for setting many flags at once. Individual flags can override the mode settings.

Flags are preceded by + or -. When a flag is preceded by + we say it is on; when it is preceded by - it is off. The precise meaning of on and off depends on the type of flag.

The +/- flag settings are used for consistency and clarity, but contradict standard UNIX usage and is easy to accidentally use the wrong one. To reduce the likelihood of using the wrong flag, LCLint issues warnings when a flag is set in an unusual way. Warnings are issued when a flag is redundantly set to the value it already had (these errors are not reported if the flag is set using a stylized comment), if a mode flag or special flag is set after a more specific flag that will be set by the general flag was already set, if value flags are given unreasonable values, of if flags are set in an inconsistent way. The -warnflags flag suppresses these warnings.

Default flag settings will be read from ~/.lclintrc if it is readable. If there is a .lclintrc file in the working directory, settings in this file will be read next and its settings will override those in ~/.lclintrc. Command-line flags override settings in either file. The syntax of the .lclintrc file is the same as that of command-line flags, except that flags may be on separate lines and the # character may be used to indicate that the remainder of the line is a comment. The -nof flag prevents the ~/.lclintrc file from being loaded. The -f <filename> flag loads options from filename.

2.3 Stylized Comments

Stylized comments are used to provide extra information about a type, variable or function interface to improve checking, or to control flag settings locally.

All stylized comments begin with /*@ and are closed by the end of the comment. The role of the @ may be played by any printable character. Use -commentchar <char> to select a different stylized comment marker.

2.3.1 Annotations

Annotations are stylized comments that follow a definite syntax. Although they are comments, they may only be used in fixed grammatical contexts (e.g., like a type qualifier).

Syntactic comments for function interfaces are described in Section 4; comments for declaring constants in Section 8.1. and comments for declaring iterators in Section 8.4. Sections 3-8 include descriptions of annotations for expressing assumptions about variables, parameters, return values, structure fields and type definitions. A summary of annotations is found in Apppendix D.

2.3.2 Control Comments

Unlike annotations, control comments may appear between any two tokens in a C program.[3] Syntactically, they are no different from standard comments. Control comments are used to provide source-level control of LCLint checking. They may be used to suppress spurious messages, set flags, and control checking locally in other ways. A complete description of control comments is found in Apppendix E.

Most flags (all except those characterized as "global" in Apppendix C) can be set locally using control comments. A control comment can set flags locally to override the command line settings. The original flag settings are restored before processing the next file. The syntax for setting flags in control comments is the same as that of the command line, except that flags may also be preceded by = to restore their setting to the original command-line value. For instance,

/*@+boolint -modifies =showfunc@*/
sets boolint on (this makes bool and int indistinguishable types), sets modifies off (this prevents reporting of modification errors), and sets showfunc to its original setting (this controls whether or not the name of a function is displayed before a message).

3. Abstract Types

Traditionally, programming books wax mathematical when they arrive at the topic of abstract data types… Such books make it seem as if you'd never actually use an abstract data type except as a sleep aid.

- Steve McConnell

Information hiding is a technique for handling complexity. By hiding implementation details, programs can be understood and developed in distinct modules and the effects of a change can be localized. One technique for information hiding is data abstraction. An abstract type is used to represent some natural program abstraction. It provides functions for manipulating instances of the type. The module that implements these functions is called the implementation module. We call the functions that are part of the implementation of an abstract type the operations of the type. Other modules that use the abstract type are called clients.

Clients may use the type name and operations, but should not manipulate or rely on the actual representation of the type. Only the implementation module may manipulate the representation of an abstract type. This hides information, since implementers and maintainers of client modules should not need to know anything about how the abstract type is implemented. It provides modularity, since the representation of an abstract type can be changed without having to change any client code.

LCLint supports abstract types by detecting places where client code depends on the concrete representation of an abstract type.

To declare an abstract type, the abstract annotation is added to a typedef. For example (in mstring.h),

typedef /*@abstract@*/ char *mstring;
declares mstring as an abstract type. It is implemented using a char *, but clients of the type should not depend on or need to be aware of this. If it later becomes apparent that a better representation such as a string table should be used, we should be able to change the implementation of mstring without having to change or inspect any client code.

In a client module, abstract types are checked by name, not structure. LCLint reports an error if an instance of mstring is passed as a char * (for instance, as an argument to strlen), since the correctness of this call depends on the representation of the abstract type. LCLint also reports errors if any C operator except assignment (=) or sizeof is used on an abstract type. The assignment operator is allowed since its semantics do not depend on the representation of the type.[4] The use of sizeof is also permitted, since this is the only way for clients to allocate pointers to the abstract type. Type casting objects to or from abstract types in a client module is an abstraction violation and will generate a warning message.

Normally, LCLint will assume a type definition is not abstract unless the /*@abstract@*/ qualifier is used. If instead you want all user-defined types to be abstract types unless they are marked as concrete, the +impabstract flag can be used. This adds an implicit abstract annotation to any typedef that is not marked with /*@concrete@*/.

Some examples of abstraction violations detected by LCLint are shown in Figure 2.

3.1 Access

Where code may manipulate the representation of an abstract type, we say the code has access to that type. If code has access to an abstract type, the representation of the type and the abstract type are indistinguishable. Usually, an abstract type is implemented by a single program module that is the only code that has access to the type representation. Sometimes, more complicated access control is desired if the implementation of an abstract type is split across program files, or particular client code needs to access the representation.

There are a several ways of selecting what code has access the representation of an abstract type:

3.2 Mutability

We can view types as being mutable or immutable. A type is mutable if the value of an instance of the type can be changed by passing it as a parameter to a function call.[7] For example, the primitive type int is immutable. If i is a local variable of type int and no variables point to the location where i is stored, the value of i must be the same before and after the call f(i). Structure and union types are also immutable, since they are copied when they are passed as arguments. On the other hand, pointer types are mutable. If x is a local variable of type int *, the value of *x (and hence, the value of the object x) can be changed by the function call g(x).

The mutability of a concrete type is determined by its type definition. For abstract types, mutability does not depend on the type representation but on what operations the type provides. If an abstract type has operations that may change the value of instances of the type, the type is mutable. If not, it is immutable. The value of an instance of an immutable type never changes. Since object sharing is noticeable only for mutable types, they are checked differently from immutable types.

The /*@mutable@*/ and /*@immutable@*/ annotations are used to declare an abstract type as mutable or immutable. (If neither is used, the abstract type is assumed to be mutable.) For example,

typedef /*@abstract@*/ /*@mutable@*/ char *mstring; typedef /*@abstract@*/ /*@immutable@*/ int weekDay; declares mstring as a mutable abstract type and weekDay as an immutable abstract type.

Clients of a mutable abstract type need to know the semantics of assignment. After the assignment expression s = t, do s and t refer to the same object (that is, will changes to the value of s also change the value of t)?

LCLint prescribes that all abstract types have sharing semantics, so s and t would indeed be the same object. LCLint will report an error if a mutable type is implemented with a representation (e.g., a struct) that does not provide sharing semantics (controlled by mutrep flag).

The mutability of an abstract type is not necessarily the same as the mutability of its representation. We could use the immutable concrete type int to represent mutable strings using an index into a string table, or declare mstring as immutable as long as no operations are provided that modify the value of an mstring.

3.3 Boolean Types

Standard C has no boolean representation - the result of a comparison operator is an integer, and no type checking is done for test expressions. Many common errors can be detected by introducing a distinct boolean type and stronger type checking.

Use the -booltype name flag to select the type name used to represent boolean values[8] Relations, comparisons and certain standard library functions are declared to return bool types.

LCLint checks that the test expression in an if, while, or for statement or an operand to &&, || or ! is a boolean. If the type of a test expression is not a boolean, LCLint will report an error depending on the type of the test expression and flag settings. If the test expression has pointer type, LCLint reports an error if predboolptr is on (this can be used to prevent messages for the idiom of testing if a pointer is not null without a comparison). If it is type int, an error is reported if predboolint is on. For all other types, LCLint reports an error if predboolothers is on.

Since using = instead of == is such a common bug, reporting of test expressions that are assignments is controlled by the separate predassign flag. The message can be suppressed by adding extra parentheses around the test expression.

Apppendix C (page 50) describes other flags for controlling boolean checking.

Figure 3. Boolean Checking

3.4 Primitive C Types

Two types have compatible type if their types are the same.
- ANSI C, 3.1.2.6.

Two types need not be identical to be compatible.
- ANSI C, footnote to 3.1.2.6.

LCLint supports stricter checking of primitive C types. The char and enum types can be checked as distinct types, and the different numeric types can be type-checked strictly.

3.4.1 Characters

The primitive char type can be type-checked as a distinct type. If char is used as a distinct type, common errors involving assigning ints to chars are detected.

The +charint flag can be used for checking legacy programs where char and int are used interchangeably. If charint is on, char types indistinguishable from ints. To keep char and int as distinct types, but allow chars to be used to index arrays, use +charindex.

3.4.2 Enumerators

Standard C treats user-declared enum types just like integers. An arbitrary integral value may be assigned to an enum type, whether or not it was listed as an enumerator member. LCLint checks each user-defined enum type as distinct type. An error is reported if a value that is not an enumerator member is assigned to the enum type, or if an enum type is used as an operand to an arithmetic operator.

If the enumint flag is on, enum and int types may be used interchangeably. Like charindex, if the enumindex flag is on, enum types may be used to index arrays.

3.4.3 Numeric Types

LCLint reports where numeric types are used in dangerous or inconsistent ways. With the strictest checking, LCLint will report an error anytime numeric types do not match exactly. If the relaxquals flag is on, only those inconsistencies which may corrupt values are reported. For example, if an int is assigned to a variable of type long (or passed as a long formal parameter), LCLint will not report an error if relaxquals is on since a long must have at least enough bits to store an int without data loss. On the other hand, an error would be reported if the long were assigned to an int, since the int type may not have enough bits to store the long value.

Similarly, if a signed value is assigned to an unsigned, LCLint will report an error since an unsigned type cannot represent all signed values correctly. If the ignoresigns flag is on, checking is relaxed to ignore all sign qualifiers in type comparisons (this is not recommended, since it will suppress reporting of real bugs, but may be necessary for quickly checking certain legacy code).

3.4.4 Arbitrary Integral Types

LCLint supports three different kinds of arbitrary integral types:

/*@integraltype@*/
An arbitrary integral type. The actual type may be any one of short, int, long, unsigned short, unsigned, or unsigned long.

/*@unsignedintegraltype@*/

An arbitrary unsigned integral type. The actual type may be any one of unsigned short, unsigned, or unsigned long.

/*@signedintegraltype@*/

An arbitrary signed integral type. The actual type may be any one of short, int, or long.

LCLint reports an error if the code depends on the actual representation of a type declared as an arbitrary integral. The match-any-integral flag relaxes checking and allows an arbitrary integral type is allowed to match any integral type.

Other flags set the arbitrary integral types to a concrete type. These should only be used if portability to platforms that may use different representations is not important. The long-integral and long-unsigned-integral flags set the type corresponding to /*@integraltype@*/ to be unsigned long and long respectively. The long-unsigned-unsigned-integral flag sets the type corresponding to /*@unsignedintegraltype@*/ to be unsigned long. The long-signed-integral flag sets the type corresponding to /*@signedintegraltype@*/ to be long.

4. Function Interfaces

Functions communicate with their calling environment through an interface. The caller communicates the values of actual parameters and global variables to the function, and the function communicates to the caller through the return value, global variables and storage reachable from the actual parameters. By keeping interfaces narrow (i.e., restricting the amount of information visible across a function interface), we can understand and implement functions independently.

A function prototype documents the interface to a function. It serves as a contract between the function and its caller. In early versions of C, the function "prototype" was very limited. It described the type returned by the function but nothing about its parameters. The main improvement provided by ANSI C was the ability to add information on the number and types of parameter to a function. LCLint provides the means to express much more about a function interface: what global variable the function may use, what values visible to the caller it may modify, if a pointer parameter may be a null pointer or point to undefined storage, if storage pointed to by a parameter is deallocated before the function returns, if the function may create new aliases to a parameter, can the caller modify or deallocate the return value, etc.

The extra interface information places constraints on both how the function may be called and how it may be implemented. LCLint reports places where these constrains are not satisfied. Typically, these indicate bugs in the code or errors in the interface documentation.

This section describes syntactic comments may be added to a function declaration to document what global variables the function implementation may use and what values visible to its caller it may modify. Sections 5-7 describe annotations may be added to parameters to constrain valid arguments to a function and how these arguments may be used after the call and to the return value to constrain results.

4.1 Modifications

The modifies clause lists what values visible to the caller may be modified by a function. Modifies clauses limit what values a function may modify, but they do not require that listed values are always modified. The declaration,
int f (int *p, int *q) /*@modifies *p@*/;
declares a function f that may modify the value pointed to by its first argument but may not modify the value of its second argument or any global state.

LCLint checks that a function does not modify any caller-visible value not encompassed by its modifies clause and does modify all values listed in its modifies clause on some possible execution of the function. Figure 4 shows an example of modifies checking done by LCLint.

4.1.1 Special Modifications

A few special names are provided for describing function modifications:

internalState

The function modifies some internal state (that is, the value of a static variable). Even though a client cannot access the internal state directly, it is important to know that something may be modified by the function call both for clear documentation and for checking undefined order of evaluation (Section 10.1) and side-effect free parameters (Section 8.2.1).

fileSystem
The function modifies the file system. Any modification that may change the system state is considered a file system modification. All functions that modify an object of type pointer to FILE also modify the file system. In addition, functions that do not modify a FILE pointer but modify some state that is visible outside this process also modify the file system (e.g., rename). The flag mod-file-system controls reporting of undocumented file system modifications.
nothing
The function modifies nothing (i.e., it is side-effect free).

The syntactic comment, /*@*/ in a function declaration or definition (after the parameter list, before the semi-colon or function body) denotes a function that modifies nothing and does not use any global variables (see Section 4.2).

Figure 4. Modifies checking.

4.1.2 Missing Modifies Clauses

LCLint is designed so programs with many functions that are declared without modifies clauses can be checked effectively. Unless modnomods is in on, no modification errors are reported checking a function declared with no modifies clause.

A function with no modifies clause is an unconstrained function since there are no documented constraints on what it may modify. When an unconstrained function is called, it is checked differently from a function declared with a modifies clause. To prevent spurious errors, no modification error is reported at the call site unless the moduncon flag is on. Flags control whether errors involving unconstrained functions are reported for other checks that depend on modifications (side-effect free macro parameters (Section 8.2.1), undefined evaluation order (Section 10.1), and likely infinite loops (Section 10.2.1).)

4.1.3 Limitations

Determining whether a function modifies a particular parameter or global is in general an undecidable[9] problem. To enable useful checking, certain simplifying assumptions are necessary. LCLint assumes an object is modified when it appears on the left hand side of an assignment or it is passed to a function as a parameter which may be modified by that function (according to the called function's modifies clause). Hence, LCLint will report spurious modification errors for assignments that do not change the value of an object or modifications that are always reversed before a procedure returns. The /*@-mods@*/ and /*@=mods@*/ control comments can be used around these modifications to suppress the message.

4.2 Global Variables

Another aspect of a function's interface, is the global variables it uses. A globals list in a function declaration lists external variables that may be used in the function body. LCLint checks that global variables used in a procedure match those listed in its globals list. A global is used in a function if it appears in the body directly, or it is in the globals list of a function called in the body. LCLint reports if a global that is used in a procedure is not listed in its globals list, and if a listed global is not used in the function implementation.

Figure 5 shows an example function definition with a globals list and associated checking done by LCLint.

4.2.1 Controlling Globals Checking

Whether on not an error is reported for a use of a global variable in a given function depends on the scope of the variable (file static or external), the checking annotation used in the variable declaration or the implicit annotation if no checking annotation is used, whether or not the function is declared with a globals list, and flag settings.

A global or file static variable declaration may be preceded by an annotation to indicate how the variable should be checked. In order of decreasing checks, the annotations are:

/*@checkedstrict@*/

Strictest checking. Undocumented uses and modifications of the variable are reported in all functions whether or not they have a globals list (unless checkstrictglobs is off).
/*@checked@*/
Undocumented use of the variable is reported in a function with a globals list, but not in a function declared with no globals (unless globnoglobs is on).

/*@checkmod@*/
Undocumented uses of the variable are not reported, but undocumented modifications are reported. (If modglobsnomods is on, errors are reported even in functions declared with no modifies clause or globals list.)
/*@unchecked@*/
No messages are reported for undocumented use or modification of this global variable. If a variable has none of these annotations, an implicit annotation is determined by the flag settings.

Different flags control the implicit annotation for variables declared with global scope and variables declared with file scope (i.e., using the static storage qualifier). To set the implicit annotation for global variables declared in context (globs for external variables or statics for file static variable) to be annotation (checked, checkmod, checkedstrict) use imp<annotation><context>. For example, +impcheckedstrictstatics makes the implicit checking on unqualified file static variables checkedstrict. (See Apppendix C, page 51, for a complete list of globals checking flags.)

4.3 Declaration Consistency

LCLint checks that function declarations and definitions are consistent. The general rule is that the first declaration of a function imply all later declarations and definitions. If a function is declared in a header file, the first declaration processed is its first declaration (if it is declared in more than one header file an error is reported if redecl is set). Otherwise, the first declaration in the file defining the function is its first declaration.

Later declarations may not include variables in the globals list that were not included in the first declaration. The exception to this is when the first declaration is in a header file and the later declaration or definition includes file static variables. Since these are not visible in the header file, they can not be included in the header file declaration. Similarly, the modifies clause of a later declaration may not include objects that are not modifiable in the first declaration. The later declaration may be more specific. For example, if the header declaration is:

extern void setName (employee e, char *s) /*@modifies e@*/;
the later declaration could be,
   void setName (employee e, char *) /*@modifies e->name@*/;
If employee is an abstract type, the declaration in the header should not refer to a particular implementation (i.e., it shouldn't rely on there being a name field), but the implementation declaration can be more specific.

This rule also applies to file static variables. The header declaration for a function that modifies a file static variable should use modifies internalState since file static variables are not visible to clients. The implementation declaration should list the actual file static variables that may be modified.

5. Memory Management

About half the bugs in typical C programs can be attributed to memory management problems. Memory management bugs are notoriously difficult to detect through traditional techniques. Often, the symptom of the bug is far removed from its actual source. Memory management bug often only appear sporadically and some bugs may only be apparent when compiler optimizations are turned on or the code is compiled on a different platform. Run-time tools offer some help, but are cumbersome to use and limited to detecting errors that occur when test cases are run. By detecting these errors statically, we can be confident that certain types of errors will never occur and provide verified documentation on the memory management behavior of a program.

LCLint can detect many memory management errors at compile time including:

Most of these checks rely heavily on annotations added to programs to document assumptions related to memory management and pointer values. By documenting these assumptions for function interfaces, variables, type definitions and structure fields, memory management bugs can be detected at their source -- where an assumption is violated. In addition, precise documentation about memory management decisions makes its easier to change code.

5.1 Storage Model[10]

Yea, from the table of my memory I'll wipe away all trivial fond records, all saws of books, all forms, all pressures past, that youth and observation copied there.
- Hamlet prefers garbage collection (Shakespeare, Hamlet. Act I, Scene v)
This section describes execution-time concepts for describing the state of storage more precisely than can be done using standard C terminology. Certain uses of storage are likely to indicate program bugs, and are reported as anomalies.

LCL assumes a CLU-like object storage model.[11] An object is a typed region of storage. Some objects use a fixed amount of storage that is allocated and deallocated automatically by the compiler.

Other objects use dynamic storage that must be managed by the program.

Storage is undefined if it has not been assigned a value, and defined after it has been assigned a value. An object is completely defined if all storage that may be reached from it is defined. What storage is reachable from an object depends on the type and value of the object. For example, if p is a pointer to a structure, p is completely defined if the value of p is NULL, or if every field of the structure p points to is completely defined.

When an expression is used as the left side of an assignment expression we say it is used as an lvalue. Its location in memory is used, but not its value. Undefined storage may be used as an lvalue since only its location is needed. When storage is used in any other way, such as on the right side of an assignment, as an operand to a primitive operator (including the indirection operator, *),[12] or as a

function parameter, we say it is used as an rvalue. It is an anomaly to use undefined storage as an rvalue.

A pointer is a typed memory address. A pointer is either live or dead. A live pointer is either NULL or an address within allocated storage. A pointer that points to an object is an object pointer. A pointer that points inside an object (e.g., to the third element of an allocated block) is an offset pointer. A pointer that points to allocated storage that is not defined is an allocated pointer. The result of dereferencing an allocated pointer is undefined storage. Hence, it is an anomaly to use it as an rvalue. A dead (or "dangling") pointer does not point to allocated storage. A pointer becomes dead if the storage it points to is deallocated (e.g., the pointer is passed to the free library function.) It is an anomaly to use a dead pointer as an rvalue.

There is a special object null corresponding to the NULL pointer in a C program. A pointer that may have the value NULL is a possibly-null pointer. It is an anomaly to use a possibly-null pointer where a non-null pointer is expected (e.g., certain function arguments or the indirection operator).

5.2 Deallocation Errors

There are two kinds of deallocation errors with which we are concerned: deallocating storage when there are other live references to the same storage, or failing to deallocate storage before the last reference to it is lost. To handle these deallocation errors, we introduce a concept of an obligation to release storage. Every time storage is allocated, it creates an obligation to release the storage. This obligation is attached to the reference to which the storage is assigned.[13] Before the scope of the reference is exited or it is assigned to a new value, the storage to which it points must be released. Annotations can be used to indicate that this obligation is transferred through a return value, function parameter or assignment to an external reference.

5.2.1 Unshared References

`Tis in my memory lock'd, and you yourself shall keep the key of it.
- Ophelia prefers explicit deallocation (Hamlet. Act I, Scene iii)
The only annotation is used to indicate a reference is the only pointer to the object it points to. We can view the reference as having an obligation to release this storage. This obligation is satisfied by transferring it to some other reference in one of three ways:

After the release obligation is transferred, the original reference is a dead pointer and the storage it points to may not be used.

All obligations to release storage stem from primitive allocation routines (e.g., malloc), and are ultimately satisfied by calls to free. The standard library declared the primitive allocation and deallocation routines.

The basic memory allocator, malloc, is declared:[14]

/*@only@*/ void *malloc (size_t size);
It returns an object that is referenced only by the function return value.

The deallocator, free, is declared:[15]

void free (/*@only@*/ void *ptr);

The parameter to free must reference an unshared object. Since the parameter is declared using only, the caller may not use the referenced object after the call, and may not pass in a reference to a shared object. There is nothing special about malloc and free -- their behavior can be described entirely in terms of the provided annotations.

Figure 6. Deallocation errors.

5.2.2 Temporary Parameters

The temp annotation is used to declare a function parameter that is used temporarily by the function. An error is reported if the function releases the storage associated with a temp formal parameter or creates new aliases it that are visible after the function returns. Any storage may be passed as a temp parameter, and it satisfies its original memory constraints after the function returns.

5.2.3 Owned and Dependent References

In real programs it is sometimes necessary to have storage that is shared between several possibly references. The owned and dependent annotations provide a more flexible way of managing storage, at the cost of less checking. The owned annotation denotes a reference with an obligation to release storage. Unlike only, however, other external references marked with dependent annotations may share this object. It is up to the programmer to ensure that the lifetime of a dependent reference is contained within the lifetime of the corresponding owned reference.

5.2.4 Kept Parameters

The keep annotation is similar to only, except the caller may use the reference after the call. The called function must assign the keep parameter to an only reference, or pass it as a keep parameter to another function. It is up to the programmer to make sure that the calling function does not use this reference after it is released. The keep annotation is useful for adding an object to a collection (e.g., a symbol table), where it is known that it will not be deallocated until the collection is.

5.2.5 Shared References

If LCLint is used to check a program designed to be used in a garbage-collected environment, there may be storage that is shared by one or more references and never explicitly released. The shared annotation declares storage that may be shared arbitrarily, but never released.

5.2.6 Stack References

Local variables that are not allocated dynamically are store on a call stack. When a function returns, its stack frame is deallocated, destroying the storage associated with the function's local variables. A memory error occurs if a pointer into this storage is live after the function returns. LCLint detects errors involving stack references exported from a function through return values or assignments to references reachable from global variables or actual parameters. No annotations are needed to detect stack reference errors, since it is clear from a declaration if storage is allocated on the function stack.

Figure 7. Stack references.

5.2.7 Inner Storage

An annotation always applies to the outermost level of storage. For example,
/*@only@*/ int **x;
declares x as an unshared pointer to a pointer to an int. The only annotation applies to x, but not to *x. To apply annotations to inner storage a type definition may be used:
  typedef /*@only@*/ int *oip;
  /*@only@*/ oip *x;
Now, x is an only pointer to an oip, which is an only pointer to an int.

When annotations are use in type definitions, they may be overridden in instance declarations. For example,

/*@dependent@*/ oip x;
makes x a dependent pointer to an int.

5.3 Implicit Memory Annotations

Since it is important that LCLint can check unannotated programs effectively, the meaning of declarations with no memory annotations is chosen to minimize the number of annotations needed to get useful checking on an unannotated program.

An implicit memory management annotation may be assumed for declarations with no explicit memory management annotation. Implicit annotations are checked identically to the corresponding explicit annotation, except error messages indicate that they result from an implicit annotation.

Unannotated function parameters are assumed to be temp. This means if memory checking is turned on for an unannotated program, all functions that release storage referenced by a parameter or assign a global variable to alias the storage will produce error messages. (Controlled by paramimptemp.)

Unannotated return values, structure fields and global variables are assumed to be only. With implicit annotations (on by default), turning on memory checking for an unannotated program will produce errors for any function that does not return unshared storage or assignment of shared storage to a global variable or structure field.[16] (Controlled by retimponly, structimponly and globimponly. The codeimponly flag sets all of the implicit only flags.)

Figure 8. Implicit annotations

5.4 Reference Counting

Another approach to memory management is to add a field to a type to explicitly keep track of the number of references to that storage. Every time a reference is added or lost the reference count is adjusted accordingly; if it would become zero, the storage is released. Reference counting it difficult to do without automatic checking since it is easy to forget to increment or decrement the reference count, and exceedingly difficult to track down these errors.

LCLint supports reference counting by using annotations to constrain the use of reference counted storage in a manner similar to other memory management annotations.

A reference counted type is declared using the refcounted annotation. Only pointer to struct types may be declared as reference counted, since reference counted storage must have a field to count the references. One field in the structure (or integral type) is preceded by the refs annotation to indicate that the value of this field is the number of live references to the structure.

For example (in rstring.h),

     typedef /*@abstract@*/ /*@refcounted@*/ struct {
	  /*@refs@*/ int refs;
        char *contents;
      } *rstring;
declares rstring as an abstract, reference-counted type. The refs field counts the number of references and the contents field holds the contents of a string.

All functions that return refcounted storage must increase the reference count before returning. LCLint cannot determine if the reference count was increased, so any function that directly returns a reference to refcounted storage will produce an error. This is avoided, by using a function to return a new reference (e.g., rstring_ref in Figure 9).

A reference counted type may be passed as a temp or dependent parameter. It may not be passed as an only parameter. Instead, the killref annotation is used to denote a parameter whose reference is eliminated by the function call. Like only parameters, an actual parameter corresponding to a killref formal parameter may not be used in the calling function after the call. LCLint checks that the implementation of a function releases all killref parameters, either by passing them as killref parameters, or assigning or returning them without increasing the reference count.

Figure 9. Reference counting.

6. Sharing

Errors involving unexpected sharing of storage can cause serious problems. Undocumented sharing may lead to unpredictable modifications, and some library calls (e.g., strcpy) have undefined behavior if parameters share storage. Another class of sharing errors occurs when clients of an abstract type may obtain a reference to mutable storage that is part of the abstract representation. This exposes the representation of the abstract type, since clients may modify an instance of the abstract type indirectly through this shared storage.

6.1 Aliasing

LCLint detects errors involving dangerous aliasing of parameters. Some of these errors are already detected through the standard memory annotations (e.g., only parameters may not be aliases.) Two additional annotations are provided for constraining aliasing of parameters and return values.

6.1.1 Unique Parameters

The unique annotation denotes a parameter that may not be aliased by any other storage reachable from the function implementation -- that is, any storage reachable through the other parameters or global variables used by the function. The unique annotation places similar constraints on function parameters as the only annotation, but it does not transfer the obligation to release storage.

LCLint will report an error if a unique parameter may be aliased by another parameter or global variable.

Figure 10. Unique parameters.

6.1.2 Returned Parameters

LCLint reports an error if a function returns a reference to storage reachable from one of its parameters (if retalias is on) since this may introduce unexpected aliases in the body of the calling function when the result is assigned.

The returned annotation denotes a parameter that may be aliased by the return value. LCLint checks the call assuming the result may be an alias to the returned parameter. Figure 11 shows an example use of a returned annotation.

6.2 Exposure

LCLint detects places where the representation of an abstract type is exposed. This occurs if a client has a pointer to storage that is part of the representation of an instance of the abstract type. The client can then modify or examine the storage this points to, and manipulate the value of the abstract type instance without using its operations.

There are three ways a representation may be exposed:

  1. Returning (or assigning to a global variable) an object that includes a pointer to a mutable component of an abstract type representation. (Controlled by retexpose).
  2. Assigning a mutable component of an abstract object to storage reachable from an actual parameter or a global variable that may be used after the call. This means the client may manipulate the abstract object using the actual parameter after the call. Note that if the corresponding formal parameter is declared only, the caller may not use the actual parameter after the call so the representation is not exposed. (Controlled by assignexpose).
  3. Casting mutable storage to or from an abstract type. (Controlled by castexpose).
Annotations may be used to allow exposed storage to be returned safely by restricting how the caller may use the returned storage.

6.2.1 Read-Only Storage

It is often useful for a function to return a pointer to internal storage (or a instance of a mutable abstract type) that is intended only as an observer. The caller may use the result, but should not modify the storage it points to. For example, consider an naïve implementation of the employee_getName operation for the abstract employee type:
   typedef /*@abstract@*/ struct {
      char *name;
      int id;
   } *employee;
   ...
   char *employee_getName (employee e) { return e->name; }
LCLint produces a message to indicate that the return value exposes the representation. One solution would be to return a fresh copy of e->name. This is expensive, though, especially if we expect employee_getName is used mainly just to get a string for searching or printing. Instead, we could change the declaration of employee_getName to:
extern /*@observer@*/ char *employee_getName (employee e);
Now, the original implementation is correct. The declaration indicates that the result may not be modified by the caller, so it is acceptable to return shared storage.[17] LCLint checks that the return value is not modified by the caller. An error is reported if observer storage is modified directly, passed as a function parameter that may be modified, assigned to a global variable or reference derivable from a global variable that is not declared with an observer annotation, or returned as a function result or a reference derivable from the function result that is not annotation with an observer annotation.

String Literals

A program that attempts to modify a string literal has undefined behavior [ANSI, Section 3.1.4]. This is not enforced by most C compilers, and can lead to particularly pernicious bugs that only appear when optimizations are turned on and the compiler attempts to minimize storage for string literals. LCLint can be used to check that string literals are not modified, by treating them as observer storage. If readonlystrings is on (default in standard mode), LCLint will report an error if a string literal is modified.

6.2.2 Exposed Storage

Sometimes it is necessary to expose the representation of an abstract type. This may be evidence of a design flaw, but in some cases is justified for efficiency reasons. The exposed annotation denotes storage that is exposed. It may be used on a return value for results that reference storage internal to an abstract representation, on a parameter value to indicate a parameter that may be assigned directly to part of an abstract representation,[18] or on a field of an abstract representation to indicate that external references to the storage may exist. An error is reported if exposed storage is released, but unlike an observer, no error is reported if it is modified.

Figure 12 shows examples of exposure problems detected by LCLint.

7. Value Constraints

LCLint can be used to constrain values of parameters, function results, global variables, and derived storage such as structure fields. These constraints are checked at interface points -- where a function is called or returns. Section 7.1 describes how to constrain parameters, return values and structures to detect use before definition errors. A similar approach is used for restricting the use of possibly null pointers in Section 7.2. To do both well, and avoid spurious errors, information about when and if a function returns if useful. Annotations for documenting execution control are described in Section 7.3.

7.1 Use Before Definition

Like many static checkers, LCLint detects instances where the value of a location is used before it is defined. This analysis is done at the procedural level. If there is a path through the procedure that

uses a local variable before it is defined, a use before definition error is reported. Use before definition checking is controlled by the usedef flag.

LCLint can do more checking than standard checkers though, because the annotations can be used to describe what storage must be defined and what storage may be undefined at interface points. Unannotated references are expected to be completely defined at interface points. This means all storage reachable from a global variable, parameter to a function, or function return value is defined before and after a function call.

7.1.1 Undefined Parameters

Sometimes, function parameters or return values are expected to reference undefined or partially defined storage. For example, a pointer parameter may be intended only as an address to store a result, or a memory allocator may return allocated but undefined storage. The out annotation denotes a pointer to storage that may be undefined.

LCLint does not report an error when a pointer to allocated but undefined storage is passed as an out parameter. Within the body of a function, LCLint will assume an out parameter is allocated but not necessarily bound to a value, so an error is reported if its value is used before it is defined.

LCLint reports an error if storage reachable by the caller after the call is not defined when the function returns. This can be suppressed by -mustdefine. When checking a call, an actual parameter corresponding to an out parameter is assumed to be completely defined after the call returns.

When checking unannotated programs, many spurious use before definition errors may be reported If impouts is on, no error is reported when an incompletely-defined parameter is passed to a formal parameter with no definition annotation, and the actual parameter is assumed to be defined after the call. The /*@in@*/ annotation can be used to denote a parameter that must be completely defined, even if impouts is on. If impouts is off, there is an implicit in annotation on every parameter with no definition annotation.

Figure 13. Use before definition.

7.1.2 Relaxing Checking

The reldef annotation relaxes definition checking for a particular declaration. Storage declared with a reldef annotation is assumed to be defined when it is used, but no error is reported if it is not defined before it is returned or passed as a parameter.

It is up to the programmer to check reldef fields are used correctly. They should be avoided in most cases, but may be useful for fields of structures that may or may not be defined depending on other constraints.

7.1.3 Partially Defined Structures

The partial annotated can be used to relax checking of structure fields. A structure with undefined fields may be passed as a partial parameter or returned as a partial result. Inside a function body, no error is reported when the field of a partial structure is used. After a call, all fields of a structure that is passed as a partial parameter are assumed to be completely defined.

7.1.4 Global Variables

Special annotations can be used in the globals list of a function declaration (Section 4.2) to describe the states of global variables before and after the call.

If a global is preceded by undef, it is assumed to be undefined before the call. Thus, no error is reported if the global is not defined when the function is called, but an error is reported if the global is used in the function body before it is defined.

The killed annotation denotes a global variable that may be undefined when the call returns. For globals that contain dynamically allocated storage, a killed global variable is similar to an only parameter (Section 5.2). An error is reported if it contains the only reference to storage that is not released before the call returns.

Figure 14. Annotated globals lists.

7.2 Null Pointers

A common cause of program failures is when a null pointer is dereferenced. LCLint detects these errors by distinguishing possibly NULL pointers at interface boundaries.

The null annotation is used to indicate that a pointer value may be NULL. A pointer declared with no null annotation, may not be NULL. If null checking is turned on (controlled by null), LCLint will report an error when a possibly null pointer is passed as a parameter, returned as a result, or assigned to an external reference with no null qualifier.

If a pointer is declared with the null annotation, the code must check that it is not NULL on all paths leading to the a dereference of the pointer (or the pointer being returned or passed as a value with no null annotation). Dereferences of possibly null pointers may be protected by conditional statements or assertions (to see how assert is declared see Section 7.3) that check the pointer is not NULL.

Consider two implementations of firstChar in Figure 15. For firstChar1, LCLint reports an error since the pointer that is dereferenced is declared with a null annotation. For firstChar2, no error is reported since the true branch of the s == NULL if statement returns, so the dereference of s is only reached if s is not NULL.

7.2.1 Predicate Functions

Another way to protect null dereference, is to declare a function using falsenull or truenull and call the function in a conditional statement before the null-annotated pointer is dereferenced. The falsenull and truenull annotations may only be used on return values for functions that return a boolean[19] result and whose first argument is a possibly null pointer.

A function is annotated with truenull is assumed to return TRUE if its first parameter is NULL and FALSE otherwise. For example, if isNull is declared as,

   /*@truenull@*/ bool isNull (/*@null@*/ char *x);
we could write firstChar2:
   char firstChar2 (/*@null@*/ char *s)
   {
      if (isNull (s)) return '\0';
      return *s;
   }
No error is reported since the dereference of s is only reached if isNull(s) is false, and since isNull is declared with the truenull annotation this means s must not be null.

The falsenull annotation is not quite the opposite of truenull. If a function declared with falsenull returns TRUE, it means its parameter is not NULL. If it returns FALSE, the parameter may or may not be NULL.

For example, we could define isNonEmpty to return TRUE if its parameter is not NULL and has least one character before the NUL terminator:

   /*@falsenull@*/ bool isNonEmpty (/*@null@*/ char *x)
   {
     return (x != NULL && *x != `\0');
   }
LCLint does not check that the implementation of a function declared with falsenull or truenull is consistent with its annotation, but assumes the annotation is correct when code that calls the function is checked.

7.2.2 Overriding Null Types

The null annotation may be used in a type definition to indicate that all instances of the type may be NULL. For declarations of a type declared using null, the null annotation in the type definition may be overridden with notnull. This is particularly useful for parameters to hidden static operations of abstract types where the null test has already been done before the function is called, or function results of the type which are never NULL. For an abstract type, notnull may not be used for parameters to external functions, since clients should not be aware of when the concrete representation may by NULL. Parameters to static functions in the implementation module, however, may be declared using notnull, since they may only be called from places where the representation is accessible. Return values for static or external functions may be declared using notnull.

Figure 16. Using notnull.

7.2.3 Relaxing Null Checking

An additional annotation, relnull may be used to relax null checking (relnull is analogous to reldef for definition checking). No error is reported when a relnull value is dereferenced, or when a possibly null value is assigned to an identifier declared using relnull.

This is generally used for structure fields that may or may not be null depending on some other constraint. LCLint does not report and error when NULL is assigned to a relnull reference, or when a relnull reference is dereferenced. It is up to the programmer to ensure that this constraint is satisfied before the pointer is dereferenced.

7.3 Execution

To detect certain errors and avoid spurious errors, it is important to know something about the control flow behavior of called functions. Without additional information, LCLint assumes that all functions eventually return and execution continues normally at the call site.

The exits annotation is used to denote a function that never returns. For example,

extern /*@exits@*/ void fatalerror (/*@observer@*/ char *s);
declares fatalerror to never return. This allows LCLint to correctly analyze code like,
   if (x == NULL) fatalerror ("Yikes!");
   *x = 3;
Other functions may exit, but sometimes (or usually) return normally. The mayexit annotation denotes a function that may or may not return. This doesn't help checking much, since LCLint must assume that a function declared with mayexit returns normally when checking the code.

To be more precise, the trueexit and falseexit annotations may be used Similar to truenull and falsenull (see Section 7.2.1), trueexit and falseexit mean that a function always exits if the value of its first argument is TRUE or FALSE respectively. They may be used only on functions whose first argument has a boolean type.

A function declared with trueexit must exit if the value of its argument is TRUE, and a function declared with falseexit must exit if the value of its argument is FALSE. For example, the standard library declares assert as[20]:

/*@falseexit@*/ void assert (/*@sef@*/ bool /*@alt int@*/ pred);
This way, code like,
   assert (x != NULL);

*x = 3;

is checked correctly, since the falseexit annotation on assert means the deference of x is not reached is x != NULL is false.

7.4 Special Clauses

Sometimes it is necessary to specify function interfaces at a lower level than is possible with the standard annotations. For example, if a function defines some fields of a returned structure but does not define all the fields. The /*@special@*/ annotation is used to mark a parameter, global variable, or return value that is described using special clauses. The usual implicit definition rules do not apply to a special declaration.

Special clauses may be used to constrain the state of a parameter or return value before or after a call. One or more special clauses may appear in a function declaration, before the modifies or globals clauses. Special clauses may be listed in any order, but the same special clause should not be used more than once. Parameters used in special clauses must be annotated with /*@special@*/ in the function header. In a special clause list, result is used to refer to the return value of the function. If result appears in a special clause, the function return value must be annotated with /*@special@*/.

The following special clauses are used to describe the definition state or parameters before and after the function is called and the return value after the function returns:

/*@uses references@*/

References in the uses clause must be completely defined before the function is called. They are assumed to be defined at function entrance when the function is checked.
/*@sets references@*/
References in the sets clause must be allocated before the function is called. They are completely defined after the function returns. When the function is checked, they are assumed to be allocated at function entrance and an error is reported if there is a path on which they are not defined before the function returns.
/*@defines references@*/
References in the defines clause must not refer to unshared, allocated storage before the function is called. They are completely defined after the function returns. When the function is checked, they are assumed to be undefined at function entrance and an error is reported if there is a path on which they are not defined before the function returns.
/*@allocates references@*/
References in the allocates clause must not refer to unshared, allocated storage before the function is called. They are allocated but not necessarily defined after the function returns. When the function is checked, they are assumed to be undefined at function entrance and an error is reported if there is a path on which they are not allocated before the function returns.
/*@releases references@*/
References in the releases clause are deallocated by the function. They must correspond to storage which could be passed as an only parameter before the function is called, and are dead pointers after the function returns. When the function is checked, they are assumed to be allocated at function entrance and an error is reported if they refer to live, allocated storage at any return point.

Additional generic special clauses can be used to describe other aspects of the state of inner storage before or after a call. Generic special clauses have the form state:constraint. The state is either pre (before the function is called), or post (after the function is called). The constraint is similar to an annotation. The following constraints are supported:

Aliasing Annotations

pre:only, post:only
pre:shared, post:shared
pre:owned, post:owned
pre:dependent, post:dependent
References refer to only, shared, owned or dependent storage before (pre) or after (post) the call.

Exposure Annotations

pre:observer, post:observer
pre:exposed, post:exposed
References refer to observer or exposed storage before (pre) or after (post) the call.

Null State Annotations

pre:isnull, post:isnull
References have the value NULL before (pre) or after (post) the call. Note, this is not the same name or meaning as the null annotation (which means the value may be NULL.)
pre:notnull, post:notnull
References do not have the value NULL before (pre) or after (post) the call.
Some examples of special clauses are shown in Figure 17. The defines clause for record_new indicates that the id field of the structure pointed to by the result is defined, but the name field is not. So, record_create needs to call record_setName to define the name field. Similarly, the releases clause for record_clearName indicates that no storage is associated with the name field of its parameter after the return, so no failure to deallocate storage message is produced for the call to free in record_free.

8. Macros

Macros are commonly used in C programs to implement constants or to mimic functions without the overhead of a function call. Macros that are used to implement functions are a persistent source of bugs in C programs, since they may not behave like the intended function when they are invoked with certain parameters or used in certain syntactic contexts.

LCLint eliminates most of the potential problems by detecting macros with dangerous implementations and dangerous macro invocations. Whether or not a macro definition is checked or expanded normally depends on flag settings and control comments (see Section 8.3). Stylized macros can also be used to define control structures for iterating through many values (see Section 8.4).

8.1 Constant Macros

Macros may be used to implement constants. To get type-checking for constant macros, use the constant syntactic comment:
/*@constant null char *mstring_undefined@*/
Declared constants are not expanded and are checked according to the declaration. A constant with a null annotation may be used as only storage.

8.2 Function-like Macros

Using macros to imitate functions is notoriously dangerous. Consider this broken macro for squaring a number:

# define square(x) x * x
This works fine for a simple invocation like square(i). It behaves unexpectedly, though, if it is invoked with a parameter that has a side effect.

For example, square(i++) expands to i++ * i++. Not only does this give the incorrect result, it has undefined behavior since the order in which the operands are evaluated is not defined. (See Section 10.1 for more information on how expressions exhibiting undefined evaluation order behavior are detected by LCLint.) To correct the problem we either need to rewrite the macro so that its parameter is evaluated exactly once, or prevent clients from invoking the macro with a parameter that has a side-effect.

Another possible problem with macros is that they may produce unexpected results because of operator precedence rules. The invocation, square(i+1) expands to i+1*i+1, which evaluates to i+i+1 instead of the square of i+1. To ensure the expected behavior, the macro parameter should be enclosed in parentheses where it is used in the macro body.

Macros may also behave unexpectedly if they are not syntactically equivalent to an expression. Consider the macro definition,

# define incCounts()  ntotal++; ncurrent++;
This works fine, unless it is used as a statement. For example,
if (x < 3) incCounts();
increments ntotal if x < 3 but always increments ncurrent.

One solution is to use the comma operator to define the macro:

# define incCounts()  (ntotal++, ncurrent++)
More complicated macros can be written using a do … while construction:

  # define incCounts() \
     do { ntotal++; ncurrent++; } while (FALSE)
LCLint detects these pitfalls in macro definitions, and checks that a macro behaves as much like a function as possible. A client should only be able to tell that a function was implemented by a macro if it attempts to use the macro as a pointer to a function.

These checks are done by LCLint on a macro definition corresponding to a function:

At the call site, a macro is checked like any other function call.

8.2.1 Side-Effect Free Parameters

Suppose we really do want to implement square as a macro, but want do so in a safe way. One way to do this is to require that it is never invoked with a parameter that has a side-effect. LCLint will check that this constraint holds, if the parameter is annotated to be side-effect free. That is, the expression corresponding to this parameter must not modify any state, so it does not matter how many times it is evaluated. The sef annotation is used to denote a parameter that may not have any side-effects:
   extern int square (/*@sef@*/ int x);
   # define square(x) ((x) *(x))
Now, LCLint will not report an error checking the definition of square even though x is used more than once.

A message will be reported, however, if square is invoked with a parameter that has a side-effect.

For the code fragment,

square (i++)
LCLint produces the message:
   Parameter 1 to square is declared sef, but the argument may modify i: i++
It is also an error to pass a non-sef macro parameter as a sef macro parameter in the body of a macro definition. For example,
   extern int sumsquares (int x, int y);
   # define sumsquares(x,y) (square(x) + square(y))
Although x only appears once in the definition of sumsquares it will be evaluated twice since square is expanded. LCLint reports an error when a non-sef macro parameter is passed as a sef parameter.

A parameter may be passed as a sef parameter without an error being reported, if LCLint can determine that evaluating the parameter has no side-effects. For function calls, the modifies clause is used to determine if a side-effect is possible.[22] To prevent many spurious errors, if the called function has no modifies clause, LCLint will report an error only if sefuncon is on. Justifiably paranoid programmers will insist on setting sefuncon on, and will add modifies clauses to unconstrained functions that are used in sef macro arguments.

8.2.2 Polymorphism

One problem with our new definition of square is that while the original macro would work for parameters of any numeric type, LCLint will now report an error is the new version is used with a non-integer parameter.

We can use the /*@alt type;,+@> syntax to indicate that an alternate type may be used. For example,

  extern int /*@alt float@*/ square (/*@sef@*/ int /*@alt float@*/ x);
  # define square(x) ((x) *(x))
declares square for both ints and floats.

Alternate types are also useful for declaring functions for which the return value may be safely ignored (see Section 10.3.2).

8.3 Controlling Macro Checking

By default, LCLint expands macros normally and checks the resulting code after macros have been expanded. Flags and control comments may be used to control which macros are expanded and which are checked as functions or constants.

If the fcnmacros flag is on, LCLint assumes all macros defined with parameter lists implement functions and checks them accordingly. Parameterized macros are not expanded and are checked as functions with unknown result and parameter types (or using the types in the prototype, if one is given). The analogous flag for macros that define constants is constmacros. If it is on, macros with no parameter lists are assumed to be constants, and checked accordingly. The allmacros flag sets both fcnmacros and constmacros. If the macrofcndecl flag is set, a message reports parameterized macros with no corresponding function prototype. If the macroconstdecl flag is set, a similar message reports macros with no parameters with no corresponding constant declaration.

The macro checks described in the previous sections make sense only for macros that are intended to replace functions or constants. When fcnmacros or constmacros is on, more general macros need to be marked so they will not be checked as functions or constants, and will be expanded normally. Macros which are not meant to behave like functions should be preceded by the /*@notfunction@*/ comment. For example,

   /*@notfunction@*/
   # define forever for(;;)
Macros preceded by notfunction are expanded normally before regular checking is done. If a macro that is not syntactically equivalent to a statement without a semi-colon (e.g., a macro which enters a new scope) is not preceded by notfunction, parse errors may result when fcnmacros or constmacros is on.

8.4 Iterators

It is often useful to be able to execute the same code for many different values. For example, we may want to sum all elements in an intSet that represents a set of integers. If intSet is an abstract type, there is no easy way of doing this in a client module without depending on the concrete representation of the type. Instead, we could provide such a mechanism as part of the type's implementation. We call a mechanism for looping through many values an iterator.

The C language provides no mechanism for creating user-defined iterators. LCLint supports a stylized form of iterators declared using syntactic comments and defined using macros.

Iterator declarations are similar to function declarations except instead of returning a value, they assign values to their yield parameters in each iteration. For example, we could add this iterator declaration to intSet.h:

/*@iter intSet_elements (intSet s, yield int el);@*/
The yield annotation means that the variable passed as the second actual argument is declared as a local variable of type int and assigned a value in each loop iteration.

Defining Iterators

An iterator is defined using a macro. Here's one (not particularly efficient) way of defining intSet_elements:
   typedef /*@abstract@*/ struct {
      int nelements;
      int *elements;
   } intSet;
   ...
   # define intSet_elements(s,m_el) \
     { int m_i; \ 
       for (m_i = (0); m_i <= ((s)->nelements); m_i++) { \
           int m_el = (s)->elements[(m_i)];

   # define end_intSet_elements }}
Each time through the loop, the yield parameter m_el is assigned to the next value. After all values have been assigned to m_el for one iteration, the loop terminates. Variables declared by the iterator macro (including the yield parameter) are preceded by the macro variable namespace prefix m_ (see Section 8.2) to avoid conflicts with variables defined in the scope where the iterator is used.

Using Iterators

The general structure for using an iterator is,

iter (<params>) stmt; end_iter

For example, a client could use intSet_elements to sum the elements of an intSet:

   intSet s;
   int sum = 0;
   ...
   intSet_elements (s, el) { 
      sum += el; 
   } end_intSet_elements;
The actual parameter corresponding to a yield parameter, el, is not declared in the function scope. Instead, it is declared by the iterator and assigned to an appropriate value for each iteration.

LCLint will do the following checks for uses of stylized iterators:

Iterators are a bit awkward to implement, but they enable compact, easily understood client code. For abstract collection types, an iterator can be used to enable clients to operate on elements of the collection without breaking data abstraction.

9. Naming Conventions

Naming conventions tend to be a religious issue. Generally, it doesn't matter too much what naming convention is followed as long as one is chosen and followed religiously. There are two kinds of naming conventions supported by LCLint. Type-based naming conventions (Section 9.1) constrain identifier names according to the abstract types that are accessible where the identifier is defined. Prefix naming conventions (Section 9.2) constrain the initial characters of identifier names according to what is being declared and its scope. Naming conventions may be combined or different conventions may be selected for different kinds of identifiers. In addition, LCLint supports checking that names do not conflict with names reserved for the standard library or implementation (Section 9.3) and that names are sufficiently distinguishable from other names.

9.1 Type-Based Naming Conventions

Generic naming conventions constrain valid names of identifiers. By limiting valid names, namespaces may be preserved and programs may be more easily understood since the name gives clues as to how and where the name is defined and how it should be used.

Names may be constrained by the scope of the name (external, file static, internal), the file in which the identifier is defined, the type of the identifier, and global constraints.

9.1.1 Czech Names

Of course, this is a complete jumble to the uninitiated, and that's the joke.
- Charles Simonyi, on the Hungarian naming convention

Czech[23] names denote operations and variables of abstract types by preceding the names by <type>_. The remainder of the name should begin with a lowercase character, but may use any other character besides the underscore. Types may be named using any non-underscore characters.

The Czech naming convention is selected by the czech flag. If accessczech is on, a function, variable, constant or iterator named <type>_<name> has access to the abstract type <type>.

Reporting of violations of the Czech naming convention is controlled by different flags depending on what is being declared:

czechfcns

Functions and iterators. An error is reported for a function name of the form <prefix>_<name> where <prefix> is not the name of an accessible type. Note that if accessczech is on, a type named <prefix> would be accessible in a function beginning with <prefix>_. If accessczech is off, an error is reported instead. An error is reported for a function name that does not have an underscore if any abstract types are accessible where the function is defined.
czechvars
czechconstants
czechmacros
Variables, constants and expanded macros. An error is reported if the identifier name starts with <prefix>_ and prefix is not the name of an accessible abstract type, or if an abstract type is accessible and the identifier name does not begin with <type>_ where type is the name of an accessible abstract type. If accessczech is on, the representation of the type is visible in the constant or variable definition.
czechtypes
User-defined types. An error is reported if a type name includes an underscore character.

9.1.2 Slovak Names

Slovak names are similar to Czech names, except they are spelled differently. A Slovak name is of the form <type><Name>. The type prefix may not use uppercase characters. The remainder of the name starts with the first uppercase character.

The slovak flag selects the Slovak naming convention. Like Czech names, it may be used with accessslovak to control access to abstract representations. The slovakfcns, slovakvars, slovakconstants, and slovakmacros flags are analogous to the similar Czech flags. If slovaktype is on, an error is reported if a type name includes an uppercase letter.

9.1.3 Czechoslovak Names

Czechoslovak names are a combination of Czech names and Slovak names. Operations may be named either <type>_ followed by any sequence of non-underscore characters, or <type> followed by an uppercase letter and any sequence of characters. Czechoslovak names have been out of favor since 1993, but may be necessary for checking legacy code. The czechoslovakfcns, czechoslovakvars, czechoslovakmacros, and czechoslovakconstants flags are analogous to the similar Czech flags. If czechoslovaktype is on, an error is reported if a type name contains either an uppercase letter or an underscore character.

9.2 Namespace Prefixes

Another way to restrict names is to constrain the leading character sequences of various kinds of identifiers. For example, a the names of all user-defined types might begin with "T" followed by an uppercase letter and all file static names begin with an uppercase letter. This may be useful for enforcing a namespace (e.g., all names exported by the X-windows library should begin with "X") or just making programs easier to understand by establishing an enforced convention. LCLint can be used to constrain identifiers in this way to detect identifiers inconsistent with prefixes.

All namespace flags are of the form, -<context>prefix <string>. For example, the macro variable namespace restricting identifiers declared in macro bodies to be preceded by "m_" would be selected by -macrovarprefix "m_". The string may contain regular characters that may appear in a C identifier. These must match the initial characters of the identifier name. In addition, special characters (shown in Table 1) can be used to denoted a class of characters.[24] The * character may be used at the end of a prefix string to specify the rest of the identifier is zero or more characters matching the character immediately before the *. For example, the prefix string "T&*" matches "T" or "TWINDOW" but not "Twin".

^    Any uppercase letter, A-Z                                                 
&    Any lowercase letter, a-z                                                 
%    Any character that is not an uppercase letter (allows lowercase           
     letters, digits and underscore)                                           
~    Any character that is not a lowercase letter (allows uppercase letters,   
     digits and underscore)                                                    
$    Any letter (a-z, A-Z)                                                     
/    Any letter or digit (A-Z, a-z, 0-9)                                       
?    Any character valid in a C identifier                                     
#    Any digit, 0-9                                                            

Table 1. Prefix character codes.

Different prefixes can be selected for the following identifier contexts:

macrovarprefix

Any variable declared inside a macro body
uncheckedmacroprefix
Any macro that is not checked as a function or constant (see Section 8.4)
tagprefix
Tags for struct, union and enum declarations
enumprefix
Members of enum types
typeprefix
Name of a user-defined type
filestaticprefix
Any identifier with file static scope
globvarprefix
Any variable (not of function type) with global variables scope
externalprefix
Any exported identifier

If an identifier is in more than one of the namespace contexts, the most specific defined namespace prefix is used (e.g., a global variable is also an exported identifier, so if globalvarprefix is set, it is checked against the variable name; if not, the identifier is checked against the externalprefix.)

For each prefix flag, a corresponding flag named <prefixname>exclude controls whether errors are reported if identifiers in a different namespace match the namespace prefix. For example, if macrovarprefixexclude is on, LCLint checks that no identifier that is not a variable declared inside a macro body uses the macro variable prefix.

Here is a (somewhat draconian) sample naming convention:

-uncheckedmacroprefix "~*"
unchecked macros have no lowercase letters
-typeprefix "T^&*"
all type typenames begin with T followed by an uppercase letter. The rest of the name is all lowercase letters.
+typeprefixexclude
no identifier that does not name a user-defined type may begin with the type name prefix (set above)
-filestaticprefix"^&&&"
file static scope variables begin with an uppercase letter and three lowercase letters
-globvarprefix "G"
all global variables variables start with G
+globvarprefixexclude
no identifier that is not a global variable starts with G

9.3 Naming Restrictions

Additional naming restrictions can be used to check that names do no conflict with names reserved for the standard library, and that identifier are sufficiently distinct (either for the compiler and linker, or for the programmer.) Restrictions may be different for names that are needed by the linker (external names) and names that are only needed during compilations (internal names). Names of non-static functions and global variables are external; all other names are internal.

9.3.1 Reserved Names

Many names are reserved for the implementation and standard library. A complete list of reserved names can be found in [vdL, p. 126-128] or [ANSI, Section 4]. Some name prefixes such as str followed by a lowercase character are reserved for future library extensions. Most C compilers do not detect naming conflicts, and they can lead to unpredictable program behavior. If ansireserved is on, LCLint reports errors for external names that conflict with reserved names. If ansireservedinternal is on, errors are also reported for internal names.

9.3.2 Distinct Identifiers

The decision to retain the old six-character case-insensitive restriction on significance was most painful.
- ANSI C Rationale

LCLint can check that identifiers differ within a given number of characters, optionally ignoring alphabetic case and differences between characters that look similar. The number of significant characters may be different for external and internal names.

Using +distinctexternalnames sets the number of significant characters for external names to six and makes alphabetical case insignificant for external names. This is the minimum significance acceptable in an ANSI-conforming compiler. Most modern compilers exceed these minimums (which are particularly hard to follow if one uses the Czech or Slovak naming convention). The number of significant characters can be changed using the externalnamelength <number> flag. If externalnamecaseinsensitive is on, alphabetical case is ignored in comparing external names. LCLint reports identifiers that differ only in alphabetic case.

For internal identifiers, a conforming compiler must recognize at least 31 characters and treat alphabetical cases distinctly. Nevertheless, it may still be useful to check that internal names are more distinct then required by the compiler to minimize the likelihood that identifiers are confused in the program. Analogously to external names, the internalnamelength <number> flag sets the number of significant characters in an internal name and internalnamecaseinsensitive sets the case sensitivity. The internalnamelookalike flag further restricts distinctions between identifiers. When set, similar-looking characters match -- the lowercase letter "l" matches the uppercase letter "I" and the number "1"; the letter "O" or "o" matches the number "0"; "5" matches "S"; and "2" matches "Z". Identifiers that are not distinct except for look-alike characters will produce an error message. External names are also internal names, so they must satisfy both the external and internal distinct identifier checks.

Figure 18 illustrates some of the name checking done by LCLint.

10. Other Checks

The section describes other errors detected by LCLint that are not directly related to extra information provided in annotations. Many of the checks are significantly improved, however, because of the extra information that is known about the program.

10.1 Undefined Evaluation Order

The order in which side effects take place in a C program is not entirely defined by the code. Certain execution points are known as sequence points -- a function call (after the arguments have been evaluated), the end of a full expression (an initializer, expression in an expression statement, the control expression of an if, switch, while or do statement, each expression of a for statement, and the expression in a return statement), and after the first operand or a &&, ||, ? or , operand.

All side effects before a sequence point must be complete before the sequence point, and no evaluations after the sequence point shall have taken place [ANSI, Section 2.1.2.3]. Between sequence points, side effects and evaluations may take place in any order. Hence, the order in which expressions or arguments are evaluated is not specified. Compilers are free to evaluate function arguments and parts of expressions (that do not contain sequence points) in any order. The behavior of code that uses a value that is modified by another expression that is not required to be evaluated before or after the other use is undefined.

LCLint detects instances where undetermined order of evaluation produces undefined behavior. If modifies clauses and globals lists are used, this checking is enabled in expressions involving function calls. Evaluation order checking is controlled by the evalorder flag.

When checking systems without modifies and globals information, evaluation order checking may report errors when unconstrained functions are called in procedure arguments. Since LCLint has no annotations to constrain what these functions may modify, it cannot be guaranteed that the evaluation order is defined if another argument calls an unconstrained function or uses a global variable or storage reachable from a parameter to the unconstrained function. Its best to add modifies and globals clauses to constrain the unconstrained functions in ways that eliminate the possibility of undefined behavior. For large legacy systems, this may require too much effort. Instead, the -evalorderuncon flag may be used to prevent reporting of undefined behavior due to the order of evaluation of unconstrained functions.

Figure 19. Evaluation order

10.2 Problematic Control Structures

A number of control structures that are syntactically legal may indicate likely bugs in programs. LCLint can detect errors involving likely infinite loops (Section 10.2.1), fall through cases and missing cases in switch statements (Section 10.2.2), break statements within deeply nested loops or switches (Section 10.2.3), clauses of if, while or for statements that are empty statements or unblocked single statements (Section 10.2.4) and incomplete if-else logic (Section 10.2.5). Although any of these may appear in a correct program, depending on the programming style used they may indicate likely bugs or style violations that should be detected and eliminated.

10.2.1 Likely Infinite Loops

LCLint reports an error if it detects a loop that appears to be infinite. An error is reported for a loop which does not modify any value used in its condition test inside the body of the loop or in the condition test itself. This checking is enhanced by modifies clauses and globals lists since they provide more information about what global variable may be used in the condition test and what values may be modified by function calls in the loop body.

Figure 20 shows examples of infinite loops detected by LCLint. An error is reported for the loop in line 14, since neither of the values used in the loop condition (x directly and glob1 through the call to f) is modified by the body of the loop. If the declaration of g is changed to include glob1 in the modifies clause no error is reported. (In this example, if we assume the annotations are correct, then the programmer has probably called the wrong function in the loop body. This isn't surprising, given the horrible choices of function and variable names!)

If an unconstrained function is called within the loop body, LCLint will assume that it modifies a value used in the condition test and not report an infinite loop error, unless infloopsuncon is on. If infloopsuncon is on, LCLint will report infinite loop errors for loops where there is no explicit modification of a value used in the condition test, but where they may be an undetected modification through a call to an unconstrained function (e.g., line 15 in Figure 20).

10.2.2 Switches

The automatic fall-through of C switch statements is almost never the intended behavior.[25] LCLint detects case statements with code that may fall through to the next case. The casebreak flag controls reporting of fall through cases. A single fall through case may be marked by preceding the case keyword with /*@fallthrough@*/ to indicate explicitly that execution falls through to this case.

For switches on enum types, LCLint reports an error if a member of the enumerator does not appear as a case in the switch body (and there is no default case). (Controlled by misscase.)

An example of switch checking is shown in Figure 21.

10.2.3 Deep Breaks

There is no syntax provided by C (other than goto) for breaking out of a nested loop. All break and continue statements act only on the innermost surrounding loop or switch. This often leads to serious problems[26] when a programmer intends to break the outer loop or switch instead. LCLint optionally reports errors for break and continue statements in nested contexts.

Since continue only makes sense within loops, errors are only reported for continue statements within nested loops. (Controlled by looploopcontinue.) A safe inner continue may be precede by /*@innercontinue@*/ to suppress error messages locally. The deepbreak flag sets all nested break and continue checking flags.

LCLint reports an error if the marker preceding a break is not consistent with its effect. An error is reported if innerbreak precedes a break that is not breaking an inner loop, switchbreak precedes a break that is not breaking a switch, or loopbreak precedes a break that is not breaking a loop.

10.2.4 Loop and If Bodies

An empty statement after an if, while or for often indicates a potential bug. A single statement (i.e., not a compound block) after an if, while or for is not likely to indicate a bug, but make the code harder to read and edit. LCLint can report errors for if or loop statements with empty bodies or bodies that are not compound statements. Separate flags control checking for statements following an if, while or for:

The if statement checks also apply to the body of the else clause. An ifblock error is not reported if the body of the else clause is an if statement, to allow else if chains.

10.2.5 Complete if-else Logic

Although it may be perfectly reasonable in many contexts, an if-else chain with no final else may indicate missing logic or forgetting to check error cases. If elseifcomplete is on, LCLint reports errors when an if statement that is the body of an else clause does not have a matching else clause. For example, the code,
   if (x == 0) { return "nil"; }
   else if (x == 1) { return "many"; }
produces an error message since the second if has no matching else branch.

10.3 Suspicious Statements

LCLint detects errors involving statements with no apparent effects (Section 10.3.1) and statements that ignore the result of a called function (Section 10.3.2).

10.3.1 Statements with No Effects

LCLint can report errors for statements that have no effect. (Controlled by noeffect.) Because of modifies clauses, LCLint can detect more errors than traditional checkers. Unless the noeffectuncon flag is on, errors are not reported for statements that involve calls to unconstrained functions since the unconstrained function may cause a modification.

Figure 22. Statements with no effect.

10.3.2 Ignored Return Values

LCLint reports an error when a return value is ignored. Checking may be controlled based on the type of the return value: retvalint controls reporting of ignored return values of type int, and retvalbool for return values of type bool, and retvalothers for all other types. A function statement may be cast to void to prevent this error from being reported.

Alternate types (Section 8.2.2) can be used to declare functions that return values that may safely be ignored by declaring the result type to alternately by void. Several functions in the standard library are specified to alternately return void to prevent ignored return value errors for standard library functions (e.g., strcpy) where the result may be safely ignored (see Apppendix F).

Figure 23 shows example of ignored return value errors reported by LCLint.

10.4 Unused Declarations

LCLint detects constants, functions, parameters, variables, types, enumerator members, and structure or union fields that are declared but never used. The flags constuse, fcnuse, paramuse, varuse, typeuse, enummemuse and fielduse control whether unused declaration errors are reported for each kind of declaration. Errors for exported declarations are reported only if topuse is on (see Section 10.5).

The /*@unused@*/ annotation can be used before a declaration to indicate that the item declared need not be used. Unused declaration errors are not reported for identifiers declared with unused.

10.5 Complete Programs

LCLint can be used on both complete and partial programs. When checking complete programs, additional checks can be done to ensure that every identifier declared by the program is defined and used, and that functions that do not need to be exported are declared static.

LCLint checks that all declared variables and functions are defined (controlled by compdef). Declarations of functions and variables that are defined in an external library, may be preceded by /*@external@*/ to suppress undefined declaration errors.

LCLint reports external declarations which are unused (Controlled by topuse). Which declarations are reported also depends on the declaration use flags (see Section 10.4).

The partial flag sets flags for checking a partial system. Top-level unused declarations, undefined declarations, and unnecessary external names are not reported if partial is set.

10.5.1 Unnecessary External Names

LCLint can report variables and functions that are declared with global scope (i.e., without using static), that are not used outside the file in which they are defined. In a stand-alone system, these identifiers should usually be declared using static to limit their scope. If the exportstatic flag is on, LCLint will report declarations that could have file scope. It should only be used when all relevant source files are listed on the LCLint command line; otherwise, variables and functions may be incorrectly identified as only used in the file scope since LCLint did not process the other file in which they are used.

10.5.2 Declarations Missing from Headers

A common practice in C programming styles, is that every function or variable exported by M.c is declared in M.h. If the exportheader flag is on, LCLint will report exported declarations in M.c that are not declared in M.h.

10.6 Compiler Limits

The ANSI Standard includes limits on minimum numbers that a conforming compiler must support. Whether of not a particular compiler exceeds these limits, it is worth checking that a program does not exceed them so that it may be safely compiled by other compilers. In addition, exceeding a limit may indicate a problem in the code (e.g., it is too complex if the control nest depth limit is exceeded) that should be fixed regardless of the compiler. The following limits are checked by LCLint. For each limit, the maximum value may be set from the command line (or locally using a stylized comment). If the ansilimits flag is on, all limits are checked with the minimum values of a conforming compiler.

includenest

Maximum nesting depth of file inclusion (#include). (ANSI minimum is 8)
controlnestdepth
Maximum nesting of compound statements, control structures. (ANSI minimum is 15)
numenummembers
Number of members in an enum declaration. (ANSI minimum is 127)
numstructfields
Number of fields in a struct or union declaration. (ANSI minimum is 127)






Since human beings themselves are not fully debugged yet, there will be bugs in your code no matter what you do.
- Chris Mason, Zero-defects memo (Microsoft Secrets, Cusumano and Selby)

Appendix A Availability

The web home page for LCLint is http://lclint.cs.virginia.edu/index.html. It includes a this guide in postscript format, samples demonstrating LCLint, and links to related web sites.

LCLint can be downloaded from http://lclint.cs.virginia.edu/download.html.

Several platforms are supported and source code is provided for other platforms.

LCLint can also be run remotely using a form at http://lclint.cs.virginia.edu/run.html

Appendix B Communication

LCLint development is largely driven by suggestions and comments from users. We are also very interested in hearing about your experiences using LCLint in developing or maintaining programs, enforcing coding standards, or teaching courses. For general information, suggestions, and questions on LCLint send mail to lclint@cs.virginia.edu.

To report a bug in LCLint send a message to lclint-bug@cs.virginia.edu.

There are two mailing lists associated with LCLint:

lclint-announce@virginia.edu

Reserved for announcements of new releases and bug fixes. (Everyone who sends mail regarding LCLint is added to this list.)
lclint-interest@virginia.edu
Informal discussions on the use and development of LCLint. To subscribe, send a majordomo message to majordomo@virginia.edu containing the body subscribe lclint-interest.

LCLint discussions relating to checks enabled by specifications or annotations are welcome in the comp.specification.larch usenet group. Messages more focused on C-specific checking would be more appropriate for the lclint-interest list of one of the C language groups.

Appendix C Flags

Flags can be grouped into four major categories:

Global flags can be used in initialization files and at the command line; all other flags may also be used in control comments.

Global Flags

Global flags can be set at the command line or in an options file, but cannot be set locally using stylized comments. These flags control on-line help, initialization files, pre-processor flags, libraries and output.

Help

On-line help provides documentation on LCLint operation and flags. When a help flag is used, no checking is done by LCLint. Help flags may be preceded by - or +.

help

Display general help overview, including list of additional help topics.
help <topic>
Display help on <topic>. Available topics:
annotationsdescribe annotations
commentsdescribe control comments
flagssummarize flag categories
flags <category>all flags pertaining to <category> (one of the categories listed by lclint -help flags)
flags alphaall flags in alphabetical order
flags fullprint a full description of all flags
mailprint information on mailing lists
modesflags settings in modes
prefixcodescharacter codes for setting namespace prefixes
referencesprint references to relevant papers and web sites
varsdescribe environment variables
versionprint maintainer and version information
help <flag>
Describe flag <flag>. (May list several flags.)
warnflags
Display a warning when a flag is set in a surprising way. An error is reported if an obsolete (LCLint Version 1.4 or earlier) flag is set, a flag is set to its current value (i.e., the + or - may be wrong), or a mode selector flag is set after mode checking flags that will be reset by the mode were set. By default, warnflags is on. To suppress flag warnings, use -warnflags.

Initialization

These flags control directories and files used by LCLint. They may be used from the command line or in an options file, but may not be used as control comments in the source code. Except where noted. they have the same meaning preceded by - or +.

tmpdir <directory>

Set directory for writing temp files. Default is /tmp/.
I<directory>
Add directory to path searched for C include files. Note there is no space after the I, to be consistent with C preprocessor flags.
S<directory>
Add directory to path search for .lcl specification files.

f <file>

Load options file <file>. If this flag is used from the command line, the default ~/.lclintrc file is not loaded. This flag may be used in an options file to load in another options file.
nof
Prevents the default options files (./.lclintrc and ~/.lclintrc) from being loaded. (Setting -nof overrides +nof, causing the options files to be loaded normally.)
systemdirs
Set directories for system files (default is "/usr/include"). Separate directories with colons (e.g., "/usr/include:/usr/local/lib"). Flag settings propagate to files in a system directory. If -systemdirerrors is set, no errors are reported for files in system directories.

Pre-processor

These flags are used to define or undefine pre-processor constants. The -I<directory> flag is also passed to the C pre-processor.

D<initializer>

Passed to the C pre-processor.
U<initializer>
Passed to the C pre-processor

Libraries

These flags control the creation and use of libraries.

dump <file>

Save state in <file> for loading. The default extension .lcd is added if <file> has no extension.
load <file>
Load state from <file> (created by -dump). The default extension .lcd is added if <file> has no extension. Only one library file may be loaded.

By default, the standard library is loaded if the -load flag is not used to load a user library. If no user library is loaded, one of the following flags may be used to select a different standard library. Precede the flag by + to load the described library (or prevent a library from being loaded using nolib). See Apppendix F for information on the provided libraries.

nolib

Do not load any library. This prevents the standard library from being loaded.
ansi-lib
Use the ANSI standard library (selected by default).
strict-lib
Use strict version of the ANSI standard library.
posix-lib
Use the POSIX standard library.
posix-strict-lib
Use the strict version of the POSIX standard library.
unix-lib
Use UNIX version of standard library.
unix-strict-lib
Use the strict version of the UNIX standard library.

Output

These flags control what additional information is printed by LCLint. Setting +<flag> causes the described information to be printed; setting -<flag> prevents it. By default, all these flags are off.

usestderr

Send error messages to standard error (instead of standard out).
showsummary
Show a summary of all errors reported and suppressed. Counts of suppressed errors are not necessarily correct since turning a flag off may prevent some checking from being done to save computation, and errors that are not reported may propagate differently from when they are reported.
showscan
Show file names are they are processed.
showalluses
Show list of uses of all external identifiers sorted by number of uses.
stats
Display number of lines processed and checking time.
timedist
Display distribution of where checking time is spent.
quiet
Suppress herald and error count. (If quiet is not set, LCLint prints out a herald with version information before checking begins, and a line summarizing the total number of errors reported.)
whichlib
Print out the standard library filename and creation information.
limit <number>
At most <number> similar errors are reported consecutively. Further errors are suppressed, and a message showing the number of suppressed messages is printed.

Expected Errors

Normally, LCLint will expect to report no errors. The exit status will be success (0) if no errors are reported, and failure if any errors are reported. Flags can be used to set the expected number of reported errors. Because of the provided error suppression mechanisms, these options should probably not be used for final checking real programs but may be useful in developing programs using make.

expect <number>

Exactly <number> code errors are expected. LCLint will exit with failure exit status unless <number> code errors are detected.

Message Format

These flags control how messages are printed. They may be set at the command line, in options files, or locally in syntactic comments. The linelen and limit flags may be preceded by + or - with the same meaning; for the other flags, + turns on the describe printing and - turns it off. The box to the left of each flag gives its default value.

showcolumn

Show column number where error is found. Default: +
showfunc
Show name of function (or macro) definition containing error. The function name is printed once before the first message detected in that function. Default: +

showallconjs
Show all possible alternate types (see Section 8.2.2). Default: -
paren-file-format
Use file(line) format in messages.
hints
Provide hints describing an error and how a message may be suppressed for the first error reported in each error class. Default: +
forcehints
Provide hints for all errors reported, even if the hint has already been displayed for the same error class. Default: -
linelen <number>
Set length of maximum message line to <number> characters. LCLint will split messages longer than <number> characters long into multiple lines. Default: 80

Mode Selector Flags

Mode selects flags set the mode checking flags to predefined values. They provide a quick coarse-grain way of controlling what classes of errors are reported. Specific checking flags may be set after a mode flag to override the mode settings. Mode flags may be used locally, however the mode settings will override specific command line flag settings. A warning is produced if a mode flag is used after a mode checking flag has been set.

These are brief descriptions to give a general idea of what each mode does. To see the complete flag settings in each mode, use lclint -help modes. A mode flag has the same effect when used with either + or -.

weak

Weak checking, intended for typical unannotated C code. No modifies checking, macro checking, rep exposure, or clean interface checking is done. Return values of type int may be ignored. The types bool, int, char and user-defined enum types are all equivalent. Old style declarations are unreported.
standard
The default mode. All checking done by weak, plus modifies checking, global alias checking, use all parameters, using released storage, ignored return values or any type, macro checking, unreachable code, infinite loops, and fall-through cases. The types bool, int and char are distinct. Old style declarations are reported.
checks
Moderately strict checking. All checking done by standard, plus must modification checking, rep exposure, return alias, memory management and complete interfaces.
strict
Absurdly strict checking. All checking done by checks, plus modifications and global variables used in unspecified functions, strict standard library, and strict typing of C operators. A special reward will be presented to the first person to produce a real program that produces no errors with strict checking.
Checking Flags

These flags control checking done by LCLint. They may be set locally using syntactic comments, from the command line, or in an options file. Some flags directly control whether a certain class of message is reported. Preceding the flag by + turns reporting on, and preceding the flag by - turns reporting off. Other flags control checking less directly by determining default values (what annotations are implicit), making types equivalent (to prevent certain type errors), controlling representation access, etc. For these flags, the effect of + is described, and the effect of - is the opposite (or explicitly explained if there is no clear opposite). The organization of this section mirrors Sections 3-10.

Key

Under each flag name is a flag descriptor encoding the what kind of flag it is and its default value. The descriptions are:

plain: -

A plain flag. The value after the colon gives the default setting (e.g., this flag is off.)
m: --++
A mode checking flag. The value of the flag is set by the mode selector. The four signs give the setting in the weak, standard, checks and strict modes. (e.g., this flag is off in the weak and standard modes, and on in the checks and strict modes.)
shortcut
A shortcut flag. This flag sets other flags, so it has no default value.

Types

Abstract Types

plain: -
impabstract

Implicit abstract annotation for type declarations that do not use concrete.

m: -+++
mutrep

Representation of mutable type has sharing semantics.

Access (Section 3.1)

plain: +
accessmodule

An abstract type defined in M.h (or specified in M.lcl) is accessible in M.c.

plain: +
accessfile

An abstract type named type is accessible in files named type.<extenstion>.

plain: +
accessczech

An abstract type named type may be accessible in a function named type_name. (see Section 9.1.1)

plain: -
accessslovak

An abstract type named type may be accessible in a function named typeName. (see Section.9.1.2)

plain: -
accessczechoslovak

An abstract type named type may be accessible in a function named type_name or typeName. (see Section 9.1.3)

shortcut
accessall

Sets accessmodule, accessfile and accessczech.

Boolean Types (Section 3.3)

These flags control the type name used to represent booleans, and whether the boolean type is abstract.

plain: -
bool

Boolean type is an abstract type.

plain: unset
booltype <name>

Set name of boolean type to <name>.

plain: FALSE
boolfalse <name>

Set name of boolean false to <name>.

plain: TRUE
booltrue <name>

Set name of boolean true to <name>.

Predicates

m: --++
predboolptr

Type of condition test is a pointer.

m: -+++
predboolint

Type of condition test is an integral type.

m: ++++
predboolothers

Type of condition test is not a boolean, pointer or integral type.

shortcut
predbool

Sets predboolint, predboolptr and preboolothers.

plain: +
predassign

The condition test is an assignment expression. If an assignment is intended, add an extra parentheses nesting (e.g., if ((a = b)) ...).

Primitive Operations

m: ---+
ptrarith

Arithmetic involving pointer and integer.

m: ++--
ptrnegate

Allow the operand of the ! operator to be a pointer.

m: ---+
bitwisesigned

An operand to a bitwise operator is not an unsigned values. This may have unexpected results depending on the signed representations.

m: -+++
shiftsigned

An operand to a shift operator is not unsigned values. This may have unexpected results depending on the signed representations.

m: ---+
strictops

Primitive operation does not type check strictly.

m: ---+
sizeoftype

Operand of sizeof operator is a type. (Safer to use expression, int *x = sizeof (*x); instead of sizeof (int).)

Format Codes

plain: +
formatcode

Invalid format code in format string for printflike or scanflike function.

plain: +
formattype

Type-mismatch in parameter corresponding to format code in a printflike or scanflike function.

Main

plain: +
maintype

Type of main does not match expected type (function returning an int, taking no parameters or two parameters of type int and char **.)

Comparisons

m: -+++
boolcompare

Comparison between boolean values. This is dangerous since there may be multiple TRUE values if any non-zero value is interpreted at TRUE.

m: -+++
realcompare

Comparison involving float or double values. This is dangerous since it may produce unexpected results because floating point representations are inexact.

m: -+++
ptrcompare

Comparison between pointer and number.

Type Equivalence

m: +---
voidabstract

Allow void * to match pointers to abstract types. (Casting a pointer to an abstract type to a pointer to void is okay if +voidabstract is set.)

plain: +
castfcnptr

A pointer to a function is cast to (or used as) a pointer to void (or vice versa).

m: +---
forwarddecl

Forward declarations of pointers to abstract representation match abstract type.

m: -+++
imptype

A variable declaration has no explicit type. The type is implicitly int.

m: +
incompletetype

A formal parameter is declared with an incomplete type (e.g., int[][]).

m: +---
charindex

Allow char to index arrays.

m: ----
enumindex

Allow members of enum type to index arrays.

m: +---
boolint

Make bool and int types equivalent. (No type errors are reported when a boolean is used where an integral type is expected and vice versa.)

m: +---
charint

Make char and int types equivalent.

m: ++--
enumint

Make enum and int types equivalent.

m: +---
floatdouble

Make float and double types equivalent.

m: ----
ignorequals

Ignore type qualifiers (long, short, unsigned).

m: ++--
relaxquals

Report qualifier mismatches only if dangerous (information may be lost since a larger type is assigned to (or passed as) a smaller one or a comparison uses signed and unsigned values.)

m: ----
ignoresigns

Ignore signs in type comparisons (unsigned matches signed).

plain: -
longintegral

Allow long type to match an arbitrary integral type (e.g., size_t).

m: +---
longunsignedintegral

Allow long unsigned type to match an arbitrary integral type (e.g., dev_t).

plain: -
matchanyintegral

Allow any integral type to match an arbitrary integral type (e.g., dev_t).

m: +---
long-unsigned-unsigned-integral

Allow unsigned long type to match an arbitrary unsigned integral type (e.g., size_t).

m: +---
long-signed-integral

Allow long type to match an arbitrary signed integral type (e.g., ssize_t).

m: ++++
num-literal

Integer literals can be used as floats.

plain: +
char-int-literal

A character constant may be used as an int.

m: ++++
zeroptr

Literal 0 may be used as a pointer.

m: ----
relaxtypes

Allow all numeric types to match.

plain: +
fullinitblock

Initializer does not set every field in the structure.

Function Interfaces

Modification (Section 4.1)

m: ++++
modifies

Undocumented modification of caller-visible state. Without +moduncon, modification errors are only reported in the definitions of functions declared with a modifies clause (or specified).

m: --++
mustmod

Documented modification is not detected. An object listed in the modifies clause for a function, is not modified by the implementation.

shortcut
mod-uncon

Report modification errors in functions declared without a modifies clause.(Sets modnomods, mod-globs-nomods and mod-strict-globs-nomods.)

m: ---+
mod-nomods

Report modification errors (not involving global variables) in functions declared without a modifies clause.

m: ---+
mod-uncon-nomods

An unconstrained function is called in a function body where modifications are checked. Since the unconstrained function may modify anything, there may be undetected modifications in the checked function.

m: ---+
mod-internal-strict

A function that modifies internalState is called from a function that does not list internalState in its modifies clause.

m: ---+
mod-file-sys

A function modifies the file system but does not list fileSystem in its modifies clause.

Global Variables (Section 4.2)

Errors involving the use and modification of global and file static variables are reported depending on flag settings, annotations where the global variable is declared, and whether or not the function where the global is used was declared with a globals clause.

m: ++++
globs

Undocumented use of a checked global variable in a function with a globals list.

m: ++++
globuse

A global listed in the globals list is not used in the implementation.

m: ---+
globnoglobs

Use of a checked global in a function with no globals list.

m: ---+
internalglobs

Undocumented use of internal state (should have globals internalState).

m: ---+
internalglobsnoglobs

Use of internal state in function with no globals list.

m: -+++
globstate

A function returns with global in inconsistent state (null or undefined)

m: --++
allglobs

Report use and modification errors for globals not annotated with unchecked.

m: ++++
checkstrictglobs

Report use and modification errors for checkedstrict globals.

Modification of Global Variables

m: -+++
modglobs

Undocumented modification of a checked global variable.

m: ---+
modglobsunchecked

Undocumented modification of an unchecked global variable.

m: ---+
modglobsnomods

Undocumented modification of a checked global variable in a function with no modifies clause.

m: ---+
modstrictglobsnomods

Undocumented modification of a checkedstrict global variable in a function declared with no modifies clause.

Globals Lists and Modifies Clauses

m: ---+
warnmissingglobs

Global variable used in modifies clause is not listed in globals list. (The global is added to the globals list.)

m: ---+
warnmissingglobsnoglobs

Global variable used in modifies clause of a function with no globals list.

m: --++
globsimpmodsnothing

A function declared with a globals list but no modifies clause is assumed to modify nothing.

m: ----
modsimpnoglobs

A function declared with a modifies clause but no globals list is assumed to use no globals.

Implicit Checking Qualifiers

m: ----
impcheckedglobs

Implicit checked qualifier on global variables with no checking annotation.

m: ----
impcheckedstatics

Implicit checked qualifier file static scope variables with no checking annotation.

m: ----
impcheckmodglobs

Implicit checkmod qualifier on global variables with no checking annotation.

m: ----
impcheckmodstatics

Implicit checkmod qualifier file static scope variables with no checking annotation.

m: ---+
impcheckedstrictglobs

Implicit checked qualifier on global variables with no checking annotation.

m: ---+
impcheckedstrictstatics

Implicit checked qualifier file static scope variables with no checking annotation.

m: --++
impcheckmodinternals

Implicit checkmod qualifier on function scope static variables with no checking annotation.

m: -+++
impglobsweak

Global Aliasing

shortcut
globalias

Function returns with global aliasing external state (sets checkstrictglobalias, checkedglobalias, checkmodglobalias and uncheckedglobalias).

m: -+++
checkstrictglobalias

Function returns with a checkstrict global aliasing external state.

m: -+++
checkedglobalias

Function returns with a checked global aliasing external state.

m: -+++
checkmodglobalias

Function returns with a checkmod global aliasing external state.

m: --++
uncheckedglobalias

Function returns with an unchecked global aliasing external state.

Declaration Consistency (Section 4.3)

m: -+++
incondefs

Identifier redeclared or redefined with inconsistent type.

m: -+++
incondefslib

Identifier defined in a library is redefined with inconsistent type

m: ----
overload

Standard library function overloaded.

m: -+++
matchfields

A struct or enum type is redefined with inconsistent fields or members.

Memory Management

Reporting of memory management errors is controlled by flags setting checking and implicit annotations and code annotations.

Deallocation Errors (Section 5.2)

m: -+++
usereleased

Storage used after it may have been released.

m: ---+
strictusereleased

An array element used after it may have been released.

Inconsistent Branches

m: -+++
branchstate

Storage has inconsistent states of alternate paths through a branch (e.g., it is released in the true branch of an if-statement, but there is no else branch.)

m: ---+
strictbranchstate

Storage through array fetch has inconsistent states of alternate paths through a branch. Since array elements are not checked accurately, this may lead to spurious errors.

m: --++
deparrays

Treat array elements as dependent storage. Checking of array elements cannot be done accurately by LCLint. If deparrays is not set, array elements are assumed to be independent, so code that releases the same element more than once will produce no error. If deparrays is set, array elements are assumed to be dependent, so code that releases the same element more that once will produce an error, but so will code that releases different elements correctly will produce a spurious error.

Memory Leaks

m: -+++
mustfree

Allocated storage was not released before return or scope exit Errors are reported for only, fresh or owned storage.

m: -+++
compdestroy

All only references derivable from out only parameter of type void * must be released. (This is the type of the parameter to free, but may also be used for user-defined deallocation functions.)

m: ---+
strictdestroy

Report complete destruction errors for array elements that may have been released. (If strictdestroy is not set, LCLint will assume that if any array element was released, the entire array was correctly released.)

Transfer Errors

A transfer error is reported when storage is transferred (by an assignment, passing a parameter, or returning) in a way that is inconsistent.

shortcut
memtrans

Sets all memory transfer errors flags.

m: -+++
onlytrans

Only storage transferred to non-only reference (memory leak).

m: -+++
ownedtrans

Owned storage transferred to non-owned reference (memory leak).

m: -+++
freshtrans

Newly-allocated storage transferred to non-only reference (memory leak).

m: -+++
sharedtrans

Shared storage transferred to non-shared reference.

m: -+++
dependenttrans

Inconsistent dependent transfer. Dependent storage is transferred to a non-dependent reference.

m: -+++
temptrans

Temporary storage (associated with a temp formal parameter) is transferred to a non-temporary reference. The storage may be released or new aliases created.

m: -+++
kepttrans

Kept storage transferred to non-temporary reference.

m: -+++
keeptrans

Keep storage is transferred in a way that may add a new alias to it, or release it.

m: -+++
refcounttrans

Reference counted storage is transferred in an inconsistent way.

m: -+++
newreftrans

A new reference transferred to a reference counted reference (reference count is not set correctly).

m: -+++
immediatetrans

An immediate address (result of &) is transferred inconsistently.

m: -+++
statictrans

Static storage is transferred in an inconsistent way.

m: -+++
exposetrans

Inconsistent exposure transfer. Exposed storage is transferred to a non-exposed, non-observer reference.

m: -+++
observertrans

Inconsistent observer transfer. Observer storage is transferred to a non-observer reference.

m: -+++
unqualifiedtrans

Unqualified storage is transferred in an inconsistent way.

Initializers

m: --++
onlyunqglobaltrans

Only storage transferred to an unqualified global or static reference. This may lead to a memory leak, since the new reference is not necessarily released.

m: --++
staticinittrans

Static storage is used as an initial value in an inconsistent way.

m: --++
unqualifiedinittrans

Unqualified storage is used as an initial value in an inconsistent way.

Derived Storage

m: -+++
compmempass

Storage derivable from a parameter does not match the alias kind expected for the formal parameter.

Stack References

m: ++++
stackref

A stack reference is pointed to by an external reference when the function returns. Since the call frame will be destroyed when the function returns the return value will point to dead storage. (Section 5.2.6)

Implicit Memory Annotations (Section 5.3)

plain: +
globimponly

Assume unannotated global storage is only.

plain: +
paramimptemp

Assume unannotated parameter is temp.

plain: +
retimponly

Assume unannotated returned storage is only.

plain: +
structimponly

Assume unannotated structure or union field is only.

shortcut
codeimponly

Sets globimponly, retimponly and structimponly.

m: -+++
memimp

Report memory errors for unqualified storage.

m: ----
passunknown

Passing a value as an unannotated parameter clears its annotation. This will prevent many spurious errors from being report for unannotated programs, but eliminates the possibility of detecting many errors.

Sharing

Aliasing (Section 6)

m: -+++
aliasunique

A actual parameter that is passed as a unique formal parameter is aliased by another parameter or global variable.

m: -+++
mayaliasunique

A actual parameter that is passed as a unique formal parameter may be aliased by another parameter or global variable.

m: -+++
mustnotalias

An alias has been added to a temp-qualifier parameter or global that is visible externally when the function returns.

m: --++
retalias

A function returns an alias to parameter or global.

Exposure (Section 6.2)

shortcut
repexpose

The internal representation of an abstract type is visible to the caller. This means clients may have access to a pointer into the abstract representation. (Sets assignexpose, retexpose, and castexpose.)

m: --++
assignexpose

Abstract representation is exposed by an assignment or passed parameter.

m: --++
castexpose

Abstract representation is exposed through a cast.

m: --++
retexpose

Abstract representation is exposed by a return value.

Observer Modifications

plain: +
modobserver

Possible modification of observer storage.

m: ---+
modobserveruncon

Storage declared with observer may be modified through a call to an unconstrained function.

String Literals (Section 6.2.1)

m: --++
readonlytrans

Report memory transfer errors for initializations to read-only string literals

m: -+++
readonlystrings

String literals are read-only (ANSI semantics). An error is reported if a string literal may be modified or released.

Use Before Definition (Section 7.1)

m: -+++
usedef

The value of a location that may not be initialized on some execution path is used.

m: ----
impouts

Allow unannotated pointer parameters to functions to be implicit out parameters.

m: -+++
compdef

Storage derivable from a parameter, return value or global variable is not completely defined.

m: -+++
uniondef

No field of a union is defined. (No error is reported if at least one union field is defined.)

m: -+++
mustdefine

Parameter declared with out is not defined before return or scope exit.

Null Pointers (Section 7.2)

m: -+++
nullderef

A possibly null pointer may be dereferenced.

m: -+++
nullpass

A possibly null pointer is passed as a parameter not annotated with null.

m: -+++
nullret

A possibly null pointer is return as a result not annotated with null.

m: -+++
nullstate

Possibly null pointer reachable from a reference with no null annotation.

m: -+++
nullassign

Inconsistent assignment or initialization involving null pointer.

shortcut
null

Sets all null checking flags.

Macros (Section 8)

These flags control expansion and checking of macro definitions and invocations.

Macro Expansion

These flags control which macros are checked as functions or constants, and which are expanded in the pre-processing phase. Macros preceded by /*@notfunction@*/ are never expanded regardless of these flag settings. These flags may be used in source-file control comments.

plain: -
fcnmacros

Macros defined with parameter lists are not expanded and are checked as functions.

plain: -
constmacros

Macros defined without parameter lists are not expanded and are checked as constants.

shortcut
allmacros

Sets allfcnmacros and allconstmacros.

plain: -
libmacros

Macros defining identifiers declared in a loaded library are not expanded and are checked according to the library information.

Macro Definitions

These flags control what errors are reported in macro definitions.

m: -+++
macrostmt

Macro definition is not syntactically equivalent to function. This means if the macro is used as a statement (e.g., if (test) macro();) unexpected behavior may result. One fix is to surround the macro body with do { ...; } while (FALSE).

m: -+++
macroparams

A macro parameter is not used exactly once in all possible invocations of the macro.

m: -+++
macroassign

A macro parameter is used as the left side of an assignment expression.

m: -+++
macroparens

A macro parameter is used without parentheses (in potentially dangerous context).

m: ---+
macroempty

Macro definition of a function is empty.

m: -+++
macroredef

Macro is redefined. There is another macro defined with the same name.

m: -+++
macrounrecog

An unrecognized identifier appears in a macro definition. Since the identifier may be defined where the macro is used, this could be okay, but LCLint will not be able to check the unrecognized identifier appropriately.

Corresponding Declarations

m: ++++
macromatchname

A iter or constant macro is defined using a different name from the one used in the previous syntactic comment.

shortcut
macrodecl

A macro definition has no corresponding declaration. (Sets macrofcndecl and macroconstdecl.)

m: -+++
macrofcndecl

Macro definition with parameter list has no corresponding function prototype. Without a prototype, the types of the macro result and parameters is unknown.

m: -+++
macroconstdecl

A macro definition without parameter list has no corresponding constant declaration.

plain: +
nextlinemacros

A constant or iter declaration is not immediately followed by a macro definition.

Side-Effect Free Parameters (Section 8.2.1)

These flags control error reporting for parameters with inconsistent side-effects in invocations of checked function macros and function calls.

m: -+++
sefparams

An actual parameter with side-effects is passed as a formal parameter declared with sef.

m: --++
sefuncon

An actual parameter involving a call to an unconstrained function (declared without modifies clause) that may modify anything is passed as a sef parameter.

Iterators

plain: -
hasyield

An iterator has been declared with no parameters annotated with yield.

Naming Conventions

plain: +
namechecks

Turns all name checking on or off without changing other settings.

Type-Based Naming Conventions (Section 9.1)

Czech Naming Convention

shortcut
czech

Selects complete Czech naming convention (sets accessczech, czechfcns, czechvars, czechconsts, czechmacros, and czechtypes).

plain: +
accessczech

Allow access to abstract types following Czech naming convention. The representation of an abstract type named t is accessible in the definition of a function or constant named t_name.

plain: -
czechfcns

Function or iterator name is not consistent with Czech naming convention.

plain: -
czechvars

Variable name is not consistent with Czech naming convention.

plain: -
czechmacros

Expanded macro name is not consistent with Czech naming convention.

plain: -
czechconsts

Constant name is not consistent with Czech naming convention.

plain: -
czechtypes

Type name is not consistent with Czech naming convention. Czech type names must not use the underscore character.

Slovak Naming Convention

shortcut
slovak

Selects complete Slovak naming convention (sets accessslovak, slovakfcns, slovakvars, slovakconsts, slovakmacros, and slovaktypes).

plain: -
accessslovak

Allow access to abstract types following Slovak naming convention. The representation of an abstract type named t is accessible in the definition of a function or constant named tName.

plain: -
slovakfcns

Function or iterator name is not consistent with Slovak naming convention.

plain: -
slovakmacros

Expanded macro name is not consistent with Slovak naming convention.

plain: -
slovakvars

Variable name is not consistent with Slovak naming convention.

plain: -
slovakconsts

Constant name is not consistent with Slovak naming convention.

plain: -
slovaktypes

Type name is not consistent with Slovak naming convention. Slovak type names may not include uppercase letters.

Czechoslovak Naming Convention

shortcut
czechoslovak

Selects complete Czechoslovak naming convention (sets accessczechoslovak, czechoslovakfcns, czechoslovakvars, czechoslovakconsts, czechoslovakmacros, and czechoslovaktypes).

plain: -
accessczechoslovak

Allow access to abstract types by Czechoslovak naming convention. The representation of an abstract type named t is accessible in the definition of a function or constant named t_name or tName.

plain: -
czechoslovakfcns

Function name is not consistent with Czechoslovak naming convention.

plain: -
czechoslovakmacros

Expanded macro name is not consistent with Czechoslovak naming convention.

plain: -
czechoslovakvars

Variable name is not consistent with Czechoslovak naming convention.

plain: -
czechoslovakconsts

Constant name is not consistent with Czechoslovak naming convention.

plain: -
czechoslovaktypes

Type name is not consistent with Czechoslovak naming convention. Czechoslovak type names may not include uppercase letters or the underscore character.

Namespace Prefixes (Section 9.2)

macrovarprefix <prefix string>

Set namespace prefix for variables declared in a macro body. (Default is m_.)

plain: +
macrovarprefixexclude

A variable declared outside a macro body starts with the macrovarprefix.

tagprefix <prefix string>

Set namespace prefix of struct, union or enum tag identifiers.

plain: -
tagprefixexclude

An identifier that is not a tag starts with the tagprefix.

enumprefix <prefix string>

Set namespace prefix for enum members.

plain: -
enumprefixexclude

An identifier that is not an enum member starts with the enumprefix.

filestaticprefix <prefix string>

Set namespace prefix for file static declarations.

plain: -
filestaticprefixexclude

An identifier that is not file static starts with the filestaticprefix.

globalprefix <prefix string>

Set namespace prefix for global variables.

plain: -
globalprefixexclude

An identifier that is not a global variable starts with the globalprefix.

typeprefix <prefix string>

Set namespace prefix for user-defined types.

plain: -
typeprefixexclude

An identifier that is not a type name starts with the typeprefix.

externalprefix <prefix string>

Set namespace prefix for external identifiers.

plain: -
externalprefixexclude

An identifier that is not external starts with the externalprefix.

localprefix <prefix string>

Set namespace prefix for local variables.

plain: -
localprefixexclude

An identifier that is not a local variable starts with the localprefix.

uncheckedmacroprefix <prefix string>

Set namespace prefix for unchecked macros.

plain: -
uncheckedmacroprefixexclude

An identifier that is not the name of an unchecked macro starts with the uncheckedmacroprefix.

constprefix <prefix string>

Set namespace prefix for constants.

plain: -
constprefixexclude

An identifier that is not a constant starts with the constantprefix.

iterprefix <prefix string>

Set namespace prefix for iterators.

plain: -
iterprefixexclude

An identifier that is not a iter starts with the iterprefix.

proto-param-prefix <prefix string>

Set namespace prefix for parameters in function prototypes.

plain: -
proto-param-prefix-exclude

An identifier that is not a parameter in a function prototype starts with the protoprarmprefix.

m: --++
proto-param-name

A parameter in a function prototype has a name (can interfere with macro definitions).

m: ---+
proto-param-match

The name of a parameter in a function definition does not match the corresponding name of the parameter in a function prototype (after removing the protoparamprefix).

Naming Restrictions (Section 9.3)

m: -+++
shadow

Declaration reuses name visible in outer scope.

Reserved Names

m: --++
ansi-reserved

External name conflicts with name reserved for the compiler or standard library.

m: ---+
ansi-reserved-internal

Internal name conflicts with name reserved for the compiler or standard library.

m: --++
cpp-names

Internal or external name conflicts with a C++ reserved word. (Will cause problems if program is compiled with a C++ compiler.)

Distinct External Names

plain: -
distinct-external-names

An external name is not distinguishable from another external name using externalnamelen significant characters.

external-name-len<number>

Sets the number of significant characters in an external name (ANSI default minimum is 6). Sets +distinctexternalnames.

plain: -
external-name-case-insensitive

Make alphabetic case insignificant in external names. According to ANSI standard, case need not be significant in an external name. If +distinctexternalnames is not set, sets +distinctexternalnames with unlimited external name length.

Distinct Internal Names

m: ----
distinct-internal-names

An internal name is not distinguishable from another internal name using internalnamelen significant characters. (Also effected by internalnamecaseinsensitive and internalnamelookalike.)

internal-name-len <number>

Set the number of significant characters in an internal name. Sets +distinctinternalnames.

plain: -
internal-name-case-insensitive

Set whether case is significant an internal names (-internalnamecaseinsensitive means case is significant). If +distinctinternalnames is not set, sets +distinct-internal-names with unlimited internal name length.

plain: -
internalnamelookalike

Set whether similar looking characters (e.g., "1" and "l") match in internal names.

Other Checks

Undefined Evaluation Order (Section 10.1)

m: -+++
evalorder

Behavior of an expression is undefined because sub-expressions contain interfering side effects that may be evaluated in any order.

m: ---+
evalorderuncon

An expression may be undefined because a sub-expression contains a call to an unconstrained function (no modifies clause) that may modify something that may be modified or used by another sub-expression.

Problematic Control Structures (Section 10.2)

m: -+++
infloops

Likely infinite loop is detected (Section 10.2.1).

m: --++
infloopsuncon

Likely infinite loop is detected. Loop test or body calls an unconstrained function, that may produce an undetected modification.

m: ---+
elseifcomplete

There is no final else following an else if construct (Section 10.2.5).

m: -+++
casebreak

These is a non-empty case in a switch not followed by a break (Section 10.2.2).

m: -+++
misscase

A switch on an enum type is missing a case for a member of the enumerator.

m: ----
loopexec

Assume all loops execute at least once. This effects use-before-definition and memory checking. It should probably not be used globally, but may be used surrounding a particular loop that is known to always execute to prevent spurious messages.

Deep Break (Section 10.2.3)

shortcut
deepbreak

Report errors for break statements inside a nested while, for or switch. (Sets all nested break and continue flags.)

m: --++
looploopbreak

There is a break inside a while, for or iterator loop that is inside a while, for or iterator loop. Mark with /*@innerbreak@*/ to suppress the message.

m: --++
switchloopbreak

There is a break inside a while, for or iterator loop that is inside a switch statement. Mark with /*@loopbreak@*/.

m: ---+
loopswitchbreak

There is a break inside a switch statement that is inside a while, for or iterator loop. Mark with /*@switchbreak@*/.

m: ---+
switchswitchbreak

There is a break inside a switch statement that is inside another switch statement. Mark with /*@innerbreak@*/.

m: ---+
looploopcontinue

There is a continue inside a while, for or iterator loop that is inside a while, for or iterator loop. Mark with /*@innercontinue@*/.

Loop and if Bodies (Section 10.2.4)

shortcut
allempty

An if, while or for statement has no body (sets ifempty, whileempty and forempty.)

shortcut
allblock

The body of an if, while or for statement is not a block (sets ifblock, whileblock and forblock.)

m: --++
whileempty

A while statement has no body.

m: ---+
whileblock

The body of a while statement is not a block

m: ---+
forempty

A for statement has no body.

m: ---+
forblock

The body of a for statement is not a block.

m: ++++
ifempty

An if statement has no body.

m: ---+
ifblock

The body of an if statement is not a block.

Suspicious Statements (Section 10.3)

m: -+++
unreachable

Code is not reached on any possible execution.

m: -+++
noeffect

Statement has no effect.

m: ---+
noeffectuncon

Statement involving call to unconstrained function may have no effect.

m: -+++
noret

There is a path with no return in a function declared to return a non-void value.

Ignored Return Values (Section 10.3.2)

These flags control when errors are reported for function calls that do not use the return value. Casting the function call to void or declaring the called function to return /*@alt void@*/.

m: -+++
retvalbool

Return value of type bool ignored.

m: -+++
retvalint

Return value of type int ignored.

m: ++++
retvalother

Return value of type other than bool or int ignored.

shortcut
retval

Return value ignored (Sets retvalbool, retvalint, retvalother.)

Unused Declarations (Section 10.4)

These flags control when errors are reported for declarations that are never used. The unused annotation can be used to prevent unused errors from being report for a particular declaration.

m: ---+
topuse

A external declaration is not used in any file.

m: -+++
constuse

Constant never used.

m: -+++
enummemuse

Member of enumerator never used.

m: ++++
varuse

Variable never used.

m: -+++
paramuse

Function parameter never used.

m: ++++
fcnuse

Function is never used.

m: ++++
typeuse

Defined type never used.

m: -+++
fielduse

Field of structure or union type is never used.

m: ---+
unusedspecial

Declaration in a special file (corresponding to .l or .y file) is unused.

Complete Programs (Section 10.5)

m: --++
declundef

Function, variable, iterator or constant declared but never defined.

shortcut
partial

Check as partial system (sets -declundef, -exportlocal and prevents checking of macros in headers without corresponding .c files.)

Exports

m: ---+
export-local

A declaration is exported but not used outside this module. (Declaration can use the static qualifier.)

m: --++
export-header

A declaration (other than a variable) is exported but does not appear in a header file.

m: --++
export-header-var

A variable declaration is exported but does not appear in a header file.

Unrecognized Identifiers

plain: +
unrecog

An unrecognized identifier is used.

plain: +
sys-unrecog

Report unrecognized identifiers that start with the system prefix, __ (two underscores).

plain: -
repeat-unrecog

Report multiple messages for unrecognized identifiers. If repeatunrecog is not set, an error is reported only the first time a particular unrecognized identifier appears in the file.

Multiple Definition and Declarations

plain: +
redef

A function or variable is defined more than once.

m: --++
redecl

An identifier is declared more than once.

m: -+++
nested-extern

An extern declaration is used inside a function body.

ANSI C Conformance

m: --++
noparams

A function is declared without a parameter list prototype.

m: ---+
oldstyle

Function definition is in old style syntax. Standard prototype syntax is preferred.

m: -+++
exitarg

Argument to exit has implementation defined behavior. The only valid arguments to exit are EXIT_SUCCESS, EXIT_FAILURE and 0. An error is reported if LCLint can detect statically that the argument to exit is not one of these.

Limits (Section 10.6)

shortcut
ansilimits

Check for violations of standard limits (Sets controlnestdepth, stringliterallen, includenest, numstructfields, and numenummembers).

m: ---+

controlnestdepth <number>

Set maximum nesting depth of compound statements, iteration control structures, and selection control structures (ANSI minimum is 15).

m: ---+
stringliterallen <number>

Set maximum length of string literals (ANSI minimum is 509).

m: ---+
numstructfields <number>

Set maximum number of fields in a struct or union (ANSI minimum is 127).

m: ---+
numenummembers <number>

Set maximum number of members of an enum type (ANSI minimum is 127).

m: --++
includenest <number>

Set maximum number of nested #include files (ANSI minimum is 8).

Header Inclusion (Apppendix F)

plain: +
skip-ansi-headers

Prevent inclusion of header files in a system directory with names that match standard ANSI headers. The symbolic information in the standard library is used instead. In effect only if a library that includes the ANSI library is used. The ANSI headers are: assert, ctype, errno, float, limits, locale, math, setjmp, signal, stdarg, stddef, stdio, stdlib, strings, string, time, and wchar.

plain: +
skip-posix-headers

Prevent inclusion of header files in a system directory with names that match standard POSIX headers. The symbolic information in the standard library is used instead. In effect only if a library that includes the POSIX library is used. The POSIX headers are: dirent, fcntl, grp, pwd, termios, sys/stat, sys/times, sys/types, sys/utsname, sys/wait, unistd, and utime.

plain: +
warn-posix-headers

Report use of a POSIX header when checking a program with a non-POSIX library.

skipsysheaders

Do not include header files in system directories (as set by -sysdirs)

plain: +
sys-dir-expand-macros

Expand macros in system directories regardless of other settings, except for macros corresponding to names defined in a load library.

m: ---+
sys-dir-errors

Report errors in files in system directories (set by -systemdirs).

single-include

Optimize header inclusion to only include each header file once.

never-include

Use library information instead of including header files.

Comments

These flags control how syntactic comments are interpreted (see Apppendix E).

commentchar <char>

Set the marker character for syntactic comments. Comments beginning with /*<char> are interpreted by LCLint. Default: @

plain: -
noaccess

Ignore access comments.

plain: -
nocomments

Ignore all stylized comments.

plain: +
supcounts

Actual number of errors does not match number in /*@i<n>@*/

plain: +
lintcomments

Interpret traditional lint comments (/*FALLTHROUGH*/, /*NOTREACHED*/, /*PRINTLIKE*/).

m: -+++
warnlintcomments

Print a warning and suggest an alternative when a traditional lint comment is used.

plain: +
unrecogcomments

Stylized comment is unrecognized.

Parsing

plain: -
continue-comment

A line continuation marker (\) appears inside a comment on the same line as the comment close. Preprocessors should handle this correctly, but it causes problems for some preprocessors. plain: +
nest-comment

A comment open sequence (/*) appears inside a comment. This usually indicates that an earlier comment was not closed.

plain: +
duplicate-quals

Report duplicate type qualifiers (e.g., long long). Duplicate type qualifiers not supported by ANSI, but some compilers (e.g., gcc) do support duplicate qualifiers.

plain: +
gnu-extensions

Support some GNU (gcc) and Windows language extensions.

Array Formal Parameters

These flags control reporting of common errors caused by confusion about the semantics of array formal parameters.

General Checks

plain: +
sizeof-formal-array

The sizeof operator is used on a parameter declared as an array. (In many instances this has unexpected behavior, since the result is the size of a pointer to the element type, not the number of elements in the array.)

plain: +
fixed-formal-array

An array formal parameter is declared with a fixed size (e.g., int x[20]). This is likely to be confusing, since the size is ignored.

plain: -
formal-array

A formal parameter is declared as an array. This is probably not a problem, but can be confusing since it is treated as a pointer.

These flags should probably not be set globally since the turn off general checks that should always be done. They may be used locally to suppress spurious errors.

plain: +
abstract

A data abstraction barrier is violated.

plain: +
control

A control flow error is detected.

plain: +
syntax

Parse error.

plain: -
trytorecover

Try to recover from a parse error. If trytorecover is not set, LCLint will abort checking after a parse error is detected. If it is set, LCLint will attempt to recover, but LCLint does performs only minimal error recovery.

plain: +
type

Type mismatch.

Flag Name Abbreviations

Within a flag name, abbreviations may be used. Table 2 shows the flag name abbreviations. The expanded and short forms are interchangeable in flag names.

For example, globsimpmodsnothing and globalsimpliesmodifiesnothing denote the same flag. Abbreviations in flag names allow pronounceable, descriptive names to be used without making flag names excessively long (although one must admit even globsimpmodsnothing is a bit of a mouthful.)

Expanded FormShort Form
constant const
declaration decl
function fcn
global glob
implicit, implied imp
iterator iter
length len
modifies mods
modify mod
memory mem
parameter param
pointer ptr
return ret
variable var
unconstrained, unconst uncon
Table 2. Flag name abbreviations.

Appendix D Annotations

The grammar below is the C syntax from [K&R,A13] modified to show the syntax of syntactic comments. Only productions effected by LCLint annotations are shown. In the annotations, the @ represents the comment marker char, set by -commentchar (default is @).

Functions

direct-declarator:

	  direct-declarator (parameter-type-list_opt) globals_opt modifies_opt
	| direct-declarator (identifier-list_opt) globals_opt modifies_opt

globals: (Section 4.2)

	  /*@globals globitem,+ ;_opt @*/
	| /*@globals declaration-list_opt ;_opt @*/
globitem:

	   globannot* identifier
        |  internalState
	|  systemState

globannot: undef | killed

modifies: (Section 4.1)

           /*@modifies moditem,+;_opt @*/	
         | /*@modifies nothing ;_opt @*/
         | /*@*/  (Abbreviation for no globals and modifies nothing.)

moditem:

	   expression	
         | internalState
         | systemState

Iterators (Section 8.4)

The globals and modifies clauses for an iterator are the same as those for a function, except they are not enclosed by a comment, since the iterator is already a comment.

direct-declarator:

       /*@iter identifier (parameter-type-list_opt) globals_opt
modifies_opt @*/

Constants (Section 8.1)

external-declaration:

       /*@constant declaration ;_opt @*/

Alternate Types (Section 8.2.2)

Alternate types may be used in the type specification of parameters and return values.

extended-type:

        type-specifier alt-type_opt

alt-type:

        /*@alt basic-type,+ @*/

Declarator Annotations

General annotations appear after storage-class-specifiers and before type-specifiers. Multiple annotations may be used in any order. Here, annotations are without the surrounding comment. In a declaration, the annotation would be surrounded by /*@ and @*/. In a globals or modifies clause or iterator or constant declaration, no surrounding comment would be used since they are within a comment.

Type Definitions (Section 3)

A type definition may use any either abstract or concrete, either mutable or immutable, and refcounted. Only a pointer to a struct may be declared with refcounted. Mutability annotations may not be used with concrete types since concrete types inherit their mutability from the actual type.

abstract

Type is abstract (representation is hidden from clients).

concrete

Type is concrete (representation is visible to clients).

immutable

Instances of the type cannot change value. (Section 3.2)

mutable

Instances of the type can change value. (Section 3.2)

refcounted

Reference counted type. (Section 5.4)

Global Variables (Section 4.2.1)

One check annotation may be used on a global or file-static variable declaration.

unchecked

Weakest checking for global use.

checkmod

Check modification by not use of global.

checked

Check use and modification of global.

checkedstrict

Check use of global, even in functions with no global list.

Memory Management (Section 5)

dependent

A reference to externally-owned storage. (Section 5.2.2)

keep

A parameter that is kept by the called function. The caller may use the storage after the call, but the called function is responsible for making sure it is deallocated. (Section 5.2.4)

killref

A refcounted parameter. This reference is killed by the call. (Section 5.4)

only

A unshared reference. Associated memory must be released before reference is lost. (Section 5.2)

owned

Storage may be shared by dependent references, but associated memory must be released before this reference is lost. (Section 5.2.2)

shared

Shared reference that is never deallocated. (Section 5.2.5)

temp

A temporary parameter. May not be released, and new aliases to it may not be created. (Section 5.2.2)

Aliasing (Section 6)

Both alias annotations may be used on a parameter declaration.

unique

Parameter that may not be aliased by any other reference visible to the function. (Section 6.1.1)

returned

Parameter that may be aliased by the return value. (Section 6.1.2)

Exposure (Section 6.2)

observer

Reference that cannot be modified. (Section 6.2.1)

exposed

Exposed reference to storage in another object. (Section 6.2.1)

Definition State (Section 7.1)

out

Storage reachable from reference need not be defined.

in

All storage reachable from reference must be defined.

partial

Partially defined. A structure may have undefined fields. No errors reported when fields are used.

reldef

Relax definition checking. No errors when reference is not defined, or when it is used.

Global State (Section 7.1.4)

These annotations may only be used in globals lists. Both annotations may be used for the same variable, to mean the variable is undefined before and after the call.

undef

Variable is undefined before the call.

killed

Variable is undefined after the call.

Null State (Section 7.2)

null

Possibly null pointer.

notnull

Non-null pointer.

relnull

Relax null checking. No errors when NULL is assigned to it, or when it is used as a non-null pointer.

Null Predicates (Section 7.2.1)

A null predicate annotation may be used of the return value of a function returning a boolean type, taking a possibly-null pointer for its first argument.

truenull

If result is TRUE, first parameter is NULL.

falsenull

If result is TRUE, first parameter is not NULL.

Execution (Section 7.3)

The exits, mayexit and neverexits annotations may be used on any function. The trueexit and falseexit annotations may only be used on functions whose first argument is a boolean.

exits

Function never returns.

mayexit

Function may or may not return.

trueexit

Function does not return if first parameter is TRUE.

falseexit

Function does not return if first parameter if FALSE.

neverexit

Function always returns.

Side-Effects (Section 8.2.1)

sef

Corresponding actual parameter has no side effects.

Declaration

These annotations can be used on a declaration to control unused or undefined error reporting.

unused

Identifier need not be used (no unused errors reported.) (Section 10.4)

external

Identifier is defined externally (no undefined error reported.) (Section 10.5)

Case

fallthrough

Fall-through case. No message is reported if the previous case may fall-through into the one immediately after the fallthrough.

Break (Section 10.2.3)

These annotations are used before a break or continue statement.

innerbreak

Break is breaking an inner loop or switch.

loopbreak

Break is breaking a loop.

switchbreak

Break is breaking a switch.

innercontinue

Continue is continuing an inner loop.

Unreachable Code

This annotation is used before a statement to prevent unreachable code errors.

notreached

Statement may be unreachable.

Special Functions (Apppendix E)

These annotations are used immediately before a function declaration.

printflike

Check variable arguments like printf library function.

scanflike

Check variable arguments like scanf library function.

Appendix E Control Comments

Error Suppression

Several comments are provided for suppressing messages. In general, it is usually better to use specific flags to suppress a particular error permanently, but the general error suppression flags may be more convenient for quickly suppressing messages for code that will be corrected or documented later.

ignore
end

No errors will be reported in code regions between /*@ignore@*/ and /*@end@*/. These comments can be used to easily suppress an unlimited number of messages, but are dangerous since if real errors are introduced in the ignore ... end region they will not be reported. The ignore and end comments must be matched - a warning is printed if the file ends in an ignore region or if ignore is used inside ignore region.
i
No errors will be reported from an /*@i@*/ comment to the end of the line.

i<n>
No errors will be reported from an /*@i<n>@*/ (e.g., /*@i3@*/) comment to the end of the line. If there are not exactly n errors suppressed from the comment point to the end of the line, LCLint will report an error. This is more robust than i or ignore since a message is generated if the expected number errors is not present. Since errors are not necessarily detected until after this file is processed (for example, and unused variable error), suppress count errors are reported after all files have been processed. The -supcounts flag may be used to suppress these errors. This is useful when a system if being rechecked with different flag settings.
t
t<n>
Like i and i<n>, except controlled by +tmpcomments flag. These can be used to temporarily suppress certain errors. Then, -tmpcomments can be set to find them again.

Type Access

Control comments may also be used to override type access settings. The syntax /*@access <type>,+@*/ allows the following code to access the representation of <type>. Similarly, /*@noaccess <type>,+@*/ restricts access to the representation of <type>. The type in a noaccess comment must have been declared as an abstract type. Type access applies from the point of the comment to the end of the file or the next access control comment for this type.

Macro Expansion

The /*@notfunction@*/indicates that the next macro definition is not intended to be a function, and should be expanded in line instead of checked as a macro function definition.

Traditional Lint Comments

Some of the control comments supported by most standard UNIX lints are supported by LCLint so legacy systems can be checked more easily. These comments are not lexically consistent with LCLint comments, and their meanings are less precise (and may vary between different lint programs), so we recommend that LCLint comments are used instead except for checking legacy systems already containing standard lint comments.

These standard lint comments supported by LCLint:

/*FALLTHROUGH*/ (alternate misspelling, /*FALLTHRU*/)

Prevents errors for fall-through cases. Same meaning as /*@fallthrough@*/.
/*NOTREACHED*/
Prevents errors about unreachable code (until the end of the function). Same meaning as /*@notreached@*/.
/*PRINTFLIKE*/
Arguments similar to the printf library function (there didn't seem to be much of a consensus among standard lints as to exactly what this means). LCLint supports:

/*@printflike@*/

Function takes zero or more arguments of any type, an unmodified char * format string argument and zero of more arguments of type and number dictated by the format string. Format codes are interpreted identically to the printf standard library function. May return a result of any type. (LCLint interprets /*PRINTFLIKE*/ as /*@printflike@*/.)

/*@scanflike@*/
Like printflike, except format codes are interpreted as in the scanf library function.
/*ARGSUSED*/
Turns off unused parameter messages for this function. The control comment, /*@-paramuse@*/ can be used to the same effect, or /*@unused@*/ can be used in individual parameter declarations.

LCLint will ignore standard lint comments if -lintcomments is used. If +warnlintcomments is used, LCLint generates a message for standard lint comments and suggest replacements,

Appendix F Libraries

Libraries can be used to record interface information. A library containing information about the Standard C Library is used to enable checking of library calls. Program libraries can be created to enable fast checking of single modules in a large program.

Standard Libraries

In order to check calls to library functions, LCLint uses an annotated standard library. This contains more information about function interfaces then is available in the system header files since it uses annotations. Further, it contains only those functions documented in the ANSI Standard. Many systems include extra functions in their system libraries; programs that use these functions cannot be compiled on other systems that do not provide them. Certain types defined by the library are treated as abstract types (e.g., a program should not rely on how the FILE type is implemented). When checking source code, LCLint does include system headers according to include directive in the source code, but instead uses the library description of the standard library.

The LCLint distribution includes several different standard libraries: the ANSI standard library, the POSIX standard library , and an ad hoc UNIX library. Each library comes in two versions: the standard version and the strict version.

ANSI Library

The default behavior of LCLint is to use the ANSI standard library (loaded from ansi.lcd). This library is based on the standard library described in the ANSI C Standard. It includes functions and types added by Amendment 1 to the ANSI C Standard.

POSIX Library

The POSIX library is selected by the +posixlib flag. The POSIX library is based on the IEEE Std 1003.1-1990.

UNIX Library

The UNIX library is selected by the +unixlib flag. This library is an ad hoc attempt to capture additional functionality provided by many UNIX platforms. Unfortunately, UNIX systems vary widely and very few are consistent with the ANSI Standard.

The differences between the UNIX library and the POSIX library are:

Code checked using the UNIX library can probably be ported to some UNIX systems without difficulty. To enhance the likelihood that a program is portable, the POSIX library should be used instead.

Strict Libraries

Stricter versions of the libraries are used if the -ansi-strict, posix-strict-lib or unix-strct-lib flag is used. These libraries use a stricter interpretation of the library. They will detect more errors in some programs, but may to produce many spurious errors for typical code.

The differences between the standard libraries and the strict libraries are:

Generating the Standard Libraries

The standard libraries are generated from header files included in the LCLint distribution. Some libraries are generated from more than one header file. Since the POSIX library includes the ANSI library, the headers for the ANSI and POSIX libraries are combined to produce the POSIX library. Similarly, the UNIX library is composed of the ANSI, POSIX and UNIX headers. The header files include some sections that are conditionally selected by defining STRICT.

The commands to generate the standard libraries are:

lclint -nolib ansi.h -dump ansi 
lclint -nolib -DSTRICT ansi.h -dump ansistrict
lclint -nolib ansi.h posix.h -dump posix
lclint -nolib -DSTRICT ansi.h posix.h -dump posixstrict
lclint -nolib ansi.h posix.h unix.h -dump unix
lclint -nolib -DSTRICT ansi.h posix.h unix.h -dump unixstrict

User Libraries

To enable running LCLint on large systems, mechanisms are provided for creating libraries containing necessary information. This means source files can be checked independently, after a library has been created. The command line option -dump library stores information in the file library (the default extension, .lcd[27], is added). Then, -load library loads the library. The library contains interface information from the files checked when the library was created.

Header File Inclusion

The standard behavior of LCLint on encountering

   #include <X.h>
is to search for a file named X.h on the include search path (set using -I) and then the system base include path (usually /usr/include, default is set when LCLint is compiled). If X.h is the name of a header file in a loaded standard library (either ANSI or POSIX) and X.h is found in a directory that is a system directory (as set by the -sysdirs flag; the default is /usr/include), X.h will not be included if skip-ansi-headers or skip-posix-headers (depending on whether X.h is an ANSI or POSIX header file) is on (both are on by default). To force all headers to be included normally, use -skip-ansi-headers and -skip-posix-headers.

Sometimes headers in system directories contain non-standard syntax which LCLint is unable to parse. The +skip-sys-headers flag may be used to prevent any include file in a system directory from being included.

LCLint is fast enough that it can be run on medium-size (10,000 line) programs without performance concerns. It takes about one second to process a thousand source lines on a DEC Alpha. Libraries can be used to enable efficient checking of small modules in large programs. To further improve performance, header file inclusion can be optimized.

When processing a complete system in which many files include the same headers, a large fraction of processing time is wasted re-reading header files unnecessarily. If you are checking a 100-file program, and every file includes utils.h, LCLint will have to process utils.h 100 times (as would most C compilers). If the +singleinclude flag is used, each header file is processed only once. Single header file processing produces a significant efficiency improvement when checking large programs split into many files, but is only safe if the same header file included in different contexts always has the same meaning (i.e., it does not depend on preprocessor variable defined differently at different inclusion sites).

When processing a single file in a large system, a large fraction of the time is spent processing included header files. This can be avoided if the information in the header files is stored in a library instead. If +neverinclude is set, inclusion of files ending in .h is prevented. Files with different suffixes are included normally. To do this the header files must not include any expanded macros. That is, the header file must be processed with +allmacros, and there must be no /*@notfunction@*/ control comments in the header. Then, the +neverinclude flag may be used to prevent inclusion of header files. Alternately, non-function macros can be moved to a different file with a name that does not end in .h. Remember, that this file must be included directly from the .c file, since if it is included from a .h file indirectly, that .h file is ignored so the other file is never included.

These options can be used for significant performance improvements on large systems. The performance depends on how the code is structured, but checking a single module in a large program is several times faster if libraries and +neverinclude are used.

Appendix G Specifications

Another way of providing more information about programs is to use formal specifications. Although this document has largely ignored specifications, LCLint was originally designed to use the information in LCL specifications instead of source-code annotations. This document focuses on annotations since it takes less effort to add annotations to source code than to maintain an additional specification file. Annotations can express everything that can be expressed in LCL specifications that is relevant to LCLint checking. However, LCL specifications can provide more precise documentation on program interfaces than is possible with LCLint annotations. This appendix (extracted from [Evans94]) is a very brief introduction to LCL Specifications. For more information, consult [GH93].

The Larch family of languages is a two-tiered approach to formal specification. A specification is built using two languages -- the Larch Shared Language (LSL), which is independent of the implementation language, and a Larch Interface Language designed for the specific implementation language. An LSL specification defines sorts, analogous to abstract types in a programming language, and operators, analogous to procedures. It expresses the underlying semantics of an abstraction.

The interface language specifies an interface to an abstraction in a particular programming language. It captures the details of the interface needed by a client using the abstraction and places constraints on both correct implementations and uses of the module. The semantics of the interface are described using primitives and sorts and operators defined in LSL specifications. Interface languages have been designed for several programming languages.

LCL [GH93, Tan94] is a Larch interface language for Standard C. LCL uses a C-like syntax. Traditionally, a C module M consists of a source file, M.c, and a header file, M.h. The header file contains prototype declarations for functions, variables and constants exported by M, as well as those macro definitions that implement exported functions or constants, and definitions of exported types. When using LCL, a module includes two additional files - M.lcl, a formal specification of M, and M.lh, which is derived by LCLint (if the lh flag is on) from M.lcl. Clients use M.lcl for documentation, and should not need to look at any implementation file. The derived file, M.lh, contains include directives (if M depends on other specified modules), prototypes of functions and declarations of variables as specified in M.lcl. The file M.h should include M.lh and retain the implementation aspects of the old M.h, but is no longer used for client documentation.

The LCLint release package includes a grammar for LCL and examples of LCL specifications.

Specification Flags

These flags are relevant only when LCLint is used with LCL specifications.

Global Flags

lcs

Generate .lcs files containing symbolic state of .lcl files (used for imports). By default .lcs files are generated for each .lcl file processed. Use -lcs to prevent generation of .lcs files.

lh

Generate .lh files. By default, -lh is set and no .lh files are generated. Use +lh to enable .lh file generation.

i <file>

Set LCL initialization file to <file>. The LCL initialization file is read if any .lcl files are listed on the command line. The default file is lclinit.lci, found on the LARCH_PATH.

lclexpect <number>

Exactly <number> specification errors are expected. Specification errors are errors detected when checking the specifications. They do not depend on the source code.

Implicit Globals Checking Qualifiers

m: -+++
impcheckedspecglobs
Implicit checked qualifier on global variables specified in an LCL file with no checking annotation.

m: ----
impcheckmodspecglobs
Implicit checkmod qualifier on global variables specified in an LCL file with no checking annotation.

m: ---+
impcheckedstrictspecglobs

Implicit checked qualifier on global variables specified in an LCL file with no checking annotation.

Implicit Annotations

plain: -
specglobimponly

Implicit only annotation on global variable declaration in an LCL file with no allocation annotation.

plain: -
specretimponly

Implicit only annotation on return value declaration in an LCL file with no allocation annotation.

plain: -
specstructimponly

Implicit only annotation on structure field declarations in an LCL file with no allocation annotation.

shortcut
specimponly

Sets specglobimponly, specretimponly and specstructimponly.

Macro Expansion

plain: +
specmacros

Macros defining specified identifiers are not expanded and are checked according to the specification.

Complete Programs and Specifications

m: -+++
specundef

Function, variable, iterator or constant specified but never defined.

plain: -
specundecl

Function, variable, iterator or constant specified but never declared.

plain: -
needspec

There is information in the specification that is not duplicated in syntactic comments. Normally, this is not an error, but it may be useful to detect it to make sure checking incomplete systems without the specifications will still use this information.

shortcut
exportany

An error is reported for any identifier that is exported but not specified. (Sets all export flags below.)

m: ---+
exportconst

Constant exported but not specified.

m: ---+
exportvar

Variable exported but not specified.

m: ---+
exportfcn

Function exported but not specified.

m: ---+
exportiter

Iterator exported but not specified.

m: ---+
exportmacro

An expanded macro exported but not specified

m: ---+
exporttype

Type definition exported but not specified

Appendix H Emacs

LCLint can be used most productively with the emacs text editor. The release package includes emacs files for running LCLint and editing code with annotations. Editing Abbreviations

An additional file, emacs/lclint-abbrevs contains abbreviations for LCLint syntactic comments and annotations. If it is loaded, the comment surrounding an LCLint annotation will be added automatically. For example, typing "only" and a space, will produce "/*@only@*/ ". Abbreviations are provided for each LCLint syntactic comment. The abbreviation of /*@null@*/ is nll (not null), since it is often necessary to type NULL.

Abbreviations are loaded and used when a .c or .h file is edited by adding these lines to your .emacs file:

(quietly-read-abbrev-file "<directory>/lclint-abbrevs")

(setq c-mode-hook (function (lambda nil (abbrev-mode 1))))

References

LCLint

[Evans94] David Evans. Using specifications to check source code. MIT/LCS/TR 628, Laboratory for Computer Science, MIT, June 1994.

SM Thesis. Describes research behind LCLint, focusing on how specifications can be exploited to do lightweight checking. Includes case studies using LCLint.

[EGHT94] David Evans, John Guttag, Jim Horning and Yang Meng Tan. LCLint: A tool for using specifications to check code. SIGSOFT Symposium on the Foundations of Software Engineering, December 1994.

Introduction to LCLint. Shows how LCLint is used to find errors in a sample program.

[Evans96] David Evans. Static Detection of Dynamic Memory Errors. In SIGPLAN Conference on Programming Language Design and Implementation (PLDI '96), Philadelphia, PA., May 1996.

Describes approach for exploiting annotations added to code to detect a wide class of errors. Focuses on checks described in Sections 5-7 of this guide.

Larch

[GH93] Guttag, John V. and Horning, James J., with Stephen J. Garland, Kevin D. Jones, Andrés Modet, and Jeannette M. Wing, Larch: Languages and Tools for Formal Specification, Springer-Verlag, Texts and Monographs in Computer Science, 1993.

Overview of the Larch family of specification languages and related tools. Includes a chapter on LCL, the Larch C interface language, on which LCLint is based.

[Tan95] Tan, Yang Meng. Formal Specification Techniques for Engineering Modular C, Kluwer International Series in Software Engineering, Volume 1, Kluwer Academic Publishers, Boston, 1995.

Modified and updated version of MIT Ph. D. dissertation, previously published as MIT/LCS/TR-619, 1994. Includes presentation of the semantics of LCL and a case study using LCL.

C

[ANSI] American National Standard for Information Systems, Programming Language, C. ANSI X3.159-1989. (Believed to be identical to ISO/IEC 9899:1990).
Specification for C programming language. LCLint aims to be consistent with this document.
[Hat95] Hatton, Les. Safer C: Developing Software for High-integrity and Safety-critical Systems. McGraw-Hill International Series in Software Engineering, 1995.
A broad work on all aspects of developing safety-critical software, focusing on the C language. Provides good justification for the use of C in safety-critical systems, and the necessity of tool-supported programming standards. LCLint users will be interested to see how many of the errors listed as only being dynamically detectable can be detected statically by LCLint.
[KR88] Kernighan, Brian W. and Ritchie, Dennis M. The C Programming Language, second edition. Prentice Hall, New Jersey, 1988.
Standard reference for ANSI C. If you haven't heard of this one, you probably didn't get this far (unless you started at the back).
[vdL94] Van der Linden, Peter. Expert C Programming: Deep C Secrets. SunSoft Press, Prentice Hall, New Jersey, 1994.
Filled with useful information on the darker corners of C, as well as lots of industry anecdotes and humor. LCLint's reserved name checking is loosely based on the list of reserved names in this book.

Abstract Types

[LG86] Liskov, Barbara. and Guttag, John V. Abstraction and Specification in Program Development, MIT Press, Cambridge, MA, 1986.
Describes a programming methodology using abstract types and specified interfaces. Much of the methodology upon which LCLint is based comes from this book. Uses the CLU programming language.
Acknowledgements

Acknowledgements

John Guttag and Jim Horning had the original idea for LCLint, have provided valuable advice on its functionality and design, and been instrumental in its development. This work has also benefited greatly from discussions with Mike Burrows, Stephen Garland, Colin Godfrey, Steve Harrison, Daniel Jackson, Angelika Leeb, Ulana Legedza, Raymie Stata, Yang Meng Tan, and Mark Vandevoorde. I especially thank Angelika Leeb for many constructive comments on improving the User's Guide, Raymie Stata for help designing and setting up the web site and Mark Vandevoorde for technical assistance.

Much of LCLint's development has been driven by feedback from users in academia and industry. Many more people than I can mention here have made contributions by suggesting improvements, reporting bugs, porting early versions of LCLint to other platforms. Particularly heroic contributions have been made by Eric Bloodworth, Jutta Degener, Rick Farnbach, Chris Flatters, Huver Hu, John Gerard Malecki, Thomas G. McWilliams, Michael Meskes, Richard O'Keefe, Jens Schweikhardt, and Albert L. Ting.

LCLint incorporates the original LCL checker developed by Yang Meng Tan. This was built on the DECspec Project (Joe Wild, Gary Feldman, Steve Garland, and Bill McKeeman). The LSL checker used by LCLint was developed by Steve Garland. The original C grammar for LCLint was provided by Nate Osgood.

This research was supported by grants from ARPA (N0014-92-J-1795), NSF (9115797-CCR) and DEC ERP. David Evans is supported by an Intel Foundation Fellowship. LCLint is developed on DEC 3000/500 and DECmips machines provided by Digital Equipment Corporation and Pentium Pro (tm) machines donated by Intel Corporation. This document was produced using Pentium (tm) and Pentium Pro (tm) Computers donated by Intel Corporation and Microsoft Office (tm) software donated by Microsoft.

If you are listed here, and want me to add a link to your homepage, send me email.

Footnotes

1 Lint is a common programming tool for detecting anomalies in C programs. The original lint was developed by S. C. Johnson in the late seventies, mainly because early versions of C did not support function prototypes.

2 Another way to provide extra information about code is to use formal specifications (Appendix G).

3 Unlike regular C comments, control comments should not be used within a single token. They may introduce new separators in the code during parsing.

4 For abstract types whose instances can change value, a client does need to know if assignment has copy or sharing semantics (see Section 3.2).

5 Does not apply to HTML version. italics.

6 The meta-notation, item,+ is used to denote a comma separated list of items. For example, /*@access mstring, intSet@*/ provides access to the representations of both mstring and intSet.)

7 Through the parameter. Modifications using some other variable that has a pointer to the location of this parameter are not considered.

8 The flag -booltype can be used to select a different name for the boolean type. To change the names of TRUE and FALSE, use -booltrue and -boolfalse. The LCLint distribution includes an implementation of bool, in lib/bool.h. However, it isn't necessary to use this implementation to get the benefits of boolean checking.

9 This means that theoreticians can prove that no algorithm exists that solves the problem correctly for all possible programs.

10 This section is largely based on [Evans96]. It semi-formally defines some of the terms needed to describe memory management checking; if you are satisfied with an intuitive understanding of these terms, this section may be skipped.

11 This is similar to the LISP storage model, except that objects are typed.

12 Except sizeof, which does not need the value of its argument.

13 If the storage is not assigned to a reference, an internal reference is created to track the storage.

14 The full declaration of malloc also includes a null annotation (Section 7.2) to indicate that the result may be NULL (as it is when the requested storage cannot be allocated) and an out annotation (Section 7.1) to indicate that the result points to undefined storage.

15 The full declaration of free also has out and null annotations on the parameter to indicate that the argument may be NULL and need not point to defined storage. According to [ANSI, 4.10.3.2], NULL may be passed to free without an error. On some UNIX platforms, passing NULL to free causes a program crash so the UNIX version of the standard library (Appendix F) specifies free without the null annotation on its parameter. To check that allocated objects are completely destroyed (e.g., all unshared objects inside a structure are deallocated before the structure is deallocated), LCLint checks that any parameter passed as an out only void * does not contain references to live, unshared objects. This makes sense, since such a parameter could not be used sensibly in any way other than deallocating its storage.

16 If an exposure qualifier is used (see Section 6.2), the implied dependent annotation is used instead of the more generally implied only annotation.

17 Strictly, we should also check that the returned observer storage is not used again after any other calls to the abstract type module using the same parameter. LCLint does not attempt to check this, and in practice it is not usually a problem.

18 Note that if the parameter is annotated with only, it is not an error to assign it to part of an abstract representation, since the caller may not use the storage after the call returns.

19 That is, the return type is bool, or int if +boolint is used.

20 The sef annotation denotes a parameter as side-effect free (see Section 8.2.1). By declaring the argument to assert to be side-effect free, LCLint will report errors if the parameter to assert produces a side-effect. This is especially pertinent if assertions are turned off when the production version is compiled. The bool /*@alt int@*/ type specifier for the parameter means the parameter type must match either bool or int. Alternate types are described in Section 8.2.2.

21 To be completely correct, all the macro parameters should be evaluated before the macro has any side-effects. Since checking this would require extensive analysis for occasional modest gain, it was not considered worth implementing.

22 Note that functions which do not produce to the same result each time they are called with the same arguments should be declared to modify internalState so they will lead to errors if they are passed as sef parameters.

23 The most renown C naming convention is the Hungarian naming convention, introduced by Charles Simonyi [Simonyi, Charles, and Martin Heller. "The Hungarian Revolution." BYTE, August 1991, p. 131-38]. The names for LCLint naming conventions follow the tradition of using Central European nationalities as mnemonics for naming conventions. The LCLint conventions are similar to the Hungarian naming convention in that they encode type information in names, except that the LCLint conventions encode the names of accessible abstract types instead of the type of the declaration of return value. Prefixes used in the Hungarian naming convention are not supported by LCLint.

24 Namespace prefixes should probably be described by regular expressions. LCLint uses a simpler, more limited means for describing names, which is believed to be adequate for describing most useful naming conventions. If there is sufficient interest, regular expressions may be supported in a future version of LCLint.

25 Peter van der Linden estimates that default fall through is the wrong behavior 97% of the time. [vdL95, p. 37]

26 "Software Glitch Cripples AT&T Network", Telephony, 22 January 1990.

27 In earlier versions of LCLint, the default extension .lldmp was used. This has been shortened to .lcd.


Contents LCLint Home Page David Evans
University of Virginia, Computer Science
evans@cs.virginia.edu