WO1995008801A1 - Execution of data processing instructions - Google Patents

Execution of data processing instructions Download PDF

Info

Publication number
WO1995008801A1
WO1995008801A1 PCT/GB1994/001793 GB9401793W WO9508801A1 WO 1995008801 A1 WO1995008801 A1 WO 1995008801A1 GB 9401793 W GB9401793 W GB 9401793W WO 9508801 A1 WO9508801 A1 WO 9508801A1
Authority
WO
WIPO (PCT)
Prior art keywords
instruction
memory
data
executed
data processing
Prior art date
Application number
PCT/GB1994/001793
Other languages
French (fr)
Inventor
David Vivian Jaggar
Original Assignee
Advanced Risc Machines Limited
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Advanced Risc Machines Limited filed Critical Advanced Risc Machines Limited
Priority to RU96107418A priority Critical patent/RU2137182C1/en
Priority to EP94923797A priority patent/EP0721619B1/en
Priority to US08/619,647 priority patent/US5961633A/en
Priority to KR1019960701507A priority patent/KR100335785B1/en
Priority to JP50961895A priority patent/JP3553946B2/en
Priority to DE69414592T priority patent/DE69414592T2/en
Publication of WO1995008801A1 publication Critical patent/WO1995008801A1/en

Links

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
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/38—Concurrent instruction execution, e.g. pipeline, look ahead
    • 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
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30003—Arrangements for executing specific machine instructions
    • G06F9/30072—Arrangements for executing specific machine instructions to perform conditional operations, e.g. using predicates or guards

Definitions

  • CPU central processor
  • RAM random access memory
  • a previously proposed example of such a data processor is the ARM6 processor, described in the "ARM6 Data Sheet" published by Advanced Rise Machines Limited, 1993- During a memory-.access, the memory management unit may generate an abort signal indicating that the current memory access cannot be completed. Abort signals may be generated for a number of reasons.
  • an aborted memory access can occur in a data processing system employing virtual memory in which data are swapped between a RAM and slower disk storage to give the illusion that the addressable memory space is greater than the amount of RAM provided.
  • virtual memory in which data are swapped between a RAM and slower disk storage to give the illusion that the addressable memory space is greater than the amount of RAM provided.
  • data corresponding to a required virtual address are currently held in the disk storage rather than the RAM, there will be a delay before those data are accessible, during which delay the data have to be transferred from the disk storage into the RAM. In this case, the current memory access is aborted, and an attempt is made later to access those data.
  • the abort signal supplied from the memory management unit is generated too late to stop execution of the instruction which initiate. the failed memory access, but can instead be used to cancel executic of the following data processing instruction, i.e. the data processing instruction after the one which initiated the failed memory access. This is useful because subsequent instructions may rely on the memory access having been successful.
  • the use of the abort signal to cancel execution of the instruction immediately following the instruction which initiated the failed memory access places stringent requiremer on the timing of the abort signal. Alternatively, a complex mechanism must be provided to 'undo' the results of the execution of the immediately following instruction, after execution of that instruction has been completed.
  • Figure 1 of the accompanying drawings is a schematic timing diagram illustrating the timing requirements of the abort signal during a data write operation (in which data are written to RAM) by the previously proposed data processor referred to above.
  • a clock signal 10 controls the execution of data processing instructions by the data processor.
  • a memory address 20 is supplied by the data processor to a memory management unit, and one half-cycle of the clock signal later, the data 30 to be written to that address are output by the data processor.
  • an abort signal 40 is generated by the memory management unit and supplied to the data processor.
  • the next instruction after a data write instruction is executed straight away, since there is no need (under normal circumstances) to await a response from the memory management unit after the data to be written have been placed on the data bus. Accordingly, in order for the abort signal to arrive in time to cancel execution of the immediately following instruction, the previously proposed data processor referred to above requires the abort signal to be valid one half-cycle of the clock signal before the data to be written are output by the data processor.
  • this timing constraint is difficult to achieve, and requires particularly fast operation of the memory management unit (with a correspondingly high power consumption by that unit) .
  • the previously proposed data processor referred to above also provides conditional execution of its entire instruction set. This is achieved by comparing the current state of up to four processing flags with respective states defined by a condition code included in each instruction. This comparison takes place concurrently with execution of the instruction. Complete execution of an instruction is then prevented if the state of the processing flags does not match that specified by the condition code.
  • This invention provides a data processing apparatus in which successive data processing instructions are executed, the apparatus comprising: memory accessing means for accessing a data memory in response to one or more of the instructions, the memory accessing means comprising means for detecting whether each memory access is invalid; condition test means, responsive to a processing state of the apparatus generated by previously executed instructions and operable during execution of each instruction, for detecting whether that instruction should be executed; and conditional control means, responsive to the memory accessing means and to the condition test means, for preventing complete execution of a current instruction if either the memory accessing means detects that a memory access initiated by the preceding instruction is inval- . or the condition test means detects that the current instruction should not be executed.
  • a fully conditional instruction set is employed, and the mechanism for conditionally preventing complete execution of each instruction is also used to handle memory aborts.
  • memory abort signals relating to a memory access initiated by the preceding instruction can be received at the same t: , during execution of each instruction, as the detection by the condition test means of whether that instruction should be executed. This can allow memory aborts to be processed at a later time for each instruction than that allowed for the previously proposed data processor described above.
  • the apparatus comprises one or more processing flags for storing data indicative of a current processing state of the data processing apparatus; each instruction includes a condition code defining a state of the processing flags required for that instruction to be executed; and the condition test means is operable to compare the required state of the processing flags defined by the condition code in each instruction with the actual state of the processing flags.
  • the condition codes may specify that a particular processing flags should be set to a particular logical state, or that the state of that processing f1,g has no influence on whether the current instruction should be executed. In an extreme case, a possible condition code may specify that a particular instruction should be executed regardless of the state of any of the processing flags.
  • the processing flags could specify various features of the processing state of the apparatus.
  • the apparatus comprises four processing flags respectively denoting:
  • the memory accessing means comprises: means for transmitting a memory address to the data memory; and means for subsequently transmitting data to the data memory or receiving data from the data memory.
  • conditional control means can conveniently be made responsive to both the memory accessing means and to the condition test means, it is preferred that: the memory accessing means is operable to generate an abort control signal to indicate that a memory access is invalid; the condition test means is operable to generate a condition failure control signal to indicate that the current instruction should not be executed; and the apparatus comprises means for combining the abort control signal and the condition failure control signal to generate a combined control signal for supply to the condition control means.
  • the means for combining comprises a logical OR gate.
  • data processing operations of the apparatus are controlled by a clock signal.
  • this invention provides an integrated circuit comprising apparatus as defined above.
  • this invention provides a method of data processing in which successive data processing instructions are conditionally executed, the method comprising the steps of: accessing a data memory in response to one or more of the instructions; detecting whether each memory access is invalid; detecting, during execution of each instruction, whether that instruction should be executed, in dependence on a processing state of the apparatus generated by previously executed instructions; and preventing complete execution of a current instruction if it is detected either that a memory access initiated by the preceding instruction is invalid or that the current instruction should not be executed.
  • Figure 1 is a schematic timing diagram illustrating the timing requirements of an abort signal during a data write operation by a previously proposed data processor
  • Figure 2 is a schematic block diagram of a data processing apparatus according to an embodiment of the invention
  • Figure 3 is a schematic block diagram of a part of a central processing unit.
  • Figure 4 is a schematic timing diagram illustrating the timing requirements of an abort signal during a data write operation by the data processing apparatus of Figure 2.
  • the apparatus comprises a central processing unit (CPU)
  • MMU memory management unit
  • RAM random access memory
  • An abort control line 150 is also provided from the MMU 110 to the CPU 100 to carry an abort signal indicative of a failed or invalid memory access.
  • a clock signal is supplied to the CPU 100 and to the MMU 110 to control the operations of both of these units.
  • Each data processing instruction is executed by the CPU 100 in a particular number of cycles of the clock signal (depending on the nature of that instruction) and memory access by the MMU 110 is performed in synchronism with the clock signal.
  • the MMU 110 operates under the control of the CPU 100 and the clock signal to access data stored in the RAM 120. Accordingly, the MMU is linked to the RAM by a plurality of address and data lines 160.
  • FIG. 3 is a schematic block diagram of a part of the central processing unit 100.
  • the CPU 100 employs instruction pipelining, to allow the processing and memory operations to be performed substantially continuously. Typically, while one data processing instruction is being executed, its successor is being decoded and a third instruction is being fetched from memory. This arrangement is referred to as a three-stage execution pipeline.
  • data processing instructions which have been fetched from memory and then decoded are passed for execution.
  • the data processing instructions are 32-bit data words, of which bits 28 to 31 form a 4-bit condition field.
  • the remaining bits (bits 0 to 27) define the operation to be performed in response to that instruction and, in some cases an operand on which the operation is to be performed.
  • condition field (bits 28 to 31) is passed to a condition tester 200 which compares the bits of the condition field with 16 pre- defined condition codes.
  • the 16 condition codes define the state of one or more of four processor flags 210, referred to as the N, Z, C and V flags. These flags represent a processing state of the CPU 100 generated by previously executed instructions.
  • the N flag denotes a negative result from the previous arithmetic operation of the CPU 100; the Z flag denotes a zero (equal) result from the previous CPU operation; the C flag denotes whether a carry bit was set during the previous CPU operation; and the V flag indicates an arithmetic overflow occurring during the previous CPU operation.
  • 0001 NE Z clear (not equal)
  • 0100 MI N set (negative)
  • 0101 PL N clear (positive or zero )
  • the condition tester 200 tests the state of the flags listed above, in dependence on which bits are set in the condition field of the current instruction. This comparison takes place during execution of the current instruction. The current instruction is allowed to complete its execution only if the appropriate flags are set to the states specified by the condition field.
  • the instruction will be executed irrespective of the flags.
  • the never (NV) condition code prevents execution of the instruction irrespective of the state of the flags 210.
  • the condition tester 200 generates an output signal 220 indicating whether the current instruction should be completely executed.
  • the output signal 220 is combined with the abort signal supplied on the abort control line 150 from the MMU 110 to the CPU 100, using an OR-gate 230.
  • the output of the OR-gate 230 is therefore set if either the condition tester 200 indicates that the current instruction should not be completely executed, or the abort signal is asserted by the MMU 110.
  • An instruction execution unit 240 receives bits 0 to 27 of each instruction, defining the operation to be performed and, in some cases, an operand on which the operation is to be performed. During execution of the current instruction (i.e. after the propagation and processing delays of the condition tester 200 and the OR-gate 230) , the instruction execution unit 240 receives the output of the OR-gate 230 indicating whether the current instruction should be completely executed. If the output of the OR-gate 230 indicates that the current instruction should not be completely executed, the instruction is cancelled without changing the state of any registers or memory locations associated with the apparatus.
  • the condition tester 200 and the instruction execution unit 240 may be of the same form as the corresponding components of the previously proposed ARM 6 processor referred to above.
  • the result of using the apparatus of Figure 3 is that the stringent timing of the abort signal is greatly relaxed, so that the abort signal relating to a memory access initiated by the preceding instruction may be set at a late stage during execution of each instruction.
  • Data read operations are followed by a non-memory-accessing (internal) processor cycle, to allow time for the data which has been read from memory to be loaded into the appropriate processor register. Accordingly, if an abort signal is received as a result of a failed data read operation, the internal cycle following the read operation allows time for any data supplied from memory by the MMU 110 (which data may well be erroneous) to be ignored and not stored in the intended location (e.g. a processor register).
  • the timing relationship of the instruction execution and the receipt of the abort signal for a data write operation is illustrated in Figure 4, in which a clock signal 300 which controls instruction execution by the CPU 100 is illustrated along with the states of the address bus 130 and the data bus l4 ⁇ .
  • an address 310 is placed on the address bus by the CPU 100.
  • data 320 are placed on the data bus 140 by the CPU 100. Once this has been done, the CPU 100 is able to execute the next instruction during the following cycle of the clock signal 300.

Abstract

Data processing apparatus in which successive data processing instructions are executed comprises: memory accessing means for accessing a data memory in response to one or more of the instructions, the memory accessing means comprising means for detecting whether each memory access is invalid; condition test means, responsive to a processing state of the apparatus generated by previously executed instructions and operable during execution of each instruction, for detecting whether that instruction should be executed; and conditional control means, responsive to the memory accessing means and to the condition test means, for preventing complete execution of a current instruction if either the memory accessing means detects that a memory access initiated by the preceding instruction is invalid or the condition test means detects that the current instruction should not be executed.

Description

EXECUTION OF DATA PROCESSING INSTRUCTIONS
BACKGROUND OF THE INVENTION Field of the Invention This inventic** relates to the execution of data processing instructions. Description of the Prior Art
Some data processors comprise a central processor ur t (CPU) which is able, under the control of a currently exec =d data processing instruction, to access data stored in a random access memory (RAM) via an intermediate memory management unit. A previously proposed example of such a data processor is the ARM6 processor, described in the "ARM6 Data Sheet" published by Advanced Rise Machines Limited, 1993- During a memory-.access, the memory management unit may generate an abort signal indicating that the current memory access cannot be completed. Abort signals may be generated for a number of reasons. In one example, an aborted memory access can occur in a data processing system employing virtual memory in which data are swapped between a RAM and slower disk storage to give the illusion that the addressable memory space is greater than the amount of RAM provided. In such a system, if data corresponding to a required virtual address are currently held in the disk storage rather than the RAM, there will be a delay before those data are accessible, during which delay the data have to be transferred from the disk storage into the RAM. In this case, the current memory access is aborted, and an attempt is made later to access those data.
The abort signal supplied from the memory management unit is generated too late to stop execution of the instruction which initiate. the failed memory access, but can instead be used to cancel executic of the following data processing instruction, i.e. the data processing instruction after the one which initiated the failed memory access. This is useful because subsequent instructions may rely on the memory access having been successful. The use of the abort signal to cancel execution of the instruction immediately following the instruction which initiated the failed memory access places stringent requiremer on the timing of the abort signal. Alternatively, a complex mechanism must be provided to 'undo' the results of the execution of the immediately following instruction, after execution of that instruction has been completed. Figure 1 of the accompanying drawings is a schematic timing diagram illustrating the timing requirements of the abort signal during a data write operation (in which data are written to RAM) by the previously proposed data processor referred to above.
Referring to Figure 1, a clock signal 10 controls the execution of data processing instructions by the data processor. When a data write operation is initiated, a memory address 20 is supplied by the data processor to a memory management unit, and one half-cycle of the clock signal later, the data 30 to be written to that address are output by the data processor.
If the memory management unit detects that the memory address 20 is invalid (for example, because data corresponding to that address are currently held in disk storage in a virtual memory system) , an abort signal 40 is generated by the memory management unit and supplied to the data processor.
The next instruction after a data write instruction is executed straight away, since there is no need (under normal circumstances) to await a response from the memory management unit after the data to be written have been placed on the data bus. Accordingly, in order for the abort signal to arrive in time to cancel execution of the immediately following instruction, the previously proposed data processor referred to above requires the abort signal to be valid one half-cycle of the clock signal before the data to be written are output by the data processor.
In practice, this timing constraint is difficult to achieve, and requires particularly fast operation of the memory management unit (with a correspondingly high power consumption by that unit) .
The previously proposed data processor referred to above also provides conditional execution of its entire instruction set. This is achieved by comparing the current state of up to four processing flags with respective states defined by a condition code included in each instruction. This comparison takes place concurrently with execution of the instruction. Complete execution of an instruction is then prevented if the state of the processing flags does not match that specified by the condition code. SUMMARY OF THE INVENTION
This invention provides a data processing apparatus in which successive data processing instructions are executed, the apparatus comprising: memory accessing means for accessing a data memory in response to one or more of the instructions, the memory accessing means comprising means for detecting whether each memory access is invalid; condition test means, responsive to a processing state of the apparatus generated by previously executed instructions and operable during execution of each instruction, for detecting whether that instruction should be executed; and conditional control means, responsive to the memory accessing means and to the condition test means, for preventing complete execution of a current instruction if either the memory accessing means detects that a memory access initiated by the preceding instruction is inval- . or the condition test means detects that the current instruction should not be executed.
In a data processing apparatus according to the invention, a fully conditional instruction set is employed, and the mechanism for conditionally preventing complete execution of each instruction is also used to handle memory aborts. Thus, memory abort signals relating to a memory access initiated by the preceding instruction can be received at the same t: , during execution of each instruction, as the detection by the condition test means of whether that instruction should be executed. This can allow memory aborts to be processed at a later time for each instruction than that allowed for the previously proposed data processor described above.
In a preferred embodiment, the apparatus comprises one or more processing flags for storing data indicative of a current processing state of the data processing apparatus; each instruction includes a condition code defining a state of the processing flags required for that instruction to be executed; and the condition test means is operable to compare the required state of the processing flags defined by the condition code in each instruction with the actual state of the processing flags. The condition codes may specify that a particular processing flags should be set to a particular logical state, or that the state of that processing f1,g has no influence on whether the current instruction should be executed. In an extreme case, a possible condition code may specify that a particular instruction should be executed regardless of the state of any of the processing flags.
The processing flags could specify various features of the processing state of the apparatus. In a preferred embodiment the apparatus comprises four processing flags respectively denoting:
(i) whether a previous data processing operation of the apparatus generated a negative result;
(ii) whether a previous data processing operation of the apparatus generated a zero result; (iϋ) whether a carry bit was set by a previous data processing operation of the apparatus; and
(iv) whether an arithmetic overflow occurred during a previous data processing operation of the apparatus.
Preferably the memory accessing means comprises: means for transmitting a memory address to the data memory; and means for subsequently transmitting data to the data memory or receiving data from the data memory.
In order that the conditional control means can conveniently be made responsive to both the memory accessing means and to the condition test means, it is preferred that: the memory accessing means is operable to generate an abort control signal to indicate that a memory access is invalid; the condition test means is operable to generate a condition failure control signal to indicate that the current instruction should not be executed; and the apparatus comprises means for combining the abort control signal and the condition failure control signal to generate a combined control signal for supply to the condition control means.
In an advantageously simple embodiment, the means for combining comprises a logical OR gate. In a convenient embodiment, data processing operations of the apparatus are controlled by a clock signal.
Viewed from a second aspect this invention provides an integrated circuit comprising apparatus as defined above.
Viewed from a third aspect this invention provides a method of data processing in which successive data processing instructions are conditionally executed, the method comprising the steps of: accessing a data memory in response to one or more of the instructions; detecting whether each memory access is invalid; detecting, during execution of each instruction, whether that instruction should be executed, in dependence on a processing state of the apparatus generated by previously executed instructions; and preventing complete execution of a current instruction if it is detected either that a memory access initiated by the preceding instruction is invalid or that the current instruction should not be executed.
BRTEF DESCRIPTION OF THE DRAWINGS
The invention will now be described by way of example with reference to the accompanying drawings, throughout which like parts are referred to by like references, and in which:
Figure 1 is a schematic timing diagram illustrating the timing requirements of an abort signal during a data write operation by a previously proposed data processor; Figure 2 is a schematic block diagram of a data processing apparatus according to an embodiment of the invention;
Figure 3 is a schematic block diagram of a part of a central processing unit; and
Figure 4 is a schematic timing diagram illustrating the timing requirements of an abort signal during a data write operation by the data processing apparatus of Figure 2.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
Referring now to Figure 2, a schematic block diagram of a data processing apparatus according to an embodiment of the invention is illustrated. The apparatus comprises a central processing unit (CPU)
100, a memory management unit (MMU) 110 and a random access memory
(RAM) 120. The CPU 100 and the MMU 110 are linked by an address bus
130 for the exchange of memory addresses and a data bus 140 for the exchange of data. An abort control line 150 is also provided from the MMU 110 to the CPU 100 to carry an abort signal indicative of a failed or invalid memory access.
A clock signal is supplied to the CPU 100 and to the MMU 110 to control the operations of both of these units. Each data processing instruction is executed by the CPU 100 in a particular number of cycles of the clock signal (depending on the nature of that instruction) and memory access by the MMU 110 is performed in synchronism with the clock signal. The MMU 110 operates under the control of the CPU 100 and the clock signal to access data stored in the RAM 120. Accordingly, the MMU is linked to the RAM by a plurality of address and data lines 160.
Figure 3 is a schematic block diagram of a part of the central processing unit 100. The CPU 100 employs instruction pipelining, to allow the processing and memory operations to be performed substantially continuously. Typically, while one data processing instruction is being executed, its successor is being decoded and a third instruction is being fetched from memory. This arrangement is referred to as a three-stage execution pipeline.
In the part of the CPU 100 illustrated in Figure 3, data processing instructions which have been fetched from memory and then decoded are passed for execution. The data processing instructions are 32-bit data words, of which bits 28 to 31 form a 4-bit condition field. The remaining bits (bits 0 to 27) define the operation to be performed in response to that instruction and, in some cases an operand on which the operation is to be performed.
The condition field (bits 28 to 31) is passed to a condition tester 200 which compares the bits of the condition field with 16 pre- defined condition codes. The 16 condition codes define the state of one or more of four processor flags 210, referred to as the N, Z, C and V flags. These flags represent a processing state of the CPU 100 generated by previously executed instructions.
The N flag denotes a negative result from the previous arithmetic operation of the CPU 100; the Z flag denotes a zero (equal) result from the previous CPU operation; the C flag denotes whether a carry bit was set during the previous CPU operation; and the V flag indicates an arithmetic overflow occurring during the previous CPU operation.
The relation between the four bits of the condition field and the conditions applied to the N, Z, C and V flags are shown in the list below:
0000 = EQ Z set (equal)
0001 = NE Z clear (not equal) 0 0001100 = = C CSS - C set (unsigned higher or same)
0011 cc C clear (unsigned lower)
0100 = MI N set (negative) 0101 = PL N clear (positive or zero )
0110 = VS V set (overflow)
0111 = c V clear (no overflow)
1000 = HI C set and Z clear (unsigned higher)
1001 = LS C clear or Z set (unsigned lower or same)
1010 GE N set and V set, or N clear and V clear
(greater or equal)
1011 LT N set and V clear, or N clear and V set
(less than)
1100 GT Z clear, and either N set and V set, or
N clear and V clear (greater than)
1101 LE Z set, or N set and V clear, or N clear and V set (less than or equal)
1110 AL always 1111 NV never
The condition tester 200 tests the state of the flags listed above, in dependence on which bits are set in the condition field of the current instruction. This comparison takes place during execution of the current instruction. The current instruction is allowed to complete its execution only if the appropriate flags are set to the states specified by the condition field.
If the always (AL) condition is specified, the instruction will be executed irrespective of the flags. The never (NV) condition code prevents execution of the instruction irrespective of the state of the flags 210.
The condition tester 200 generates an output signal 220 indicating whether the current instruction should be completely executed. The output signal 220 is combined with the abort signal supplied on the abort control line 150 from the MMU 110 to the CPU 100, using an OR-gate 230. The output of the OR-gate 230 is therefore set if either the condition tester 200 indicates that the current instruction should not be completely executed, or the abort signal is asserted by the MMU 110.
An instruction execution unit 240 receives bits 0 to 27 of each instruction, defining the operation to be performed and, in some cases, an operand on which the operation is to be performed. During execution of the current instruction (i.e. after the propagation and processing delays of the condition tester 200 and the OR-gate 230) , the instruction execution unit 240 receives the output of the OR-gate 230 indicating whether the current instruction should be completely executed. If the output of the OR-gate 230 indicates that the current instruction should not be completely executed, the instruction is cancelled without changing the state of any registers or memory locations associated with the apparatus. The condition tester 200 and the instruction execution unit 240 may be of the same form as the corresponding components of the previously proposed ARM 6 processor referred to above.
The result of using the apparatus of Figure 3 is that the stringent timing of the abort signal is greatly relaxed, so that the abort signal relating to a memory access initiated by the preceding instruction may be set at a late stage during execution of each instruction.
If an abort signal is received as a result of a failed instruction fetch operation, then that instruction is simply discarded at a later stage in the three-stage instruction pipeline referred to above.
Data read operations are followed by a non-memory-accessing (internal) processor cycle, to allow time for the data which has been read from memory to be loaded into the appropriate processor register. Accordingly, if an abort signal is received as a result of a failed data read operation, the internal cycle following the read operation allows time for any data supplied from memory by the MMU 110 (which data may well be erroneous) to be ignored and not stored in the intended location (e.g. a processor register). The timing relationship of the instruction execution and the receipt of the abort signal for a data write operation is illustrated in Figure 4, in which a clock signal 300 which controls instruction execution by the CPU 100 is illustrated along with the states of the address bus 130 and the data bus l4θ. In order to initiate the data write operation an address 310 is placed on the address bus by the CPU 100. One half-cycle of the clock signal 300 later, data 320 are placed on the data bus 140 by the CPU 100. Once this has been done, the CPU 100 is able to execute the next instruction during the following cycle of the clock signal 300.
At the same time as the data 320 are placed on the data bus, if an abort signal is received (indicated in Figure 4 as 330) or if the condition tester 200 determines that the next instruction should not be executed (which is determined by the condition tester 200 during the half-cycle after the data has been placed on the data bus and is indicated as 340), execution of the next instruction is abandoned.
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 of the invention as defined by the appended claims.

Claims

1. Data processing apparatus in which successive data processing instructions are executed, said apparatus comprising: memory accessing means for accessing a data memory in response to one or more of said instructions, said memory accessing means comprising means for detecting whether each memory access is invalid; condition test means, responsive to a processing state of said apparatus generated by previously executed instructions and operable during execution of each instruction, for detecting whether that instruction should be executed; and conditional control means, responsive to said memory accessing means and to said condition test means, for preventing complete execution of a current instruction if either said memory accessing means detects that a memory access initiated by a preceding instruction is invalid or said condition test means detects that said current instruction should not be executed.
2. Apparatus according to claim 1, in which: said apparatus comprises one or more processing flags for storing data indicative of a current processing state of said data processing apparatus; each instruction includes a condition code defining a state of said processing flags required for that instruction to be executed; said condition test means is operable to compare a required state of the processing flags defined by said condition code in each instruction with an actual state of said processing flags.
3. Apparatus according to claim 2, comprising four processing flags respectively denoting:
(i) whether a previous data processing operation of said apparatus generated a negative result;
(ii) whether a previous data processing operation of said apparatus generated a zero result; (iϋ) whether a carry bit was set by a previous data processing operation of said apparatus; and
(iv) whether an arithmetic overflow occurred during a previous data processing operation of said apparatus.
4. Apparatus according to any one of claims 1 to 3. in which said memory accessing means comprises: means for transmitting a memory address to said data memory; and means for subsequently transmitting data to said data memory or receiving data from said data memory.
5. Apparatus according to any one of the preceding claims, in which: said memory accessing means is operable to generate an abort control signal to indicate that a memory access is invalid; said condition test means is operable to generate a condition failure control signal to indicate that said current instruction should not be executed; and said apparatus comprises means for combining said abort control signal and said condition failure control signal to generate a combined control signal for supply to said condition control means.
6. Apparatus according to claim 5. in which said means for combining comprises a logical OR gate.
7. Apparatus according t~- any one of the preceding claims, in which data processing operations of the apparatus are controlled by a clock signal.
8. An integrated circuit comprising apparatus according to any one of the preceding claims.
9. A method of data processing in which successive data processing instructions are conditionally executed, said method comprising the steps of: accessing a data memory in response to one or more of said instructions; detecting whether each memory access is invalid; detecting, during execution of each instruction, whether that instruction should be executed, in dependence on a processing state of said apparatus generated by previously executed instructions; and' preventing complete execution of a current instruction if it is detected either that a memory access initiated by a preceding instruction is invalid or that said current instruction should not be executed.
PCT/GB1994/001793 1993-09-23 1994-08-16 Execution of data processing instructions WO1995008801A1 (en)

Priority Applications (6)

Application Number Priority Date Filing Date Title
RU96107418A RU2137182C1 (en) 1993-09-23 1994-08-16 Execution of data processing instruction
EP94923797A EP0721619B1 (en) 1993-09-23 1994-08-16 Execution of data processing instructions
US08/619,647 US5961633A (en) 1993-09-23 1994-08-16 Execution of data processing instructions
KR1019960701507A KR100335785B1 (en) 1993-09-23 1994-08-16 Execution of data processing instructions
JP50961895A JP3553946B2 (en) 1993-09-23 1994-08-16 Execution of data processing instructions
DE69414592T DE69414592T2 (en) 1993-09-23 1994-08-16 EXECUTION OF DATA PROCESSING COMMANDS

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
GB9319662.4 1993-09-23
GB9319662A GB2282245B (en) 1993-09-23 1993-09-23 Execution of data processing instructions

Publications (1)

Publication Number Publication Date
WO1995008801A1 true WO1995008801A1 (en) 1995-03-30

Family

ID=10742425

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/GB1994/001793 WO1995008801A1 (en) 1993-09-23 1994-08-16 Execution of data processing instructions

Country Status (13)

Country Link
US (1) US5961633A (en)
EP (1) EP0721619B1 (en)
JP (1) JP3553946B2 (en)
KR (1) KR100335785B1 (en)
CN (1) CN1099633C (en)
DE (1) DE69414592T2 (en)
GB (1) GB2282245B (en)
IL (1) IL110799A (en)
IN (1) IN189692B (en)
MY (1) MY121544A (en)
RU (1) RU2137182C1 (en)
TW (1) TW332266B (en)
WO (1) WO1995008801A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8056072B2 (en) 2005-10-31 2011-11-08 Microsoft Corporation Rebootless display driver upgrades

Families Citing this family (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH1049368A (en) * 1996-07-30 1998-02-20 Mitsubishi Electric Corp Microporcessor having condition execution instruction
GB2406684B (en) * 2002-12-12 2005-08-24 Advanced Risc Mach Ltd Processing activity masking in a data processing system
US20040230781A1 (en) * 2003-05-16 2004-11-18 Via-Cyrix, Inc. Method and system for predicting the execution of conditional instructions in a processor
US9645822B2 (en) 2011-04-07 2017-05-09 Via Technologies, Inc Conditional store instructions in an out-of-order execution microprocessor
US8880851B2 (en) 2011-04-07 2014-11-04 Via Technologies, Inc. Microprocessor that performs X86 ISA and arm ISA machine language program instructions by hardware translation into microinstructions executed by common execution pipeline
US9274795B2 (en) 2011-04-07 2016-03-01 Via Technologies, Inc. Conditional non-branch instruction prediction
US9336180B2 (en) 2011-04-07 2016-05-10 Via Technologies, Inc. Microprocessor that makes 64-bit general purpose registers available in MSR address space while operating in non-64-bit mode
US9043580B2 (en) 2011-04-07 2015-05-26 Via Technologies, Inc. Accessing model specific registers (MSR) with different sets of distinct microinstructions for instructions of different instruction set architecture (ISA)
US9244686B2 (en) 2011-04-07 2016-01-26 Via Technologies, Inc. Microprocessor that translates conditional load/store instructions into variable number of microinstructions
US9146742B2 (en) 2011-04-07 2015-09-29 Via Technologies, Inc. Heterogeneous ISA microprocessor that preserves non-ISA-specific configuration state when reset to different ISA
WO2012138950A2 (en) 2011-04-07 2012-10-11 Via Technologies, Inc. Conditional load instructions in an out-of-order execution microprocessor
US9292470B2 (en) 2011-04-07 2016-03-22 Via Technologies, Inc. Microprocessor that enables ARM ISA program to access 64-bit general purpose registers written by x86 ISA program
US9128701B2 (en) 2011-04-07 2015-09-08 Via Technologies, Inc. Generating constant for microinstructions from modified immediate field during instruction translation
US9141389B2 (en) 2011-04-07 2015-09-22 Via Technologies, Inc. Heterogeneous ISA microprocessor with shared hardware ISA registers
US8880857B2 (en) 2011-04-07 2014-11-04 Via Technologies, Inc. Conditional ALU instruction pre-shift-generated carry flag propagation between microinstructions in read-port limited register file microprocessor
US8924695B2 (en) 2011-04-07 2014-12-30 Via Technologies, Inc. Conditional ALU instruction condition satisfaction propagation between microinstructions in read-port limited register file microprocessor
US9317288B2 (en) 2011-04-07 2016-04-19 Via Technologies, Inc. Multi-core microprocessor that performs x86 ISA and ARM ISA machine language program instructions by hardware translation into microinstructions executed by common execution pipeline
US9176733B2 (en) 2011-04-07 2015-11-03 Via Technologies, Inc. Load multiple and store multiple instructions in a microprocessor that emulates banked registers
US9032189B2 (en) 2011-04-07 2015-05-12 Via Technologies, Inc. Efficient conditional ALU instruction in read-port limited register file microprocessor
US9378019B2 (en) 2011-04-07 2016-06-28 Via Technologies, Inc. Conditional load instructions in an out-of-order execution microprocessor
US9898291B2 (en) 2011-04-07 2018-02-20 Via Technologies, Inc. Microprocessor with arm and X86 instruction length decoders

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3766527A (en) * 1971-10-01 1973-10-16 Sanders Associates Inc Program control apparatus
US4385365A (en) * 1978-02-13 1983-05-24 Hitachi, Ltd. Data shunting and recovering device

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB1480209A (en) * 1974-07-03 1977-07-20 Data Loop Ltd Digital computers
JPS6247746A (en) * 1985-08-27 1987-03-02 Fujitsu Ltd Interruption control system
JPH01229326A (en) * 1988-03-09 1989-09-13 Toshiba Corp Information processor
JPH01310443A (en) * 1988-06-09 1989-12-14 Nec Corp Information processor
US5202967A (en) * 1988-08-09 1993-04-13 Matsushita Electric Industrial Co., Ltd. Data processing apparatus for performing parallel decoding and parallel execution of a variable word length instruction
JPH0335323A (en) * 1989-06-30 1991-02-15 Toshiba Corp Control system for execution of instruction

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3766527A (en) * 1971-10-01 1973-10-16 Sanders Associates Inc Program control apparatus
US4385365A (en) * 1978-02-13 1983-05-24 Hitachi, Ltd. Data shunting and recovering device

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
L. C. GARCIA ET AL.: "Storage Access-Exception Detection for Pipelined Execution Units", IBM TECHNICAL DISCLOSURE BULLETIN., vol. 25, no. 12, May 1983 (1983-05-01), NEW YORK US, pages 6711 - 12 *
M. MULLER: "ARM 6: A High Performance Low Power Consumption Macrocell", COMPCON SPRING, vol. 93, pages 80 - 87
M. MULLER: "ARM6 a High Performance Low Power Consumption Macrocell", COMPCON SPRING '93, February 1993 (1993-02-01), SAN FRANCISCO, CA, US, pages 80 - 87 *
S. B. FURBER: "VLSI RISC Achitecture and Organization", 1989, MARCEL DEKKER, INC., NEW YORK, US. *
W. C. BRANTLEY ET AL.: "Exception Handling in a Highly Overlapped Machine", IBM TECHNICAL DISCLOSURE BULLETIN., vol. 27, no. 5, October 1984 (1984-10-01), NEW YORK US, pages 2757 - 59 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8056072B2 (en) 2005-10-31 2011-11-08 Microsoft Corporation Rebootless display driver upgrades
US8302089B2 (en) 2005-10-31 2012-10-30 Microsoft Corporation Rebootless display driver upgrades

Also Published As

Publication number Publication date
EP0721619B1 (en) 1998-11-11
KR960705271A (en) 1996-10-09
JP3553946B2 (en) 2004-08-11
IN189692B (en) 2003-04-12
JPH09503876A (en) 1997-04-15
DE69414592T2 (en) 1999-05-06
CN1099633C (en) 2003-01-22
DE69414592D1 (en) 1998-12-17
RU2137182C1 (en) 1999-09-10
US5961633A (en) 1999-10-05
EP0721619A1 (en) 1996-07-17
CN1134193A (en) 1996-10-23
GB9319662D0 (en) 1993-11-10
TW332266B (en) 1998-05-21
MY121544A (en) 2006-02-28
KR100335785B1 (en) 2002-11-30
GB2282245A (en) 1995-03-29
IL110799A0 (en) 1994-11-11
GB2282245B (en) 1998-04-15
IL110799A (en) 1997-09-30

Similar Documents

Publication Publication Date Title
US5961633A (en) Execution of data processing instructions
US4811215A (en) Instruction execution accelerator for a pipelined digital machine with virtual memory
US5509119A (en) Fast comparison method and apparatus for error corrected cache tags
US5515518A (en) Two-level branch prediction cache
US5127091A (en) System for reducing delay in instruction execution by executing branch instructions in separate processor while dispatching subsequent instructions to primary processor
EP0377990B1 (en) Data processing systems
JP2539199B2 (en) Digital processor controller
US5860107A (en) Processor and method for store gathering through merged store operations
US5027270A (en) Processor controlled interface with instruction streaming
EP0380858A2 (en) Method and apparatus for detecting and correcting errors in a pipelined computer system
US6959367B2 (en) System having read-modify-write unit
US5297281A (en) Multiple sequence processor system
JPH05100905A (en) System having hardware support break-point function and method of providing said function
US6760835B1 (en) Instruction branch mispredict streaming
JP3242508B2 (en) Microcomputer
JP3707581B2 (en) Data processing system having self-aligned stack pointer and method thereof
US5678016A (en) Processor and method for managing execution of an instruction which determine subsequent to dispatch if an instruction is subject to serialization
US5732005A (en) Single-precision, floating-point register array for floating-point units performing double-precision operations by emulation
US20020144041A1 (en) Early exception detection
US5784606A (en) Method and system in a superscalar data processing system for the efficient handling of exceptions
EP0374598B1 (en) Control store addressing from multiple sources
US5421026A (en) Data processor for processing instruction after conditional branch instruction at high speed
WO2005033862A2 (en) Method and system for processing a loop of instructions
JPH0418648A (en) Data processor equipped with cache and data access method for the processor
CA1233271A (en) Cache disable for a data processor

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 94194017.9

Country of ref document: CN

AK Designated states

Kind code of ref document: A1

Designated state(s): CN JP KR RU US

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE CH DE DK ES FR GB GR IE IT LU MC NL PT SE

DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
121 Ep: the epo has been informed by wipo that ep was designated in this application
WWE Wipo information: entry into national phase

Ref document number: 1994923797

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 08619647

Country of ref document: US

WWP Wipo information: published in national office

Ref document number: 1994923797

Country of ref document: EP

WWG Wipo information: grant in national office

Ref document number: 1994923797

Country of ref document: EP