SPRUJ53B April 2024 – September 2024 TMS320F28P550SJ , TMS320F28P559SJ-Q1
Convert 32-Bit Integer to 32-Bit Floating-Point Value
MRa | CLA floating-point destination register (MR0 to MR3) |
mem32 | 32-bit memory source for the MMOV32 operation. |
LSW: mmmm mmmm mmmm mmmm
MSW: 0111 0100 01aa addr
Convert the 32-bit signed integer indicated by mem32 to a 32-bit floating-point value and store the result in MRa.
MRa = MI32TOF32[mem32];
This instruction does not affect any flags:
Flag | TF | ZF | NF | LUF | LVF |
---|---|---|---|---|---|
Modified | No | No | No | No | No |
This is a single-cycle instruction.
; Given X, M and B are IQ24 numbers:
; X = IQ24(+2.5) = 0x02800000
; M = IQ24(+1.5) = 0x01800000
; B = IQ24(-0.5) = 0xFF800000
;
; Calculate Y = X * M + B
;
; Convert M, X, and B from IQ24 to float
;
_Cla1Task3:
MI32TOF32 MR0, @_M ; MR0 = 0x4BC00000
MI32TOF32 MR1, @_X ; MR1 = 0x4C200000
MI32TOF32 MR2, @_B ; MR2 = 0xCB000000
MMPYF32 MR0, MR0, #0x3380 ; M = 1/(1*2^24) * iqm = 1.5 (0x3FC00000)
MMPYF32 MR1, MR1, #0x3380 ; X = 1/(1*2^24) * iqx = 2.5 (0x40200000)
MMPYF32 MR2, MR2, #0x3380 ; B = 1/(1*2^24) * iqb = -.5 (0xBF000000)
MMPYF32 MR3, MR0, MR1 ; M*X
MADDF32 MR2, MR2, MR3 ; Y=MX+B = 3.25 (0x40500000)
; Convert Y from float32 to IQ24
MMPYF32 MR2, MR2, #0x4B80 ; Y * 1*2^24
MF32TOI32 MR2, MR2 ; IQ24(Y) = 0x03400000
MMOV32 @_Y, MR2 ; store result
MSTOP ; end of task