amoadd.h.aqrl
Atomic fetch-and-add halfword (acquire-release)
This instruction is defined by:
Synopsis
Atomically with acquire and release ordering:
-
Load the halfword at address xs1
-
Write the sign-extended value into xd
-
Add the least-significant halfword of register xs2 to the loaded value
-
Write the result to the address in xs1
Decode Variables
Bits<5> xs2 = $encoding[24:20];
Bits<5> xs1 = $encoding[19:15];
Bits<5> xd = $encoding[11:7];Execution
-
Pruned, XLEN == 64
-
Original
memory_model_acquire();
XReg virtual_address = X[xs1];
X[xd] = amo(16, virtual_address, X[xs2][15:0], AmoOperation::Add, 1'b1, 1'b1, $encoding);
memory_model_release();if (implemented?(ExtensionName::A) && MISA_CSR_IMPLEMENTED && CSR[misa].A == 1'b0) {
reserved_instruction($encoding);
}
memory_model_acquire();
XReg virtual_address = X[xs1];
X[xd] = amo(16, virtual_address, X[xs2][15:0], AmoOperation::Add, 1'b1, 1'b1, $encoding);
memory_model_release();