US20020069402A1 - Scheduling control within a system having mixed hardware and software based instruction execution - Google Patents
Scheduling control within a system having mixed hardware and software based instruction execution Download PDFInfo
- Publication number
- US20020069402A1 US20020069402A1 US09/731,060 US73106000A US2002069402A1 US 20020069402 A1 US20020069402 A1 US 20020069402A1 US 73106000 A US73106000 A US 73106000A US 2002069402 A1 US2002069402 A1 US 2002069402A1
- Authority
- US
- United States
- Prior art keywords
- instruction
- execution unit
- empty
- state
- tos
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
- 238000012545 processing Methods 0.000 claims abstract description 29
- 230000001960 triggered effect Effects 0.000 claims abstract description 5
- 238000000034 method Methods 0.000 claims description 6
- 238000013519 translation Methods 0.000 abstract description 29
- 238000013459 approach Methods 0.000 abstract description 8
- 230000007246 mechanism Effects 0.000 abstract description 4
- 238000013507 mapping Methods 0.000 description 22
- 230000009471 action Effects 0.000 description 11
- 230000007704 transition Effects 0.000 description 9
- 230000008859 change Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 230000001419 dependent effect Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 230000003111 delayed effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000002250 progressing effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 210000003813 thumb Anatomy 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/30098—Register arrangements
- G06F9/30101—Special purpose registers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/30145—Instruction analysis, e.g. decoding, instruction word fields
- G06F9/30149—Instruction analysis, e.g. decoding, instruction word fields of variable length instructions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/3017—Runtime instruction translation, e.g. macros
- G06F9/30174—Runtime instruction translation, e.g. macros for non-native instruction set, e.g. Javabyte, legacy code
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/38—Concurrent instruction execution, e.g. pipeline, look ahead
- G06F9/3802—Instruction prefetching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/38—Concurrent instruction execution, e.g. pipeline, look ahead
- G06F9/3802—Instruction prefetching
- G06F9/3814—Implementation provisions of instruction buffers, e.g. prefetch buffer; banks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/38—Concurrent instruction execution, e.g. pipeline, look ahead
- G06F9/3836—Instruction issuing, e.g. dynamic instruction scheduling or out of order instruction execution
- G06F9/3851—Instruction issuing, e.g. dynamic instruction scheduling or out of order instruction execution from multiple instruction streams, e.g. multistreaming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/38—Concurrent instruction execution, e.g. pipeline, look ahead
- G06F9/3861—Recovery, e.g. branch miss-prediction, exception handling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45504—Abstract machines for programme code execution, e.g. Java virtual machine [JVM], interpreters, emulators
Definitions
- This invention relates to the field of data processing systems. More particular, this invention relates to data processing systems having both a hardware based instruction execution unit and a software based instruction execution unit and in which it is desired to perform scheduling operations.
- Multitasking operating systems require processing resources to be shared between several different programs that may be simultaneously active and multithreaded computer programs similarly require processing resources to be shared between different active threads. It is known to control processing operations using a counter based approach whereby program instructions being executed are counted and a scheduling operation initiated each time a predetermined program instruction count level is reached.
- An alternative approach is to adopt timer based scheduling in which a scheduling operation is initiated at a regular time interval in a manner similar to servicing an interrupt request.
- a simple timer based scheduling approach may suffer from the disadvantage that scheduling operations may be inappropriately triggered at points part way through the software interpretation of a complex program instruction in a manner that could cause a loss of data integrity should an inappropriate context switch occur.
- Counter based scheduling systems suffer from the disadvantage of the need to provide for the exchange of counter values between hardware executed program instructions and software executed program instructions. This represents a disadvantageous overhead.
- the present invention provides apparatus for processing data operable to execute operations specified in a stream of program instructions, said apparatus comprising:
- said hardware based execution unit includes scheduling support logic operable to generate a scheduling signal for triggering a scheduling operation to be performed between program instructions irrespective of whether a preceding program instruction was executed by said hardware based execution unit or said software based execution unit.
- the invention simplifies the provision of scheduling support by providing a system in which program instructions are sent to the hardware based instruction execution unit and forwarded from there to the software based instruction execution unit if they cannot be deal with by the hardware based instruction execution unit. In this way, by routing all the program instructions through the hardware based instruction execution unit, this unit is able to keep track of the execution of instructions and accordingly generate a scheduling signal for triggering a scheduling operation irrespective of whether the preceding instructions have been executed by hardware or software.
- the scheduling support logic within the hardware based instruction execution unit includes a counter that can count program instructions executed by both the hardware and the software based approaches and generate an appropriate scheduling signal to trigger a scheduling operation when a predetermined count value is reached.
- the count value needed to trigger a scheduling operation may be user programmed to fine-tune the scheduling operation concerned, or in some embodiments provide a debugging tool by combining a debugging operation with a scheduling operation in a manner that could, if desired, support single step debugging at one extreme.
- a timer based approach may be used with the signal generated by a timer being logically combined (“qualified”) with the scheduling signal generated within the hardware based instruction execution unit so as to ensure that scheduling operations are started at safe points between the execution of program instructions.
- the invention is particularly useful in embodiments in which the hardware based instruction execution unit is a hardware instruction translator and the software based instruction execution unit is a software interpreter.
- the use of hardware instruction translation in combination with software based instruction interpretation provides accelerated execution of high level program instructions whilst maintaining comprehensive support for the more complex operations that may be specified within such high level program instructions.
- the instruction translation and interpretation operations could be simple, but often require multiple lower level operations to be performed and in this context the accurate control of points at which scheduling operations may be triggered is particularly significant.
- FIGS. 1 and 2 schematically represent example instruction pipeline arrangements
- FIG. 3 illustrates in more detail a fetch stage arrangement
- FIG. 4 schematically illustrates the reading of variable length non-native instructions from within buffered instruction words within the fetch stage
- FIG. 5 schematically illustrates a data processing system for executing both processor core native instructions and instructions requiring translation
- FIG. 6 schematically illustrates, for a sequence of example instructions and states the contents of the registers used for stack operand storage, the mapping states and the relationship between instructions requiring translation and native instructions;
- FIG. 7 schematically illustrates the execution of a non-native instruction as a sequence of native instructions
- FIG. 8 is a flow diagram illustrating the way in which the instruction translator may operate in a manner that preserves interrupt latency for translated instructions;.
- FIG. 9 schematically illustrates the translation of Java bytecodes into ARM opcodes using hardware and software techniques
- FIG. 10 schematically illustrates the flow of control between a hardware based translator, a software based interpreter and software based scheduling
- FIGS. 11 and 12 illustrate another way of controlling scheduling operations using a timer based approach
- FIG. 13 is a signal diagram illustrating the signals controlling the operation of the circuit of FIG. 12.
- FIG. 1 shows a first example instruction pipeline 30 of a type suitable for use in an ARM processor based system.
- the instruction pipeline 30 includes a fetch stage 32 , a native instruction (ARM/Thumb instructions) decode stage 34 , an execute stage 36 , a memory access stage 38 and a write back stage 40 .
- the execute stage 36 , the memory access stage 38 and the write back stage 40 are substantially conventional.
- Downstream of the fetch stage 32 , and upstream of the native instruction decode stage 34 there is provided an instruction translator stage 42 .
- the instruction translator stage 42 is a finite state machine that translates Java bytecode instructions of a variable length into native ARM instructions.
- the instruction translator stage 42 is capable of multi-step operation whereby a single Java bytecode instruction may generate a sequence of ARM instructions that are fed along the remainder of the instruction pipeline 30 to perform the operation specified by the Java bytecode instruction.
- Simple Java bytecode instructions may required only a single ARM instruction to perform their operation, whereas more complicated Java bytecode instructions, or in circumstances where the surrounding system state so dictates, several ARM instructions may be needed to provide the operation specified by the Java bytecode instruction.
- This multi-step operation takes place downstream of the fetch stage 32 and accordingly power is not expended upon fetching multiple translated ARM instructions or Java bytecodes from a memory system.
- the Java bytecode instructions are stored within the memory system in a conventional manner such that additional constraints are not provided upon the memory system in order to support the Java bytecode translation operation.
- the instruction translator stage 42 is provided with a bypass path.
- the instruction pipeline 30 may bypass the instruction translator stage 42 and operate in an essentially unaltered manner to provide decoding of native instructions.
- the instruction translator stage 42 is illustrated as generating translator output signals that fully represent corresponding ARM instructions and are passed via a multiplexer to the native instruction decoder 34 .
- the instruction translator 42 also generates some extra control signals that may be passed to the native instruction decoder 34 .
- Bit space constraints within the native instruction encoding may impose limitations upon the range of operands that may be specified by native instructions. These limitations are not necessarily shared by the non-native instructions. Extra control signals are provided to pass additional instruction specifying signals derived from the non-native instructions that would not be possible to specify within native instructions stored within memory.
- a native instruction may only provide a relatively low number of bits for use as an immediate operand field within a native instruction, whereas the non-native instruction may allow an extended range and this can be exploited by using the extra control signals to pass the extended portion of the immediate operand to the native instruction decoder 34 outside of the translated native instruction that is also passed to the native instruction decoder 34 .
- FIG. 2 illustrates a further instruction pipeline 44 .
- the system is provided with two native instruction decoders 46 , 48 as well as a non-native instruction decoder 50 .
- the non-native instruction decoder 50 is constrained in the operations it can specify by the execute stage 52 , the memory stage 54 and the write back stage 56 that are provided to support the native instructions. Accordingly, the non-native instruction decoder 50 must effectively translate the non-native instructions into native operations (which may be a single native operation or a sequence of native operations) and then supply appropriate control signals to the execute stage 52 to carry out these one or more native operations.
- an instruction fetched by the fetch stage 58 is selectively supplied to one of the instruction decoders 46 , 48 or 50 in dependence upon the particular processing mode using the illustrated demultiplexer.
- the maximum instruction length of a Java bytecode instruction is three bytes. Accordingly, three multiplexers are provided that enable any three neighbouring bytes within either side of the word buffer 62 to be selected and supplied to the instruction translator 64 .
- the word buffer 62 and the instruction translator 64 are also provided with a bypass path 66 for use when native instructions are being fetched and decoded.
- each instruction word is fetched from memory once and stored within the word buffer 62 .
- a single instruction word may have multiple Java bytecodes read from it as the instruction translator 64 performs the translation of Java bytecodes into ARM instructions.
- Variable length translated sequences of native instructions may be generated without requiring multiple memory system reads and without consuming memory resource or imposing other constraints upon the memory system as the instruction translation operations are confined within the instruction pipeline.
- a program counter value is associated with each Java bytecode currently being translated. This program counter value is passed along the stages of the pipeline such that each stage is able, if necessary, to use the information regarding the particular Java bytecode it is processing.
- the program counter value for a Java bytecode that translates into a sequence of a plurality of ARM instruction operations is not incremented until the final ARM instruction operation within that sequence starts to be executed. Keeping the program counter value in a manner that continues to directly point to the instruction within the memory that is being executed advantageously simplifies other aspects of the system, such as debugging and branch target calculation.
- FIG. 4 schematically illustrates the reading of variable length Java bytecode instructions from the instruction buffer 62 .
- a Java bytecode instruction having a length of one is read and decoded.
- the next stage is a Java bytecode instruction that is three bytes in length and spans between two adjacent instruction words that have been fetched from the memory. Both of these instruction words are present within the instruction buffer 62 and so instruction decoding and processing is not delayed by this spanning of a variable length instruction between instruction words fetched.
- the refill of the earlier fetched of the instruction words may commence as subsequent processing will continue with decoding of Java bytecodes from the following instruction word which is already present.
- the final stage illustrated in FIG. 4 illustrates a second three bytecode instruction being read. This again spans between instruction words. If the preceding instruction word has not yet completed its refill, then reading of the instruction may be delayed by a pipeline stall until the appropriate instruction word has been stored into the instruction buffer 62 . In some embodiments the timings may be such that the pipeline never stalls due to this type of behaviour. It will be appreciated that the particular example is a relatively infrequent occurrence as most Java bytecodes are shorter than the examples illustrated and accordingly two successive decodes that both span between instruction words is relatively uncommon. A valid signal may be associated with each of the instruction words within the instruction buffer 62 in a manner that is able to signal whether or not the instruction word has appropriately been refilled before a Java bytecode has been read from it.
- FIG. 5 shows a data processing system 102 including a processor core 104 and a register bank 106 .
- An instruction translator 108 is provided within the instruction path to translate Java Virtual Machine instructions to native ARM instructions (or control signals corresponding thereto) that may then be supplied to the processor core 104 .
- the instruction translator 108 may be bypassed when native ARM instructions are being fetched from the addressable memory.
- the addressable memory may be a memory system such as a cache memory with further off-chip RAM memory. Providing the instruction translator 108 downstream of the memory system, and particularly the cache memory, allows efficient use to be made of the storage capacity of the memory system since dense instructions that require translation may be stored within the memory system and only expanded into native instructions immediately prior to being passed to the processor core 104 .
- the register bank 106 in this example contains sixteen general purpose 32-bit registers, of which four are allocated for use in storing stack operands, i.e. the set of registers for storing stack operands is registers R0, R1, R2 and R3.
- the set of registers may be empty, partly filled with stack operands or completely filled with stack operands.
- the particular register that currently holds the top of stack operand may be any of the registers within the set of registers. It will thus be appreciated that the instruction translator may be in any one of seventeen different mapping states corresponding to one state when all of the registers are empty and four groups of four states each corresponding to a respective different number of stack operands being held within the set of registers and with a different register holding the top of stack operand. Table 1 illustrates the seventeen different states of the state mapping for the instruction translator 108 .
- mapping states can very considerably depending upon the particular implementation and Table 1 is only given as an example of one particular implementation.
- the first three bits of the state value indicate the number of non-empty registers within the set of registers.
- the final two bits of the state value indicate the register number of the register holding the top of stack operand.
- the state value may be readily used to control the operation of a hardware translator or a software translator to take account of the currently occupancy of the set of registers and the current position of the top of stack operand.
- a stream of Java bytecodes J 1 , J 2 , J 3 is fed to the instruction translator 108 from the addressable memory system.
- the instruction translator 108 then outputs a stream of ARM instructions (or equivalent control signals, possibly extended) dependent upon the input Java bytecodes and the instantaneous mapping state of the instruction translator 8 , as well as other variables.
- the example illustrated shows Java bytecode J 1 being mapped to ARM instructions A 1 1 and A 1 2 .
- Java bytecode J 2 maps to ARM instructions A 2 1 , A 2 2 and A 2 3 .
- Java bytecode J 3 maps to ARM instruction A 3 1 .
- Each of the Java bytecodes may require one or more stack operands as inputs and may produce one or more stack operands as an output.
- the processor core 104 in this example is an ARM processor core having a load/store architecture whereby only data values held within registers may be manipulated
- the instruction translator 108 is arranged to generate ARM instructions that, as necessary, fetch any required stack operands into the set of registers before they are manipulated or store to addressable memory any currently held stack operands within the set of registers to make room for result stack operands that may be generated.
- each Java bytecode may be considered as having an associated “require full” value indicating the number of stack operands that must be present within the set of registers prior to its execution together with a “require empty” value indicating the number of empty registers within the set of registers that must be available prior to execution of the ARM instructions representing the Java opcode.
- Table 2 illustrates the relationship between initial mapping state values, require full values, final state values and associated ARM instructions.
- the initial state values and the final state values correspond to the mapping states illustrated in Table 1.
- the instruction translator 108 determines a require full value associated with the particular Java bytecode (opcode) it is translating.
- the instruction translator ( 108 ) in dependence upon the initial mapping state that it has, determines whether or not more stack operands need to be loaded into the set of registers prior to executing the Java bytecode.
- Table 1 shows the initial states together with tests applied to the require full value of the Java bytecode that are together applied to determine whether a stack operand needs to be loaded into the set of registers using an associated ARM instruction (an LDR instruction) as well as the final mapping state that will be adopted after such a stack cache load operation.
- an LDR instruction an LDR instruction
- Table 1 shows the initial states together with tests applied to the require full value of the Java bytecode that are together applied to determine whether a stack operand needs to be loaded into the set of registers using an associated ARM instruction (an LDR instruction) as well as the final mapping state that will be adopted after such a stack cache load operation.
- an LDR instruction an LDR instruction
- Table 3 in a similar manner illustrates the relationship between initial state, require empty value, final state and an associated ARM instruction for emptying a register within the set of registers to move between the initial state and the final state if the require empty value of a particular Java bytecode indicates that it is necessary given the initial state before the Java bytecode is executed.
- the particular register values stored off to the addressable memory with an STR instruction will vary depending upon which of the registers is the current top of stack operand.
- the require full and require empty conditions are mutually exclusive, that is to say only one of the require full or require empty conditions can be true at any given time for a particular Java bytecode which the instruction translator is attempting to translate.
- the instruction templates used by the instruction translator 108 together with the instructions it is chosen to support with the hardware instruction translator 108 are selected such that this mutually exclusive requirement may be met. If this requirement were not in place, then the situation could arise in which a particular Java bytecode required a number of input stack operands to be present within the set of registers that would not allow sufficient empty registers to be available after execution of the instruction representing the Java bytecode to allow the results of the execution to be held within the registers as required.
- a given Java bytecode will have an overall nett stack action representing the balance between the number of stack operands consumed and the number of stack operands generated upon execution of that Java bytecode. Since the number of stack operands consumed is a requirement prior to execution and the number of stack operands generated is a requirement after execution, the require full and require empty values associated with each Java bytecode must be satisfied prior to execution of that bytecode even if the nett overall action would in itself be met.
- Table 4 illustrates the relationship between an initial state, an overall stack action, a final state and a change in register use and relative position of the top of stack operand (TOS).
- the relationships between the different states, conditions, and nett actions may be used to define a hardware state machine (in the form of a finite state machine) for controlling this aspect of the operation of the instruction translator 108 .
- these relationships could be modelled by software or a combination of hardware and software.
- FIG. 1 An example execution sequence is illustrated below of a single Java bytecode executed by a hardware translation unit 108 in accordance with the techniques described above.
- the execution sequence is shown in terms of an initial state progressing through a sequence of states dependent upon the instructions being executed, generating a sequence of ARM instructions as a result of the actions being performed on each state transition, the whole having the effect of translating a Java bytecode to a sequence of ARM instructions.
- State Transition 00000 >0 00100
- FIG. 6 illustrates in a different way the execution of a number of further Java bytecode instructions.
- the top portion of FIG. 6 illustrates the sequence of ARM instructions and changes of mapping states and register contents that occur upon execution of an iadd Java bytecode instruction.
- the initial mapping state is 00000 corresponding to all of the registers within the set of registers being empty.
- the first two ARM instructions generated serve to POP two stack operands into the registers storing stack operands with the top of stack “TOS” register being R0.
- the third ARM instruction actually performs the add operation and writes the result into register R3 (which now becomes the top of stack operand) whilst consuming the stack operand that was previously held within register R 1 , thus producing an overall stack action of ⁇ 1.
- Processing then proceeds to execution of two Java bytecodes each representing a long load of two stack operands.
- the require empty condition of 2 for the first Java bytecode is immediately met and accordingly two ARM LDR instructions may be issued and executed.
- the mapping state after execution of the first long load Java bytecode is 01101. In this state the set of registers contains only a single empty register.
- the next Java bytecode long load instruction has a require empty value of 2 that is not met and accordingly the first action required is a PUSH of a stack operand to the addressable memory using an ARM STR instruction. This frees up a register within the set of registers for use by a new stack operand which may then be loaded as part of the two following LDR instructions.
- the instruction translation may be achieved by hardware, software, or a combination of the two.
- DCD 0 . . . DCD do_iconst_0 ; Opcode 0x03 . . . DCD do_lload_0 ; Opcode 0x1e . . . DCD do_iastore ; Opcode 0x4f DCD do_lastore ; Opcode 0x50 . . . DCD do_iadd ; Opcode 0x60 .
- DCD State_00100_do_iastore Opcode 0x4f DCD State_00100_do_lastore ; Opcode 0x50 . . . DCD State_00100_do_iadd ; Opcode 0x60 . . . DCD State_00100_do_ineg ; Opcode 0x74 . . . DCD State_00100_do_land ; Opcode 0x7f . . .
- FIG. 7 illustrates a Java bytecode instruction “laload” which has the function of reading two words of data from within a data array specified by two words of data starting at the top of stack position. The two words read from the data array then replace the two words that specified their position and to form the topmost stack entries.
- the Java bytecode instruction is specified as having a require empty value of 2, i.e. two of the registers within the register bank dedicated to stack operand storage must be emptied prior to executing the ARM instructions emulating the “laload” instruction. If there are not two empty registers when this Java bytecode is encountered, then store operations (STRs) may be performed to PUSH stack operands currently held within the registers out to memory so as to make space for the temporary storage necessary and meet the require empty value for the instruction.
- STRs store operations
- the instruction also has a require full value of 2 as the position of the data is specified by an array location and an index within that array as two separate stack operands.
- the drawing illustrates the first state as already meeting the require full and require empty conditions and having a mapping state of “01001”.
- the “laload” instruction is broken down into three ARM instructions. The first of these loads the array reference into a spare working register outside of the set of registers acting as a register cache of stack operands. The second instruction then uses this array reference in conjunction with an index value within the array to access a first array word that is written into one of the empty registers dedicated to stack operand storage.
- mapping state of the system is not changed and the top of stack pointer remains where it started with the registers specified as empty still being so specified.
- the final instruction within the sequence of ARM instructions loads the second array word into the set of registers for storing stack operands.
- the mapping state changes to “01011” which places the new top of stack pointer at the second array word and indicates that the input variables of the array reference and index value are now empty registers, i.e. marking the registers as empty is equivalent to removing the values they held from the stack.
- mapping state swap has nevertheless occurred.
- the change of mapping state performed upon execution of the final operation is hardwired into the instruction translator as a function of the Java bytecode being translated and is indicated by the “swap” parameter shown as a characteristic of the “laload” instruction.
- FIG. 8 is a flow diagram schematically illustrating the above technique.
- a Java bytecode is fetched from memory.
- the require full and require empty values for that Java bytecode are examined. If either of the require empty or require full conditions are not met, then respective PUSH and POP operations of stack operands (possibly multiple stack operands) may be performed with steps 14 and 16 . It is will be noted that this particular system does not allow the require empty and require full conditions to be simultaneously unmet. Multiple passes through steps 14 and 16 may be required until the condition of step 12 is met.
- step 18 the first ARM instruction specified within the translation template for the Java bytecode concerned is selected.
- step 20 a check is made as to whether or not the selected ARM instruction is the final instruction to be executed in the emulation of the Java bytecode fetched at step 10 . If the ARM instruction being executed is the final instruction, then step 21 serves to update the program counter value to point to the next Java bytecode in the sequence of instructions to be executed.
- the ARM instruction is the final instruction, then it will complete its execution irrespective of whether or not an interrupt now occurs and accordingly it is safe to update the program counter value to the next Java bytecode and restart execution from that point as the state of the system will have reached that matching normal, uninterrupted, full execution of the Java bytecode. If the test at step 20 indicates that the final bytecode has not been reached, then updating of the program counter value is bypassed.
- Step 22 executes the current ARM instruction.
- a test is made as to whether or not there are any more ARM instructions that require executing as part of the template. If there are more ARM instructions, then the next of these is selected at step 26 and processing is returned to step 20 . If there are no more instructions, then processing proceeds to step 28 at which any mapping change/swap specified for the Java bytecode concerned is performed in order to reflect the desired top of stack location and full/empty status of the various registers holding stack operands.
- FIG. 8 also schematically illustrates the points at which an interrupt if asserted is serviced and then processing restarted after an interrupt.
- An interrupt starts to be serviced after the execution of an ARM instruction currently in progress at step 22 with whatever is the current program counter value being stored as a return point with the bytecode sequence. If the current ARM instruction executing is the final instruction within the template sequence, then step 21 will have just updated the program counter value and accordingly this will point to the next Java bytecode (or ARM instruction should an instruction set switch have just been initiated).
- the program counter value will still be the same as that indicated at the start of the execution of the Java bytecode concerned and accordingly when a return is made, the whole Java bytecode will be re-executed.
- FIG. 9 illustrates a Java bytecode translation unit 68 that receives a stream of Java bytecodes and outputs a translated stream of ARM instructions (or corresponding control signals) to control the action of a processor core.
- the Java bytecode translator 68 translates simple Java bytecodes using instruction templates into ARM instructions or sequences of ARM instructions.
- a counter value within scheduling control logic 70 is decremented.
- this counter value reaches 0, then the Java bytecode translation unit 68 issues an ARM instruction branching to scheduling code that manages scheduling between threads or tasks as appropriate.
- Java bytecode translation unit 68 Whilst simple Java bytecodes are handled by the Java bytecode translation unit 68 itself providing high speed hardware based execution of these bytecodes, bytecodes requiring more complex processing operations are sent to a software interpreter provided in the form of a collection of interpretation routines (examples of a selection of such routines are given earlier in this description). More specifically, the Java bytecode translation unit 68 can determined that the bytecode it has received is not one which is supported by hardware translation and accordingly a branch can be made to an address dependent upon that Java bytecode where a software routine for interpreting that bytecode is found or referenced. This mechanism can also be employed when the scheduling logic 70 indicates that a scheduling operation is needed to yield a branch to the scheduling code.
- FIG. 10 illustrates the operation of the embodiment of FIG. 9 in more detail and the split of tasks between hardware and software.
- All Java bytecodes are received by the Java bytecode translation unit 68 and cause the counter to be decremented at step 72 .
- a check is made as to whether or not the counter value has reached 0. If the counter value has reached 0 (counting down from either a predetermined value hardwired into the system or a value that may be user controlled/programmed), then a branch is made to scheduling code at step 76 . Once the scheduling code has completed at step 76 , control is returned to the hardware and processing proceeds to step 72 , where the next Java bytecode is fetched and the counter again decremented. Since the counter reached 0, then it will now roll round to a new, non-zero value. Alternatively, a new value may be forced into the counter as part of the exiting of the scheduling process at step 76 .
- step 78 fetches the Java bytecode.
- step 80 a determination is made as to whether the fetched bytecode is a simple bytecode that may be executed by hardware translation at step 82 or requires more complex processing and accordingly should be passed out for software interpretation at step 84 . If processing is passed out to software interpretation, then once this has completed control is returned to the hardware where step 72 decrements the counter again to take account of the fetching of the next Java bytecode.
- FIG. 11 illustrates an alternative control arrangement.
- an instruction signal (scheduling signal) is deasserted.
- a fetched Java bytecode is examined to see if it is a simple bytecode for which hardware translation is supported. If hardware translation is not supported, then control is passed out to the interpreting software at step 90 which then executes a ARM instruction routine to interpret the Java bytecode. If the bytecode is a simple one for which hardware translation is supported, then processing proceeds to step 92 at which one or more ARM instructions are issued in sequence by the Java bytecode translation unit 68 acting as a form of multi-cycle finite state machine.
- step 94 the instruction signal is asserted for a short period prior to being deasserted at step 86 .
- the assertion of the instruction signal indicates to external circuitry that an appropriate safe point has been reached at which a timer based scheduling interrupt could take place without risking a loss of data integrity due to the partial execution of an interpreted or translated instruction.
- FIG. 12 illustrates example circuitry that may be used to respond to the instruction signal generated in FIG. 11.
- a timer 96 periodically generates a timer signal after expiry of a given time period. This timer signal is stored within a latch 98 until it is cleared by a clear timer interrupt signal.
- the output of the latch 98 is logically combined by an AND gate 100 with the instruction signal asserted at step 94 .
- an interrupt is generated as the output of the AND gate 100 and is used to trigger an interrupt that performs scheduling operations using the interrupt processing mechanisms provided within the system for standard interrupt processing. Once the interrupt signal has been generated, this in turn triggers the production of a clear timer interrupt signal that clears the latch 98 until the next timer output pulse occurs.
- FIG. 13 is a signal diagram illustrating the operation of the circuit of FIG. 12.
- the processor core clock signals occur at a regular frequency.
- the timer 96 generates timer signals at predetermined periods to indicate that, when safe, a scheduling operation should be initiated.
- the timer signals are latched. Instruction signals are generated at times spaced apart by intervals that depend upon how quickly a particular Java bytecode was executed.
- a simple Java bytecode may execute in a single processor core clock cycle, or more typically two or three, whereas a complex Java bytecode providing a high level management type function may take several hundred processor clock cycles before its execution is completed by the software interpreter.
- a pending asserted latched timer signal is not acted upon to trigger a scheduling operation until the instruction signal issues indicating that it is safe for the scheduling operation to commence.
- the simultaneous occurrence of a latched timer signal and the instruction signal triggers the generation of an interrupt signal followed immediately thereafter by a clear signal that clears the latch 98 .
Abstract
A processing system provides both hardware instruction translation (68) and software instruction interpretation (84) mechanisms for supporting high level program instructions. All of the program instructions are supplied to the hardware translation unit (68) which forwards those instructions it does not itself support to the software interpretation mechanism (84). By routing all program instructions through the hardware translation unit (68), the hardware translation unit (86) is able to monitor when it is appropriate and safe to trigger a scheduling operation for controlling multitasking or multithreaded operations. The scheduling operations may be triggered based upon a count of executed program instructions or by using a timer based scheduling approach with the timer signal being qualified by a signal indicating an appropriate point within the cycle of execution of program instructions.
Description
- 1. Field of the Invention
- This invention relates to the field of data processing systems. More particular, this invention relates to data processing systems having both a hardware based instruction execution unit and a software based instruction execution unit and in which it is desired to perform scheduling operations.
- 2. Description of the Prior Art
- Within modern data processing systems the ability to reliably perform scheduling between tasks or threads is an important capability. Multitasking operating systems require processing resources to be shared between several different programs that may be simultaneously active and multithreaded computer programs similarly require processing resources to be shared between different active threads. It is known to control processing operations using a counter based approach whereby program instructions being executed are counted and a scheduling operation initiated each time a predetermined program instruction count level is reached. An alternative approach is to adopt timer based scheduling in which a scheduling operation is initiated at a regular time interval in a manner similar to servicing an interrupt request.
- In order to provide support for execution of higher level computer program languages, it is known to use mixed hardware based execution units and software based execution units. Simple instructions within a hardware based execution unit may be executed under control of that hardware based execution unit, whereas more complex program instructions trigger the execution of a software routine, typically written in a low level directly executable program language, which interprets the complex instructions. Whilst such systems are able to provide comprehensive and yet relatively high speed execution of high level program instructions. they pose difficulties in also supporting scheduling.
- A simple timer based scheduling approach may suffer from the disadvantage that scheduling operations may be inappropriately triggered at points part way through the software interpretation of a complex program instruction in a manner that could cause a loss of data integrity should an inappropriate context switch occur. Counter based scheduling systems suffer from the disadvantage of the need to provide for the exchange of counter values between hardware executed program instructions and software executed program instructions. This represents a disadvantageous overhead.
- Examples of known systems for translation between instruction sets and other background information may be found in the following: U.S. Pat. No. 5,805,895; U.S. Pat. No. 3,955,180; U.S. Pat. No. 5,970,242; U.S. Pat. No. 5,619,665; U.S. Pat. No. 5,826,089; U.S. Pat. No. 5,925,123; U.S. Pat. No. 5,875,336; U.S. Pat. No. 5,937,193; U.S. Pat. No. 5,953,520; U.S. Pat. No. 6,021,469; U.S. Pat. No. 5,568,646; U.S. Pat. No. 5,758,115; U.S. Pat. No. 5,367,685; IBM Technical Disclosure Bulletin, March 1988, pp308-309, “System/370 Emulator Assist Processor For a Reduced Instruction Set Computer”; IBM Technical Disclosure Bulletin, July 1986, pp548-549, “Full Function Series/1 Instruction Set Emulator”; IBM Technical Disclosure Bulletin, March 1994, pp605-606, “Real-Time CISC Architecture HW Emulator On A RISC Processor”; IBM Technical Disclosure Bulletin, March 1998, p272, “Performance Improvement Using An EMULATION Control Block”; IBM Technical Disclosure Bulletin, January 1995, pp537-540, “Fast Instruction Decode For Code Emulation on Reduced Instruction Set Computer/Cycles Systems”; IBM Technical Disclosure Bulletin, February 1993, pp231-234, “High Performance Dual Architecture Processor”; IBM Technical Disclosure Bulletin, August 1989, pp40-43, “System/370 I/O Channel Program Channel Command Word Prefetch”; IBM Technical Disclosure Bulletin, June 1985, pp305-306, “Fully Microcode-Controlled Emulation Architecture”; IBM Technical Disclosure Bulletin, March 1972, pp3074-3076, “Op Code and Status Handling For Emulation”; IBM Technical Disclosure Bulletin, August 1982, pp954-956, “On-Chip Microcoding of a Microprocessor With Most Frequently Used Instructions of Large System and Primitives Suitable for Coding Remaining Instructions”; IBM Technical Disclosure Bulletin, April 1983, pp5576-5577, “Emulation Instruction”; the book ARM System Architecture by S Furber; the book Computer Architecture: A Quantitative Approach by Hennessy and Patterson; and the book The Java Virtual Machine Specification by Tim Lindholm and Frank Yellin 1st and 2nd Editions.
- The ability to reliably and efficiently support scheduling within mixed hardware and software based instruction execution systems is strongly desirable.
- Viewed from one aspect the present invention provides apparatus for processing data operable to execute operations specified in a stream of program instructions, said apparatus comprising:
- (i) a hardware based instruction execution unit operable to execute program instructions; and
- (ii) a software based instruction execution unit operable to execute program instructions; wherein
- (ii) program instructions to be executed are sent to said hardware based execution unit for execution;
- (iii) program instructions received by said hardware based execution unit for which execution is not supported by said hardware based execution unit are forwarded to said software based execution unit for execution with control being returned to said hardware based execution unit for a next program instruction to be executed; and
- (iv) said hardware based execution unit includes scheduling support logic operable to generate a scheduling signal for triggering a scheduling operation to be performed between program instructions irrespective of whether a preceding program instruction was executed by said hardware based execution unit or said software based execution unit.
- The invention simplifies the provision of scheduling support by providing a system in which program instructions are sent to the hardware based instruction execution unit and forwarded from there to the software based instruction execution unit if they cannot be deal with by the hardware based instruction execution unit. In this way, by routing all the program instructions through the hardware based instruction execution unit, this unit is able to keep track of the execution of instructions and accordingly generate a scheduling signal for triggering a scheduling operation irrespective of whether the preceding instructions have been executed by hardware or software.
- In one preferred embodiment the scheduling support logic within the hardware based instruction execution unit includes a counter that can count program instructions executed by both the hardware and the software based approaches and generate an appropriate scheduling signal to trigger a scheduling operation when a predetermined count value is reached.
- Preferably the count value needed to trigger a scheduling operation may be user programmed to fine-tune the scheduling operation concerned, or in some embodiments provide a debugging tool by combining a debugging operation with a scheduling operation in a manner that could, if desired, support single step debugging at one extreme.
- In an alternative preferred embodiment a timer based approach may be used with the signal generated by a timer being logically combined (“qualified”) with the scheduling signal generated within the hardware based instruction execution unit so as to ensure that scheduling operations are started at safe points between the execution of program instructions.
- The invention is particularly useful in embodiments in which the hardware based instruction execution unit is a hardware instruction translator and the software based instruction execution unit is a software interpreter. The use of hardware instruction translation in combination with software based instruction interpretation provides accelerated execution of high level program instructions whilst maintaining comprehensive support for the more complex operations that may be specified within such high level program instructions.
- The instruction translation and interpretation operations could be simple, but often require multiple lower level operations to be performed and in this context the accurate control of points at which scheduling operations may be triggered is particularly significant.
- Whilst the program instruction language within which scheduling is being supported could take many different forms, the invention is particularly well suited to embodiments in which the program is a Java Virtual Machine instruction program involving a mix of Java bytecode hardware translation and Java bytecode software interpretation;
- The above, and other objects, features and advantages of this invention will be apparent from the following detailed description of illustrative embodiments which is to be read in connection with the accompanying drawings.
- FIGS. 1 and 2 schematically represent example instruction pipeline arrangements;
- FIG. 3 illustrates in more detail a fetch stage arrangement;
- FIG. 4 schematically illustrates the reading of variable length non-native instructions from within buffered instruction words within the fetch stage;
- FIG. 5 schematically illustrates a data processing system for executing both processor core native instructions and instructions requiring translation;
- FIG. 6 schematically illustrates, for a sequence of example instructions and states the contents of the registers used for stack operand storage, the mapping states and the relationship between instructions requiring translation and native instructions;
- FIG. 7 schematically illustrates the execution of a non-native instruction as a sequence of native instructions;
- FIG. 8 is a flow diagram illustrating the way in which the instruction translator may operate in a manner that preserves interrupt latency for translated instructions;.
- FIG. 9 schematically illustrates the translation of Java bytecodes into ARM opcodes using hardware and software techniques;
- FIG. 10 schematically illustrates the flow of control between a hardware based translator, a software based interpreter and software based scheduling;
- FIGS. 11 and 12 illustrate another way of controlling scheduling operations using a timer based approach; and
- FIG. 13 is a signal diagram illustrating the signals controlling the operation of the circuit of FIG. 12.
- FIG. 1 shows a first
example instruction pipeline 30 of a type suitable for use in an ARM processor based system. Theinstruction pipeline 30 includes a fetchstage 32, a native instruction (ARM/Thumb instructions)decode stage 34, an executestage 36, amemory access stage 38 and a write backstage 40. The executestage 36, thememory access stage 38 and the write backstage 40 are substantially conventional. Downstream of the fetchstage 32, and upstream of the nativeinstruction decode stage 34, there is provided aninstruction translator stage 42. Theinstruction translator stage 42 is a finite state machine that translates Java bytecode instructions of a variable length into native ARM instructions. Theinstruction translator stage 42 is capable of multi-step operation whereby a single Java bytecode instruction may generate a sequence of ARM instructions that are fed along the remainder of theinstruction pipeline 30 to perform the operation specified by the Java bytecode instruction. Simple Java bytecode instructions may required only a single ARM instruction to perform their operation, whereas more complicated Java bytecode instructions, or in circumstances where the surrounding system state so dictates, several ARM instructions may be needed to provide the operation specified by the Java bytecode instruction. This multi-step operation takes place downstream of the fetchstage 32 and accordingly power is not expended upon fetching multiple translated ARM instructions or Java bytecodes from a memory system. The Java bytecode instructions are stored within the memory system in a conventional manner such that additional constraints are not provided upon the memory system in order to support the Java bytecode translation operation. - As illustrated, the
instruction translator stage 42 is provided with a bypass path. When not operating in an instruction translating mode, theinstruction pipeline 30 may bypass theinstruction translator stage 42 and operate in an essentially unaltered manner to provide decoding of native instructions. - In the
instruction pipeline 30, theinstruction translator stage 42 is illustrated as generating translator output signals that fully represent corresponding ARM instructions and are passed via a multiplexer to thenative instruction decoder 34. Theinstruction translator 42 also generates some extra control signals that may be passed to thenative instruction decoder 34. Bit space constraints within the native instruction encoding may impose limitations upon the range of operands that may be specified by native instructions. These limitations are not necessarily shared by the non-native instructions. Extra control signals are provided to pass additional instruction specifying signals derived from the non-native instructions that would not be possible to specify within native instructions stored within memory. As an example, a native instruction may only provide a relatively low number of bits for use as an immediate operand field within a native instruction, whereas the non-native instruction may allow an extended range and this can be exploited by using the extra control signals to pass the extended portion of the immediate operand to thenative instruction decoder 34 outside of the translated native instruction that is also passed to thenative instruction decoder 34. - FIG. 2 illustrates a
further instruction pipeline 44. In this example, the system is provided with twonative instruction decoders non-native instruction decoder 50. Thenon-native instruction decoder 50 is constrained in the operations it can specify by the executestage 52, thememory stage 54 and the write backstage 56 that are provided to support the native instructions. Accordingly, thenon-native instruction decoder 50 must effectively translate the non-native instructions into native operations (which may be a single native operation or a sequence of native operations) and then supply appropriate control signals to the executestage 52 to carry out these one or more native operations. It will be appreciated that in this example the non-native instruction decoder does not produce signals that form a native instruction, but rather provides control signals that specify native instruction (or extended native instruction) operations. The control signals generated may not match the control signals generated by thenative instruction decoders - In operation, an instruction fetched by the fetch
stage 58 is selectively supplied to one of theinstruction decoders - FIG. 3 schematically illustrates the fetch stage of an instruction pipeline in more detail. Fetching
logic 60 fetches fixed length instruction words from a memory system and supplies these to aninstruction word buffer 62. Theinstruction word buffer 62 is a swing buffer having two sides such that it may store both a current instruction word and a next instruction word. Whenever the current instruction word has been fully decoded and decoding has progressed onto the next instruction word, then the fetchlogic 60 serves to replace the previous current instruction word with the next instruction word to be fetched from memory, i.e. each side of the swing buffer will increment by two in an interleaved fashion the instruction words that they successively store. - In the example illustrated, the maximum instruction length of a Java bytecode instruction is three bytes. Accordingly, three multiplexers are provided that enable any three neighbouring bytes within either side of the
word buffer 62 to be selected and supplied to theinstruction translator 64. Theword buffer 62 and theinstruction translator 64 are also provided with abypass path 66 for use when native instructions are being fetched and decoded. - It will be seen that each instruction word is fetched from memory once and stored within the
word buffer 62. A single instruction word may have multiple Java bytecodes read from it as theinstruction translator 64 performs the translation of Java bytecodes into ARM instructions. Variable length translated sequences of native instructions may be generated without requiring multiple memory system reads and without consuming memory resource or imposing other constraints upon the memory system as the instruction translation operations are confined within the instruction pipeline. - A program counter value is associated with each Java bytecode currently being translated. This program counter value is passed along the stages of the pipeline such that each stage is able, if necessary, to use the information regarding the particular Java bytecode it is processing. The program counter value for a Java bytecode that translates into a sequence of a plurality of ARM instruction operations is not incremented until the final ARM instruction operation within that sequence starts to be executed. Keeping the program counter value in a manner that continues to directly point to the instruction within the memory that is being executed advantageously simplifies other aspects of the system, such as debugging and branch target calculation.
- FIG. 4 schematically illustrates the reading of variable length Java bytecode instructions from the
instruction buffer 62. At the first stage a Java bytecode instruction having a length of one is read and decoded. The next stage is a Java bytecode instruction that is three bytes in length and spans between two adjacent instruction words that have been fetched from the memory. Both of these instruction words are present within theinstruction buffer 62 and so instruction decoding and processing is not delayed by this spanning of a variable length instruction between instruction words fetched. Once the three Java bytecodes have been read from theinstruction buffer 62, the refill of the earlier fetched of the instruction words may commence as subsequent processing will continue with decoding of Java bytecodes from the following instruction word which is already present. - The final stage illustrated in FIG. 4 illustrates a second three bytecode instruction being read. This again spans between instruction words. If the preceding instruction word has not yet completed its refill, then reading of the instruction may be delayed by a pipeline stall until the appropriate instruction word has been stored into the
instruction buffer 62. In some embodiments the timings may be such that the pipeline never stalls due to this type of behaviour. It will be appreciated that the particular example is a relatively infrequent occurrence as most Java bytecodes are shorter than the examples illustrated and accordingly two successive decodes that both span between instruction words is relatively uncommon. A valid signal may be associated with each of the instruction words within theinstruction buffer 62 in a manner that is able to signal whether or not the instruction word has appropriately been refilled before a Java bytecode has been read from it. - FIG. 5 shows a data processing system102 including a
processor core 104 and aregister bank 106. Aninstruction translator 108 is provided within the instruction path to translate Java Virtual Machine instructions to native ARM instructions (or control signals corresponding thereto) that may then be supplied to theprocessor core 104. Theinstruction translator 108 may be bypassed when native ARM instructions are being fetched from the addressable memory. The addressable memory may be a memory system such as a cache memory with further off-chip RAM memory. Providing theinstruction translator 108 downstream of the memory system, and particularly the cache memory, allows efficient use to be made of the storage capacity of the memory system since dense instructions that require translation may be stored within the memory system and only expanded into native instructions immediately prior to being passed to theprocessor core 104. - The
register bank 106 in this example contains sixteen general purpose 32-bit registers, of which four are allocated for use in storing stack operands, i.e. the set of registers for storing stack operands is registers R0, R1, R2 and R3. - The set of registers may be empty, partly filled with stack operands or completely filled with stack operands. The particular register that currently holds the top of stack operand may be any of the registers within the set of registers. It will thus be appreciated that the instruction translator may be in any one of seventeen different mapping states corresponding to one state when all of the registers are empty and four groups of four states each corresponding to a respective different number of stack operands being held within the set of registers and with a different register holding the top of stack operand. Table 1 illustrates the seventeen different states of the state mapping for the
instruction translator 108. It will be appreciated that with a different number of registers allocated for stack operand storage, or as a result of constraints that a particular processor core may have in the way it can manipulate data values held within registers, the mapping states can very considerably depending upon the particular implementation and Table 1 is only given as an example of one particular implementation.TABLE 1 STATE 00000 R0 = EMPTY R1 = EMPTY R2 = EMPTY R3 = EMPTY STATE 00100 STATE 01000 STATE 01100 STATE 10000 R0 = TOS R0 = TOS R0 = TOS R0 = TOS R1 = EMPTY R1 = EMPTY R1 = EMPTY R1 = TOS-3 R2 = EMPTY R2 = EMPTY R2 = TOS-2 R2 = TOS-2 R3 = EMPTY R3 = TOS-1 R3 = TOS-1 R3 = TOS-1 STATE 00101 STATE 01001 STATE 01101 STATE 10001 R0 = EMPTY R0 = TOS-1 R0 = TOS-1 R0 = TOS-1 R1 = TOS R1 = TOS R1 = TOS R1 = TOS R2 = EMPTY R2 = EMPTY R2 = EMPTY R2 = TOS-3 R3 = EMPTY R3 = EMPTY R3 = TOS-2 R3 = TOS-2 STATE 00110 STATE 01010 STATE 01110 STATE 10010 R0 = EMPTY R0 = EMPTY R0 = TOS-2 R0 = TOS-2 R1 = EMPTY R1 = TOS-1 R1 = TOS-1 R1 = TOS-1 R2 = TOS R2 = TOS R2 = TOS R2 = TOS R3 = EMPTY R3 = EMPTY R3 = EMPTY R3 = TOS-3 STATE 00111 STATE 01011 STATE 01111 STATE 10011 R0 = EMPTY R0 = EMPTY R0 = EMPTY R0 = TOS-3 R1 = EMPTY R1 = EMPTY R1 = TOS-2 R1 = TOS-2 R2 = EMPTY R2 = TOS-1 R2 = TOS 1 R2 = TOS-1 R3 = TOS R3 = TOS R3 = TOS R3 = TOS - Within Table 1 it may be observed that the first three bits of the state value indicate the number of non-empty registers within the set of registers. The final two bits of the state value indicate the register number of the register holding the top of stack operand. In this way, the state value may be readily used to control the operation of a hardware translator or a software translator to take account of the currently occupancy of the set of registers and the current position of the top of stack operand.
- As illustrated in FIG. 5 a stream of Java bytecodes J1, J2, J3 is fed to the
instruction translator 108 from the addressable memory system. Theinstruction translator 108 then outputs a stream of ARM instructions (or equivalent control signals, possibly extended) dependent upon the input Java bytecodes and the instantaneous mapping state of the instruction translator 8, as well as other variables. The example illustrated shows Java bytecode J1 being mapped to ARM instructions A1 1 andA 1 2. Java bytecode J2 maps to ARM instructions A2 1, A2 2 andA 2 3. Finally, Java bytecode J3 maps toARM instruction A 3 1. Each of the Java bytecodes may require one or more stack operands as inputs and may produce one or more stack operands as an output. Given that theprocessor core 104 in this example is an ARM processor core having a load/store architecture whereby only data values held within registers may be manipulated, theinstruction translator 108 is arranged to generate ARM instructions that, as necessary, fetch any required stack operands into the set of registers before they are manipulated or store to addressable memory any currently held stack operands within the set of registers to make room for result stack operands that may be generated. It will be appreciated that each Java bytecode may be considered as having an associated “require full” value indicating the number of stack operands that must be present within the set of registers prior to its execution together with a “require empty” value indicating the number of empty registers within the set of registers that must be available prior to execution of the ARM instructions representing the Java opcode. - Table 2 illustrates the relationship between initial mapping state values, require full values, final state values and associated ARM instructions. The initial state values and the final state values correspond to the mapping states illustrated in Table 1. The
instruction translator 108 determines a require full value associated with the particular Java bytecode (opcode) it is translating. The instruction translator (108), in dependence upon the initial mapping state that it has, determines whether or not more stack operands need to be loaded into the set of registers prior to executing the Java bytecode. Table 1 shows the initial states together with tests applied to the require full value of the Java bytecode that are together applied to determine whether a stack operand needs to be loaded into the set of registers using an associated ARM instruction (an LDR instruction) as well as the final mapping state that will be adopted after such a stack cache load operation. In practice, if more than one stack operand needs to be loaded into the set of registers prior to execution of the Java bytecode, then multiple mapping state transitions will occur, each with an associated ARM instruction loading a stack operand into one of the registers of the set of registers. In different embodiments it may be possible to load multiple stack operands in a single state transition and accordingly make mapping state changes beyond those illustrated in Table 2.TABLE 2 INITIAL REQUIRE FINAL STATE FULL STATE ACTIONS 00000 >0 00100 LDR R0, [Rstack, #-4]! 00100 >1 01000 LDR R3, [Rstack, #-4]! 01001 >2 01101 LDR R3, [Rstack, #-4]! 01110 >3 10010 LDR R3, [Rstack, #-4]! 01111 >3 10011 LDR R0, [Rstack, #-4]! 01100 >3 10000 LDR R1, [Rstack, #-4]! 01101 >3 10001 LDR R2, [Rstack, #-4]! 01010 >2 01110 LDR R0, [Rstack, #-4]! 01011 >2 01111 LDR R1, [Rstack, #-4]! 01000 >2 01100 LDR R2, [Rstack, #-4]! 00110 >1 01010 LDR R1, [Rstack, #-4]! 00111 >1 01011 LDR R2, [Rstack, #-4]! 00101 >1 01001 LDR R0, [Rstack, #-4]! - As will be seen from Table 2, a new stack operand loaded into the set of registers storing stack operands will form a new top of stack operand and this will be loaded into a particular one of the registers within the set of registers depending upon the initial state.
- Table 3 in a similar manner illustrates the relationship between initial state, require empty value, final state and an associated ARM instruction for emptying a register within the set of registers to move between the initial state and the final state if the require empty value of a particular Java bytecode indicates that it is necessary given the initial state before the Java bytecode is executed. The particular register values stored off to the addressable memory with an STR instruction will vary depending upon which of the registers is the current top of stack operand.
TABLE 3 INITIAL REQUIRE FINAL STATE FULL STATE ACTIONS 00100 >3 00000 STR R0, [Rstack], #4 01001 >2 00101 STR R0, [Rstack], #4 01110 >1 01010 STR R0, [Rstack], #4 10011 >0 01111 STR R0, [Rstack], #4 10000 >0 01100 STR R1, [Rstack], #4 10001 >0 01101 STR R2, [Rstack], #4 10010 >0 01110 STR R3, [Rstack], #4 01111 >1 01011 STR R1, [Rstack], #4 01100 >1 01000 STR R2, [Rstack], #4 01101 >1 01001 STR R3, [Rstack], #4 01010 >2 00110 STR R1, [Rstack], #4 01011 >2 00111 STR R2, [Rstack], #4 01000 >2 00100 STR R3, [Rstack], #4 00110 >3 00000 STR R2, [Ratack], #4 00111 >3 00000 STR R3, [Rstack], #4 00101 >3 00000 STR R1, [Rstack], #4 - It will be appreciated that in the above described example system the require full and require empty conditions are mutually exclusive, that is to say only one of the require full or require empty conditions can be true at any given time for a particular Java bytecode which the instruction translator is attempting to translate. The instruction templates used by the
instruction translator 108 together with the instructions it is chosen to support with thehardware instruction translator 108 are selected such that this mutually exclusive requirement may be met. If this requirement were not in place, then the situation could arise in which a particular Java bytecode required a number of input stack operands to be present within the set of registers that would not allow sufficient empty registers to be available after execution of the instruction representing the Java bytecode to allow the results of the execution to be held within the registers as required. - It will be appreciated that a given Java bytecode will have an overall nett stack action representing the balance between the number of stack operands consumed and the number of stack operands generated upon execution of that Java bytecode. Since the number of stack operands consumed is a requirement prior to execution and the number of stack operands generated is a requirement after execution, the require full and require empty values associated with each Java bytecode must be satisfied prior to execution of that bytecode even if the nett overall action would in itself be met. Table 4 illustrates the relationship between an initial state, an overall stack action, a final state and a change in register use and relative position of the top of stack operand (TOS). It may be that one or more of the state transitions illustrated in Table 2 or Table 3 need to be carried out prior to carrying out the state transitions illustrated in Table4 in order to establish the preconditions for a given Java bytecode depending on the require full and require empty values of the Java bytecode.
TABLE 4 INITIAL STACK FINAL STATE ACTION STATE ACTIONS 00000 +1 00101 R1 <− TOS 00000 +2 01010 R1 <− TOS−1, R2 <− TOS 00000 +3 01111 R1 <− TOS−2, R2 <− TOS−1, R3 <− TOS 00000 +4 10000 R0 <− TOS, R1 <− TOS−3, R2 <− TOS−2, R3 <− TOS−1 00100 +1 01001 R1 <− TOS 00100 +2 01110 R1 <− TOS−1, R2 <− TOS 00100 +3 10011 R1 <− TOS−2, R2 <− TOS−1, R3 <− TOS 00100 −1 00000 R0 < EMPTY 01001 +1 01110 R2 <− TOS 01001 +2 10011 R2 <− TOS−1, R3 <− TOS 01001 −1 00100 R1 <− EMPTY 01001 −2 00000 R0 <− EMPTY, R1 <− EMPTY 01110 +1 10011 R3 <− TOS 01110 −1 01001 R2 <− EMPTY 01110 −2 00100 R1 <− EMPTY, R2 <− EMPTY 01110 −3 00000 R0 <− EMPTY, R1 <− EMPTY, R2 <− EMPTY 10011 −1 01110 R3 <− EMPTY 10011 −2 01001 R2 <− EMPTY, R3 <− EMPTY 10011 −3 00100 R1 <− EMPTY, R2 <− EMPTY, R3 <− EMPTY 10011 −4 00000 R0 <− EMPTY, R1 <− EMPTY, R2 <− EMPTY, R3 <− EMPTY 10000 −1 01111 R0 <− EMPTY 10000 −2 01010 R0 <− EMPTY, R3 <− EMPTY 10000 −3 00101 R0 <− EMPTY, R2 <− EMPTY, R3 <− EMPTY 10000 −4 00000 R0 <− EMPTY, R1 <− EMPTY, R2 <− EMPTY, R3 <− EMPTY 10001 −1 01100 R1 <− EMPTY 10001 −2 01011 R0 <− EMPTY, R1 <− EMPTY 10001 −3 00110 R0 <− EMPTY, R1 <− EMPTY, R3 <− EMPTY 10001 −4 00000 R0 <− EMPTY, R1 <− EMPTY, R2 <− EMPTY, R3 <− EMPTY 10010 −1 01101 R2 <− EMPTY 10010 −2 01000 R1 <− EMPTY, R2 <− EMPTY 10010 −3 00111 R0 <− EMPTY, R1 <− EMPTY, R2 <− EMPTY 10010 −4 00000 R0 <− EMPTY, R1 <− EMPTY, R2 <− EMPTY, R3 <− EMPTY 01111 +1 10000 R0 <− TOS 01111 −1 01010 R3 <− EMPTY 01111 −2 00101 R2 <− EMPTY, R3 <− EMPTY 01111 −3 00000 R1 <− EMPTY, R2 <− EMPTY, R3 <− EMPTY 01100 +1 10001 R1 <− TOS 01100 −1 01011 R0 <− EMPTY 01100 −2 00110 R0 <− EMPTY, R3 <− EMPTY 01100 −3 00000 R0 <− EMPTY, R2 <− EMPTY, R3 <− EMPTY 01101 +1 10010 R2 <− TOS 01101 −1 01000 R1 <− EMPTY 01101 −2 00111 R0 <− EMPTY, R1 <− EMPTY 01101 −3 00000 R0 <− EMPTY, R1 <− EMPTY, R3 <− EMPTY 01010 +1 01111 R3 <− TOS 01010 +2 10000 R3 <− TOS−1, R0 <− TOS 01010 −1 00101 R2 <− EMPTY 01010 −2 00000 R1 <− EMPTY, R2 <− EMPTY 01011 +1 01100 R0 <− TOS 01011 +2 10001 R0 <− TOS−1, R1 <− TOS 01011 −1 00110 R3 <− EMPTY 01011 −2 00000 R2 <− EMPTY, R3 <− EMPTY 01000 +1 01101 R1 <− TOS 01000 +2 10010 R1 <− TOS−1, R2 <− TOS 01000 −1 00111 R0 <− EMPTY 01000 −2 00000 R0 <− EMPTY, R3 <− EMPTY 00110 +1 01011 R3 <− TOS 00110 +2 01100 R0 <− TOS, R3 <− TOS−1 00110 +3 10001 R1 <− TOS, R0 <− TOS−1, R3 <− TOS−2 00110 −1 00000 R2 <− EMPTY 00111 +1 01000 R0 <− TOS 00111 +2 01101 R0 <− TOS−1, R1 <− TOS 00111 +3 10010 R0 <− TOS−2, R1 <− TOS−1, R2 <− TOS 00111 −1 00000 R3 <− EMPTY 00101 +1 01010 R2 <− TOS 00101 +2 01111 R2 <− TOS−1, R3 <− TOS 00101 +3 10000 R2 <− TOS−2, R3 <− TOS−1, R1 <− TOS 00101 −1 00000 R1 <− EMPTY - It will be appreciated that the relationships between states and conditions illustrated in Table 2, Table 3 and Table 4 could be combined into a single state transition table or state diagram, but they have been shown separately above to aid clarity.
- The relationships between the different states, conditions, and nett actions may be used to define a hardware state machine (in the form of a finite state machine) for controlling this aspect of the operation of the
instruction translator 108. Alternatively, these relationships could be modelled by software or a combination of hardware and software. - There follows below an example of a subset of the possible Java bytecodes that indicates for each Java bytecode of the subset the associated require full, require empty and stack action values for that bytecode which may be used in conjunction with Tables 2, 3 and 4.
--- iconst_0 Operation: Push inc constant Stack: . . . => . . ., 0 Require-Full = 0 Require-Empty = 1 Stack-Action = +1 --- iadd Operation: Add mt Stack: . . ., value1, value2 => . . ., result Require-Full = 2 Require-Empty = 0 Stack-Action = −1 --- lload_O Operation: Load long from local variable Stack: . . . => . . ., value.word1, value.word2 Require-Full = 0 Require-Empty = 2 Stack-Action = +2 --- lastore Operation: Store into long array Stack: . . ., arrayref, index, value.word1, value.word2 => . . . Require-Full = 4 Require-Empty = 0 Stack-Action = −4 --- land Operation Boolean AND long Stack: . . ., value1.word1, value1.word2, value2.word1, value2.word2 => . . ., result.word1, result.word2 Require-Full = 4 Require-Empty = 0 Stack-Action = −2 --- iastore Operation: Store into int array Stack: . . ., arrayref, index, value => Require-Full = 3 Require-Empty = 0 Stack-Action = −3 --- ineg Operation: Negate int Stack: . . ., value => . . ., result Require-Full = 1 Require-Empty = 0 Stack-Action = 0 - There also follows example instruction templates for each of the Java bytecode instructions set out above. The instructions shown are the ARM instructions which implement the required behaviour of each of the Java bytecodes. The register field “TOS−3”, “TOS−2”, “TOS−1”, “TOS”, “TOS+1” and “TOS+2” may be replaced with the appropriate register specifier as read from Table 1 depending upon the mapping state currently adopted. The denotation “TOS+n” indicates the Nth register above the register currently storing the top of stack operand starting from the register storing the top of stack operand and counting upwards in register value until reaching the end of the set of registers at which point a wrap is made to the first register within the set of registers.
iconst_0 MOV tos+1, #0 lload_0 LDR tos+2, [vars, #4] LDR tos+1, [vars, #0] iastore LDR Rtmp2, [tos-2, #4] LDR Rtmp1, [tos-2, #0] CMP tos-1, Rtmp2, LSR # 5BLXCS Rexc STR tos, [Rtmp1, tos-1, LSL #2] lastore LDR Rtmp2, [tos-3, #4] LDR Rtmp1, [tos-3, #0] CMP tos-2, Rtmp2, LSR # 5BLXCS Rexc STR tos-1, [Rtmp1, tos-2, LSL #3]! STR tos, [Rtmp1, #4] iadd ADD tos-1, tos-1, tos ineg RSB tos, tos, #0 land AND tos-2, tos-2, tos AND tos-3, tos-3, tos-1 - An example execution sequence is illustrated below of a single Java bytecode executed by a
hardware translation unit 108 in accordance with the techniques described above. The execution sequence is shown in terms of an initial state progressing through a sequence of states dependent upon the instructions being executed, generating a sequence of ARM instructions as a result of the actions being performed on each state transition, the whole having the effect of translating a Java bytecode to a sequence of ARM instructions.Initial state: 00000 Instruction: iadd (Require−Full=2, Require−Empty=0, Stack−Action=−1) Condition: Require−Full>0 State Transition: 00000 >0 00100 ARM Instruction(s): LDR R0, [Rstack, #−4] ! Next state: 00100 Instruction: iadd (Require−Full=2, Require−Empty=0, Stack−Action=−1) Condition: Require−Full>1 State Transition: 00100 >1 01000 ARM Instructions(s): LDR R3, [Rstack, #−4] ! Next state: 01000 Instruction: iadd (Require−Fu11=2, Require−Empty=0, Stack−Action=−1) Condition: Stack−Action=−1 State Transition: 01000 −1 00111 Instruction template: ADD tos−1, tos−1, tos ARM Instructions(s) (after substitution): ADD R3, R3, R0 Next state: 00111 - FIG. 6 illustrates in a different way the execution of a number of further Java bytecode instructions. The top portion of FIG. 6 illustrates the sequence of ARM instructions and changes of mapping states and register contents that occur upon execution of an iadd Java bytecode instruction. The initial mapping state is 00000 corresponding to all of the registers within the set of registers being empty. The first two ARM instructions generated serve to POP two stack operands into the registers storing stack operands with the top of stack “TOS” register being R0. The third ARM instruction actually performs the add operation and writes the result into register R3 (which now becomes the top of stack operand) whilst consuming the stack operand that was previously held within register R1, thus producing an overall stack action of −1.
- Processing then proceeds to execution of two Java bytecodes each representing a long load of two stack operands. The require empty condition of 2 for the first Java bytecode is immediately met and accordingly two ARM LDR instructions may be issued and executed. The mapping state after execution of the first long load Java bytecode is 01101. In this state the set of registers contains only a single empty register. The next Java bytecode long load instruction has a require empty value of 2 that is not met and accordingly the first action required is a PUSH of a stack operand to the addressable memory using an ARM STR instruction. This frees up a register within the set of registers for use by a new stack operand which may then be loaded as part of the two following LDR instructions. As previously mentioned, the instruction translation may be achieved by hardware, software, or a combination of the two. Given below is a subsection of an example software interpreter generated in accordance with the above described techniques.
Interpret: LDRB Rtmp, [Rjpc, #1] ! LDR pc, [pc, Rtmp, lsl #2] DCD 0 . . . DCD do_iconst_0 ; Opcode 0x03 . . . DCD do_lload_0 ; Opcode 0x1e . . . DCD do_iastore ; Opcode 0x4f DCD do_lastore ; Opcode 0x50 . . . DCD do_iadd ; Opcode 0x60 . . . DCD do_ineg ; Opcode 0x74 . . . DCD do_land ; Opcode 0x7f . . . do_iconst_0 MOV RO, #0 STR RO, [Rstack], #4 B Interpret do_lload_0 LDMIA Rvars, {R0, R1} STMIA Rstack!, {R0, R1} B Interpret do_iastore LDMDB Rstack!, {R0, R1, R2} LDR Rtmp2, [r0, #4] LDR Rtmp1, [r0, #0] CMP R1, Rtmp2, LSR # 5BCS ArrayBoundException STR R2, [Rtmp1, R1, LSL #2] B Interpret do_lastore LDMDB Rstack!, {R0, R1, R2, R3} LDR Rtmp2, [r0, #4] LDR Rtmp1, [r0, #0] CMP R1, Rtmp2, LSR # 5BCS ArrayBoundException STR R2, [Rtmp1, R1, LSL #3] ! STR R3, [Rtmp1, #4] B Interpret do_iadd LDMDB Rstack!, {r0, r1} ADD r0, r0, r1 STR r0, [Rstack], #4 B Interpret do_ineg LDR r0, [Rstack, #−4] ! RSB tos, tos, #0 STR r0, [Rstack], #4 B Interpret do_land LDMDB Rstack!, {r0, r1, r2, r3} AND r1, r1, r3 AND r0, r0, r2 STMIA Rstack!, {r0, r1} B Interpret State_00000_Interpret LDRB Rtmp, [Rjpc, #1] ′ LDR pc, [pc, Rtmp, lsl #2] DCD 0 . . . DCD State_00000_do_iconst_0 ; Opcode 0x03 . . . DCD State_00000_do_lload_0 ; Opcode 0x1e . . . DCD State_00000_do_iastore ; Opcode 0x4f DCD State_00000_do_lastore ; Opcode 0x50 . . . DCD State_00000_do_iadd ; Opcode 0x60 . . . DCD State_00000_do_ineg ; Opcode 0x74 . . . DCD State_00000_do_land ; Opcode 0x7f . . . State_00000_do_iconst_0 MOV R1, #0 B State_00101_Interpret State_00000_do_lload_0 LDMIA Rvars, {R1, R2} B State_01010_Interpret State_00000_do_iastore LDMDB Rstack!, {R0, R1, R2} LDR Rtmp2, [r0, #4] LDR Rtmp1, [r0, #0] CMP R1, Rtmp2, LSR # 5BCS ArrayBoundException STR R2, [Rtmp1, R1, LSL #2] B State_00000_Interpret State_00000_do_lastore LDMDB Rstack!, {R0, R1, R2, R3} LDR Rtmp2, [r0, #4] LDR Rtmp1, [r0, #0] CMP R1, Rtmp2, LSR # 5BCS ArrayBoundException STR R2, [Rtmp1, R1, LSL #3] ! STR R3, [Rtmp1, #4] B State_00000_Interpret State_00000_do_iadd LDMDB Rstack!, {R1, R2} ADD r1, r1, r2 B State_00101_Interpret State_00000_do_ineg LDR r1, [Rstack, #−4] ! RSB r1, r1, #0 B State_00101_Interpret State_00000_do_land LDR r0, [Rstack, #−4] ! LDMDB Rstack!, {r1, r2, r3} AND r2, r2, r0 AND r1, r1, r3 B State_01010_Interpret State_00100_Interpret LDRB Rtmp, [Rjpc, #1] ! LDR pc, [pc, Rtmp, lsl #2] DCD 0 . . . DCD State_00100_do_iconst_0 ; Opcode 0x03 . . . DCD State_00100_do_lload_0 ; Opcode 0x1e . . . DCD State_00100_do_iastore ; Opcode 0x4f DCD State_00100_do_lastore ; Opcode 0x50 . . . DCD State_00100_do_iadd ; Opcode 0x60 . . . DCD State_00100_do_ineg ; Opcode 0x74 . . . DCD State_00100_do_land ; Opcode 0x7f . . . State_00100_do_iconst_0 MOV R1, #0 B State_01001_Interpret State_00100_do_lload_0 LDMIA Rvars, {r1, R2} B State_01110_Interpret State_00100_do_iastore LDMDB Rstack!, {r2, r3} LDR Rtmp2, [r2, #4] LDR Rtmp1, [r2, #0] CMP R3, Rtmp2, LSR # 5BCS ArrayBoundException STR R0, [Rtmp1, R3, lsl #2] B State_00000_Interpret State_00100_do_lastore LDMDB Rstack!, {r1, r2, r3} LDR Rtmp2, [r1, #4] LDR Rtmp1, [r1, #0] CMP r2, Rtmp2, LSR # 5BCS ArrayBoundException STR r3, [Rtmp1, r2, lsl #3] STR r0, [Rtmp1, #4] B State_00000_Interpret State_00100_do_iadd LDR r3, [Rstack, #−4] ! ADD r3, r3, r0 B State_00111_Interpret State_00100_do_ineg RSB r0, r0, #0 B State_00100_Interpret State_00100_do_land LDMDB Rstack!, {r1, r2, r3} AND r2, r2, r0 AND r1, r1, r3 B State_01010_Interpret State_01000_Interpret LDRB Rtmp, [Rjpc, #1] ! LDR pc, [pc, Rtmp, lsl #2] DCD 0 . . . DCD State_01000_do_iconst_0 ; Opcode 0x03 . . . DCD State_01000_do_lload_0 ; Opcode 0x1e . . . DCD State_01000_do_iastore ; Opcode 0x4f DCD State_01000_do_lastore ; Opcode 0x50 . . . DCD State_01000_do_iadd ; Opcode 0x60 . . . DCD State_01000_do_ineg ; Opcode 0x74 . . . DCD State_01000_do_land ; Opoode 0x7f . . . State_01000_do_iconst_0 MOV R1, #0 B State_01101_Interpret State_01000_do_lload_0 LDMIA Rvars, {r1, r2} B State_10010_Interpret State_01000_do_iastore LDR r1, [Rstack, #−4] ! LDR Rtmp2, [R3, #4] LDR Rtmp1, [R3, #0] CMP r0, Rtmp2, LSR # 5BCS ArrayBoundException STR r1, [Rtmp1, r0, lsl #2] B State_00000_Interpret State_01000_do_lastore LDMDB Rstack!, {r1, r2} LDR Rtmp2, {r3, #4} LDR Rtmp1, {R3, #0} CMP r0, Rtmp2, LSR # 5BCS ArrayBoundException STR r1, [Rtmp1, r0, lsl #3] ! STR r2, [Rtmp1, #4] B State_00000_Interpret State_01000_do_iadd ADD r3, r3, r0 B State_00111_Interpret State_01000_do_ineg RSB r0, r0, #0 B State_01000_Interpret State_01000_do_land LDMDB Rstack!, {r1, r2} AND R0, R0, R2 AND R3, R3, R1 B State_01000_Interpret State_01100_Interpret . . . State_10000_Interpret . . . State_00101_Interpret . . . State_01001_Interpret . . . State_01101_Interpret . . . State_10001_Interpret . . . State_00110_Interpret . . . State_01010_Interpret . . . State_01110_Interpret . . . State_10010_Interpret . . . State_00111_Interpret . . . State_01011_Interpret . . . State_01111_Interpret . . . State_10011_Interpret . . . - FIG. 7 illustrates a Java bytecode instruction “laload” which has the function of reading two words of data from within a data array specified by two words of data starting at the top of stack position. The two words read from the data array then replace the two words that specified their position and to form the topmost stack entries.
- In order that the “laload” instruction has sufficient register space for the temporary storage of the stack operands being fetched from the array without overwriting the input stack operands that specify the array and position within the array of the data, the Java bytecode instruction is specified as having a require empty value of 2, i.e. two of the registers within the register bank dedicated to stack operand storage must be emptied prior to executing the ARM instructions emulating the “laload” instruction. If there are not two empty registers when this Java bytecode is encountered, then store operations (STRs) may be performed to PUSH stack operands currently held within the registers out to memory so as to make space for the temporary storage necessary and meet the require empty value for the instruction.
- The instruction also has a require full value of 2 as the position of the data is specified by an array location and an index within that array as two separate stack operands. The drawing illustrates the first state as already meeting the require full and require empty conditions and having a mapping state of “01001”. The “laload” instruction is broken down into three ARM instructions. The first of these loads the array reference into a spare working register outside of the set of registers acting as a register cache of stack operands. The second instruction then uses this array reference in conjunction with an index value within the array to access a first array word that is written into one of the empty registers dedicated to stack operand storage.
- It is significant to note that after the execution of the first two ARM instructions, the mapping state of the system is not changed and the top of stack pointer remains where it started with the registers specified as empty still being so specified.
- The final instruction within the sequence of ARM instructions loads the second array word into the set of registers for storing stack operands. As this is the final instruction, if an interrupt does occur during it, then it will not be serviced until after the instruction completes and so it is safe to change the input state with this instruction by a change to the mapping state of the registers storing stack operands. In this example, the mapping state changes to “01011” which places the new top of stack pointer at the second array word and indicates that the input variables of the array reference and index value are now empty registers, i.e. marking the registers as empty is equivalent to removing the values they held from the stack.
- It will be noted that whilst the overall stack action of the “laload” instruction has not changed the number of stack operands held within the registers, a mapping state swap has nevertheless occurred. The change of mapping state performed upon execution of the final operation is hardwired into the instruction translator as a function of the Java bytecode being translated and is indicated by the “swap” parameter shown as a characteristic of the “laload” instruction.
- Whilst the example of this drawing is one specific instruction, it will be appreciated that the principles set out may be extended to many different Java bytecode instructions that are emulated as ARM instructions or other types of instruction.
- FIG. 8 is a flow diagram schematically illustrating the above technique. At step10 a Java bytecode is fetched from memory. At step 12 the require full and require empty values for that Java bytecode are examined. If either of the require empty or require full conditions are not met, then respective PUSH and POP operations of stack operands (possibly multiple stack operands) may be performed with
steps steps - At
step 18, the first ARM instruction specified within the translation template for the Java bytecode concerned is selected. Atstep 20, a check is made as to whether or not the selected ARM instruction is the final instruction to be executed in the emulation of the Java bytecode fetched atstep 10. If the ARM instruction being executed is the final instruction, then step 21 serves to update the program counter value to point to the next Java bytecode in the sequence of instructions to be executed. It will be understood that if the ARM instruction is the final instruction, then it will complete its execution irrespective of whether or not an interrupt now occurs and accordingly it is safe to update the program counter value to the next Java bytecode and restart execution from that point as the state of the system will have reached that matching normal, uninterrupted, full execution of the Java bytecode. If the test atstep 20 indicates that the final bytecode has not been reached, then updating of the program counter value is bypassed. -
Step 22 executes the current ARM instruction. At step 24 a test is made as to whether or not there are any more ARM instructions that require executing as part of the template. If there are more ARM instructions, then the next of these is selected atstep 26 and processing is returned to step 20. If there are no more instructions, then processing proceeds to step 28 at which any mapping change/swap specified for the Java bytecode concerned is performed in order to reflect the desired top of stack location and full/empty status of the various registers holding stack operands. - FIG. 8 also schematically illustrates the points at which an interrupt if asserted is serviced and then processing restarted after an interrupt. An interrupt starts to be serviced after the execution of an ARM instruction currently in progress at
step 22 with whatever is the current program counter value being stored as a return point with the bytecode sequence. If the current ARM instruction executing is the final instruction within the template sequence, then step 21 will have just updated the program counter value and accordingly this will point to the next Java bytecode (or ARM instruction should an instruction set switch have just been initiated). If the currently executing ARM instruction is anything other than the final instruction in the sequence, then the program counter value will still be the same as that indicated at the start of the execution of the Java bytecode concerned and accordingly when a return is made, the whole Java bytecode will be re-executed. - FIG. 9 illustrates a Java
bytecode translation unit 68 that receives a stream of Java bytecodes and outputs a translated stream of ARM instructions (or corresponding control signals) to control the action of a processor core. As described previously, theJava bytecode translator 68 translates simple Java bytecodes using instruction templates into ARM instructions or sequences of ARM instructions. When each Java bytecode has been executed, then a counter value within scheduling control logic 70 is decremented. When this counter value reaches 0, then the Javabytecode translation unit 68 issues an ARM instruction branching to scheduling code that manages scheduling between threads or tasks as appropriate. - Whilst simple Java bytecodes are handled by the Java
bytecode translation unit 68 itself providing high speed hardware based execution of these bytecodes, bytecodes requiring more complex processing operations are sent to a software interpreter provided in the form of a collection of interpretation routines (examples of a selection of such routines are given earlier in this description). More specifically, the Javabytecode translation unit 68 can determined that the bytecode it has received is not one which is supported by hardware translation and accordingly a branch can be made to an address dependent upon that Java bytecode where a software routine for interpreting that bytecode is found or referenced. This mechanism can also be employed when the scheduling logic 70 indicates that a scheduling operation is needed to yield a branch to the scheduling code. - FIG. 10 illustrates the operation of the embodiment of FIG. 9 in more detail and the split of tasks between hardware and software. All Java bytecodes are received by the Java
bytecode translation unit 68 and cause the counter to be decremented at step 72. At step 74 a check is made as to whether or not the counter value has reached 0. If the counter value has reached 0 (counting down from either a predetermined value hardwired into the system or a value that may be user controlled/programmed), then a branch is made to scheduling code atstep 76. Once the scheduling code has completed atstep 76, control is returned to the hardware and processing proceeds to step 72, where the next Java bytecode is fetched and the counter again decremented. Since the counter reached 0, then it will now roll round to a new, non-zero value. Alternatively, a new value may be forced into the counter as part of the exiting of the scheduling process atstep 76. - If the test at step74 indicated that the counter did not equal 0, then step 78 fetches the Java bytecode. At step 80 a determination is made as to whether the fetched bytecode is a simple bytecode that may be executed by hardware translation at step 82 or requires more complex processing and accordingly should be passed out for software interpretation at
step 84. If processing is passed out to software interpretation, then once this has completed control is returned to the hardware where step 72 decrements the counter again to take account of the fetching of the next Java bytecode. - FIG. 11 illustrates an alternative control arrangement. At the start of processing at step86 an instruction signal (scheduling signal) is deasserted. At
step 88, a fetched Java bytecode is examined to see if it is a simple bytecode for which hardware translation is supported. If hardware translation is not supported, then control is passed out to the interpreting software atstep 90 which then executes a ARM instruction routine to interpret the Java bytecode. If the bytecode is a simple one for which hardware translation is supported, then processing proceeds to step 92 at which one or more ARM instructions are issued in sequence by the Javabytecode translation unit 68 acting as a form of multi-cycle finite state machine. Once the Java bytecode has been properly executed either atstep 90 or at step 92, then processing proceeds to step 94 at which the instruction signal is asserted for a short period prior to being deasserted at step 86. The assertion of the instruction signal indicates to external circuitry that an appropriate safe point has been reached at which a timer based scheduling interrupt could take place without risking a loss of data integrity due to the partial execution of an interpreted or translated instruction. - FIG. 12 illustrates example circuitry that may be used to respond to the instruction signal generated in FIG. 11. A timer96 periodically generates a timer signal after expiry of a given time period. This timer signal is stored within a
latch 98 until it is cleared by a clear timer interrupt signal. The output of thelatch 98 is logically combined by an ANDgate 100 with the instruction signal asserted atstep 94. When the latch is set and the instruction signal is asserted, then an interrupt is generated as the output of the ANDgate 100 and is used to trigger an interrupt that performs scheduling operations using the interrupt processing mechanisms provided within the system for standard interrupt processing. Once the interrupt signal has been generated, this in turn triggers the production of a clear timer interrupt signal that clears thelatch 98 until the next timer output pulse occurs. - FIG. 13 is a signal diagram illustrating the operation of the circuit of FIG. 12. The processor core clock signals occur at a regular frequency. The timer96 generates timer signals at predetermined periods to indicate that, when safe, a scheduling operation should be initiated. The timer signals are latched. Instruction signals are generated at times spaced apart by intervals that depend upon how quickly a particular Java bytecode was executed. A simple Java bytecode may execute in a single processor core clock cycle, or more typically two or three, whereas a complex Java bytecode providing a high level management type function may take several hundred processor clock cycles before its execution is completed by the software interpreter. In either case, a pending asserted latched timer signal is not acted upon to trigger a scheduling operation until the instruction signal issues indicating that it is safe for the scheduling operation to commence. The simultaneous occurrence of a latched timer signal and the instruction signal triggers the generation of an interrupt signal followed immediately thereafter by a clear signal that clears the
latch 98. - Although illustrative embodiments of the invention have been described in detail herein with reference to the accompanying drawings, it is to be understood that the invention is not limited to those precise embodiments, and that various changes and modifications can be effected therein by one skilled in the art without departing from the scope and spirit of the invention as defined by the appended claims.
Claims (16)
1. Apparatus for processing data operable to execute operations specified in a stream of program instructions, said apparatus comprising:
(i) a hardware based instruction execution unit operable to execute program instructions; and
(ii) a software based instruction execution unit operable to execute program instructions; wherein
(iii) program instructions to be executed are sent to said hardware based execution unit for execution;
(iv) program instructions received by said hardware based execution unit for which execution is not supported by said hardware based execution unit are forwarded to said software based execution unit for execution with control being returned to said hardware based execution unit for a next program instruction to be executed; and
(v) said hardware based execution unit includes scheduling support logic operable to generate a scheduling signal for triggering a scheduling operation to be performed between program instructions irrespective of whether a preceding program instruction was executed by said hardware based execution unit or said software based execution unit.
2. Apparatus as claimed in claim 1 , wherein said scheduling support logic includes a counter with a value that is changed in response to a program instruction sent to said hardware based execution unit.
3. Apparatus as claimed in claim 2 , wherein said counter triggers generation of said scheduling signal when a predetermined count value is reached.
4. Apparatus as claimed in claim 3 , wherein said counter may be programmed to start from a user programmable start value.
5. Apparatus as claimed in claim 3 , wherein said counter counts up to said predetermined value.
6. Apparatus as claimed in claim 3 , wherein said counter counts down to said predetermined value.
7. Apparatus as claimed in claim 1 , wherein a debug operation is triggered by said scheduling signal.
8. Apparatus as claimed in claim 1 , further comprising timer logic operable to generate a timer signal indicative of a time since a last scheduling operation.
9. Apparatus as claimed in claim 8 , wherein said scheduling signal is combined with said timer signal to trigger said scheduling operation.
10. Apparatus as claimed in claim 8 , wherein a scheduling operation is triggered upon generation of said scheduling signal after said timer signal has reached a predetermined value indicating a predetermined period time since a last scheduling operation has expired.
11. Apparatus as claimed in claim 1 , further comprising a processor core operable to execute operations as specified by instructions of a first instruction set.
12. Apparatus as claimed in claim 11 , where said hardware based instruction execution unit includes an instruction translator operable to translate instructions of a second instruction set into translator output signals corresponding to instructions of said first instruction set.
13. Apparatus as claimed in claim 12 , wherein
(i) at least one instruction of said second instruction set specifies a multi-step operation that requires a plurality of operations that may be specified by instructions of said first instruction set in order to be performed by said processor core; and
(ii) said instruction translator is operable to generate a sequence of translator output signals to control said processor core to perform said multi-step operation.
14. Apparatus as claimed in claim 1 , wherein said software based execution unit is a software based interpreter.
15. Apparatus as claimed in claim 1 , wherein said program instructions are Java Virtual Machine instructions.
16. A method of processing data by executing operations specified in a stream of program instructions, said method comprising the steps of:
(i) executing program instructions with a hardware based instruction execution unit; and
(ii) executing program instructions with a software based instruction execution unit; wherein
(iii) program instructions to be executed are sent to said hardware based execution unit for execution;
(iv) program instructions received by said hardware based execution unit for which execution is not supported by said hardware based execution unit are forwarded to said software based execution unit for execution with control being returned to said hardware based execution unit for a next program instruction to be executed; and
(v) said hardware based execution unit generates a scheduling signal for triggering a scheduling operation to be performed between program instructions irrespective of whether a preceding program instruction was executed by said hardware based execution unit or said software based execution unit.
Applications Claiming Priority (10)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
GB0024399.8 | 2000-10-05 | ||
GB0024399A GB2367652B (en) | 2000-10-05 | 2000-10-05 | Scheduling control within a system having mixed hardware and software based instruction execution |
GB0024396.4 | 2000-10-05 | ||
GB0024402A GB2367653B (en) | 2000-10-05 | 2000-10-05 | Restarting translated instructions |
GB0024404A GB2367654B (en) | 2000-10-05 | 2000-10-05 | Storing stack operands in registers |
GB0024402.0 | 2000-10-05 | ||
GB0024404.6 | 2000-10-05 | ||
GB0024396A GB2367651B (en) | 2000-10-05 | 2000-10-05 | Hardware instruction translation within a processor pipeline |
GB0028249A GB2367658B (en) | 2000-10-05 | 2000-11-20 | Intercalling between native and non-native instruction sets |
GB0028249.1 | 2000-11-20 |
Publications (1)
Publication Number | Publication Date |
---|---|
US20020069402A1 true US20020069402A1 (en) | 2002-06-06 |
Family
ID=27515982
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US09/731,060 Abandoned US20020069402A1 (en) | 2000-10-05 | 2000-12-07 | Scheduling control within a system having mixed hardware and software based instruction execution |
US09/887,561 Expired - Lifetime US7134119B2 (en) | 2000-10-05 | 2001-06-25 | Intercalling between native and non-native instruction sets |
Family Applications After (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US09/887,561 Expired - Lifetime US7134119B2 (en) | 2000-10-05 | 2001-06-25 | Intercalling between native and non-native instruction sets |
Country Status (2)
Country | Link |
---|---|
US (2) | US20020069402A1 (en) |
JP (1) | JP4938187B2 (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020099930A1 (en) * | 2000-12-04 | 2002-07-25 | Mitsubishi Denki Kabushiki Kaisha | Data processor having translator and interpreter that execute non-native instructions |
US20040031022A1 (en) * | 2002-08-02 | 2004-02-12 | Masayuki Kabasawa | Information processing device for multiple instruction sets with reconfigurable mechanism |
US20060101427A1 (en) * | 2004-10-29 | 2006-05-11 | Tetsuya Yamada | Handover between software and hardware accelarator |
US20070106889A1 (en) * | 2001-02-21 | 2007-05-10 | Mips Technologies, Inc. | Configurable instruction sequence generation |
US7711763B2 (en) | 2001-02-21 | 2010-05-04 | Mips Technologies, Inc. | Microprocessor instructions for performing polynomial arithmetic operations |
US20100257338A1 (en) * | 2009-04-07 | 2010-10-07 | Spracklen Lawrence A | Methods and mechanisms to support multiple features for a number of opcodes |
US7860911B2 (en) | 2001-02-21 | 2010-12-28 | Mips Technologies, Inc. | Extended precision accumulator |
US20110231438A1 (en) * | 2008-09-19 | 2011-09-22 | Continental Automotive Gmbh | Infotainment System And Computer Program Product |
WO2013072038A1 (en) * | 2011-11-15 | 2013-05-23 | Giesecke & Devrient Gmbh | Method for controlling program execution |
US9535699B2 (en) | 2012-03-09 | 2017-01-03 | Panasonic Intellectual Property Management Co., Ltd. | Processor, multiprocessor system, compiler, software system, memory control system, and computer system |
Families Citing this family (31)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6691308B1 (en) * | 1999-12-30 | 2004-02-10 | Stmicroelectronics, Inc. | Method and apparatus for changing microcode to be executed in a processor |
US6694435B2 (en) * | 2001-07-25 | 2004-02-17 | Apple Computer, Inc. | Method of obfuscating computer instruction streams |
US7577944B2 (en) * | 2002-03-18 | 2009-08-18 | Hewlett-Packard Development Company, L.P. | Unbundling, translation and rebundling of instruction bundles in an instruction stream |
GB0209670D0 (en) * | 2002-04-26 | 2002-06-05 | Easics Nv | Efficient packet processing pipelining device and method |
US20050232303A1 (en) * | 2002-04-26 | 2005-10-20 | Koen Deforche | Efficient packet processing pipeline device and method |
US8108843B2 (en) * | 2002-09-17 | 2012-01-31 | International Business Machines Corporation | Hybrid mechanism for more efficient emulation and method therefor |
US9043194B2 (en) * | 2002-09-17 | 2015-05-26 | International Business Machines Corporation | Method and system for efficient emulation of multiprocessor memory consistency |
US7953588B2 (en) * | 2002-09-17 | 2011-05-31 | International Business Machines Corporation | Method and system for efficient emulation of multiprocessor address translation on a multiprocessor host |
GB2393270B (en) | 2002-09-19 | 2005-07-27 | Advanced Risc Mach Ltd | Executing variable length instructions stored within a plurality of discrete memory address regions |
AU2002363920A1 (en) * | 2002-10-29 | 2004-05-25 | Freescale Semiconductor, Inc. | Method and apparatus for selectively optimizing interpreted language code |
US7210121B2 (en) * | 2003-02-07 | 2007-04-24 | Sun Microsystems, Inc. | Method and system for generating first class citizen application implementing native software application wrapper |
CN100395709C (en) * | 2003-02-20 | 2008-06-18 | 皇家飞利浦电子股份有限公司 | Translation of a series of computer instructions |
JP4224430B2 (en) * | 2003-07-07 | 2009-02-12 | 株式会社ルネサステクノロジ | Information processing device |
US7617490B2 (en) * | 2003-09-10 | 2009-11-10 | Intel Corporation | Methods and apparatus for dynamic best fit compilation of mixed mode instructions |
US9189230B2 (en) | 2004-03-31 | 2015-11-17 | Intel Corporation | Method and system to provide concurrent user-level, non-privileged shared resource thread creation and execution |
US7921425B2 (en) * | 2005-03-14 | 2011-04-05 | Cisco Technology, Inc. | Techniques for allocating computing resources to applications in an embedded system |
GB2426840A (en) * | 2005-06-04 | 2006-12-06 | Transitive Ltd | Method of executing program code where a portion of the target code calls a native code portion which then calls a second target code portion. |
KR100781340B1 (en) * | 2006-09-18 | 2007-11-30 | 삼성전자주식회사 | System and method for processing user defined extended operation |
US8683453B2 (en) * | 2006-11-30 | 2014-03-25 | Motorola Mobility Llc | System for overriding interpreted byte-code with native code |
US7979685B1 (en) | 2007-11-27 | 2011-07-12 | Oracle America, Inc. | Multiple instruction execution mode resource-constrained device |
US9003377B2 (en) | 2010-01-07 | 2015-04-07 | Microsoft Technology Licensing, Llc | Efficient resumption of co-routines on a linear stack |
US8997049B1 (en) | 2010-05-21 | 2015-03-31 | Cadence Design Systems, Inc. | Method and system for debugging of compiled code using an interpreter |
WO2013112791A1 (en) * | 2012-01-26 | 2013-08-01 | Unisys Corporation | Translation of non-native code based on meta-operators |
US10146545B2 (en) | 2012-03-13 | 2018-12-04 | Nvidia Corporation | Translation address cache for a microprocessor |
US9880846B2 (en) | 2012-04-11 | 2018-01-30 | Nvidia Corporation | Improving hit rate of code translation redirection table with replacement strategy based on usage history table of evicted entries |
US10241810B2 (en) | 2012-05-18 | 2019-03-26 | Nvidia Corporation | Instruction-optimizing processor with branch-count table in hardware |
US9384001B2 (en) * | 2012-08-15 | 2016-07-05 | Nvidia Corporation | Custom chaining stubs for instruction code translation |
US20140189310A1 (en) | 2012-12-27 | 2014-07-03 | Nvidia Corporation | Fault detection in instruction translations |
US10108424B2 (en) | 2013-03-14 | 2018-10-23 | Nvidia Corporation | Profiling code portions to generate translations |
US10459600B2 (en) * | 2015-06-24 | 2019-10-29 | Microsoft Technology Licensing, Llc | Conversion of platform-independent accessibility logic into platform-specific accessibility functionality |
JP6352486B2 (en) * | 2017-04-12 | 2018-07-04 | インテル コーポレイション | ISA bridging by callback |
Citations (43)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3889243A (en) * | 1973-10-18 | 1975-06-10 | Ibm | Stack mechanism for a data processor |
US4263204A (en) * | 1977-11-29 | 1981-04-21 | Montedison S.P.A. | Monoazoic hydrosoluble cationic dyes |
US4587632A (en) * | 1980-05-27 | 1986-05-06 | At&T Bell Laboratories | Lookahead stack oriented computer |
US4922414A (en) * | 1982-12-17 | 1990-05-01 | Symbolics Inc. | Symbolic language data processing system |
US4969091A (en) * | 1987-08-06 | 1990-11-06 | Mueller Otto | Apparatus for stack control employing mixed hardware registers and memory |
US5136696A (en) * | 1988-06-27 | 1992-08-04 | Prime Computer, Inc. | High-performance pipelined central processor for predicting the occurrence of executing single-cycle instructions and multicycle instructions |
US5455775A (en) * | 1993-01-25 | 1995-10-03 | International Business Machines Corporation | Computer design system for mapping a logical hierarchy into a physical hierarchy |
US5619665A (en) * | 1995-04-13 | 1997-04-08 | Intrnational Business Machines Corporation | Method and apparatus for the transparent emulation of an existing instruction-set architecture by an arbitrary underlying instruction-set architecture |
US5638525A (en) * | 1995-02-10 | 1997-06-10 | Intel Corporation | Processor capable of executing programs that contain RISC and CISC instructions |
US5659703A (en) * | 1989-08-03 | 1997-08-19 | Patriot Scientific Corporation | Microprocessor system with hierarchical stack and method of operation |
US5740461A (en) * | 1994-05-03 | 1998-04-14 | Advanced Risc Machines Limited | Data processing with multiple instruction sets |
US5742802A (en) * | 1996-02-16 | 1998-04-21 | International Business Machines Corporation | Method and system for efficiently mapping guest instruction in an emulation assist unit |
US5752035A (en) * | 1995-04-05 | 1998-05-12 | Xilinx, Inc. | Method for compiling and executing programs for reprogrammable instruction set accelerator |
US5838948A (en) * | 1995-12-01 | 1998-11-17 | Eagle Design Automation, Inc. | System and method for simulation of computer systems combining hardware and software interaction |
US5875336A (en) * | 1997-03-31 | 1999-02-23 | International Business Machines Corporation | Method and system for translating a non-native bytecode to a set of codes native to a processor within a computer system |
US5892966A (en) * | 1997-06-27 | 1999-04-06 | Sun Microsystems, Inc. | Processor complex for executing multimedia functions |
US5926832A (en) * | 1996-09-26 | 1999-07-20 | Transmeta Corporation | Method and apparatus for aliasing memory data in an advanced microprocessor |
US5925123A (en) * | 1996-01-24 | 1999-07-20 | Sun Microsystems, Inc. | Processor for executing instruction sets received from a network or from a local memory |
US5937193A (en) * | 1996-11-27 | 1999-08-10 | Vlsi Technology, Inc. | Circuit arrangement for translating platform-independent instructions for execution on a hardware platform and method thereof |
US5953741A (en) * | 1996-11-27 | 1999-09-14 | Vlsi Technology, Inc. | Stack cache for stack-based processor and method thereof |
US6003126A (en) * | 1997-07-01 | 1999-12-14 | International Business Machines | Special instruction register including allocation field utilized for temporary designation of physical registers as general registers |
US6009509A (en) * | 1997-10-08 | 1999-12-28 | International Business Machines Corporation | Method and system for the temporary designation and utilization of a plurality of physical registers as a stack |
US6009499A (en) * | 1997-03-31 | 1999-12-28 | Sun Microsystems, Inc | Pipelined stack caching circuit |
US6014723A (en) * | 1996-01-24 | 2000-01-11 | Sun Microsystems, Inc. | Processor with accelerated array access bounds checking |
US6021469A (en) * | 1996-01-24 | 2000-02-01 | Sun Microsystems, Inc. | Hardware virtual machine instruction processor |
US6031992A (en) * | 1996-07-05 | 2000-02-29 | Transmeta Corporation | Combining hardware and software to provide an improved microprocessor |
US6038643A (en) * | 1996-01-24 | 2000-03-14 | Sun Microsystems, Inc. | Stack management unit and method for a processor having a stack |
US6070173A (en) * | 1997-11-26 | 2000-05-30 | International Business Machines Corporation | Method and apparatus for assisting garbage collection process within a java virtual machine |
US6088786A (en) * | 1997-06-27 | 2000-07-11 | Sun Microsystems, Inc. | Method and system for coupling a stack based processor to register based functional unit |
US6122638A (en) * | 1997-11-26 | 2000-09-19 | International Business Machines Corporation | Object-oriented processor and method for caching intermediate data in an object-oriented processor |
US6148391A (en) * | 1998-03-26 | 2000-11-14 | Sun Microsystems, Inc. | System for simultaneously accessing one or more stack elements by multiple functional units using real stack addresses |
US6298434B1 (en) * | 1997-10-02 | 2001-10-02 | U.S. Philips Corporation | Data processing device for processing virtual machine instructions |
US6317872B1 (en) * | 1997-07-11 | 2001-11-13 | Rockwell Collins, Inc. | Real time processor optimized for executing JAVA programs |
US6332215B1 (en) * | 1998-12-08 | 2001-12-18 | Nazomi Communications, Inc. | Java virtual machine hardware for RISC and CISC processors |
US6338134B1 (en) * | 1998-12-29 | 2002-01-08 | International Business Machines Corporation | Method and system in a superscalar data processing system for the efficient processing of an instruction by moving only pointers to data |
US6338160B1 (en) * | 1998-12-08 | 2002-01-08 | Nazomi Communications, Inc. | Constant pool reference resolution method |
US6349377B1 (en) * | 1997-10-02 | 2002-02-19 | U.S. Philips Corporation | Processing device for executing virtual machine instructions that includes instruction refeeding means |
US6374286B1 (en) * | 1998-04-06 | 2002-04-16 | Rockwell Collins, Inc. | Real time processor capable of concurrently running multiple independent JAVA machines |
US6502237B1 (en) * | 1996-01-29 | 2002-12-31 | Compaq Information Technologies Group, L.P. | Method and apparatus for performing binary translation method and apparatus for performing binary translation |
US6513057B1 (en) * | 1996-10-28 | 2003-01-28 | Unisys Corporation | Heterogeneous symmetric multi-processing system |
US6564179B1 (en) * | 1999-07-26 | 2003-05-13 | Agere Systems Inc. | DSP emulating a microcontroller |
US6606743B1 (en) * | 1996-11-13 | 2003-08-12 | Razim Technology, Inc. | Real time program language accelerator |
US6826749B2 (en) * | 1998-12-08 | 2004-11-30 | Nazomi Communications, Inc. | Java hardware accelerator using thread manager |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4236204A (en) * | 1978-03-13 | 1980-11-25 | Motorola, Inc. | Instruction set modifier register |
GB2307072B (en) * | 1994-06-10 | 1998-05-13 | Advanced Risc Mach Ltd | Interoperability with multiple instruction sets |
US5805895A (en) * | 1996-06-09 | 1998-09-08 | Motorola, Inc. | Method and apparatus for code translation optimization |
US6066181A (en) * | 1997-12-08 | 2000-05-23 | Analysis & Technology, Inc. | Java native interface code generator |
US6199202B1 (en) * | 1998-01-06 | 2001-03-06 | Hewlett-Packard Company | Method and apparatus for the inter-operation of differing architectural and run time conventions |
US6351844B1 (en) * | 1998-11-05 | 2002-02-26 | Hewlett-Packard Company | Method for selecting active code traces for translation in a caching dynamic translator |
JP2001195250A (en) * | 2000-01-13 | 2001-07-19 | Mitsubishi Electric Corp | Instruction translator and instruction memory with translator and data processor using the same |
DE10104043A1 (en) * | 2000-02-23 | 2001-08-30 | Ibm | Method and system for improving the use of in-built macro-languages makes it possible for existing applications to use languages other than their in-built macro-languages without changing an existing application. |
-
2000
- 2000-12-07 US US09/731,060 patent/US20020069402A1/en not_active Abandoned
-
2001
- 2001-06-25 US US09/887,561 patent/US7134119B2/en not_active Expired - Lifetime
- 2001-08-29 JP JP2001259954A patent/JP4938187B2/en not_active Expired - Lifetime
Patent Citations (47)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3889243A (en) * | 1973-10-18 | 1975-06-10 | Ibm | Stack mechanism for a data processor |
US4263204A (en) * | 1977-11-29 | 1981-04-21 | Montedison S.P.A. | Monoazoic hydrosoluble cationic dyes |
US4587632A (en) * | 1980-05-27 | 1986-05-06 | At&T Bell Laboratories | Lookahead stack oriented computer |
US4922414A (en) * | 1982-12-17 | 1990-05-01 | Symbolics Inc. | Symbolic language data processing system |
US4969091A (en) * | 1987-08-06 | 1990-11-06 | Mueller Otto | Apparatus for stack control employing mixed hardware registers and memory |
US5136696A (en) * | 1988-06-27 | 1992-08-04 | Prime Computer, Inc. | High-performance pipelined central processor for predicting the occurrence of executing single-cycle instructions and multicycle instructions |
US5809336A (en) * | 1989-08-03 | 1998-09-15 | Patriot Scientific Corporation | High performance microprocessor having variable speed system clock |
US5784584A (en) * | 1989-08-03 | 1998-07-21 | Patriot Scientific Corporation | High performance microprocessor using instructions that operate within instruction groups |
US5659703A (en) * | 1989-08-03 | 1997-08-19 | Patriot Scientific Corporation | Microprocessor system with hierarchical stack and method of operation |
US5455775A (en) * | 1993-01-25 | 1995-10-03 | International Business Machines Corporation | Computer design system for mapping a logical hierarchy into a physical hierarchy |
US5740461A (en) * | 1994-05-03 | 1998-04-14 | Advanced Risc Machines Limited | Data processing with multiple instruction sets |
US5638525A (en) * | 1995-02-10 | 1997-06-10 | Intel Corporation | Processor capable of executing programs that contain RISC and CISC instructions |
US5752035A (en) * | 1995-04-05 | 1998-05-12 | Xilinx, Inc. | Method for compiling and executing programs for reprogrammable instruction set accelerator |
US5619665A (en) * | 1995-04-13 | 1997-04-08 | Intrnational Business Machines Corporation | Method and apparatus for the transparent emulation of an existing instruction-set architecture by an arbitrary underlying instruction-set architecture |
US5838948A (en) * | 1995-12-01 | 1998-11-17 | Eagle Design Automation, Inc. | System and method for simulation of computer systems combining hardware and software interaction |
US6014723A (en) * | 1996-01-24 | 2000-01-11 | Sun Microsystems, Inc. | Processor with accelerated array access bounds checking |
US6125439A (en) * | 1996-01-24 | 2000-09-26 | Sun Microsystems, Inc. | Process of executing a method on a stack-based processor |
US6038643A (en) * | 1996-01-24 | 2000-03-14 | Sun Microsystems, Inc. | Stack management unit and method for a processor having a stack |
US5925123A (en) * | 1996-01-24 | 1999-07-20 | Sun Microsystems, Inc. | Processor for executing instruction sets received from a network or from a local memory |
US6026485A (en) * | 1996-01-24 | 2000-02-15 | Sun Microsystems, Inc. | Instruction folding for a stack-based machine |
US6021469A (en) * | 1996-01-24 | 2000-02-01 | Sun Microsystems, Inc. | Hardware virtual machine instruction processor |
US6502237B1 (en) * | 1996-01-29 | 2002-12-31 | Compaq Information Technologies Group, L.P. | Method and apparatus for performing binary translation method and apparatus for performing binary translation |
US5742802A (en) * | 1996-02-16 | 1998-04-21 | International Business Machines Corporation | Method and system for efficiently mapping guest instruction in an emulation assist unit |
US6031992A (en) * | 1996-07-05 | 2000-02-29 | Transmeta Corporation | Combining hardware and software to provide an improved microprocessor |
US5926832A (en) * | 1996-09-26 | 1999-07-20 | Transmeta Corporation | Method and apparatus for aliasing memory data in an advanced microprocessor |
US6513057B1 (en) * | 1996-10-28 | 2003-01-28 | Unisys Corporation | Heterogeneous symmetric multi-processing system |
US6606743B1 (en) * | 1996-11-13 | 2003-08-12 | Razim Technology, Inc. | Real time program language accelerator |
US5953741A (en) * | 1996-11-27 | 1999-09-14 | Vlsi Technology, Inc. | Stack cache for stack-based processor and method thereof |
US5937193A (en) * | 1996-11-27 | 1999-08-10 | Vlsi Technology, Inc. | Circuit arrangement for translating platform-independent instructions for execution on a hardware platform and method thereof |
US6009499A (en) * | 1997-03-31 | 1999-12-28 | Sun Microsystems, Inc | Pipelined stack caching circuit |
US5875336A (en) * | 1997-03-31 | 1999-02-23 | International Business Machines Corporation | Method and system for translating a non-native bytecode to a set of codes native to a processor within a computer system |
US6088786A (en) * | 1997-06-27 | 2000-07-11 | Sun Microsystems, Inc. | Method and system for coupling a stack based processor to register based functional unit |
US5892966A (en) * | 1997-06-27 | 1999-04-06 | Sun Microsystems, Inc. | Processor complex for executing multimedia functions |
US6003126A (en) * | 1997-07-01 | 1999-12-14 | International Business Machines | Special instruction register including allocation field utilized for temporary designation of physical registers as general registers |
US6317872B1 (en) * | 1997-07-11 | 2001-11-13 | Rockwell Collins, Inc. | Real time processor optimized for executing JAVA programs |
US6298434B1 (en) * | 1997-10-02 | 2001-10-02 | U.S. Philips Corporation | Data processing device for processing virtual machine instructions |
US6349377B1 (en) * | 1997-10-02 | 2002-02-19 | U.S. Philips Corporation | Processing device for executing virtual machine instructions that includes instruction refeeding means |
US6009509A (en) * | 1997-10-08 | 1999-12-28 | International Business Machines Corporation | Method and system for the temporary designation and utilization of a plurality of physical registers as a stack |
US6122638A (en) * | 1997-11-26 | 2000-09-19 | International Business Machines Corporation | Object-oriented processor and method for caching intermediate data in an object-oriented processor |
US6070173A (en) * | 1997-11-26 | 2000-05-30 | International Business Machines Corporation | Method and apparatus for assisting garbage collection process within a java virtual machine |
US6148391A (en) * | 1998-03-26 | 2000-11-14 | Sun Microsystems, Inc. | System for simultaneously accessing one or more stack elements by multiple functional units using real stack addresses |
US6374286B1 (en) * | 1998-04-06 | 2002-04-16 | Rockwell Collins, Inc. | Real time processor capable of concurrently running multiple independent JAVA machines |
US6332215B1 (en) * | 1998-12-08 | 2001-12-18 | Nazomi Communications, Inc. | Java virtual machine hardware for RISC and CISC processors |
US6338160B1 (en) * | 1998-12-08 | 2002-01-08 | Nazomi Communications, Inc. | Constant pool reference resolution method |
US6826749B2 (en) * | 1998-12-08 | 2004-11-30 | Nazomi Communications, Inc. | Java hardware accelerator using thread manager |
US6338134B1 (en) * | 1998-12-29 | 2002-01-08 | International Business Machines Corporation | Method and system in a superscalar data processing system for the efficient processing of an instruction by moving only pointers to data |
US6564179B1 (en) * | 1999-07-26 | 2003-05-13 | Agere Systems Inc. | DSP emulating a microcontroller |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020099930A1 (en) * | 2000-12-04 | 2002-07-25 | Mitsubishi Denki Kabushiki Kaisha | Data processor having translator and interpreter that execute non-native instructions |
US6820252B2 (en) * | 2000-12-04 | 2004-11-16 | Renesas Technology Corp. | Selective conversion to native code using hardware translator, software translator, and software interpreter |
US20070106889A1 (en) * | 2001-02-21 | 2007-05-10 | Mips Technologies, Inc. | Configurable instruction sequence generation |
US7711763B2 (en) | 2001-02-21 | 2010-05-04 | Mips Technologies, Inc. | Microprocessor instructions for performing polynomial arithmetic operations |
US8447958B2 (en) | 2001-02-21 | 2013-05-21 | Bridge Crossing, Llc | Substituting portion of template instruction parameter with selected virtual instruction parameter |
US7860911B2 (en) | 2001-02-21 | 2010-12-28 | Mips Technologies, Inc. | Extended precision accumulator |
US20040031022A1 (en) * | 2002-08-02 | 2004-02-12 | Masayuki Kabasawa | Information processing device for multiple instruction sets with reconfigurable mechanism |
US20060101427A1 (en) * | 2004-10-29 | 2006-05-11 | Tetsuya Yamada | Handover between software and hardware accelarator |
US7853776B2 (en) * | 2004-10-29 | 2010-12-14 | Renesas Technology Corp. | Handover between software and hardware accelerator |
US20110231438A1 (en) * | 2008-09-19 | 2011-09-22 | Continental Automotive Gmbh | Infotainment System And Computer Program Product |
US8195923B2 (en) * | 2009-04-07 | 2012-06-05 | Oracle America, Inc. | Methods and mechanisms to support multiple features for a number of opcodes |
US20100257338A1 (en) * | 2009-04-07 | 2010-10-07 | Spracklen Lawrence A | Methods and mechanisms to support multiple features for a number of opcodes |
WO2013072038A1 (en) * | 2011-11-15 | 2013-05-23 | Giesecke & Devrient Gmbh | Method for controlling program execution |
US9535699B2 (en) | 2012-03-09 | 2017-01-03 | Panasonic Intellectual Property Management Co., Ltd. | Processor, multiprocessor system, compiler, software system, memory control system, and computer system |
Also Published As
Publication number | Publication date |
---|---|
US7134119B2 (en) | 2006-11-07 |
JP4938187B2 (en) | 2012-05-23 |
JP2002116908A (en) | 2002-04-19 |
US20020108103A1 (en) | 2002-08-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7003652B2 (en) | Restarting translated instructions | |
US20020069402A1 (en) | Scheduling control within a system having mixed hardware and software based instruction execution | |
US7000094B2 (en) | Storing stack operands in registers | |
US20020083302A1 (en) | Hardware instruction translation within a processor pipeline | |
US4879646A (en) | Data processing system with a pipelined structure for editing trace memory contents and tracing operations during system debugging | |
US5944841A (en) | Microprocessor with built-in instruction tracing capability | |
US8578351B2 (en) | Hybrid mechanism for more efficient emulation and method therefor | |
US5889985A (en) | Array prefetch apparatus and method | |
Goudge et al. | Thumb: reducing the cost of 32-bit RISC performance in portable and consumer applications | |
US4173782A (en) | Return and link mechanism | |
US7356673B2 (en) | System and method including distributed instruction buffers for storing frequently executed instructions in predecoded form | |
GB2367652A (en) | Scheduling control within a system having mixed hardware and software based instruction execution | |
GB2367658A (en) | Intercalling between native and non-native instruction sets | |
WO1999023549A1 (en) | Direct cache accessing primary operations hierarchically organized to snippets and threads implemented in isa processor |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: ARM LIMITED, UNITED KINGDOM Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:NEVILL, EDWARD C.;ROSE, ANDREW C.;REEL/FRAME:011542/0630 Effective date: 20001206 |
|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- AFTER EXAMINER'S ANSWER OR BOARD OF APPEALS DECISION |