NAME

SYNOPSIS

bool
Q_SIGNED(NTYPE n);

bool
Q_LTZ(NTYPE n);

bool
Q_PRECEQ(QTYPE a, QTYPE b);

bool
Q_QLTQ(QTYPE a, QTYPE b);

bool
Q_QLEQ(QTYPE a, QTYPE b);

bool
Q_QGTQ(QTYPE a, QTYPE b);

bool
Q_QGEQ(QTYPE a, QTYPE b);

bool
Q_QEQ(QTYPE a, QTYPE b);

bool
Q_QNEQ(QTYPE a, QTYPE b);

bool
Q_OFLOW(QTYPE q, ITYPE iv);

int
Q_RELPREC(QTYPE a, QTYPE b);

DESCRIPTION

Q_LTZ() returns true if the numeric value passed in as n is negative (requires types which use the MSB as the sign bit), or false otherwise.

Q_PRECEQ() returns true if the number of a and b fractional bits is the same, false otherwise.

The Q_QLTQ(), Q_QLEQ(), Q_QGTQ(), Q_QGEQ(), Q_QEQ() and Q_QNEQ() functions compare two Q numbers, returning true if a is less than, less than or equal to, greater than, greater than or equal to, equal to, or not equal to b respectively, or false otherwise. The integral and fractional values are used to perform the comparison, without explicit concern for the underlying number of integer versus fractional bits.

Q_OFLOW() returns true if integer value iv cannot be stored in q without truncation, or false otherwise.

Q_RELPREC() returns the relative precision of a versus b. In terms of Qm.n notation, this function returns the difference between the n values of a and b. For example, a return value of +4 means that a has an additional 4 bits of fractional precision compared to b.

All of those functions operate on the following data types: s8q_t, u8q_t, s16q_t, u16q_t, s32q_t, u32q_t, s64q_t, and u64q_t, which are referred to generically as QTYPE. The ITYPE refers to the stdint(7) integer types. NTYPE is used to refer to any numeric type and is therefore a superset of QTYPE and ITYPE.

For more details, see qmath(3).

RETURN VALUES

Q_RELPREC() returns the relative precision difference as a signed integer.

SEE ALSO

HISTORY

AUTHORS