Multiple operations employing divided arithmetic logic unit and multiple flags register

• US 5,592,405 A
• Filed: 06/07/1995
• Issued: 01/07/1997
• Est. Priority Date: 11/17/1989
• Status: Expired due to Term

• Current Assignee: Texas Instruments, Inc.
• Original Assignee: Texas Instruments, Inc.
• Inventors: Guttag, Karl M., Balmer, Keith, Gove, Robert J., Ing-Simmons, Nicholas K.
• Primary Examiner(s): Envall, Jr., Roy N.
• Assistant Examiner(s): Ngo, Chuong D.
• Application Number:
  US08/484,579
• Time in Patent Office: 580 Days
• Field of Search: 364/715.06, 364/736, 364/736.5, 364/737, 364/749, 364/750.5, 364/758, 364/786, 364/787, 364/788
• US Class Current: 708/518
• CPC Class Codes: G06F 12/0284 Multiple user address space allocation, e.g. using different base addresses interprocessor communication G06F15/163
• CPC Class Codes: G06F 15/17375 One dimensional, e.g. linear array, ring

Abstract

A data processing apparatus includes an arithmetic logic unit is divided into a plurality of sections. Each section generates at a corresponding output a digital resultant signal representing a combination of respective subsets of first and second multibit digital inputs. The arithmetic logic unit includes a status detector generating a single bit status signal indicative of said digital resultant signal of a corresponding section of the arithmetic logic unit. These single bit status signals are stored in predetermined locations within a multiple flags register. An options register stores an indication of the number of sections selected from a plurality of possible number of sections into which the arithmetic logic unit is divided. The arithmetic logic unit is further connected to the multiple flags register so that each section selects for output either corresponding bits of the first multibit digital input or the second multibit digital input dependent upon the digital state of a corresponding single status bit in the multiple flags register. This technique permits a variety of functions such as add with saturation, maximum, pixel transparency and color expansion.

Claims

1. A data processing apparatus comprising:

an options register storing an indication of a number of sections selected from a plurality of possible number of sections;

an arithmetic logic unit connected to said options register having a first N-bit data input for a first N-bit digital input and a second N-bit data input for a second N-bit digital input, said arithmetic logic unit being divisible into said plurality of possible number of sections and divided into a plurality of sections equal in number to said indication of said number of sections stored in said options register, each section generating at a corresponding output a digital resultant signal representing a combination of respective subsets of said first and second multibit digital inputs of said inputs, and said arithmetic logic unit including a status detector generating a plurality of single bit status signals equal in number to said indication of said number of sections stored in said options register, each single bit status signal indicative of said digital resultant signal of a corresponding section of said arithmetic logic unit; and

a multiple flags register connected to said options register and said status detector having a number of bit storage locations greater than a greatest possible number of sections of said arithmetic logic unit, said multiple flags register connected to said status detector of said arithmetic logic unit for storing in predetermined locations within said multiple flags register said plurality of single bit status signals equal in number to said indication of said number of sections stored in said options register.

2. The data processing apparatus of claim 1, wherein:

said status detector generates said single bit status signals indicating whether respective sections of said arithmetic logic unit generate a carry output.

3. The data processing apparatus of claim 1, wherein:

said status detector generates said single bit status signals indicating whether respective sections of said arithmetic logic unit generate a borrow output.

4. The data processing apparatus of claim 1, wherein:

said status detector generates said single bit status signals indicating whether respective sections of said arithmetic logic unit generate digital resultant signals equal to zero.

5. The data processing apparatus of claim 1, wherein:

said arithmetic logic unit includes 32 bits; and

said options register stores said indication of a number of sections including a either a first indication of 4 sections of 8 bits each, a second indication of 2 sections of 16 bits each or a third indication of 1 section of 32 bits.

6. The data processing apparatus of claim 5, wherein:

said options register stores said indication of a number of sections in three bits, said first indication being stored as "000", said second indication being stored as "001", said third indication being stored as "010", said fourth indication being stored as "011" and said fifth indication being stored as "100".

7. The data processing apparatus of claim 1, wherein:

said arithmetic logic unit includes 32 bits; and

said options register stores said indication of a number of sections including a either a first indication of 16 sections of 2 bits each, a second indication of 8 sections of 4 bits each, a third indication of 4 sections of 8 bits each, a fourth indication of 2 sections of 16 bits each or a fifth indication of 1 section of 32 bits.

8. The data processing apparatus of claim 1, wherein:

said arithmetic logic unit performs an add multiple instruction by adding respective sections of said first and second multibit digital inputs, said status detector generating said single bit status signals indicating whether respective sections of said arithmetic logic unit generate a carry output.

9. The data processing apparatus of claim 1, wherein:

said arithmetic logic unit performs a subtract multiple instruction by subtracting respective sections of said second multibit digital input from said first multibit digital input, said status detector generating said single bit status signals indicating whether respective sections of said arithmetic logic unit generate a borrow output.

10. The data processing apparatus of claim 1, wherein:

said arithmetic logic unit performs a compare multiple instruction by comparing respective sections of said first and second multibit digital inputs, said status detector generates said single bit status signals indicating whether respective sections of said arithmetic logic unit generate digital resultant signals equal to zero.

11. The data processing apparatus of claim 1, wherein:

said arithmetic logic unit being further connected to said multiple flags register, said arithmetic logic unit performing a merge multiple instruction by selecting for each section said digital resultant signal equal to either corresponding bits of said first multibit digital input or corresponding bits of said second multibit digital input dependent upon the digital state of a corresponding one of a set of bits in said multiple flags register equal in number to said indication of said number of sections stored in said options register.

12. The data processing apparatus of claim 11, wherein:

said arithmetic logic unit performs said merge multiple instruction by selecting for each section said digital resultant signal equal to said corresponding bits of said first multibit digital input if said corresponding bit of said multiple flags register is "0" and equal to said corresponding bits of said second multibit digital input if said corresponding bit of said multiple flags register is "1".

13. The data processing apparatus of claim 11, wherein:

said arithmetic logic unit includes 32 bits; and

said options register stores said indication of a number of sections including a either a first indication of 4 sections of 8 bits each, a second indication of 2 sections of 16 bits each or a third indication of 1 section of 32 bits.

14. The data processing apparatus of claim 11, wherein:

said arithmetic logic unit includes 32 bits; and

said options register stores said indication of a number of sections including a either a first indication of 16 sections of 2 bits each, a second indication of 8 sections of 4 bits each, a third indication of 4 sections of 8 bits each, a fourth indication of 2 sections of 16 bits each or a fifth indication of 1 section of 2 bits.

15. The data processing apparatus of claim 14, wherein:

said options register stores said indication of a number of sections in three bits, said first indication being stored as "000", said second indication being stored as "001", said third indication being stored as "010", said fourth indication being stored as "011" and said fifth indication being stored as "100".

16. The data processing apparatus of claim 1, further comprising:

a data register file including

a plurality of data registers for storing data,

a first output connected to said first data input of said arithmetic logic unit for recalling from a first specified data register data stored therein, thereby forming said first multibit digital input,

a second output connected to said second data input of said arithmetic logic unit for recalling form a second specified data register data stored therein, thereby forming said second multibit digital input,

a first input connected to said output of said arithmetic logic unit for storing in a fourth specified one of said data registers said digital resultant signal; and

a bus connected to said multiple flags register and said data register file for moving data from an instruction specified one of said plurality of data registers to said multiple flags register in response to a register move instruction specifying one of said plurality of data registers as a source and said multiple flags register as said destination.

17. The data processing apparatus of claim 1, further comprising:

a data register file including

a plurality of data registers for storing data,

a first output connected to said first data input of said arithmetic logic unit for recalling from a first specified data register data stored therein, thereby forming said first multibit digital input,

a second output connected to said second data input of said arithmetic logic unit for recalling form a second specified data register data stored therein, thereby forming said second multibit digital input,

a first input connected to said output of said arithmetic logic unit for storing in a fourth specified one of said data registers said digital resultant signal; and

a bus connected to said multiple flags register and said data register file for moving data from said multiple flags register to an instruction specified one of said plurality of data registers in response to a register move instruction specifying said multiple flags register as a source and one of said plurality of data registers as said destination.

18. The data processing apparatus of claim 1, further comprising:

a data memory having an address input and a data port for storing data at a plurality of addressable memory storage locations;

an address generator connected to said address input of said data memory, said address generator generating an address corresponding to one of said plurality of addressable memory storage locations of said data memory; and

a data bus connected to said multiple flags register and said data port of said data memory for moving data from one of said plurality of addressable memory storage locations of said memory corresponding to said address generated by said address generator to said multiple flags register in response to a register load instruction specifying said multiple flags register as said destination.

19. The data processing apparatus of claim 1, further comprising:

a data memory having an address input and a data port for storing data at a plurality of addressable memory storage locations;

an address generator connected to said address input of said data memory, said address generator generating an address corresponding to one of said plurality of addressable memory storage locations of said data memory; and

a data bus connected to said multiple flags register and said data port of said data memory for moving data from said multiple flags register to one of said plurality of addressable memory storage locations of said memory corresponding to said address generated by said address generator in response to a register store instruction specifying said multiple flags register as said destination.

20. The data processing apparatus of claim 1, wherein:

said predetermined locations within said multiple flags register where said single bit status signals are stored are a set of least significant bits of said multiple flags register.

21. The data processing apparatus of claim 20, wherein:

said multiple flags register consists of N bit storage locations.

22. A data processing apparatus comprising:

an options register storing an indication of a number of sections selected from a plurality of possible number of sections;

a multiple flags register connected to said options register, said multiple flags register having a number of bit storage locations greater than a greatest. possible number of sections of said indication of said number of sections stored in said options register, for storing in predetermined locations a plurality of single status bits equal in number to said number of indications of said options register; and

an arithmetic logic unit connected to said options register and said multiple flags register having a first N-bit data input for a first N-bit digital input and a second N-bit data input for a second N-bit digital input, said arithmetic logic unit being divisible into said plurality of possible number of sections and divided into a plurality of sections equal in number to said indication of said number of sections stored in said options register, each section generating at a corresponding output a digital resultant signal equal to either corresponding bits of said first multibit digital input or corresponding bits of said second multibit digital input dependent upon the digital state of a corresponding one of said plurality of single status bits in said multiple flags register equal in number to said indication of said number of sections stored in said options register.

23. The data processing apparatus of claim 22, wherein:

said arithmetic logic unit performs said merge multiple instruction by selecting for each section said digital resultant signal equal to said corresponding bits of said first multibit digital input if said corresponding bit of said multiple flags register is "0" and equal to said corresponding bits of said second multibit digital input if said corresponding bit of said multiple flags register is "1".

24. The data processing apparatus of claim 22, wherein:

said arithmetic logic unit includes 32 bits; and

said options register stores said indication of a number of sections including a either a first indication of 4 sections of 8 bits each, a second indication of 2 sections of 16 bits each or a third indication of 1 section of 32 bits.

25. The data processing apparatus of claim 22, wherein:

said arithmetic logic unit includes 32 bits; and

said options register stores said indication of a number of sections including a either a first indication of 16 sections of 2 bits each, a second indication of 8 sections of 4 bits each, a third indication of 4 sections of 8 bits each, a fourth indication of 2 sections of 16 bits each or a fifth indication of 1 section of 32 bits.

26. The data processing apparatus of claim 25, wherein:

said options register stores said indication of a number of sections in three bits, said first indication being stored as "000", said second indication being stored as "001", said third indication being stored as "010", said fourth indication being stored as "011" and said fifth indication being stored as "100".

27. The data processing apparatus of claim 22, further comprising:

a data register file including

a plurality of data registers for storing data,

a first output connected to said first data input of said arithmetic logic unit for recalling from a first specified data register data stored therein, thereby forming said first multibit digital input,

a second output connected to said second data input of said arithmetic logic unit for recalling form a second specified data register data stored therein, thereby forming said second multibit digital input,

a first input connected to said output of said arithmetic logic unit for storing in a fourth specified one of said data registers said digital resultant signal; and

a bus connected to said multiple flags register and said data register file for moving data from an instruction specified one of said plurality of data registers to said multiple flags register in response to a register move instruction specifying one of said plurality of data registers as a source and said multiple flags register as said destination.

28. The data processing apparatus of claim 22, further comprising:

a data register file including

a plurality of data registers for storing data,

a first output connected to said first data input of said arithmetic logic unit for recalling from a first specified data register data stored therein, thereby forming said first multibit digital input,

a second output connected to said second data input of said arithmetic logic unit for recalling form a second specified data register data stored therein, thereby forming said second multibit digital input,

a first input connected to said output of said arithmetic logic unit for storing in a fourth specified one of said data registers said digital resultant signal; and

a bus connected to said multiple flags register and said data register file for moving data from said multiple flags register to an instruction specified one of said plurality of data registers in response to a register move instruction specifying said multiple flags register as a source and one of said plurality of data registers as said destination.

29. The data processing apparatus of claim 22, further comprising:

a data memory having an address input and a data port for storing data at a plurality of addressable memory storage locations;

an address generator connected to said address input of said data memory, said address generator generating an address corresponding to one of said plurality of addressable memory storage locations of said data memory; and

a data bus connected to said multiple flags register and said data port of said data memory for moving data from one of said plurality of addressable memory storage locations of said memory corresponding to said address generated by said address generator to said multiple flags register in response to a register load instruction specifying said multiple flags register as said destination.

30. The data processing apparatus of claim 22, further comprising:

a data memory having an address input and a data port for storing data at a plurality of addressable memory storage locations;

an address generator connected to said address input of said data memory, said address generator generating an address corresponding to one of said plurality of addressable memory storage locations of said data memory; and

a data bus connected to said multiple flags register and said data port of said data memory for moving data from said multiple flags register to one of said plurality of addressable memory storage locations of said memory corresponding to said address generated by said address generator in response to a register store instruction specifying said multiple flags register as said destination.

31. The data processing apparatus of claim 22, wherein:

said multiple flags register includes more bits than a maximum number of sections into which said arithmetic logic unit may be divided and said predetermined location within said multiple flags register where said single bit status signals are stored are a set of least significant bits of said multiple flags register.

32. The data processing apparatus of claim 31, wherein:

said multiple flags register consists of N bit storage locations.

33. The method of addition with saturation comprising the steps of:

storing an indication of a number of sections selected from a plurality of possible number of sections;

dividing an arithmetic logic unit into a plurality of sections equal in number to said stored indication of said selected number of sections, each section forming a digital resultant signal for respective subsets of a first multibit digital input and a second multibit digital input;

employing said divided arithmetic logic unit to form a sum for respective subsets of a first multibit digital input and a second multibit digital input;

generating a plurality of single bit status signal equal in number to said stored indication of said selected number of sections, each indicative of whether a corresponding one of said plurality of sums generates a carry out;

storing said plurality of single bit status signals corresponding to each of said plurality of sums in a register having a number of bit storage locations greater than a greatest possible number of sections of said stored indication; and

employing said divided arithmetic logic unit to select for each section either said corresponding sum if said stored single bit status signals equal in number to said stored indication of said selected number of sections indicates said corresponding sum failed to generate a carry out or a corresponding saturated value if said stored single bit status signals equal in number to said stored indication of said selected number of sections indicates said corresponding sum generated a carry out.

34. The method of addition with saturation of claim 33, wherein:

said saturated value consists of all "1'"'"'s" for each section.

35. The method of determining a maximum value comprising the steps of:

storing an indication of a number of sections selected from a plurality of possible number of sections;

dividing an arithmetic logic unit into a plurality of sections equal in number to said stored indication equal in number to said stored indication of said selected number of sections, each section forming a digital resultant signal for respective subsets of a first multibit digital input and a second multibit digital input;

employing said divided arithmetic logic unit to form a difference for respective subsets of a first multibit digital input and a second multibit digital input;

generating a plurality of single bit status signal equal in number to said stored indication of said selected number of sections, each indicative of whether a corresponding one of said plurality of differences generates a borrow out;

storing said plurality of single bit status signals corresponding to each of said plurality of differences in a register having a number of bit storage locations greater than a greatest possible number of sections of said stored indication; and

employing said divided arithmetic logic unit to select for each section either said respective bits of said second multibit digital input if said stored single bit status signals equal in number to said stored indication of said selected number of sections indicates said corresponding difference failed to generate a borrow out or said respective bits of said second multibit digital input if said stored single bit status signals equal in number to said stored indication of said selected number of sections indicates said corresponding difference generated a borrow out.

36. The method of pixel transparency comprising the steps of:

storing an indication of a number of sections selected from a plurality of possible number of sections;

dividing an arithmetic logic unit into a plurality of sections equal in number to said stored indication of said selected number of sections, each section forming a digital resultant signal for respective subsets of a first multibit digital input and a second multibit digital input;

employing said divided arithmetic logic unit to compare for respective subsets of a first multibit digital input representing source pixel color codes and a second multibit digital input in which each section is a transparent color code;

generating a plurality of single bit status signal equal in number to said stored indication of said selected number of sections, each indicative of whether a corresponding one of said corresponding source pixel color code equals said transparent color code;

storing said plurality of single bit status signals corresponding to each of said plurality of comparisons in a register having a number of bit storage locations greater than a greatest possible number of sections of said stored indication; and

employing said divided arithmetic logic unit to select for each section either said respective bits of said source pixel color code if said stored single bit status signals equal in number to said stored indication of said selected number of sections indicates said corresponding comparison failed to generate an equal signal or said respective bits of a destination pixel color code if said stored single bit status signals equal in number to said stored indication of said selected number of sections indicates said corresponding comparison generated an equal signal.

37. A method for guided copying comprising the steps of:

storing an indication of a number of sections selected from a plurality of possible number of sections;

loading a multiple flags register with a plurality of single status bits equal in number to said stored indication of said selected number of sections, said multiple flags register having a number of bit storage locations greater than a greatest possible number of sections of said stored indication;

dividing an arithmetic logic unit into a plurality of sections equal in number to said stored indication of said selected number of sections, each section forming a digital resultant signal for respective subsets of a first multibit digital input and a second multibit digital input; and

employing said divided arithmetic logic unit to generate at a corresponding output a digital resultant signal equal to either corresponding bits of said first multibit digital input or corresponding bits of said second multibit digital input dependent upon the digital state of a corresponding one of said plurality of single status bits in said multiple flags register equal in number to said stored indication of said selected number of sections.

38. The method of guided copying of claim 37, wherein:

said arithmetic logic unit selects for each section said digital resultant signal equal to said corresponding bits of said first multibit digital input if said corresponding bit of said multiple flags register is "0" and equal to said corresponding bits of said second multibit digital input if said corresponding bit of said multiple flags register is "1".

39. The method of guided copying of claim 38, wherein:

said first multibit digital input consists of a first pixel color code for each section; and

said second multibit digital input consists of a second pixel color code for each section.