- •Introduction
- •ARM7TDMI Architecture
- •The THUMB Concept
- •THUMB’s Advantages
- •ARM7TDMI Block Diagram
- •ARM7TDMI Core Diagram
- •ARM7TDMI Functional Diagram
- •Key to signal types
- •Processor Operating States
- •Switching State
- •Entering THUMB state
- •Entering ARM state
- •Memory Formats
- •Big endian format
- •Little endian format
- •Instruction Length
- •Data Types
- •Operating Modes
- •Registers
- •The ARM state register set
- •The THUMB state register set
- •The relationship between ARM and THUMB state registers
- •Accessing Hi registers in THUMB state
- •The Program Status Registers
- •The condition code flags
- •The control bits
- •Exceptions
- •Action on entering an exception
- •Action on leaving an exception
- •Exception entry/exit summary
- •Notes
- •Abort
- •Software interrupt
- •Undefined instruction
- •Exception vectors
- •Exception priorities
- •Not all exceptions can occur at once:
- •Interrupt Latencies
- •Reset
- •Instruction Set Summary
- •Format summary
- •Instruction summary
- •The Condition Field
- •Branch and Exchange (BX)
- •Instruction cycle times
- •Assembler syntax
- •Using R15 as an operand
- •Examples
- •Branch and Branch with Link (B, BL)
- •The link bit
- •Instruction cycle times
- •Assembler syntax
- •Examples
- •Data Processing
- •CPSR flags
- •Shifts
- •Instruction specified shift amount
- •Register specified shift amount
- •Immediate operand rotates
- •Writing to R15
- •Using R15 as an operand
- •TEQ, TST, CMP and CMN opcodes
- •Instruction cycle times
- •Assembler syntax
- •where:
- •Examples
- •PSR Transfer (MRS, MSR)
- •Operand restrictions
- •Reserved bits
- •Example
- •Instruction cycle times
- •Assembler syntax
- •Examples
- •Multiply and Multiply-Accumulate (MUL, MLA)
- •If the operands are interpreted as signed
- •If the operands are interpreted as unsigned
- •Operand restrictions
- •CPSR flags
- •Instruction cycle times
- •Assembler syntax
- •Examples
- •Multiply Long and Multiply-Accumulate Long (MULL,MLAL)
- •Operand restrictions
- •CPSR flags
- •Instruction cycle times
- •For signed instructions SMULL, SMLAL:
- •For unsigned instructions UMULL, UMLAL:
- •Assembler syntax
- •where:
- •Examples
- •Single Data Transfer (LDR, STR)
- •Offsets and auto-indexing
- •Shifted register offset
- •Bytes and words
- •Little endian configuration
- •Big endian configuration
- •Restriction on the use of base register
- •Example:
- •Data aborts
- •Instruction cycle times
- •Assembler syntax
- •Examples
- •Halfword and Signed Data Transfer(LDRH/STRH/LDRSB/LDRSH)
- •Offsets and auto-indexing
- •Halfword load and stores
- •Signed byte and halfword loads
- •Endianness and byte/halfword selection
- •Little endian configuration
- •Big endian configuration
- •Data aborts
- •Instruction cycle times
- •Assembler syntax
- •Examples
- •Block Data Transfer (LDM, STM)
- •The register list
- •Addressing modes
- •Address alignment
- •LDM with R15 in transfer list and S bit set (Mode changes)
- •STM with R15 in transfer list and S bit set (User bank transfer)
- •R15 not in list and S bit set (User bank transfer)
- •Use of R15 as the base
- •Inclusion of the base in the register list
- •Data aborts
- •Aborts during STM instructions
- •Aborts during LDM instructions
- •Instruction cycle times
- •Assembler syntax
- •Addressing mode names
- •Examples
- •Single Data Swap (SWP)
- •Bytes and words
- •Data aborts
- •Instruction cycle times
- •Assembler syntax
- •Examples
- •Software Interrupt (SWI)
- •Return from the supervisor
- •Comment field
- •Instruction cycle times
- •Assembler syntax
- •Examples
- •Supervisor code
- •Coprocessor Data Operations (CDP)
- •The coprocessor fields
- •Instruction cycle times
- •Assembler syntax
- •Examples
- •Coprocessor Data Transfers (LDC, STC)
- •The coprocessor fields
- •Addressing modes
- •Address alignment
- •Data aborts
- •Instruction cycle times
- •Assembler syntax
- •Examples
- •Coprocessor Register Transfers (MRC, MCR)
- •The coprocessor fields
- •Transfers to R15
- •Transfers from R15
- •Instruction cycle times
- •Assembler syntax
- •Examples
- •Undefined Instruction
- •Instruction cycle times
- •Assembler syntax
- •Instruction Set Examples
- •Using the conditional instructions
- •Using conditionals for logical OR
- •Absolute value
- •Multiplication by 4, 5 or 6 (run time)
- •Combining discrete and range tests
- •Division and remainder
- •Overflow detection in the ARM7TDMI
- •Pseudo-random binary sequence generator
- •Multiplication by constant using the barrel shifter
- •Multiplication by 2^n (1,2,4,8,16,32..)
- •Multiplication by 2^n+1 (3,5,9,17..)
- •Multiplication by 2^n-1 (3,7,15..)
- •Multiplication by 6
- •Multiply by 10 and add in extra number
- •General recursive method for Rb := Ra*C, C a constant:
- •Loading a word from an unknown alignment
- •Format Summary
- •Opcode Summary
- •Format 1: move shifted register
- •Operation
- •Instruction cycle times
- •Examples
- •Format 2: add/subtract
- •Operation
- •Instruction cycle times
- •Examples
- •Format 3: move/compare/add/subtract immediate
- •Operations
- •Instruction cycle times
- •Examples
- •Format 4: ALU operations
- •Operation
- •Instruction cycle times
- •Examples
- •Format 5: Hi register operations/branch exchange
- •Operation
- •Instruction cycle times
- •The BX instruction
- •Examples
- •Using R15 as an operand
- •Format 6: PC-relative load
- •Operation
- •Instruction cycle times
- •Examples
- •Format 7: load/store with register offset
- •Operation
- •Instruction cycle times
- •Examples
- •Format 8: load/store sign-extended byte/halfword
- •Operation
- •Instruction cycle times
- •Examples
- •Format 9: load/store with immediate offset
- •Operation
- •Instruction cycle times
- •Examples
- •Format 10: load/store halfword
- •Operation
- •Instruction cycle times
- •Examples
- •Format 11: SP-relative load/store
- •Operation
- •Instruction cycle times
- •Examples
- •Format 12: load address
- •Operation
- •Instruction cycle times
- •Examples
- •Format 13: add offset to Stack Pointer
- •Operation
- •Instruction cycle times
- •Examples
- •Format 14: push/pop registers
- •Operation
- •Instruction cycle times
- •Examples
- •Format 15: multiple load/store
- •Operation
- •Instruction cycle times
- •Examples
- •Format 16: conditional branch
- •Operation
- •Instruction cycle times
- •Examples
- •Format 17: software interrupt
- •Operation
- •Instruction cycle times
- •Examples
- •Format 18: unconditional branch
- •Operation
- •Examples
- •Format 19: long branch with link
- •Operation
- •Instruction cycle times
- •Examples
- •Instruction Set Examples
- •Multiplication by a constant using shifts and adds
- •General purpose signed divide
- •Thumb code
- •ARM code
- •Division by a constant
- •Explanation of divide-by-constant ARM code
- •ARM code
- •THUMB code
- •Overview
- •Cycle Types
- •Address Timing
- •Data Transfer Size
- •Instruction Fetch
- •Memory Management
- •Locked Operations
- •Stretching Access Times
- •The ARM Data Bus
- •The External Data Bus
- •The unidirectional data bus
- •The bidirectional data bus
- •Example system: The ARM7TDMI Testchip
- •Overview
- •Interface Signals
- •Coprocessor present/absent
- •Busy-waiting
- •Pipeline following
- •Data transfer cycles
- •Register Transfer Cycle
- •Privileged Instructions
- •Idempotency
- •Undefined Instructions
- •Debug Interface
- •Overview
- •Debug Systems
- •Debug Interface Signals
- •Entry into debug state
- •Entry into debug state on breakpoint
- •Entry into debug state on watchpoint
- •Entry into debug state on debug-request
- •Action of ARM7TDMI in debug state
- •Scan Chains and JTAG Interface
- •Scan limitations
- •Scan chain 0
- •Scan chain 1
- •Scan Chain 2
- •The JTAG state machine
- •Reset
- •Pullup Resistors
- •Instruction Register
- •Public Instructions
- •EXTEST (0000)
- •SCAN_N (0010)
- •INTEST (1100)
- •IDCODE (1110)
- •BYPASS (1111)
- •CLAMP (0101)
- •HIGHZ (0111)
- •CLAMPZ (1001)
- •SAMPLE/PRELOAD (0011)
- •RESTART (0100)
- •Test Data Registers
- •Bypass register
- •ARM7TDMI device identification (ID) code register
- •Operating mode:
- •Instruction register
- •Scan chain select register
- •Scan chains 0,1 and 2
- •Scan chain 0 and 1
- •Scan chain 0
- •Scan chain 1
- •Scan chain 3
- •ARM7TDMI Core Clocks
- •Clock switch during debug
- •Clock switch during test
- •Determining the Core and System State
- •Determining the core’s state
- •Determining system state
- •Exit from debug state
- •The PC’s Behaviour During Debug
- •Breakpoint
- •Watchpoints
- •Watchpoint with another exception
- •Debug request
- •System speed access
- •Summary of return address calculations
- •Priorities / Exceptions
- •Breakpoint with prefetch abort
- •Interrupts
- •Data aborts
- •Scan Interface Timing
- •Debug Timing
- •Overview
- •The Watchpoint Registers
- •Programming and reading watchpoint registers
- •Using the mask registers
- •The control registers
- •Programming Breakpoints
- •Hardware breakpoints:
- •Software breakpoints:
- •Hardware breakpoints
- •Software breakpoints
- •Setting the breakpoint
- •Clearing the breakpoint
- •Programming Watchpoints
- •The Debug Control Register
- •Debug Status Register
- •Coupling Breakpoints and Watchpoints
- •Example
- •CHAINOUT signal
- •RANGEOUT signal
- •Example
- •Disabling ICEBreaker
- •ICEBreaker Timing
- •Programming Restriction
- •Debug Communications Channel
- •Debug comms channel registers
- •Communications via the comms channel
- •Introduction
- •Branch and Branch with Link
- •THUMB Branch with Link
- •Branch and Exchange (BX)
- •Data Operations
- •Multiply and Multiply Accumulate
- •Load Register
- •Store Register
- •Load Multiple Registers
- •Store Multiple Registers
- •Data Swap
- •Software Interrupt and Exception Entry
- •Coprocessor Data Operation
- •Coprocessor Data Transfer (from memory to coprocessor)
- •Coprocessor Data Transfer (from coprocessor to memory)
- •Coprocessor Register Transfer (Load from coprocessor)
- •Coprocessor Register Transfer (Store to coprocessor)
- •Undefined Instructions and Coprocessor Absent
- •Unexecuted Instructions
- •Instruction Speed Summary
- •Timing Diagrams
Register Transfer Cycle
The coprocessor register transfer cycle is the one case when ARM7TDMI requires the data bus without requiring the memory to be active. The memory system is informed that the bus is required by ARM7TDMI taking both nMREQ and SEQ HIGH. When the bus is free, DBE should be taken HIGH to allow ARM7TDMI or the coprocessor to drive the bus, and an MCLK cycle times the transfer.
Privileged Instructions
The coprocessor may restrict certain instructions for use in privileged modes only. To do this, the coprocessor will have to track the nTRANS output.
As an example of the use of this facility, consider the case of a floating point coprocessor (FPU) in a multi-tasking system. The operating system could save all the floating point registers on every task switch, but this is inefficient in a typical system where only one or two tasks will use floating point operations. Instead, there could be a privileged instruction which turns the FPU on or off. When a task switch happens, the operating system can turn the FPU off without saving its registers. If the new task attempts an FPU operation, the FPU will appear to be absent, causing an undefined instruction trap. The operating system will then realise that the new task requires the FPU, so it will reenable it and save FPU registers. The task can then use the FPU as normal. If, however, the new task never attempts an FPU operation (as will be the case for most tasks), the state saving overhead will have been avoided.
Coprocessor
Idempotency
A consequence of the implementation of the coprocessor interface, with the interruptible busy-wait state, is that all instructions may be interrupted at any point up to the time when the coprocessor goes not-busy. If so interrupted, the instruction will normally be restarted from the beginning after the interrupt has been processed. It is therefore essential that any action taken by the coprocessor before it goes not-busy must be idempotent, ie must be repeatable with identical results.
For example, consider a FIX operation in a floating point coprocessor which returns the integer result to an ARM7TDMI register. The coprocessor must stay busy while it performs the floating point to fixed point conversion, as ARM7TDMI will expect to receive the integer value on the cycle immediately following that where it goes not-busy. The coprocessor must therefore preserve the original floating point value and not corrupt it during the conversion, because it will be required again if an interrupt arises during the busy period.
The coprocessor data operation class of instruction is not generally subject to idempotency considerations, as the processing activity can take place after the coprocessor goes not-busy. There is no need for ARM7TDMI to be held up until the result is generated, because the result is confined to stay within the coprocessor.
Undefined Instructions
Undefined instructions are treated by ARM7TDMI as coprocessor instructions. All coprocessors must be absent (ie CPA and CPB must be HIGH) when an undefined instruction is presented. ARM7TDMI will then take the undefined instruction trap. Note that the coprocessor need only look at bit 27 of the instruction to differentiate undefined instructions (which all have 0 in bit 27) from coprocessor instructions (which all have 1 in bit 27)
Note that when in THUMB state, coprocessor instructions are not supported but undefined instructions are. Thus, all coprocessors must monitor the state of the TBIT output from ARM7TDMI. When ARM7TDMI is in THUMB state, coprocessors must appear absent (ie they must drive CPA and CPB HIGH) and the instructions seen on the data bus must be ignored. In this way, coprocessors will not erroneously execute THUMB instructions, and all undefined instructions will be handled correctly.
137
138 Coprocessor
This chapter describes the ARM7TDMI advanced debug interface.
Overview
The ARM7TDMI debug interface is based on IEEE Std. 1149.1- 1990, “Standard Test Access Port and Boundary-Scan Architecture”. Please refer to this standard for an explanation of the terms used in this chapter and for a description of the TAP controller states.
ARM7TDMI contains hardware extensions for advanced debugging features. These are intended to ease the user’s development of application software, operating systems, and the hardware itself.
The debug extensions allow the core to be stopped either on a given instruction fetch (breakpoint) or data access (watchpoint), or asynchronously by a debug-request. When this happens, ARM7TDMI is said to be in debug state. At this point, the core’s internal state and the system’s external state may be examined. Once examination is complete, the core and system state may be restored and program execution resumed.
ARM7TDMI is forced into debug state either by a request on one of the external debug interface signals, or by an internal functional unit known as ICEBreaker. Once in debug state, the core isolates itself from the memory system. The core can then be examined while all other system activity continues as normal.
ARM7TDMI’s internal state is examined via a JTAG-style serial interface, which allows instructions to be serially inserted into the core’s pipeline without using the external data bus. Thus, when in debug state, a store-multiple (STM) could be inserted into the instruction pipeline and this would dump the contents of ARM7TDMI’s registers. This data can be serially shifted out without affecting the rest of the system.
Debug Interface |
139 |
Debug Systems
The ARM7TDMI forms one component of a debug system that interfaces from the high-level debugging performed by the user to the lowlevel interface supported by ARM7TDMI. Such a system typically has three parts:
1.The Debug Host
This is a computer, for example a PC, running a software debugger such as ARMSD. The debug host allows the user to issue high level commands such as “set breakpoint at location XX”, or “examine the contents of memory from 0x0 to 0x100”.
2.The Protocol Converter
The Debug Host will be connected to the ARM7TDMI development system via an interface (an RS232, for example). The messages broadcast over this connection must be converted to the interface signals of the ARM7TDMI, and this function is performed by the protocol converter.
3.ARM7TDMI
ARM7TDMI, with hardware extensions to ease debugging, is the lowest level of the system. The debug extensions allow the user to stall the core from program execution, examine its internal state and the state of the memory system, and then resume program execution.
Figure 75. Typical Debug System
Debug |
Host computer running ARMSD |
Host |
|
|
|
Protocol
Converter
Debug Development System
Target Containing
ARM7TDMI
The anatomy of ARM7TDMI is shown in Figure 77. The major blocks are:
ARM7TDMI This is the CPU core, with hardware support for debug.
ICEBreaker This is a set of registers and comparators used to generate debug exceptions (eg breakpoints). This unit is described in ICEBreaker Module on page 163.
TAP controller This controls the action of the scan chains via a JTAG serial interface.
The Debug Host and the Protocol Converter are system dependent. The rest of this chapter describes the ARM7TDMI’s hardware debug extensions.
140 |
Debug |
|
|
||
|