Analog Devices ADSP-2106x SHARC User Manual page 551

Compute Operations
Syntax:
Rn = MRF – Rx * Ry mod2
Rn = MRB – Rx * Ry mod2
MRF = MRF – Rx * Ry mod2
MRB = MRB – Rx * Ry mod2
Function:
Multiplies the fixed-point fields in registers Rx and Ry, and subtracts the
product from the specified MR register value. If rounding is specified
(fractional data only), the result is rounded. The result is placed either in
the fixed-point field in register Rn or one of the MR accumulation
registers, which must be the same MR register that provided the input. If
Rn is specified, only the portion of the result that has the same format as
the inputs is transferred (bits 31-0 for integers, bits 63-32 for fractional).
The floating-point extension field in Rn is set to all 0s. If MRF or MRB is
specified, the entire 80-bit result is placed in MRF or MRB.
Status flags:
MN Is set if the result is negative, otherwise cleared
MV Is set if the upper bits are not all zeros (signed or unsigned result) or
ones (signed result). Number of upper bits depends on format. For a
signed result, fractional=33, integer=49. For an unsigned result,
fractional=32, integer=48.
MU Is set if the upper 48 bits of a fractional result are all zeros (signed or
unsigned result) or ones (signed result) and the lower 32 bits are not
all zeros. Integer results do not underflow.
MI Is cleared
www.BDTIC.com/ADI
Multiplier Fixed-Point
Rn|MR = MR – Rx * Ry
B
B – 49