Unit 08 Advanced Microprocessor

Transcription

1 Unit 08 Advanced Microprocessor 1. Features of The microprocessor is an enhanced version of the microprocessor Memory-management unit is enhanced to provide memory paging. The also includes 32-bit extended registers and a 32-bit address and data bus. These extended registers include EAX, EBX, ECX, EDX, EBP, ESP, EDI, ESI, EIP and EFLAGS. The has a physical memory size of 4GBytes that can be addressed as a virtual memory with up to 64TBytes. The is operated in the pipelined mode, it sends the address of the next instruction or memory data to the memory system prior to completing the execution of the current instruction This allows the memory system to begin fetching the next instruction or data before the current is completed. This increases access time. The instruction set of the is enhanced to include instructions that address the 32-bit extended register set. The memory manager is similar to the 80286, except the physical addresses generated by the MMU are 32 bits wide instead of 24-bits. The concept of paging is introduced in support three operating modes: 1) Real Mode (default) 2) Protected Virtual Address Mode (PVAM) 3) Virtual Mode The memory management section of supports virtual memory, paging and four levels of protection. The includes special hardware for task switching Architecture The internal architecture of the includes six functional units that operate in parallel. The parallel operation is called as pipeline processing. Fetching, decoding execution, memory management, and bus access for several instructions are performed simultaneously. The six functional units of the are 1) Bus Interface Unit 2) Code Pre-fetch Unit 3) Instruction Decoder Unit 4) Execution Unit 5) Segmentation Unit 6) Paging Unit 1

2 Unit 08 Advanced Microprocessor Figure: Architecture The Bus Interface Unit connects the with memory and I/O. Based on internal requests for fetching instructions and transferring data from the code pre-fetch unit, the generates the address, data and control signals for the current bus cycles. The code pre-fetch unit pre-fetches instructions when the bus interface unit is not executing the bus cycles. It then stores them in a 16-byte instruction queue for decoding by the instruction decode unit. The instruction decode unit translates instructions from the pre-fetch queue into microcodes. The decoded instructions are then stored in an instruction queue (FIFO) for processing by the execution unit. The execution unit processes the instructions from the instruction queue. It contains a control unit, a data unit and a protection test unit. The control unit contains microcode and parallel hardware for fast multiply, divide and effective address calculation. The unit includes a 32-bit ALU, 8 general purpose registers and a 64-bit barrel shifter for performing multiple bit shifts in one clock. The data unit carries out data operations requested by the control unit. The protection test unit checks for segmentation violations under the control of microcode. The segmentation unit calculates and translates the logical address into linear addresses at the request of the execution unit. The translated linear address is sent to the paging unit. Upon enabling the paging mechanism, the translates these linear addresses into physical addresses. 2

3 Unit 08 Advanced Microprocessor If paging is not enabled, the physical address is identical to the linear address and no translation is necessary. 3. Register organization of The Register organization of is as follows: Figure:80386 General Purpose, Index and Pointer Register 3.1. General Purpose Register Registers EAX, EBX, ECX, EDX, EBP, EDI and ESI are regarded as general purpose or multipurpose registers. EAX (ACCUMULATOR):The accumulator is used for instructions such as multiplication, division and some of the adjustment instructions. In and above, the EAX register may also hold the offset address of a location in memory system. EBX (BASE INDEX): This can hold the offset address of a location in the memory system in all version of the microprocessor. It the and above EBX also can address memory data. ECX (count): This acts as a counter for various instructions. EDX (data): EDX is a general-purpose registers that holds a part of the result for multiplication or part of the division. In the and above this register can also address memory data Pointer and Index Register EBP (Base Pointer): EBP points to a memory location in all version of the microprocessor for memory data transfers. 3

4 Unit 08 Advanced Microprocessor ESP (Stack Pointer): ESP addresses an area of memory called the stack. The stack memory is a data LIFO data structure. The register is referred to as SP if used in 16 bit mode and ESP if referred to as a 32 bit register. EDI (Destination index): EDI often addresses string destination data for the string instruction. It also functions as either a 32-bit (EDI) or 16-bit (DI) general-purpose register. ESI (Source index): ESI can either be used as ESI or SI. It is often used to the address source string data for the string instructions. Like EDI ESI also functions as a generalpurpose registers. Figure:80386 Segment Register CS (Code): The code segment is a section of memory that holds the code used by the microprocessor. The code segment registers defines the starting address of the section of memory holding code. SS (Stack): The stack segment defines the area of memory used for the stack. The stack entry point is determined by the stack segment and stack pointer registers. The BP registers also addresses data within the stack segment. DS (Data): The data section contains most data used by a program. Data are accessed in the data segment by an offset address of the contests of other registers that hold the offset address. ES (extra): The extra segment is used to hold information about string transfer and manipulation FS and GS: These are supplementing segment registers available in the and above microprocessors to allow two additional memory segments for access by programs. EIP (Instruction Pointer): EIP addresses the next instruction in a section of memory defined as a code segment. This register is IP (16bit) when microprocessor operates in the real mode and EIP (32 bits) when and above operate in protected mode 4

5 3.3. Flag Register: Unit 08 Advanced Microprocessor Figure: Instruction Pointer Indicates the condition of the microprocessor and controls its operations. Flag registers are also upward compatible since the have 16bit registers and the and above have EGLAF register (32 bits) Figure: Flag Register IOPL (I/O Privilege level): IOPL is used in protected mode operation to select the privilege level for I./O devices. IF the current privilege level is higher or more trusted than the IOPL, I/O executed without hindrance. If the IOPL is lover than the current privilege level, an interrupt occurs, causing execution to suspend. Note that an IPOL is 00 is the highest or more trusted; if IOPL is 11, it s the lowest or least trusted. NT (Nested Task): The nested task flag is used to indicate that the current task is nested within another task in protected mode operation. This flag is when the task I nested by software. RF (Resume): The resume flag is used with debugging to control the resumption of execution after the next instruction. VM (Virtual Mode): The VM flag bit selects virtual mode operation in a protected mode system. Note: All the other flag bit is having similar description as in 8086 flag register System Address Register: Four memory management registers are used to specify the locations of data structures which control segmented memory management. 5

6 Unit 08 Advanced Microprocessor GDTR (Global Descriptor Table Register) and IDTR (Interrupt Descriptor Table Register) be loaded with instructions which get a 6 byte data item from memory LDTR (Local Descriptor Table Register) and TR (Task Register) can be loaded with instructions which take a 16-bit segment selector as an operand Special Register Control Register: Four Control Register (CR0-CR3) Debug Register: Eight Debug Register (DR0-DR7) Test Register: Two Test Register (TR6-TR7) Control Register in detail. The has three 32 bit control registers CR0, CR2 and CR3 to hold global machine status independent of the executed task. Load and store instructions are available to access these registers Common tasks performed by control registers include interrupt control, switching the addressing mode, paging control, and coprocessor control. CR0: Figure: Control Register CR0 contains system control flags, which control or indicate conditions that apply to the system as a whole, not to an individual task. PE (Protection Enable, bit 0): Setting PE causes the processor to begin executing in protected mode. Resetting PE returns to real-address mode. 6

7 Unit 08 Advanced Microprocessor CR2: CR3: MP (Math Present, bit 1): MP controls the function of the WAIT instruction, which is used to coordinate a coprocessor. EM (Emulation, bit 2):EM indicates whether coprocessor functions are to be emulated. TS (Task Switched, bit 3):The processor sets TS with every task switch and tests TS when interpreting coprocessor instructions. ET (Extension Type, bit 4):ET indicates the type of coprocessor present in the system (80287 or 80387). PG (Paging, bit 31): PG indicates whether the processor uses page tables to translate linear addresses into physical addresses. CR2 is used for handling page faults when PG is set. The processor stores in CR2 the linear address that triggers the fault. Contains a value called Page Fault Linear Address (PFLA). When a page fault occurs, the address the program attempted to access is stored in the CR2 register. Used when virtual addressing is enabled, hence when the PG bit is set in CR0, CR3 enables the processor to translate linear addresses into physical addresses by locating the page directory and page tables for the current task. Typically, the upper 20 bits of CR3 become the page directory base register (PDBR), which stores the physical address of the first page directory entry. Note: Reserved field shown in figure are kept reserved by manufacturer Debug Register Debugging of allows data access breakpoints as well as code execution breakpoints contains 6 debug registers to specify 1) 4 breakpoints 2) Breakpoint Control options 3) Breakpoint Status 7

8 Unit 08 Advanced Microprocessor Linear Breakpoint Address Registers: Figure: Debug Register The breakpoint addresses specified are 32-bit linear addresses While debugging, Intel 386 h/w continuously compares the linear breakpoint addresses in DR0-DR3 with the linear addresses generated by executing software. Debug Control Register: Figure: Debug Control Register LENi(i=0-3): Breakpoint Length Specification Bits: 1) 2 bit field for each breakpoint 2) Specifies length of breakpoint fields 3) The choices of data breakpoints are 1byte, 2bytes & 4bytes 4) For instruction execution breakpoint, the length is 1(beginning byte address) 8

9 Unit 08 Advanced Microprocessor Table: LENi Encoding RWi(i=0-3): Memory Access Qualifier Bit 1) 2 bit field for each breakpoint 2) Specifies the type of usage which must occur inorder to activate the associated breakpoint GD: Global Debug Register Access Detect Table: Table: RWii Encoding Debug registers can only be accessed in real mode or at privilege level 0 in protected mode 9

10 Unit 08 Advanced Microprocessor GD bit, when set, provides extra protection against any Debug Register access even in Real Mode or at privilege level 0 in Protected Mode. The GD bit, when set, causes an exception 1 fault if an instruction attempts to read or write any Debug Register. The GD bit is then automatically cleared when the exception 1 handler is invoked, allowing the exception 1 handler free access to the debug registers. GE and LE bit: Exact data breakpoint match, global and local Gi and Li(i=0-3): Breakpoint Enable, global and local If either Gi or Li is set then the associated breakpoint is enabled. Debug Status Register: A Debug Status Register allows the exception 1 handler to easily determine why it was invoked. It can be invoked as a result of one of several events: 1) DR0 Breakpoint fault/trap. 2) DR1 Breakpoint fault/trap. 3) DR2 Breakpoint fault/trap. 4) DR3 Breakpoint fault/trap. 5) Single-step (TF) trap. 6) Task switch trap. 7) Fault due to attempted debug register access when GD = 1. Bi : Debug fault/trap due to breakpoint 0-3 Figure: Debug Status Register Four breakpoint indicator flags, B0-B3, correspond one-to-one with the breakpoint registers in DR0-DR3. A flag Bi is set when the condition described by DRi, LENi, and RWi occurs. BD: Debug fault due to attempted register access when GD bit is set This bit is set if the exception 1 handler was invoked due to an instruction attempting to read or write to the debug registers when GD bit was set. BS: Debug trap due to single step This bit is set if the exception 1 handler was invoked due to the TF bit in the flag register being set BT: Debug trap due to task switch 10

11 Unit 08 Advanced Microprocessor This bit is set if the exception 1 handler was invoked due to a task switch occurring to a task having an Intel Test Register Two more test register are provided by for page caching They are used to control the testing of Translation Look-aside Buffer(TLB) of Intel386. TR6 is the command test register TR7 is the data register which contains the data of Translation Look-aside buffer test. Figure: Test Register 3.6. Program-Invisible registers The global and local descriptor tables are found in the memory system. In order to access and specify the address of these tables, the program invisible registers used. The program invisible registers are not directly addressed by software so they are given name. The GDTR (global descriptor table register) and IDTR (interrupt descriptor table register) contain the base addresses of the descriptor table and its limit. The limit of each descriptor table is 16 bits because the maximum table length is 64 Kbytes. When the protected mode operation is desired, the address of the global descriptor table and its limit are loaded into the GDTR. 4. Real Modes of Default Mode After reset, the starts from the memory location FFFFFFF0 H under real address mode. In real address mode, works as a fast 8086 with 32 bit registers and data types. Real-address mode is in effect after a signal on the RESET pin. Even if the system is going to be used in protected mode, the start-up program will execute in real-address mode temporarily while initializing for protected mode. The addressing techniques, memory size, interrupt handling in this mode of are similar to the real addressing mode of In real address mode, the default operand size is 16 bit but 32 bit operands and addressing modes may be used with the help of override prefixed. Maximum physical memory = 1Mega byte (1MB) The only way to leave real-address mode is to switch to protected mode. 5. PVAM of

12 Unit 08 Advanced Microprocessor Introduced in the processor. 32-bit address bus => access up to 232 bytes = B = 4 GB Base address => 32-bit value Offset =>16-bit or 32-bit value Linear address = base address + offset Linear address physical address with paging In protected mode, the segment registers contain an index into a table of segment descriptors. Each segment descriptor contains the start address of the segment, to which the offset is added to generate the address. In addition, the segment descriptor contains memory protection information. This includes an offset limit and bits for write and read permission. This allows the processor to prevent memory accesses to certain data. Protected mode is accessed by placing a logic 1 into the PE bit of CR0 This system contains one data segment descriptor and one code segment descriptor with each segment set to 4G bytes in length. PVAM mode support memory management, virtual memory, multitasking, protection, debugging, segmentation and paging. 6. Virtual Mode of In its protected mode of operation, 80386DX provides a virtual 8086 operating environment to execute the 8086 programs. The real mode can also use to execute the 8086 programs along with the capabilities of 80386, like protection and a few additional instructions. Once the enters the protected mode from the real mode, it cannot return back to the real mode without a reset operation. Thus, the virtual 8086 mode of operation of 80386, offers an advantage of executing 8086 programs while in protected mode. The address forming mechanism in virtual 8086 mode is exactly identical with that of 8086 real mode. In virtual mode, 8086 can address 1Mbytes of physical memory that may be anywhere in the 4Gbytes address space of the protected mode of Like real mode, the addresses in virtual 8086 mode lie within 1Mbytes of memory. In virtual mode, the paging mechanism and protection capabilities are available at the service of the programmers. The supports multiprogramming, hence more than one programmer may be using the CPU at a time. 12

13 Unit 08 Advanced Microprocessor Paging unit may not be necessarily enable in virtual mode, but may be needed to run the 8086 programs which require more than 1Mbyts of memory for memory management function. In virtual mode, the paging unit allows only 256 pages, each of 4Kbytes size. Each of the pages may be located anywhere in the maximum 4Gbytes physical memory. The virtual mode allows the multiprogramming of 8086 applications. The virtual 8086 mode executes all the programs at privilege level 3. Any of the other programmers may deny access to the virtual mode programs or data. Even in the virtual mode, all the interrupts and exceptions are handled by the protected mode interrupt handler. To return to the protected mode from the virtual mode, any interrupt or execution may be used. As a part of interrupt service routine, the VM bit may be reset to zero to pull back the into protected mode. 7. Segment in In protected mode the contents of segment register is called selectors shown below:- Figure: Segment Selector Descriptor Index: Selects any one of the 8192(213) descriptor from a descriptor table TI Table Indicator TI=1 Local Descriptor Table TI=0 Global Descriptor Table RPL (Requested Privilege Level)(2-bit): Refers the privilege of that segment. They are used while protection checks to indicate whether access to segment is allowed or not allowed Descriptor Need of Descriptor Segments are the memory area defined by programmer and can be code, stack and data segment. 13

14 Unit 08 Advanced Microprocessor segments are also assigned to have attributes viz. privilege level, segment type, segment limit. So, it s not possible to use a 16 bit segment register to represent all the info regarding a segment. therefore the Solution is Descriptor Segment sizes vary from 1byte to 4GB in Figure: Segment Descriptor in Protected Mode Offset Bit Description 40 A(Accessed) sets this bit to 1 whenever a memory reference to a segment defined by this descriptor is made TYPE 3-bit field describes the type of segment 000 Data segment, read only 001 Data segment, read/write 010 Stack segment, read only 011 Stack segment, read/write 100 Code segment, execute-only, non-conforming 101 Code segment, execute/read, non-conforming 110 Code segment, execute-only, conforming 111 Code segment, execute/read, conforming 14

15 Unit 08 Advanced Microprocessor 44 S(System) 0 Specifies system segment descriptor E.g. LDS, TSS, Gates 1 Specifies non system segment descriptor E.g. code, stack or data DPL (Descriptor Privilege level) Indicates privilege level associated with memory space. 0 - most privileged 3 least privileged 47 Present If set 0, indicates that the address range that is specified by the descriptor is temporarily not present. 52 U(User) Undefined bit. 53 X Reserved by Intel 54 D(Default Size) 0 Operands within this segments are assumed to be 16 bits. 1 Operands assumed to be 32 bits 55 G(Granularity bit) 0 Unit of limit field is 1 byte. 1 Unit of limit field is of 4096 bytes (or 1 page). The Above figure is the format of the code/data descriptor; one descriptor is 64-bit long. As we can see, a descriptor actually includes a 32-bit base address and a 20-bit limit and some attributes, the 32-bit base address indicate where the segment starts, and the 20-bit limit indicates the length of the segment. However, a problem comes up, 20-bit limit can only represent 2^20 = 1MB memory, to access a 4GB memory space, descriptor uses G bit to indicate whether the limit use 4K or 1 byte for one unit, that means if G bit is set then we get 2^20*4K = 4GB memory, if it is unset then we only use a memory space under 1MB Descriptor tables in The segment descriptors that we defined are grouped together and placed in a continuous memory location. The group arrangement called Descriptor Table. Each descriptor requires 8byte in order to store the info of a particular segment. Descriptor table can contain 8192(8K) descriptors at the max. 15

16 Unit 08 Advanced Microprocessor The maximum length of a descriptor table is a 64Kbytes. The descriptor tables define all the segments used in the when it operates in the protected mode. There are three types of descriptor tables: the global descriptor table (GDT), the local descriptor table (LDT), and the interrupt descriptor table (IDT). The registers used by the 80386to address these three tables are called the global descriptor table register (GDTR), the local descriptor table register (LDTR), and the interrupt descriptor table register (IDTR). These registers are loaded with the LGDT, LLDT, and LIDT instructions, respectively. The local and global descriptor tables hold up to 8192 entries each, and the interrupt descriptor table holds up to 256 entries. A descriptor is indexed from either the local or global descriptor table by the selector that appears in a segment register. Whenever a new selector is placed into one of the segment registers, the accesses one of the descriptor tables and automatically loads the descriptor into a program-invisible cache portion of the segment register. 1) Global Descriptor Table (GDT): This is the main table of descriptors. The same GDT can be used by all programs to refer to segment of memory can have many LDT s but only one GDT. Figure: Global Descriptor Table 2) Local Descriptor Table (LDT) : A multi-tasking system is defined on a per task basis. The main purpose of an LDT would be combined with GDT in order to expand the total number of available descriptors. Generally, each task can have its own LDT and can also be shared with other task. 8. Paging 16

17 Unit 08 Advanced Microprocessor Paging is one of the memory management techniques The paging mechanism allows any linear (logical) address, as it is generated by a program, to be placed into any physical memory page, as generated by the paging mechanism. A linear memory page is a page that is addressed with a selector and an offset in either the real or protected mode. A physical memory page is a page that exists at some actual physical memory location. For example, linear memory location 20000H could be mapped into physical memory location 30000H, or any other location, with the paging unit. This means that an instruction that accesses location 20000H actually accesses location 30000H. Each memory page is 4K bytes long. Paging allows the system software to be placed at any physical address with the paging mechanism. Three components are used in page address translation: the page directory, the page table, and the actual physical memory page. The segmentation scheme may divide the physical memory into a variable size segments but the paging divides the memory into a fixed size pages. The segments are supposed to be the logical segments of the program. The pages are just fixed size portions of the program module or data. Advantage: The advantage of paging scheme is that the complete segment of a task need not be in the physical memory at any time. Only a few pages of the segments, which are required currently for the execution need to be available in the physical memory. Thus the memory requirement of the task is substantially reduced, relinquishing the available memory for other tasks Page Directory The page directory contains the location of up to 1024 page translation tables, which are each four bytes long. Each page translation table translates a logical address into a physical address. The page directory is stored in the memory and accessed by the page descriptor address register (CR3). Control register CR3 holds the base address of the page directory, which starts at any 4Kbyte boundary in the memory system. Each entry in the page directory translates the leftmost 10 bits of the memory address. This 10-bit portion of the linear address is used to locate different page tables for different page table entries Page Table 17

18 Unit 08 Advanced Microprocessor The page table contains 1024 physical page addresses, accessed to translate a linear address into a physical address Page Directory Entry Total Page Directory Entries are1024 Each directory entry is of 4byte 8.4. Page Table Entry Figure: Page Directory Entry The page table entries contain the starting address of the page and the statistical information about the page. Total Entries are1024 Each page table entry is of 4byte Figure: Page Table Entry D-bit: Dirty bit is undefined for page table directory entries by the microprocessor and is provided for use by the operating system. A-bit: Accessed bit is set to logic 1 whenever the microprocessor accesses the page Directory entry. R/W (Read/write) & U/S (user/supervisor) are both used in the protection scheme. Both bits combine to develop paging priority level protection for level 3, the lowest user level. P-bit: Present bit, if logic1 indicates that the entry can be used in address translation. If P = 0, the entry cannot be used for translation. When P = 0, the remaining bits of the entry can be used to indicate the location of the page on the disk memory system. Difference between page directory and page table entry: 18

19 Unit 08 Advanced Microprocessor The main difference is that the page directory entry contains the physical address of a page table, while the page table entry contains the physical address of a 4K-bytephysical page of memory. The other difference is the D (dirty bit), which has no function in thepage directory entry, but indicates that a page has been written to in a page table entry Page Translation Mechanism in A page frame is a 4K-byte unit of contiguous addresses of physical memory. Pages begin on byte boundaries and are fixed in size. A linear address refers indirectly to a physical address by specifying a page table, a page within that table, and an offset within that page Figure: Format of Linear Address The below figure of page translation shows, how processor converts the DIR, PAGE, and OFFSET fields of a linear address into the physical address by consulting two levels of page tables. The addressing mechanism uses the DIR field as an index into a page directory, uses the PAGE field as an index into the page table determined by the page directory, and uses the OFFSET field to address a byte within the page determined by the page table. In the second phase of address transformation, the transforms a linear address into a physical address. This phase of address transformation implements the basic features needed for pageoriented virtual-memory systems and page-level protection. Page translation is in effect only when the PG bit of CR0 is set. Figure: Page Translation 19

20 9. Features of Unit 08 Advanced Microprocessor The 32-bit is the next evolutionary step up from the One of the most obvious feature included in a is a built in math coprocessor. This coprocessor is essentially the same as the processor used with a 80386, but being integrated on the chip allows it to execute math instructions about three times as fast as a 80386/387 combination is an 8Kbyte code and data cache. To make room for the additional signals, the is packaged in a 168 pin, pin grid array package instead of the 132 pin PGA used for the Operates on 25MHz, 33 MHz, 50 MHz, 60 MHz, 66 MHz or 100MHz. It consists of parity generator/checker unit in order to implement parity detection and generation for memory reads and writes. Supports burst memory reads and writes to implement fast cache fills. Three mode of operation: real, protected and virtual 8086 mode. The microprocessor is a highly integrated device, containing well over 1.2 million transistors. 10. Pentium Processor Features It consists of all the features of The additional enhancements that Pentium provides are: 1) Wider data bus width : It has 64 bit data bus and 32 bit address bus. It allows 8 byte of data info to be transferred to and from memory. Bus cycle pipelining has been added to allow two bus cycles to be in progress simultaneously. 2) Improved Cache Structure: 8KB dedicated instruction cache which gives instruction to its execution units and floating point unit via dual instruction pipeline. Cache is organized in a 2 way set associate cache with 32 byte line(256 lines). 8KB data cache which gives data to its execution unit. This allows 32 byte transfer from cache to pre-fetch buffer which is of 64 bytes. 3) Two parallel integer execution unit : It allows the execution of two instructions to be executed simultaneously in a single processor clock. 4) Faster floating point unit : The floating point unit has been completely redesigned over

21 Unit 08 Advanced Microprocessor Faster algorithms provide up to ten times speed up for common operations including add, multiply etc. 5) Branch prediction logic: The Pentium uses tech called branch prediction. To implement this Pentium has two pre-fetch buffers, one to pre-fetch code in linear fashion, and one that pre-fetches code according to the Branch Target Buffer (BTB). Therefore, needed code is almost pre-fetched before it is required for execution. 6) Data Integrity and Error Detection: The Pentium have added significant data integrity and error detection capability. Data parity checking is still byte-by-byte basis. Address parity checking has also been added. 7) Functional Redundancy Checking: (provides maximum error detection) Two or more Pentium Processor can participate in functional redundancy checking. One processor (the master) fetching the instruction and executes the instruction in normal fashion. Other processor (the checker) (connected directly to the master processor s buses) verify correctness of master processor. Checker executes the instruction same as the master but doesn t drive the buses. Checker samples master s output and compares the values with the internal computed values. An error signal is asserted in case if mismatch occurs. 8) Super Scalar Architecture: Processor is capable of parallel instruction execution of multiple instructions are known as superscalar processors. Pentium is capable in some cases of executing two integer of two floating point instruction simultaneously and thus support superscalar architecture Architecture 21

22 Unit 08 Advanced Microprocessor Features of Figure: Pentium Architecture The 32-bit is the next evolutionary step up from the One of the most obvious feature included in a is a built in math coprocessor. This coprocessor is essentially the same as the processor used with a 80386, but being integrated on the chip allows it to execute math instructions about three times as fast as a 80386/387 combination is an 8Kbyte code and data cache. To make room for the additional signals, the is packaged in a 168 pin, pin grid array package instead of the 132 pin PGA used for the Operates on 25MHz, 33 MHz, 50 MHz, 60 MHz, 66 MHz or 100MHz. It consists of parity generator/checker unit in order to implement parity detection and generation for memory reads and writes. Supports burst memory reads and writes to implement fast cache fills. Three mode of operation: real, protected and virtual 8086 mode. The microprocessor is a highly integrated device, containing well over 1.2 million transistors. 22

23 Unit 08 Advanced Microprocessor Explain Architecture of Processor. The architecture of the 80486DX is almost identical to the Added to the architecture inside the 80486DX is a math coprocessor and an 8K-byte level 1 cache memory. BIU: Figure: Architecture BIU generates address, data and control signals for a bus cycle it is supported with an additional parity detection/generation for memory reads and writes. During memory write operation, the 486 generates even parity bit for each byte outputs these bits. These bits will be stored in a separate parity memory bank. During read operation, stored parity bits will be read from the parity memory checks the parities of data bytes read and compare them with the DP0 DP3 signals and generates parity check error, if it occurs. Instruction Pre-fetch Unit : It pre-fetches the instruction bytes in advance and holds them in a 32 byte code queue. Instruction Decoder : Decodes the instructions in the queue and passes the control and protection test unit. Execution Unit: Executes the instruction with the help of Barrel Shifter, ALU and Register bank. 23

24 Unit 08 Advanced Microprocessor Segmentation Unit and Paging Unit : They are part of MMU(which manages virtual memory of system). Helpful in generation of Physical Address. Work same as they work in Floating Point Unit: Responsible for performing the floating point operations. The has a built in math co-processor. It performs the floating point operations. It executes math instructions 3 times faster than the 386/387 combination. Cache Unit: 8KB cache Additional high speed cache memory provides a way of improving overall system performance. It contains the recently used instructions, data or both. The main aim is that the microprocessor unit access code and data in the cache most of time, instead from the main memory. EFLAG Register The extended flag register EFLAG is illustrated in the figure below:- The only new flag bit is the AC alignment check, used to indicate that the microprocessor has accessed a word at an odd address or a double word boundary. Figure: EFLAG Register of The

25 08 Advanced Microprocessor 1. Features of 8086 microprocessor General Features: 1) 8086 is a 16-bit processor, which implies that 16-bit data bus 16-bit ALU, that perform 16-bit operation at the same time 16-bit registers 2) Speed of processor: There are three versions based on the frequency of operation > 5MHz (5 million cycles per second) > 8MHz (8 million cycles per second) > 10MHz (10 million cycles per second) 3) 8086 has a 20 bit address bus can access up to 2 20 memory locations.( 2 20 = bytes = 1 MB) 4) It can support up to 64K I/O ports. (216 I/O ports-> 216=65536 bytes=64 Kb) 5) In 8086 two Unit works in parallel: Bus Interface Unit (BIU) Execution Unit (EU) 6) 8086 has 256 vectored interrupt. 7) 8086 contains powerful instruction set, that supports Multiply and Divide operation (this operation were not possible in previous versions of 8086) 8) 8086 can perform operation on bit, byte (8-bit), word (16-bit) or string types of data. Special Features: 1) 8086 is a pipelined processor 2) 8086 is two staged pipelined architecture: Fetch Stage: It pre-fetch up to 6 bytes of instruction and store them in the queue. Execute stage: Executes the instruction 3) 8086 can operate in 2 modes Minimum mode: A system with only one processor i.e.8086 Maximum mode: A system with multiple processors. e.g math co-processor(8087), I/O processor (8089), Multiple 8086 processors 4) 8086 uses memory bank In 8086 entire data is not stored in single sequential memory of 1MB. The memory is divided into two banks of 512KB each. I. Lower Bank/ Even Bank: Stores the data types at even locations (0,2,4 ) 1

26 08 Advanced Microprocessor II. Higher Bank/ Odd Bank: Stores the data types at odd locations (1,3,5 ) 5) 8086 uses memory segmentation Segmentation means dividing memory into logical components. In 8086 memory is divided into 16 segments of capacity 216 bytes each and used as code, stack, data and extra segment ARCHITECTURE The Block diagram for the internal architecture of 8086 is as follows:- Figure 1 architecture of 8086 In 8086 CPU is divided into two independent functional parts BIU and EU. Dividing the work between these two units speeds up the processing. 1) BIU (Bus Interface Unit) 2

27 08 Advanced Microprocessor Components of BIU Instruction queue - It holds the instruction bytes of the next instruction to be executed by EU Segment Registers - Four 16-bit register that provides powerful memory management mechanism - ES (extra segment), CS (code segment), SS (stack segment), DS (data segment). - The size of each register is 64kb. Instruction pointer (IP) - Register that holds 16-bit address or offset of next code byte within code segment Address Generation and bus control - Generation of 20-bit physical address Task carried out by BIU Fetch instructions from memory Read/ Write instruction from / to the memory Input/ Output (I/O) of data from / to peripheral ports Write the data to memory. Address generation for memory reference Queuing of instruction (The instruction bytes are transferred to the instruction queue) Thus, BUI handles all transfer of data and address on the buses for Execution unit. BIU works in synchronous with machine cycles 2) EU (Execution Unit) Components of EU ALU (Arithmetic logic Unit) - Contains 16-bit ALU, that performs add, subtract, increment, decrement, compliment, shift binary numbers, AND, OR, XOR etc. CU (Control Unit) - Directs internal operation Flag Register - 16-bit flag register - EU contains 9 active flags General Purpose Registers (GPR) - EU has 4 general purpose 16-bit register - i.e. AX, BX, CX, DX - each register is the combination of two 8-bit register 3

28 08 Advanced Microprocessor - AH, AL, BH, BL, CH, CL, DH, DL where L means Lower byte and H means higher byte. Index Register - 16-bit Register is SI (source index) and DI (destination index). - Both the register used for string related operation and for moving block of memory from one location to the other. Pointers - 16-bit Register. - i.e. SP (stack pointer), BP (base pointer) - BP : is used when we need to pass parameter through stack - SP: It always points to the top of the stack. Used for sequential access of stack segment. Decoder (instruction decoder) - Translates the instruction fetched from into series of action which EU carries out Task carried out by EU - Decodes the instruction - It executes instructions (executes decoded instructions) - Tells BIU from where to fetch the instruction - Decodes instruction (decode the fetched instruction) - EU takes care of performing operation on the data - EU is also known as execution heart of the processor REGISTERS The 8086 microprocessor has a total of fourteen registers that are accessible to the programmer as follows:- 1. General Purpose Register AX: - Accumulator register consists of two 8-bit registers AL and AH, which can be combined together and used as a 16-bit register AX. AX works as an intermediate register in memory and I/O operation. Accumulator is used for the instruction such as MUL and DIV. BX: -Base register consists of two 8-bit registers BL and BH, which can be combined together and used as a 16-bit register BX. BX register usually contains a data pointer used for based, based indexed or register indirect addressing. CX:-Count Register consists of two 8-bit registers CL and CH, which can be combined together and used as a 16-bit register CX. 4

29 08 Advanced Microprocessor Count register can be used in Loop, shift/rotate instructions and as a counter in string manipulation. DX: - Data register can be used together with AX register to execute MUL and DIV instruction. Data register can be used as a port number in I/O operations. 2. Segment Register Types of Segment registers are as follows:- Code Segment (CS): The CS register is used for addressing a memory location in the Code Segment of the memory, where the executable program is stored. Data Segment (DS): The DS contains most data used by program. Data are accessed in the Data Segment by an offset address or the content of other register that holds the offset address. Stack Segment (SS): SS defined the area of memory used for the stack. Extra Segment (ES): ES is additional data segment that is used by some of the string to hold the destination data 3. Pointer Registers The pointers IP, BP, SP usually contain offsets within the code, data and stack segments respectively. Stack Pointer (SP): SP is a 16-bit register pointing to program stack in stack segment. Base Pointer (BP): BP is a 16-bit register pointing to data in stack segment. BP register is usually used for based, based indexed or register indirect addressing. Instruction Pointer (IP): IP is a 16-bit register pointing to next instruction to be executed. 4. Index registers The Index Registers are as follows:- Source Index (SI): SI is a 16-bit register used for indexed, based indexed and register indirect addressing, as well as a source data addresses in string manipulation instructions. Destination Index (DI): DI is a 16-bit register. DI is used for indexed, based indexed and register indirect addressing, as well as a destination data addresses in string manipulation instructions. 5

30 08 Advanced Microprocessor 5. Flag Registers The 16-bit flag register of 8086 contains 9 active flags (six conditional & 3 control flags), other 7 flags are undefined. Status Flags: It indicates certain condition that arises during the execution. They are controlled by the processor. Control Flags: It controls certain operations of the processor. They are deliberately set/reset by the user U U U U OF DF IF TF SF ZF U AF U PF U CF U- Undefined CF- Carry Flag TF- Trap Flag PF- Parity Flag IF- Interrupt Flag AF- Auxiliary Flag DF- Direction Flag ZF- Zero Flag OF- Overflow Flag SF- Sign Flag The Flag Registers are classified as follows:- CONTROL FLAGS Control flags are set or reset deliberately to control the operations of the execution unit. Trap Flag (TF): It is used for single step control. It allows user to execute one instruction of a program at a time for debugging. When trap flag is set, program can be run in single step mode. Interrupt Flag (IF): It is an interrupt enable/disable flag. If it is set, the mask able interrupt of 8086 is enabled and if it is reset, the interrupt is disabled. It can be set by executing instruction sit and can be cleared by executing CLI instruction. Direction Flag (DF): It is used in string operation. 6

31 08 Advanced Microprocessor If it is set, string bytes are accessed from higher memory address to lower memory address. When it is reset, the string bytes are accessed from lower memory address to higher memory address. STATUS FLAG Carry Flag (CF): This flag indicates an overflow condition for unsigned integer arithmetic. It is also used in multiple-precision arithmetic. Auxiliary Flag (AF): If an operation performed in ALU generates a carry/barrow from lower nibble (i.e. D0 D3) to upper nibble (i.e. D4 D7), the AF flag is set i.e. carry given by D3 bit to D4 is AF flag. This is not a general-purpose flag, it is used internally by the processor to perform Binary to BCD conversion. Parity Flag (PF): This flag is used to indicate the parity of result. If lower order 8-bits of the result contains even number of 1 s, the Parity Flag is set and for odd number of 1 s, the Parity Flag is reset. Zero Flag (ZF): It is set; if the result of arithmetic or logical operation is zero else it is reset. Sign Flag (SF): In sign magnitude format the sign of number is indicated by MSB bit. If the result of operation is negative, sign flag is set. Overflow Flag (OF): It occurs when signed numbers are added or subtracted. OF=1 indicates that the result has exceeded the capacity of machine. 4. Addressing Mode There are 12 addressing modes in 8086 as follows:- 1) Immediate Mode 7

32 08 Advanced Microprocessor If a source operand is part of the instruction instead of the contents of a register or memory location, it represents what is called the immediate operand. In other word it is constant data contained in an instruction If source operand is the part of instruction rather than register or memory, then referred as immediate addressing mode. Operand = address field E.g. MOV AL, 05H; instruction copies immediate number 05H to AL register Immediate data may be 8-bits or 16-bits in length Instruction Opcode Immediate Operand Advantage: Operand can be accessed quickly as they are directly available in instruction queue. No need of External bus or bus-cycles to obtain data. No memory reference to fetch data Comparatively Faster execution Limitation: The operand can only be used as a source operand. Value of the operand will remain Constant 2) Register Addressing Mode The operand is stored in one of the CPU register. Register may be used as source operand or destination operand or both E.g. ADD r1, r2; adding r1 and r2 and store the result in r2 E.g. MOV AX,BX ; move value from BX to AX register Advantage: This mode is normally preferred, as the execution of instruction is faster and compact, because all the registers reside on the same chip. Therefore, data transfer is within the chip and no External bus is required. Limitation: Number of CPU registers are limited 8

33 08 Advanced Microprocessor 3) Direct Addressing Mode Address field of Instruction contains address of operand Advantage: Single memory reference is required to access data No additional calculations to work out effective address Equation: Physical Address PA ={starting address of Segment Register } + {Direct memory address} Example: MOV [30540], TEMP; moving value of TEMP to memory location [30540] DS=3060 and direct memory address= ( 0 is added to LSB by BIU) ( direct address) (physical address) 9

34 08 Advanced Microprocessor 4) Register Indirect Addressing In this addressing mode effective address of memory is calculated from base register (BX) or index register (SI, DI), specified in the instruction. Then it is added to the segment register to generate physical address. Equation: Physical Address PA ={starting address of Segment Register } + { [BX] or index register } E.g. MOV [DI], BX; value of BX is moved to the memory location specified in DI MOV [BX], AX ; value of AX is moved to the memory location specified in BX Advantage: Large address space 10

35 08 Advanced Microprocessor Limitation: Slower access, as multiple memory accesses is needed to find operand. 5) Based Addressing Mode When memory is accessed PA is computed from BX and DS, when the stack is accessed PA is computed from BP and SS. Equation: PA ={starting address of Segment Register } + { [BX] or [BP] }+ (8 or 16 bit) displacement Example: MOV AL, TEMP [BX]; segment register address+ BX+ offset MOV AL, TEMP [BP]; segment register address+ BP+ offset Assume DS=3060, BX=0050 and displacement= H (starting address of segment register) H(base register) H(offset) H(physical address) 6) Indexed Addressing Mode In this addressing mode offset address is added to index register and finally the sum is added to segment register. Example. MOV BH, TEMP [SI] ; TEMP is the offset address 11

36 08 Advanced Microprocessor Equation: PA ={starting address of Segment Register } + { [SI] or [DI] }+ (8 or 16 bit) displacement Assume DS=3060, SI=0050 and displacement= H (starting address of segment register) H (Source Index) H (offset) H (physical address) 7) Based Indexed Addressing Mode This mode generates the effective address, which is sum of Base address +Index Address + (8 or 16 bit) displacement address. This sum is added to segment register to generate effective address. Example : MOV CX, TEMP[BX] [SI] Equation: PA ={starting address of Segment Register } + {[BX] or [BP]}+({ [SI] or [DI] }+ (8 or 16 bit) displacement Assume DS=5000, BX=3000, SI=0400 and displacement= H (starting address of segment register) H (base register) H (Source Index) H (offset) 12

37 08 Advanced Microprocessor H (physical address) 8) String Addressing Mode The string instructions automatically assume SI to point to the first byte or word of the source operand and DI to point to the first byte or word of the destination operand. The contents of SI and DI are automatically incremented (by clearing DF to 0 by CLD instruction) to point to the next byte or word. Source address : 20500, assume it contains 38 PA : [DS] + [SI] = =20500 Destination address : [ES] + [DI] = After executing MOV SI, DI [40300] = 38 [SI] = 0501 incremented [DI] = 03 9) Relative addressing mode: Relative address means relative to IP (Instruction Pointer). Example : JNC START ; jump to label if no carry is generated - If CY=O, then PC is loaded with current PC contents plus 8 bit signed value of START, otherwise the next instruction is executed. 13

38 08 Advanced Microprocessor - Displacement is calculated on the basis of next location to be executed. 10) Implied addressing mode: Instructions using this mode have no operands. Example: CLC; which clears carry flag to zero. STC; set the carry flag 11) I/O mode (direct) : Port number is an 8 bit immediate operand. Example : OUT 05 H, AL Outputs value of AL to 8 bit port 05 H IN and OUT instructions are allowed to use only AL or AX registers. 12) I/O mode (indirect): I/O port address is provided in DX register. Port address ranges from 0000 to FFFFH. The port number is taken from DX. Example 1 : IN AL, DX IN and OUT instructions are allowed to use only AL or AX registers. 5. SEGMENTATION IN 8086 In Segmentation, the total memory size is divided into segments of various sizes. Segment is just an area in memory. The process of dividing memory into segments of various sizes is called Segmentation. Memory is huge collection of bytes. In order to organize these bytes in an efficient manner Segmentation is used. The complete 1Mbyte memory is divided into 16 logical segments. In memory, data is stored as bytes. Each byte has a specific address. Intel 8086 has 20 lines address bus. With 20 address lines, the memory that can be addressed is 220 bytes. 20 lines= 220 = 1,048,576 bytes =1 MB= = FFFFF H At a time 8086 can only access 4 segments. i.e 64KB * 4 = 256KB of segments can be accessed at a time. 14

39 08 Advanced Microprocessor 8086 memory with address ranging from H to FFFFFF H. Size of each Segment Register is 16-bit 216 = bytes = 64K [size of each segment] Total number of segments: = Total memory available/size of each segment = 1MB/64KB = 1024KB/64KB = 16 segments. Segment Registers are used to hold the upper 16-bit of the starting address for each of the segment. The 16-bit of the starting address is the starting address of the segment from where the BIU is currently fetching instruction code bytes. The BIU always inserts zeros for the LSB of the 20-bit address for a segment. Because the segment registers cannot store 20 bits, they only store the upper 16 bits. The 20-bit address of a byte is called its Physical Address (PA). PA= Base Address : Offset Offset is the displacement of the memory location from the starting location of the segment. The value of Data Segment Register (DS) is 2222 H. To convert this 16-bit address into 20-bit, the BIU appends 0H to the LSBs of the base address. After appending, the starting address of the Data Segment becomes 22220H. If the data at any location has an address specified as: 2222H: 0016 H where the number 0016 H is an offset. To calculate the effective address of the memory, BIU uses the following formula: 15

40 08 Advanced Microprocessor Physical Address = Starting Address of Segment + Offset To find the starting address of the segment, BIU appends the contents of Segment Register with 0H and then, it adds offset to it. EA = H H H 16

41 Unit 08 Advanced Microprocessor The was designed for multi-user systems with multitasking applications, including communications and real-time process control. It had 134,000 transistors and consisted of four independent units: address unit, bus unit, instruction unit and execution unit. These were organized into a pipeline, significantly increasing performance. It was produced in a 68-pin package including PLCC (Plastic Leaded Chip Carrier), LCC (Leadless chip carrier) and PGA (Pin Grid Array) packages. The Intel had a 24-bit address bus and was able to address up to 16 MB of RAM, compared to 1 MB for its predecessor. However cost and initial rarity of software using the memory above 1 MB meant that computers were rarely shipped with more than one megabyte of RAM Architecture contains 4 separate processing units. (1) Bus Unit (BU) (2) Instruction Unit (IU) (3) Address Unit (AU) (4) Execution Unit (EU) Fig. 1: Architecture of Intel Bus Unit (BU): It has address latches, data transceivers, bus interface and circuitry, instruction pre-fetcher, processor extension interface and 6 byte instruction queue. Functions : To perform all memory and I/O read and write. 1

42 Unit 08 Advanced Microprocessor To pre-fetch the instruction bytes. To control the transfer of data to and from processor extension devices like math coprocessor. Whenever BU is not using the buses for the operation, it pre-fetches the instruction bytes and put them is a 6 byte pre-fetch queue. Instruction Unit (IU): It has 3 decoded instruction queue and instruction decoder. Functions: It fully decodes up to three pre-fetched instructions and holds them in a queue. So that EU can access them. It helps the processor to speed up, as pipelining of instruction is done. Execution Unit (EU): It includes ALU, registers and the Control unit. Registers are general purpose, index, pointer, flag register and 16 bit Machine Status Word (MSW). Functions: To sequentially execute the instructions received from the instruction unit. ALU result is either stored in register bank or sent over the data bus. Address Unit (AU): It consists of segment registers, offset address and a physical address adder. Functions: Compute the physical address that will be sent out to the memory or I/O by BU operate in two different modes Real address mode Protected virtual address mode. When used in Real address mode, AU computes the address with segment base and offset like Segment register are CS, DS, ES and SS hold base address. IP, BP, SI, DI, SP hold offset. Maximum physical space allowed in this mode is 1MB. When operate in protected mode, the address unit acts as MMU. All 24 address lines used and can access up to 16MB of physical memory. If descriptor table scheme is used it can address up to 1GB of virtual memory. 2. REGISTER ORGANIZATION OF The CPU contains almost the same set of registers, as in 8086, namely (1) Eight 16-bit general purpose registers (AX, BX, CX, DX) (2) Four 16-bit segment registers (CS, SS, DS, ES) 2

43 (3) Status and control registers (SP, BP, SI, DI) (4) Instruction Pointer (IP) (5) Two 16-bit register - FLAGS, MSW (6) Two 16-bit register - LDTR and TR (7) Two 48-bit register -GDTR and IDTR Unit 08 Advanced Microprocessor Fig.2: Register Set 3

44 Unit 08 Advanced Microprocessor Fig. 3: Flag Register Flag register is of 32 bit. 15th bit undefined/reserved. IOPL I/O Privilege Level flag: 2 bits are used in protected mode. It holds the privilege level from 0 to 3. 0 assigns to highest privilege whereas 3 assigns to lower privilege level. NT: Nested Task flag: It is used in protected mode. Bit is set when one task invokes another task. MSW machine status word There are four Processor Status bit in 80286:- Fig. 4 : Machine Status Word 1) PE - Protection Enable The PE bit is set to enable the Protected Mode. If PE is reset, the processor operates again in Real Mode. 2) MP - Monitor Processor Extension The MP bit is used in conjunction with the TS bit to determine if the WAIT opcode will generate a Coprocessor Not Available fault when TS=1. When both MP = 1 and TS = 1, the WAIT opcode generates a trap. Otherwise, the WAIT opcode does not generate a trap. Note that TS is automatically set whenever a task switch operation is performed. 3) EM - Processor Extension Emulator The EMulate coprocessor bit is set to cause all coprocessor opcodes to generate a Coprocessor Not Available fault. It is reset to allow coprocessor opcodes to be executed on an actual Intel387 DX coprocessor. Note that the WAIT opcode is not affected by the EM bit setting. 4) TS Task Switch 4

45 Unit 08 Advanced Microprocessor TS is automatically set whenever a task switch operation is performed. 3. Salient Features of The is the first member of the family of advanced microprocessors with memory management and protection abilities. The CPU, with its 24-bit address bus is able to address 16 Mbytes of physical memory. Various versions of are available that runs on 12.5 MHz, 10 MHz and 8 MHz clock frequencies is upwardly compatible with 8086 in terms of instruction set have two operating modes namely real address mode and virtual address mode. In real address mode, the can address up to 1Mb of physical memory address like In virtual address mode, it can address up to 16 Mb of physical memory address space and 1 GB of virtual memory address space. The instruction set of includes the instructions of 8086 and have some extra instructions to support operating system and memory management. In protected virtual address mode, it is source code compatible with The performance of is five times faster than the standard Memory Management Unit In advanced microprocessor, memory management becomes extremely important. Memory management is required due to the following two reasons. 1. Limitation of physical memory A microprocessor has limited number of address lines. Hence the physical memory addressability is limited. Increasing the number of address lines is not attractive as it makes the architecture and design complex without significant gain. Packaging becomes difficult and expensive. Memory Management Unit (MMU) solves this problem by translating the virtual memory address into the physical memory address. Virtual memory can be many times larger than the physical memory. Only the programs currently required are brought from the secondary storage such as a hard disk to, the physical memory (RAM) for execution. 2. Need for Protection In a multi-user operating system, there is a possibility that a user program can corrupt the operating system area or the area of some other user unless a protection mechanism is 5

46 Unit 08 Advanced Microprocessor built. Hence each user should be protected from other users and the operating system should be protected from other user (task). The user (task) should be allowed to have a controlled access to the operating system resources. Hence various privilege levels are defined. For example, in a situation having 4 privilege levels, 0 is the highest privilege and 3 has the lowest privilege as shown in the figure. The figure shows a typical UNIX operating system layout. It has to be noted that the user is at the lowest privilege level, i.e., privilege level 3 and the operating system Kernel is at the highest privilege level, i.e., privilege level 0. Role of Memory Management Unit (MMU): The virtual address space of a microprocessor may be many times larger than the actual physical address space. This is desirable as a microprocessor is supposed to store large programs and data which cannot be accommodated in the physical memory space. Usually programs and data are stored in a secondary storage such as a hard disk. The hard disk is in the virtual or logical address space but not in the physical address space. Faster memory such as RAM is used as the physical memory (Primary Storage). When a microprocessor is to execute a program, it checks whether the program is available in the physical memory (RAM). If the program is not available in the physical memory, it is brought from the secondary memory to the physical memory for execution. If available space is inadequate in the physical memory, some less important or unused program can be swapped back to the secondary memory to create space. 5. OPERATING MODES OF works in two operating mode: 1. Real Address Mode (Just act as a fast 8086). Instruction set is upwardly compatible Because of extra pipelining and other circuit level improvements, in real address mode also, the operates at a much faster rate than 8086, although functionally they work in an identical fashion. As in 8086, the physical memory is organized in terms of segments of 64Kbyte maximum size. 6

47 Unit 08 Advanced Microprocessor In the real mode the first 1Kbyte of memory starting from address 0000H to 003FFH is reserved for interrupt vector table. The addresses from FFFF0H to FFFFFH are reserved for system initialization. When the is reset, it always starts the execution in real address mode. In real address mode, it initializes the IP and other registers of Fig.5 : Real Address Mode Address Calculation 2. Protected Virtual Address Mode (PVAM) is the first processor to support the concepts of virtual memory and memory management. The concept of Virtual Memory is implemented using Physical memory that the CPU can directly access and secondary memory that is used as storage for data and program, which are stored in secondary memory initially. The complete virtual memory is mapped on to the 16Mbyte physical memory. If a program larger than 16Mbyte is stored on the hard disk and is to be executed, if it is fetched in terms of data or program segments of less than 16Mbyte in size into the program memory by swapping sequentially as per sequence of execution. The is able to address 1 GB (230 bytes) of virtual memory uses the 16-bit content of a segment register as a selector to address a descriptor stored in the physical memory. 7

48 Unit 08 Advanced Microprocessor The descriptor is a block of contiguous memory locations containing information of a segment, like segment base address, segment limit, segment type, privilege level, segment availability in physical memory descriptor type and segment. Hardware reset is the only way to come out of protected mode Fig. 6 : Physical Address Calculation in PVAM 6. Privilege level There are four types of privilege levels kernel level (highest privilege level) OS services OS extensions Applications (lowest privilege level) 8

49 Unit 08 Advanced Microprocessor Fig. 7 : Privilege Level Each task assigned a privilege level, which indicates the priority or privilege of that task. It can only change by transferring the control, using gate descriptors, to a new segment. A task executing at level 0, the most privileged level, can access the entire data segment defined in GDT and LDT of the task. A task executing at level 3, the least privileged level, will have the most limited access to data and other descriptors. The use of rings allows for system software to restrict tasks from accessing data. In most environments, the operating system and some device drivers run in ring 0 and applications run in ring DESCRIPTOR TABLE GDT GDT, LDT, IDT and TSS are all data structures specified by Intel architecture in memory management module. Descriptor is an identifier of a program segment or page. GDT, Global Descriptor Table, is used to define the characteristics of the various memory areas used during program execution, including the base address, the size and access privileges like execute ability and write ability. 9

50 Unit 08 Advanced Microprocessor LDT These memory areas are called segments in Intel terminology. Segment is a term for memory management in Intel architecture, which is also used collaboratively with paging mechanism. LDT, Local Descriptor Table, acts similar to GDT, which also saves segments descriptor. The LDT is the sibling of the Global Descriptor Table (GDT) and defines up to 8192 memory segments accessible to programs. Difference between LDT and GDT USE The main differences between GDT and LDT is: 1. GDT have only one copy in system while LDT can have many 2. GDT may not change during execution which LDT often changes when task switches 3. Entry of LDT is save in GDT. Entries in GDT and LDT have the same structure. A segment cannot be accessed, if its descriptor does not exist in either LDT or GDT. Set of descriptor (descriptor table) arranged in a proper sequence describes the complete program. The descriptor is a block of contiguous memory location containing information of a segment, like 1) Segment base address 2) Segment limit 3) Segment type 4) Privilege level prevents unauthorized access 5) Segment availability in physical memory 6) Descriptor type 7) Segment use by another task Requirement of Descriptor Table The descriptor describes the location, length, and access rights of the segment of memory. The selector, located in the segment register, selects one of descriptors from one of two tables of descriptors. 10

51 Unit 08 Advanced Microprocessor 1. Features of Fig. 8 : Protected Mode Addressing with Descriptor Table The microprocessor is an enhanced version of the microprocessor Memory-management unit is enhanced to provide memory paging. The also includes 32-bit extended registers and a 32-bit address and data bus. These extended registers include EAX, EBX, ECX, EDX, EBP, ESP, EDI, ESI, EIP and EFLAGS. The has a physical memory size of 4GBytes that can be addressed as a virtual memory with up to 64TBytes. The is operated in the pipelined mode, it sends the address of the next instruction or memory data to the memory system prior to completing the execution of the current instruction This allows the memory system to begin fetching the next instruction or data before the current is completed. This increases access time. The instruction set of the is enhanced to include instructions that address the 32-bit extended register set. The memory manager is similar to the 80286, except the physical addresses generated by the MMU are 32 bits wide instead of 24-bits. The concept of paging is introduced in support three operating modes: 1) Real Mode (default) 2) Protected Virtual Address Mode (PVAM) 3) Virtual Mode 11

52 Unit 08 Advanced Microprocessor The memory management section of supports virtual memory, paging and four levels of protection. The includes special hardware for task switching Architecture The internal architecture of the includes six functional units that operate in parallel. The parallel operation is called as pipeline processing. Fetching, decoding execution, memory management, and bus access for several instructions are performed simultaneously. The six functional units of the are 1) Bus Interface Unit 2) Code Pre-fetch Unit 3) Instruction Decoder Unit 4) Execution Unit 5) Segmentation Unit 6) Paging Unit Figure: Architecture The Bus Interface Unit connects the with memory and I/O. Based on internal requests for fetching instructions and transferring data from the code pre-fetch unit, the generates the address, data and control signals for the current bus cycles. 12

53 Unit 08 Advanced Microprocessor The code pre-fetch unit pre-fetches instructions when the bus interface unit is not executing the bus cycles. It then stores them in a 16-byte instruction queue for decoding by the instruction decode unit. The instruction decode unit translates instructions from the pre-fetch queue into microcodes. The decoded instructions are then stored in an instruction queue (FIFO) for processing by the execution unit. The execution unit processes the instructions from the instruction queue. It contains a control unit, a data unit and a protection test unit. The control unit contains microcode and parallel hardware for fast multiply, divide and effective address calculation. The unit includes a 32-bit ALU, 8 general purpose registers and a 64-bit barrel shifter for performing multiple bit shifts in one clock. The data unit carries out data operations requested by the control unit. The protection test unit checks for segmentation violations under the control of microcode. The segmentation unit calculates and translates the logical address into linear addresses at the request of the execution unit. The translated linear address is sent to the paging unit. Upon enabling the paging mechanism, the translates these linear addresses into physical addresses. If paging is not enabled, the physical address is identical to the linear address and no translation is necessary. 3. Register organization of The Register organization of is as follows: Figure:80386 General Purpose, Index and Pointer Register 13

54 3.1. General Purpose Register Unit 08 Advanced Microprocessor Registers EAX, EBX, ECX, EDX, EBP, EDI and ESI are regarded as general purpose or multipurpose registers. EAX (ACCUMULATOR):The accumulator is used for instructions such as multiplication, division and some of the adjustment instructions. In and above, the EAX register may also hold the offset address of a location in memory system. EBX (BASE INDEX): This can hold the offset address of a location in the memory system in all version of the microprocessor. It the and above EBX also can address memory data. ECX (count): This acts as a counter for various instructions. EDX (data): EDX is a general-purpose registers that holds a part of the result for multiplication or part of the division. In the and above this register can also address memory data Pointer and Index Register EBP (Base Pointer): EBP points to a memory location in all version of the microprocessor for memory data transfers. ESP (Stack Pointer): ESP addresses an area of memory called the stack. The stack memory is a data LIFO data structure. The register is referred to as SP if used in 16 bit mode and ESP if referred to as a 32 bit register. EDI (Destination index): EDI often addresses string destination data for the string instruction. It also functions as either a 32-bit (EDI) or 16-bit (DI) general-purpose register. ESI (Source index): ESI can either be used as ESI or SI. It is often used to the address source string data for the string instructions. Like EDI ESI also functions as a generalpurpose registers. Figure:80386 Segment Register CS (Code): The code segment is a section of memory that holds the code used by the microprocessor. The code segment registers defines the starting address of the section of memory holding code. 14

55 Unit 08 Advanced Microprocessor SS (Stack): The stack segment defines the area of memory used for the stack. The stack entry point is determined by the stack segment and stack pointer registers. The BP registers also addresses data within the stack segment. DS (Data): The data section contains most data used by a program. Data are accessed in the data segment by an offset address of the contests of other registers that hold the offset address. ES (extra): The extra segment is used to hold information about string transfer and manipulation FS and GS: These are supplementing segment registers available in the and above microprocessors to allow two additional memory segments for access by programs. EIP (Instruction Pointer): EIP addresses the next instruction in a section of memory defined as a code segment. This register is IP (16bit) when microprocessor operates in the real mode and EIP (32 bits) when and above operate in protected mode 3.3. Flag Register: Figure: Instruction Pointer Indicates the condition of the microprocessor and controls its operations. Flag registers are also upward compatible since the have 16bit registers and the and above have EGLAF register (32 bits) Figure: Flag Register IOPL (I/O Privilege level): IOPL is used in protected mode operation to select the privilege level for I./O devices. IF the current privilege level is higher or more trusted than the IOPL, I/O executed without hindrance. If the IOPL is lover than the current privilege level, an interrupt occurs, causing execution to suspend. Note that an IPOL is 00 is the highest or more trusted; if IOPL is 11, it s the lowest or least trusted. 15

56 Unit 08 Advanced Microprocessor NT (Nested Task): The nested task flag is used to indicate that the current task is nested within another task in protected mode operation. This flag is when the task I nested by software. RF (Resume): The resume flag is used with debugging to control the resumption of execution after the next instruction. VM (Virtual Mode): The VM flag bit selects virtual mode operation in a protected mode system. Note: All the other flag bit is having similar description as in 8086 flag register System Address Register: Four memory management registers are used to specify the locations of data structures which control segmented memory management. GDTR (Global Descriptor Table Register) and IDTR (Interrupt Descriptor Table Register) be loaded with instructions which get a 6 byte data item from memory LDTR (Local Descriptor Table Register) and TR (Task Register) can be loaded with instructions which take a 16-bit segment selector as an operand Special Register Control Register: Four Control Register (CR0-CR3) Debug Register: Eight Debug Register (DR0-DR7) Test Register: Two Test Register (TR6-TR7) Control Register in detail. The has three 32 bit control registers CR0, CR2 and CR3 to hold global machine status independent of the executed task. Load and store instructions are available to access these registers 16

57 Unit 08 Advanced Microprocessor Common tasks performed by control registers include interrupt control, switching the addressing mode, paging control, and coprocessor control. Figure: Control Register CR0: CR2: CR3: CR0 contains system control flags, which control or indicate conditions that apply to the system as a whole, not to an individual task. PE (Protection Enable, bit 0): Setting PE causes the processor to begin executing in protected mode. Resetting PE returns to real-address mode. MP (Math Present, bit 1): MP controls the function of the WAIT instruction, which is used to coordinate a coprocessor. EM (Emulation, bit 2):EM indicates whether coprocessor functions are to be emulated. TS (Task Switched, bit 3):The processor sets TS with every task switch and tests TS when interpreting coprocessor instructions. ET (Extension Type, bit 4):ET indicates the type of coprocessor present in the system (80287 or 80387). PG (Paging, bit 31): PG indicates whether the processor uses page tables to translate linear addresses into physical addresses. CR2 is used for handling page faults when PG is set. The processor stores in CR2 the linear address that triggers the fault. Contains a value called Page Fault Linear Address (PFLA). When a page fault occurs, the address the program attempted to access is stored in the CR2 register. Used when virtual addressing is enabled, hence when the PG bit is set in CR0, CR3 enables the processor to translate linear addresses into physical addresses by locating the page directory and page tables for the current task. Typically, the upper 20 bits of CR3 become the page directory base register (PDBR), which stores the physical address of the first page directory entry. Note: Reserved field shown in figure are kept reserved by manufacturer 17

58 Unit 08 Advanced Microprocessor Debug Register Debugging of allows data access breakpoints as well as code execution breakpoints contains 6 debug registers to specify 1) 4 breakpoints 2) Breakpoint Control options 3) Breakpoint Status Linear Breakpoint Address Registers: Figure: Debug Register The breakpoint addresses specified are 32-bit linear addresses While debugging, Intel 386 h/w continuously compares the linear breakpoint addresses in DR0-DR3 with the linear addresses generated by executing software. Debug Control Register: Figure: Debug Control Register LENi(i=0-3): Breakpoint Length Specification Bits: 1) 2 bit field for each breakpoint 2) Specifies length of breakpoint fields 3) The choices of data breakpoints are 1byte, 2bytes & 4bytes 4) For instruction execution breakpoint, the length is 1(beginning byte address) 18

59 Unit 08 Advanced Microprocessor Table: LENi Encoding RWi(i=0-3): Memory Access Qualifier Bit 1) 2 bit field for each breakpoint 2) Specifies the type of usage which must occur inorder to activate the associated breakpoint GD: Global Debug Register Access Detect Table: Table: RWii Encoding Debug registers can only be accessed in real mode or at privilege level 0 in protected mode 19

60 Unit 08 Advanced Microprocessor GD bit, when set, provides extra protection against any Debug Register access even in Real Mode or at privilege level 0 in Protected Mode. The GD bit, when set, causes an exception 1 fault if an instruction attempts to read or write any Debug Register. The GD bit is then automatically cleared when the exception 1 handler is invoked, allowing the exception 1 handler free access to the debug registers. GE and LE bit: Exact data breakpoint match, global and local Gi and Li(i=0-3): Breakpoint Enable, global and local If either Gi or Li is set then the associated breakpoint is enabled. Debug Status Register: A Debug Status Register allows the exception 1 handler to easily determine why it was invoked. It can be invoked as a result of one of several events: 1) DR0 Breakpoint fault/trap. 2) DR1 Breakpoint fault/trap. 3) DR2 Breakpoint fault/trap. 4) DR3 Breakpoint fault/trap. 5) Single-step (TF) trap. 6) Task switch trap. 7) Fault due to attempted debug register access when GD = 1. Bi : Debug fault/trap due to breakpoint 0-3 Figure: Debug Status Register Four breakpoint indicator flags, B0-B3, correspond one-to-one with the breakpoint registers in DR0-DR3. A flag Bi is set when the condition described by DRi, LENi, and RWi occurs. BD: Debug fault due to attempted register access when GD bit is set This bit is set if the exception 1 handler was invoked due to an instruction attempting to read or write to the debug registers when GD bit was set. BS: Debug trap due to single step This bit is set if the exception 1 handler was invoked due to the TF bit in the flag register being set BT: Debug trap due to task switch 20

61 Unit 08 Advanced Microprocessor This bit is set if the exception 1 handler was invoked due to a task switch occurring to a task having an Intel Test Register Two more test register are provided by for page caching They are used to control the testing of Translation Look-aside Buffer(TLB) of Intel386. TR6 is the command test register TR7 is the data register which contains the data of Translation Look-aside buffer test. Figure: Test Register 3.6. Program-Invisible registers The global and local descriptor tables are found in the memory system. In order to access and specify the address of these tables, the program invisible registers used. The program invisible registers are not directly addressed by software so they are given name. The GDTR (global descriptor table register) and IDTR (interrupt descriptor table register) contain the base addresses of the descriptor table and its limit. The limit of each descriptor table is 16 bits because the maximum table length is 64 Kbytes. When the protected mode operation is desired, the address of the global descriptor table and its limit are loaded into the GDTR. 4. Real Modes of Default Mode After reset, the starts from the memory location FFFFFFF0 H under real address mode. In real address mode, works as a fast 8086 with 32 bit registers and data types. Real-address mode is in effect after a signal on the RESET pin. Even if the system is going to be used in protected mode, the start-up program will execute in real-address mode temporarily while initializing for protected mode. The addressing techniques, memory size, interrupt handling in this mode of are similar to the real addressing mode of In real address mode, the default operand size is 16 bit but 32 bit operands and addressing modes may be used with the help of override prefixed. Maximum physical memory = 1Mega byte (1MB) The only way to leave real-address mode is to switch to protected mode. 5. PVAM of

62 Unit 08 Advanced Microprocessor Introduced in the processor. 32-bit address bus => access up to 232 bytes = B = 4 GB Base address => 32-bit value Offset =>16-bit or 32-bit value Linear address = base address + offset Linear address physical address with paging In protected mode, the segment registers contain an index into a table of segment descriptors. Each segment descriptor contains the start address of the segment, to which the offset is added to generate the address. In addition, the segment descriptor contains memory protection information. This includes an offset limit and bits for write and read permission. This allows the processor to prevent memory accesses to certain data. Protected mode is accessed by placing a logic 1 into the PE bit of CR0 This system contains one data segment descriptor and one code segment descriptor with each segment set to 4G bytes in length. PVAM mode support memory management, virtual memory, multitasking, protection, debugging, segmentation and paging. 6. Virtual Mode of In its protected mode of operation, 80386DX provides a virtual 8086 operating environment to execute the 8086 programs. The real mode can also use to execute the 8086 programs along with the capabilities of 80386, like protection and a few additional instructions. Once the enters the protected mode from the real mode, it cannot return back to the real mode without a reset operation. Thus, the virtual 8086 mode of operation of 80386, offers an advantage of executing 8086 programs while in protected mode. The address forming mechanism in virtual 8086 mode is exactly identical with that of 8086 real mode. In virtual mode, 8086 can address 1Mbytes of physical memory that may be anywhere in the 4Gbytes address space of the protected mode of Like real mode, the addresses in virtual 8086 mode lie within 1Mbytes of memory. In virtual mode, the paging mechanism and protection capabilities are available at the service of the programmers. The supports multiprogramming, hence more than one programmer may be using the CPU at a time. 22

63 Unit 08 Advanced Microprocessor Paging unit may not be necessarily enable in virtual mode, but may be needed to run the 8086 programs which require more than 1Mbyts of memory for memory management function. In virtual mode, the paging unit allows only 256 pages, each of 4Kbytes size. Each of the pages may be located anywhere in the maximum 4Gbytes physical memory. The virtual mode allows the multiprogramming of 8086 applications. The virtual 8086 mode executes all the programs at privilege level 3. Any of the other programmers may deny access to the virtual mode programs or data. Even in the virtual mode, all the interrupts and exceptions are handled by the protected mode interrupt handler. To return to the protected mode from the virtual mode, any interrupt or execution may be used. As a part of interrupt service routine, the VM bit may be reset to zero to pull back the into protected mode. 7. Segment in In protected mode the contents of segment register is called selectors shown below:- Figure: Segment Selector Descriptor Index: Selects any one of the 8192(213) descriptor from a descriptor table TI Table Indicator TI=1 Local Descriptor Table TI=0 Global Descriptor Table RPL (Requested Privilege Level)(2-bit): Refers the privilege of that segment. They are used while protection checks to indicate whether access to segment is allowed or not allowed Descriptor Need of Descriptor Segments are the memory area defined by programmer and can be code, stack and data segment. 23

64 Unit 08 Advanced Microprocessor segments are also assigned to have attributes viz. privilege level, segment type, segment limit. So, it s not possible to use a 16 bit segment register to represent all the info regarding a segment. therefore the Solution is Descriptor Segment sizes vary from 1byte to 4GB in Figure: Segment Descriptor in Protected Mode Offset Bit Description 40 A(Accessed) sets this bit to 1 whenever a memory reference to a segment defined by this descriptor is made TYPE 3-bit field describes the type of segment 000 Data segment, read only 001 Data segment, read/write 010 Stack segment, read only 011 Stack segment, read/write 100 Code segment, execute-only, non-conforming 101 Code segment, execute/read, non-conforming 110 Code segment, execute-only, conforming 111 Code segment, execute/read, conforming 24

65 Unit 08 Advanced Microprocessor 44 S(System) 0 Specifies system segment descriptor E.g. LDS, TSS, Gates 1 Specifies non system segment descriptor E.g. code, stack or data DPL (Descriptor Privilege level) Indicates privilege level associated with memory space. 0 - most privileged 3 least privileged 47 Present If set 0, indicates that the address range that is specified by the descriptor is temporarily not present. 52 U(User) Undefined bit. 53 X Reserved by Intel 54 D(Default Size) 0 Operands within this segments are assumed to be 16 bits. 1 Operands assumed to be 32 bits 55 G(Granularity bit) 0 Unit of limit field is 1 byte. 1 Unit of limit field is of 4096 bytes (or 1 page). The Above figure is the format of the code/data descriptor; one descriptor is 64-bit long. As we can see, a descriptor actually includes a 32-bit base address and a 20-bit limit and some attributes, the 32-bit base address indicate where the segment starts, and the 20-bit limit indicates the length of the segment. However, a problem comes up, 20-bit limit can only represent 2^20 = 1MB memory, to access a 4GB memory space, descriptor uses G bit to indicate whether the limit use 4K or 1 byte for one unit, that means if G bit is set then we get 2^20*4K = 4GB memory, if it is unset then we only use a memory space under 1MB Descriptor tables in The segment descriptors that we defined are grouped together and placed in a continuous memory location. The group arrangement called Descriptor Table. Each descriptor requires 8byte in order to store the info of a particular segment. Descriptor table can contain 8192(8K) descriptors at the max. 25

66 Unit 08 Advanced Microprocessor The maximum length of a descriptor table is a 64Kbytes. The descriptor tables define all the segments used in the when it operates in the protected mode. There are three types of descriptor tables: the global descriptor table (GDT), the local descriptor table (LDT), and the interrupt descriptor table (IDT). The registers used by the 80386to address these three tables are called the global descriptor table register (GDTR), the local descriptor table register (LDTR), and the interrupt descriptor table register (IDTR). These registers are loaded with the LGDT, LLDT, and LIDT instructions, respectively. The local and global descriptor tables hold up to 8192 entries each, and the interrupt descriptor table holds up to 256 entries. A descriptor is indexed from either the local or global descriptor table by the selector that appears in a segment register. Whenever a new selector is placed into one of the segment registers, the accesses one of the descriptor tables and automatically loads the descriptor into a program-invisible cache portion of the segment register. 1) Global Descriptor Table (GDT): This is the main table of descriptors. The same GDT can be used by all programs to refer to segment of memory can have many LDT s but only one GDT. Figure: Global Descriptor Table 2) Local Descriptor Table (LDT) : A multi-tasking system is defined on a per task basis. The main purpose of an LDT would be combined with GDT in order to expand the total number of available descriptors. Generally, each task can have its own LDT and can also be shared with other task. 8. Paging 26

67 Unit 08 Advanced Microprocessor Paging is one of the memory management techniques The paging mechanism allows any linear (logical) address, as it is generated by a program, to be placed into any physical memory page, as generated by the paging mechanism. A linear memory page is a page that is addressed with a selector and an offset in either the real or protected mode. A physical memory page is a page that exists at some actual physical memory location. For example, linear memory location 20000H could be mapped into physical memory location 30000H, or any other location, with the paging unit. This means that an instruction that accesses location 20000H actually accesses location 30000H. Each memory page is 4K bytes long. Paging allows the system software to be placed at any physical address with the paging mechanism. Three components are used in page address translation: the page directory, the page table, and the actual physical memory page. The segmentation scheme may divide the physical memory into a variable size segments but the paging divides the memory into a fixed size pages. The segments are supposed to be the logical segments of the program. The pages are just fixed size portions of the program module or data. Advantage: The advantage of paging scheme is that the complete segment of a task need not be in the physical memory at any time. Only a few pages of the segments, which are required currently for the execution need to be available in the physical memory. Thus the memory requirement of the task is substantially reduced, relinquishing the available memory for other tasks Page Directory The page directory contains the location of up to 1024 page translation tables, which are each four bytes long. Each page translation table translates a logical address into a physical address. The page directory is stored in the memory and accessed by the page descriptor address register (CR3). Control register CR3 holds the base address of the page directory, which starts at any 4Kbyte boundary in the memory system. Each entry in the page directory translates the leftmost 10 bits of the memory address. This 10-bit portion of the linear address is used to locate different page tables for different page table entries Page Table 27

68 Unit 08 Advanced Microprocessor The page table contains 1024 physical page addresses, accessed to translate a linear address into a physical address Page Directory Entry Total Page Directory Entries are1024 Each directory entry is of 4byte 8.4. Page Table Entry Figure: Page Directory Entry The page table entries contain the starting address of the page and the statistical information about the page. Total Entries are1024 Each page table entry is of 4byte Figure: Page Table Entry D-bit: Dirty bit is undefined for page table directory entries by the microprocessor and is provided for use by the operating system. A-bit: Accessed bit is set to logic 1 whenever the microprocessor accesses the page Directory entry. R/W (Read/write) & U/S (user/supervisor) are both used in the protection scheme. Both bits combine to develop paging priority level protection for level 3, the lowest user level. P-bit: Present bit, if logic1 indicates that the entry can be used in address translation. If P = 0, the entry cannot be used for translation. When P = 0, the remaining bits of the entry can be used to indicate the location of the page on the disk memory system. Difference between page directory and page table entry: 28

69 Unit 08 Advanced Microprocessor The main difference is that the page directory entry contains the physical address of a page table, while the page table entry contains the physical address of a 4K-bytephysical page of memory. The other difference is the D (dirty bit), which has no function in thepage directory entry, but indicates that a page has been written to in a page table entry Page Translation Mechanism in A page frame is a 4K-byte unit of contiguous addresses of physical memory. Pages begin on byte boundaries and are fixed in size. A linear address refers indirectly to a physical address by specifying a page table, a page within that table, and an offset within that page Figure: Format of Linear Address The below figure of page translation shows, how processor converts the DIR, PAGE, and OFFSET fields of a linear address into the physical address by consulting two levels of page tables. The addressing mechanism uses the DIR field as an index into a page directory, uses the PAGE field as an index into the page table determined by the page directory, and uses the OFFSET field to address a byte within the page determined by the page table. In the second phase of address transformation, the transforms a linear address into a physical address. This phase of address transformation implements the basic features needed for pageoriented virtual-memory systems and page-level protection. Page translation is in effect only when the PG bit of CR0 is set. Figure: Page Translation 29

70 9. Features of Unit 08 Advanced Microprocessor The 32-bit is the next evolutionary step up from the One of the most obvious feature included in a is a built in math coprocessor. This coprocessor is essentially the same as the processor used with a 80386, but being integrated on the chip allows it to execute math instructions about three times as fast as a 80386/387 combination is an 8Kbyte code and data cache. To make room for the additional signals, the is packaged in a 168 pin, pin grid array package instead of the 132 pin PGA used for the Operates on 25MHz, 33 MHz, 50 MHz, 60 MHz, 66 MHz or 100MHz. It consists of parity generator/checker unit in order to implement parity detection and generation for memory reads and writes. Supports burst memory reads and writes to implement fast cache fills. Three mode of operation: real, protected and virtual 8086 mode. The microprocessor is a highly integrated device, containing well over 1.2 million transistors. 10. Pentium Processor Features It consists of all the features of The additional enhancements that Pentium provides are: 1) Wider data bus width : It has 64 bit data bus and 32 bit address bus. It allows 8 byte of data info to be transferred to and from memory. Bus cycle pipelining has been added to allow two bus cycles to be in progress simultaneously. 2) Improved Cache Structure: 8KB dedicated instruction cache which gives instruction to its execution units and floating point unit via dual instruction pipeline. Cache is organized in a 2 way set associate cache with 32 byte line(256 lines). 8KB data cache which gives data to its execution unit. This allows 32 byte transfer from cache to pre-fetch buffer which is of 64 bytes. 3) Two parallel integer execution unit : It allows the execution of two instructions to be executed simultaneously in a single processor clock. 4) Faster floating point unit : The floating point unit has been completely redesigned over

71 Unit 08 Advanced Microprocessor Faster algorithms provide up to ten times speed up for common operations including add, multiply etc. 5) Branch prediction logic: The Pentium uses tech called branch prediction. To implement this Pentium has two pre-fetch buffers, one to pre-fetch code in linear fashion, and one that pre-fetches code according to the Branch Target Buffer (BTB). Therefore, needed code is almost pre-fetched before it is required for execution. 6) Data Integrity and Error Detection: The Pentium have added significant data integrity and error detection capability. Data parity checking is still byte-by-byte basis. Address parity checking has also been added. 7) Functional Redundancy Checking: (provides maximum error detection) Two or more Pentium Processor can participate in functional redundancy checking. One processor (the master) fetching the instruction and executes the instruction in normal fashion. Other processor (the checker) (connected directly to the master processor s buses) verify correctness of master processor. Checker executes the instruction same as the master but doesn t drive the buses. Checker samples master s output and compares the values with the internal computed values. An error signal is asserted in case if mismatch occurs. 8) Super Scalar Architecture: Processor is capable of parallel instruction execution of multiple instructions are known as superscalar processors. Pentium is capable in some cases of executing two integer of two floating point instruction simultaneously and thus support superscalar architecture Architecture 31

72 Unit 08 Advanced Microprocessor Features of Figure: Pentium Architecture The 32-bit is the next evolutionary step up from the One of the most obvious feature included in a is a built in math coprocessor. This coprocessor is essentially the same as the processor used with a 80386, but being integrated on the chip allows it to execute math instructions about three times as fast as a 80386/387 combination is an 8Kbyte code and data cache. To make room for the additional signals, the is packaged in a 168 pin, pin grid array package instead of the 132 pin PGA used for the Operates on 25MHz, 33 MHz, 50 MHz, 60 MHz, 66 MHz or 100MHz. It consists of parity generator/checker unit in order to implement parity detection and generation for memory reads and writes. Supports burst memory reads and writes to implement fast cache fills. Three mode of operation: real, protected and virtual 8086 mode. The microprocessor is a highly integrated device, containing well over 1.2 million transistors. 32

73 Unit 08 Advanced Microprocessor Explain Architecture of Processor. The architecture of the 80486DX is almost identical to the Added to the architecture inside the 80486DX is a math coprocessor and an 8K-byte level 1 cache memory. BIU: Figure: Architecture BIU generates address, data and control signals for a bus cycle it is supported with an additional parity detection/generation for memory reads and writes. During memory write operation, the 486 generates even parity bit for each byte outputs these bits. These bits will be stored in a separate parity memory bank. During read operation, stored parity bits will be read from the parity memory checks the parities of data bytes read and compare them with the DP0 DP3 signals and generates parity check error, if it occurs. Instruction Pre-fetch Unit : It pre-fetches the instruction bytes in advance and holds them in a 32 byte code queue. Instruction Decoder : Decodes the instructions in the queue and passes the control and protection test unit. Execution Unit: Executes the instruction with the help of Barrel Shifter, ALU and Register bank. 33

74 Unit 08 Advanced Microprocessor Segmentation Unit and Paging Unit : They are part of MMU(which manages virtual memory of system). Helpful in generation of Physical Address. Work same as they work in Floating Point Unit: Responsible for performing the floating point operations. The has a built in math co-processor. It performs the floating point operations. It executes math instructions 3 times faster than the 386/387 combination. Cache Unit: 8KB cache Additional high speed cache memory provides a way of improving overall system performance. It contains the recently used instructions, data or both. The main aim is that the microprocessor unit access code and data in the cache most of time, instead from the main memory. EFLAG Register The extended flag register EFLAG is illustrated in the figure below:- The only new flag bit is the AC alignment check, used to indicate that the microprocessor has accessed a word at an odd address or a double word boundary. Figure: EFLAG Register of The

75 Unit 10 ARM Processor 1. ARM architecture Feature: The ARM architecture has been designed to allow very small, yet high-performance implementations. The architectural simplicity of ARM processors leads to very small implementations, and small implementations allow devices with very low power consumption. The ARM is a Reduced Instruction Set Computer (RISC), as it incorporates these typical RISC architecture features: A large uniform register file A load/store architecture, where data-processing operations only operate on register contents, not Directly on memory contents Simple addressing modes, with all load/store addresses being determined from register contents and instruction fields only Uniform and fixed-length instruction fields, to simplify instruction decode. In addition, the ARM architecture gives you: Control over both the Arithmetic Logic Unit (ALU) and shifter in every data-processing instruction to maximize the use of an ALU and a shifter Auto-increment and auto-decrement addressing modes to optimize program loops Load and Store Multiple instructions to maximize data throughput Conditional execution of all instructions to maximize execution throughput. 2. ARM7TDMI Processor The ARM7TDMI core is a member of the ARM family of general-purpose 32-bit microprocessors. The ARM family offers high performance for very low-power consumption and gate count. The ARM architecture is based on Reduced Instruction Set Computer (RISC) principles. The RISC instruction set, and related decode mechanism are much simpler than those of Complex Instruction Set Computer (CISC) designs. This simplicity gives: a high instruction throughput an excellent real-time interrupt response a small, cost-effective, processor macrocell. The ARM7TDMI processor uses a pipeline to increase the speed of the flow of instructions to the processor. This enables several operations to take place simultaneously, and the processing, and memory systems to operate continuously. 1

76 Unit 10 ARM Processor Figure: ARM7TDMI Core A three-stage pipeline is used, so instructions are executed in three stages: Fetch Decode Execute. The ARM7TDMI processor has a Von Neumann architecture, with a single 32-bit data bus carrying both instructions and data. Only load, store, and swap instructions can access data from memory. Data can be 8-bit bytes, 16-bit half-words, or 32-bit words. Words must be aligned to 4- byte boundaries. Half-words must be aligned to 2-byte boundaries. The ARM7TDMI processor has two instruction sets: The 32-bit ARM instruction set The 16-bit Thumb instruction set 2

77 Unit 01 Introduction to Microprocessor 1. Introduction Microprocessor (MP) is a programmable logic device that is capable of data handling and performing data processing operations. More elaborately, it is Digital device: understands binary. Programmable device: It can perform multiple tasks and can be instructed (i.e. programmed) to perform specific task (within its capability) Clock driven: requires clock for its operation. Capable of data handling: Storing data, communicating data with other devices, etc. Capable of data processing: Performing various arithmetic and logic operations like addition, subtraction, magnitude comparison, ANDing, ORing, etc. Available in the form of an Integrated circuit (IC). 2. Components of Microprocessor Primary component of any programmable system, e.g. computer, and is also referred as Central Processing Unit (CPU). CPU popularly consist of three main units viz. Arithmetic and logic unit (ALU), Register array (memory limited in size) and Control unit (CU). Figure1: Central processing unit ARITHMETIC AND LOGIC UNIT (ALU) This unit performs all the logical and arithmetic operations. Various arithmetic operations are: addition, subtraction, increment and decrement etc. Various logical operations are: AND, OR, NOT, XOR, etc. TIMING AND CONTROL UNIT This unit controls the entire operations being performed by the system. It controls the operations of ALU, input/output devices and memory unit. This unit interprets the instructions and generates various timing and control signals. 1

78 Unit 01 Introduction to Microprocessor REGISTERS A register is a very small amount of very fast memory that is built into the CPU in order to store the current data and instructions which are being executed by the CPU. SYSTEM BUS A bus in a microprocessor-based system is defined as a group of separate wires which work together to perform a particular task. A microprocessor-based system, or microcomputer, has three types of buses which combine to transfer information between the microprocessor and other parts of the system, such as memory or input/output devices. Typical tasks performed by these buses include selecting the source or destination location address for a data transfer, actually moving the data from one part of the system to another, and finally, controlling and synchronizing the electronic devices involved in the data transfer process. The three microprocessor buses are called the address bus, data bus and control bus, and these names give a strong clue as to their function: 1) Address Bus The address bus contains the connections between the microprocessor and memory that carry the addresses at which the CPU is processing at that time, such as the locations that the CPU is reading from or writing to. The width of the address bus corresponds to the maximum addressing capacity of the bus, or the largest address within memory that the bus can work with. The addresses are transferred in binary format, with each line of the address bus carrying a single binary digit. Therefore the maximum address capacity is equal to two to the power of the number of lines present (2^lines). 2) Data Bus This is used for the exchange of data between the processor, memory and peripherals, and is bi-directional so that it allows data flow in both directions along the wires. The number of wires used in the data bus (sometimes known as the 'width') can differ. Each wire is used for the transfer of signals corresponding to a single bit of binary data. As such, a greater width allows greater amounts of data to be transferred at the same time. 3) Control Bus The control bus carries the signals relating to the control and co-ordination of the various activities across the computer, which can be sent from the control unit within the CPU. Each line is used to perform a specific task. For instance, different, specific lines are used for each of read, write and reset requests. 2

79 Unit 01 Introduction to Microprocessor 3. Microprocessor System with Bus Organization To design any meaningful application microprocessor requires support of other auxiliary devices. In most simplified form a microprocessor based system consist of a microprocessor, I/O (input/output) devices and memory. These components are interfaced (connected) with microprocessor over a common communication path called system bus. Typical structure of a microprocessor based system is shown in Figure 2. Figure 2: Microprocessor based system Here, microprocessor is master of the system and responsible for executing the program and coordinating with connected peripherals as required. Memory is responsible for storing program as well as data. System generally consists of two types of memories ROM (Read only and non-volatile) and RAM (Read/Write and volatile). I/O devices are used to communicate with the environment. Keyboard can be example of input devices and LED, LCD or monitor can be example of output device. Depending on the application level of sophistication varies in a microprocessor based systems. For example: washing machine, computer. 3

80 03 Microprocessor Architecture 1. THE MAIN FEATURES OF 8085 MICROPROCESSOR: It is an 8 bit microprocessor. It is manufactured with N-MOS technology. It has 16-bit address bus and hence can address up to 2 16 = bytes (64KB) memory locations through A0-A15 The first 8 lines of address bus and 8 lines of data bus are multiplexed AD0 AD7 Data bus is a group of 8 lines D0 D7 It supports external interrupt request.. A 16 bit program counters (PC) A 16 bit stack pointer (SP) Six 8-bit general purpose register arranged in pairs: BC, DE, HL. It requires a signal +5V power supply and operates at 3.2 MHZ single phase clock. It is enclosed with 40 pins DIP (Dual in line package). 1

81 Microprocessor architecture: 03 Microprocessor Architecture Fig. 1 Block Diagram of

82 03 Microprocessor Architecture The architecture of microprocessor 8085 can be divided into five parts as follows: 2.1 Register Unit: Fig. 2 Registers of 8085 General Purpose Data Register 8085 has six general purpose data registers to store 8-bit data. These registers are named as B, C, D, E, H and L as shown in fig. 1. The user can use these registers to store or copy a data temporarily during the execution of a program by using data transfer instructions. These registers are of 8 bits but whenever the microprocessor has to handle 16-bit data, these registers can be combined as register pairs BC, DE and HL. There are two internal registers W and X. These registers are only for internal operation like execution of CALL and XCHG instructions and not available to the user. Program Counter (PC) 16-bit register deals with sequencing the execution of instructions. This register is a memory pointer. Memory locations have 16-bit addresses which are why this is a 16-bit register. The microprocessor uses this register to sequence the execution of the instructions. The function of the program counter is to point to the memory address from which the next byte is to be fetched. When a byte (machine code) is being fetched, the program counter is incremented by one to point to the next memory location. Stack Pointer (SP) SP is also a 16-bit register used as a memory pointer. It points to a memory location in R/W memory, called the stack. The beginning of the stack is defined by loading 16-bit address in the stack pointer. 2.2 MUX/DEMUX unit This unit is used to select a register out of all the available registers. This unit behaves as a MUX when data is going from the register to the internal data bus. 3

83 03 Microprocessor Architecture It behaves as a DEMUX when data is coming to a register from the internal data bus of the microprocessor. The register select will behave as the function selection lines of the MUX/DEMUX. 2.3Address Buffer Register & Data/Address Buffer Register These registers hold the address/data, received from PC/internal data bus and then load the external address and data buses. These registers actually behave as the buffer stage between the microprocessor and external system buses. 2.4 Control Unit: The control unit generates signals within microprocessor to carry out the instruction, which has been decoded. In reality it causes connections between blocks of the microprocessor to be opened or closed, so that the data goes where it is required and the ALU operations occur. The control unit itself consists of three parts; the instruction registers (IR), instruction decoder and machine cycle encoder and timing and control unit. Instruction Register This register holds the machine code of the instruction. When microprocessor executes a program it reads the opcode from the memory, this opcode is stored in the instruction register. Instruction Decoder & Machine Cycle Encoder The IR sends the machine code to this unit. This unit, as its name suggests, decodes the opcode and finds out what is to be done in response of the coming opcode and how many machine cycles are required to execute this instruction. Timing & Control unit The control unit generates signals within microprocessor to carry out the instruction, which has been decoded. In reality, it causes certain connections between blocks of the microprocessor to be opened or closed, so that the data goes where it is required and the ALU operations occur. 1.5 Arithmetic & Logical Unit: The ALU performs the actual numerical and logical operation such as add, subtract, AND, OR, etc. ALU uses data from memory and from accumulator to perform the arithmetic operations and always stores the result of the operation in accumulator. ALU consists of accumulator, flag register and temporary register. 4

84 03 Microprocessor Architecture Accumulator The accumulator is an 8-bit register that is a part of ALU. This register is used to store 8-bit data and perform arithmetical and logical operations. The result of an operation is stored in the accumulator. It is also identified as register A. Flags register Flag register includes five flip-flops, which are set or reset after an operation according to the data conditions of the result in the accumulator and other registers. They are called zero (Z), carry (CY), sign (S), parity (P) and auxiliary carry (AC) flags; their bit positions in the flag register are shown in fig. 3. The microprocessor uses these flags to set and test data conditions. Fig. 3 Flag Register of 8085 The flags are stored in the 8-bit register so that the programmer can examine these flags by accessing the register through an instruction. These flags have critical importance in the decision-making process of the microprocessor. The conditions (set or reset) of the flags are tested through the software instructions. For instance, JC (jump on carry) is implemented to change the sequence of a program when CY flag is set. Z (Zero) Flag: This flag indicates whether the result of mathematical or logical operation is zero or not. If the result of the current operation is zero, then this flag will be set, otherwise reset. CY (Carry) Flag: This flag indicates, whether, during an addition or subtraction operation, carry or borrow is generated or not, if generated then this flag bit will be set. AC (Auxiliary Carry) Flag: It shows carry propagation from D3 position to D4 position. Fig. 4 Auxiliary Carry Flag 5

85 03 Microprocessor Architecture As shown in the fig., a carry is generated from D3 bit position and propagates to the D4 position. This carry is called auxiliary carry. S (Sign) Flag: Sign flag indicates whether the result of a mathematical operation is negative or positive. If the result is positive, then this flag will reset and if the result is negative this flag will be set. This bit, in fact, is a replica of the D7 bit. P (Parity) Flag: Parity is the number of 1 s in a number. If the number of 1 s in a number is even then that number is known as even parity number. If the number of 1 s in a number is odd then that number is known as an odd parity number. This flag indicates whether the current result is of even parity (set) or of odd parity (reset). 1.6 Interrupt Control The interrupt control unit has 5 interrupt inputs TRAP,RST 7.5, RST 6.5, RST 5.5 & INTR and one acknowledge signal INTA. It controls the interrupt activity of 8085 microprocessor. 1.7 Serial IO control 8085 serial IO control provides two lines, SOD and SID for serial communication. The serial output data (SOD) line is used to send data serially and serial input data line (SID) is used to receive data serially. 6

86 03 Microprocessor Architecture PIN DIAGRAM Fig. 5 pin diagram of 8085 Pin Function: The following section describes pin functions 7

87 ALE (Address Latch Enable): RD (Read): WR (Write): 03 Microprocessor Architecture This is a positive going pulse generated every time the 8085 begins an operation (machine cycle); it indicates that the bits on AD7-AD0 are address bits. This signal is used primarily to latch the low-order address from the multiplexed bus and generate a separate set of eight address lines, A7-A0. This is a Read control signal (Active Low). This signal indicates that the selected memory or I/O device is to be read and data are available on the data bus. This is a write control signal (Active Low). This signal indicates that the data on the data bus are to be written into a selected memory or I/O location. IO/M: S1 & S0: This is a status signal used to differentiate between I/O and memory operations. When it is high, it indicates an I/O operation. When it is low, it indicates a memory operation. This signal is combined with RD (read) and WR (Write) to generate I/O and memory control signals. These status signals, similar to IO/M, can be identify various operations, but they are rarely used in small systems. Encoded status of the bus cycle: AD8 AD15Address Bus S1 S0 Operation 0 0 HALT 0 1 WRITE 1 0 READ 1 1 FETCH Table 1 bus cycle status The most significant 8 bits of the memory address or the 8 bits of the I/0 address,3 stated during Hold and Halt modes. 8

88 03 Microprocessor Architecture READY (Input): If Ready is high during a read or write cycle, it indicates that the memory or peripheral is ready to send or receive data. If Ready is low, the CPU will wait for Ready to go high before completing the read or write cycle. HOLD (Input): This pin indicates that another Master is requesting the use of the Address and Data Buses. The CPU, upon receiving the Hold request will relinquish the use of buses as soon as the completion of the current machine cycle. Internal processing can continue. The processor can regain the buses only after the Hold is removed. When the Hold is acknowledged, the Address, Data, RD, WR, and IO/M lines are 3stated. HLDA (Output): HOLD ACKNOWLEDGE Indicates that the CPU has received the Hold request and it will relinquish the buses in the next clock cycle. HLDA goes low after the Hold request is removed. The CPU takes the buses one half clock cycles after HLDA goes low. INTR (Input): INTERRUPT REQUEST It is used as a general purpose interrupt. It is sampled only during the next to the last clock cycle of the instruction. If it is active, the Program Counter (PC) will be inhibited from incrementing and an INTA will be issued. During this cycle a RESTART or CALL instruction can be inserted to jump to the interrupt service routine. The INTR is enabled and disabled by software. It is disabled by Reset and immediately after an interrupt is accepted. INTA (Output) :INTERRUPT ACKNOWLEDGE; It is used instead of (and has the same timing as) RD during the Instruction cycle after an INTR is accepted. It can be used to activate the 8259 Interrupt chip or some other interrupt port. 9

89 03 Microprocessor Architecture RESTART INTERRUPTS; RST 5.5 RST 6.5 RST 7.5 These three inputs have the same timing as INTR except they cause an internal RESTART to be automatically inserted. RST 7.5 has Highest Priority RST 6.5 & RST 5.5 has Lower Priority. The priority order of these interrupts is as shown above. These interrupts have a higher priority than the INTR. TRAP (Input): Trap interrupt is a nonmaskable restart interrupt. It is recognized at the same time as INTR. It is unaffected by any mask or Interrupt Enable. It has the highest priority of any interrupt. RESET IN (Input) Reset sets the Program Counter to zero and resets the Interrupt Enable and HLDA flipflops. None of the other flags or registers (except the instruction register) is affected. The CPU is held in the reset condition as long as Reset is applied. RESET OUT (Output) Indicates CPU is being reset and can be used as a system RESET. The signal is synchronized to the processor clock. SID (Input) :Serial input data line The data on this line is loaded into accumulator bit 7 whenever a RIM instruction is executed. SOD (output) :Serial output data line. The output SOD is set or reset as specified by the SIM instruction. 10

90 03 Microprocessor Architecture 4. Basic Definitions: Instruction Cycle Machine Cycle T-state It is defined as the time required to complete the execution of an instruction. The 8085 instruction cycle consists of one to six machine cycles or one to six operations It is defined as the time required completing one operation of accessing memory, I/O, or acknowledging an external request. This cycle may consist of three to six T-states. It is defined as one subdivision of the operation performed in one clock period. These subdivisions are internal states synchronized with the system clock, and each T-state is precisely equal to one clock period. 11

91 Unit 06 Stack & Subroutines 1. Stack: The stack is a group of memory locations in Read/Write (R/W) Memory of microcomputer and is used to store the contents of a register, operand and memory address. The starting location of the stack is defined by loading 16-bit address into the stack pointer, that space is reserved, usually at the top of the memory map. Theoretically, the size of the stack is unlimited, but it is restricted only by the available R/W memory in a microcomputer system. The stack can be initialized anywhere in the user memory map, but the stack is initialized at the highest user memory location so that there will not be any interference with the program. In 8085, the beginning of the stack is defined in the program by using the instruction LXI SP, 16 bit. The 16-bit address is loaded into the stack pointer register. Then the contents can be stored in stack memory locations by using the instruction PUSH and can be retrieved from the stack i by using the POP instruction. The microprocessor keeps track of the next available stack memory location by incrementing or decrementing the address in the stack pointer. The address in the stack pointer (register) always points to the top of the stack and indicates that the next memory location (SP + 1) is available to store information. This method of information storage looks like the process of stacking books one above another. Therefore, data is always retrieved from the top of the stack. So data are stored in the stack on Last-In-First-Out (LIFO) principle. The syntax of stack instructions to store data on and retrieve data from the stack are given below: Kishan.K. Govani, EC Department Microprocessor and Interfacing 1

92 Unit 06 Stack & Subroutines Instruction LXI SP, 16-bit address PUSH Rp E.g.: PUSH PSW PUSH B PUSH D PUSH H Description Load 16-bit address into the stack pointer register. This is a load instruction, similar to other 16-bit load instructions. This is a 1-byte instruction and copies the contents of the specified register or index register onto the stack. The instruction first decrements the stack pointer (register) and copies the high-order byte of the register pair of the index register on the stack location SP 1. Then it again decrements the stack pointer and copies the low-order byte of the register pair or the index register onto the stack location SP 2. Instruction POP Rp E.g.: POP PSW POP B POP D POP H Description This is a 1-byte instruction and copies the contents of the top two locations of the stack into the specified register pair or the index register. First, the instruction copies the contents of the stack indicated by SP into the low-order register (for e.g., register C of the BC pair) or as a low-order byte into the index register and then increments the stack pointer to SP + 1. It copies the contents of the SP + 1 location into the high order register (for e.g., register B of the BC pair) or as a high-order byte into the index register and increments the stack pointer to SP + 2. Fig. 1 shows a stack and stack top location. The SP register holds the address of stack top location, i.e. 8004H. Example: A program is stored in memory locations starting from 7000H and the stack is initialized at the location 8004H. Program 7000H LXI SP, 8004H 7001H PUSH D 7002H POP D 7003H HLT Kishan.K. Govani, EC Department Microprocessor and Interfacing 2

93 Unit 06 Stack & Subroutines The position of a stack before PUSH operation is depicted in fig. 1. When the program is executed, the contents of the register pair DE must be pushed to the stack. After the PUSH operation, the stack position has been changes to 8001 H. In the same way, the POP operation is used to transfer the contents from the stack to the register. The stack position before and after the PUSH operation has been given in fig. 2 (a) and (b) respectively. Fig. 3 (a) and (b) shows the stack position before and after the POP operation correspondingly. Fig. 1 Stack and Stack Top Location Kishan.K. Govani, EC Department Microprocessor and Interfacing 3

94 Unit 06 Stack & Subroutines Fig. 2 (a) Stack before PUSH operation (b) Stack after PUSH operation Fig. 3 (a) Stack before POP instruction (b) Stack after POP instruction Kishan.K. Govani, EC Department Microprocessor and Interfacing 4

95 Unit 06 Stack & Subroutines From the above example, following points can be summarized: 1. During the executing of a program, a 16-bit address (8004H) is given to the SP register. The stack space grows upward in the numerically decreasing order of memory addresses. The contents of DE register pairs can be stored beginning from the next location (SP 1). 2. The PUSH instructions are used to store contents of register pairs on the stack, and the POP instructions are used to retrieve the information from the stack. The address in the stack pointer register always points to the top of the stack, and the address is decremented as information is stored or retrieved, respectively. 3. The storage and retrieval of the content of registers on the stack should follow the LIFO sequence. 4. Information in the stack locations may not be destroyed until new information is stored in that memory location. 2. Subroutine: When some operations/functions like multiplication and division are repeatedly performed in a main program like there is time delay between two operations and similar programs. The groups of instructions are written to perform these operations and these groups of instructions are known as subroutine, which will be called by the main program whenever required. When a main program calls a subroutine, the program execution is transferred to the subroutine and after the completion of the subroutine, the program execution returns to the main program. The microprocessor uses the stack to store the return address of the subroutine. For e.g., generally subroutines are written for sine, cosine, logarithms, square root, time delay multiplication functions in 8085 microprocessors. The subroutine is implemented with two associated instructions, namely, CALL and RETURN. CALL is used to call a subroutine and the CALL instruction is written in the main program. RETURN is used to return from the subroutine and the Return instruction is written in the subroutine to return to the main program. When a subroutine is called, the contents of the program counter are stored on the stack, and the program execution is transferred to the subroutine address. When the RETURN instruction is executed at the end of the subroutine, the memory address stored in the stack is retrieved and the sequence of execution is resumed in the main program. The syntax of CALL and RET are given below: Kishan.K. Govani, EC Department Microprocessor and Interfacing 5

96 Unit 06 Stack & Subroutines Instruction Description CALL 16-bit address Call subroutine unconditionally located at the memory address specified by 16-bit operand This instruction places the address of the next instruction on the stack and transfers the program execution to the subroutine address. RET Return unconditionally from the subroutine. This instruction locates the return address on the top of the stack and transfers the program execution back to the calling program. The general characteristics of CALL and RET instructions are given below: 1. The CALL instructions are 3-byte instruction; the second byte specifies the low-order byte, and the third byte specifies the high-order byte of the subroutine address. 2. The RET instructions are 1-byte instructions. 3. A CALL instruction must be used in conjunction with RET instruction (conditional or unconditional) in the subroutine. The following types of subroutines generally are used in microprocessors: 2.1 Multiple Call Subroutines Fig. 1 shows the basic concept of multiple CALL subroutines. This is a subroutine called from many locations in the main program. For example, the DELAY routine is a multiple call subroutine. These types of routines are easy to trace and need minimal stack space. Initially, stack pointer content is XX55H so that the return address can be stored on the stack. When the CALL instruction starts to execute, the subroutine is called from the 8050H memory location. The return address is stored on the stack and the stack pointer is decremented by two locations to XX53H. Kishan.K. Govani, EC Department Microprocessor and Interfacing 6

97 Unit 06 Stack & Subroutines Fig. 1 Multiple Call Subroutines Kishan.K. Govani, EC Department Microprocessor and Interfacing 7

98 2.2 Nested Subroutines Unit 06 Stack & Subroutines When the subroutine is called by another subroutine, it is called nested subroutine. When a subroutine calls another subroutine, all return addresses are stored on the stack. Therefore, only the number of available stack locations limits the extent of nesting. The structure of a nested subroutine is depicted in fig. 2. The main program calls Subroutine I from location 8050H. The address of the next instruction, 8053H, is placed on the stack, and the program is transferred to Subroutine I at 8150H. During the execution of Subroutine I, Subroutine I calls Subroutine II from location 8190H. The address 8193H is placed on the stack and the program is transferred to Subroutine II. The sequence of execution and returns to the main program as shown in fig. 2. Kishan.K. Govani, EC Department Microprocessor and Interfacing 8

99 Unit 06 Stack & Subroutines Fig. 2 Nested Subroutines 2.3 Multiple Ending Subroutines When a subroutine can be terminated at more than one place, it is called a multiple ending subroutine, as illustrated in fig.3. The subroutine has conditional such as RZ Return On Zero and RC Return On Carry. This subroutine has an unconditional return RET. While the Z flag is set, the subroutine returns from location 8050H, and if the CY flag is set, it returns from location 8090H. If neither flag is set, the subroutine returns from location 80A0H. Fig. 3 Multiple Ending Subroutine Kishan.K. Govani, EC Department Microprocessor and Interfacing 9

100 Unit 06 Stack & Subroutines 3. Time Delay Loops Microprocessors perform different operations in sequence and one operation at a time. To complete an operation, some time is required. When some time delay is required between two operations, a time delay loop is used to provide it. 3.1 Time delay using single Register Time delay can be generated using a register. Initially, a register is loaded with an operand or number and then the number is decremented until it reaches zero. So a conditional jump instruction is used in a delay loop to come out from the loop. The time delay depends on the number, which is loaded in the register. Fig. 1 shows the flow chart of time delay loop using one register. Fig. 1 Flowchart for time delay using one register Calculation of Time Delay using One Register The typical instructions of a time delay loop are given below: Kishan.K. Govani, EC Department Microprocessor and Interfacing 10

101 Unit 06 Stack & Subroutines Memory Address Machine Codes Labels Mnemonics Comments T-states 8000H 06H, 08H MVI B, 80H Initialize the B register H 05H LOOP: DCR B Decrement B register H C2H, 02H, 08H JNZ LOOP Jump not zero to LOOP 10/7 MVI B, 80H requires seven clock cycles. DCR B requires 4 clock pulses. JNZ requires 10 clock pulses during execution. When these instructions are executed, MVI B, 80H instruction is executed once and the next two instructions are executed for N times, where N = 80 H = 128 in decimal Hence, DCR B is executed 128 times. The JNZ is also executed for 128 times, out of which (N 1) = 127 times the program jumps to the level LOOP as the content of B register is not equal to zero and takes 10 states each time. When the content of B register becomes zero, JNZ is executed and the program will out from LOOP. For the last execution of JNZ instruction, only 7 T states are required. The detailed execution of instructions and T states are given below: Instructions No. of times instruction executed T states MVI B, 80H 1 7 DCR B 128 (128 x 4) JNZ 128 (127 x 10) + (1 x 7) Total T states = 1796 Kishan.K. Govani, EC Department Microprocessor and Interfacing 11

102 Unit 06 Stack & Subroutines The no. of T states for execution of LOOP is = (N x T states for DCR B) + (N 1) x T states for JNZ + 7 = (128 x 4) + [(128-1) x 10] + [7 x 1] = 1789 T states The delay time to execute the LOOP instruction is, TL = T x no. of T states for execution of LOOP, where T is system clock period When the microprocessor operator in 5 MHz clock frequency, TL = (1/5) x (10-6 ) x 1789 = μs The total time delay TD is calculated from the summation of time to execute instruction of outside LOOP, TOL and time to execute LOOP instruction, TL. TD = TOL + TL = [(1/5) x 10-6 x 7] + [(1/5) x 10-6 x 1789] = μs 3.2 Time delay using Register Pair For example, a 16-bit operand is loaded in the DE register pair. Then DE register pair is decremented by one using DCX D instruction. The DCX instruction does not set the zero flag. Therefore, additional testing will be done using some extra instructions as the JNZ instruction is executed only when the zero flag is set. Fig. 2 Flowchart for time delay using register pair The typical instructions of time delay loop using a register pair are given below: Kishan.K. Govani, EC Department Microprocessor and Interfacing 12

103 Unit 06 Stack & Subroutines Memory Address Machine Codes Labels Mnemonics Comments T- states 8100H 11H, 00H, 80H LXI D, 8000H Initialize the DE register pair H 1BH LOOP: DCX D Decrement DE register pair H 7BH MOV A, E Copy the content of E register in accumulator 8105H B2H ORA D ORing of D with accumulator H C2, 03H, 81H JNZ LOOP Jump not zero to LOOP 10/7 Calculation of Time Delay using Register Pair In time delay loop using register pair, LXI D, 8000H is executed once and the other instructions DCX D, MOV A,E ORA D and JNZ are executed many times. Instructions No. of times instruction executed T states LXI D, 8000H 1 10 DCX D x 4 MOV A, E x 4 ORA D x 4 JNZ [( ) x 10] + [1 x 7] Total T states = The no. of T-states for execution of LOOP is, TL = (N x T states for DCX D) + (T states for MOV A, E) + (N x ORA D) + (N 1 x T states for JNZ) + 7 {where, N = 8000H = } = [32768 x ( )] + [( ) x 10] + [1 x 7] = If the microprocessor clock frequency is 5 MHz, time delay in LOOP is equal to TL. TL = T x no. of T states for execution of LOOP = (1/5) x 10-6 x Kishan.K. Govani, EC Department Microprocessor and Interfacing 13

104 Unit 06 Stack & Subroutines = msec The total time delay TD is calculated from the summation of time to execute instruction of outside LOOP, TOL, and time to execute LOOP instruction, TL. TD = TOL + TL = [(1/5) x 10-6 x 10] + [(1/5) x 10-6 x ] = msec 3.3Time delay using two loops The time delay can also be generated by using two loops as depicted in Fig. 3. The C register is used in the inner loop (LOOP 1) and B register is used in the external loop (LOOP 2). Here, both B and C registers are loaded with numbers. Kishan.K. Govani, EC Department Microprocessor and Interfacing 14

105 Unit 06 Stack & Subroutines Fig. 3 Flowchart for time delay using two loops The C register is decremented until it becomes zero. When the content of C register is zero, decrement B register. If the content of B register is not zero, load the C register with initial value and repeat the process. The example of time delay using two loops is given below: Memory Address Machine Codes Label s Mnemonics Comments T- states 8200H 06H,08H MVI B, 80H Initialize B register H 0EH,FFH LOOP H 0DH LOOP 1 MVI C, FFH Initialize C register 7 DCR C Decrement C register H C2H,04H,82H JNZ LOOP1 Jump not zero to LOOP1 10/7 8208H 05H DCR B Decrement B register H C2H,02H,82H JNZ LOOP2 Jump not zero to LOOP2 10/7 820CH C9H RET Return to main program 10 In time delay generation using two loops as given above MVI B, 80H is executed once and the other instructions MVI C, FFH, DCR C, JNZ, DCR B and JNZ are executed for many times. The detail execution of instructions and T states are given below: Instructions No. of times instruction executed T states MVI B, 80H 1 7 x 1 MVI C, FFH x 7 DCR C 255 x x 128 x 4 JNZ LOOP1 255 x 128 [(255-1) x 128 x 10] + [1 x 7] DCR B x 4 JNZ LOOP2 128 [(128-1) x 10] + [1 x 7] Kishan.K. Govani, EC Department Microprocessor and Interfacing 15

106 Unit 06 Stack & Subroutines RET 1 10 x 1 Total T states = Time delay will be calculated based on the time delay for LOOP1 and LOOP2. Total no. of T states for LOOP1 and LOOP2 are equal to = [128 x 7] + [255 x 128 x 4] + [(255-1) x 128 x 10] + [1 x 7] + [128 x 4] + [(128 1) x 10] + [1 x 7] = If the microprocessor clock frequency is 5 MHz, time delay in LOOP1 and LOOP2 is equal to TL, TL= T x no. of T states for execution of LOOP1 and LOOP2 = (1/5) x 10-6 x = msec Total delay time, TD = TOL + TL = [(1/5) x 10-6 x 17] msec = msec 4. BCD Arithmetic 4.1 BCD to Binary Conversion Convert a 2 digit BCD number stored at memory address 2200H into its binary equivalent number and store the result in memory location 2300H. Kishan.K. Govani, EC Department Microprocessor and Interfacing 16

107 Unit 06 Stack & Subroutines Flowchart Program: LDA add MOV B,A ANI 0FH MOV C,A MOV A.B ANI F0H RRC RRC RRC RRC MOV B,A XRA A MVI D, 0AH X: ADD D DCR B JNZ X Kishan.K. Govani, EC Department Microprocessor and Interfacing 17

108 Unit 06 Stack & Subroutines ADD C STA 2300H HLT 4.2 Binary to BCD Conversion Flowchart Program: START: LXI SP,STACK LXI H,BINBYT MOV A,M CALL PWRTEN HLT PWRTEN: LXI H,OUTBUF Kishan.K. Govani, EC Department Microprocessor and Interfacing 18

109 Unit 06 Stack & Subroutines BINBCD: NXTBUF: MVI B, 64H CALL BINBCD MVI B, 0AH CALL BINBCD MOV M,A RET MVI M,FFH INR M SUB B JNC NXTBUF ADD B INX H RET 4.3 BCD to Seven Segment Code Conversion A set of five packed BCD numbers are stored in memory location starting at 6000H. The seven segment codes of digits 0 to 9 for LED are stored in memory location starting at 6200H and output store at 7000h and select an appropriate seven segment code for each digit. Program: BACK : LXI H,6200H LXI D,6000H LXI B,7000H LDAX D MOV L,A MOV A,M STAX B INX D INX B MOV A,C CPI O5H JNZ BACK HLT Kishan.K. Govani, EC Department Microprocessor and Interfacing 19

110 Unit 06 Stack & Subroutines Kishan.K. Govani, EC Department Microprocessor and Interfacing 20