fsw
Floating-Point Store Single-Precision
The fsw instruction stores a single-precision floating-point value in fs2 to memory at address xs1 + imm.
It does not modify the bits being transferred; in particular, the payloads of non-canonical NaNs are preserved.
Decode Variables
signed Bits<12> imm = sext({$encoding[31:25], $encoding[11:7]}, 12);
Bits<5> fs2 = $encoding[24:20];
Bits<5> xs1 = $encoding[19:15];
Execution
check_f_ok($encoding);
XReg virtual_address = X[xs1] + $signed(imm);
write_memory(32, virtual_address, F[fs2][31:0], $encoding);
Exceptions
This instruction may result in the following synchronous exceptions:
-
IllegalInstruction
-
LoadAccessFault
-
StoreAmoAccessFault
-
StoreAmoAddressMisaligned
-
StoreAmoPageFault