Intelメーカー80C186XLの使用説明書/サービス説明書
ページ先へ移動 of 405
80C186XL/80C188XL Microprocessor User ’ s Manual.
80C186XL/80C188XL Microprocessor User’s Manual 1995.
Informat ion in this docume nt is provided solely to enable use of Intel products. Intel assumes no liability whatsoe ver, includin g infringe ment of any pat ent or copyrigh t, for sale and use of In tel pro ducts exce pt as pr ovid ed in Int el’s Terms and Con ditio ns of Sale for such produ cts.
iii CONTEN TS CHAPTER 1 INTRODUCTION 1.1 HOW TO USE THIS MANUAL.... ...................... ....................... ...................... ........... ..... 1-2 1.2 RELATED DOCUMENTS ........... ................. ...................... ................
CONTENTS iv 2.3 INTERRUPTS AND EXCEPTI ON HANDLING ............ ....................... ........... ........... ... 2-39 2.3.1 Interru pt/Exc eption Processi ng ........ ........... ....................... ...................... ...............2-39 2.
v CONTENTS CHAPTER 4 PERIPHERAL CONTROL BLOCK 4.1 PERIPHERAL CONTROL REGISTERS ........... ....................... ................ ...................... 4-1 4.2 PCB RELOCATION REGISTER ............ ...................... ............ ........... ...
CONTENTS vi 6.4.5 Memory or I/O Bus Cycle Decodi ng .................. ........... ........... ........... ............ .........6-17 6.4.6 Progra mming Con s ideratio ns ................. ...................... ...................... ................
vii CONTENTS 8.4 PROGRAMMING THE INTERRUPT CONTROL UNIT ...................... ...................... ... 8-11 8.4.1 Interru pt Contro l Regis ters ............... ...... ...................... ...................... .....................8-12 8.4.2 Interru pt Requ est Register .
CONTENTS viii 10.1.3 DMA Requests ........... ........... ................. ...................... ...................... .....................10-3 10.1.4 Extern a l Requ e sts ...................... ....................... ...................... ....
ix CONTENTS 11.3.1.4 Trans cendenta l Instruction s .................... ................. ........... ........... ........... .... 11-5 11.3.1.5 Const ant Ins tructions .................... ...................... ...................... ..............
CONTENTS x FIGURES Figure Pa ge 2-1 Simplif ied Function al Bloc k Diagram of the 80C186 Family CPU ............... ...... ..... ......2-2 2-2 Physical Addre s s Generati on .................... ...................... ................. ..... ...... ..
xi CONTENTS FIGURES Figure Pa ge 3-15 Generating a Normally No t-Ready Bus Signal ................ ...................... .....................3-16 3-16 Generating a Normally Re ady Bus Signa l ...... ....................... ...................... ...
CONTENTS xii FIGURES Figure Pa ge 6-11 Wait State and Ready Con trol Functions ....... ....................... ...................... ...............6-16 6-12 Using Chi p-Selects Durin g HOLD ............. ........... ........... ........... .........
xiii CONTENTS FIGURES Figure Pa ge 10-3 Source-Sy nchronized Transfe rs ................ ..... ....................... ...................... ...............10-5 10-4 Destinat ion-Synchroni z ed Tra nsfers .............. ....................... ......
CONTENTS xiv TABLES Table Pa ge 1-1 Comparis on of 80C186 Mod ular Core Family Products ........ ........... ........... ........... ...... 1-2 1-2 Relate d Docume nts and Software .................. ....................... ...................... ....
xv CONTENTS TABLES Table Pa ge C-1 Instruction Format Vari ables.......... ............ ...................... ...................... ...................... C-1 C-2 Instruction Opera nds ..................... ....................... ..................
CONTENTS xvi EXA MPLES Examp le Page 5-1 Initiali zing the Power Management Unit for Power-Save Mode .................. ...... .........5-14 6-1 Initiali zing the Chip -Select Unit ............ ........... ....................... .....................
1 Introduction.
.
1-1 CHAPTER 1 INTRO DUCTION The 8086 micropro cessor was first introduced in 1978 and g ained rapid support as the microco m- puter engine of cho ice. There are literally millions of 8086/8088-based systems in the world to- day. The amo unt of software written for the 8086/8088 is rivaled by n o other architecture.
INTRODUCTION 1-2 The 80C186 Mod ular Core family is the direct result of ten years of Intel develop ment. It offers the designer the p eace of mind of a well-established architecture with the benef its of state-of-the- art tec hnolo gy.
1-3 INTRODUCTION Each chap ter covers a specific section of t he device, beg i nning with the CPU core. Each perip h- eral chapter includ es programmin g ex amples intended to aid in you r under standing of device o p- eration.
INTRODUCTION 1-4 1.3 ELECTRONIC SUPP ORT SYSTEM S Intel’s FaxBack* service and app li cation BB S provide up-to -date technical information. Intel also maintains several fo rums on Compu S erve and offer s a variety of infor mati on on the World Wide Web.
1-5 INTRODUCTION The fo l lowing catalog s and informatio n are available at the ti me of publication: 1. Solutions OEM subscription form 2. Microcontr oller and flash catalog 3. Developmen t tools catalog 4. Systems catalog 5. Multimedia catalog 6. Multib us and iRMX ® software catalog an d BBS file listings 7.
INTRODUCTION 1-6 1.3.2.1 How to Find Ap BUILDER Software a nd Hypertext Doc uments on the BBS The latest Ap BUILDER files and hypertext man uals and data sheets are available first from the BBS. To access the files, comp l ete these steps: 1. Ty pe F from the BBS Main menu.
1-7 INTRODUCTION 1.5 PRODUCT LITERATURE You can order product literature f rom the followin g Intel literature cen ters. 1-800-468-8118, ext. 28 3 U.S. and Canada 708-296-93 33 U.S. (from overseas) 44(0)17 93-431155 Europe (U.K.) 44(0)1793-421333 Germany 44(0)1793-421777 France 81(0)120-47-88- 32 Japan (fax only) 1.
.
2 Overview of the 80C186 Family Architecture.
.
2-1 CHAPTER 2 OVERVIEW OF THE 80C186 FAMI LY ARCHITECTU RE The 80C1 86 Modular Micro processor Cor e shares a common base architecture with the 8086, 8088 , 80186, 8018 8, 80286, Intel386™ and Intel486™ pro cessors.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-2 Figure 2-1. Simplifie d Functio nal Bl ock Diagram of the 80C1 86 Fami ly CPU 2.1.1 Execution Unit The Execution U nit executes all instructions, provides data and addr es ses to the Bus Interface Unit and manipu l ates the general registers and the Processor Status Word.
2-3 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE The Execution Unit does not con nect directly to the system bus. It o btains instructions from a queue mainta ined by the Bus Interface Un it. When an instruction req uires access to memory or a peripheral device, th e Execution Unit requests the Bus Interface Unit to read and wr ite data .
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-4 During per iods when the Execution Unit is busy executing instructions, the Bus Interface Unit sequentially prefetches instructions from memo ry. As long as the prefetch queue is partially full, the Execu tion Unit fetches instructions.
2-5 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE The data registers can be addressed by their upper or lower halves. Each data reg ister can be used interchang eably as a 16-bit register or two 8-bit reg i sters. The pointer registers are always acces s- ed as 16-bit values.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-6 Figure 2-4 . Segment Registe r s 2.1.5 Instruction Pointer The Bus Interface Unit updates the 16-bit Instr uction Pointer (IP) register so it contains the offset of the next instruction to be fetched.
2-7 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE 2.1.6 Flags The 80C186 Modu lar Core family has six status f l ag s (see F igure 2-5) that the Execution Unit posts as the result of arithmetic or logical operation s. Program branch instru ctions allow a pro- gram to alter its execution depending on con ditions flagged by a prior operation.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-8 2.1.7 Memory Segmentation Programs f or the 80 C186 Modular Core family view the 1 Mby te memory space as a gro up of user-defin ed se gments. A segment is a lo gical unit of memory th at can be up to 6 4 Kbytes long.
2-9 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE Figure 2-5 . Processor Stat us Word Register Name: Processor Status Word Register Mnemonic: PSW (FLAGS) Register Function: Posts CPU statu s inf ormation. Bit Mnemo nic B it Name Res et State Func tion OF O verf low Flag 0 If OF is set, an arithmetic overflow has occurred.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-10 Figure 2 -6. Segm ent Locations in P hysical Memory The four segm ent registers point to four “currently ad dressable” segments (see Figure 2-7). The curren t ly add ressable segments pro vide a work space consis ting of 64 Kbytes for co de, a 64 Kbytes for stack and 128 Kbytes for data storag e.
2-11 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE Figure 2-7. Curren tly Address able Seg ments The segment register is automatically selected according to th e rules in Table 2-2. All information in one segment ty pe generally shares the same lo gical attributes (e.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-12 Figure 2 -8. Logi cal and Physica l Address Physical Address Segment Base Logical Addresses 2C4H 2C3H 2C2H 2C1H 2C0H 2BFH 2BEH 2BDH 2BCH 2BBH .
2-13 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE Instructions are always fetched from the current code segmen t. The IP register contain s the in- struction’s off s et from th e beginning of the segment. Stack instruction s always o perate on the cur- rent stack segment.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-14 Figure 2 -9 . Dynam ic Code Relocation To be dynamically reloc at able, a pro gram must not load or alter its segment registers and must not transfer directly to a lo cation o utside the curr ent code segment.
2-15 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE 2.1.10 Stack Implementation Stacks in the 80C186 Modu lar Core family reside in memo ry space. They are located by the Stack Segment reg ister (SS) and the Stack Pointer (SP). A system can have multiple stacks, but only one stack i s directly addre s sable at a time.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-16 Figure 2-10. St ack Operati on A1013-0 A 1060 1062 105E 105B 105A 1058 1056 1054 1052 1050 22 00 44 66 88 AA 34 45 89 CD 33 11 55 77 99 BB 12 67 AB EF .
2-17 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE 2.2 SOFTWARE OVERV IEW All 80C186 Modular Core family m embers execute the same instruction s . This includes all the 8086 /8088 instruction s plus several addition s and enhancem ents (see Append ix A, “8 0C186 In- struction Set Additions and E xtensions”).
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-18 2.2.1.1 Data Tra nsfer Inst r uctions The instruction set contains 14 data transfer instructions. Th es e instructions move single by tes and words b etw een m emory and registers. They also m ove single b ytes and words between the AL or AX register and I/O ports.
2-19 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE Figure 2-11. F lag Storage Format 2.2.1.2 Arithmeti c Instructi ons The arithmetic in structions (see Table 2-4) op erate on four types of n u m bers: .
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-20 Table 2-5 shows the interpr etations of variou s bit patterns according to number type. Binary nu m- bers can be 8 or 16 bits long . Decimal numbers are stored in bytes, two digits per by te for pack ed decimal and one digit per byte fo r unpacked decim al.
2-21 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE 2.2.1.3 Bit Manipulatio n Instructio ns There are three grou ps of instructions for manipulating bits within byte s and words. These three groups are logical, shifts and rotates. Table 2-6 lists the bit manipulation instructions and their function s.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-22 Individual bits in bytes and words can also be rotated. The pr ocessor does not discard the bits ro- tated out of an operand.
2-23 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE String instruction s automatically update the SI reg i ster, the DI re gister, or both, bef ore processing the next string element. The Direction Flag (DF) determines whether the ind ex registers are auto- incremented (DF = 0 ) or auto-decremen ted (DF = 1).
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-24 Uncond i tional transfer instructions can t ransfer co ntrol either to a tar get instruction within the curren t code segment ( i ntrasegment transfer) or to a different cod e segment (intersegmen t trans- fer).
2-25 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE Table 2-9 . Progra m Transfer Instructio ns Conditional Transfers JA/JNBE Jum p if above/not below nor equal JAE/JNB Jum p if above or equal/not be low.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-26 Iteration control instruction s can be used to regulate the repetition of software loops. These in- structions use the CX register as a counter.
2-27 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE 2.2.1.6 Processor Con tr ol Instructi ons Processor control instr uctions (see Table 2-11) allow programs to control var ious CPU functions.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-28 Immediate operand s are constant d ata contained in an in struction. Imm ediate data can be either 8 or 16 b i ts in leng t h. Immediate operands ar e available directly from the inst ruction qu eue and can be accessed quickly.
2-29 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE Figure 2-12 . Memory Address Comp utation The displacement is an 8- or 16-bit n u mber contained in the instruction. The displacement gen- erally i s derived from th e position of th e operand ’s name (a variable or label) in the prog ram.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-30 The BX or BP register can be specified as the base register for an e ffective a ddress calculation. Similarly, either the SI or the DI register can be specified as the inde x register. The displacement value is a constant.
2-31 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE Figure 2 -14. Reg ister Indirect Address ing Figure 2 -15. Bas ed Addressing Based add res sing pro vides a s imple way to ad dress data s tructures that may be located in different places in memory (see Figure 2-1 6).
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-32 Figure 2-16. Acc essing a Structure with Based Add ressing With indexed addressing, the effective add ress is calculated by summing a displacement and the contents of an index register (SI or DI, s ee Figure 2-1 7).
2-33 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE Figure 2-17. In dexed Addressing Figure 2-1 8. Accessin g an Array with Indexed Addre ssing EA DI SI Opcode Mod R/M Displacement or + A1020-0A Displace.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-34 Based in dex addressing gen erates an effective address that i s the sum of a base register , an index register and a displacement (see Figure 2-1 9). The two address componen ts can be determ ined at execution ti me, making this a v ery flexible addressing mo de.
2-35 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE Figure 2- 20. Acce ssing a Stacked Array with Base d Index Addressing Displacement EA High Address Index Register 6 12 Base Register Displacement EA In.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-36 Figure 2-21. String Operand 2.2.2.3 I/O Port Addressin g Any memory o peran d addressing modes can be used to access an I/O port if the p ort is memor y- mapped. String instructions can also be used to transfer data to memo ry-mapped ports with an appro priate hardware interface.
2-37 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE 2.2.2.4 Data Ty pes Use d in the 80C186 Modu lar Core Family The 80C1 86 Modu lar Co re family supp orts the data types described in Tab le 2-1 2 an d illustrated in Figure 2 -23. In general, ind ividual data elements must fit within defined segment limits.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-38 Figure 2 -23. 80 C186 Modular Core Fami ly Supported Data Typ es 15 +1 87 0 0 16 +2 23 24 +3 31 NOTE: *Directly supported if the system contains an 80C187.
2-39 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE 2.3 INTERRUPTS AND EXCEPTION HAND LING Interru pts and exception s alter program execu t ion in re s ponse to an exter nal event or an error condition . An interrupt han dles asynchrono us external events, for ex ample an NMI.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-40 Figure 2-25. I nterrupt Vecto r Table When an interru pt is acknowledg ed, a common event sequence (Fig ure 2-26) allows the proces- sor to execu te the interrupt service ro utine. 1. The processor saves a partial machine status by pushing th e Processor Status Word onto the stack.
2-41 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE 2. The Trap Flag bit and Interrupt Enable bit are cleared in the Processor Status Word. This prevent s maskable interrupts or single step ex ceptions from interrupting the processor during the in terrupt service ro utine.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-42 Figure 2-2 6. Interrupt Sequ ence 2.3.1 .1 Non-Maskable Interrupts The Non-Mask able Interrupt (NMI) is the highest priority interrup t . It is usually reserved for a catastrophic ev ent such as impending po wer failure.
2-43 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE 2.3.1.2 Maskab le Interrupts Maskable inter rup ts are the mo st com mon way to serv ice external har dware interr upts. Software can glo bally enable or disable maskable interru pts. This is done by setting or clearing t he Inter- rupt Enable b it in the Processor Status Word.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-44 Breakpoint Interrupt — Type 3 The Breakpo i nt Interrupt is a single-byte v ersi on of the INT instruction . It is commonly used by software debu ggers to set breakpoints in RAM. Because the instruction is only one byte long, it can substitute for any instruction.
2-45 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE 2.3.2 Software Interrupts A Software I nterrupt is cau sed by execu ting an “I NT n ” instru ction. The n p arameter corr espond s to the specific interrupt type to be executed. The interrup t type can be any number between 0 and 255.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-46 2.3.4 Interrupt Response Time Interrupt r es ponse time is the time fr om the CPU recognizing an interru pt u ntil the fir st instruction in the service routine is executed . Interrupt response time is less for inter rupts or exception s which sup ply their own vector type.
2-47 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE Only the single step exception can occu r concurrently with another excep tion. At most, two ex- ceptions can occur at the same in structio n bou ndary and one of tho se excep tions must be the sin- gle step.
OVERVIEW OF THE 80C18 6 FAMILY ARCHITECTURE 2-48 Single step priority is a special case. If an interrupt (NMI or maskable) occu rs at the same instruc- tion boun dary as a sing le s tep , th e interr upt vector is taken first, then is followed immediately b y the sing le step vector.
2-49 OVERVIEW OF THE 80C1 86 FAMILY ARCHITECTURE Figure 2 -30. Simultan eous NMI, Single Step a nd Maskable Interrupt A1034-0A NMI Push PSW, CS, IP Fetch Divide Error Vector IRET Divide Timer Inter.
.
3 Bus Interface Unit.
.
3-1 CHAPTER 3 BUS INTERF ACE UNIT The Bus Interface Unit ( BIU) generates bus cycles that prefetch instructions from memory, pass data to and fr om the execution unit, and pass data to an d from th e integrated peripheral units. The BIU driv es addr ess, data, status an d control infor mation to define a b us cycle.
BUS INTERFACE UNIT 3-2 Figure 3-1 . Physical Da ta Bus Models Byte transfers to ev en ad dresses transfer inf orm ation over the lower half o f the data b us (see Fig- ure 3-2). A0 low enab les the lower bank, while BHE hig h disables the upper bank. The data value from the upper bank is ign ored during a bus read cycle.
3-3 BUS INTERFACE UNIT Figure 3- 2. 16-Bit Data Bus Byte Transfers Even Byte Transfer Odd Byte Transfer A19:1 D15:8 D7:0 A0 (Low) BHE (High) A19:1 D15:8 D7:0 A0 (High) BHE (Low) Y (X) Y + .
BUS INTERFACE UNIT 3-4 Figure 3-3. 16-Bit Data Bus Even Word Transfers During a byte read oper ation, the BIU floats the en t ire 16-bit data bus, even though the transfer occurs on only one h alf of the bus. This action simplifies the decoding r equirements for read-only devices (e.
3-5 BUS INTERFACE UNIT Figure 3-4 . 16-Bit Da ta Bus Odd Word Tra nsfers 3.2.2 8-Bit Data Bus The memo ry address space on an 8-b it data bus is physically implem ented as one bank of 1 Mbyte (see Figure 3-1 on page 3-2). Add ress lines A19:0 select a specific byte within the b ank.
BUS INTERFACE UNIT 3-6 For word tran sfers, the word address defines the f irst byte transferre d. The second byte tr ansfer occurs from the word address plus one. Figure 3-5 illustr ates a word transfer on an 8- bit bus in- terface. Figure 3-5 . 8-Bit Data Bus Word T ransfers 3.
3-7 BUS INTERFACE UNIT 3.3.1 16-Bit Bus Memory a nd I/O Requirements A 16-bit bus has certain as sumptions that mu st be met to operate proper ly. Memor y used to store instru ct ion oper ands (i.e., the program) and immediate data must be 16 bits wide.
BUS INTERFACE UNIT 3-8 Figure 3 -6. Typi cal Bus Cycle Figure 3-7 . T-State Relation to CLKOUT Figure 3-8 shows the BIU state diagram. Typically a bus cycle consists of four consecutive T- states labeled T1, T2, T3 and T4. A TI (idle) state occu rs when no bu s cycle is pending.
3-9 BUS INTERFACE UNIT The address/status phase starts just before T1 and continues through T1. The data phase starts at T2 and con t inues through T4.
BUS INTERFACE UNIT 3-10 Figure 3-9 . T-State and Bu s Phases 3.4.1 Address/Sta tus Phase Figure 3-1 0 shows signal timing relationships for the addr ess/status phase of a b us cycle. A bus cycle begins with the tran sition of ALE and S2:0 . These signals transition during phase 2 of the T-state just prio r to T1.
3-11 BUS INTERFACE UNIT Figure 3-10. Ad dress/Statu s Phase Signal Relationship s ALE AD15:0 A19:16 CLKOUT S2:0 BHE T4 or TI T1 T2 1 4 2 3 5 6 Valid Valid NOTES: 1. T CHLH T CHSV : Clock high to ALE high, S2:0 valid. 2. T CLAV : Clock low to address valid, BHE valid.
BUS INTERFACE UNIT 3-12 Figure 3-11 . Demultiplex ing Addres s Informat ion Table 3-1 . Bus Cycle Type s Status B it Op eration S2 S1 S0 0 0 0 Interrupt Acknowledge 001 I / O R e a d 0 1 0 I/O Write 0.
3-13 BUS INTERFACE UNIT 3.4.2 Data Phase Figure 3-1 2 shows the timing relationships for the data phase of a bus cycle. Th e only bus cycle type that does not h ave a data phase is a bus halt.
BUS INTERFACE UNIT 3-14 Figure 3 -12. Data Phase Signal Relat ionships AD15:0 Write AD15:0 Read S2:0 CLKOUT T2 T3 or TW T4 or TI RD/ WR 1 4 2 3 5 6 7 Valid Read Data Valid Write Data NOTES : 1. T CLRL/CLWL, T CLOV : Clock low to valid RD/WR active, write data valid.
3-15 BUS INTERFACE UNIT Figure 3-13. T ypical Bu s Cycle with Wait States Figure 3 -14. ARDY and SRDY Pin Block Diagram ALE S2:0 A19:16 AD15:0 READY WR CLKOUT T1 T2 T3 TW TW T4 Valid Address Address V.
BUS INTERFACE UNIT 3-16 A normally not-ready system is one in which ARDY and SRDY remain low at all times except to signal a ready co ndition. For any bus cycle, only the selected d evice drives either read y input high to comp lete the bus cycle.
3-17 BUS INTERFACE UNIT Figure 3-1 6. Generat ing a Normally Ready Bus Signal The ARDY input ha s two major timing concerns that can affect whethe r a normally ready or nor- mally not-r eady signal may be required . Two latches capture the state of the ARDY input ( s ee Figure 3-14 on page 3-15).
BUS INTERFACE UNIT 3-18 Figure 3 -17. Norma lly Not-Ready System Timing A valid not-ready input can be generated as late as phase 1 of T3 to insert w ait states in a normally ready system. A normally not-read y system must run wait states if the not-ready condition can not be met in time.
3-19 BUS INTERFACE UNIT Figure 3 -18. Norma lly Ready System Timings Conditions causing the BIU to become idle in clude the following. • The instruction pref etch queue is full. • An effective address calculation is in progress. • The bus cy cle inherently requ ires idle states (e.
BUS INTERFACE UNIT 3-20 An idle bus state m ay o r may not dr ive the bus. An idle bus state following a bu s read cycle con- tinues to float the bus. An idle bus state follo w ing a bus write cycle continues to driv e t he bus. The BIU drives no con trol strobes active in an idle state except to indicate the start of another bu s cycle.
3-21 BUS INTERFACE UNIT T OE , T ACC and T CE d efine the maximum data access requirements for the mem ory device. These device parameters must be l ess than the value calculated in the equ at ion column. An equal to or greater than result ind icat es that wait states must be inserted into the bus cycle.
BUS INTERFACE UNIT 3-22 3.5.1.1 Refresh Bus Cycles A refresh b us cycle operates similar ly to a normal read bus cycle except for the following : • For a 16-b it data bus, address bit A0 and BHE drive to a 1 (high) and the data value on the bus is ignored.
3-23 BUS INTERFACE UNIT Figure 3- 21. Typ ical Write Bus Cycle Figure 3-2 2 illustrates a typical 16-bit interface conn ection to a read/write device. Write bus cy- cles have many parameters that must be eva lu ated in deter mining th e compatibility of a memory (or I/O) device.
BUS INTERFACE UNIT 3-24 Most memory and per ipheral devices latch data on the rising edge of the write strobe. Address, chip-select and d ata must be v alid (set up) prior to the rising edge of W R . T AW , T CW an d T DW de- fine the minimum data se t up requirem ents .
3-25 BUS INTERFACE UNIT The minimum device data hold time (fro m WR high) is defined by T DH . The calculated value must be greater than the minimum device requ irements; however, the value can be changed only by decreasing the clock rate. T WC and T WP define the minimum time (maximum freq uency ) a device can process write bus cy- cles.
BUS INTERFACE UNIT 3-26 Figure 3-2 3. Interrupt Ackn owledge Bus Cycle T1 T2 T3 T4 CLKOUT ALE TI TI T1 T2 T3 AD15:0 [AD7:0] RD, WR BHE DEN DT/R LOCK S2:0 INTA0 INTA1 A19:16 [A15:8] NOTE: Vector Type is read from AD7:0 only. INTA occurs during T2 in slave mode.
3-27 BUS INTERFACE UNIT Figure 3-24 sho ws a typical 82C59A interface example. Bus ready must be provided to terminate both bus cycles in the interrupt acknowledge sequence.
BUS INTERFACE UNIT 3-28 3.5.4 HALT Bus Cycle Suspending the CPU reduces device power consumptio n and potentially reduces interru pt latency time. The HLT in struction initiates two events: 1. Suspends the Ex ecution Unit. 2. Instructs the BIU to execute a HALT b us cycle.
3-29 BUS INTERFACE UNIT Figure 3-2 5. HALT Bus Cycle Table 3 -6. HALT Bus Cycle Pin States Pin(s) Pin S tate AD15:0 ( AD7:0 f or 8-bit) Float A15:8 (8-b it) Drive Address A19:16 Drive 8H or Zero BHE (16-b it) Drive Last V alue RD , WR , DEN , DT/R , RFSH (8-bit), S2: 0 Drive One 011 CLKOUT ALE S2:0 AD15:0 [AD7:0] [A15:8] A19:16 NOTES: 1.
BUS INTERFACE UNIT 3-30 3.5.5 Temporarily Exiti ng the HALT Bus State A DMA request, refresh request or bus hold req uest causes the BIU to exit the HALT bus state temporar i ly. This can o ccur only w hen in the Active or Idle power managemen t mode.
3-31 BUS INTERFACE UNIT Figure 3-2 7. Returning to HALT After a Refresh Bus Cycle CLKOUT AD15:0 [AD7:0] ALE [A15:8] A19:16 Address Note 1 Note 2 Note 3 NOTES: 1. Previous bus cycle value. 2. Only occurs for BHE on the first refresh bus cycle after entering HALT.
BUS INTERFACE UNIT 3-32 Figure 3-28 . Returning to HALT After a DM A Bus Cycle 3.5.6 Exiting HALT An NMI or an y unmasked INT n interrupt causes the BIU to ex it HALT. The first bus operation s to occur after exiting HALT are read cycles to reload the CS :IP registers.
3-33 BUS INTERFACE UNIT Figure 3 -29 . Exiti ng HALT 3.6 SYSTEM DE SIGN ALTERNATIVE S Most system d es igns require no signals other than those already provided by the BIU.
BUS INTERFACE UNIT 3-34 3.6.1 Buffering the Data Bus The BIU generates two contr ol sign als, DEN an d DT/R , to contro l bidirectional buffers or trans- ceivers. The timing relationship of DEN and DT/R is shown in Figure 3-30. The following con- ditions requ ire transceivers: • The capacitive lo ad on the address/data bus gets too large.
3-35 BUS INTERFACE UNIT Figure 3-3 1. Buffered AD Bus Syste m In a fully b uffered system, DEN directly drives the transceiver o utput enable. A partially buff ered system requires that DEN be qualified with another signal to prevent the transceiver from going active for local bus accesses.
BUS INTERFACE UNIT 3-36 Figure 3-32 . Qualifying DEN with Chip-Selects 3.6.2 Synchronizing Software and Hardware Even ts The executio n sequen ce of a p rogr am and hardware events occur ring within a system are often asynchro nous to each other.
3-37 BUS INTERFACE UNIT The WAIT instruction suspend s program execution until one of two events occurs: an interrupt is generated, o r the TEST inpu t pin is sampled low. Unlike interrupts, the TEST input pin doe s not require th at program execu t io n b e transferred to a new location (i.
BUS INTERFACE UNIT 3-38 In general, p refix bytes (such as LOCK) are considered extensions o f the instructio ns they pre- cede. Inter rupts, DMA requ ests and refresh req uests that occur dur ing exe.
3-39 BUS INTERFACE UNIT Figure 3- 33. Queue Status Ti ming 3.7 MULTI-MASTER BUS SYSTE M DESIGNS The BIU suppor ts protocols for tr ansferring control of the local bus between itself and other de- vices capable of acting as bus masters.
BUS INTERFACE UNIT 3-40 Figure 3 -34. Ti ming Sequence Ente ring HOLD 3.7.1.1 HOLD Bus La tency The duration between the time that the external device asserts HOLD and the time that the BIU asserts HLDA is known as bus laten cy . In Figure 3-34 , the two-clock delay between HOLD and HLDA represents the shortest bus latency .
3-41 BUS INTERFACE UNIT The major factors that influence bu s latency are listed below (in order fro m l ongest delay to short- est delay). 1. Bus Not Ready — As long as the bus remains no t ready, a bus hold requ est cannot be serviced. 2. Locked Bus Cycle — A s long as LOCK remains asserted, a bus hold request cannot be serviced.
BUS INTERFACE UNIT 3-42 Figure 3-35 . Refresh Request Du r ing HOLD The devic e requesting a bu s hold must be able to detect a HLDA pu lse that is one clock in dura- tion.
3-43 BUS INTERFACE UNIT Figure 3-3 6. Latching HLDA The removal of HOLD must be detected for at l east one clo ck c ycle to all ow the BIU to regain the bus and execute a refresh bu s cycle. Sho uld HOLD go active before the refresh bus cycle i s complete, the BIU will release the b us and gen erate HLDA.
BUS INTERFACE UNIT 3-44 Figure 3-3 7. Exitin g HOLD 3.8 BUS CYCLE P RIORITIES The BIU arbitrates req uests for bus cycles from th e Execution Unit, the integrated p eripherals (e.g., Interru pt Control Unit) and external bus masters (i.e., bus hold requests).
3-45 BUS INTERFACE UNIT 6. Internal er ror (e.g ., divide error, over flow) interrupt vector ing sequence. 7. Hardware (e. g., INT0, DMA) i nterrupt vector ing sequence. 8. 80C187 Math Copro cessor error interrup t vectoring sequence. 9 . DMA bus cycles.
.
4 Peripheral Control Block.
.
4-1 CHAPTER 4 PERIPHERAL CO NTROL BL OCK All integrated perip herals in the 80C186 Mod ular Core family are con trolled by sets of register s within an integr ated Peripheral Control Block (P CB).
PERIPHERAL CONTROL BLOCK 4-2 Figure 4-1 . PCB Relocation Register Register Name: PCB Relocati on Reg ister Register Mnemonic: RELREG Register Function: Relocates the PCB within mem ory or I/O space. Bit Mnemo nic Bit Name Reset State Fu nct ion ET Escape Trap 0 If ET is set, the CPU will trap when an ESC instruction is execut ed.
4-3 PERIPHERAL CONTROL BLOCK Table 4-1. Peripheral Control Block PCB Offset Funct ion PCB Offset Functio n PCB Offset Fun cti on PCB Offset Fu nct ion 00H Reserved 40H Reserved 80H Reserved C0H D0SRC .
PERIPHERAL CONTROL BLOCK 4-4 4.3 RESERVE D LOCATIONS Many locations within the Peripheral Contr ol Block are not assigned to any peripheral. Unused locations are r ese rved. Reading from these location s yields an undefined result. If reserved reg- isters are written (for example, during a block MOV instruction) they must be set to 0H.
4-5 PERIPHERAL CONTROL BLOCK 4.4.3 F-Bus Operation The F-Bus functions d ifferently than the external data bus for by te and word accesses. All write transfers on the F-Bus occur as wo rds, regardless of how they are encoded.
PERIPHERAL CONTROL BLOCK 4-6 4.4.3.1 Writing the PCB Re location Register Whenever map ping the Peripheral Contr ol Block to anoth er location, the user sh o uld pr ogram the Relocation Register with a byte write (i.
4-7 PERIPHERAL CONTROL BLOCK As an examp le, to relocate the Peripheral Con trol Block to the me mory range 1 0000-100FFH, the user would program the PCB Relocation Register with th e value 1100H. Since the Relocation Register is part of the Periph eral Control Block , it relocates to word 10000H plus its fixed offset.
.
5 Clock Generation and Power Management.
.
5-1 CHAPTER 5 CLOCK G ENERATION AND POWER MANAGEMENT The clo ck generation and dis tribution circuits prov i de unifo rm clock signals for the Execution Unit, the Bu s Interface Un it and all integrated peripherals.
CLOCK GENERATION AND POWER MANAGEMENT 5-2 5.1.1.1 Oscillator Opera tion A phase shift oscillator operates through positive feedb ack, where a non-inver ted, amplified ver- sion of th e input conn ects back to the inpu t. A 3 60° phase shift around the loop wi ll sustain the feedback in the oscillator.
5-3 CLOCK GENERATION AND POWER MANAGEMENT Choose C 1 an d L 1 comp onent values in the third overtone crystal circuit to satisfy the fo ll owing condition s: • The LC compon ents form an equivalent ser ies resonant circuit at a freq uency below the fund amental frequency .
CLOCK GENERATION AND POWER MANAGEMENT 5-4 To examine th e parallel resonant frequ ency, refer to Figur e 5-3(c), an equivalen t circuit to Figure 5-3(b).
5-5 CLOCK GENERATION AND POWER MANAGEMENT 5.1.1.2 Selec ting Crystal s When specifying crystals, consider these parameters: • Resonance and Load Capacitance — Crystals car ry a parallel or series resonance specifi- cation. The two typ es do not dif fer in constru ct ion, just in t est conditions and expected circuit application.
CLOCK GENERATION AND POWER MANAGEMENT 5-6 An impo rtant consideration when u sing crystals is that the oscillator st art correctly over the volt- age and temper ature ranges expected in operation.
5-7 CLOCK GENERATION AND POWER MANAGEMENT Reset may be either cold (power-up) or warm. Figure 5-6 illu strates a cold reset. Assert the RES input dur ing power supply and oscillator startup. The processor’s pins assume their reset pin states a maximum of 28 X1 per iods after X1 and V CC stabilize.
CLOCK GENERATION AND POWER MANAGEMENT 5-8 Figure 5-6. Cold Reset Wa veform RES AD15:0, S2:0 RD, WR, DEN DT/R, LOCK V cc cc V cc and X1 stable to RES high, approximately 32 X1 periods. UCS, LCS MCS3:0, NCS TMR OUT0 TMR OUT1 PCS6:0 NOTE: CLKOUT synchronization occurs 1 1/2 X1 periods after RES is sampled low.
5-9 CLOCK GENERATION AND POWER MANAGEMENT Figure 5-7. Wa rm Reset W aveform At the second falling CLKOUT edge after samp li n g RES inactive, the processor deasserts RE- SET. Bus activity starts 6½ CLKOUT periods after recognition of RES in the logic high state.
CLOCK GENERATION AND POWER MANAGEMENT 5-10 Figure 5-8. Clock Synchron iza tion at Res et 5.2 POWER MANAGEME NT Many VLSI devices available tod ay use dynamic circuitr y. A dynamic circuit uses a capacitor (usually parasitic gate or diffusion capacitan ce) to st o re information.
5-11 CLOCK GENERATION AND POWER MANAGEMENT 5.2.1 Power-Save Mode Power-Save mo de is a means for reducing o perating current. Power- S ave mode en ables a pro- grammable clock divider in the clock generation circuit. NOTE Power-Save mode can be used to stretch bus cycles as an alternative to wait states.
CLOCK GENERATION AND POWER MANAGEMENT 5-12 Figure 5-9 . Power-Save Register Register Name: Power Save Reg ister Register Mnemonic: PWRS AV Register Function: Enables and sets clock di vision factor.
5-13 CLOCK GENERATION AND POWER MANAGEMENT Figure 5-10. Po wer-Save Cloc k Transition 5.2.1.2 Leav ing Power-Save Mode Power-Save mo de continues until one o f three even ts occurs: execution clears the PSEN bit in the Power-Save Register, an unmasked interr upt occurs o r an NMI occurs.
CLOCK GENERATION AND POWER MANAGEMENT 5-14 Examp le 5-1. Initializ ing the P ower Manageme nt Unit for Power-Save M ode $mod186 name example_PSU_code ;FUNCTION: This function reduces CPU power consumption ; by dividing the CPU operating frequency by a ; divisor.
6 Chip-Select Unit.
.
6-1 CHAPTER 6 CHIP-SELECT UNIT Every system require s some fo rm of component-selection m echanism to enab l e the C P U to ac- cess a specific memory or peripheral device. The signal that selects the memory or periph eral de- vice is referred to as a chip-select.
CHIP-SELECT UNIT 6-2 Figure 6-1 . Common Chip-S elect Generation Method s 6.3 CHIP-SELECT UNI T FUNCTIONAL OVERV IEW The Chip-Select Unit (CSU) decodes bus cycle address and status information an d enables the appropriate chip-select. Figure 6-3 illustrates the timing of a chip-select during a bus cycle.
6-3 CHIP-SELECT UNIT Figure 6 -2. Chip-Se lect Bloc k Diagram = Block Size UCS = Block Size LCS MCS3 MCS2 MCS1 MCS0 = Base Base + 0 Base + 128 Base + 256 Base + 384 Base + 512 Base + 640 Base + 768 PC.
CHIP-SELECT UNIT 6-4 UCS Mapped only to the upper memory address space; selects the BOOT memory device (EPROM or Flash memo ry types). LCS Mapped only to the lower mem ory address space; selects a static memory (SRAM) device that stores the interrupt vector table, local stack, lo cal data, a n d scratch pad data.
6-5 CHIP-SELECT UNIT By combin i ng LCS , UCS and MCS3:0 , you can cover up to 786 Kb ytes of memor y address space. Methods such as those shown in Figure 6-1 on page 6-2 can be used to decode the remaining 256 Kbytes. The PCS6:0 chip-selects access a contiguous, 89 6-byte block of memo ry or I/O address space.
CHIP-SELECT UNIT 6-6 6.4 PROGRAMMING Four registers determine the ope rating char acter istics of the chip-selects. The Peripheral Control Block defines the location of the Chip- Select Unit registers. Table 6-1 lists the registers and their associated progr amming names.
6-7 CHIP-SELECT UNIT Figure 6 -5. UMCS Registe r Definition Register Name: UCS Control Re giste r Register Mnemonic: UMCS Register Function: Controls the op eration o f the UCS chip-select. Bit Mnemonic Bit Name Reset State Fu nction U17:10 Start Address 0FFH Defines the start ing address for the chip-s elect.
CHIP-SELECT UNIT 6-8 Figure 6 -6. LMCS Regis ter Definition Register Name: LCS Con trol Regist er Register Mnemonic: LMCS Register Function: Controls the op eration o f the LCS chip-select. Bit Mnemonic Bit Name Reset State Fu nction U17:10 Ending Address 00H Defines the ending address for the chip-select.
6-9 CHIP-SELECT UNIT Figure 6 -7. MMCS Registe r Definition Register Name: MCS Con trol Regist er Register Mnemonic: MMCS Register Function: Controls the op eration o f the MCS chip-selects. Bit Mnemonic Bit Name Reset State Fu nction U19:13 Start Address XXH Defines the start ing address for the block of MCS chip-select s.
CHIP-SELECT UNIT 6-10 Figure 6- 8. PACS Register Defi nition Register Name: PCS Co ntrol Register Register Mnemonic: PACS Register Function: Controls the op eration o f the PCS chip-selects. Bit Mnemonic Bit Name Reset State Fu nction U19:13 Start Address XXH Defines the start ing address for the block of PCS chip-select s.
6-11 CHIP-SELECT UNIT Figure 6-9 . MPCS Register Definition Register Name: MCS an d PCS Alternate Con trol Regi ster Register Mnemonic: MPCS Register Function: Controls operati on of the MCS and PCS chip- selects. Bit Mnemonic Bit Name Reset State Fu nction M6:0 Block Size XXH Defines the block siz e for the MCS chip-selects.
CHIP-SELECT UNIT 6-12 The UMCS and LMCS registers can be programm ed in any sequence. To pro gram th e MCS and PCS chip-selects, follow this sequence: 1. Program the MPCS register 2. Program the MMCS register to enable the MCS chip-selects. 3. Program the PACS register to enable the PCS chip-selects.
6-13 CHIP-SELECT UNIT 6.4.2 .2 LCS Active Range The LCS starting add ress is fixed at zero in m emory address space; its ending address is the pr o- grammed block size m inus one. Tab le 6.3 defin es the acceptable valu es for the field (U17:1 0) in the LMCS register th at determines the LCS block size and ending address.
CHIP-SELECT UNIT 6-14 Figure 6 -10. MCS3:0 Active Rang es Table 6- 5. MCS Blo ck Size and Start Address Restrictions MPCS Block Si ze Bit s Block Size (Kbytes) MMCS Sta rt Add ress Restriction s M6 M5 M4 M3 M2 M1 M0 0000001 8 None 000001 X 1 6 U13 must be zero.
6-15 CHIP-SELECT UNIT 6.4.2.4 PCS Active Range Each PCS chip-select starts at an offset abo ve the base address progr ammed in the PACS register and is active for 1 28 bytes. The base address can start on any 1 Kbyte memory or I/O address location. Tab le 6-6 lists the active range for each PCS chip-select.
CHIP-SELECT UNIT 6-16 Figure 6 -11. Wa it State and Ready Co ntrol Functions The R2 control b it determines whether the bu s cycle completes normally (r equires bus ready) or unconditionally ( ignor es bus ready ). The R1:0 bits defin e the numb er of wait states to insert into the bus cycle.
6-17 CHIP-SELECT UNIT For examp le, assume MCS3 overlaps UCS . MCS3 is programmed f or two wait states and re- quires bus ready, while UCS is programmed for no wait states and ignores bus ready. An access to the overlapped regio n has two wait states and requires bus read y (the values program med in the R2:0 bits in the MPCS register).
CHIP-SELECT UNIT 6-18 6.5 CHIP-SELECTS AND BUS HOLD The Chip-Select Un it decodes only internally generated ad dress and bus state information. An ex- ternal bus master cannot make use of the Chip-Select Unit. During HLDA, all ch i p-selects remain inactive.
6-19 CHIP-SELECT UNIT Figure 6 -13. Typ ical System L a t c h Processor ALE AD Bus Addr Bus PCS1 DRQ CE D R A M 256K ARDY 20 MCS3:0 UCS PCS0 LCS SRAM 32K Floppy Disk.
CHIP-SELECT UNIT 6-20 Exam ple 6-1. Initializ ing the Chip-Select Unit $ TITLE (Chip-Select Unit Initialization) $ MOD186XREF NAME CSU_EXAMPLE_1 ; External reference from this module $ include(PCBMAP.
6-21 CHIP-SELECT UNIT Example 6-1. In itializing the Chip -Select Unit (Continued ) DRAM_BASE EQU 256 ;window start address in Kbytes DRAM_SIZE EQU 256 ;window size in Kbytes DRAM_WAIT EQU 0 ;wait states DRAM_RDY EQU INTRDY ;ignore bus ready ;The MPCS register is used to program both the MCS and PCS chip-selec ts.
CHIP-SELECT UNIT 6-22 Example 6-1. In itializing the Chip -Select Unit (Continued ) mov dx, MPCS_REG ;ready for PCS lines 4-6 mov ax, MPCS_VAL ;as well as MCS programming out dx, al mov dx, MMCS_REG ;.
7 Refresh Control Unit.
.
7-1 CHAPTER 7 REFRESH CO NTROL UNIT The Ref re s h Control Unit (RCU) simplifies dynamic memory con t roller design with its in tegrat- ed address and clock counters. Figur e 7-1 shows the relationship between the Bus Interface Unit and the Refresh Control Unit.
REFRESH CONTROL UNIT 7-2 7.1 THE ROLE OF THE REFRE SH CONTROL UNIT Like a DMA controller, the Refresh Con trol Unit runs bus cycles independen t of CPU execution. Unlike a DMA co ntroller, however, the Refresh Control Un it does n ot run bus cycle bu rsts nor does it transfer data.
7-3 REFRESH CONTROL UNIT Figure 7-2. Re fresh Control Unit Operation Flow Chart The nine-b it refr esh clock co unter does not wait u ntil the BI U services the refresh r equest to co n- tinue cou nting. This operation ensures that refresh requests occur at the correct interval.
REFRESH CONTROL UNIT 7-4 The BIU does not queue DRAM refresh requ es ts. If the Refresh Control Unit generates ano ther request before the BIU handle s the present req uest, the BIU loses the present request. Howev er, the addr ess associated with the request is not lost.
7-5 REFRESH CONTROL UNIT 7.5 REFRESH BUS CYCLES Refresh bus cycles look exactly like or dinar y memory read bus cycles ex cept fo r the con trol sig- nals listed in Table 7-1. These signals can be ANDed in a DRAM controller to detect a refresh bus cycle.
REFRESH CONTROL UNIT 7-6 Figure 7 -4. Suggeste d DRAM Control Signal Tim ing Relati onships The cycle begins with presentation of the row address. RAS should go active on the falling edge of T2. At the rising edge of T2, the address lines should switch to a colum n address.
7-7 REFRESH CONTROL UNIT 7.7 PROGRAMMING THE RE FRESH CONTROL UNIT Given a specific processor operating frequency and information ab out the DRAMs in the system, the user can program the Refresh Control Unit registers.
REFRESH CONTROL UNIT 7-8 7.7.2.1 Refresh Base Address Registe r The Refresh Base Address Register (Figure 7-6) programs the base (upper seven bits) of the re- fresh add ress. Seven-bit mapping places the refresh address at any 4 Kbyte boundary within the 1 Mbyte a ddre ss sp ace.
7-9 REFRESH CONTROL UNIT Figure 7 -7. Refresh Cloc k Interval Registe r 7.7.2.3 Refresh Control Register Figure 7-8 shows the Refresh Contro l Register. The user may read or write the REN bit at any time to turn the Refr es h Control Unit on or off. The lower nine bits contain the cu rrent nine-b i t down- counter valu e.
REFRESH CONTROL UNIT 7-10 Figure 7-8 . Refresh Control Register 7.7.3 Programming Exa mple Example 7-1 contains sample co de to initialize the Refresh Control Unit. Example 5-2 on page 5-14 shows the additional code to reprogram the Refresh Control Unit upo n entering Po wer-Save mode.
7-11 REFRESH CONTROL UNIT Exam ple 7-1. Initializ ing the Refresh Control Unit $mod186 name example_80C186_RCU_code ; FUNCTION: This function initializes the DRAM Refresh ; Control Unit to refresh the DRAM starting at dram_addr ; at clock_time intervals.
REFRESH CONTROL UNIT 7-12 Example 7-1. Initial izing the Refresh Co ntrol Unit (Continued) 7.8 REFRESH OPE RATION AND BUS HOLD When anoth er b us master controls th e bus, the processor keeps HLDA active as long as the HOLD input remains activ e.
7-13 REFRESH CONTROL UNIT Figure 7 -9 . Regain ing Bus Control to Run a DRA M Refresh Bus Cy cle HLDA CLKOUT HOLD NOTES: 1. HLDA is deasserted; signaling need to run DRAM refresh cycles less than T CLHAV . 2. External bus master terminates use of the bus.
.
8 Interrupt Control Unit.
.
8-1 CHAPTER 8 INTERRU PT CONTROL UNIT The 80C186 Modular Core has a single maskable interrup t input. (See “Interr upts and Excep tion Handling” on pag e 2-39. ) The Interrup t Contro l Unit (ICU) expands th e interr upt capabilities be- yond a single inp ut.
INTERRUPT CONTROL UNIT 8-2 Interrupts eliminate the need for polling by signalling the CPU that a peripheral device requires servicing. Th e C PU then stops executing the main task, saves its state and transfers executio n to the perip heral-servicing code (the interrupt handler ).
8-3 INTERRUPT CONTROL UNIT 8.2.1.1 Interrupt Maskin g There are circu mst ances in which a pro grammer may need to d isable an interrupt so urce tempo- rarily (fo r example, while executin g a ti me-critical section of cod e or servicing a high-pr iority task).
INTERRUPT CONTROL UNIT 8-4 The priority of each source is pro grammable. The Interrupt Con trol register enables the programmer to assign each source a priority that differ s from the default. The prior ity must still be between zero (h ighest) and seven (lowest) .
8-5 INTERRUPT CONTROL UNIT 8.3 FUNCTIONAL OPERATION IN MAS TER MODE This section cove rs the process in which the In terrupt Contro l Unit receives interrup ts and asserts the maskable inter rupt requ es t to the CPU.
INTERRUPT CONTROL UNIT 8-6 8.3.2.1 Priority Re solution Example This example illustrates prior ity resolution. Assume these initial conditions: • the Interr upt Control Unit h as been initialized .
8-7 INTERRUPT CONTROL UNIT 8.3.2.2 Interrupts That Sha r e a Single Source Multiple interru pt requests can share a single interrupt input to the Interrup t Control Unit. (For example, the three timers share a single input.) Althou gh these interrupts share an input, each ha s its own interrupt vector.
INTERRUPT CONTROL UNIT 8-8 Figure 8 -2. Using External 8259A Module s in Cascade Mode 8.3.3.1 Specia l Fully Nested Mode Special fully nested mode is an optional feature normally used with cascade mode. It is applicable only to I NT0 and INT1. In special f ully nested mode, an interrupt req uest is serviced even if its In-Service bit is set.
8-9 INTERRUPT CONTROL UNIT 8.3.4 Interrupt Acknowledge Sequence During the interru pt acknowledge sequence, the Inter rupt C ontrol Unit passes the interrupt type to the CPU. The CPU then multiplies the interrupt type by four to derive the interrupt vector ad- dress in the interrupt vector table.
INTERRUPT CONTROL UNIT 8-10 8.3.6 Edge and Level Trigger ing The exter nal interrupts (INT3:0) can b e programmed f or either ed ge or level triggerin g (see “In- terrupt Control Regi sters” on page 8-12). Both types of triggering are active high.
8-11 INTERRUPT CONTROL UNIT Figure 8-3. Inte rrupt Contro l Unit Late ncy and Response Time 8.4 PROGRAMMING THE INTE RRUPT CONTROL UNIT Table 8-3 lists th e Interrupt Control U nit registers in master mod e with their Peripher al Control Block offset add ress es.
INTERRUPT CONTROL UNIT 8-12 8.4.1 Interrupt Control Registers Each interru pt source has its own Interrupt Control register. The Interrup t Control register allows you to defin e the behavior of each interrupt source.
8-13 INTERRUPT CONTROL UNIT Figure 8-4. Interrupt Control Registe r for Internal Sources Register Name: Interrupt Co ntrol Register (int ernal sources) Register Mnemonic: TCUCON, DMA0CON, DMA1CON Regi.
INTERRUPT CONTROL UNIT 8-14 . Figure 8-5. Interrupt Control Registe r for Noncasc adable External Pi ns Register Name: Interrupt Co ntrol Register (non -cascadable pi ns) Register Mnemonic: I2CON, I3C.
8-15 INTERRUPT CONTROL UNIT Figure 8 -6. Interrupt Control Register for Cascada ble Interrupt Pins Register Name: Interrupt Co ntrol Register (cascad able pins) Register Mnemonic: I0CON, I1CON Registe.
INTERRUPT CONTROL UNIT 8-16 8.4.2 Interrupt Request Register The Interrup t Request register (Figure 8- 7) has one bit for each interrup t s ource. When a source requests an interrupt, its Interrupt Requ es t bit is set (without regard to whether the interru pt is masked).
8-17 INTERRUPT CONTROL UNIT Figure 8-8 . Interrupt Mask Re gister 8.4.4 Priority Mask Register The Priority Mask register (Figure 8-9) contains a three-leve l field that holds a priority value. This register allows you to mask interrupts based on their priority levels.
INTERRUPT CONTROL UNIT 8-18 Figure 8-9 . Priority Mask Regis ter 8.4.5 In-Service Regis ter The In-Service register has a bit for each interrupt source.
8-19 INTERRUPT CONTROL UNIT Figure 8-1 0. In-Serv ice Register 8.4.6 Poll and Poll Status R egisters The Poll and Poll Status registers allow you to poll the I nterrup t Control Unit and service inter- rupts throu gh software. You can read these registers to determine whether an interrupt is pending and, if so, the inter rupt type.
INTERRUPT CONTROL UNIT 8-20 Reading the Poll register (Figure 8-11) acknowledges the pending interru pt, just as if th e C PU had started the in terrupt vectoring seq uence. The Interrupt Co ntrol Unit updates the Interrup t Re- quest, In-Ser vice, Poll, and Poll Status registers, as it does in the norm al i nterrupt ackno wl edge sequence.
8-21 INTERRUPT CONTROL UNIT Figure 8 -12. Poll Status Register 8.4.7 End-of-Interrupt (EOI) Register The End- of- Interrupt reg ister (Figure 8-1 3) i ssues an End-of -Inter rup t (EOI) command to the I n- terrupt Co ntrol Unit, which clears the In-Service bit f or the associated interrupt.
INTERRUPT CONTROL UNIT 8-22 Figure 8-1 3. End-of-Interrupt Register 8.4.8 Interrupt Status Register The Interrupt Status register (Figure 8 -14) contains the DMA Halt bit and one bit for each timer interrupt. The CPU sets the DMA Halt bit to suspend DMA transfers while an NMI is processed.
8-23 INTERRUPT CONTROL UNIT Figure 8-14 . Interrupt Status Regi ster NOTE Do not write to the DHLT bit while Timer/Counter Unit interrupts are enabled . A conflict with the intern al use of the reg ister may cause incorrect processing of timer interru pts.
INTERRUPT CONTROL UNIT 8-24 Figure 8 -15. Interrupt Co ntrol Unit in Slave Mode 8259A/ 82C59A INT INTA Cascade Address Decode INT0 INTA Select 80186 Modular Core V CC IRQ A1194-A0.
8-25 INTERRUPT CONTROL UNIT Figure 8-1 6. Interrup t Sources in Sla ve Mode 8.5.1 Slave Mode P rogramming Some reg isters differ between Slave mod e and Master m ode.
INTERRUPT CONTROL UNIT 8-26 8.5.1.1 Interrupt Vecto r Register The Interrup t Vector Register is used only in Slave mode. In Master mode, the interr upt vector types are fixed ; in Slave m ode they are programmab le. The Interrupt Vector Register is used to specify the five most-signif icant bits of the interrupt vector type.
8-27 INTERRUPT CONTROL UNIT Figure 8-1 7. Interrupt Vector Regis ter (Slave Mode Only) 8.5.1.2 End-Of-Interru pt Register The End- of-Interrupt (EOI) register has the same fu nction in Slave m ode as in Master mo de. However, non-spe cific EOI commands are not suppo rted, so the NSPEC bit is omitted from the register.
INTERRUPT CONTROL UNIT 8-28 Figure 8-1 8. End-of- Interrupt Register i n Slave Mode 8.5.1.3 Other Regis ters The Priority Mask regi s ter is identical in Slave mode and Master mode.
8-29 INTERRUPT CONTROL UNIT 8.5.2 Interrupt Vectorin g in Slave Mode In Slav e mode, the ex t ernal 8259A module acts as the m as ter interrupt con troller. Therefore, in- terrupt ackn owledge cycles are required for every interrupt, includ ing those from integrated pe- ripherals.
INTERRUPT CONTROL UNIT 8-30 External interrupt ack nowledge cycles must be run for every maskable interrupt. Therefo re, the interrup t response time for every interru pt will be 5 5 clocks, as sho wn in Figure 8-21. Figure 8-2 1. Interrup t Response Tim e in Slave Mode 8.
8-31 INTERRUPT CONTROL UNIT 5. Set the mask bit in th e Interrupt Mask register for any interrupts that you wish to disab le. Example 8-1 shows sample code to initialize the Interrup t C ontrol Unit.
.
9 Timer/Counter Unit.
.
9-1 CHAPTER 9 TIMER/COUNT ER UNIT The Timer/Counter Unit can be used in many application s . Some of these applications include a real-time clock, a squar e-wave generator and a digital one-shot. All of these can be implemented in a system design. A real-time clock can be used to update time-dependent memory variables.
TIMER/COUNTER UNIT 9-2 Figure 9-1. Timer/Counter Unit Block Diagram Transition Latch/ Synchronizer Transition Latch/ Synchronizer Timer 0 Registers Timer 1 Registers Timer 2 Registers O.
9-3 TIMER/COUNTER UNIT Figure 9 -2. Counte r Element Multiple xing and Timer Inp ut Synchronization T1IN T1OUT NOTES: 1. T0IN resolution time (setup time met). 2. T1IN resolution time (setup time not met). 3. Modified count value written into Timer 0 count register.
TIMER/COUNTER UNIT 9-4 Figure 9-3 . Timers 0 and 1 Flow Chart Timer Enabled (EN = 1) ? Clear Count Register Start No Yes No Yes Lo to Hi transition on input pin since last service.
9-5 TIMER/COUNTER UNIT Figure 9 -3. Tim ers 0 and 1 F low Chart (Continued) No (Use"B") No Counter = Compare "A" ? Alternating Maxcount Regs (ALT = 1) ? Using.
TIMER/COUNTER UNIT 9-6 When conf igured for internal clocking, the Tim er/C ounter Unit uses the in put pins either to en- able timer counting or to retrigger the associated timer. Extern ally, a timer increments o n low-to- high transition s on its input pin (up to ¼ C LKOUT frequen cy).
9-7 TIMER/COUNTER UNIT Figure 9 -5. Time r 0 and Timer 1 Contro l Registers Register Na me: Timer 0 and 1 Control Registers Register Mnemonic: T0CON, T 1CON Register Function: Defines Ti mer 0 an d 1 operation . Bit Mnemoni c Bit Name Reset State Functio n EN Ena ble 0 Set to enable the timer.
TIMER/COUNTER UNIT 9-8 Figure 9-5. Timer 0 and Tim er 1 Control Registers (Co ntinued) Register Na me: Timer 0 and 1 Control Registers Register Mnemonic: T0CON, T 1CON Register Function: Defines Ti mer 0 an d 1 operation .
9-9 TIMER/COUNTER UNIT Figure 9-6 . Timer 2 Control Regis ter Register Na me: Timer 2 Control Register Register Mnemonic: T2CON Register Function: Defines Ti mer 2 ope ration. Bit Mnemonic Bit Name Reset State Fu nction EN Enable 0 Set to enable the timer.
TIMER/COUNTER UNIT 9-10 Figure 9 -7. Timer Coun t Registers Register Name: Timer Count Re gister Register Mnemonic: T0CNT, T1CNT, T2CNT Register Function: Contains the current timer coun t. Bit Mnemonic Bit Name Reset State Fu nction TC15:0 Timer Count Value XXXXH Contains the current count of the associated timer.
9-11 TIMER/COUNTER UNIT Figure 9-8. Timer Maxcount Comp are Registers 9 .2.1 Initialization Sequence When initializing the Timer / Counter Unit, t he followin g sequence is suggested: 1. If timer in t errup ts will be used, pr ogram interrupt v ectors into the Interrup t Vect or Table.
TIMER/COUNTER UNIT 9-12 9 .2.2 Clock Sourc es The 16-bit Timer Co unt register increments once for each t imer event. A timer even t can be a low-to-hig h transition o n a timer inpu t pin ( Timers 0 and 1), a pulse gener ated ever y fourth CPU clock (all timers) or a timeout of Timer 2 ( Tim ers 0 and 1).
9-13 TIMER/COUNTER UNIT The timer cou nting from its initial count ( usually zero) to its maximum coun t (either Maxco unt Compare A or B) and resetting to zero defines one timing cycle. A Maxcount Compare value of 0 implies a maximum cou nt of 65 536 , a Maxcount Compare value of 1 implies a maximum count of 1, etc.
TIMER/COUNTER UNIT 9-14 When the EXT bit is clear and the RTG bit is set, every low-to- high tran sition on the timer input pin causes the Count reg ister to reset to zero. After the timer is enabled , counting begins only after the first low-to-h igh transition on the input pin.
9-15 TIMER/COUNTER UNIT Figure 9-9 . TxOUT Sign al Timing In dual maximum count mode, the timer outpu t pin indicates which Maxcount Compar e register is currently in use. A low outp ut indicates Maxcount Compar e B, and a high output indicates Maxcoun t Co mpar e A (see Figure 9-4 on page 9-6).
TIMER/COUNTER UNIT 9-16 The inpu t pins for Timers 0 and 1 provide an alternate metho d for enabling and disabling timer counting . When using internal clocking , the input pin can be programmed either to enable the tim- er or to reset the timer co unt, depending on the state o f the Retrigger (RTG) bit in the contro l reg- ister.
9-17 TIMER/COUNTER UNIT 9 .3.2 Sy nchronization and Maximum Frequency All timer inputs ar e latched and synchron ized with the CPU clock. B ecause of the internal logic required to synchronize th e external signals, and th e multiplexing of the counter element, the Timer/Counter Unit can op erate only up to ¼ of the CLKOUT frequency .
TIMER/COUNTER UNIT 9-18 Examp le 9-1. Configuring a Real-Tim e Clock $mod186 name example_80186_family_timer_code ;FUNCTION: This function sets up the timer and interrupt controller ; to cause the timer to generate an interrupt every ; 10 milliseconds and to service interrupts to ; implement a real time clock.
9-19 TIMER/COUNTER UNIT Example 9-1. Con figuring a Real-Time Clock (Continued) lib_80186 segment public ’code’ assume cs:lib_80186, ds:data public _set_time _set_time proc far push bp ;save calle.
TIMER/COUNTER UNIT 9-20 Example 9-1. Con figuring a Real-Time Clock (Continued) sti ;enable interrupts pop si ;restore saved registers pop dx pop ax pop bp ;restore caller’s bp ret _set_time endp timer_2_interrupt_routine proc far push ax ;save registers used push dx cmp _msec, 99 ;has 1 sec passed? jae bump_second ;if a bove or equal.
9-21 TIMER/COUNTER UNIT Exam ple 9-2. Configuring a Square-W ave Generator $mod186 name example_timer1_square_wave_code ;FUNCTION: This function generates a square wave of given ; frequency and duty cycle on Timer 1 output pin. ; ; SYNTAX: extern void far clock(int mark, int space) ; ; INPUTS: mark - This is the mark (1) time.
TIMER/COUNTER UNIT 9-22 Exampl e 9-2. Configu ring a Square-Wa ve Generator (Contin ued) Examp le 9-3. Configuring a Digi tal One-Shot pop dx ;restore saved registers pop bx pop ax pop bp ;restore cal.
9-23 TIMER/COUNTER UNIT Example 9-3. Configurin g a Digital One-Shot (Conti nued) _CMPB equ word ptr[bp+6] ;get parameter off the stack push ax ;save registers that will be push dx ;modified mov dx, T.
.
10 Direct Memory Access Unit.
.
10-1 CHAPTER 10 DIRECT MEMOR Y ACCESS UNIT In many applicatio ns, large blocks of data must be transferred between memory and I/O space. A disk drive, for example, usually read s and writes data in b locks that may be th ousands of bytes long.
DIRECT MEMORY ACCESS UNIT 10-2 When the DMA requ est is granted, the Bus In terface Unit provides the bu s signals for the DMA transfer, while th e DMA channel provides the ad dress information for the source an d destination devices.
10-3 DIRECT MEMORY ACCESS UNIT 10.1.1.1 DMA Transfer Directions The source an d destination ad dress es for a DMA transfer are programmable and can b e in either memory or I/O space.
DIRECT MEMORY ACCESS UNIT 10-4 10.1.4 External Requests External DMA requests ar e asserted on the DRQ pins. The DRQ p ins are sampled on th e falli ng edge of CLKOUT. It takes a minimum of four clocks befor e the DMA cycle is initiated by the BIU (see Figure 10-2).
10-5 DIRECT MEMORY ACCESS UNIT 10.1.4.1 Source Synchroniz atio n A typical source-syn chronized transfer is shown in Figu re 1 0-3. Most DMA-driven p eripherals deassert their DRQ line only after the DMA transfer has begun.
DIRECT MEMORY ACCESS UNIT 10-6 Figure 10-4 . Destination-S ynchroniz ed Trans fers 10.1.5 Internal Requests Internal DMA requests can come fro m either Timer 2 or the system software.
10-7 DIRECT MEMORY ACCESS UNIT 10.1.6 DMA Transfer Counts Each DMA Unit maintains a prog rammable 16-b it transfer count v alue that controls the total numb er of transfers th e channel runs. Th e transfer count is decremented by one af ter each transfer (regardless of data size).
DIRECT MEMORY ACCESS UNIT 10-8 10.1.8 DMA Unit Int errupts Each DMA channel can be programmed to generate an interrupt request when its transfer count reaches zero. 10.1.9 DMA Cycles and the BIU The DMA Unit uses the Bus Interface Unit to perform its transfers.
10-9 DIRECT MEMORY ACCESS UNIT The last point is extr emely importan t when the two channels use different synchronization. For example, con s ider the case in which chann el 1 is programm ed for high priority and destination synchro nization and channel 0 is pr ogrammed for lo w priority and so urce synchro nization.
DIRECT MEMORY ACCESS UNIT 10-10 Figure 1 0-6. Exampl es of DMA Priority 10.1.10. 1.2 Rotating Priority Channel prior it y rotates when the chan nels are program med as both high or both low priority . The highest priority is initially assigned to channel 1 of the module.
10-11 DIRECT MEMORY ACCESS UNIT Two 16-bit Periph eral Control Block regi sters d efine each of the 20-bit pointers. Figures 10.7 and 10.8 show th e layout of the DMA Source Pointer ad dress registers, and Figu res 10.9 and 10.10 show the layout o f the DMA Destination Pointer address registers.
DIRECT MEMORY ACCESS UNIT 10-12 Figure 10-8 . DMA Source Pointer (Low-Orde r Bits) The address space referenced by the source and destination poin ter s is progr ammed in the DMA Control Register for the channel (see Figure 10- 11 on page 10- 15) .
10-13 DIRECT MEMORY ACCESS UNIT Figure 1 0-9 . DMA Destin ation Pointer (High-Order Bits) Register Name: DMA Destinati on Ad dress Pointer (Hi gh) Register Mnemonic: DxDSTH Register Function: Contains the upper 4 bits of the D MA Destination poi nter.
DIRECT MEMORY ACCESS UNIT 10-14 Figure 10-10 . DMA Destination Pointer (L ow-O rder Bits) 10.2.1.2 Selecting By te or Word Si z e Trans fers The WORD bit in the DMA Contro l Register (Figure 10- 11) con tr ols the d ata size for a channel. When WORD is set, the channel transfers data in 16-b i t words.
10-15 DIRECT MEMORY ACCESS UNIT Figure 1 0-11. DMA Contro l Register Register Name: DMA Control Regist er Register Mnemonic: DxCO N Register Function: Controls DMA chan nel pa rameters. Bit Mnemonic Bit Name Reset State Fu nction DMEM Destination Address Space Select X Selects memory or I/O space for the destination pointer.
DIRECT MEMORY ACCESS UNIT 10-16 Figure 10 -11. DMA Control R egister (Cont inued) Register Name: DMA Control Regist er Register Mnemonic: DxCO N Register Function: Controls DMA chan nel pa rameters. Bit Mnemonic Bi t Nam e Reset State Fu nction TC Term inal Count X Set TC to terminate transfe rs on Terminal Count.
10-17 DIRECT MEMORY ACCESS UNIT Figure 10 -11. DMA Control R egister (Cont inued) 10.2.1.3 Selecting t he Source of DMA Requests DMA requests can com e from either an internal source ( Timer 2) or an external source.
DIRECT MEMORY ACCESS UNIT 10-18 10.2.1.4 Arming the DMA Channel Each DMA channel mu s t be armed befo re it can recognize DMA requests. A channel is armed by setting its STRT (Start) bit in the DMA Contro l Register (Figure 10-11 on page 1 0 -15). The STRT bit can be modified only if the CHG (Change Start) bit is set at the same time.
10-19 DIRECT MEMORY ACCESS UNIT Figure 10 -12. Transfer Co unt Regist er The TC bit, when set, instructs the DMA ch annel to disarm itself (by clearing the STRT bit) when the transf er count reaches zero. If the TC bit is cleared, the chann el continues to perform transfer s regardless of the state of th e Transfer Coun t Reg ister.
DIRECT MEMORY ACCESS UNIT 10-20 10.2.2 Suspension of DMA Transfer s Whenever the CPU receives an NMI, all DMA activity is suspended at the end of th e current transfer. The CPU suspends DMA activity by setting the DHLT bit in the Interrupt Status Regis- ter (Figur e 8-14 on page 8-23 ).
10-21 DIRECT MEMORY ACCESS UNIT 10.3.2 DMA Latency DMA Latency is the delay between a DMA request being asserted and the DMA cycle being run. The DMA latency f or a channel is contro ll ed by many factor s: • Bus H OLD — Bus HOLD takes preced ence over internal DMA requests.
DIRECT MEMORY ACCESS UNIT 10-22 10.3.4 Generating a DMA Acknowle dge The DMA channels do not provide a d istinct DMA acknowledge sign al. A chip-select line can be prog rammed to activ at e f or the memory or I/O range that requires th e acknowled ge.
10-23 DIRECT MEMORY ACCESS UNIT Exam ple 10-1 . Initializing the DMA Unit $MOD186 name DMA_EXAMPLE_1 ; This example shows code necessary to set up two DMA channels. ; One channel performs an unsynchronized transfer from memory to memory. ; The second channel is used by a hard disk controller located in ; I/O space.
DIRECT MEMORY ACCESS UNIT 10-24 Exampl e 10-1. Initializ ing the DMA Unit (Continued) MOV DX, D0DSTH MOV AX, BX ; GET HIGH NIBBLE OUT DX, AX ; THE POINTER ADDRESSES HAVE BEEN SET UP. NOW WE SET UP THE TRANSFER COUNT. MOV AX, 29 ; THE MESSAGE IS 29 BYTES LONG.
10-25 DIRECT MEMORY ACCESS UNIT Exampl e 10-1. Initializ ing the DMA Unit (Continued) MOV AX, 512 ; THE DISK READS IN 512 BYTE SECTORS MOV DX, D1TC ; XFER COUNT REG OUT DX, AX ; NOW WE NEED TO SET THE.
DIRECT MEMORY ACCESS UNIT 10-26 Example 10-2. Timed DMA Transfers $mod186 name DMA_EXAMPLE_1 ; This example sets up the DMA Unit to perform a transfer from me mory to ; I/O space every 22 uS. The data is sent to an A/D converter. ; It is assumed that the constants for PCB register addresses are ; defined elsewhere with EQUates.
10-27 DIRECT MEMORY ACCESS UNIT Example 10-2. Timed DMA Transf ers (Continued) ; NOW WE NEED TO SET THE PARAMETERS FOR THE CHANNEL AS FOLLOWS: ; ; DESTINATION SOURCE ; ----------- ------ ; I/O SPACE M.
.
11 Math Coprocessing.
.
11-1 CHAPTER 11 MATH C OPROCESSING The 80C186 Mod ular Core Family meets the need for a general-pur pose embedded microproces- sor. In most data control ap plicat ions, efficient data mov ement and control instruc tions are fore- most and ar ithmetic performed on the data is simple.
MATH COPROCESSING 11-2 The core has an Escape Trap (ET) bit in the P CB Relocation Register (Figur e 4-1 on page 4-2) to contro l the availability of math coprocessing. If the ET bit is set, an attempted numerics execution results in a Typ e 7 interrupt.
11-3 MATH COPROCESSING 11.3.1.1 Data Transfe r Instruction s Data transfer instructio ns move op erands between elements of the 80C187 reg ister stack or be- tween stack top an d memory. Instruc tions can co nver t any d ata type to tempo rary real and load it onto the stack in a single operation .
MATH COPROCESSING 11-4 Available data types include temp orary real, long real, sho rt real, short integ er and word integer. The 80 C1 87 performs autom atic type conver sion to temporary re al.
11-5 MATH COPROCESSING 11.3.1.3 Comparison Instruc tion s Each com parison instruction (see Table 11-3) analyzes the stack top element, of t en in relationship to ano t her operan d.
MATH COPROCESSING 11-6 11.3.1.5 Constant Ins tructions Each constant instruction ( s ee Table 11-5) loads a common ly used constant onto the stack. The values have full 80 -bit precision and are accurate to about 19 decimal digi t s.
11-7 MATH COPROCESSING 11.3.2 80C187 Data Types The micropro ces sor/math coprocessor combination supports seven data types: • Word Integer — A signed 16-bit numeric value. All operations assume a 2’ s complement representation. • Short Integ er — A signed 32-bit numeric value (doub le word).
MATH COPROCESSING 11-8 Figure 11-1. 8 0C187-Sup ported Data Types Increasing Significance Word Integer Packed Decimal Short Real Temporary Real (Two's Complement) Short Integer Long Integer Long .
11-9 MATH COPROCESSING Figure 11-2 . 80C186 Modular Core Family/80C187 Sy stem Configuration ALE PEREQ RESET PEREQ 80C187 CKM NPS2 80C186 Modular Core Latch D15:0 External Oscillator CLKOUT R.
MATH COPROCESSING 11-10 11.4.1 Clocking the 80C187 The micropr ocessor and math copr ocess or operate asynchron ously, and their clock rates may dif- fer. The 80C187 has a CKM pin that determines whether it uses the inp ut cl ock directly or divided by two.
11-11 MATH COPROCESSING Bus cycles involving the 80C187 Math Coprocessor behave ex actly like other I/O bus cycles with respect to the processor’s co ntrol pins. See “System Design Tips” for inform ati on on integrating the 80C18 7 into the ov erall syst em.
MATH COPROCESSING 11-12 Figure 11- 3. 80C187 Configura tion with a Partially Buffe red Bus ALE PEREQ RESET PEREQ EN 80C187 CKM NPS2 80C186 Modular Core Latch D15:0 External Oscillator CLKOUT .
11-13 MATH COPROCESSING 11.4.4 Exception Trapping The 80C187 d etects six error condition s that can occur durin g instruction execu tion. The 80C18 7 can ap ply default fix- ups or signal ex ceptions to th e microprocessor’s ERROR pin .
MATH COPROCESSING 11-14 Figure 11 -4. 80C18 7 Exception Trapp ing via Processor In terrupt Pin INT x CLKOUT D15:0 CMD1 CMD0 PEREQ BUSY ALE A19:A16 AD15:0 RESET CKM NPS2 Q D Q D EN D15:0 A19:0 A2 A1.
11-15 MATH COPROCESSING Example 11-1. I nitialization Sequ ence for 80C187 Math Coprocessor $mod186 name example_80C187_init ; ;FUNCTION: This function initializes the 80C187 numerics coprocessor.
MATH COPROCESSING 11-16 Example 11-2. Floating Poi nt Math Routine Using FSINCOS $mod186 $modc187 name example_80C187_proc ;DESCRIPTION: This code section uses the 80C187 FSINCOS transcendental ; instruction to convert the locus of a point from polar ; to Cartesian coordinates.
12 ONCE Mode.
.
12-1 CHAPTER 12 ONCE MODE ONCE (pronounced “ahn ce”) Mode provides the ability to three-state all output, bidirectional, or weakly held high/low pin s except OSCOUT. To allow device operation with a crystal network, OSCOUT does not three-state. ONCE Mode electrically isolates the device from the rest of the board logic.
ONCE MODE 12-2 Figure 1 2-1. Enteri ng/Leaving ONCE Mode RES UCS LCS All output, bidirectional, weakly held pins except OSCOUT NOTES: 1. Entering ONCE Mode.
A 80C186 Instruction Set Additions and Extensions.
.
A-1 APPENDIX A 80C186 INST RUCT ION SET ADDITIO NS AND EXTENSIONS The 80C18 6 Mo dular Core family instruction set differ s from the orig inal 808 6/8088 instru ction set in two ways. First, several instructions that were n ot available in the 8086/8088 instruction set have been added.
80C186 INST RU CTION SET ADDITIONS AND EXTENSIONS A-2 A.1.2 Stri ng Instructions INS source_string, port INS (in string) perf orms block input from an I/O port to memor y. The port address is placed in the DX register. The memory addr es s is placed in the DI register.
A-3 80C186 INSTRUCTION SET ADDITIONS AND EXTENSIONS Figure A-1. Formal De finition of ENTER ENTER treats a reentrant p roced ure as a proce dure calling another pro cedure at the same lexical level. A reentran t procedure can ad dress only its own variables and variables of higher-level call- ing procedu res.
80C186 INST RU CTION SET ADDITIONS AND EXTENSIONS A-4 Figure A-2. Variable Access in Nested Procedure s The first ENTER, executed in the Main Program , al locates dynamic storage space for M ain, but no pointers are co pied.
A-5 80C186 INSTRUCTION SET ADDITIONS AND EXTENSIONS Figure A-4 . Stack Frame for Procedure A at Lev el 2 After Procedu re A calls Procedure B, ENTER creates the display for Procedure B. Th e first word of the display points to the prev i ous value of BP (BPA).
80C186 INST RU CTION SET ADDITIONS AND EXTENSIONS A-6 Figure A-5. Stack Fra me for Procedure B at Level 3 Calle d from A A1004-0A Old BP BP SP 15 0 BPM BPM BPM Display B Dynamic Storage B BPA BPM BPA .
A-7 80C186 INSTRUCTION SET ADDITIONS AND EXTENSIONS Figure A-6. Stack Fra me for Procedure C at Level 3 Calle d from B LEAVE LEAVE reverses the action of the mo s t recent ENTER instruction .
80C186 INST RU CTION SET ADDITIONS AND EXTENSIONS A-8 BOUND register, address BOUND verifies that the signed value in the specified r egis ter lies wi thin specified limits. If the value does not lie within the bounds, an ar ray bounds exceptio n (type 5) occurs.
A-9 80C186 INSTRUCTION SET ADDITIONS AND EXTENSIONS A.2.2 Arithmetic Instructions IMUL desti nation, source, data IMUL (integer immediate multiply, sig ned) allows a value to be multip lied by an immediate op- erand. IMUL requires three oper ands. The first, destination , i s the register where the result will be placed.
80C186 INST RU CTION SET ADDITIONS AND EXTENSIONS A-10 A.2.3.2 Rotate Instruc tions ROL destination, co unt ROL (immediate rotate left) rotates the destination byte or word left by an immediate value. ROL has two operan ds. The first, d estination , is the effective ad dress to be ro tated.
B Input Synchronization.
.
B-1 APPENDIX B INPUT SYNCH RONIZATIO N Many input signals to an embed ded processor are asynchro n ous. Asynchr onous signals do not re- quire a specified setup or h old time to ensur e the device d oes not incur a failure. Ho wever, asyn- chron ous setup and hold times are specified in the data sheet to en su re recognition .
INPUT SYNCHRONIZATION B-2 A synchronization failure can occur when the output of the first latch does not meet the setup and hold requ irements of the input of the second latch.
C Instruction Set Descriptions.
.
C-1 APPENDIX C INSTRUCT IO N SET DESCRIPT IONS This appendix pro vides reference info rmation for the 80C186 Modular Core family instruction set. Tables C-1 through C- 3 defin e the variables u s ed in Table C-4, wh ich lists the instruction s with their descriptions and operations.
INSTRUCTION SET DESCRIPTIONS C-2 Table C-2. Instruc tion Operand s Operand Descrip tion reg An 8- or 16-bit general regist er. reg16 An 16-bit general register. seg-reg A segment register. accum Register AX or AL immed A constant in the range 0–FFFFH.
C-3 INSTRUCTION SET DESCRIPTIONS Table C-3. Fla g Bit Functi ons Name F unctio n AF Auxiliary Flag: Set on carry from or borrow to the low order four bits of AL; cleared other wise. CF Carry Flag: Set on high-order bit carry or borrow; cleared otherwise.
INSTRUCTION SET DESCRIPTIONS C-4 Table C-4. Instruction Se t Name Description Operation Flags Affected AAA ASCII Ad just fo r Add ition : AAA Changes the contents of register A L to a valid unpacked decimal number; the high-order half-byt e is zeroed.
C-5 INSTRUCTION SET DESCRIPTIONS AAS ASCII Ad just for Sub tracti on : AAS Corrects the result of a previous subtractio n of two val id unpacked decimal operands (the destination operand must have been specified as register AL). Changes the c ontent of AL to a valid unpacked decimal number; the high-order half-byte is zeroed.
INSTRUCTION SET DESCRIPTIONS C-6 ADD Additi on : ADD dest , src Sums two operands, which may be bytes or words, rep laces the destination operand. Both operands may be signed or unsigned binary numbers (see AAA and DAA) .
C-7 INSTRUCTION SET DESCRIPTIONS BOUND Detect Value Out of Range : BOUND dest , src Provides array bounds chec king in hardware. The calculated array index is placed in one of the gener al purpose registers, and the upper and lower bounds of the array are placed in two consecutive m emory locations.
INSTRUCTION SET DESCRIPTIONS C-8 CBW Con vert Byte to Word : CBW Extends the sign of the byte in register AL throughout register AH. Use t o produce a double-lengt h (word) dividend from a byte prior to performing byte division.
C-9 INSTRUCTION SET DESCRIPTIONS CLI Clear Interrupt-en abl e Flag : CLI Zeroes the interrupt-enab le flag (IF). When the int errupt-enable flag is cleared, the 8086 and 8088 do not recognize an external int errupt request that appears on the INT R line; in other words maskable interrupts are disabled.
INSTRUCTION SET DESCRIPTIONS C-10 CMP Comp are : CMP dest , src Subtracts the s ource from the desti- nation, which may be bytes or words, but does not return the resu lt. The operands are unchanged, but the flags are updated and can be tested by a subsequent con ditional jump instruction.
C-11 INSTRUCTION SET DESCRIPTIONS CWD Convert W o rd to Dou blew ord : CWD Extends the sign of the word in register AX throughout register DX. Use t o produce a double-length (doubleword) dividend from a word prior to performing word division.
INSTRUCTION SET DESCRIPTIONS C-12 DEC Decrement : DEC dest Subtracts one from the dest ination operand. The operand may be a byte or a word and is treated as an unsigned binary num ber (see AAA and DAA). Instructi o n Operan d s: DEC reg DEC mem (dest) ← (dest) – 1 AF CF – DF – IF – OF PF SF TF – ZF Table C-4.
C-13 INSTRUCTION SET DESCRIPTIONS DIV Divide : DIV src Performs an unsigned division of the accumulator (and it s extension) by t he source operand. If the source operand is a byte, it is divided into the two-by te dividend assumed to be in registers AL and AH.
INSTRUCTION SET DESCRIPTIONS C-14 ENTER Procedure Entry : ENTER locals, levels Executes the calling sequence for a high-level language. It saves the current frame point er in BP, copies the frame poin.
C-15 INSTRUCTION SET DESCRIPTIONS HLT Halt : HLT Causes the CPU to enter the halt state. The process or leaves the halt state upon activation of the RESET line, upon receipt of a non-mask able interrupt request on NMI, or upon receipt of a maskab le interrupt request on INTR (if interrupts are ena bled).
INSTRUCTION SET DESCRIPTIONS C-16 IDIV Integer Divi d e : IDIV src Performs a signed division of the accumulator (and its extension) by the source operand.
C-17 INSTRUCTION SET DESCRIPTIONS IMUL In teger M ultip ly : IMUL src Performs a signed mu ltiplication of the source operand and the accumulator. If the source is a byte, then it is multiplied by register AL, and the double-length result is returned in AH and AL.
INSTRUCTION SET DESCRIPTIONS C-18 INC Increment : INC dest Adds one t o the destinat ion operand. The operand may be byte or a word and is treated as an unsigned binary number (see AAA a nd DAA).
C-19 INSTRUCTION SET DESCRIPTIONS INT Interrupt : INT interrupt-type Activates the in terru pt procedur e specified by the int errupt-type operand. De crements the stack pointer by two, pushes the flags onto the stack, and clears the trap (TF) and interrupt-enable (IF) flags to disable single-step and maskable interrupts.
INSTRUCTION SET DESCRIPTIONS C-20 INTO Interrupt o n Overflow : INTO Generates a soft ware interrupt if the overflow flag (OF) is set; otherwise control proceeds to the following instruction without act ivating an interrupt procedure.
C-21 INSTRUCTION SET DESCRIPTIONS JAE JNB Jump on Above or Equal : Jump on Not Belo w : JAE disp8 JNB disp8 Transfers control to t he target locat ion if the tested condition (CF = 0) is true.
INSTRUCTION SET DESCRIPTIONS C-22 JCXZ Jump if CX Zero : JCXZ disp8 Transfers cont rol to the targ et location if CX is 0. Useful at the beginning of a loop to bypass the loop if CX has a zero value, i.
C-23 INSTRUCTION SET DESCRIPTIONS JL JNGE Jump on Less Th an : Jump on Not Greater Than or Eq u al : JL disp8 JNGE disp8 Transfers control to t he target locat ion if the condition tested (SF ≠ OF) is true .
INSTRUCTION SET DESCRIPTIONS C-24 JNE JNZ Jump on Not Equ al : Jump o n Not Zero : JNE disp 8 JNZ disp8 Transfers cont rol to the targ et location if the tested condition (ZF = 0) is true.
C-25 INSTRUCTION SET DESCRIPTIONS JO Jump on Overflow : JO disp8 Transfers control to t he target locat ion if the tested condition (OF = 1) is true. Instruction Op eran ds : JO short- labe l if (OF) .
INSTRUCTION SET DESCRIPTIONS C-26 LDS Load Pointer Usi ng DS : LDS dest, src Transfers a 32-bit pointer variable from the source operand, which must be a memory operand, to the dest ination operand and regist er DS. The offs et word of the po inter is transferred to th e destination operand, which may be any 16-bit general register.
C-27 INSTRUCTION SET DESCRIPTIONS LES Load P ointer U sing E S : LES dest, src Transfers a 32-bit pointer variable from the source operand to the destination operand and register ES. The offset word of the pointer is transf erred to the destination op erand.
INSTRUCTION SET DESCRIPTIONS C-28 LODS Lo ad Stri ng (Byte or Wo rd) : LODS src-string Transfers t he byte or word string element addressed by SI to register AL or AX and updates SI to point to the next e lement in the string.
C-29 INSTRUCTION SET DESCRIPTIONS LOOPNE LOOP NZ Loop Whi le Not Equal : Loop W hile Not Zero : LOOPNE disp8 LOOP NZ disp8 Decrements CX by 1 and transfers control to t he target loca tion if CX is not 0 and if ZF is clear; otherwise the next sequenti al instruction is ex ecute d.
INSTRUCTION SET DESCRIPTIONS C-30 MOVS Move String : MOVS dest-string, src -st ring Transfers a byte or a word from the source string (addressed by SI ) to the destination string (addresse d by DI) and updates SI and DI to point to the next string element.
C-31 INSTRUCTION SET DESCRIPTIONS NEG Negate : NEG dest Subtracts the dest ination operand, which may be a byte or a word, from 0 and returns the result to the desti- nation. This forms the tw o's complement of the num ber, effect ively reversing the s ign of an integer.
INSTRUCTION SET DESCRIPTIONS C-32 OR Logical O R : OR dest,src Performs the logical "inclusive or" of the two operands (bytes or words) and returns the result to the dest ination operand. A bit in the result is set if either or both corresponding bits in t he original operands are s et; otherwise the result bit is cleared.
C-33 INSTRUCTION SET DESCRIPTIONS OUTS Out St rin g : OUTS port, src_st ring Performs block o utput from memory to an I/O port. The port address is placed in the DX re gister. T he mem ory address is placed in the SI register. This instruction uses the DS segm ent register, but this may be changed with a segment ov erride instruc tion.
INSTRUCTION SET DESCRIPTIONS C-34 POPA Po p All : POPA Pops all data, pointer, and index registers off of the stack. The SP value popped is discarded. Instructi o n Operan d s : none (DI) ← ((S P) +.
C-35 INSTRUCTION SET DESCRIPTIONS PUSHA Push All : PUSHA Pushes all data, pointer, and index registers onto the stack . The order in which the registers are saved is: AX, CX, DX, BX, SP, BP, SI, and DI. The SP value pushed is the SP value before the first register ( AX) is pushed.
INSTRUCTION SET DESCRIPTIONS C-36 RCL Rotate Thro u gh Carry Left : RCL dest, count Rotates the bits in the byte or word destination operand to the left by the number of bits specified in the count operand.
C-37 INSTRUCTION SET DESCRIPTIONS REP REPE REPZ REPNE REPNZ Repeat : Repeat Wh ile Eq ual : Repeat Wh ile Zero : Repeat Wh ile Not Equal : Repeat Wh ile Not Zero : Controls subsequent string instruction repetition. The different mnem on ics are provided to improve program clarity.
INSTRUCTION SET DESCRIPTIONS C-38 RET Return : RET optional-pop-value Transfers cont rol from a procedure back to the instruction following the CALL that activated t he procedure. The assem bler generates an intra- segment RET if the programme r has defined the procedure near, or an intersegment RE T if the procedure has been defined as far.
C-39 INSTRUCTION SET DESCRIPTIONS ROR Rotate Rig h t : ROR dest, count Operates s imilar t o ROL except that the bits in the destination byte or word are rotated righ t instead of left .
INSTRUCTION SET DESCRIPTIONS C-40 SHL SAL Shift Logical Left : Shift Arith met ic Lef t : SHL dest, count SAL dest, count Shifts the des tination byte or word left by the number of bits specified in the count operand. Zeros are shifted in on the right.
C-41 INSTRUCTION SET DESCRIPTIONS SBB Sub tract With Borro w : SBB dest, src Subtracts the sourc e from the desti- nation, subtracts one if CF is set, and returns the result to the destination operand.
INSTRUCTION SET DESCRIPTIONS C-42 SCAS Scan Stri n g : SCAS dest-string Subtracts the des tination string element (byte or word) addressed by DI from the co ntent of AL ( byte st ring) or AX (word string) and updates the flags, but does not alt er the destination string or t he accum ulator.
C-43 INSTRUCTION SET DESCRIPTIONS SHR Shif t Lo gical Righ t : SHR dest, src Shifts the bits in the destina tion operand (byte or word) to t he right by the number of bits specified in the count operand. Zeros are shifted in on the left. If the sign bit retains its original value, then OF is cleared.
INSTRUCTION SET DESCRIPTIONS C-44 STI Set Interrupt-en able Flag : STI Sets IF to 1, enabling process or recognition of maskable interrupt requests appearing on the INTR l ine. Note however, that a pending int errupt will not actually be recognized unt il the instruction f ollowing ST I has exe cuted.
C-45 INSTRUCTION SET DESCRIPTIONS SUB Subtract : SUB dest, src The source operand is subtract ed from the destination oper and, and th e result replaces the destination operand. The operands may be bytes or words. Both operands may be signed or unsigned binary numbers (see AAS and DAS).
INSTRUCTION SET DESCRIPTIONS C-46 WAIT W ait : WAIT Causes the CPU t o enter the wa it state while its test line is not active. Instructi o n Operan d s : none None AF – CF – DF – IF – OF – PF – SF – TF – ZF – XCHG Exchan ge : XCHG dest, src Switches the contents of the source and destination operands (byt es or words).
C-47 INSTRUCTION SET DESCRIPTIONS XLAT Transl ate : XLAT translate-table Replaces a byte in the AL regist er with a byte from a 256-byte, user-c oded translation table.
.
D Instruction Set Opcodes and Clock Cycles.
.
D-1 APPENDIX D INSTRUC TION SET OPCO DES AND CLOCK C YCLES This appen dix provides reference information for the 80C186 Modu lar Core family instruction set. Table D-1 def ines the variables used in Table D-2, w hich lists the instru ctions with the ir for- mats and execution times.
INSTRUCTION SET OPCODES AND CLOCK CYCLES D-2 Table D-2. Inst ruction Set Summary Function Format C locks Notes DATA TRANSFER INSTRUCTIONS MOV = Move register to r egister/me mory 1 0 0 0 1 0 0 w mod r.
D-3 INSTRUCTION SET OPCODES AND CLOCK CYCLES DATA TRANSFER INSTRUCTIONS (Continued) LEA = Load EA to r egister 1 0 0 0 1 1 0 1 mod reg r /m 6 LDS = Load pointer to DS 1 1 0 0 0 1 0 1 mod reg r/m (mod .
INSTRUCTION SET OPCODES AND CLOCK CYCLES D-4 ARITHMETIC INSTRUCTIONS (Cont inued) SUB = Subtr act reg/memo ry with register to either 0 0 1 0 1 0 d w mod reg r / m 3/10 immediate fr om register/memory.
D-5 INSTRUCTION SET OPCODES AND CLOCK CYCLES ARITHMETIC INSTRUCTIONS (Cont inued) AAM = A SCII adjust for multip ly 1 1 0 1 0 1 0 0 0 0 0 0 1 0 1 0 19 DIV = Divide (unsigned) 1 1 1 1 0 1 1 w mod 110 r.
INSTRUCTION SET OPCODES AND CLOCK CYCLES D-6 BIT MANIPULATION INSTRUCTIONS ( Continued) TEST = And function to f l ags, no result register/me mory and r egister 1 0 0 0 0 1 0 w mod reg r/m 3/10 immedi.
D-7 INSTRUCTION SET OPCODES AND CLOCK CYCLES PROGRAM TRANSFER INSTRUCTIONS Conditio nal Trans fers — jump if: JE/JZ = equal/zero 0 1 1 1 0 1 0 0 disp 4/13 (2) JL/JNGE = less/ not greater or equal 0 .
INSTRUCTION SET OPCODES AND CLOCK CYCLES D-8 PROGRAM TRANSFER INSTRUCTIONS (Continued) RET = Return from pro cedure within segment 1 1 0 0 0 0 1 1 16 within segment a dding immed to SP 1 1 0 0 0 0 1 0.
D-9 INSTRUCTION SET OPCODES AND CLOCK CYCLES PROCESSOR CONTROL INSTRUCTIONS CLC = Clear carry 1 1 1 1 1 0 0 0 2 CMC = Complement carry 1 1 1 1 0 1 0 1 2 STC = Set carry 1 1 1 1 1 0 0 1 2 CLD = Clear d.
INSTRUCTION SET OPCODES AND CLOCK CYCLES D-10 09 0000 10 01 mod reg r /m (disp-lo),(disp-h i) or reg16/me m16,reg16 0A 0000 1010 mod reg r/m (disp-lo),(disp-h i ) or reg8,reg8/mem 8 0B 0000 1011 mod r.
D-11 INSTRUCTION SET OPCODES AND CLOCK CYCLES 2E 0010 1110 DS: (segment override pr efix) 2F 0010 1111 das 30 0011 00 00 mod reg r /m (disp-lo),(disp-h i) xor reg8/mem8,re g8 31 0011 00 01 mod reg r /.
INSTRUCTION SET OPCODES AND CLOCK CYCLES D-12 53 0101 00 11 push BX 54 0101 01 00 push SP 55 0101 01 01 push BP 56 0101 01 10 push SI 57 0101 01 11 push DI 58 0101 10 00 pop AX 59 0101 10 01 pop CX 5A.
D-13 INSTRUCTION SET OPCODES AND CLOCK CYCLES 7E 0111 11 10 IP-inc-8 jle/jng short-label 7F 0111 1111 IP -inc-8 jnle/jg short-label 80 1000 00 00 mod 000 r /m (disp-lo),(disp-h i ), data-8 add reg8/me.
INSTRUCTION SET OPCODES AND CLOCK CYCLES D-14 87 1000 01 11 mod reg r /m (disp-lo),(disp-h i) xchg reg16,reg 16/mem16 88 1000 01 00 mod reg r /m (disp-lo),(disp-h i) mov reg8/mem 8,reg8 89 1000 10 01 .
D-15 INSTRUCTION SET OPCODES AND CLOCK CYCLES AA 1010 10 10 stos dest- str8 AB 1010 10 11 stos dest- str16 AC 1010 11 00 lods src-str 8 AD 1010 11 01 lods src-str16 AE 1010 11 10 scas dest-str8 AF 101.
INSTRUCTION SET OPCODES AND CLOCK CYCLES D-16 mod 111 r / m data-8 sar reg16/mem16, immed8 C2 1100 00 10 data-lo data-hi ret immed16 (intrase gment) C3 1100 00 11 ret (intrasegment) C4 1100 01 00 mod .
D-17 INSTRUCTION SET OPCODES AND CLOCK CYCLES D1 1101 00 01 mod 000 r /m (disp-lo),( disp-hi) rol reg16/mem16,1 mod 001 r / m (disp-lo),(disp-h i) ror reg1 6/mem16,1 D1 1101 00 01 mod 010 r /m (disp-l.
INSTRUCTION SET OPCODES AND CLOCK CYCLES D-18 E1 1110 0001 IP -inc-8 loope/loopz short-label E2 1110 0010 IP -inc-8 loop short-label E3 1110 0011 IP-inc-8 jcxz short-label E4 1110 0100 data-8 in AL,im.
D-19 INSTRUCTION SET OPCODES AND CLOCK CYCLES F8 1111 1000 clc F9 1111 1001 stc FA 1111 10 10 cli FB 1111 10 11 sti FC 1111 11 00 cld FD 1111 1101 std FE 1111 11 10 mod 000 r /m (disp-lo),(disp-h i) i.
INSTRUCTION SET OPCODES AND CLOCK CYCLES D-20 Table D-4. Mnemon ic Encodi ng Matrix (Left Half) x0 x1 x2 x 3 x4 x5 x6 x7 0x ADD b,f,r/m ADD w,f,r/m ADD b,t,r/m ADD w,t,r/m ADD b,ia ADD w,ia PUSH ES PO.
D-21 INSTRUCTION SET OPCODES AND CLOCK CYCLES Table D-4. Mnemon ic Encoding Matrix (Righ t Half) x8 x9 xA xB xC xD xE xF OR b,f,r/m OR w,f,r/m OR b,t,r/m OR w,t,r/m OR b,i OR w,i PUSH CS 0x SBB b,f,r/.
INSTRUCTION SET OPCODES AND CLOCK CYCLES D-22 Table D-5. Abbreviations for Mnemonic Encoding Matrix Abbr Defi nition Abbr De finiti on Ab br Def initio n Abbr D efinit ion b byte operation ia immedia .
Index.
.
Index -1 80C18 7 Math Cop rocessor, 10-2 –1 0-8 accessing, 10-10–10-11 arithm etic instructio ns, 10-3 – 10-4 bus cycles, 10-11 clockin g , 10-1 0 code exampl e s, 10-13–10-1 6 comparison inst.
INDEX Inde x-2 and ch ip-selects, 6-5 HALT state, exi ting, 3- 3 0 idle states, 3-18 instruction p refe tch, 3-20 interrupt ackno wledg e (INTA) cycles, 3-6, 3-25 –3-26, 8-9 and ch ip-selects, 6-5 i.
Index -3 INDEX Data sheets, obtaining from BBS, 1-5 Data transfers, 3 -1–3-6 instruct io ns, 2-18 PCB consideratio ns, 4-5 PSW flag sto rage formats, 2-19 See also Bus cycles Data types, 2-37–2-38.
INDEX Inde x-4 F Fault exceptio ns, 2 -43 FaxBack service, 1-4 F-Bus and P CB, 4-5 operation, 4-5 Flags‚ Se e Processor Status Word (PSW) Floati ng Poin t, defined, 2-3 7 H HALT bus cycle‚ See Bus.
Index -5 INDEX maskable, 2-43 masking, 8 -3, 8-12, 8-16 priority -based, 8-17 multip lexed, 8 - 7 nest ing, 8-4 NMI, 2-42 nonmaskable, 2-45 overview, 8 -1, 8-2 prio rity, 2 -46–2-49, 8 -3 default , .
INDEX Inde x-6 Poll ing, 8-1, 8 -9 POPA instru c tion, A-1 Power co nsumption‚ r e ducing, 3-28 Power manag ement, 5-10–5-14 Power management mod e s and HAL T bus cy c les, 3- 30 Powerd own mod e.
Index -7 INDEX SI regi s ter, 2 -1, 2- 5, 2-13, 2- 22, 2-23 , 2-30, 2-3 2, 2-34 Sign Flag (SF), 2-7, 2-9 Single-step trap (Type 1 exception), 2-43 Software code example 80C1 87 fl oating-p oint routin.
INDEX Inde x-8 and PCB accesses, 4-4 and RE ADY input, 3-13 Word integer, defin ed , 10-7 World Wide Web, 1 -6 Write b us cycle, 3-22 Z Zero Flag (ZF), 2-7, 2-9 , 2-23.
デバイスIntel 80C186XLの購入後に（又は購入する前であっても）重要なポイントは、説明書をよく読むことです。その単純な理由はいくつかあります：
Intel 80C186XLをまだ購入していないなら、この製品の基本情報を理解する良い機会です。まずは上にある説明書の最初のページをご覧ください。そこにはIntel 80C186XLの技術情報の概要が記載されているはずです。デバイスがあなたのニーズを満たすかどうかは、ここで確認しましょう。Intel 80C186XLの取扱説明書の次のページをよく読むことにより、製品の全機能やその取り扱いに関する情報を知ることができます。Intel 80C186XLで得られた情報は、きっとあなたの購入の決断を手助けしてくれることでしょう。
Intel 80C186XLを既にお持ちだが、まだ読んでいない場合は、上記の理由によりそれを行うべきです。そうすることにより機能を適切に使用しているか、又はIntel 80C186XLの不適切な取り扱いによりその寿命を短くする危険を犯していないかどうかを知ることができます。
ですが、ユーザガイドが果たす重要な役割の一つは、Intel 80C186XLに関する問題の解決を支援することです。そこにはほとんどの場合、トラブルシューティング、すなわちIntel 80C186XLデバイスで最もよく起こりうる故障・不良とそれらの対処法についてのアドバイスを見つけることができるはずです。たとえ問題を解決できなかった場合でも、説明書にはカスタマー・サービスセンター又は最寄りのサービスセンターへの問い合わせ先等、次の対処法についての指示があるはずです。
