Hardware and Computer Organization- P12:Today, we often take for granted the impressive array of computing machinery that surrounds us and helps us manage our daily lives. Because you are studying computer architecture and digital hardware, you no doubt have a good understanding of these machines, and you’ve probably written countless programs on your PCs and workstations. | Chapter 11 value in that field may produce unpredictable Another feature of the compare instructions is that they will always set the flags so the state of the S bit is ignored. The next group of data processing instructions is the very powerful set of multiplication instructions. There are size multiplication instructions as shown in the following table Mnemonic Description Syntax MUL Multiply two 32-bit numbers produce a 32-bit result Rd Rm Rs MUL cond S Rd Rm Rn MLA Multiply two 32-bit numbers and add 3rd number for a 32-bit result Rd Rn Rm Rs MLA cond S Rd Rm Rn Rs UMULL Multiply two unsigned 32-bit numbers produce an unsigned 64-bit resulted in two registers RdHi RdLo Rm Rs UMULL cond S RdLo RdHi Rm Rs UMLAL Multiply two unsigned 32-bit numbers and add an unsigned 64-bit number in two registers to produce an unsigned 64-bit resulted in two registers RdHi RdLo RdHi RdLo Rm Rs UMLAL cond S RdLo RdHi Rm Rs SMULL Multiply two signed 32-bit numbers produce a signed 64-bit result in two registers SMULL cond S RdLo RdHi Rm Rs SMLAL Multiply two signed 32-bit numbers and add a signed 64-bit number in two registers to produce a signed 64-bit resulted in two registers RdHi RdLo RdHi RdLo Rm Rs SMLAL cond S RdLo RdHi Rm Rs As a class of instructions the multiple instructions also take longer than one cycle to execute. Finally it may surprise you that the ARM instruction set does not contain any division instructions. Sloss et al7 describe approximation methods that may be used to convert division operations to multiplications. 2. Load Store Instructions All data transfers between registers and memory use the load and store class of instructions. All memory addresses are generated using a base register pointer summed with an additional immediate offset value register values or scaled register values. In addition the calculated memory address pointer may be used without updating the base register pointer with the new address value. Finally the address calculation
