Vorlesung Rechnerarchitektur - Fachbereich Informatik
Vorlesung Rechnerarchitektur - Fachbereich Informatik Vorlesung Rechnerarchitektur - Fachbereich Informatik
B eis piele für einen 16/32 B it Ins truktions s a tz (A R M Thum b2) 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 opcode opcode 00xxxx Instruction or instruction class Shift (immediate), add, subtract, move, and compare 010000 Data processing 010001 Special data instructi 01001x 0101xx 011xxx 100xxx 10100x 10101x 1011xx 11000x 11001x 1101xx 11100x Load from Literal Pool, see LDR (literal) Load/store single data item Generate PC-relative address, see ADR Generate SP-relative address, see ADD (SP plus immediate) Miscellaneous 16-bit instructi Store multiple registers, see STM / STMIA / STMEA Load multiple registers, see LDM / LDMIA / LDMFD Conditional branch, and supervisor call Unconditional Branch, see B Vorlesung Rechnerarchitektur © Gerhard Raffius, WS 2009/10, h_da - Fachbereich Informatik 76
32 B it Ins truktionen 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 1 1 1 op1 op2 op Vorlesung Rechnerarchitektur © Gerhard Raffius, WS 2009/10, h_da - Fachbereich Informatik 77
- Seite 25 und 26: D er M ic roprog ra m C ounter S te
- Seite 27 und 28: D er Fetc h Z yklus im M u1 MU1 Dat
- Seite 29 und 30: D er S ta c k A push Speicher Der S
- Seite 31 und 32: U nterprog ra m m e Hauptprogramm U
- Seite 33 und 34: Indirekte A dres s ierung A STR Spe
- Seite 35 und 36: B es c hreibung des M ic roc odes d
- Seite 37 und 38: B es c hreibung des M ic roc odes d
- Seite 39 und 40: Z us ta nds bes c hreibung des La d
- Seite 41 und 42: D ie O pera tion P us h, der A kkum
- Seite 43 und 44: eding te S prüng e jge ACC15 = 1 1
- Seite 45 und 46: der S top B efehl stop 1 Der Stop B
- Seite 47 und 48: P roblem e des M u1 D es ig ns Der
- Seite 49 und 50: P roblem e des M U 0 D es ig ns Der
- Seite 51 und 52: K om plexe A dres s berec hnung en
- Seite 53 und 54: ela tive S prüng e .L1: .L2: movs
- Seite 55 und 56: K om plexe A dres s berec hnung : B
- Seite 57 und 58: R eg is tera rc hitektur Die Ergebn
- Seite 59 und 60: K ons equenzen a us der R eg is ter
- Seite 61 und 62: E ine V erbes s erte A LU Einsatz v
- Seite 63 und 64: V erbes s erung des internen B us s
- Seite 65 und 66: Fetc h Z yklus im M u5 In Aout steh
- Seite 67 und 68: D a tentra ns fer Adress Data In Ad
- Seite 69 und 70: V erbes s erte A dres s berec hnung
- Seite 71 und 72: B eis piel: s pekula tive B efehls
- Seite 73 und 74: P ipeline Die Pipeline erlaubt es d
- Seite 75: D ie H a rva rd A rc hitektur Durch
- Seite 79 und 80: A rithm etis c he und log is c he B
- Seite 81 und 82: D er A dditions befehl 15 9 8 7 6 5
- Seite 83 und 84: S ta c k-A rc hitektur Stacks werde
- Seite 85 und 86: R eg is ter-R eg is ter-A rc hitekt
32 B it Ins truktionen<br />
15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0<br />
1 1 1 op1 op2 op<br />
<strong>Vorlesung</strong> <strong>Rechnerarchitektur</strong><br />
© Gerhard Raffius, WS 2009/10, h_da - <strong>Fachbereich</strong> <strong>Informatik</strong><br />
77