WO1999061997A1 - Recirculating register file - Google Patents

Recirculating register file Download PDF

Info

Publication number
WO1999061997A1
WO1999061997A1 PCT/GB1999/000707 GB9900707W WO9961997A1 WO 1999061997 A1 WO1999061997 A1 WO 1999061997A1 GB 9900707 W GB9900707 W GB 9900707W WO 9961997 A1 WO9961997 A1 WO 9961997A1
Authority
WO
WIPO (PCT)
Prior art keywords
registers
register
data
instruction
coprocessor
Prior art date
Application number
PCT/GB1999/000707
Other languages
French (fr)
Inventor
Christopher Neal Hinds
David Vivian Jaggar
David Terrence Matheny
David James Seal
Original Assignee
Arm 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 Arm Limited filed Critical Arm Limited
Priority to EP99907748A priority Critical patent/EP1080421B1/en
Priority to JP2000551329A priority patent/JP2002517038A/en
Priority to IL13924999A priority patent/IL139249A/en
Priority to DE69901556T priority patent/DE69901556T2/en
Priority to PCT/GB1999/000707 priority patent/WO1999061997A1/en
Publication of WO1999061997A1 publication Critical patent/WO1999061997A1/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30003Arrangements for executing specific machine instructions
    • G06F9/3004Arrangements for executing specific machine instructions to perform operations on memory
    • G06F9/30043LOAD or STORE instructions; Clear instruction
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30003Arrangements for executing specific machine instructions
    • G06F9/30007Arrangements for executing specific machine instructions to perform operations on data operands
    • G06F9/30032Movement instructions, e.g. MOVE, SHIFT, ROTATE, SHUFFLE
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30003Arrangements for executing specific machine instructions
    • G06F9/30007Arrangements for executing specific machine instructions to perform operations on data operands
    • G06F9/30036Instructions to perform operations on packed data, e.g. vector, tile or matrix operations
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30003Arrangements for executing specific machine instructions
    • G06F9/30076Arrangements for executing specific machine instructions to perform miscellaneous control operations, e.g. NOP
    • G06F9/30087Synchronisation or serialisation instructions
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30098Register arrangements
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30098Register arrangements
    • G06F9/30105Register structure
    • G06F9/30109Register structure having multiple operands in a single register
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30098Register arrangements
    • G06F9/30105Register structure
    • G06F9/30112Register structure comprising data of variable length
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30098Register arrangements
    • G06F9/3012Organisation of register space, e.g. banked or distributed register file
    • G06F9/3013Organisation of register space, e.g. banked or distributed register file according to data content, e.g. floating-point registers, address registers
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30181Instruction operation extension or modification
    • G06F9/30185Instruction operation extension or modification according to one or more bits in the instruction, e.g. prefix, sub-opcode
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/32Address formation of the next instruction, e.g. by incrementing the instruction counter
    • G06F9/322Address formation of the next instruction, e.g. by incrementing the instruction counter for non-sequential address
    • G06F9/325Address formation of the next instruction, e.g. by incrementing the instruction counter for non-sequential address for loops, e.g. loop detection or loop counter
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/34Addressing or accessing the instruction operand or the result ; Formation of operand address; Addressing modes
    • G06F9/345Addressing or accessing the instruction operand or the result ; Formation of operand address; Addressing modes of multiple operands or results
    • G06F9/3455Addressing or accessing the instruction operand or the result ; Formation of operand address; Addressing modes of multiple operands or results using stride
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/38Concurrent instruction execution, e.g. pipeline, look ahead
    • G06F9/3885Concurrent instruction execution, e.g. pipeline, look ahead using a plurality of independent parallel functional units

Definitions

  • This invention relates to the field of data processing. More particularly, this invention relates to data processing systems having a register bank and supporting vector operations. It is known to provide data processing systems having a register bank and supporting vector operations. Examples of such systems are the Cray 1 and Digital Equipment Corporation MultiTitan processors.
  • the Cray 1 processor has separate vector and scalar register banks. If the opcode of the instruction being executed indicates a vector operation, then a sequence of data values are returned from the vector register bank in dependence upon a length value stored in a length register and a mask stored in a mask register. The length specifies how many data values are in the sequence and mask specifies which data values are returned from among a plurality of data values associated with the vector register indicated in the instruction.
  • the MultiTitan processor has a single register bank the registers of which can act serve either as scalars or vectors.
  • the instruction itself includes flags that indicate whether a register specified is a scalar or a vector and a length field indicating the number of data values within the sequence when a vector register is used.
  • Vector instructions themselves are desirable as they allow code density to be increased since a single instruction can specify a plurality of data processing operations.
  • Digital signal processing such as audio or graphics processing is particularly well suited to exploiting vector operations as there is often a requirement to perform the same operation upon a sequence of related data values, e.g. performing a filter operation by multiply a sequence of signal values by tap coefficients of a digital filter. It is also desirable to perform data processing operations as quickly and efficiently as possible.
  • One way of helping to increase speed and efficiency is to avoid having to reload or reposition data values that have already been stored within the register bank.
  • a problem in achieving this is that instruction code that is able to reuse data values directly tends to be longer and more complex. If more instructions are needed to specify the operation required then this tends to slow down the processing and negate the purpose of seeking to reuse the data values within the register bank.
  • special purpose digital signal processing circuits are often provided with the specific role of supporting a small number of digital signal processing operations.
  • a common technique is to store the data values required within a large memory and then fetch the data values required for each manipulation as needed. The data values need not be reloaded or reposition within the large memory as the order and sequencing of their use is controlled by manipulation of the addresses used to access the large memory.
  • a problem with this approach is that the circuits have to be specifically designed to match the operation being performed and so lack the flexibility and ease of integration with other functions that is provided by the use of a more typical general purpose processor. It is an object of the present invention to provide efficient and fast data processing whilst maintaining the flexibility of a general purpose processor using a register bank and instruction decoder supporting vector operations.
  • the present invention provides an apparatus for processing data, said apparatus comprising: a register bank having a plurality of registers; and an instruction decoder responsive to at least one data processing instruction specifying a vector operation that executes a data processing operation a plurality of times using data values from a sequence of registers within said register bank; wherein said register bank includes at least one subset of registers, said sequence of registers being within said subset; and said instruction decoder controls said sequence of registers to wrap within said subset of registers.
  • register wrapping within a subset (i.e. less than all) of the registers of the register bank allows compact code to be written that reuses data values within the register bank without requiring reloading or moving of the data values.
  • the instruction codes for each use can start at different points within the subset and so act upon the data values in a different order with the hardware taking care of the wrapping needed without having to provide extra instructions to split the vector sequence.
  • performing a vector operation upon a subset of registers that wrap upon themselves allows for the possibility to concurrently perform data transfers upon data values within registers not within the subset.
  • the register wrapping can also be thought of as providing hardware support for a ring (circular) buffer type arrangement where, for example, data is loaded in and multiplied out of the buffer at point that chase each other round and round the buffer.
  • said vector operation executes said data processing operation using a plurality of respective data values from a corresponding plurality of sequences of registers; said register bank contains a plurality of subsets of registers, said plurality of sequences of registers being within respective subsets; and said instruction decoder controls said plurality of sequences of registers to wrap within respective subsets of registers.
  • subsets could be formed of registers from positions intermixed with registers not within the subsets.
  • programming and implementation are made easier when the subsets are a range of consecutively numbered registers.
  • the ranges could be spaced apart within the register bank, but in preferred embodiments the ranges are contiguous as this makes more efficient use of the register space available.
  • the ability of the invention to more effectively use vector operations is complemented in preferred embodiments which further comprise a memory and a transfer controller for controlling transfers of data values between said memory and registers within said register bank, said transfer controller being responsive to multiple transfer instructions to transfer a sequence of data values between said memory and a sequence of registers within said register bank.
  • the ability to transfer data values to and form blocks of registers within the register bank is well matched to the ability of the invention to efficiently use vector operations as it allows a block of registers to be reused several times and then swapped out with a single instruction.
  • the use of a sequence of registers in the vector operation and the division of the register bank into predefined subsets of registers may be efficiently implemented in preferred embodiments in which each range is addressed via an incrementer that wraps between the end points of that range.
  • sequence of registers used in the vector operation could take many forms, e.g. every alternate register within the subset, however the most commonly useful situation is the one in which the sequence is a sequence of consecutive registers.
  • the invention provides a method of processing data, said method comprising the steps of: storing data values within a plurality of registers of a register bank; and in response to at least one data processing instruction specifying a vector operation, executing a data processing operation a plurality of times using data values from a sequence of registers within said register bank; wherein said register bank includes at least one subset of registers, said sequence of registers being within said subset; and during said executing, said sequence of registers wraps within said subset of registers.
  • the technique is particularly useful in efficiently providing FIR filter operations in which tap coefficient values and signal values are reused several times with the relative offset between them changing for each vector operation.
  • Figure 1 schematically illustrates a data processing system
  • Figure 2 illustrates a floating point unit supporting both scalar and vector registers
  • Figure 3 is a flow diagram illustrating how, for single precision operation, it is determined whether a given register is a vector or scalar register
  • Figure 4 is a flow diagram illustrating how, for double precision operation, it is determined whether a given register is a vector or a scalar;
  • Figure 5 illustrates the division of the register bank into subsets with wrapping within each subset during single precision operation;
  • Figure 6 illustrates the division of the register bank into subsets with wrapping within each subset during double precision operation
  • Figures 7A to 7C illustrate a main processor view of a coprocessor instruction, a single and double precision coprocessor view of the coprocessor instruction and a single precision coprocessor view of the coprocessor instruction respectively;
  • Figure 8 illustrates a main processor controlling a single and double precision coprocessor
  • Figure 9 illustrates the main processor controlling a single precision coprocessor
  • Figure 10 illustrates the circuit within the single and double precision coprocessor that determines whether an accept signal should be returned to the main processor for a received coprocessor instruction
  • Figure 11 illustrates the circuit within the single precision coprocessor that determines whether an accept signal should be returned to the main processor for a received coprocessor instruction
  • Figure 12 illustrates undefined instruction exception handling within the main processor
  • Figure 13 is a block diagram illustrating elements of a coprocessor in accordance with preferred embodiments of the present invention
  • Figure 14 is a flow diagram illustrating operation of the register control and instruction issue logic in accordance with preferred embodiments of the present invention
  • Figure 15 provides an example of the contents of the floating point register in accordance with preferred embodiments of the present invention
  • Figure 16 illustrates the register bank within a Cray 1 processor
  • Figure 17 illustrates the register bank within a MultiTitan processor.
  • Figure 1 illustrates a data processing system 22 comprising a main processor 24, a floating point unit coprocessor 26, a cache memory 28, a main memory 30 and an input/output system 32.
  • the main processor 24, the cache memory 28, the main memory 30 and the input/output system 32 are linked via a main bus 34.
  • a coprocessor us 36 links the main processor 24 to the floating point unit coprocessor 26.
  • the main processor 24 (also referred to as the ARM core) executes a stream of data processing instructions that control data processing operations of a general type including interactions with the cache memory 28, the main memory 30 and the input/output system 32. Embedded within the stream of data processing instructions are coprocessor instructions.
  • the main processor 24 recognises these coprocessor instructions as being of a type that should be executed by an attached coprocessor. Accordingly, the main processor 24 issues these coprocessor instructions on the coprocessor bus 36 from where they are received by any attached coprocessors.
  • the floating point unit coprocessor 26 will accept and execute any received coprocessor instructions that it detects are intended for it. This detection is via a coprocessor number field within the coprocessor instruction.
  • FIG. 2 schematically illustrates the floating point unit coprocessor 26 in more detail.
  • the floating point unit coprocessor 26 includes a register bank 38 that is formed of 32 32-bit registers (less shown in Figure 2). These registers can operate individually as single precision registers each storing a 32-bit data value or as pairs that together store a 64-bit data value.
  • a pipelined multiply accumulate unit 40 and a load store control unit 42 are provided within the floating point unit coprocessor 26.
  • the multiply accumulate unit 40 and the load store control unit 42 can operate concurrently with the multiply accumulate unit 40 performing arithmetic operations (that include multiply accumulate operations as well as other operations) upon data values within the register bank 38 whilst the load store control unit 42 transfers data values not being used by the multiply accumulate unit 40 to and from the floating point unit coprocessor 26 via the main processor 24.
  • arithmetic operations that include multiply accumulate operations as well as other operations
  • a coprocessor instruction that is accepted is latched within an instruction register 44.
  • the coprocessor instruction can in this simplified view be considered to be formed of an opcode portion followed by three register specifying fields Rl, R2 and R3 (in fact these fields may be split and spread around differently within a full instruction). These register specifying fields Rl, R2 and R3 respectively correspond to the registers within the register bank 38 that serve as the destination, first source and second source for the data processing operation being performed.
  • a vector control register 46 (which may be part of a larger register serving additional functions) stores a length value and a stride value for the vector operations that may be performed by the floating point unit coprocessor 26.
  • the vector control register 46 may be initialised and updated with length and stride values in response to a vector control register load instruction.
  • the vector length and stride values apply globally within the floating point unit coprocessor 26 thereby allowing these values to be dynamically altered on a global basis without having to resort to self-modifying code.
  • a register control and instruction issue unit 48, the load store control unit 42 and a vector control unit 50 can together be considered to perform a main part of the role of instruction decoder.
  • the register control and instruction issue unit 48 is responsive to the opcode and the three register specifying fields Rl, R2 and R3 and first outputs the initial register access (address) signals to the register bank 38 without performing any decode upon the opcode or needing to use the vector control unit 50. Having direct access to the initial register values in this way assists in achieving a faster implementation. If a vector register is specified, then the vector control unit 50 serves to generate the necessary sequence of register access signals using 3 -bit incrementers (adders) 52.
  • the vector control unit 50 is responsive to the length value and the stride value stored within the vector control register 46 in performing its addressing of the register bank 38.
  • a register scoreboard 54 is provided to perform register locking such that the pipelined multiply accumulate unit 40 and concurrently operating load store control unit 42 do not give rise to any data consistency problems (the register scoreboard 54 may alternatively be considered to be part of the register control and instruction issue unit 48).
  • the opcode within the instruction register 44 specifies the nature of the data processing operation to be performed (e.g. whether the instruction is an add, subtract, multiply, divide, load, store... etc.). This is independent of the vector or scalar nature of the register being specified. This further simplifies the instruction decoding and set-up of the multiply accumulate unit 40.
  • the first register specifying value Rl and the second register specifying value R2 together encode the vector/scalar nature of the operation specified by the opcode.
  • V V op V
  • some instructions e.g. a compare, a compare with zero or an absolute value
  • may have no destination registers e.g. outputs are the condition flags
  • fewer input operands a compare with zero has only one input operand.
  • the register control and instruction issue unit 48 and the vector control unit 50 that together perform the main part of the role of instruction decoder are responsive to the first register specifying field Rl and the second register specifying field R2 to determine and then control the vector/scalar nature of the data processing operation specified. It will be noted that if the length value stored within the vector control register 46 indicates a length of one (corresponding to a stored value of zero), then this can be used as an early indication of a purely scalar operation.
  • Figure 3 shows a flow diagram indicating the processing logic used to decode the vector/scalar nature from the register specifying values in single precision mode.
  • a test is made as to whether the vector length is globally set as one (length value equals zero). If the vector length is one, then all registers are treated as scalars in step 58.
  • V V op V.
  • step 60 is useful in recognising an all scalar operation when mixed vector and scalar instructions are being used. It will also be noticed that when operating in a mixed vector and scalar mode, if the first operand is a scalar, then it may be any of SO to S7, whilst if the first operand is a vector, then it may be any of S8 to S31. Providing three times the number of registers to be available within the register bank for the first operand being a vector is an adaptation to the generally greater number of registers needed to hold sequences of data values when using vector operations.
  • a common operation one may wish to perform is a graphics transformation.
  • the transformation to be performed may be represented by a 4*4 matrix.
  • the operand reuse in such calculations means that it is desirable that the matrix values be stored in registers that may be manipulated as vectors.
  • an input pixel value is usually stored in 4 registers that again should be able to be manipulated as a vector to aid reuse.
  • the output of the matrix operation will usually be scalars (accumulating the separate vector line multiplies) stored in 4 registers. If it is desired to double pump the input and output values, then you arrive at a requirement for 24 (16 + 4 + 4) vector registers and 8 (4 + 4) scalar registers.
  • Figure 4 is a flow diagram corresponding to that of Figure 3, but in this case illustrating double precision mode.
  • the register slots within the register bank 38 act as pairs to store 16 64-bit data values in logical registers DO to D15.
  • the encoding of the vector/scalar nature of the registers is modified from that of Figure 3 in that the tests of steps 60 and 66 now become "Is the destination one of DO to D3?" and "Is the first operand one of DO to D3?” at steps 72 and 74 respectively.
  • Figure 5 illustrates the wrapping of vectors within subsets of the register bank 38.
  • the register bank is split into 4 ranges of registers with addresses SO to S7, S8 to SI 5, S16 to S23 and S24 to S31. These ranges are disjoint and contiguous.
  • the wrapping function for these subsets containing eight registers may be provided by employing 3 -bit incrementers (adders) 52 within the vector control unit 50. In this way, when a subset boundary is crossed, the incrementers will wrap back. This simple implementation is facilitated by the alignment of the subsets on eight word boundaries within the register address space.
  • the first vector operation specifies a start register S2, a vector length of 4 (indicated by a length value within the vector control register 46 of 3) and a stride of one (indicated by a stride value within the vector control register 46 of zero). Accordingly, when an instruction is executed that is decoded to refer to register S2 as a vector with these global vector control parameters set, then the instruction will be executed 4 times respectively using the data values within the registers S2, S3, S4 and S5. As this vector does not cross a subset boundary, there is no vector wrapping.
  • the starting register is SI 4
  • the length is 6
  • the stride is one. This will result in the instruction being executed 6 times starting with register S 14.
  • the next register used will be S 15.
  • the register increments by the stride again, then instead of the register used being SI 6, it will wrap to be register S8.
  • the instruction is then executed further 3 times to complete the full sequence of SI 4, S15, S8, S9, SlO and Sl l.
  • FIG. 5 shows a starting register of S25. a length of 8 and a stride of 2.
  • the first register used will be S25 and this will be followed by S27, S29 and S31 in accordance with the stride value of 2.
  • the next register value will wrap back to the start of the subset, pass over register S24 in view of the stride of 2, and execute the operation using register S25.
  • the incrementers 52 can take the form of 3-bit adders that add the stride to the current value when moving between vector registers. Accordingly, the stride can be adjusted by supplying a different stride value to the adder.
  • Figure 6 illustrates the wrapping of the register bank 38 within double precision mode.
  • the subsets of registers comprises DO to D3, D4 to D7, D8 to Dl 1 and D12 to Dl 5.
  • the minimum value input to the adder serving as the incrementer 52 in double precision mode will be 2 - corresponding to a double precision stride of one.
  • a double precision stride of two will require an input of 4 to the adder.
  • the first example illustrated in Figure 6 has a start register of DO, a length of 4 and a stride one. This will result in a vector register sequence of DO, Dl, D2 and D3. As no subset boundaries are crossed, there is no wrapping in this example.
  • the start register is D15
  • the length is 2
  • the stride is 2. This results in a vector register sequence of D15 and D13.
  • the load store control unit 42 has a 5-bit incrementer at its output and that load/store multiple operations are not subject to the register wrapping applied to vector operations. This enables a single load/store multiple instruction to access as many consecutive registers as it requires.
  • An example of an operation that makes good use of this wrapping arrangement is an FIR filter split into units of 4 signal values and 4 taps. If the syntax R8-R11 op R16-R19 represents the vector operations R8opR16, R9opR17, R10opR18 and Rl 1 opRl 9, then the FIR filter operation may be performed as:
  • Figure 7A schematically illustrates how the main processor 24 views a coprocessor instruction.
  • the main processor uses a bit combination of a field 76 (which may be split) within the instruction to identify the instruction as a coprocessor instruction.
  • a coprocessor instruction includes a coprocessor number field 78 that the coprocessor(s) attached to the main processor use to identify if a particular coprocessor instruction is targeted at them.
  • Different types of coprocessor such as a DSP coprocessor (e.g. the Piccolo coprocessor produced by ARM) or a floating point unit coprocessor, can be allocated different coprocessor numbers and so separately addressed within a single system using the same coprocessor bus 36.
  • the coprocessor instructions also include an opcode that is used by the coprocessor and three 5 -bit fields respectively specifying the destination, first operand and second operand from among the coprocessor registers.
  • the main processor at least partially decodes the coprocessor instruction such that the coprocessor and main processor can together complete the desired data processing operation.
  • the main processor may also be responsive to the data type encoded within the coprocessor number as part of the instruction decode it performs in such circumstances.
  • Figure 7B illustrates how a coprocessor supporting both double and single precision operations interprets a coprocessor instruction it receives.
  • a coprocessor is allocated two adjacent coprocessor numbers and uses the most significant 3 bits of the coprocessor number to identify whether it is the target coprocessor.
  • the least significant bit of the coprocessor number is redundant for the purpose of identifying the target coprocessor and can instead be used to specify the data type to be used in executing that coprocessor instruction.
  • the data type corresponds to the data size being either single or double precision.
  • the number of registers is effectively reduced from 32 to 16. Accordingly, it would be possible to decrease the register field size, but in that case the decode of which register to use would not be available directly from a self-contained field in a known position within the coprocessor instruction and would be dependent upon the decoding of other portions of the coprocessor instruction. This would disadvantageously complicate and possibly slow the operation of the coprocessor.
  • Using the least significant bit of the coprocessor number to encode the data type means that the opcode can be completely independent of data type which also simplifies and speeds its decode.
  • Figure 7C illustrates how a coprocessor supporting only a single data type that is a subset of the data types supported by the Figure 7B coprocessor interpreters the coprocessor instructions.
  • the full coprocessor number is used to determine whether or not to accept the instruction.
  • the main processor 24 can then fall back on undefined instruction exception handling to emulate the operation on the unsupported data type.
  • Figure 8 illustrates a data processing system comprising an ARM core 80 serving as a main processor and communicating via a coprocessor bus 82 with a coprocessor 84 that supports both single and double precision data type.
  • the coprocessor instruction including the coprocessor number, is issued from the ARM core 80 on the coprocessor bus 82 when it is encountered within the instruction stream.
  • the coprocessor 84 then compares the coprocessor number with its own numbers and if a match occurs issues an accept signal back to the ARM core 80. If no accept signal is received, then the ARM core recognises an undefined instruction exception and refers to exception handling code stored in the memory system 86.
  • Figure 9 illustrates the system of Figure 8 modified by replacing the coprocessor 84 with a coprocessor 88 that supports only single precision operations.
  • the coprocessor 88 recognises only a single coprocessor number. Accordingly, double precision coprocessor instructions within the original instruction stream that would be executed by the coprocessor 84 of Figure 8 are not accepted by the single precision coprocessor 88.
  • the undefined exception handling code within the memory system 86 can include a double precision emulation routine.
  • the single precision coprocessor 88 can be smaller and less expensive than the double precision equivalent 84 and a net benefit gained if double position instructions are sufficiently rare.
  • Figure 10 illustrates the instruction latch circuit within the coprocessor 84 that supports both single and double precision instructions and has two adjacent coprocessor numbers.
  • the most significant 3 bits CP#[3:1] of the coprocessor number within the coprocessor instruction are compared with those allocated for that coprocessor 84.
  • this comparison can be achieved by matching the most significant the bits of the coprocessor number CP#[3:1] against binary 101. If a match occurs, then an accept signal is returned to the ARM core 80 and the coprocessor instruction is latched for execution.
  • Figure 11 illustrates the equivalent circuit within the single precision coprocessor 88 of Figure 9. In this case only a single coprocessor number will be recognised and single precision operations used by default. The comparison made in determining whether to accept and latch the coprocessor instruction is between the full 4 bits of the coprocessor number CP#[3:0] and the single embedded coprocessor number of binary 1010.
  • Figure 12 is a flow illustrating how the undefined exception handling routine of the Figure 9 embodiment may be triggered to run the double precision emulation code. This is achieved by detecting (step 90) if the instruction that gave rise to the undefined instruction exception is a coprocessor instruction with a coprocessor number of binary 1011. If yes, then this was intended as a double precision instruction and so can be emulated at step 92 before returning to the main program flow. Other exception types may be detected and handled by further steps if not trapped by step 90.
  • Figure 13 illustrates the use of a format register, FPREG 200, to store information identifying the type of data stored in each 32-bit register, or data slot, of the register bank 220.
  • each data slot can operate individually as a single precision register for storing a 32-bit data value (a data word), or can be paired with another data slot to provide a double precision register for storing a 64-bit data value (2 data words).
  • the FPREG register 200 is arranged to identify whether any particular data slot has single precision or double precision data stored therein.
  • the 32 data slots in the register bank 220 are arranged to provide 16 pairs of data slots. If a first data slot has a single precision data value stored therein, then in preferred embodiments the other data slot in that pair will be arranged to only store a single precision data value, and will not be linked with any other data slot in order to store a double precision data value. This ensures that any particular pair of data slots is arranged to store either two single precision data values, or one double precision data value.
  • This information can be identified by a single bit of information associated with each pair of data slots in the register bank 220, and hence in preferred embodiments the FPREG register 200 is arranged to store 16 bits of information to identify the type of data stored in each pair of data slots of the register bank 220. It will be appreciated that the register FPREG 200 can hence be embodied as a 16-bit register, or, for consistency with other registers within the FPU coprocessor 26, can be embodied as a 32-bit register having 16 spare bits of information.
  • Figure 15 illustrates six pairs of data slots within the register bank 220, which can in accordance with preferred embodiments be used to store six double precision data values or twelve single precision data values.
  • An example of data which may be stored within these data slots is shown in Figure 15, DH representing the 32 most significant bits of a double precision data value, DL indicating the 32 lowest significant bits of a double precision data value, and S representing a single precision data value.
  • the corresponding entries within the FPREG register 200 in accordance with preferred embodiments of the present invention are also illustrated in Figure 15.
  • the value "1" is stored in the FPREG register 200 to indicate that the associated pair of data slots contains a double precision data value
  • the value "0" is used to indicate that at least one of the corresponding pair of data slots contains a single precision data value, or that both data slots are uninitialised.
  • the FPU coprocessor 26 of preferred embodiments may be used to process either single precision or double precision data values, and coprocessor instructions issued by the main processor 24 will identify whether any particular instruction is a single precision or a double precision instruction (see Figure 7B and associated description). If an instruction is accepted by the coprocessor, it will be 5 passed to the register control and instruction issue unit 48 for decoding and execution. If the instruction is a load instruction, the register control and instruction issue logic 48 will instruct the load store control unit 42 to retrieve the identified data from memory, and to store that data in the specified data slots of the register bank 220. At this stage, the coprocessor will know whether single precision or double precision data values are
  • the load store control logic 42 will either pass 32-bit single precision data values, or 64-bit double precision data values, over path 225 to the register bank input logic 230 for storing in the register bank 220.
  • data is also provided to the format register FPREG 200 to enable the necessary bits of information to be added to identify whether each pair of data slots receiving data is storing single precision or double precision data.
  • this data is stored in the format register FPREG 200 before data is loaded into the register bank, so that this information is available to the register bank input logic
  • the internal format of the data in the register bank 220 is the same as the external format, and hence single precision data values are stored as 32-bit data values, and double precision data values are stored as 64-bit data values within the register bank 220. Since the register bank input logic 230 has access to the
  • the register bank input logic 230 merely arranges the data received over path 225 for storing in the appropriate data slot(s) of the register bank 220.
  • the register bank input logic 230 would be arranged to perform the necessary conversion. For example, a number is typically represented as l.abc... multiplied by a base value raised to the power of an exponent.
  • typical single and double precision representations do not use a data bit to represent the 1 to the left of the decimal point, but rather the 1 is taken as implied.
  • the register bank input logic 230 would perform the necessary conversion of the data.
  • the data slots would typically be somewhat bigger than 32 bits in order to accommodate the additional data generated by the register bank input logic 230.
  • the load store control unit 42 may also load data into one or more system registers of the coprocessor 26, for example a user status and control register FPSCR 210.
  • the FPSCR register 210 contains user accessible configuration bits and exception status bits, and is discussed in more detail in the architectural description of the floating point unit provided at the end of the preferred embodiment description.
  • the register control and instruction issue unit 48 receives a store instruction identifying particular data slots in the register bank 220 whose contents are to be stored to memory, then the load store control unit 42 is instructed accordingly, and the necessary data words are read out from the register bank 220 to the load store control unit 42 via the register bank output logic 240.
  • the register bank output logic 240 has access to the FPREG register 200 contents in order to determine whether the data being read out is single or double precision data. It then applies appropriate data conversion to reverse any data conversion applied by the register bank input logic 230, and provides the data to the load store control logic 42 over path 235.
  • the coprocessor 26 can be considered to be operating in a second mode of operation where instructions are applied to double precision data values. Since double precision data values contain an even number of data words, then any store instruction issued in the second mode of operation would typically identify an even number of data slots whose contents are to be stored to memory. However, in accordance with preferred embodiments of the present invention, if an odd number of data slots are specified, then the load store control unit 42 is arranged to read the contents of FPREG register 200 and to first store those contents to memory prior to storing the identified even number of data slots from the register bank 220.
  • the data slots to be transferred are identified by a base address identifying a particular data slot in the register bank, followed by a number indicating the number of data slots (i.e. number of data words), counting from the identified data slot, that are to be stored.
  • the store instruction gives as a base address the first data slot in the register bank 220, and specifies 33 data slots, this will cause the contents of all 32 data slots to be stored to memory, but, since the specified number of data slots is odd, it will also cause the contents of the FPREG register 200 to be stored to memory.
  • a single instruction can be used to store both the contents of the register bank to memory, and the contents of the FPREG register 200 identifying the data types stored within the various data slots of the register bank 220.
  • this technique can be taken one stage further to enable additional system registers, such as the FPSCR register 210, to also be stored to memory, if required, using a single instruction.
  • the FPREG register 200 will be stored to memory in addition to the contents of the 32 data slots in the register bank 220.
  • the load store control unit 42 can be interpreted by the load store control unit 42 as a requirement to also store the contents of the FPSCR register 210 to memory in addition to the contents of FPREG register 200 and the data slots of the register bank 220.
  • the coprocessor may also include further system registers, for example exception registers identifying exceptions that have occurred during processing of instructions by the coprocessor. If a different odd number is identified in a store instruction, for example 37, then this can be interpreted by the load store control unit 42 as a requirement to additionally store the contents of the one or more exception registers in addition to the contents of the FPSCR register 210 the FPREG register 200, and the register bank 220.
  • This technique is particularly useful when the code initiating the store or load instruction is not aware of the register bank content, and the register bank content is only temporarily stored to memory for subsequent retrieval into the register bank. If the code were aware of the register bank content, then it may not be necessary for the contents of FPREG register 200 to also be stored to memory. Typical examples of code which may be unaware of the register bank content are context switch code and procedure call entry and exit routines. In such cases, the contents of the FPREG register 200 can be efficiently stored to memory in addition to the contents of the register bank, and indeed, as discussed above, certain other system registers can also be stored as required.
  • the load store control unit 42 upon receiving a double precision load instruction specifying an odd number of data slots, will be arranged to cause the contents of FPREG register 200 to be loaded into the FPREG register 200, followed by the contents of any system registers indicated by the number of slots identified in the load instruction, followed by an even number of data words to be stored in the specified data slots of the register bank 220.
  • the number of data slots specified in the load instruction is 33, then the FPREG register contents will be loaded into the FPREG register 200, followed by the contents of the 32 data slots.
  • FIG. 14 is a flow diagram illustrating operation of the register control and instruction issue logic 48 in accordance with preferred embodiments of the present invention when executing store and load instructions.
  • the number of data words (which is identical to the number of data slots in preferred embodiments) is read from the instruction, along with the first register number, i.e. the base register, identified in the instruction.
  • the instruction is a double precision instruction, as mentioned previously this information being available to the coprocessor at this stage since the instruction identifies whether it is a double precision or a single precision instruction.
  • step 320 it is determined whether the number of words specified in the instruction is odd. Assuming for the sake of this embodiment that the technique is not used to selectively transfer various system registers in addition to the FPREG register 200, then if the number of words is odd, this will indicate that the contents of the FPREG register 200 should be transferred, and accordingly at step 325, the contents of the FPREG register are transferred by the load store control unit 42. Then, the number of words is decremented by 1 at step 327, and the process proceeds to step 330. If, at step 320, the number of words was determined to be even, then the process proceeds directly to step 330.
  • step 330 it is determined whether the number of words is greater than zero. If not, then the instruction is deemed completed, and the process exits at step 340.
  • step 332 a double precision data value (i.e. the contents of two data slots) is transferred to or from the first specified register number.
  • step 334 the number of words is decremented by 2
  • step 336 the register number is incremented by 1.
  • a register actually consists of two data slots, and hence incrementing the register count by one is equivalent to incrementing the data slot number by 2.
  • step 330 the procedure returns to step 330, where it is determined whether the number of words is still greater than zero, and if so the process is repeated. When the number of words reaches zero, then the process is exited at step 340.
  • step 310 If at step 310, it was determined that the instruction was not a double precision instruction, then the process proceeds to step 350, where it is again determined whether the number of words is greater than zero. If so, the process proceeds to step 352, where a single precision data value is transferred to or from the first register number identified in the instruction. Then, at step 354, the number of words is decremented by one, and at step 356 the register number count is incremented by one so as to point at the next data slot. Then the process returns to step 350, where it is determined whether the number of words is still greater than zero. If so, the process is repeated, until such time as the number of words is equal to zero, at which time the process is exited at step 360.
  • the above approach provides a great deal of flexibility when executing code which is unaware of the register bank contents, for example context switch code or procedure call entry and exit sequences. In these cases, the operating system is not aware of the contents of the registers, and it is desirable not to have to treat the registers differently, dependent on their contents.
  • the above approach allows these code routines to be written with a single store or load instruction specifying an odd number of data words. If the coprocessor requires the use of the register content information, it will interpret the odd number of data words in the instruction as a requirement to also store to memory or load from memory the format information required to identify the contents of the data in the register bank. This flexibility removes the need for unique operating system software to support coprocessors that require the register content information.
  • This technique also removes the necessity for loading and storing the register content information in a separate operation within the code. Since the option to load and store the register content information is incorporated in the instruction, no additional memory access is required. This reduces the code length and potentially saves time.
  • VFPvl is a floating point system (FPS) architecture designed to be implemented as a coprocessor for use with ARM processor modules. Implementations of this architecture may incorporate features in either hardware or software, or an implementation may use software to complement the functionality or provide IEEE 754 compatibility. This specification intends to achieve full IEEE 754 compatibility using a combination of hardware and software support.
  • FPS floating point system
  • VFPvl Two coprocessor numbers are used by VFPvl; 10 is used for operations with single precision operands, while 11 is used for operations with double precision operands. Conversion between single and double precision data is accomplished with 2 conversion instructions which operate in the source operand coprocessor space.
  • VFPvl architecture features of the VFPvl architecture include:
  • Denormal handling option selects between IEEE 754 compatibility (with intended support from the floating point emulation package) or fast flush-to-zero capability.
  • Implementers may choose to implement the VFPvl completely in hardware or utilize a combination of hardware and support code.
  • the VFPvl may be implemented completely in software.
  • ConvertToUnsignedlnteger(Fm) Conversion of the contents in Fm to a unsigned 32-bit integer value.
  • the result is dependent on the rounding mode for final rounding and handling of floating point values outside the range of a 32-bit unsigned integer.
  • the INVALID exception is possible if the floating point input value is negative or too large for a 32-bit unsigned integer.
  • ConvertToSignedlnteger(Fm) Conversion of the contents in Fm to a signed 32-bit integer value.
  • the result is dependent on the rounding mode for final rounding and handling of floating point values outside the range of a 32-bit signed integer.
  • the INVALID exception is possible if the floating point input value is too large for a 32- bit signed integer.
  • Denormalized value A representation of a value in the range (-2 mm ⁇ x ⁇ 2 Emm ).
  • a denormalized value, or denormal has a zero exponent and the leading significand bit is 0 rather than 1.
  • the IEEE 754-1985 specification requires that the generation and manipulation of denormalized operands be performed with the same precision as with normal operands.
  • Disabled exception An exception which has its associated Exception Enable bit in the FPCSR set to 0 is referred to as 'disabled.' For these exceptions the IEEE 754 specification defines the correct result to be returned. An operation which generates an exception condition may bounce to the support code to produce the IEEE 754 defined result. The exception will not be reported to the user exception handler
  • Enabled exception An exception with the respective exception enable bit set to 1. In the event of an occurrence of this exception a trap to the user handler will be taken. An operation which generates an exception condition may bounce to the support code to produce the IEEE 754 defined result. The exception will then be reported to the user exception handler.
  • Exponent - The component of a floating point number that normally signifies the integer power to which two is raised in determining the value of the represented number. Occasionally the exponent is called the signed or unbiased exponent.
  • IEEE 754-1985 "IEEE Standard for Binary Floating-Point Arithmetic", ANSI/IEEE Std 754-1985, The Institute of Electrical and Electronics Engineers, Inc. New York, New York, 10017.
  • the standard often referred to as the IEEE 754 standard, which defines data types, correct operation, exception types and handling, and error bounds for floating point systems.
  • Most processors are built in compliance with the standard in either hardware or a combination of hardware and software.
  • Infinity An IEEE 754 special format used to represent ⁇ . The exponent will be maximum for the precision and the significand will be all zeros.
  • Input exception - An exception condition in which one or more of the operands for a given operation are not supported by the hardware. The operation will bounce to support code for completion of the operation.
  • Intermediate result - An internal format used to store the result of a calculation before rounding. This format may have a larger exponent field and significand field than the destination format.
  • MCR - "Move to Coprocessor from ARM Register" For the FPS this includes instructions which transfer data or control registers between an ARM register and a FPS register. Only 32 bits of information may be transferred using a single MCR class instruction.
  • MRC "Move to ARM Register from Coprocessor" For the FPS this includes instructions which transfer data or control registers between the FPS and an ARM register. Only 32 bits of information may be transferred using a single MRC class instruction.
  • NaN - Not a number a symbolic entity encoded in a floating point format.
  • Signalling NaNs will cause an Invalid Operand exception if used as an operand.
  • Quiet NaNs propagate through almost every arithmetic operation without signalling exceptions.
  • the format for a NaN has the exponent field of all l's with the significand non-zero. To represent a signalling NaN the most significant bit of the fraction is zero, while a quiet NaN will have the bit set to a one.
  • Rounding Mode - The IEEE 754 specification requires all calculations to be performed as if to an infinite precision, that is, a multiply of two single precision values must calculate accurately the significand to twice the number of bits of the significand. To represent this value in the destination precision rounding of the significand is often required.
  • the IEEE 754 standard specifies four rounding modes - round to nearest (RN), round to zero, or chop (RZ), round to plus infinity (RP), and round to minus infinity (RM).
  • the first is accomplished by rounding at the halfway point, with the tie case rounding up if it would zero the lsb of the significand, making it 'even.
  • the second effectively chops any bits to the right of the significand, always rounding down, and is used by the C, C++, and Java languages in integer conversions. The later two modes are used in interval arithmetic.
  • Significand - The component of a binary floating point number that consists of an explicit or implicit leading bit to the left of its implied binary point and a fraction field to the right.
  • Support Code Software which must be used to complement the hardware to provide compatibility with the IEEE 754 standard.
  • the support code is intended to have two components: a library of routines which perform operations beyond the scope of the hardware, such as transcendental computations, as well as supported functions, such as divide with unsupported inputs or inputs which may generate an exception; and a set of exception handlers which process exceptional conditions in order to provide IEEE 754 compliance.
  • the support code is required to perform implemented functions in order to emulate proper handling of any unsupported data type or data representation (e.g., denormal values or decimal datatypes).
  • the routines may be written to utilize the FPS in their intermediate calculations if care is taken to restore the users' state at the exit of the routine.
  • the user's trap handler will be executed.
  • UNPREDICTABLE The result of an instruction or control register field value that cannot be relied upon. UNPREDICTABLE instructions or results must not represent security holes, or halt or hang the processor, or any parts of the system.
  • Unsupported Data Specific data values which are not processed by the hardware but bounced to the support code for completion. These data may include infinities, NaNs, denormal values, and zeros. An implementation is free to select which of these values will be supported in hardware fully or partially, or will require assistance from support code to complete the operation. Any exception resulting from processing unsupported data will be trapped to user code if the corresponding exception enable bit for the exception is set.
  • the architecture provides 32 single precision and 16 double precision registers, all individually addressable within a fully defined 5-bit register index as source or destination operands.
  • the 32 single precision registers are overlapped with the 16 double precision registers, i.e., a write of a double precision data to D5 will overwrite the contents of S 10 and S 11. It is the job of the compiler or the assembly language programmer to be aware of register usage conflicts between the use of a register as a single precision data storage and as half of a double precision data storage in an overlapped implementation. No hardware is provided to insure register use is limited to one precision, and the result is UNPREDICTABLE if this is violated.
  • VFPvl provides access to these registers in a scalar mode, in which one, two, or three operand registers are used to produce a result which is written into a destination register, or in vector mode, in which the operands specified refer to a group of registers.
  • VFPvl supports vector operations for up to eight elements in a single instruction for single precision operands and up to 4 elements for double precision operands.
  • Vector mode is enabled by writing a non-zero value to the LEN field. If the LEN field contains 0, the FPS operates in scalar mode, and the register fields are interpreted as addressing 32 individual single precision registers or 16 double precision registers in a flat register model. If the LEN field is non-zero, the FPS operates in vector mode, and the register fields are as addressing vectors of registers. See Table 1 for encoding of the LEN field.
  • Scalar operations may be specified while in vector mode if the destination register is in the first bank of registers (SO - S7 or DO - D3). See Section 0 for more information.
  • 3.2 Single Precision Register Usage If the LEN field in the FPSCR is 0, 32 single precision registers are available numbered SO through S31. Any of the registers may be used as a source or destination register.
  • Illustration 1 Single Precision Register Map
  • the single precision (coprocessor 10) register map may be drawn as shown in Illustration 1.
  • the register file behaves as 4 banks of 8 circulating registers, as shown in Illustration 2.
  • Double Precision Register Map Any of the registers may be used as a source or destination register.
  • the register map may be drawn as shown in Illustration 3. If the LEN field in the FPSCR is greater than 0, 4 scalar registers and 16 vector registers, in 4 banks of 4 circulating registers, are available as shown in Illustration 4.
  • the registers are addressed as scalars or according to the registers selected for each operand. See Section 0, 3.4 Register Usage, for more information.
  • the double precision registers are circulating within the four banks.
  • OP 2 may be any of the two operand operations supported by the floating point coprocessor;
  • OP 3 may be any of the three operand operations.
  • the 'first bank' of the register file is defined as registers SO - S7 for single precision operations and DO - D3 for double precision operations.
  • Destination and source registers may be any of the scalar registers, 0 through 31 for single precision operations and 0 through 15 for double precision operations. The operation will be performed only on the registers explicitly specified in the instruction.
  • the destination register is in the first bank of the register file.
  • the source scalars may be any of the other registers. This mode allows the intermixing of scalar and vector operations without having to change the LEN field in the FPSCR.
  • the LEN field in the FPSCR is greater than zero, and the destination register is not in the first bank of the register file.
  • the scalar source registers may be any register in the first bank of the register file while any of the remaining registers may be used for VectorB. Note that the behavior is UNPREDICTABLE if the source scalar register is a member of VectorB or if VectorD overlaps VectorB in less then LEN elements. I.e., Vector D and VectorB must be either the same vector or completely distinct in all members. See the summary tables in Section 0.
  • the LEN field in the FPSCR is greater than zero and the destination vector register is not in the first bank of the register file.
  • the individual elements of the VectorA vector are combined with the corresponding element in VectorB and written to VectorD. Any register not in the first bank of the register file is available for Vector A, while all vectors are available for VectorB.
  • the behavior is UNPREDICTABLE if the either of the source vectors and the destination vector overlap in less than LEN elements. They must be identical or completely distinct in all members. See the summary tables in Section 0.
  • FPS instructions may be divided into three categories:
  • the FPS specifies a single instruction that causes the FPS to busy-wait the ARM until all currently executing instructions have completed and the exception status of each is known. If an exception is pending, the serializing instruction will be aborted and exception processing will begin in the ARM.
  • the serializing instructions in the FPS is:
  • the conversion between floating point and integer data is a two step process in the FPS made up of a data transfer instruction involving the integer data and a CDP instruction performing the conversion. If any arithmetic operation is attempted on the integer data in the FPS register while in integer format the results are UNPREDICTABLE and any such operation should be avoided.
  • Integer data may be loaded into a floating point single precision register from either an ARM register, using a MCR FMOVS instruction.
  • the integer data in the FPS register may then be converted into a single or double precision floating point value with the integer-to-float family of operations and written to a destination FPS register.
  • the destination register may be the source register if the integer value is no longer needed.
  • the integer may be a signed or unsigned 32-bit quantity.
  • a value in a FPS single or double precision register may be converted to signed or unsigned 32-bit integer format with the float-to-integer family of instructions.
  • the resulting integer is placed in the destination single precision register.
  • the integer data may be stored to an ARM register using the MRC FMOVS instruction.
  • Instructions operating in single precision space will use the 5 bits available in i the instruction field for operand access.
  • the upper 4 bits are contained in the operand fields labeled Fn, Fm, or Fd; the least significant bit ofthe address is in N, M, or D, respectively.
  • the MCR operations involve the transfer or use of data in ARM registers by the FPS. This includes moving data in single precision format from an ARM register or in double precision format from a pair of ARM registers to an FPS register, loading a signed or unsigned integer value from an ARM register to a single precision FPS register, and loading a control register with the contents of an ARM register.
  • FMOV[S, HD, LD] instructions Only 32-bit data operations are supported by FMOV[S, HD, LD] instructions. Only the data in the ARM register or single precision register is moved by the FMOVS operation. To transfer a double precision operand from 2 ARM registers the FMOVLD and FMOVHD instructions will move the lower half and the upper half, respectively.
  • the MRC operations involve the transfer of data in an FPS register to an ARM register. This includes moving a single precision value or the result of a conversion of a floating point value to integer to an ARM register or a double precision FPS register to two ARM registers, and modifying the status bits ofthe CPSR with the results of a previous floating point compare operation.
  • LDC and STC operations transfer data between the FPS and memory.
  • Floating point data may be transferred in either precision in a single data transfer or in multiple data transfers, with the ARM address register updated or left unaffected. Both full descending stack and empty ascending stack structures are supported, as well as multiple operand access to data structures in the move multiple operations. See Table 10 11 for a description ofthe various options for LDC and STC.
  • Double precision operations Must be set to 0
  • the implementation may write another 32-bit data item or read another 32-bit data item, but is not required to do so. This additional data item may be used to identify the contents of the registers as they are loaded or stored. This is useful in implementations in which the register file format is different from the IEEE 754 format for the precision and each register has type information which is required to identify it in memory. If the offset is odd and the number is greater than the number of single precision registers, this may be used to initiate a context switch ofthe registers and all the system registers.
  • Load/store multiple registers from a starting address in Rn and no modification of Rn may be 1 to 16 for single precision, 1 to 8 for double precision.
  • the offset field contains the number of 32-bit transfers. This mode may be used to load a transform matrix for graphics operations and a point for the transform.
  • FLDMEQS rl2, ⁇ f8-fl 1 ⁇ loads 4 single from the address in rl2 to 4 fp registers s8, s9, slO, and rl2 is unchanged FSTMEQD r4, ⁇ fO ⁇ ; stores one double from dO to the address in r4. r4 is unchanged.
  • Type 1 Transfer Load/Store multiple with post- index of Rn and writeback
  • the offset field is the number of 32-bit transfers.
  • the writeback to Rn is Offset* 4.
  • the maximum number of words transferred in a load multiple is 16.
  • the U bit must be set to 1. This is used for storing into an empty ascending stack or loading from a full descending stack, or storing a transformed point and incrementing the pointer to the next point, and for loading and storing multiple data in a filter operation.
  • Type 2 Transfer Load/Store one register with pre-index or Rn and no writeback
  • Type 3 Transfer Load/Store multiple registers with pre-index and writeback
  • the offset field contains the number of 32-bit transfers.
  • FSTMEQDBS r9! ⁇ f27-f29 ⁇ ; store 3 singles from s27, s28. and s29 to a full descending stack with the last entry address contained in r9. r9 is updated to point to the new last entry.
  • Table 12 lists the allowable combinations for the P. W, and U bits in the LDC/STC opcode and the function of the offset field for each valid operation.
  • CDP instructions include all data processing operations which involve operands from the floating point register file and produce a result which will be written back to the register file.
  • FMAC multiply-accumulate chained
  • This operation differs from fused multiply-accumulate operations in that an IEEE rounding operation is performed on the product before the addition ofthe third operand. This allows Java code to utilize the FMAC operation to speed up multiply- accumulate operations over the separate multiply then add operations.
  • FFTOSIZ[S/D] performs a conversion ofthe contents of a single or double precision to an unsigned integer in a FPS register, using the current rounding mode in the FPSCR.
  • FFTOSI[S/D] performs the conversion to a signed integer.
  • FFTOUIZ[S/D] and FFTOSIZ[S/D] perform the same functions but override the FPSCR rounding mode for the conversion and truncates any fraction bits.
  • the functionality of FFTOSIZ[S/D] is required by C, C++, and Java in float to integer conversions.
  • the FFTOSIZ[S/D] instructions provide this capability without requiring adjustment of the rounding mode bits in the FPSCR to RZ for the conversion, reducing the cycle count for the conversion to only that ofthe FFTOSIZ[S/D] operation, saving 4 to 6 cycles.
  • Compare operations are performed using the CDP CMP instructions followed by a 5 MRC FMOVX FPSCR instruction to load the ARM CPSR flag bits with the resulting FPS flag bits (FPSCR[31:28]).
  • the compare operations are provided with and without the potential for an INVALID exception if one ofthe compare operands is a NaN.
  • the FCMP and FCMPO will not signal the INVALID if one ofthe compare operands is a NaN, while the FCMPE and FCMPEO will signal the exception.
  • the FCMPO and 10 FCMPEO compare the operand in the Fm field with 0 and set the FPS flags accordingly.
  • the ARM flags N, Z, C, and V are defined as follows after a FMOVX FPSCR operation:
  • Table 14 lists the primary opcodes for the CDP instructions. All mnemonics have the form [OPERATION] [COND] [SIO]. Table 14. CDP Opcode Specification
  • Table 15 lists the extended operations available using the Extend value in the opcode field. All instructions have the form [OPERATION] [COND] [S/D] with the exception ofthe serializing and FLSCB instructions.
  • the instruction encoding for the Extended operations is formed in the same way as the index into the register file for the Fn operand, i.e., ⁇ Fn[3:0], N ⁇ . Table 15.
  • Non-vectorizable operations The LEN field is ignored and a scalar operation is performed on the specified registers.
  • the FPSID contains the FPS architecture and implementation-defined identification
  • the FPSCR register contains user accessible configuration bits and the exception status bits.
  • the configuration options include the exception enable bits, rounding control, vector stride and length, handling of denormal operands and results, and the use of debug mode. This register is for user and operating system code to configure
  • Bits 31 through 28 contain the flag values from the most recent compare instruction, and may be accessed using a read of the FPSCR.
  • FPSCR is shown in Illustration 10.
  • Bits 31 through 28 contain the result ofthe most recent compare operation and several 20 control bits useful in specifying the arithmetic response ofthe FPS in special circumstances.
  • the format of the Compare Status and Processing Control Byte are given in Illustration 11.
  • the system control byte controls the rounding mode, vector stride and vector length fields.
  • the bits are specified as shown in Illustration 12.
  • the VFPvl architecture incorporates a register file striding mechanism for use with vector operations. If the STRIDE bits are set to 00, the next register selected in a vector operation will be the register immediately following the previous register in the register file. The normal register file wrapping mechanism is unaffected by the stride value. A STRIDE of 11 will increment all input registers and the output register by 2. For example,
  • FMULEQS F8 FMULEQS F8, F16, F24 FMULEQS F10, F18, F26 FMULEQS F12, F20, F28 FMULEQS F14, F22, F30 effectively 'striding' the operands for the multiply in the register file by 2 rather than by 1 register
  • the exception enable byte occupies bits 15:8 and contains the enables for exception traps.
  • the bits are specified as shown in Illustration 13.
  • the exception enable bits conform to the requirements ofthe IEEE 754 specification for handling of floating point exception conditions. If the bit is set, the exception is enabled, and FPS will signal a user visible trap to the operating system in the event of an occurrence of the exceptional condition on the current instruction. If the bit is cleared, the exception is not enabled, and the FPS will not signal a user visible trap to the operating system in the event ofthe exceptional condition, but will generate a mathematically reasonable result. The default for the exception enable bits is disabled. For more information on exception handling please see the IEEE 754 standard.
  • Some implementations will generate a bounce to the support code to handle exceptional conditions outside the capability ofthe hardware, even when the exception is disabled. This will be generally invisible to user code.
  • the exception status byte occupies bits 7:0 ofthe FPSCR and contains the exception status flag bits. There are five exception status flag bits, one for each floating point exception. These bits are 'sticky'; once set by a detected exception, they must be cleared by a FMOVX write to the FPSCR or a FSERIALCL instruction. The bits are specified as shown in Illustration 14. In the case of an enabled exception, the corresponding exception status bit will not be automatically set. It is the task ofthe support code to set the proper exception status bit as needed. Some exceptions may be automatic, i.e., if the exception condition is detected, the FPS will bounce on the subsequent floating point instruction regardless of how the exception enable bit is set.
  • the FPS allows implementers to select the response including the option to bounce and utilize the support code to produce the correct result and write this value to the destination register. If the underflow exception enable bit is set, the user ' s trap handler will be called after the support code has completed the operation. This code may alter the state ofthe FPS and return, or terminate the process.
  • the Register File Content Register is a privileged register containing information which may be used by a debugger to properly present the contents ofthe register as interpreted by the currently running program.
  • the FPREG contains 16 bits, one bit
  • Illustration 15 FPREG Register Encoding for each double precision register in the register file. If the bit is set, the physical register pair represented by the bit is to be displayed as a double precision register. If the bit is clear, the physical register is uninitialized or contains one or two single precision data values.
  • the FPS operates in one of two modes, a debug mode and a normal mode. If the DM bit is set in the FPSCR, the FPS operates in debug mode. In this mode the FPS executes one instruction at a time while ARM is made to wait until the exception status ofthe instruction is known. This will guarantee the register file and memory are precise with respect to instruction flow, but at the expense of much increased execution time.
  • the FPS will accept a new instruction from the ARM when resources allow, and signal exceptions upon detection ofthe exceptional condition. Exception reporting to the ARM will always be precise with respect to the floating point instruction stream except in the case of a load or store operation which follows a vector operation and executes in parallel with the vector operation. In this case the contents ofthe register file, for load operations, or memory, for store operations, may not be precise.
  • Implementations ofthe FPS may elect to be IEEE 754 compliant with a combination of hardware and software support.
  • the support code will perform the function of compliant hardware and return the result, when appropriate, to the destination register and return to the user's code without calling a user's trap handler or otherwise modifying the flow ofthe user's code. It will appear to the user that the hardware alone was responsible for the processing ofthe floating point code. Bouncing to support code to handle these features significantly increases the time to perform or process the feature, but the incidence of these situations is typically minimal in user code, embedded applications, and well written numeric applications.
  • the support code is intended to have two components: a library of routines which perform operations beyond the scope ofthe hardware, such as transcendental computations, as well as supported functions, such as divide with unsupported inputs or inputs which may generate an exception; and a set of exception handlers which process exception traps in order to provide IEEE 754 compliance.
  • the support code is required to perform implemented functions in order to emulate proper handling of any unsupported data type or data representation (e.g., denormal values).
  • the routines may be written to utilize the FPS in their intermediate calculations if care is taken to restore the users' state at the exit ofthe routine. 6.2 Exception Reporting and Processing
  • Exceptions in normal mode will be reported to the ARM on the next floating point instruction issued after the exception condition is detected.
  • the state ofthe ARM processor, the FPS register file, and memory may not be precise with respect to the offending instruction at the time the exception is taken. Sufficient information is available to the support code to correctly emulate the instruction and process any exception resulting from the instruction.
  • support code may be used to process some or all operations with special IEEE 754 data, including infinities, NaNs, denormal data, and zeros. Implementations which do so will refer to these data as unsupported, and bounce to the support code in a manner generally invisible to user code, and return with the IEEE 754 specified result in the destination register. Any exceptions resulting from the operation will abide by the IEEE 754 rules for exceptions. This may include trapping to user code if the corresponding exception enable bit is set.
  • the IEEE 754 standard defines the response to exceptional conditions for both cases ofthe exception enabled and disabled in the FPSCR.
  • the VFPvl Architecture does not specify the boundary between the hardware and software used to properly comply with the IEEE 754 specification.
  • the FPS does not support any operations with decimal data or conversion to or from decimal data. These operations are required by the IEEE 754 standard and must be provided by the support code. Any attempt to utilize decimal data will require library routines for the desired functions.
  • the FPS has no decimal data type and cannot be used to trap instructions which use decimal data.
  • the FMOVX instruction executed in SUPERVISOR or UNDEFINED mode may read and write the FPSCR or read the FPSID or FPREG when the FPS is in an exceptional state or is disabled (if the implementation supports a disable option) without causing an exception to be signalled to the ARM.

Abstract

A floating point unit having a register bank containing a plurality of registers supports vector operations that execute a specified operation a plurality of times upon a sequence of data values form different registers. The register bank is divided into subsets and with the sequence of registers used in a vector operation wrapping within a subset. The subsets comprise disjoint, contiguous ranges of register numbers. The wrapping within ranges allows compact code and efficient to be provided for performing DSP operations, such as FIR filtering and matrix transformations.

Description

RECIRCULATING REGISTER FILE
This invention relates to the field of data processing. More particularly, this invention relates to data processing systems having a register bank and supporting vector operations. It is known to provide data processing systems having a register bank and supporting vector operations. Examples of such systems are the Cray 1 and Digital Equipment Corporation MultiTitan processors.
The Cray 1 processor has separate vector and scalar register banks. If the opcode of the instruction being executed indicates a vector operation, then a sequence of data values are returned from the vector register bank in dependence upon a length value stored in a length register and a mask stored in a mask register. The length specifies how many data values are in the sequence and mask specifies which data values are returned from among a plurality of data values associated with the vector register indicated in the instruction. The MultiTitan processor has a single register bank the registers of which can act serve either as scalars or vectors. The instruction itself includes flags that indicate whether a register specified is a scalar or a vector and a length field indicating the number of data values within the sequence when a vector register is used.
Vector instructions themselves are desirable as they allow code density to be increased since a single instruction can specify a plurality of data processing operations. Digital signal processing such as audio or graphics processing is particularly well suited to exploiting vector operations as there is often a requirement to perform the same operation upon a sequence of related data values, e.g. performing a filter operation by multiply a sequence of signal values by tap coefficients of a digital filter. It is also desirable to perform data processing operations as quickly and efficiently as possible. One way of helping to increase speed and efficiency is to avoid having to reload or reposition data values that have already been stored within the register bank. A problem in achieving this is that instruction code that is able to reuse data values directly tends to be longer and more complex. If more instructions are needed to specify the operation required then this tends to slow down the processing and negate the purpose of seeking to reuse the data values within the register bank.
As an alternative to the use of general purpose processors such as the Cray 1 and MultiTitan processors, special purpose digital signal processing circuits are often provided with the specific role of supporting a small number of digital signal processing operations. Within these special purpose digital signal processing circuits, a common technique is to store the data values required within a large memory and then fetch the data values required for each manipulation as needed. The data values need not be reloaded or reposition within the large memory as the order and sequencing of their use is controlled by manipulation of the addresses used to access the large memory. A problem with this approach is that the circuits have to be specifically designed to match the operation being performed and so lack the flexibility and ease of integration with other functions that is provided by the use of a more typical general purpose processor. It is an object of the present invention to provide efficient and fast data processing whilst maintaining the flexibility of a general purpose processor using a register bank and instruction decoder supporting vector operations.
Viewed from one aspect the present invention provides an apparatus for processing data, said apparatus comprising: a register bank having a plurality of registers; and an instruction decoder responsive to at least one data processing instruction specifying a vector operation that executes a data processing operation a plurality of times using data values from a sequence of registers within said register bank; wherein said register bank includes at least one subset of registers, said sequence of registers being within said subset; and said instruction decoder controls said sequence of registers to wrap within said subset of registers.
Providing register wrapping within a subset (i.e. less than all) of the registers of the register bank allows compact code to be written that reuses data values within the register bank without requiring reloading or moving of the data values. The instruction codes for each use can start at different points within the subset and so act upon the data values in a different order with the hardware taking care of the wrapping needed without having to provide extra instructions to split the vector sequence. Furthermore, performing a vector operation upon a subset of registers that wrap upon themselves allows for the possibility to concurrently perform data transfers upon data values within registers not within the subset. The register wrapping can also be thought of as providing hardware support for a ring (circular) buffer type arrangement where, for example, data is loaded in and multiplied out of the buffer at point that chase each other round and round the buffer.
Whilst it is possible to have only a single subset of wrapping registers, it is advantageous to provide systems in which said vector operation executes said data processing operation using a plurality of respective data values from a corresponding plurality of sequences of registers; said register bank contains a plurality of subsets of registers, said plurality of sequences of registers being within respective subsets; and said instruction decoder controls said plurality of sequences of registers to wrap within respective subsets of registers.
Within digital signal processing operations it is often required to reuse the data values from two sequences (e.g. a FIR operation with taps and signal values to be multiplied and accumulated at different offsets or a matrix operation) and so multiple subsets of wrapping registers are desirable.
Whilst it is possible for the subsets to overlap, in practice the data values that require reuse in such situations are usually quite separate and so the subsets may be made disjoint. This makes the hardware implementation advantageously less complex.
It will be appreciated that the subsets could be formed of registers from positions intermixed with registers not within the subsets. However, programming and implementation are made easier when the subsets are a range of consecutively numbered registers.
The ranges could be spaced apart within the register bank, but in preferred embodiments the ranges are contiguous as this makes more efficient use of the register space available. The ability of the invention to more effectively use vector operations is complemented in preferred embodiments which further comprise a memory and a transfer controller for controlling transfers of data values between said memory and registers within said register bank, said transfer controller being responsive to multiple transfer instructions to transfer a sequence of data values between said memory and a sequence of registers within said register bank.
The ability to transfer data values to and form blocks of registers within the register bank is well matched to the ability of the invention to efficiently use vector operations as it allows a block of registers to be reused several times and then swapped out with a single instruction. The use of a sequence of registers in the vector operation and the division of the register bank into predefined subsets of registers may be efficiently implemented in preferred embodiments in which each range is addressed via an incrementer that wraps between the end points of that range.
The sequence of registers used in the vector operation could take many forms, e.g. every alternate register within the subset, however the most commonly useful situation is the one in which the sequence is a sequence of consecutive registers.
The above techniques can be used in any processor have a register bank and supporting vector operations. However, the ability to allow compact code and reuse data values within the registers has been found to be particular useful and not interfere with other considerations in embodiments in which the register bank and instruction decoder are within a floating point unit.
Viewed from another aspect the invention provides a method of processing data, said method comprising the steps of: storing data values within a plurality of registers of a register bank; and in response to at least one data processing instruction specifying a vector operation, executing a data processing operation a plurality of times using data values from a sequence of registers within said register bank; wherein said register bank includes at least one subset of registers, said sequence of registers being within said subset; and during said executing, said sequence of registers wraps within said subset of registers.
The technique is particularly useful in efficiently providing FIR filter operations in which tap coefficient values and signal values are reused several times with the relative offset between them changing for each vector operation.
Embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings in which:
Figure 1 schematically illustrates a data processing system;
Figure 2 illustrates a floating point unit supporting both scalar and vector registers;
Figure 3 is a flow diagram illustrating how, for single precision operation, it is determined whether a given register is a vector or scalar register;
Figure 4 is a flow diagram illustrating how, for double precision operation, it is determined whether a given register is a vector or a scalar; Figure 5 illustrates the division of the register bank into subsets with wrapping within each subset during single precision operation;
Figure 6 illustrates the division of the register bank into subsets with wrapping within each subset during double precision operation;
Figures 7A to 7C illustrate a main processor view of a coprocessor instruction, a single and double precision coprocessor view of the coprocessor instruction and a single precision coprocessor view of the coprocessor instruction respectively;
Figure 8 illustrates a main processor controlling a single and double precision coprocessor;
Figure 9 illustrates the main processor controlling a single precision coprocessor;
Figure 10 illustrates the circuit within the single and double precision coprocessor that determines whether an accept signal should be returned to the main processor for a received coprocessor instruction; Figure 11 illustrates the circuit within the single precision coprocessor that determines whether an accept signal should be returned to the main processor for a received coprocessor instruction;
Figure 12 illustrates undefined instruction exception handling within the main processor; Figure 13 is a block diagram illustrating elements of a coprocessor in accordance with preferred embodiments of the present invention;
Figure 14 is a flow diagram illustrating operation of the register control and instruction issue logic in accordance with preferred embodiments of the present invention; Figure 15 provides an example of the contents of the floating point register in accordance with preferred embodiments of the present invention;
Figure 16 illustrates the register bank within a Cray 1 processor; and
Figure 17 illustrates the register bank within a MultiTitan processor.
Figure 1 illustrates a data processing system 22 comprising a main processor 24, a floating point unit coprocessor 26, a cache memory 28, a main memory 30 and an input/output system 32. The main processor 24, the cache memory 28, the main memory 30 and the input/output system 32 are linked via a main bus 34. A coprocessor us 36 links the main processor 24 to the floating point unit coprocessor 26.
In operation, the main processor 24 (also referred to as the ARM core) executes a stream of data processing instructions that control data processing operations of a general type including interactions with the cache memory 28, the main memory 30 and the input/output system 32. Embedded within the stream of data processing instructions are coprocessor instructions. The main processor 24 recognises these coprocessor instructions as being of a type that should be executed by an attached coprocessor. Accordingly, the main processor 24 issues these coprocessor instructions on the coprocessor bus 36 from where they are received by any attached coprocessors. In this case, the floating point unit coprocessor 26 will accept and execute any received coprocessor instructions that it detects are intended for it. This detection is via a coprocessor number field within the coprocessor instruction.
Figure 2 schematically illustrates the floating point unit coprocessor 26 in more detail. The floating point unit coprocessor 26 includes a register bank 38 that is formed of 32 32-bit registers (less shown in Figure 2). These registers can operate individually as single precision registers each storing a 32-bit data value or as pairs that together store a 64-bit data value. Within the floating point unit coprocessor 26 there is provided a pipelined multiply accumulate unit 40 and a load store control unit 42. In appropriate circumstances, the multiply accumulate unit 40 and the load store control unit 42 can operate concurrently with the multiply accumulate unit 40 performing arithmetic operations (that include multiply accumulate operations as well as other operations) upon data values within the register bank 38 whilst the load store control unit 42 transfers data values not being used by the multiply accumulate unit 40 to and from the floating point unit coprocessor 26 via the main processor 24.
Within the floating point unit coprocessor 26, a coprocessor instruction that is accepted is latched within an instruction register 44. The coprocessor instruction can in this simplified view be considered to be formed of an opcode portion followed by three register specifying fields Rl, R2 and R3 (in fact these fields may be split and spread around differently within a full instruction). These register specifying fields Rl, R2 and R3 respectively correspond to the registers within the register bank 38 that serve as the destination, first source and second source for the data processing operation being performed. A vector control register 46 (which may be part of a larger register serving additional functions) stores a length value and a stride value for the vector operations that may be performed by the floating point unit coprocessor 26. The vector control register 46 may be initialised and updated with length and stride values in response to a vector control register load instruction. The vector length and stride values apply globally within the floating point unit coprocessor 26 thereby allowing these values to be dynamically altered on a global basis without having to resort to self-modifying code.
A register control and instruction issue unit 48, the load store control unit 42 and a vector control unit 50 can together be considered to perform a main part of the role of instruction decoder. The register control and instruction issue unit 48 is responsive to the opcode and the three register specifying fields Rl, R2 and R3 and first outputs the initial register access (address) signals to the register bank 38 without performing any decode upon the opcode or needing to use the vector control unit 50. Having direct access to the initial register values in this way assists in achieving a faster implementation. If a vector register is specified, then the vector control unit 50 serves to generate the necessary sequence of register access signals using 3 -bit incrementers (adders) 52. The vector control unit 50 is responsive to the length value and the stride value stored within the vector control register 46 in performing its addressing of the register bank 38. A register scoreboard 54 is provided to perform register locking such that the pipelined multiply accumulate unit 40 and concurrently operating load store control unit 42 do not give rise to any data consistency problems (the register scoreboard 54 may alternatively be considered to be part of the register control and instruction issue unit 48).
The opcode within the instruction register 44 specifies the nature of the data processing operation to be performed (e.g. whether the instruction is an add, subtract, multiply, divide, load, store... etc.). This is independent of the vector or scalar nature of the register being specified. This further simplifies the instruction decoding and set-up of the multiply accumulate unit 40. The first register specifying value Rl and the second register specifying value R2 together encode the vector/scalar nature of the operation specified by the opcode. The three common cases supported by the encoding are S = S * S (e.g. basic random maths as generated by a C compiler from a block of C code), V = V op S (e.g. to scale the elements of a vector) and V = V op V (e.g. matrix operations such as FIR filters and graphics transformations) (note that in this context a "op" indicates a general operation and the syntax is of the form destination = second operand op first operand). It should also be understood that some instructions (e.g. a compare, a compare with zero or an absolute value) may have no destination registers (e.g. outputs are the condition flags) or fewer input operands (a compare with zero has only one input operand). In these cases there is more opcode bit space available to specify options such as vector/scalar nature and the full range of registers could be made available for each operand (e.g. compares may always be fully scalar whatever the register).
The register control and instruction issue unit 48 and the vector control unit 50 that together perform the main part of the role of instruction decoder are responsive to the first register specifying field Rl and the second register specifying field R2 to determine and then control the vector/scalar nature of the data processing operation specified. It will be noted that if the length value stored within the vector control register 46 indicates a length of one (corresponding to a stored value of zero), then this can be used as an early indication of a purely scalar operation.
Figure 3 shows a flow diagram indicating the processing logic used to decode the vector/scalar nature from the register specifying values in single precision mode. At step 56 a test is made as to whether the vector length is globally set as one (length value equals zero). If the vector length is one, then all registers are treated as scalars in step 58. At step 60, a test is made as to whether the destination register Rl is within the range SO to S7. If this is the case, then the operation is all scalar and is of the form S = S op S, as is indicated in step 62. If step 60 returns a no, then the destination is determined to be a vector, as indicated at step 64. If the destination is a vector, then the encoding takes the second operand as also being a vector.
Accordingly, the two possibilities remaining at this stage are V = V op S and V = V op V. These to options are distinguished between by the test at step 66 that determines whether the first operand is one of SO to S7. If this is the case, then the operation is V = V op S, else the operation is V = V op V. These states are recognised in steps 68 and 70 respectively.
It should be noticed that when the vector length is set to one, then all of the 32 registers of the register bank 38 are available to be used as scalars since the scalar nature of the operation will be recognised at step 58 without having to rely upon the test of step 60 that does limit the range of registers that may be used for the destination. The test of step 60 is useful in recognising an all scalar operation when mixed vector and scalar instructions are being used. It will also be noticed that when operating in a mixed vector and scalar mode, if the first operand is a scalar, then it may be any of SO to S7, whilst if the first operand is a vector, then it may be any of S8 to S31. Providing three times the number of registers to be available within the register bank for the first operand being a vector is an adaptation to the generally greater number of registers needed to hold sequences of data values when using vector operations.
It will be appreciated that a common operation one may wish to perform is a graphics transformation. In the general case, the transformation to be performed may be represented by a 4*4 matrix. The operand reuse in such calculations means that it is desirable that the matrix values be stored in registers that may be manipulated as vectors. In the same way, an input pixel value is usually stored in 4 registers that again should be able to be manipulated as a vector to aid reuse. The output of the matrix operation will usually be scalars (accumulating the separate vector line multiplies) stored in 4 registers. If it is desired to double pump the input and output values, then you arrive at a requirement for 24 (16 + 4 + 4) vector registers and 8 (4 + 4) scalar registers.
Figure 4 is a flow diagram corresponding to that of Figure 3, but in this case illustrating double precision mode. As previously mentioned, in double precision mode the register slots within the register bank 38 act as pairs to store 16 64-bit data values in logical registers DO to D15. In this case, the encoding of the vector/scalar nature of the registers is modified from that of Figure 3 in that the tests of steps 60 and 66 now become "Is the destination one of DO to D3?" and "Is the first operand one of DO to D3?" at steps 72 and 74 respectively.
Whilst encoding the vector/scalar nature of the registers within the register specifying fields as described above provides a significant saving in instruction bit space, it does cause some difficulties for non-commutative operations such as subtract and division. Given the register configuration V = V op S, the lack of symmetry between the first and second operands for non-commutative operations can be overcome without additional instructions swapping register values by extending the instruction set to include pairs of opcodes such as SUB, RSUB and DIV, RDIV that represent the two different operand options for non-commutative operations.
Figure 5 illustrates the wrapping of vectors within subsets of the register bank 38. In particular, in single precision mode the register bank is split into 4 ranges of registers with addresses SO to S7, S8 to SI 5, S16 to S23 and S24 to S31. These ranges are disjoint and contiguous. Referring to Figure 2, the wrapping function for these subsets containing eight registers may be provided by employing 3 -bit incrementers (adders) 52 within the vector control unit 50. In this way, when a subset boundary is crossed, the incrementers will wrap back. This simple implementation is facilitated by the alignment of the subsets on eight word boundaries within the register address space.
Returning to Figure 5, a number of vector operations are illustrated to assist understanding of the wrapping of the registers. The first vector operation specifies a start register S2, a vector length of 4 (indicated by a length value within the vector control register 46 of 3) and a stride of one (indicated by a stride value within the vector control register 46 of zero). Accordingly, when an instruction is executed that is decoded to refer to register S2 as a vector with these global vector control parameters set, then the instruction will be executed 4 times respectively using the data values within the registers S2, S3, S4 and S5. As this vector does not cross a subset boundary, there is no vector wrapping.
In the second example, the starting register is SI 4, the length is 6 and the stride is one. This will result in the instruction being executed 6 times starting with register S 14. The next register used will be S 15. When the register increments by the stride again, then instead of the register used being SI 6, it will wrap to be register S8. The instruction is then executed further 3 times to complete the full sequence of SI 4, S15, S8, S9, SlO and Sl l.
The final example of Figure 5 shows a starting register of S25. a length of 8 and a stride of 2. The first register used will be S25 and this will be followed by S27, S29 and S31 in accordance with the stride value of 2. Following the use of register S31, the next register value will wrap back to the start of the subset, pass over register S24 in view of the stride of 2, and execute the operation using register S25. The incrementers 52 can take the form of 3-bit adders that add the stride to the current value when moving between vector registers. Accordingly, the stride can be adjusted by supplying a different stride value to the adder.
Figure 6 illustrates the wrapping of the register bank 38 within double precision mode. In this mode, the subsets of registers comprises DO to D3, D4 to D7, D8 to Dl 1 and D12 to Dl 5. The minimum value input to the adder serving as the incrementer 52 in double precision mode will be 2 - corresponding to a double precision stride of one. A double precision stride of two will require an input of 4 to the adder. The first example illustrated in Figure 6 has a start register of DO, a length of 4 and a stride one. This will result in a vector register sequence of DO, Dl, D2 and D3. As no subset boundaries are crossed, there is no wrapping in this example. In the second example, the start register is D15, the length is 2 and the stride is 2. This results in a vector register sequence of D15 and D13.
Referring to Figure 2, it will be noted the load store control unit 42 has a 5-bit incrementer at its output and that load/store multiple operations are not subject to the register wrapping applied to vector operations. This enables a single load/store multiple instruction to access as many consecutive registers as it requires.
An example of an operation that makes good use of this wrapping arrangement is an FIR filter split into units of 4 signal values and 4 taps. If the syntax R8-R11 op R16-R19 represents the vector operations R8opR16, R9opR17, R10opR18 and Rl 1 opRl 9, then the FIR filter operation may be performed as:
Load 8 taps in R8-R15 and 8 signal values into R16-R23
R8-R1 lopR16-R19 and put results into R24-R27 R9-R12opRl 6-R19 and accumulate the results into R24-R27
R10-R13opR16-R19 and accumulate the results into R24-R27 Rl l-R14opR16-R19 and accumulate the results into R24-R27
Reload R8-R1 1 with new taps
R12-R15opR16-R19 and accumulate the results into R24-R27 R13-R8opR16-R19 and accumulate the results into R24-R27 (R15 - > R8 wrap)
R14-R9opR16-R19 and accumulate the results into R24-R27 (R15 - > R8 wrap)
R15-R10opR16-R19 and accumulate the results into R24-R27 (R15 - > R8 wrap)
Reload Rl 2 to Rl 5 with new taps
When out of taps, reload R16-R19 with new data
R12-R15opR20-R23 and put results in R28-R31 R13-R8opR20-R23 and accumulate results into R28-R31 (R15 - > R8 wrap)
R14-R9opR20-R23 and accumulate results into R28-R31 (R15 - > R8 wrap) R15-R10opR20-R23 and accumulate results into R28-R31 (R15 - > R8 wrap)
The rest as above.
It should be noted from the above that the loads are to different registers from the multiple accumulates and so can take place in parallel (i.e. achieves double buffering).
Figure 7A schematically illustrates how the main processor 24 views a coprocessor instruction. The main processor uses a bit combination of a field 76 (which may be split) within the instruction to identify the instruction as a coprocessor instruction. Within standard ARM processor instruction set, a coprocessor instruction includes a coprocessor number field 78 that the coprocessor(s) attached to the main processor use to identify if a particular coprocessor instruction is targeted at them. Different types of coprocessor, such as a DSP coprocessor (e.g. the Piccolo coprocessor produced by ARM) or a floating point unit coprocessor, can be allocated different coprocessor numbers and so separately addressed within a single system using the same coprocessor bus 36. The coprocessor instructions also include an opcode that is used by the coprocessor and three 5 -bit fields respectively specifying the destination, first operand and second operand from among the coprocessor registers. In some instructions, such as a coprocessor load or store, the main processor at least partially decodes the coprocessor instruction such that the coprocessor and main processor can together complete the desired data processing operation. The main processor may also be responsive to the data type encoded within the coprocessor number as part of the instruction decode it performs in such circumstances.
Figure 7B illustrates how a coprocessor supporting both double and single precision operations interprets a coprocessor instruction it receives. Such a coprocessor is allocated two adjacent coprocessor numbers and uses the most significant 3 bits of the coprocessor number to identify whether it is the target coprocessor. In this way, the least significant bit of the coprocessor number is redundant for the purpose of identifying the target coprocessor and can instead be used to specify the data type to be used in executing that coprocessor instruction. In this example, the data type corresponds to the data size being either single or double precision.
It can be noted that whilst in double precision mode, the number of registers is effectively reduced from 32 to 16. Accordingly, it would be possible to decrease the register field size, but in that case the decode of which register to use would not be available directly from a self-contained field in a known position within the coprocessor instruction and would be dependent upon the decoding of other portions of the coprocessor instruction. This would disadvantageously complicate and possibly slow the operation of the coprocessor. Using the least significant bit of the coprocessor number to encode the data type means that the opcode can be completely independent of data type which also simplifies and speeds its decode.
Figure 7C illustrates how a coprocessor supporting only a single data type that is a subset of the data types supported by the Figure 7B coprocessor interpreters the coprocessor instructions. In this case, the full coprocessor number is used to determine whether or not to accept the instruction. In this way, if a coprocessor instruction is of a data type not supported, then it corresponds to a different coprocessor number and will not be accepted. The main processor 24 can then fall back on undefined instruction exception handling to emulate the operation on the unsupported data type.
Figure 8 illustrates a data processing system comprising an ARM core 80 serving as a main processor and communicating via a coprocessor bus 82 with a coprocessor 84 that supports both single and double precision data type. The coprocessor instruction, including the coprocessor number, is issued from the ARM core 80 on the coprocessor bus 82 when it is encountered within the instruction stream. The coprocessor 84 then compares the coprocessor number with its own numbers and if a match occurs issues an accept signal back to the ARM core 80. If no accept signal is received, then the ARM core recognises an undefined instruction exception and refers to exception handling code stored in the memory system 86. Figure 9 illustrates the system of Figure 8 modified by replacing the coprocessor 84 with a coprocessor 88 that supports only single precision operations. In this case the coprocessor 88 recognises only a single coprocessor number. Accordingly, double precision coprocessor instructions within the original instruction stream that would be executed by the coprocessor 84 of Figure 8 are not accepted by the single precision coprocessor 88. Thus, if it is desired to execute the same code, then the undefined exception handling code within the memory system 86 can include a double precision emulation routine.
It will be noted that whilst the need to emulate double precision instructions will make the execution of these instructions slow, the single precision coprocessor 88 can be smaller and less expensive than the double precision equivalent 84 and a net benefit gained if double position instructions are sufficiently rare.
Figure 10 illustrates the instruction latch circuit within the coprocessor 84 that supports both single and double precision instructions and has two adjacent coprocessor numbers. In this case, the most significant 3 bits CP#[3:1] of the coprocessor number within the coprocessor instruction are compared with those allocated for that coprocessor 84. In this example, if the coprocessor 84 has coprocessor numbers 10 and 11, then this comparison can be achieved by matching the most significant the bits of the coprocessor number CP#[3:1] against binary 101. If a match occurs, then an accept signal is returned to the ARM core 80 and the coprocessor instruction is latched for execution.
Figure 11 illustrates the equivalent circuit within the single precision coprocessor 88 of Figure 9. In this case only a single coprocessor number will be recognised and single precision operations used by default. The comparison made in determining whether to accept and latch the coprocessor instruction is between the full 4 bits of the coprocessor number CP#[3:0] and the single embedded coprocessor number of binary 1010.
Figure 12 is a flow illustrating how the undefined exception handling routine of the Figure 9 embodiment may be triggered to run the double precision emulation code. This is achieved by detecting (step 90) if the instruction that gave rise to the undefined instruction exception is a coprocessor instruction with a coprocessor number of binary 1011. If yes, then this was intended as a double precision instruction and so can be emulated at step 92 before returning to the main program flow. Other exception types may be detected and handled by further steps if not trapped by step 90.
Figure 13 illustrates the use of a format register, FPREG 200, to store information identifying the type of data stored in each 32-bit register, or data slot, of the register bank 220. As mentioned earlier, each data slot can operate individually as a single precision register for storing a 32-bit data value (a data word), or can be paired with another data slot to provide a double precision register for storing a 64-bit data value (2 data words). In accordance with preferred embodiments of the present invention, the FPREG register 200 is arranged to identify whether any particular data slot has single precision or double precision data stored therein.
As illustrated in Figure 13, the 32 data slots in the register bank 220 are arranged to provide 16 pairs of data slots. If a first data slot has a single precision data value stored therein, then in preferred embodiments the other data slot in that pair will be arranged to only store a single precision data value, and will not be linked with any other data slot in order to store a double precision data value. This ensures that any particular pair of data slots is arranged to store either two single precision data values, or one double precision data value. This information can be identified by a single bit of information associated with each pair of data slots in the register bank 220, and hence in preferred embodiments the FPREG register 200 is arranged to store 16 bits of information to identify the type of data stored in each pair of data slots of the register bank 220. It will be appreciated that the register FPREG 200 can hence be embodied as a 16-bit register, or, for consistency with other registers within the FPU coprocessor 26, can be embodied as a 32-bit register having 16 spare bits of information.
Figure 15 illustrates six pairs of data slots within the register bank 220, which can in accordance with preferred embodiments be used to store six double precision data values or twelve single precision data values. An example of data which may be stored within these data slots is shown in Figure 15, DH representing the 32 most significant bits of a double precision data value, DL indicating the 32 lowest significant bits of a double precision data value, and S representing a single precision data value.
The corresponding entries within the FPREG register 200 in accordance with preferred embodiments of the present invention are also illustrated in Figure 15. In accordance with the preferred embodiment, the value "1" is stored in the FPREG register 200 to indicate that the associated pair of data slots contains a double precision data value, and the value "0" is used to indicate that at least one of the corresponding pair of data slots contains a single precision data value, or that both data slots are uninitialised. Hence, if both data slots are uninitialised, if one of the data slots is uninitialised and the other data slot in the pair contains a single precision data value, or if both data slots in the pair contain a single precision data value, then a logic "0'" value will be stored in the corresponding bit of the FPREG register 200.
As mentioned earlier, the FPU coprocessor 26 of preferred embodiments may be used to process either single precision or double precision data values, and coprocessor instructions issued by the main processor 24 will identify whether any particular instruction is a single precision or a double precision instruction (see Figure 7B and associated description). If an instruction is accepted by the coprocessor, it will be 5 passed to the register control and instruction issue unit 48 for decoding and execution. If the instruction is a load instruction, the register control and instruction issue logic 48 will instruct the load store control unit 42 to retrieve the identified data from memory, and to store that data in the specified data slots of the register bank 220. At this stage, the coprocessor will know whether single precision or double precision data values are
10 being retrieved, and the load store control unit 42 will act accordingly. Hence, the load store control logic 42 will either pass 32-bit single precision data values, or 64-bit double precision data values, over path 225 to the register bank input logic 230 for storing in the register bank 220.
In addition to the data being loaded by the load store control unit 42 into the
15 register bank 220, data is also provided to the format register FPREG 200 to enable the necessary bits of information to be added to identify whether each pair of data slots receiving data is storing single precision or double precision data. In preferred embodiments, this data is stored in the format register FPREG 200 before data is loaded into the register bank, so that this information is available to the register bank input logic
20 230.
In preferred embodiments, the internal format of the data in the register bank 220 is the same as the external format, and hence single precision data values are stored as 32-bit data values, and double precision data values are stored as 64-bit data values within the register bank 220. Since the register bank input logic 230 has access to the
25 FPREG format register 200, it knows whether the data it is receiving is single or double precision, and so, in such an embodiment, the register bank input logic 230 merely arranges the data received over path 225 for storing in the appropriate data slot(s) of the register bank 220. However, if in alternative embodiments, the internal representation within the register bank is different to the external format, then the register bank input logic 230 would be arranged to perform the necessary conversion. For example, a number is typically represented as l.abc... multiplied by a base value raised to the power of an exponent. For the sake of efficiency, typical single and double precision representations do not use a data bit to represent the 1 to the left of the decimal point, but rather the 1 is taken as implied. If, for any reason, the internal representation used within the register bank 220 required the 1 to be represented explicitly, then the register bank input logic 230 would perform the necessary conversion of the data. In such embodiments, the data slots would typically be somewhat bigger than 32 bits in order to accommodate the additional data generated by the register bank input logic 230. In addition to loading data values into the register bank 220, the load store control unit 42 may also load data into one or more system registers of the coprocessor 26, for example a user status and control register FPSCR 210. In preferred embodiments, the FPSCR register 210 contains user accessible configuration bits and exception status bits, and is discussed in more detail in the architectural description of the floating point unit provided at the end of the preferred embodiment description. If the register control and instruction issue unit 48 receives a store instruction identifying particular data slots in the register bank 220 whose contents are to be stored to memory, then the load store control unit 42 is instructed accordingly, and the necessary data words are read out from the register bank 220 to the load store control unit 42 via the register bank output logic 240. The register bank output logic 240 has access to the FPREG register 200 contents in order to determine whether the data being read out is single or double precision data. It then applies appropriate data conversion to reverse any data conversion applied by the register bank input logic 230, and provides the data to the load store control logic 42 over path 235. In accordance with the preferred embodiments of the present invention, if the store instruction is a double precision instruction, then the coprocessor 26 can be considered to be operating in a second mode of operation where instructions are applied to double precision data values. Since double precision data values contain an even number of data words, then any store instruction issued in the second mode of operation would typically identify an even number of data slots whose contents are to be stored to memory. However, in accordance with preferred embodiments of the present invention, if an odd number of data slots are specified, then the load store control unit 42 is arranged to read the contents of FPREG register 200 and to first store those contents to memory prior to storing the identified even number of data slots from the register bank 220. Typically the data slots to be transferred are identified by a base address identifying a particular data slot in the register bank, followed by a number indicating the number of data slots (i.e. number of data words), counting from the identified data slot, that are to be stored. Hence, if as an example, the store instruction gives as a base address the first data slot in the register bank 220, and specifies 33 data slots, this will cause the contents of all 32 data slots to be stored to memory, but, since the specified number of data slots is odd, it will also cause the contents of the FPREG register 200 to be stored to memory. By this approach, a single instruction can be used to store both the contents of the register bank to memory, and the contents of the FPREG register 200 identifying the data types stored within the various data slots of the register bank 220. This avoids a separate instruction having to be issued to explicitly store the contents of the FPREG register 200, and hence does not so adversely affect the processing speed during a store to memory or a load from memory process. In further embodiments of the present invention, this technique can be taken one stage further to enable additional system registers, such as the FPSCR register 210, to also be stored to memory, if required, using a single instruction. Hence, considering the example of a register bank 220 having 32 data slots, then, as discussed earlier, if 33 data slots are identified in the store instruction, then the FPREG register 200 will be stored to memory in addition to the contents of the 32 data slots in the register bank 220.
However, if a different odd number exceeding the number of data slots in the register bank is identified, for example 35, then this can be interpreted by the load store control unit 42 as a requirement to also store the contents of the FPSCR register 210 to memory in addition to the contents of FPREG register 200 and the data slots of the register bank 220. The coprocessor may also include further system registers, for example exception registers identifying exceptions that have occurred during processing of instructions by the coprocessor. If a different odd number is identified in a store instruction, for example 37, then this can be interpreted by the load store control unit 42 as a requirement to additionally store the contents of the one or more exception registers in addition to the contents of the FPSCR register 210 the FPREG register 200, and the register bank 220.
This technique is particularly useful when the code initiating the store or load instruction is not aware of the register bank content, and the register bank content is only temporarily stored to memory for subsequent retrieval into the register bank. If the code were aware of the register bank content, then it may not be necessary for the contents of FPREG register 200 to also be stored to memory. Typical examples of code which may be unaware of the register bank content are context switch code and procedure call entry and exit routines. In such cases, the contents of the FPREG register 200 can be efficiently stored to memory in addition to the contents of the register bank, and indeed, as discussed above, certain other system registers can also be stored as required.
Upon receipt of a subsequent load instruction, a similar process is employed. Hence, the load store control unit 42, upon receiving a double precision load instruction specifying an odd number of data slots, will be arranged to cause the contents of FPREG register 200 to be loaded into the FPREG register 200, followed by the contents of any system registers indicated by the number of slots identified in the load instruction, followed by an even number of data words to be stored in the specified data slots of the register bank 220. Hence, considering the earlier discussed example, if the number of data slots specified in the load instruction is 33, then the FPREG register contents will be loaded into the FPREG register 200, followed by the contents of the 32 data slots. Similarly, if the number of data slots specified in the load instruction is 35, then the contents of the FPSCR register 210 will also be loaded into the FPSCR register in addition to the above mentioned contents. Finally, if the number of data slots specified is 37, then the contents of any exception registers will also be loaded into those exception registers in addition to the above mentioned contents. Clearly, it will be appreciated by those skilled in the art that the particular actions associated with particular odd numbers is entirely arbitrary, and can be varied as desired. Figure 14 is a flow diagram illustrating operation of the register control and instruction issue logic 48 in accordance with preferred embodiments of the present invention when executing store and load instructions. Firstly, at step 300, the number of data words (which is identical to the number of data slots in preferred embodiments) is read from the instruction, along with the first register number, i.e. the base register, identified in the instruction. Then, at step 310, it is determined whether the instruction is a double precision instruction, as mentioned previously this information being available to the coprocessor at this stage since the instruction identifies whether it is a double precision or a single precision instruction.
If the instruction is a double precision instruction, then the process proceeds to step 320, where it is determined whether the number of words specified in the instruction is odd. Assuming for the sake of this embodiment that the technique is not used to selectively transfer various system registers in addition to the FPREG register 200, then if the number of words is odd, this will indicate that the contents of the FPREG register 200 should be transferred, and accordingly at step 325, the contents of the FPREG register are transferred by the load store control unit 42. Then, the number of words is decremented by 1 at step 327, and the process proceeds to step 330. If, at step 320, the number of words was determined to be even, then the process proceeds directly to step 330.
At step 330, it is determined whether the number of words is greater than zero. If not, then the instruction is deemed completed, and the process exits at step 340.
However, if the number of words is greater than zero, then the process proceeds to step 332, where a double precision data value (i.e. the contents of two data slots) is transferred to or from the first specified register number. Then, at step 334, the number of words is decremented by 2, and at step 336, the register number is incremented by 1. As discussed earlier, for a double precision instruction, a register actually consists of two data slots, and hence incrementing the register count by one is equivalent to incrementing the data slot number by 2.
Then the procedure returns to step 330, where it is determined whether the number of words is still greater than zero, and if so the process is repeated. When the number of words reaches zero, then the process is exited at step 340.
If at step 310, it was determined that the instruction was not a double precision instruction, then the process proceeds to step 350, where it is again determined whether the number of words is greater than zero. If so, the process proceeds to step 352, where a single precision data value is transferred to or from the first register number identified in the instruction. Then, at step 354, the number of words is decremented by one, and at step 356 the register number count is incremented by one so as to point at the next data slot. Then the process returns to step 350, where it is determined whether the number of words is still greater than zero. If so, the process is repeated, until such time as the number of words is equal to zero, at which time the process is exited at step 360.
The above approach provides a great deal of flexibility when executing code which is unaware of the register bank contents, for example context switch code or procedure call entry and exit sequences. In these cases, the operating system is not aware of the contents of the registers, and it is desirable not to have to treat the registers differently, dependent on their contents. The above approach allows these code routines to be written with a single store or load instruction specifying an odd number of data words. If the coprocessor requires the use of the register content information, it will interpret the odd number of data words in the instruction as a requirement to also store to memory or load from memory the format information required to identify the contents of the data in the register bank. This flexibility removes the need for unique operating system software to support coprocessors that require the register content information.
This technique also removes the necessity for loading and storing the register content information in a separate operation within the code. Since the option to load and store the register content information is incorporated in the instruction, no additional memory access is required. This reduces the code length and potentially saves time.
An architectural description of a floating point unit incorporating the above described techniques is given below:
1. Introduction The VFPvl is a floating point system (FPS) architecture designed to be implemented as a coprocessor for use with ARM processor modules. Implementations of this architecture may incorporate features in either hardware or software, or an implementation may use software to complement the functionality or provide IEEE 754 compatibility. This specification intends to achieve full IEEE 754 compatibility using a combination of hardware and software support.
Two coprocessor numbers are used by VFPvl; 10 is used for operations with single precision operands, while 11 is used for operations with double precision operands. Conversion between single and double precision data is accomplished with 2 conversion instructions which operate in the source operand coprocessor space.
Features of the VFPvl architecture include:
• Full compatibility with IEEE 754 in hardware with support code. • 32 single precision registers, each addressable as a source operand or a destination register.
• 16 double precision registers, each addressable as a source operand or a destination register. (Double precision registers overlap physical single precision registers) • Vector mode provides for a significant increase in floating point code density and concurrency with load and store operations.
• 4 banks of 8 circulating single precision registers or 4 banks of 4 circulating double precision registers to enhance dsp and graphics operations.
• Denormal handling option selects between IEEE 754 compatibility (with intended support from the floating point emulation package) or fast flush-to-zero capability.
• Intended for implementation with a fully pipelined chained multiply-accumulate with IEEE 754 compatible results.
• Fast floating point to integer conversion for C, C++, and Java with the FFTOSIZ instruction.
Implementers may choose to implement the VFPvl completely in hardware or utilize a combination of hardware and support code. The VFPvl may be implemented completely in software.
2. Terminology
This specification uses the following terminology:
Automatic exception - An exceptional condition which will always bounce to the support code regardless of the value of the respective exception enable bit. The choice of which, if any, exceptions are Automatic is an implementation option. See Section 0,
6. Exception Processing.
Bounce - An exception reported to the operating system which will be handled by the support code entirely without calling user trap handlers or otherwise interrupting the normal flow of user code.
CDP - 'Coprocessor Data Processing' For the FPS, CDP operations are arithmetic operations rather than load or store operations.
ConvertToUnsignedlnteger(Fm) - Conversion of the contents in Fm to a unsigned 32-bit integer value. The result is dependent on the rounding mode for final rounding and handling of floating point values outside the range of a 32-bit unsigned integer. The INVALID exception is possible if the floating point input value is negative or too large for a 32-bit unsigned integer.
ConvertToSignedlnteger(Fm) - Conversion of the contents in Fm to a signed 32-bit integer value. The result is dependent on the rounding mode for final rounding and handling of floating point values outside the range of a 32-bit signed integer. The INVALID exception is possible if the floating point input value is too large for a 32- bit signed integer.
ConvertUnsignedIntToSingle/Double(Rd) - Conversion of the contents of an ARM register (Rd), interpreted as a 32-bit unsigned integer, to a single or double precision floating point value. If the destination precision is single, an INEXACT exception is possible in the conversion operation.
ConvertSignedIntToSingle/Double(Rd) - Conversion of the contents of an ARM register (Rd), interpreted as a 32-bit signed integer, to a single or double precision floating point value. If the destination precision is single, an INEXACT exception is possible in the conversion operation. ,
Denormalized value - A representation of a value in the range (-2 mm < x < 2Emm). In the IEEE 754 format for single and double precision operands, a denormalized value, or denormal, has a zero exponent and the leading significand bit is 0 rather than 1. The IEEE 754-1985 specification requires that the generation and manipulation of denormalized operands be performed with the same precision as with normal operands.
Disabled exception - An exception which has its associated Exception Enable bit in the FPCSR set to 0 is referred to as 'disabled.' For these exceptions the IEEE 754 specification defines the correct result to be returned. An operation which generates an exception condition may bounce to the support code to produce the IEEE 754 defined result. The exception will not be reported to the user exception handler
Enabled exception - An exception with the respective exception enable bit set to 1. In the event of an occurrence of this exception a trap to the user handler will be taken. An operation which generates an exception condition may bounce to the support code to produce the IEEE 754 defined result. The exception will then be reported to the user exception handler.
Exponent - The component of a floating point number that normally signifies the integer power to which two is raised in determining the value of the represented number. Occasionally the exponent is called the signed or unbiased exponent.
Fraction - The field of the significand that lies to the right of its implied binary point.
Flush-To-Zero Mode - In this mode all values in the range (-2 mm < x < 2 mm) after rounding are treated as zero, rather than converted to a denormalized value.
High(Fn/Fm) - The upper 32 bits [63:32] of a double precision value as represented in memory.
IEEE 754-1985 - "IEEE Standard for Binary Floating-Point Arithmetic", ANSI/IEEE Std 754-1985, The Institute of Electrical and Electronics Engineers, Inc. New York, New York, 10017. The standard, often referred to as the IEEE 754 standard, which defines data types, correct operation, exception types and handling, and error bounds for floating point systems. Most processors are built in compliance with the standard in either hardware or a combination of hardware and software. Infinity - An IEEE 754 special format used to represent ∞. The exponent will be maximum for the precision and the significand will be all zeros.
Input exception - An exception condition in which one or more of the operands for a given operation are not supported by the hardware. The operation will bounce to support code for completion of the operation.
Intermediate result - An internal format used to store the result of a calculation before rounding. This format may have a larger exponent field and significand field than the destination format.
Low(Fn/Fm) - The lower 32 bits [31 :0] of a double precision value as represented in memory.
MCR - "Move to Coprocessor from ARM Register" For the FPS this includes instructions which transfer data or control registers between an ARM register and a FPS register. Only 32 bits of information may be transferred using a single MCR class instruction.
MRC - "Move to ARM Register from Coprocessor" For the FPS this includes instructions which transfer data or control registers between the FPS and an ARM register. Only 32 bits of information may be transferred using a single MRC class instruction.
NaN - Not a number, a symbolic entity encoded in a floating point format. There are two types of NaNs, signalling and non-signalling, or quiet. Signalling NaNs will cause an Invalid Operand exception if used as an operand. Quiet NaNs propagate through almost every arithmetic operation without signalling exceptions. The format for a NaN has the exponent field of all l's with the significand non-zero. To represent a signalling NaN the most significant bit of the fraction is zero, while a quiet NaN will have the bit set to a one.
Reserved - A field in a control register or instruction format is 'reserved' if the field is to be defined by the implementation or would produce UNPREDICTABLE results if the contents of the field were not zero. These fields are reserved for use in future extensions of the architecture or are implementation specific. All Reserved bits not used by the implementation must be written as zero and will be read as zero.
Rounding Mode - The IEEE 754 specification requires all calculations to be performed as if to an infinite precision, that is, a multiply of two single precision values must calculate accurately the significand to twice the number of bits of the significand. To represent this value in the destination precision rounding of the significand is often required. The IEEE 754 standard specifies four rounding modes - round to nearest (RN), round to zero, or chop (RZ), round to plus infinity (RP), and round to minus infinity (RM). The first is accomplished by rounding at the halfway point, with the tie case rounding up if it would zero the lsb of the significand, making it 'even.' The second effectively chops any bits to the right of the significand, always rounding down, and is used by the C, C++, and Java languages in integer conversions. The later two modes are used in interval arithmetic.
Significand - The component of a binary floating point number that consists of an explicit or implicit leading bit to the left of its implied binary point and a fraction field to the right.
Support Code - Software which must be used to complement the hardware to provide compatibility with the IEEE 754 standard. The support code is intended to have two components: a library of routines which perform operations beyond the scope of the hardware, such as transcendental computations, as well as supported functions, such as divide with unsupported inputs or inputs which may generate an exception; and a set of exception handlers which process exceptional conditions in order to provide IEEE 754 compliance. The support code is required to perform implemented functions in order to emulate proper handling of any unsupported data type or data representation (e.g., denormal values or decimal datatypes). The routines may be written to utilize the FPS in their intermediate calculations if care is taken to restore the users' state at the exit of the routine.
Trap - An exceptional condition which has the respective exception enable bit set in the FPSCR. The user's trap handler will be executed.
UNDEFINED - Indicates an instruction that generates an undefined instruction trap. See the ARM Architectural Reference Manual for more information on ARM exceptions.
UNPREDICTABLE - The result of an instruction or control register field value that cannot be relied upon. UNPREDICTABLE instructions or results must not represent security holes, or halt or hang the processor, or any parts of the system.
Unsupported Data - Specific data values which are not processed by the hardware but bounced to the support code for completion. These data may include infinities, NaNs, denormal values, and zeros. An implementation is free to select which of these values will be supported in hardware fully or partially, or will require assistance from support code to complete the operation. Any exception resulting from processing unsupported data will be trapped to user code if the corresponding exception enable bit for the exception is set.
3. Register File 3.1 Introductory Notes
The architecture provides 32 single precision and 16 double precision registers, all individually addressable within a fully defined 5-bit register index as source or destination operands.
The 32 single precision registers are overlapped with the 16 double precision registers, i.e., a write of a double precision data to D5 will overwrite the contents of S 10 and S 11. It is the job of the compiler or the assembly language programmer to be aware of register usage conflicts between the use of a register as a single precision data storage and as half of a double precision data storage in an overlapped implementation. No hardware is provided to insure register use is limited to one precision, and the result is UNPREDICTABLE if this is violated.
VFPvl provides access to these registers in a scalar mode, in which one, two, or three operand registers are used to produce a result which is written into a destination register, or in vector mode, in which the operands specified refer to a group of registers. VFPvl supports vector operations for up to eight elements in a single instruction for single precision operands and up to 4 elements for double precision operands.
Table 1. LEN Bit Encodings
Figure imgf000035_0001
Figure imgf000036_0001
Vector mode is enabled by writing a non-zero value to the LEN field. If the LEN field contains 0, the FPS operates in scalar mode, and the register fields are interpreted as addressing 32 individual single precision registers or 16 double precision registers in a flat register model. If the LEN field is non-zero, the FPS operates in vector mode, and the register fields are as addressing vectors of registers. See Table 1 for encoding of the LEN field.
A means of mixing scalar and vector operations without changing the LEN field is available through the specification of the destination register. Scalar operations may be specified while in vector mode if the destination register is in the first bank of registers (SO - S7 or DO - D3). See Section 0 for more information. 3.2 Single Precision Register Usage If the LEN field in the FPSCR is 0, 32 single precision registers are available numbered SO through S31. Any of the registers may be used as a source or destination register.
Figure imgf000037_0001
Illustration 1. Single Precision Register Map The single precision (coprocessor 10) register map may be drawn as shown in Illustration 1.
If the LEN field in the FPSCR is greater than 0, the register file behaves as 4 banks of 8 circulating registers, as shown in Illustration 2. The first bank of vector registers, V0 through V7, overlap with scalar registers SO through S7, and are addressed as scalars or vectors! according to the registers selected for each operand. See Section 0, 3.4 Register Usage, for more information.
Figure imgf000038_0001
Illustration 2. Circulating Single Precision Registers For example, if the LEN in the FPSCR is set to 3, referencing vector VI 0 will cause registers SIO, SI 1, S12, and S13 to be involved in a vector operation. Similarly, V22 would involve S22, S23, SI 6, and SI 7 in the operation. When the register file is accessed in vector mode, the register following V7 in order is VO; similarly, V8 follows V15, V16 follows V23, and V24 follows V31.
3.3 Double Precision Register Usage
If the LEN field in the FPSCR is 0, 16 double precision scalar registers are available.
63 063
Figure imgf000038_0002
Illustration 3. Double Precision Register Map Any of the registers may be used as a source or destination register. The register map may be drawn as shown in Illustration 3. If the LEN field in the FPSCR is greater than 0, 4 scalar registers and 16 vector registers, in 4 banks of 4 circulating registers, are available as shown in Illustration 4. The first bank of vector registers, VO through V3, overlap with scalar registers DO through D3. The registers are addressed as scalars or according to the registers selected for each operand. See Section 0, 3.4 Register Usage, for more information.
Figure imgf000039_0001
Illustration 4. Circulating Double Precision Registers
As with the single precision examples in Section 0, the double precision registers are circulating within the four banks.
3.4 Register Usage
Three operations between scalars and vectors are supported: (OP2 may be any of the two operand operations supported by the floating point coprocessor; OP3 may be any of the three operand operations.)
For the following descriptions, the 'first bank' of the register file is defined as registers SO - S7 for single precision operations and DO - D3 for double precision operations.
• ScalarD = OP2 ScalarA or ScalarD = ScalarA OP3 ScalarB or ScalarD = ScalarA * ScalarB + ScalarD
• VectorD = OP2 ScalarA or VectorD = ScalarA OP3 VectorB or VectorD = ScalarA * VectorB + VectorD • VectorD = OP2 VectorA or VectorD = VectorA OP3 VectorB or VectorD = VectorA * VectorB + VectorD
3.4.1 Scalar Operations Two conditions will cause the FPS to operate in scalar mode:
1? The LEN field in the FPSCR is 0. Destination and source registers may be any of the scalar registers, 0 through 31 for single precision operations and 0 through 15 for double precision operations. The operation will be performed only on the registers explicitly specified in the instruction.
2? The destination register is in the first bank of the register file. The source scalars may be any of the other registers. This mode allows the intermixing of scalar and vector operations without having to change the LEN field in the FPSCR.
3.4.2 Operations Involving a Scalar and Vector Source with a Vector Destination
To operate in this mode, the LEN field in the FPSCR is greater than zero, and the destination register is not in the first bank of the register file. The scalar source registers may be any register in the first bank of the register file while any of the remaining registers may be used for VectorB. Note that the behavior is UNPREDICTABLE if the source scalar register is a member of VectorB or if VectorD overlaps VectorB in less then LEN elements. I.e., Vector D and VectorB must be either the same vector or completely distinct in all members. See the summary tables in Section 0.
3.4.3 Operations Involving Only Vector Data
To operate in this mode, the LEN field in the FPSCR is greater than zero and the destination vector register is not in the first bank of the register file. The individual elements of the VectorA vector are combined with the corresponding element in VectorB and written to VectorD. Any register not in the first bank of the register file is available for Vector A, while all vectors are available for VectorB. As in the second case, the behavior is UNPREDICTABLE if the either of the source vectors and the destination vector overlap in less than LEN elements. They must be identical or completely distinct in all members. See the summary tables in Section 0.
Note that for the FMAC family of operations the destination register or vector is always the accumulate register or vector. 3.4.4 Operation Summary Tables The following tables present the register usage options for single and double precision 2 and 3 operand instructions. 'Any' refers to availability of all registers in the precision for the specified operand.
Table 2. Single Precision 3 -Operand Register Usage
Figure imgf000041_0001
Table 3. Single Precision 2-Operand Register Usage
Figure imgf000042_0001
Table 4. Double Precision 3 -Operand Register Usage
Figure imgf000042_0002
Table 5. Double Precision 2-Operand Register Usage
Figure imgf000042_0003
4. Instruction Set
FPS instructions may be divided into three categories:
• MCR and MRC - Transfer operations between the ARM and the FPS • LDC and STC - Load and store operations between the FPS and memory
• CDP - Data processing operations
4.1 Instruction Concurrency
The intent of the FPS architectural specification is concurrency on two levels: pipelined functional units and parallel load/store operation with CDP functions. A significant performance gain is available by supporting load and store operations which do not have register dependencies with currently processing operations to execute in parallel with these operations.
4.2 Instruction Serialization
The FPS specifies a single instruction that causes the FPS to busy-wait the ARM until all currently executing instructions have completed and the exception status of each is known. If an exception is pending, the serializing instruction will be aborted and exception processing will begin in the ARM. The serializing instructions in the FPS is:
• FMOVX - read or write to a floating point system register
Any read or write to a floating point system register will be stalled until the current instructions have completed. An FMOVX to the System ID Register (FPSID) will trigger an exception caused by the preceding floating point instruction. Performing a read/modify/write (using FMOVX) on the User Status and Control Register (FPSCR) can be used to clear the exception status bits (FPSCR[4:0]).
4.3 Conversion involving Integer Data
The conversion between floating point and integer data is a two step process in the FPS made up of a data transfer instruction involving the integer data and a CDP instruction performing the conversion. If any arithmetic operation is attempted on the integer data in the FPS register while in integer format the results are UNPREDICTABLE and any such operation should be avoided.
4.3.1 Conversion of integer data to floating point data in a FPS register
Integer data may be loaded into a floating point single precision register from either an ARM register, using a MCR FMOVS instruction. The integer data in the FPS register may then be converted into a single or double precision floating point value with the integer-to-float family of operations and written to a destination FPS register. The destination register may be the source register if the integer value is no longer needed. The integer may be a signed or unsigned 32-bit quantity.
4.3.2 Conversion of floating point data in an FPS register to integer data
A value in a FPS single or double precision register may be converted to signed or unsigned 32-bit integer format with the float-to-integer family of instructions. The resulting integer is placed in the destination single precision register. The integer data may be stored to an ARM register using the MRC FMOVS instruction.
4.4 Register File Addressing
Instructions operating in single precision space (S = 0) will use the 5 bits available in i the instruction field for operand access. The upper 4 bits are contained in the operand fields labeled Fn, Fm, or Fd; the least significant bit ofthe address is in N, M, or D, respectively.
Instructions operating in double precision space (S = 1) will use only the upper 4 bits ofthe operand address. These 4 bits are contained in the Fn, Fm, and Fd fields. The N, M, and D bits must contain 0 when the corresponding operand field contains an operand address.
4.5 MCR (Move to Coprocessor from ARM Register)
The MCR operations involve the transfer or use of data in ARM registers by the FPS. This includes moving data in single precision format from an ARM register or in double precision format from a pair of ARM registers to an FPS register, loading a signed or unsigned integer value from an ARM register to a single precision FPS register, and loading a control register with the contents of an ARM register.
The format for an MCR instruction is given in Illustration 5. 28 27 24 23 21 20 19 16 15 12 11 8 7 6 5 4 3
Figure imgf000045_0001
Illustration 5. MCR Instruction Format Table 6. MCR Bit Field Definitions
Figure imgf000045_0002
Figure imgf000046_0001
Table 7. MCR Opcode Field Definition
Figure imgf000046_0002
Note: Only 32-bit data operations are supported by FMOV[S, HD, LD] instructions. Only the data in the ARM register or single precision register is moved by the FMOVS operation. To transfer a double precision operand from 2 ARM registers the FMOVLD and FMOVHD instructions will move the lower half and the upper half, respectively.
4.6 MRC (Move to ARM Register from Coprocessor / Compare Floating Registers)
The MRC operations involve the transfer of data in an FPS register to an ARM register. This includes moving a single precision value or the result of a conversion of a floating point value to integer to an ARM register or a double precision FPS register to two ARM registers, and modifying the status bits ofthe CPSR with the results of a previous floating point compare operation.
The format ofthe MRC instruction is shown in Illustration 6.
31 28 27 24 23 21 20 19 16 15 12 11 7 6 5 4 3
Figure imgf000047_0001
Illustration 6. MRC Instruction Format
Table 8. MRC Bit Field Definitions
Figure imgf000047_0002
Bit Definition Field
N Single precision operations:
Destination register lsb Double precision operations:
Must be set to 0 or operation is UNDEFINED System register moves
Reserved
M Reserved
Fn Single precision operations:
Destination register address upper 4 bits Double precision operations:
Destination register address System register moves:
0000 - FPID (Coprocessor ID number)
0001 - FPSCR (User Status and Control Register)
0100 - FPREG (Register File Content Register)
Other register encodings are Reserved and may be different on various implementations.
Fm Reserved
R Reserved
* For the FMOVX FPSCR instruction, if the Rd field contains R15 (1111), the upper 4 bits ofthe CPSR will be updated with the resulting condition codes.
Table 9. MRC Opcode Field Definition
Figure imgf000049_0001
Note: See the Note for MCR FMOV instruction. 4.7 LDC/STC (Load/Store FPS Registers)
5 LDC and STC operations transfer data between the FPS and memory. Floating point data may be transferred in either precision in a single data transfer or in multiple data transfers, with the ARM address register updated or left unaffected. Both full descending stack and empty ascending stack structures are supported, as well as multiple operand access to data structures in the move multiple operations. See Table 10 11 for a description ofthe various options for LDC and STC.
The format ofthe LDC and STC instructions is shown in Illustration 7.
1 28 27 25 24 23 22 21 20 19 16 15 12 1 1 8 7 0
COND 1 1 0 P U D W L Rn Fd 1 0 1 s Offset/Transfer No.
Illustration 7. LDC/STC Instruction Format Table 10. LDC/STC Bit Field Definitions
Bit Definition Field
Pre/Post Indexing (0=post,l=pre)
U Up/Down bit (0=down,l=up)
D Single precision operations: Source/Destination register lsb
Double precision operations: Must be set to 0
W Write-back bit (0=no writeback, 1= writeback)
L Direction bit (0=store, l=load)
Rn ARM Base register encoding
Fd Single precision operations:
Source/Destination register address upper 4 bits Double precision operations:
Source/Destination register address
Operation operand size.
0 - Single precision operands
1 - Double precision operands
Offset/ Unsigned 8-bit offset or number of single
Transfer precision (double the count of double
No. precision registers) registers to transfer for FLDM(IA/DB) and FSTM(IA/DB). The maximum number of words in a transfer is 16, allowing for 16 single precision values or 8 double precision values. 4.7.1 General notes for load and store operations
Loading and storing multiple registers will do so linearly through the register file without wrapping across 4 or 8 register boundaries used by the vector operations. Attempting to load past the end ofthe register file is UNPREDICTABLE.
If the offset for a double load or store multiple contains an odd register count 17 or less, the implementation may write another 32-bit data item or read another 32-bit data item, but is not required to do so. This additional data item may be used to identify the contents of the registers as they are loaded or stored. This is useful in implementations in which the register file format is different from the IEEE 754 format for the precision and each register has type information which is required to identify it in memory. If the offset is odd and the number is greater than the number of single precision registers, this may be used to initiate a context switch ofthe registers and all the system registers.
Table 1 1. Load and Store Addressing Mode Options
Figure imgf000051_0001
W Offset Addressing
Transfer Mode Name
No.
Load/store multiple registers from a starting address in Rn and no modification of Rn. The number of registers may be 1 to 16 for single precision, 1 to 8 for double precision. The offset field contains the number of 32-bit transfers. This mode may be used to load a transform matrix for graphics operations and a point for the transform.
Examples:
FLDMEQS rl2, {f8-fl 1 } ; loads 4 single from the address in rl2 to 4 fp registers s8, s9, slO, and rl2 is unchanged FSTMEQD r4, {fO} ; stores one double from dO to the address in r4. r4 is unchanged.
W Offset/ Addressing Transfer Mode Name
No.
Type 1 Transfer: Load/Store multiple with post- index of Rn and writeback
0 1 Number of FLDM<cond>IA<S/D> Rn!, <register Load/Store registers list> Multiple to transfer FSTM<cond>IA<S/D> Rn!. <register list>
Load/Store multiple registers from a starting address in Rn and writeback of the next address after the last transfer to Rn. The offset field is the number of 32-bit transfers. The writeback to Rn is Offset* 4. The maximum number of words transferred in a load multiple is 16. The U bit must be set to 1. This is used for storing into an empty ascending stack or loading from a full descending stack, or storing a transformed point and incrementing the pointer to the next point, and for loading and storing multiple data in a filter operation.
Example:
FLDMEQIAS rl3!. {fl2-fl5} ; loads 4 singles from the address in rl3 to 4 fp registers si 2, si 3, si 4. and si 5, updating rl3 with the address pointing to the next data the series.
W Offset/ Addressing
Transfer Mode Name
NJL
Type 2 Transfer: Load/Store one register with pre-index or Rn and no writeback
1 0 Offset FLD<cond><S/D> [Rn, #+/-offset], Load/Store
Fd with Offset
FST<cond><S/D> [Rn. #+/-offset], Fd
Load/Store single register with pre-increment of the address in Rn and no writeback. The offset value is Offset*4, and is added (U-1) or subtracted (U=0) from Rn to generate the address. This is useful for operand access into a structure and is the typical method used to access memory for floating point data.
Example:
FSTEQD f4, [r8, #+8] ; Stores a double to d4 from the address in r8 offset b> 32 (8 * 4) bytes. r8 is unchanged.
W Offset/ Addressing Transfer Mode Name fo.
Type 3 Transfer: Load/Store multiple registers with pre-index and writeback
1 Number of FLDM<cond>DB<S/D> Rn!, Load/Store registers <register list> Multiple with to transfer FSTM<cond>DB<S/D> Rn!, <register Pre- list> Decrement
Load/Store multiple registers with pre-decrement of the address in Rn and writeback of the new target address to Rn. The offset field contains the number of 32-bit transfers. The writeback value is the Offset*4. subtracted (U=0) from Rn. This mode is used for storing to a full descending stack or loading from an empty ascending stack.
Example:
FSTMEQDBS r9!,{f27-f29} ; store 3 singles from s27, s28. and s29 to a full descending stack with the last entry address contained in r9. r9 is updated to point to the new last entry.
4.7.2 LDC/STC Operation Summary
Table 12 lists the allowable combinations for the P. W, and U bits in the LDC/STC opcode and the function of the offset field for each valid operation.
Table 12. LDC/STC Operation Summary
Figure imgf000055_0001
Figure imgf000056_0001
4.8 CDP (Coprocessor Data Processing)
CDP instructions include all data processing operations which involve operands from the floating point register file and produce a result which will be written back to the register file. Of special interest is the FMAC (multiply-accumulate chained) operation, an operation performing a multiply on two of the operands and adding a third. This operation differs from fused multiply-accumulate operations in that an IEEE rounding operation is performed on the product before the addition ofthe third operand. This allows Java code to utilize the FMAC operation to speed up multiply- accumulate operations over the separate multiply then add operations.
Two instructions in the CDP group are useful in conversion of a floating point value in a FPS register to its integer value. FFTOUI[S/D] performs a conversion ofthe contents of a single or double precision to an unsigned integer in a FPS register, using the current rounding mode in the FPSCR. FFTOSI[S/D] performs the conversion to a signed integer. FFTOUIZ[S/D] and FFTOSIZ[S/D] perform the same functions but override the FPSCR rounding mode for the conversion and truncates any fraction bits. The functionality of FFTOSIZ[S/D] is required by C, C++, and Java in float to integer conversions. The FFTOSIZ[S/D] instructions provide this capability without requiring adjustment of the rounding mode bits in the FPSCR to RZ for the conversion, reducing the cycle count for the conversion to only that ofthe FFTOSIZ[S/D] operation, saving 4 to 6 cycles.
Compare operations are performed using the CDP CMP instructions followed by a 5 MRC FMOVX FPSCR instruction to load the ARM CPSR flag bits with the resulting FPS flag bits (FPSCR[31:28]). The compare operations are provided with and without the potential for an INVALID exception if one ofthe compare operands is a NaN. The FCMP and FCMPO will not signal the INVALID if one ofthe compare operands is a NaN, while the FCMPE and FCMPEO will signal the exception. The FCMPO and 10 FCMPEO compare the operand in the Fm field with 0 and set the FPS flags accordingly. The ARM flags N, Z, C, and V are defined as follows after a FMOVX FPSCR operation:
N Less than
15 Z Equal
C Greater Than or Equal or Unordered
V Unordered
The format ofthe CDP instruction is shown in Illustration 8.
28 27 24 23 22 21 20 19 16 15 12 11 7 6 5 4 3
Figure imgf000057_0001
0 pcode[3 Opcode[2 1] 0 pcode[0]
Illustration 8. CDP Instruction Format
20
Table 13. CDP Bit Field Definitions
Bit Field Definition
Opcode 4-bit FPS operation code (See Table 14)
Figure imgf000058_0001
Bit Field Definition.
Fm Single precision operations:
Source B register address upper 4 bits Double precision operations:
Source B register address
4.8.1 Opcodes
Table 14 lists the primary opcodes for the CDP instructions. All mnemonics have the form [OPERATION] [COND] [SIO]. Table 14. CDP Opcode Specification
Figure imgf000059_0001
Figure imgf000060_0001
4.8.2 Extended Operations
Table 15 lists the extended operations available using the Extend value in the opcode field. All instructions have the form [OPERATION] [COND] [S/D] with the exception ofthe serializing and FLSCB instructions. The instruction encoding for the Extended operations is formed in the same way as the index into the register file for the Fn operand, i.e., {Fn[3:0], N}. Table 15. CDP Extended Operations
Figure imgf000060_0002
Figure imgf000061_0001
Non-vectorizable operations. The LEN field is ignored and a scalar operation is performed on the specified registers.
5. System Registers
5.1 System ID Register (FPSID)
5 The FPSID contains the FPS architecture and implementation-defined identification
1 24 23 16 15 4 3 0
Implementor Architecture version Part number Revision
Illustration 9. FPSID Register Encoding
value. This word may be used to determine the model, feature set and revision ofthe FPS and the mask set number. The FPSID is read only and writes to the FPSID are ignored. See Illustration 9 for the FPSID register layout. 5.2 User Status and Control Register (FPSCR) 10 The FPSCR register contains user accessible configuration bits and the exception status bits. The configuration options include the exception enable bits, rounding control, vector stride and length, handling of denormal operands and results, and the use of debug mode. This register is for user and operating system code to configure
3 1 30 29 28 27 26 25 24 23 22 21 20 19 1 8 16 15 14 13 12 1 1 10 9 8 7 6 5 4 3 2 1 0 z c R M OD E ST RIDE R ES ER VED R ES ER VE D
Illustration 10. User Status and Control Register (FPSCR) the FPS and interrogate the status of completed operations. It must be saved and
15 restored during a context switch. Bits 31 through 28 contain the flag values from the most recent compare instruction, and may be accessed using a read of the FPSCR. The
FPSCR is shown in Illustration 10.
5.2.1 Compare Status and Processing Control Byte
Bits 31 through 28 contain the result ofthe most recent compare operation and several 20 control bits useful in specifying the arithmetic response ofthe FPS in special circumstances. The format of the Compare Status and Processing Control Byte are given in Illustration 11.
31 30 29 28 27 26 25 24
N Z C V R R R FZ
Illustration 11. FPSCR Compare Status and Processing Control Byte
Table 16. FPSCR Compare Status and Processing Control Byte Field Definitions
Figure imgf000063_0001
5,2.2 System Control Byte
The system control byte controls the rounding mode, vector stride and vector length fields. The bits are specified as shown in Illustration 12. The VFPvl architecture incorporates a register file striding mechanism for use with vector operations. If the STRIDE bits are set to 00, the next register selected in a vector operation will be the register immediately following the previous register in the register file. The normal register file wrapping mechanism is unaffected by the stride value. A STRIDE of 11 will increment all input registers and the output register by 2. For example,
FMULEQS F8, F16, F24
will perform the following non- vector operations: FMULEQS F8, F16, F24 FMULEQS F10, F18, F26 FMULEQS F12, F20, F28 FMULEQS F14, F22, F30 effectively 'striding' the operands for the multiply in the register file by 2 rather than by 1 register
23 22 21 20 19 18 16
Figure imgf000064_0001
Illustration 12. FPSCR System Control Byte
Table 17. FPSCR System Control Byte Field Definitions
Figure imgf000065_0001
5.2.3 Exception Enable Byte The exception enable byte occupies bits 15:8 and contains the enables for exception traps. The bits are specified as shown in Illustration 13. The exception enable bits conform to the requirements ofthe IEEE 754 specification for handling of floating point exception conditions. If the bit is set, the exception is enabled, and FPS will signal a user visible trap to the operating system in the event of an occurrence of the exceptional condition on the current instruction. If the bit is cleared, the exception is not enabled, and the FPS will not signal a user visible trap to the operating system in the event ofthe exceptional condition, but will generate a mathematically reasonable result. The default for the exception enable bits is disabled. For more information on exception handling please see the IEEE 754 standard.
Some implementations will generate a bounce to the support code to handle exceptional conditions outside the capability ofthe hardware, even when the exception is disabled. This will be generally invisible to user code.
15 13 12 11 10
Figure imgf000066_0001
Illustration 13. FPSCR Exception Enable Table 18. FPSCR Eχβ δon Enable Byte Fields
Figure imgf000066_0002
Figure imgf000067_0001
5.2.4 Exception Status Byte
The exception status byte occupies bits 7:0 ofthe FPSCR and contains the exception status flag bits. There are five exception status flag bits, one for each floating point exception. These bits are 'sticky'; once set by a detected exception, they must be cleared by a FMOVX write to the FPSCR or a FSERIALCL instruction. The bits are specified as shown in Illustration 14. In the case of an enabled exception, the corresponding exception status bit will not be automatically set. It is the task ofthe support code to set the proper exception status bit as needed. Some exceptions may be automatic, i.e., if the exception condition is detected, the FPS will bounce on the subsequent floating point instruction regardless of how the exception enable bit is set. This allows some ofthe more involved exception processing required by the IEEE 754 standard to be performed in software rather than in hardware. An example would be an underflow condition with the FZ bit set to 0. In this case, the correct result may be a denormalized number depending on the exponent ofthe result and the rounding mode. The FPS allows implementers to select the response including the option to bounce and utilize the support code to produce the correct result and write this value to the destination register. If the underflow exception enable bit is set, the user's trap handler will be called after the support code has completed the operation. This code may alter the state ofthe FPS and return, or terminate the process.
7 5 4 3 1 0
Figure imgf000068_0001
Illustration 14. FPSCR Exception Status Byte
Table 19. FPSCR Exception Status Byte Field Definitions
Figure imgf000068_0002
5.3 Register File Content Register (FPREG)
The Register File Content Register is a privileged register containing information which may be used by a debugger to properly present the contents ofthe register as interpreted by the currently running program. The FPREG contains 16 bits, one bit
16 15 14 13 12 1 1 10 9 8 7 6 5 4 3 2 1 0
Reserved C 15 C I4 C 13 C I2 C M C IO C9 c» C 7 C 6 C 5 C 4 C 3 C 2 c i CD
Illustration 15. FPREG Register Encoding for each double precision register in the register file. If the bit is set, the physical register pair represented by the bit is to be displayed as a double precision register. If the bit is clear, the physical register is uninitialized or contains one or two single precision data values.
Table 20. FPREG Bit Field Definitions
Figure imgf000069_0001
Figure imgf000070_0001
6. Exception Processing
The FPS operates in one of two modes, a debug mode and a normal mode. If the DM bit is set in the FPSCR, the FPS operates in debug mode. In this mode the FPS executes one instruction at a time while ARM is made to wait until the exception status ofthe instruction is known. This will guarantee the register file and memory are precise with respect to instruction flow, but at the expense of much increased execution time. The FPS will accept a new instruction from the ARM when resources allow, and signal exceptions upon detection ofthe exceptional condition. Exception reporting to the ARM will always be precise with respect to the floating point instruction stream except in the case of a load or store operation which follows a vector operation and executes in parallel with the vector operation. In this case the contents ofthe register file, for load operations, or memory, for store operations, may not be precise. 6.1 Support Code
Implementations ofthe FPS may elect to be IEEE 754 compliant with a combination of hardware and software support. For unsupported data types and automatic exceptions, the support code will perform the function of compliant hardware and return the result, when appropriate, to the destination register and return to the user's code without calling a user's trap handler or otherwise modifying the flow ofthe user's code. It will appear to the user that the hardware alone was responsible for the processing ofthe floating point code. Bouncing to support code to handle these features significantly increases the time to perform or process the feature, but the incidence of these situations is typically minimal in user code, embedded applications, and well written numeric applications.
The support code is intended to have two components: a library of routines which perform operations beyond the scope ofthe hardware, such as transcendental computations, as well as supported functions, such as divide with unsupported inputs or inputs which may generate an exception; and a set of exception handlers which process exception traps in order to provide IEEE 754 compliance. The support code is required to perform implemented functions in order to emulate proper handling of any unsupported data type or data representation (e.g., denormal values). The routines may be written to utilize the FPS in their intermediate calculations if care is taken to restore the users' state at the exit ofthe routine. 6.2 Exception Reporting and Processing
Exceptions in normal mode will be reported to the ARM on the next floating point instruction issued after the exception condition is detected. The state ofthe ARM processor, the FPS register file, and memory may not be precise with respect to the offending instruction at the time the exception is taken. Sufficient information is available to the support code to correctly emulate the instruction and process any exception resulting from the instruction.
In some implementations, support code may be used to process some or all operations with special IEEE 754 data, including infinities, NaNs, denormal data, and zeros. Implementations which do so will refer to these data as unsupported, and bounce to the support code in a manner generally invisible to user code, and return with the IEEE 754 specified result in the destination register. Any exceptions resulting from the operation will abide by the IEEE 754 rules for exceptions. This may include trapping to user code if the corresponding exception enable bit is set.
The IEEE 754 standard defines the response to exceptional conditions for both cases ofthe exception enabled and disabled in the FPSCR. The VFPvl Architecture does not specify the boundary between the hardware and software used to properly comply with the IEEE 754 specification.
6.2.1 Unsupported Operations and Formats
The FPS does not support any operations with decimal data or conversion to or from decimal data. These operations are required by the IEEE 754 standard and must be provided by the support code. Any attempt to utilize decimal data will require library routines for the desired functions. The FPS has no decimal data type and cannot be used to trap instructions which use decimal data.
6.2.2 Use of FMOVX When the FPS is Disabled or Exceptional The FMOVX instruction, executed in SUPERVISOR or UNDEFINED mode may read and write the FPSCR or read the FPSID or FPREG when the FPS is in an exceptional state or is disabled (if the implementation supports a disable option) without causing an exception to be signalled to the ARM.
Although particular embodiments ofthe invention have been described herewith, it will be apparent that the invention is not limited thereto, and that many modifications and additions may be made within the scope ofthe invention. For example, various combinations ofthe features ofthe following dependent claims could be made with the features ofthe independent claims without departing from the scope ofthe present invention.

Claims

CLAIM
1. An apparatus for processing data, said apparatus comprising: a register bank having a plurality of registers; and an instruction decoder responsive to at least one data processing instruction specifying a vector operation that executes a data processing operation a plurality of times using data values from a sequence of registers within said register bank; wherein said register bank includes at least one subset of registers, said sequence of registers being within said subset; and said instruction decoder controls said sequence of registers to wrap within said subset of registers.
2. An apparatus as claimed in claim 1, wherein said vector operation executes said data processing operation using a plurality of respective data values from a corresponding plurality of sequences of registers; said register bank contains a plurality of subsets of registers, said plurality of sequences of registers being within respective subsets; and said instruction decoder controls said plurality of sequences of registers to wrap within respective subsets of registers.
3. Apparatus as claimed in claim 2, wherein said plurality of subsets are disjoint.
4. Apparatus as claimed in any one of claims 1, 2 and 3, wherein said subset comprises a range of consecutively numbers registers.
5. Apparatus as claimed in claim 2, wherein each of said plurality of subsets comprises a range of consecutively numbered registers.
6. Apparatus as claimed in claim 5, where said plurality of subsets comprise respective contiguous ranges of consecutively numbered registers.
7. Apparatus as claimed in claim 6, comprising four contiguous ranges.
8. Apparatus as claimed in any one ofthe preceding claims, further comprising a memory and a transfer controller for controlling transfers of data values between said memory and registers within said register bank, said transfer controller being responsive to multiple transfer instructions to transfer a sequence of data values between said memory and a sequence of registers within said register bank.
9. Apparatus as claimed in claim 6, wherein each range is addressed via an incrementer that wraps between the end points of that range.
10. Apparatus as claimed in any one ofthe preceding claims, wherein said sequence is a sequence of consecutive registers.
11. Apparatus as claimed in any one ofthe preceding claims, wherein said register bank and said instruction decoder are part of a floating point unit.
12. A method of processing data, said method comprising the steps of: storing data values within a plurality of registers of a register bank; and in response to at least one data processing instruction specifying a vector operation, executing a data processing operation a plurality of times using data values from a sequence of registers within said register bank; wherein said register bank includes at least one subset of registers, said sequence of registers being within said subset; and during said executing, said sequence of registers wraps within said subset of registers.
13. A method as claimed in claim 12, wherein said vector operation executes said data processing operation using a plurality of respective data values from a corresponding plurality of sequences of registers; said register bank contains a plurality of subsets of registers, said plurality of sequences of registers being within respective subsets; and during said execution, said plurality of sequences of registers to wrap within respective subsets of registers.
14. A method as claimed in claim 13, wherein data values within one sequence are tap coefficients of a filter and data values in another sequence are signal values for filtering by said filter.
15. A method as claimed in claim 12, wherein a plurality of vectors operations are executed upon data values within said plurality of sequences with a starting point of at least one sequence changing with each vector operation.
PCT/GB1999/000707 1998-05-27 1999-03-09 Recirculating register file WO1999061997A1 (en)

Priority Applications (5)

Application Number Priority Date Filing Date Title
EP99907748A EP1080421B1 (en) 1998-05-27 1999-03-09 Recirculating register file
JP2000551329A JP2002517038A (en) 1998-05-27 1999-03-09 Recirculation register file
IL13924999A IL139249A (en) 1998-05-27 1999-03-09 Recirculating register file
DE69901556T DE69901556T2 (en) 1998-05-27 1999-03-09 RETURNING REGISTER STORAGE
PCT/GB1999/000707 WO1999061997A1 (en) 1998-05-27 1999-03-09 Recirculating register file

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US09/085,752 1998-05-27
PCT/GB1999/000707 WO1999061997A1 (en) 1998-05-27 1999-03-09 Recirculating register file

Publications (1)

Publication Number Publication Date
WO1999061997A1 true WO1999061997A1 (en) 1999-12-02

Family

ID=10845913

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/GB1999/000707 WO1999061997A1 (en) 1998-05-27 1999-03-09 Recirculating register file

Country Status (1)

Country Link
WO (1) WO1999061997A1 (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002017071A1 (en) * 2000-08-22 2002-02-28 Theis Jean Paul A configurable register file with multi-range shift register support
EP1462932A2 (en) * 2003-03-28 2004-09-29 Seiko Epson Corporation Vector processor and register addressing method
EP1550949A3 (en) * 2003-12-31 2007-10-24 STMicroelectronics, Inc. Method and system for software pipelining using a shifting register queue
US20220206805A1 (en) * 2020-12-26 2022-06-30 Intel Corporation Instructions to convert from fp16 to bf8

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2216307A (en) * 1988-03-01 1989-10-04 Ardent Computer Corp Vector register file
EP0646877A2 (en) * 1993-10-05 1995-04-05 Fujitsu Limited Computer system and method for processing vector data
US5513366A (en) * 1994-09-28 1996-04-30 International Business Machines Corporation Method and system for dynamically reconfiguring a register file in a vector processor

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2216307A (en) * 1988-03-01 1989-10-04 Ardent Computer Corp Vector register file
EP0646877A2 (en) * 1993-10-05 1995-04-05 Fujitsu Limited Computer system and method for processing vector data
US5513366A (en) * 1994-09-28 1996-04-30 International Business Machines Corporation Method and system for dynamically reconfiguring a register file in a vector processor

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002017071A1 (en) * 2000-08-22 2002-02-28 Theis Jean Paul A configurable register file with multi-range shift register support
US6718429B1 (en) * 2000-08-22 2004-04-06 Antevista Gmbh Configurable register file with multi-range shift register support
EP1462932A2 (en) * 2003-03-28 2004-09-29 Seiko Epson Corporation Vector processor and register addressing method
EP1462932A3 (en) * 2003-03-28 2006-07-05 Seiko Epson Corporation Vector processor and register addressing method
US7836279B2 (en) 2003-09-05 2010-11-16 Stmicroelectronics, Inc. Method and system for supporting software pipelining using a shifting register queue
EP1550949A3 (en) * 2003-12-31 2007-10-24 STMicroelectronics, Inc. Method and system for software pipelining using a shifting register queue
US20220206805A1 (en) * 2020-12-26 2022-06-30 Intel Corporation Instructions to convert from fp16 to bf8

Similar Documents

Publication Publication Date Title
US6189094B1 (en) Recirculating register file
EP1080420B1 (en) Mixed vector/scalar register file
US6247113B1 (en) Coprocessor opcode division by data type
US6170001B1 (en) System for transfering format data from format register to memory wherein format data indicating the distribution of single or double precision data type in the register bank
US6304963B1 (en) Handling exceptions occuring during processing of vector instructions
US7127593B2 (en) Conditional execution with multiple destination stores
US5889980A (en) Mode independent support of format conversion instructions for hexadecimal and binary floating point processing
US5812439A (en) Technique of incorporating floating point information into processor instructions
US6986025B2 (en) Conditional execution per lane
US5696709A (en) Program controlled rounding modes
EP1267258A2 (en) Setting up predicates in a processor with multiple data paths
US20030005013A1 (en) Floating point system that represents status flag information within a floating point operand
US7861071B2 (en) Conditional branch instruction capable of testing a plurality of indicators in a predicate register
KR100864890B1 (en) Locking source registers in a data processing apparatus
EP1080421B1 (en) Recirculating register file
GB2338094A (en) Vector register addressing
WO1999061997A1 (en) Recirculating register file
EP0992890A2 (en) A processor
JPH06501805A (en) RISC microprocessor architecture with multiple register sets

Legal Events

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

Ref document number: 99806670.2

Country of ref document: CN

AK Designated states

Kind code of ref document: A1

Designated state(s): CN IL IN JP KR RU

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE CH CY DE DK ES FI 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: 1999907748

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: IN/PCT/2000/00499/MU

Country of ref document: IN

WWE Wipo information: entry into national phase

Ref document number: 139249

Country of ref document: IL

WWE Wipo information: entry into national phase

Ref document number: 1020007013163

Country of ref document: KR

WWP Wipo information: published in national office

Ref document number: 1999907748

Country of ref document: EP

WWP Wipo information: published in national office

Ref document number: 1020007013163

Country of ref document: KR

WWG Wipo information: grant in national office

Ref document number: 1999907748

Country of ref document: EP

WWG Wipo information: grant in national office

Ref document number: 1020007013163

Country of ref document: KR