ARM ARM1176JZF-S - Chapter 21 VFP Instruction Execution

To Next Page

To Previous Page

ID012310 Non-Confidential, Unrestricted Access

Chapter 21

VFP Instruction Execution

This chapter describes the VFP11 instruction pipeline and its relationship with the ARM processor

instruction pipeline. It contains the following sections:

• About instruction execution on page 21-2

• Serializing instructions on page 21-3

• Interrupting the VFP11 coprocessor on page 21-4

• Forwarding on page 21-5

• Hazards on page 21-6

• Operation of the scoreboards on page 21-7

• Data hazards in full-compliance mode on page 21-13

• Data hazards in RunFast mode on page 21-16

• Resource hazards on page 21-17

• Parallel execution on page 21-20

• Execution timing on page 21-22.

Main Page

Default Chapter2
- Change History2
- Table of Contents4
Chapter 1 Introduction28
- About the Processor28
- Extensions to Armv629
- Figure29
- Trustzone Security Extensions30
- ARM1176JZF-S Architecture with Jazelle Technology32
- Components of the Processor34
  - Figure 1-1 ARM1176JZF-S Processor Block Diagram34
  - Conditional Execution35
  - Figure37
- Table 1-2 Double-Precision VFP Operations46
- Table 1-3 Flush-To-Zero Mode46
- Power Management49
- Configurable Options51
- Table 1-4 Configurable Options51
- Table 1-5 ARM1176JZF-S Processor Default Configurations51
- Pipeline Stages52
- Figure 1-2 ARM1176JZF-S Pipeline Stages52
- Typical Pipeline Operations54
- Figure 1-3 Typical Operations in Pipeline Stages54
- Figure 1-4 Typical ALU Operation54
- Figure 1-5 Typical Multiply Operation55
- Figure 1-6 Progression of an LDR/STR Operation56
- Figure 1-7 Progression of an LDM/STM Operation56
- Figure 1-8 Progression of an LDR that Misses57
- ARM1176JZF-S Instruction Set Summary58
- Table 1-7 ARM Instruction Set Summary59
- Table64
- Table 1-8 Addressing Mode 267
- Table 1-9 Addressing Mode 2P, Post-Indexed Only67
- Addressing Mode 368
  - Addressing Mode 468
- Addressing Mode 568
- Table 1-13 Operand269
- Table 1-14 Fields69
- Table 1-15 Condition Codes69
- Table 1-16 Thumb Instruction Set Summary70
- Product Revisions73
Chapter 2 Programmer's Model75
- About the Programmer's Model75
- Secure World and Non-Secure World Operation with Trustzone76
- Figure 2-1 Secure and Non-Secure Worlds76
- Figure 2-2 Memory in the Secure and Non-Secure Worlds79
- Figure 2-3 Memory Partition in the Secure and Non-Secure Worlds80
- Figure81
- Table 2-1 Write Access Behavior for System Control Processor Registers82
- Table 2-2 Secure Monitor Bus Signals84
- Processor Operating States85
- Instruction Length86
- Data Types87
- Memory Formats88
- Figure 2-4 Big-Endian Addresses of Bytes Within Words88
- Figure 2-5 Little-Endian Addresses of Bytes Within Words88
- Addresses in a Processor System89
- Table 2-3 Address Types in the Processor System89
- Operating Modes90
- Table 2-4 Mode Structure90
- Registers91
- Figure91
- Table 2-5 Register Mode Identifiers92
- Figure92
- Figure 2-6 Register Organization in ARM State93
- Figure 2-7 Processor Core Register Set Showing Banked Registers94
- Figure 2-8 Register Organization in Thumb State95
- Figure 2-9 ARM State and Thumb State Registers Relationship96
- The Program Status Registers97
- Figure 2-10 Program Status Register97
- Table 2-6 GE[3:0] Settings99
- Table 2-7 PSR Mode Bit Values101
- Additional Instructions103
- Copyright © 2004-2009 ARM Limited. All Rights Reserved103
- Figure 2-11 LDREXB Instruction103
- Figure 2-12 STREXB Instructions103
- Figure 2-13 LDREXH Instruction104
- Figure 2-14 STREXH Instruction105
- Figure 2-15 LDREXD Instruction106
- Figure 2-16 STREXD Instruction106
- Figure 2-17 CLREX Instruction107
- Figure 2-18 NOP-Compatible Hint Instruction107
- Exceptions109
- Table 2-8 Exception Entry and Exit110
- Exception Vectors121
- Table 2-9 Exception Priorities130
- Figure130
- Software Considerations132
Chapter 3 System Control Coprocessor133
- About the System Control Coprocessor134
- Table 3-1 System Control Coprocessor Register Functions135
- Figure 3-1 System Control and Configuration Registers137
- Figure 3-2 MMU Control and Configuration Registers139
- Figure 3-3 Cache Control and Configuration Registers140
- Figure 3-4 TCM Control and Configuration Registers140
- Figure 3-5 Cache Master Valid Registers141
- Figure 3-6 DMA Control and Configuration Registers141
- Figure 3-7 System Performance Monitor Registers142
- Figure 3-8 System Validation Registers143
- Figure 3-9 CP15 MRC and MCR Bit Pattern144
- System Control Processor Registers145
  - Table 3-2 Summary of CP15 Registers and Operations146
  - Table 3-3 Summary of CP15 MCRR Operations151
  - Figure 3-10 Main ID Register Format152
  - Table 3-4 Main ID Register Bit Functions152
  - Table 3-5 Results of Access to the Main ID Register152
  - Figure 3-11 Cache Type Register Format153
  - Table 3-6 Cache Type Register Bit Functions153
  - Table 3-7 Results of Access to the Cache Type Register155
  - Table 3-8 Example Cache Type Register Format155
  - Figure 3-12 TCM Status Register Format156
  - Table 3-9 TCM Status Register Bit Functions156
  - Figure 3-13 TLB Type Register Format157
  - Table 3-10 TLB Type Register Bit Functions157
  - Table 3-11 Results of Access to the TLB Type Register157
  - Figure 3-14 Processor Feature Register 0 Format158
  - Table 3-12 Processor Feature Register 0 Bit Functions158
  - Table 3-13 Results of Access to the Processor Feature Register 0159
  - Figure 3-15 Processor Feature Register 1 Format160
  - Table 3-14 Processor Feature Register 1 Bit Functions160
  - Table 3-15 Results of Access to the Processor Feature Register 1160
  - Figure 3-16 Debug Feature Register 0 Format161
  - Table 3-16 Debug Feature Register 0 Bit Functions161
  - Table 3-17 Results of Access to the Debug Feature Register 0161
  - Table 3-18 Auxiliary Feature Register 0 Bit Functions162
  - Table 3-19 Results of Access to the Auxiliary Feature Register 0162
  - Figure 3-17 Memory Model Feature Register 0 Format163
  - Table 3-20 Memory Model Feature Register 0 Bit Functions163
  - Table 3-21 Results of Access to the Memory Model Feature Register 0163
  - Figure 3-18 Memory Model Feature Register 1 Format164
  - Table 3-22 Memory Model Feature Register 1 Bit Functions164
  - Table 3-23 Results of Access to the Memory Model Feature Register 1165
  - Figure 3-19 Memory Model Feature Register 2 Format166
  - Table 3-24 Memory Model Feature Register 2 Bit Functions166
  - Figure 3-20 Memory Model Feature Register 3 Format167
  - Table 3-25 Results of Access to the Memory Model Feature Register 2167
  - Table 3-26 Memory Model Feature Register 3 Bit Functions167
  - Figure 3-21 Instruction Set Attributes Register 0 Format168
  - Table 3-27 Results of Access to the Memory Model Feature Register 3168
  - Table 3-28 Instruction Set Attributes Register 0 Bit Functions168
  - Table 3-29 Results of Access to the Instruction Set Attributes Register 0169
  - Figure 3-22 Instruction Set Attributes Register 1 Format170
  - Table 3-30 Instruction Set Attributes Register 1 Bit Functions170
  - Table 3-31 Results of Access to the Instruction Set Attributes Register 1170
  - Figure 3-23 Instruction Set Attributes Register 2 Format171
  - Table 3-32 Instruction Set Attributes Register 2 Bit Functions171
  - Figure 3-24 Instruction Set Attributes Register 3 Format172
  - Table 3-33 Results of Access to the Instruction Set Attributes Register 2172
  - Table 3-34 Instruction Set Attributes Register 3 Bit Functions173
  - Table 3-35 Results of Access to the Instruction Set Attributes Register 3173
  - Figure 3-25 Instruction Set Attributes Register 4 Format174
  - Table 3-36 Instruction Set Attributes Register 4 Bit Functions174
  - Table 3-37 Results of Access to the Instruction Set Attributes Register 4175
  - Table 3-38 Results of Access to the Instruction Set Attributes Register 5175
  - Figure 3-26 Control Register Format176
  - Figure177
  - Table 3-39 Control Register Bit Functions177
  - Figure179
  - Table 3-40 Results of Access to the Control Register179
  - Table 3-41 Resultant B Bit, U Bit, and EE Bit Values180
  - Figure 3-27 Auxiliary Control Register Format181
  - Table 3-42 Auxiliary Control Register Bit Functions181
  - Table 3-43 Results of Access to the Auxiliary Control Register182
  - Figure 3-28 Coprocessor Access Control Register Format183
  - Table 3-44 Coprocessor Access Control Register Bit Functions183
  - Table 3-45 Results of Access to the Coprocessor Access Control Register183
  - Figure 3-29 Secure Configuration Register Format184
  - Table 3-46 Secure Configuration Register Bit Functions184
  - Table 3-47 Operation of the FW and FIQ Bits185
  - Table 3-48 Operation of the AW and EA Bits185
  - Figure 3-30 Secure Debug Enable Register Format186
  - Table 3-49 Secure Debug Enable Register Bit Functions186
  - Table 3-50 Results of Access to the Coprocessor Access Control Register187
  - Figure 3-31 Non-Secure Access Control Register Format188
  - Table 3-51 Non-Secure Access Control Register Bit Functions188
  - Figure 3-32 Translation Table Base Register 0 Format189
  - Table 3-52 Results of Access to the Auxiliary Control Register189
  - Table 3-53 Translation Table Base Register 0 Bit Functions190
  - Table 3-54 Results of Access to the Translation Table Base Register 0190
  - Figure 3-33 Translation Table Base Register 1 Format191
  - Table 3-55 Translation Table Base Register 1 Bit Functions191
  - Table 3-56 Results of Access to the Translation Table Base Register 1192
  - Figure 3-34 Translation Table Base Control Register Format193
  - Table 3-57 Translation Table Base Control Register Bit Functions193
  - Table 3-58 Results of Access to the Translation Table Base Control Register194
  - Figure 3-35 Domain Access Control Register Format195
  - Table 3-59 Domain Access Control Register Bit Functions195
  - Table 3-60 Results of Access to the Domain Access Control Register195
  - Figure 3-36 Data Fault Status Register Format196
  - Table 3-61 Data Fault Status Register Bit Functions196
  - Figure 3-37 Instruction Fault Status Register Format198
  - Table 3-62 Results of Access to the Data Fault Status Register198
  - Table 3-63 Instruction Fault Status Register Bit Functions199
  - Table 3-64 Results of Access to the Instruction Fault Status Register199
  - Table 3-65 Results of Access to the Fault Address Register200
  - Table 3-66 Results of Access to the Instruction Fault Address Register201
  - Figure 3-38 Cache Operations202
  - Figure 3-39 Cache Operations with MCRR Instructions203
  - Figure 3-40 C7 Format for Set and Index204
  - Table 3-67 Functional Bits of C7 for Set and Index204
  - Table 3-68 Cache Size and S Parameter Dependency204
  - Figure 3-41 C7 Format for MVA205
  - Figure 3-42 Format of C7 for VA205
  - Table 3-69 Functional Bits of C7 for MVA205
  - Table 3-70 Functional Bits of C7 for VA Format206
  - Table 3-71 Cache Operations for Entire Cache206
  - Table 3-72 Cache Operations for Single Lines207
  - Table 3-73 Cache Operations for Address Ranges208
  - Figure209
  - Figure 3-43 Cache Dirty Status Register Format210
  - Table 3-74 Cache Dirty Status Register Bit Functions210
  - Figure 3-44 C7 Format for Flush Branch Target Entry Using MVA211
  - Table 3-75 Cache Operations Flush Functions211
  - Table 3-76 Flush Branch Target Entry Using MVA Bit Functions211
  - Figure 3-45 PA Register Format for Successful Translation212
  - Figure 3-46 PA Register Format for Aborted Translation212
  - Table 3-77 PA Register for Successful Translation Bit Functions212
  - Table 3-78 PA Register for Unsuccessful Translation Bit Functions213
  - Table 3-79 Results of Access to the Data Synchronization Barrier Operation216
  - Table 3-80 Results of Access to the Data Memory Barrier Operation217
  - Table 3-81 Results of Access to the Wait for Interrupt Operation217
  - Table 3-82 Results of Access to the TLB Operations Register218
  - Figure 3-47 TLB Operations Register MVA and ASID Format219
  - Figure 3-48 TLB Operations Register ASID Format219
  - Figure 3-49 Instruction and Data Cache Lockdown Register Formats220
  - Table 3-83 Instruction and Data Cache Lockdown Register Bit Functions220
  - Table 3-84 Results of Access to the Instruction and Data Cache Lockdown Register220
  - Figure 3-50 Data TCM Region Register Format222
  - Table 3-85 Data TCM Region Register Bit Functions222
  - Figure 3-51 Instruction TCM Region Register Format223
  - Table 3-86 Results of Access to the Data TCM Region Register223
  - Table 3-87 Instruction TCM Region Register Bit Functions224
  - Figure 3-52 Data TCM Non-Secure Control Access Register Format225
  - Table 3-88 Results of Access to the Instruction TCM Region Register225
  - Table 3-89 Data TCM Non-Secure Control Access Register Bit Functions226
  - Table 3-90 Effects of NS Items for Data TCM Operation226
  - Figure 3-53 Instruction TCM Non-Secure Control Access Register Format227
  - Table 3-91 Instruction TCM Non-Secure Control Access Register Bit Functions227
  - Table 3-92 Effects of NS Items for Instruction TCM Operation227
  - Figure 3-54 TCM Selection Register Format228
  - Table 3-93 TCM Selection Register Bit Functions228
  - Figure 3-55 Cache Behavior Override Register Format229
  - Table 3-94 Results of Access to the TCM Selection Register229
  - Table 3-95 Cache Behavior Override Register Bit Functions230
  - Table 3-96 Results of Access to the Cache Behavior Override Register230
  - Figure 3-56 TLB Lockdown Register Format232
  - Table 3-97 TLB Lockdown Register Bit Functions232
  - Table 3-98 Results of Access to the TLB Lockdown Register232
  - Figure 3-57 Primary Region Remap Register Format234
  - Table 3-99 Primary Region Remap Register Bit Functions234
  - Figure 3-58 Normal Memory Remap Register Format235
  - Table 3-100 Encoding for the Remapping of the Primary Memory Type235
  - Table 3-101 Normal Memory Remap Register Bit Functions235
  - Table 3-102 Remap Encoding for Inner or Outer Cacheable Attributes236
  - Table 3-103 Results of Access to the Memory Region Remap Registers236
  - Figure 3-59 DMA Identification and Status Registers Format238
  - Table 3-104 DMA Identification and Status Register Bit Functions238
  - Table 3-105 DMA Identification and Status Register Functions238
  - Table 3-106 Results of Access to the DMA Identification and Status Registers239
  - Figure 3-60 DMA User Accessibility Register Format240
  - Table 3-107 DMA User Accessibility Register Bit Functions240
  - Table 3-108 Results of Access to the DMA User Accessibility Register240
  - Figure 3-61 DMA Channel Number Register Format241
  - Table 3-109 DMA Channel Number Register Bit Functions241
  - Table 3-110 Results of Access to the DMA Channel Number Register241
  - Table 3-111 Results of Access to the DMA Enable Registers243
  - Figure 3-62 DMA Control Register Format244
  - Table 3-112 DMA Control Register Bit Functions244
  - Table 3-113 Results of Access to the DMA Control Register245
  - Table 3-114 Results of Access to the DMA Internal Start Address Register246
  - Table 3-115 Results of Access to the DMA External Start Address Register247
  - Table 3-116 Results of Access to the DMA Internal End Address Register248
  - Figure 3-63 DMA Channel Status Register Format249
  - Table 3-117 DMA Channel Status Register Bit Functions249
  - Table 3-118 Results of Access to the DMA Channel Status Register251
  - Figure 3-64 DMA Context ID Register Format252
  - Table 3-119 DMA Context ID Register Bit Functions252
  - Table 3-120 Results of Access to the DMA Context ID Register252
  - Figure 3-65 Secure or Non-Secure Vector Base Address Register Format253
  - Table 3-121 Secure or Non-Secure Vector Base Address Register Bit Functions253
  - Figure 3-66 Monitor Vector Base Address Register Format254
  - Table 3-122 Results of Access to the Secure or Non-Secure Vector Base Address Register254
  - Table 3-123 Monitor Vector Base Address Register Bit Functions255
  - Table 3-124 Results of Access to the Monitor Vector Base Address Register255
  - Figure 3-67 Interrupt Status Register Format256
  - Table 3-125 Interrupt Status Register Bit Functions256
  - Table 3-126 Results of Access to the Interrupt Status Register256
  - Figure 3-68 FCSE PID Register Format258
  - Table 3-127 FCSE PID Register Bit Functions258
  - Table 3-128 Results of Access to the FCSE PID Register258
  - Figure 3-69 Address Mapping with the FCSE PID Register259
  - Figure 3-70 Context ID Register Format260
  - Table 3-129 Context ID Register Bit Functions260
  - Table 3-130 Results of Access to the Context ID Register260
  - Table 3-131 Results of Access to the Thread and Process ID Registers261
  - Figure 3-71 Peripheral Port Memory Remap Register Format262
  - Table 3-132 Peripheral Port Memory Remap Register Bit Functions263
  - Table 3-133 Results of Access to the Peripheral Port Remap Register263
  - Figure 3-72 Secure User and Non-Secure Access Validation Control Register Format264
  - Table 3-134 Secure User and Non-Secure Access Validation Control Register Bit Functions264
  - Figure 3-73 Performance Monitor Control Register Format265
  - Table 3-136 Performance Monitor Control Register Bit Functions266
  - Table 3-137 Performance Monitoring Events267
  - Table 3-138 Results of Access to the Performance Monitor Control Register269
  - Table 3-139 Results of Access to the Cycle Counter Register270
  - Table 3-140 Results of Access to the Count Register 0271
  - Table 3-141 Results of Access to the Count Register 1272
- Table 3-142 System Validation Counter Register Operations272
- Table 3-143 Results of Access to the System Validation Counter Register273
- Figure 3-74 System Validation Counter Register Format for External Debug Request Counter273
- Table 3-144 System Validation Operations Register Functions274
- Table 3-145 Results of Access to the System Validation Operations Register275
- Table 3-146 System Validation Cache Size Mask Register Bit Functions277
- Figure 3-75 System Validation Cache Size Mask Register Format277
- Table 3-147 Results of Access to the System Validation Cache Size Mask Register278
- Table 3-148 TLB Lockdown Index Register Bit Functions281
- Figure 3-76 TLB Lockdown Index Register Format281
- Figure 3-77 TLB Lockdown VA Register Format281
- Table 3-149 TLB Lockdown VA Register Bit Functions282
- Table 3-150 TLB Lockdown PA Register Bit Functions282
- Figure 3-78 TLB Lockdown PA Register Format282
- Table 3-151 Access Permissions APX and AP Bit Fields Encoding283
- Table 3-152 TLB Lockdown Attributes Register Bit Functions283
- Figure 3-79 TLB Lockdown Attributes Register Format283
- Table 3-153 Results of Access to the TLB Lockdown Access Registers284
Chapter 4 Unaligned and Mixed-Endian Data Access Support287
- About Unaligned and Mixed-Endian Support287
- Unaligned Access Support288
- Table 4-1 Unaligned Access Handling289
- Endian Support291
- Figure 4-1 Load Unsigned Byte291
- Figure 4-2 Load Signed Byte291
- Figure 4-3 Store Byte292
- Figure 4-4 Load Unsigned Halfword, Little-Endian292
- Figure 4-5 Load Unsigned Halfword, Big-Endian293
- Figure 4-6 Load Signed Halfword, Little-Endian293
- Figure 4-7 Load Signed Halfword, Big-Endian294
- Figure 4-8 Store Halfword, Little-Endian294
- Figure 4-9 Store Halfword, Big-Endian295
- Figure 4-10 Load Word, Little-Endian295
- Figure 4-11 Load Word, Big-Endian296
- Figure 4-12 Store Word, Little-Endian296
- Figure 4-13 Store Word, Big-Endian297
- Operation of Unaligned Accesses298
- Table 4-2 Memory Access Types298
- Table 4-3 Unalignment Fault Occurrence When Access Behavior Is Architecturally Unpredictable299
- Mixed-Endian Access Support302
- Table 4-4 Legacy Endianness Using CP15 C1302
- Figure302
- Table 4-5 Mixed-Endian Configuration304
- Table 4-6 B Bit, U Bit, and EE Bit Settings304
- Instructions to Reverse Bytes in a General-Purpose Register305
- Instructions to Change the CPSR E Bit306
Chapter 5 Program Flow Prediction307
- About Program Flow Prediction308
- Branch Prediction310
- Return Stack313
- Memory Barriers314
- ARM1176JZF-S IMB Implementation316
Chapter 6 Memory Management Unit319
- About the MMU319
- TLB Organization321
  - Figure321
  - Figure323
- Memory Access Sequence324
- Enabling and Disabling the MMU326
- Memory Access Control328
- Table 6-1 Access Permission Bit Encoding329
- Figure330
- Memory Region Attributes331
- Figure331
- Table 6-2 TEX Field, and C and B Bit Encodings Used in Page Table Formats332
- Table 6-3 Cache Policy Bits333
- Table 6-4 Inner and Outer Cache Policy Implementation Options333
- Table 6-5 Effect of Remapping Memory with TEX Remap = 1334
- Table 6-6 Values that Remap the Shareable Attribute335
- Table 6-7 Primary Region Type Encoding335
- Table 6-8 Inner and Outer Region Remap Encoding335
- Figure336
- Memory Attributes and Types337
- Table 6-9 Memory Attributes337
- Figure340
- Figure 6-1 Memory Ordering Restrictions341
- Table 6-10 Memory Region Backwards Compatibility343
- MMU Aborts344
- MMU Fault Checking346
  - Translation Table Managed TLB Fault Checking Sequence Part 1347
  - Figure347
  - Translation Table Managed TLB Fault Checking Sequence Part 2348
- Fault Status and Address351
- Table 6-11 Fault Status Register Encoding351
- Table 6-12 Summary of Aborts352
- Hardware Page Table Translation353
- Figure 6-4 Backwards-Compatible First-Level Descriptor Format354
- Figure 6-5 Backwards-Compatible Second-Level Descriptor Format355
- Figure 6-6 Backwards-Compatible Section, Supersection, and Page Translation355
- Figure 6-7 Armv6 First-Level Descriptor Formats with Subpages Disabled356
- Figure 6-8 Armv6 Second-Level Descriptor Format357
- Figure 6-9 Armv6 Section, Supersection, and Page Translation358
- MMU Descriptors360
- Table 6-13 Translation Table Size360
- Figure 6-10 Creating a First-Level Descriptor Address361
- Table 6-14 Access Types from First-Level Descriptor Bit Values362
- Translation for a 1MB Section, Armv6 Format363
- Translation for a 1MB Section, Backwards-Compatible Format363
- Table 6-15 Access Types from Second-Level Descriptor Bit Values364
- Generating a Second-Level Page Table Address364
- Figure 6-14 Large Page Table Walk, Armv6 Format365
- Figure 6-15 Large Page Table Walk, Backwards-Compatible Format366
- Figure 6-16 4KB Small Page or 1KB Small Subpage Translations, Backwards-Compatible Format367
- Figure 6-17 4KB Extended Small Page Translations, Armv6 Format368
- Figure 6-18 4KB Extended Small Page or 1KB Extended Small Subpage Translations369
- MMU Software-Accessible Registers370
- Table 6-16 CP15 Register Functions370
- Table 6-17 CP14 Register Functions371
Chapter 7 Level One Memory System373
- About the Level One Memory System373
- Figure373
- Cache Organization374
- Figure374
- Figure 7-1 Level One Cache Block Diagram375
- Tightly-Coupled Memory378
- Table 7-1 TCM Configurations378
- Table 7-2 Access to Non-Secure TCM379
- Table 7-3 Access to Secure TCM379
- Dma381
- TCM and Cache Interactions383
- Table 7-4 Summary of Data Accesses to TCM and Caches385
- Table 7-5 Summary of Instruction Accesses to TCM and Caches386
- Write Buffer387
Chapter 8 Level Two Interface389
- About the Level Two Interface389
- Figure 8-1 Level Two Interconnect Interfaces389
- Table 8-1 AXI Parameters for the Level 2 Interconnect Interfaces390
- Dma Interface392
- Synchronization Primitives393
- AXI Control Signals in the Processor395
- Figure 8-2 Channel Architecture of Reads395
- Figure 8-3 Channel Architecture of Writes395
- Table 8-2 Axlen[3:0] Encoding397
- Table 8-3 Axsize[2:0] Encoding398
- Table 8-4 Axburst[1:0] Encoding398
- Table 8-5 Axlock[1:0] Encoding398
- Table 8-6 Axcache[3:0] Encoding399
- Table 8-7 Axprot[2:0] Encoding399
- Table 8-8 Axsideband[4:1] Encoding400
- Table 8-9 ARSIDEBANDI[4:1] Encoding400
- Instruction Fetch Interface Transfers401
- Table 8-10 AXI Signals for Cacheable Fetches401
- Table 8-11 AXI Signals for Noncacheable Fetches401
- Data Read/Write Interface Transfers402
  - Table 8-12 Linefill Behavior on the AXI Interface402
  - Table 8-13 Noncacheable LDRB403
  - Table 8-14 Noncacheable LDRH403
  - Table 8-15 Noncacheable LDR or LDM1404
  - Table 8-16 Noncacheable LDRD or LDM2404
  - Noncacheable LDRD or LDM2 from Word 7405
  - Table 8-18 Noncacheable LDM3, Strongly Ordered or Device Memory405
  - Table 8-19 Noncacheable LDM3, Noncacheable Memory or Cache Disabled405
- Noncacheable LDM3 from Word 6, or 7405
- Table 8-21 Noncacheable LDM4, Strongly Ordered or Device Memory406
- Table 8-22 Noncacheable LDM4, Noncacheable Memory or Cache Disabled406
- Noncacheable LDM4 from Word 5, 6, or 7406
- Table 8-24 Noncacheable LDM5, Strongly Ordered or Device Memory407
- Table 8-25 Noncacheable LDM5, Noncacheable Memory or Cache Disabled407
- Table 8-27 Noncacheable LDM6, Strongly Ordered or Device Memory407
- Noncacheable LDM5 from Word 4, 5, 6, or 7407
- Table 8-28 Noncacheable LDM6, Noncacheable Memory or Cache Disabled408
- Noncacheable LDM6 from Word 3, 4, 5, 6, or 7408
- Table 8-30 Noncacheable LDM7, Strongly Ordered or Device Memory408
- Table 8-31 Noncacheable LDM7, Noncacheable Memory or Cache Disabled408
- Table 8-32 Noncacheable LDM7 from Word 2, 3, 4, 5, 6, or 7409
- Noncacheable LDM8 from Word 1, 2, 3, 4, 5, 6, or 7409
- Table 8-35 Noncacheable LDM9409
- Table 8-36 Noncacheable LDM10410
- Table 8-37 Noncacheable LDM11410
- Table 8-38 Noncacheable LDM12411
- Table 8-39 Noncacheable LDM13411
- Table 8-40 Noncacheable LDM14411
- Table 8-41 Noncacheable LDM15412
- Table 8-42 Noncacheable LDM16412
- Table 8-43 Half-Line Write-Back413
- Table 8-44 Full-Line Write-Back413
- Table 8-45 Cacheable Write-Through or Noncacheable STRB414
- Table 8-46 Cacheable Write-Through or Noncacheable STRH414
- Table 8-47 Cacheable Write-Through or Noncacheable STR or STM1415
- Cacheable Write-Through or Noncacheable STRD or STM2 to Words 0, 1, 2, 3, 4, 5, or 6416
- Cacheable Write-Through or Noncacheable STM2 to Word 7416
- Cacheable Write-Through or Noncacheable STM3 to Words 0, 1, 2, 3, 4, or 5416
- Cacheable Write-Through or Noncacheable STM3 to Words 6 or 7416
- Cacheable Write-Through or Noncacheable STM4 to Word 0, 1, 2, 3, or 4417
- Cacheable Write-Through or Noncacheable STM4 to Word 5, 6, or 7417
- Cacheable Write-Through or Noncacheable STM5 to Word 0, 1, 2, or 3417
- Cacheable Write-Through or Noncacheable STM5 to Word 4, 5, 6, or 7417
- Cacheable Write-Through or Noncacheable STM6 to Word 0, 1, or 2418
- Cacheable Write-Through or Noncacheable STM6 to Word 3, 4, 5, 6, or 7418
- Cacheable Write-Through or Noncacheable STM7 to Word 0 or 1418
- Cacheable Write-Through or Noncacheable STM7 to Word 2, 3, 4, 5, 6 or 7419
- Cacheable Write-Through or Noncacheable STM8 to Word 0419
- Cacheable Write-Through or Noncacheable STM8 to Word 1, 2, 3, 4, 5, 6, or 7419
- Table 8-62 Cacheable Write-Through or Noncacheable STM9419
- Table 8-63 Cacheable Write-Through or Noncacheable STM10420
- Table 8-64 Cacheable Write-Through or Noncacheable STM11420
- Table 8-65 Cacheable Write-Through or Noncacheable STM12421
- Table 8-66 Cacheable Write-Through or Noncacheable STM13421
- Table 8-67 Cacheable Write-Through or Noncacheable STM14422
- Table 8-68 Cacheable Write-Through or Noncacheable STM15422
- Table 8-69 Cacheable Write-Through or Noncacheable STM16423
- Peripheral Interface Transfers424
- Table 8-70 Example Peripheral Interface Reads and Writes424
- Endianness425
- Figure 8-4 Swizzling of Data and Strobes in BE-32 Big-Endian Configuration425
- Locked Access426
Chapter 9 Clocking and Resets428
- About Clocking and Resets428
- Clocking and Resets with no IEM429
- Figure 9-1 Processor Clocks with no IEM429
- Figure 9-2 Read Latency with no IEM430
- Clocking and Resets with IEM431
- Figure431
- Figure 9-3 Processor Clocks with IEM432
- Figure 9-4 Processor Synchronization with IEM432
- Figure 9-5 Read Latency with IEM434
- Reset Modes436
- Table 9-1 Reset Modes436
- Figure 9-6 Power-On Reset436
Chapter 10 Power Control438
- About Power Control439
- Power Management440
- VFP Shutdown443
- Intelligent Energy Management444
- Figure444
Chapter 11 Coprocessor Interface445
- Figure 10-1 IEM Structure445
- About the Coprocessor Interface447
- Coprocessor Pipeline448
- Table 11-1 Coprocessor Instructions448
- Table 11-2 Coprocessor Control Signals449
- Figure449
- Figure 11-1 Core and Coprocessor Pipelines450
- Figure 11-2 Coprocessor Pipeline and Queues450
- Figure 11-3 Coprocessor Pipeline451
- Table 11-3 Pipeline Stage Update452
- Token Queue Management454
- Figure 11-4 Token Queue Buffers454
- Table 11-4 Addressing of Queue Buffers455
- Figure 11-5 Queue Reading and Writing455
- Figure 11-6 Queue Flushing456
- Token Queues457
- Figure 11-7 Instruction Queue457
- Data Transfer460
- Figure 11-8 Coprocessor Data Transfer460
- Figure 11-9 Instruction Iteration for Loads461
- Figure 11-10 Load Data Buffering462
- Operations464
- Table 11-5 Retirement Conditions465
- Multiple Coprocessors467
Chapter 12 Vectored Interrupt Controller Port469
- About the PL192 Vectored Interrupt Controller469
- About the Processor VIC Port470
- Table 12-1 VIC Port Signals470
- Figure 12-1 Connection of a VIC to the Processor470
- Timing of the VIC Port472
- Figure 12-2 VIC Port Timing Example472
- Interrupt Entry Flowchart474
- Figure 12-3 Interrupt Entry Sequence474
Chapter 13 Debug476
- Debug Systems476
- Figure 13-1 Typical Debug System476
- About the Debug Unit477
- Debug Registers479
- Table 13-1 Terms Used in Register Descriptions479
- Table 13-2 CP14 Debug Register Map479
- Figure 13-2 Debug ID Register Format480
- Table 13-3 Debug ID Register Bit Field Definition481
- Figure481
- Table 13-4 Debug Status and Control Register Bit Field Definitions482
- Figure 13-3 Debug Status and Control Register Format482
- Table 13-5 Data Transfer Register Bit Field Definitions486
- Figure 13-4 DTR Format486
- Figure 13-5 Vector Catch Register Format487
- Table 13-6 Vector Catch Register Bit Field Definitions488
- Table 13-7 Summary of Debug Entry and Exception Conditions488
- Table 13-8 Processor Breakpoint and Watchpoint Registers490
- Table 13-9 Breakpoint Value Registers, Bit Field Definition491
- Table 13-10 Processor Breakpoint Control Registers491
- Figure 13-6 Breakpoint Control Registers, Format491
- Table 13-11 Breakpoint Control Registers, Bit Field Definitions492
- Table 13-12 Meaning of BCR[22:20] Bits493
- Table 13-13 Processor Watchpoint Value Registers494
- Table 13-14 Watchpoint Value Registers, Bit Field Definitions495
- Table 13-15 Processor Watchpoint Control Registers495
- Table 13-16 Watchpoint Control Registers, Bit Field Definitions495
- Figure 13-7 Watchpoint Control Registers, Format495
- Table 13-17 Debug State Cache Control Register Bit Functions497
- Table 13-18 Debug State MMU Control Register Bit Functions498
- CP14 Registers Reset499
- CP14 Debug Instructions500
- Table 13-19 CP14 Debug Instructions500
- Table 13-20 Debug Instruction Execution501
- External Debug Interface502
- Table 13-21 Secure Debug Behavior502
- Changing the Debug Enable Signals505
- Debug Events506
- Table 13-22 Behavior of the Processor on Debug Events507
- Table 13-23 Setting of CP15 Registers on Debug Events508
- Debug Exception509
- Table 13-24 Values in the Link Register after Exceptions510
- 13.10 Debug State511
- Table 13-25 Read PC Value after Debug State Entry513
- Table 13-26 Example Memory Operation Sequence515
- 13.11 Debug Communications Channel516
- 13.12 Debugging in a Cached System517
- 13.13 Debugging in a System with Tlbs518
- 13.14 Monitor Debug-Mode Debugging519
- 13.15 Halting Debug-Mode Debugging524
- 13.16 External Signals526
Chapter 14 Debug Test Access Port528
- Debug Test Access Port and Debug State528
- Figure 14-1 JTAG DBGTAP State Machine Diagram528
- Synchronizing Realview ICE529
- Figure 14-2 Realview ICE Clock Synchronization529
- Entering Debug State530
- Exiting Debug State531
- The DBGTAP Port and Debug Registers532
- Table 14-1 Supported Public Instructions532
- Debug Registers534
- Figure 14-3 Bypass Register Bit Order534
- Figure 14-4 Device ID Code Register Bit Order535
- Figure 14-5 Instruction Register Bit Order535
- Figure 14-6 Scan Chain Select Register Bit Order536
- Figure 14-7 Scan Chain 0 Bit Order537
- Figure 14-8 Scan Chain 1 Bit Order537
- Figure 14-9 Scan Chain 4 Bit Order539
- Figure 14-10 Scan Chain 5 Bit Order, EXTEST Selected541
- Figure 14-11 Scan Chain 5 Bit Order, INTEST Selected541
- Figure 14-12 Scan Chain 6 Bit Order543
- Figure 14-13 Scan Chain 7 Bit Order544
- Table 14-2 Scan Chain 7 Register Map545
- Using the Debug Test Access Port547
- Figure 14-14 Behavior of the Itrsel IR Instruction548
- Debug Sequences555
- Programming Debug Events566
- 14.10 Monitor Debug-Mode Debugging568
Chapter 15 Trace Interface Port569
- About the ETM Interface570
- Table 15-1 Instruction Interface Signals570
- Table 15-2 ETMIACTL[17:0]571
- Table 15-3 ETMIASECCTL[1:0]572
- Table 15-4 Data Address Interface Signals572
- Table 15-5 ETMDACTL[17:0]573
- Table 15-6 Data Value Interface Signals574
- Table 15-7 ETMDDCTL[3:0]574
- Table 15-8 ETMPADV[2:0]574
- Table 15-9 Coprocessor Interface Signals575
- Table 15-10 ETMCPSECCTL[1:0] Format575
- Figure 15-1 ETMCPADDRESS Format575
- Table 15-11 Other Connections576
- Figure576
Chapter 16 Cycle Timings and Interlock Behavior577
- About Cycle Timings and Interlock Behavior578
- Table 16-1 Pipeline Stages579
- Conditional Instructions580
- Table 16-2 Definition of Cycle Timing Terms581
- Register Interlock Examples582
- Table 16-3 Register Interlock Examples582
- Data Processing Instructions583
- Table 16-4 Data Processing Instruction Cycle Timing Behavior if Destination Is Not PC583
- Table 16-5 Data Processing Instruction Cycle Timing Behavior if Destination Is the PC583
- QADD, QDADD, QSUB, and QDSUB Instructions585
- Table 16-6 QADD, QDADD, QSUB, and QDSUB Instruction Cycle Timing Behavior585
- Armv6 Media Data-Processing586
- Table 16-7 Armv6 Media Data-Processing Instructions Cycle Timing Behavior586
- Armv6 Sum of Absolute Differences (SAD)587
- Table 16-8 Armv6 Sum of Absolute Differences Instruction Timing Behavior587
- Table 16-9 Example Interlocks587
- Multiplies588
- Table 16-10 Example Multiply Instruction Cycle Timing Behavior588
- Branches590
- Table 16-11 Branch Instruction Cycle Timing Behavior590
- Processor State Updating Instructions591
- Table 16-12 Processor State Updating Instructions Cycle Timing Behavior591
- 16.10 Single Load and Store Instructions592
- Table 16-13 Cycle Timing Behavior for Stores and Loads, Other than Loads to the PC592
- Table 16-14 Cycle Timing Behavior for Loads to the PC593
- Table 16-15 <Addr_Md_1Cycle> and <Addr_Md_2Cycle> LDR Example Instruction Explanation593
- 16.11 Load and Store Double Instructions595
- Table 16-16 Load and Store Double Instructions Cycle Timing Behavior595
- Table 16-17 <Addr_Md_1Cycle> and <Addr_Md_2Cycle> LDRD Example Instruction Explanation595
- 16.12 Load and Store Multiple Instructions597
- Table 16-18 Cycle Timing Behavior of Load and Store Multiples, Other than Load Multiples Including the PC597
- Table 16-19 Cycle Timing Behavior of Load Multiples, Where the PC Is in the Register List598
- 16.13 RFE and SRS Instructions599
- Table 16-20 RFE and SRS Instructions Cycle Timing Behavior599
- 16.14 Synchronization Instructions600
- Table 16-21 Synchronization Instructions Cycle Timing Behavior600
- 16.15 Coprocessor Instructions601
- Table 16-22 Coprocessor Instructions Cycle Timing Behavior601
- 16.16 SVC, SMC, BKPT, Undefined, and Prefetch Aborted Instructions602
- Table 16-23 SVC, BKPT, Undefined, Prefetch Aborted Instructions Cycle Timing Behavior602
- 16.17 no Operation603
- 16.18 Thumb Instructions604
Chapter 17 AC Characteristics605
- Processor Timing Diagrams606
- Processor Timing Parameters607
- Table 17-1 Global Signals607
- Table 17-2 AXI Signals607
- Table 17-3 Coprocessor Signals609
- Table 17-4 ETM Interface Signals609
- Table 17-5 Interrupt Signals609
- Table 17-6 Debug Interface Signals610
- Table 17-7 Test Signals610
- Table 17-8 Static Configuration Signals610
- Table 17-9 Trustzone Internal Signals611
Chapter 18 Introduction to the VFP Coprocessor612
- About the VFP11 Coprocessor613
- Applications614
- Coprocessor Interface615
- VFP11 Coprocessor Pipelines616
- Figure 18-1 FMAC Pipeline617
- Figure 18-2 DS Pipeline619
- Figure 18-3 LS Pipeline620
- Modes of Operation622
- Short Vector Instructions624
- Parallel Execution of Instructions625
- VFP11 Treatment of Branch Instructions626
- Writing Optimal VFP11 Code627
- 18.10 VFP11 Revision Information628
Chapter 19 The VFP Register File629
- About the Register File630
- Figure630
- Register File Internal Formats631
- Figure 19-1 Single-Precision Data Format631
- Figure 19-2 Double-Precision Data Format632
- Decoding the Register File633
- Figure 19-3 Register File Access633
- Loading Operands from ARM11 Registers634
- Table 19-1 VFP11 MCR Instructions634
- Table 19-2 VFP11 MRC Instructions634
- Table 19-3 VFP11 MCRR Instructions634
- Table 19-4 VFP11 MRRC Instructions635
- Maintaining Consistency in Register Precision636
- Data Transfer between Memory and VFP11 Registers637
- Table 19-5 Single-Precision Data Memory Images and Byte Addresses637
- Table 19-6 Double-Precision Data Memory Images and Byte Addresses637
- Access to Register Banks in CDP Operations638
- Figure 19-4 Register Banks638
- Table 19-7 Single-Precision Three-Operand Register Usage641
- Table 19-8 Single-Precision Two-Operand Register Usage641
- Table 19-9 Double-Precision Three-Operand Register Usage641
- Table 19-10 Double-Precision Two-Operand Register Usage641
Chapter 20 VFP Programmer's Model642
- About the Programmer's Model643
- Compliance with the IEEE 754 Standard644
- Table 20-1 Default Nan Values645
- Table 20-2 Qnan and Snan Handling646
- Armv5Te Coprocessor Extensions649
- Figure 20-1 FMDRR Instruction Format649
- Figure 20-2 FMRRD Instruction Format650
- Figure 20-3 FMSRR Instruction Format651
- Figure 20-4 FMRRS Instruction Format652
- VFP11 System Registers653
- Table 20-3 VFP11 System Registers653
- Table 20-4 Accessing VFP11 System Registers654
- Figure 20-5 Floating-Point System ID Register654
- Table 20-5 FPSID Bit Fields655
- Figure 20-6 Floating-Point Status and Control Register655
- Table 20-6 Encoding of the Floating-Point Status and Control Register656
- Table 20-7 Vector Length and Stride Combinations657
- Table 20-8 Encoding of the Floating-Point Exception Register658
- Figure 20-7 Floating-Point Exception Register658
- Table 20-9 Media and VFP Feature Register 0 Bit Functions660
- Figure 20-8 Media and VFP Feature Register 0 Format660
- Table 20-10 Media and VFP Feature Register 1 Bit Functions661
- Figure 20-9 Media and VFP Feature Register 1 Format661
Chapter 21 VFP Instruction Execution662
- About Instruction Execution663
- Serializing Instructions664
- Interrupting the VFP11 Coprocessor665
- Forwarding666
- Hazards667
- Operation of the Scoreboards668
- Table 21-1 Single-Precision Source Register Locking669
- Table 21-2 Single-Precision Source Register Clearing670
- Table 21-3 Double-Precision Source Register Locking671
- Table 21-4 Double-Precision Source Register Clearing for One-Cycle Instructions672
- Table 21-5 Double-Precision Source Register Clearing for Two-Cycle Instructions672
- Data Hazards in Full-Compliance Mode674
- Copyright © 2004-2009 ARM Limited. All Rights Reserved674
- Table 21-6 FCMPS-FMSTAT RAW Hazard674
- Table 21-7 FLDM-FADDS RAW Hazard675
- Table 21-8 FLDM-Short Vector FADDS RAW Hazard675
- Table 21-9 FMULS-FADDS RAW Hazard676
- Table 21-10 Short Vector FMULS-FLDMS war Hazard676
- Data Hazards in Runfast Mode677
- Table 21-11 Short Vector FMULS-FLDMS war Hazard in Runfast Mode677
- Resource Hazards678
- Table 21-12 FLDM-FLDS-FADDS Resource Hazard679
- Table 21-13 FLDM-Short Vector FMULS Resource Hazard679
- Table 21-14 Short Vector FDIVS-FADDS Resource Hazard680
- 21.10 Parallel Execution681
- Table 21-15 Parallel Execution in All Three Pipelines682
- 21.11 Execution Timing683
- Table 21-16 Throughput and Latency Cycle Counts for VFP11 Instructions683
Chapter 22 VFP Exception Handling684
- About Exception Processing685
- Bounced Instructions686
- Support Code688
- Exception Processing691
- Table 22-1 Exceptional Short Vector FMULD Followed by Load/Store Instructions692
- Table 22-2 Exceptional Short Vector FADDS with a FADDS in the Pretrigger Slot693
- Table 22-3 Exceptional Short Vector FADDD with an FMACS Trigger Instruction694
- Input Subnormal Exception695
- Invalid Operation Exception696
- Table 22-4 Possible Invalid Operation Exceptions696
- Table 22-5 Default Results for Invalid Conversion Inputs697
- Division by Zero Exception698
- Overflow Exception699
- Table 22-6 Rounding Mode Overflow Results699
- Underflow Exception700
- 22.10 Inexact Exception701
- 22.11 Input Exceptions702
- 22.12 Arithmetic Exceptions703
- Table 22-7 LSA and USA Determination703
- Table 22-8 FADD Family Bounce Thresholds704
- Table 22-9 FMUL Family Bounce Thresholds705
- Table 22-10 FDIV Bounce Thresholds706
- Table 22-11 FCVTSD Bounce Thresholds707
- Table 22-12 Single-Precision Float-To-Integer Bounce Thresholds and Stored Results708
- Table 22-13 Double-Precision Float-To-Integer Bounce Thresholds and Stored Results709
Appendix A Signal Descriptions711
- Global Signals712
- Table A-1 Global Signals712
- Static Configuration Signals714
- Table A-2 Static Configuration Signals714
- Trustzone Internal Signals715
- Table A-3 Trustzone Internal Signals715
- Interrupt Signals, Including VIC Interface716
- Table A-4 Interrupt Signals716
- AXI Interface Signals717
- Table A-5 Port Signal Name Suffixes717
- Table A-6 Instruction Read Port AXI Signal Implementation718
- Table A-7 Data Port AXI Signal Implementation719
- Table A-8 Peripheral Port AXI Signal Implementation720
- Table A-9 DMA Port Signals721
- Coprocessor Interface Signals722
- Table A-10 Core to Coprocessor Signals722
- Table A-11 Coprocessor to Core Signals722
- Debug Interface Signals, Including JTAG724
- Table A-12 Debug Interface Signals724
- ETM Interface Signals725
- Table A-13 ETM Interface Signals725
- Test Signals726
- Table A-14 Test Signals726
Appendix B Summary of ARM1136JF-S and ARM1176JZF-S Processor Differences727
- About the Differences between the ARM1136JF-S and ARM1176JZF-S Processors728
- Summary of Differences729
  - B.2 Summary of Differences729
  - Figure730
- Table B-1 TCM for ARM1176JZF-S Processors732
- Table B-2 CP15 C15 Features Common to ARM1136JF-S and ARM1176JZF-S Processors734
- Table B-3 CP15 C15 Only Found in ARM1136JF-S Processors735
- Table C-1 Differences between Issue G and Issue H738
  - Figure751
  - Figure752

ARM ARM1176JZF-S - Chapter 21 VFP Instruction Execution

Table of Contents

Related product manuals