SPRUIW9C October 2021 – March 2024 TMS320F280033 , TMS320F280034 , TMS320F280034-Q1 , TMS320F280036-Q1 , TMS320F280036C-Q1 , TMS320F280037 , TMS320F280037-Q1 , TMS320F280037C , TMS320F280037C-Q1 , TMS320F280038-Q1 , TMS320F280038C-Q1 , TMS320F280039 , TMS320F280039-Q1 , TMS320F280039C , TMS320F280039C-Q1
32-Bit Floating-Point Compare for Equal, Less Than or Greater Than
MRa | CLA floating-point source register (MR0 to MR3) |
#16FHi | A 16-bit immediate value that represents the upper 16-bits of an IEEE 32-bit floating-point value. The low 16-bits of the mantissa are assumed to be all 0. |
LSW: IIII IIII IIII IIII
MSW: 0111 1000 1100 00aa
Compare the value in MRa with the floating-point value represented by the immediate operand. Set the ZF and NF flags on (MRa - #16FHi:0).
#16FHi is a 16-bit immediate value that represents the upper 16-bits of an IEEE 32-bit floating-point value. The low 16-bits of the mantissa are assumed to be all 0. This addressing mode is most useful for constants where the lowest 16-bits of the mantissa are 0. Some examples are 2.0 (0x40000000), 4.0 (0x40800000), 0.5 (0x3F000000), and -1.5 (0xBFC00000). The assembler accepts either a hex or float as the immediate value. That is, -1.5 can be represented as #-1.5 or #0xBFC0.
The MCMPF32 instruction is performed as a logical compare operation. This is possible because of the IEEE floating-point format offsets the exponent. Basically the bigger the binary number, the bigger the floating-point value.
Special cases for inputs:
This instruction modifies the following flags in the MSTF register:
Flag | TF | ZF | NF | LUF | LVF |
---|---|---|---|---|---|
Modified | No | Yes | Yes | No | No |
The MSTF register flags are modified as follows:
If(MRa == #16FHi:0) {ZF=1, NF=0;}
If(MRa > #16FHi:0) {ZF=0, NF=0;}
If(MRa < #16FHi:0) {ZF=0, NF=1;}
This is a single-cycle instruction
; Behavior of ZF and NF flags for different comparisons
MMOVIZ MR1, #-2.0 ; MR1 = -2.0 (0xC0000000)
MMOVIZ MR0, #5.0 ; MR0 = 5.0 (0x40A00000)
MCMPF32 MR1, #-2.2 ; ZF = 0, NF = 0
MCMPF32 MR0, #6.5 ; ZF = 0, NF = 1
MCMPF32 MR0, #5.0 ; ZF = 1, NF = 0
; X is an array of 32-bit floating-point values
; and has length elements. Find the maximum value in
; the array and store the value in Result
;
; Note: MCMPF32 and MSWAPF can be replaced with MMAXF32
;
_Cla1Task1:
MMOVI16 MAR1,#_X ; Start address
MUI16TOF32 MR0, @_len ; Length of the array
MNOP ; delay for MAR1 load
MNOP ; delay for MAR1 load
MMOV32 MR1, *MAR1[2]++ ; MR1 = X0
LOOP
MMOV32 MR2, *MAR1[2]++ ; MR2 = next element
MCMPF32 MR2, MR1 ; Compare MR2 with MR1
MSWAPF MR1, MR2, GT ; MR1 = MAX(MR1, MR2)
MADDF32 MR0, MR0, #-1.0 ; Decrememt the counter
MCMPF32 MR0 #0.0 ; Set/clear flags for MBCNDD
MNOP
MNOP
MNOP
MBCNDD LOOP, NEQ ; Branch if not equal to zero
MMOV32 @_Result, MR1 ; Always executed
MNOP ; Always executed
MNOP ; Always executed
MSTOP ; End of task