|Kernel Functions for Drivers||vsprintf(9F)
| ||vsprintf - format characters in
char *vsprintf(char *buf, const char *fmt, va_list ap);
Solaris DDI specific (Solaris DDI).
- Pointer to a character string.
- Pointer to a character string.
- Pointer to a variable argument list.
vsprintf() builds a string in buf
under the control of the format fmt. The format is
a character string with either plain characters, which are simply copied
into buf, or conversion specifications, each of which
converts zero or more arguments, again copied into buf.
The results are unpredictable if there are insufficient arguments for the
format; excess arguments are simply ignored. It is the user's responsibility
to ensure that enough storage is available for buf.
ap contains the list of arguments used by the
conversion specifications in fmt. ap
is a variable argument list and must be initialized by calling va_start(9F). va_end(9F) is used to clean up and
must be called after each traversal of the list. Multiple traversals of
the argument list, each bracketed by va_start(9F)
and va_end(9F), are possible.
Each conversion specification is introduced by the %
character, after which the following appear in sequence:
An optional decimal digit specifying a minimum field width for numeric
conversion. The converted value will be right-justified and padded with
leading zeroes if it has fewer characters than the minimum.
An optional l (ll) specifying
that a following d, D, o, O, x, X,
or u conversion character applies to a long (long long) integer argument. An l (ll) before any other conversion character
A character indicating the type of conversion to be applied:
- The integer argument
is converted to signed decimal (d, D),
unsigned octal (o, O), unsigned hexadecimal
(x, X) or unsigned decimal (u), respectively, and copied. The letters abcdef
are used for x conversion. The letters ABCDEF are used for X conversion.
- The character
value of the argument is copied.
- This conversion
uses two additional arguments. The first is an integer, and is converted
according to the base specified in the second argument. The second argument
is a character string in the form <base>[<arg>...]. The base supplies the conversion
base for the first argument as a binary value; \10 gives octal, \20
gives hexadecimal. Each subsequent <arg> is a sequence of characters,
the first of which is the bit number to be tested, and subsequent characters,
up to the next bit number or terminating null, supply the name of the bit.
- A bit number is a binary-valued
character in the range 1-32. For each bit set in the
first argument, and named in the second argument, the bit names are copied,
separated by commas, and bracketed by < and >. Thus, the following function call would generate reg=3<BitTwo,BitOne>\n in buf.
- The argument
is taken to be a string (character pointer), and characters from the string
are copied until a null character is encountered. If the character pointer
is NULL on SPARC, the
string <nullstring> is used in its place; on IA,
it is undefined.
- Copy a %; no argument is converted.
vsprintf() returns its first parameter, buf.
vsprintf() can be called from user, kernel, or
| ||Example 1. Using vsprintf
In this example, xxerror() accepts a pointer to
a dev_info_t structure dip, an error
level level, a format fmt, and a
variable number of arguments. The routine uses vsprintf()
to format the error message in buf. Note that va_start(9F) and va_end(9F)
bracket the call to vsprintf(). instance, level, name, and buf are
then passed to cmn_err(9F).
#define MAX_MSG 256
xxerror(dev_info_t *dip, int level, const char *fmt, ...)
instance = ddi_get_instance(dip);
name = ddi_binding_name(dip);
/* format buf using fmt and arguments contained in ap */
vsprintf(buf, fmt, ap);
/* pass formatted string to cmn_err(9F) */
cmn_err(level, "%s%d: %s", name, instance, buf);