Renesas M32R-FPU Benutzerhandbuch

Seite von 192
3
3-38
M32R-FPU Software Manual (Rev.1.01)
INSTRUCTIONS
3.2 Instruction description
FCMP
FCMP
src1
1101
src2
0000
dest
0000
0000
1100
FCMP  Rdest,Rsrc1,Rsrc2
[Mnemonic]
FCMP  Rdest,Rsrc1,Rsrc2
[Function]
Floating-point compare
Rdest = (comparison results of Rsrc1 and Rsrc2);
When at least one value, either Rsrc1 or Rsrc2, is SNaN, a floating-point exception (other than
Invalid Operation Exception) occurs.
[Description]
Compare the floating-point single precision values stored in Rsrc1 and Rsrc2 and store the
result in Rdest. The results of the comparison can be determined y the following methods.
The DN bit of FPSR handles the conversion of denormalized numbers. The condition bit (C)
remains unchanged.
[EIT occurrence]
Floating-Point Exceptions (FPE)
• Unimplemented Operation Exception (UIPL)
• Invalid Operation Exception (IVLD)
[Encoding]
floating point Instructions
Floating-point compare
[M32R-FPU Extended Instruction]
Rdest
Comparison Results
Typical instructions used to 
determine comparison results
b0=0
   All bits, b1 to b31, are 0.
   Rsrc1=Rsrc2
   beqz   Rdest, LABEL
   bgtz   Rdest, LABEL
   bltz   Rdest, LABEL
   Rsrc1>Rsrc2
   Rsrc1<Rsrc2
   Comparison invalid
   Bits b1 to b31 are an undefined. 
   All others
   b1 to b9=111 1111 11,
   Bits b10 to b31 are an undefined.
b0=1