Method and apparatus for implementing a flush command for an accelerated graphics port device

• US 5,990,913 A
• Filed: 07/30/1997
• Issued: 11/23/1999
• Est. Priority Date: 07/30/1997
• Status: Expired due to Term

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Harriman, David J., Langendorf, Brian K.
• Primary Examiner(s): Tung, Kee M.
• Assistant Examiner(s): Luu, Sy D.
• Application Number:
  US08/902,825
• Time in Patent Office: 846 Days
• Field of Search: 345/521, 345/522, 345/507, 345/501, 395/376, 395/561, 395/825, 395/844-845
• US Class Current: 345/531
• CPC Class Codes: G06F 3/14 Digital output to display device ; Cooperation and interconnection of the display device with other functional units
• CPC Class Codes: G06F 9/30087 Synchronisation or serialisation instructions
• CPC Class Codes: G06F 9/3834 Maintaining memory consistency

Abstract

A system and method for ensuring the execution of commands is visible throughout that system. A flush command is received and enqueued in a first queue together with a synchronization value enqueued in a second queue. The flush command is delayed from being dequeued from the first queue until both the flush command and the synchronization value are advanced to the head of their respective queues. Thereafter, the flush command is dequeued subsequently causing the return of a quad-word of random data as an acknowledge signal.

Claims

1. A method comprising:

queuing a flush command in a first queue;

queuing a synchronization value in a second queue; and

delaying the flush command from being dequeued from the first queue until both the flush command is advanced to a head of the first queue and the synchronization value is advanced to a head of the second queue.

2. The method of claim 1 further comprising delaying the synchronization value from being dequeued from the second queue until the flush command is advanced to the head of the first queue.

3. The method of claim 1 further comprising delaying the flush command from being dequeued from the first queue until at least one executable command, enqueued in the second queue prior to receiving the flush command, has been dequeued from the second queue.

4. The method of claim 1 further comprising delaying the flush command from being dequeued from the first queue until at least one executable command, enqueued in the first queue prior to receiving the flush command, has been dequeued from the first queue.

5. The method of claim 1 further comprising concurrently dequeuing the flush command and the synchronization value from the first queue and the second queue, respectively.

6. The method of claim 1 further comprising:

producing a control signal upon dequeuing the flush command; and

transferring the control signal to a memory access logic.

7. The method of claim 6 further comprising:

producing an acknowledge signal upon detecting the control signal; and

precluding access to a main memory.

8. The method of claim 1, wherein the synchronization value includes an invalid command.

9. The method of claim 1, wherein the synchronization value includes a no-operation command.

10. A system comprising:

a memory;

a graphics controller; and

a chipset coupled to both the memory and the graphics controller, the chipset including a memory controller that comprises;

a command interface including a plurality of queues including a first queue adapted to enqueue the flush command and a second queue adapted to enqueue a synchronization value that includes a no-operation command processed in coordination with the flush command, and a queue advance circuit adapted to delay the flush command from being dequeued from the first queue until the synchronization value is advanced to a head of the second queue.

11. The system of claim 10, wherein the first queue of the command interface is a read command queue.

12. The system of claim 11, wherein the second queue of the command interface is a write command queue.

13. The system of claim 10, wherein the memory controller of the chipset further comprises a command reordering logic coupled to the command interface, the command reordering logic reorders the flush command for execution and conversion to a control signal according to a selected optimization strategy.

14. The system of claim 13, wherein the memory controller of the chipset further comprises a memory access logic coupled to the command reordering logic, the memory access logic produces an acknowledge signal upon receiving the control signal in lieu of accessing the memory and returns the acknowledge signal to the graphics controller.

15. The system of claim 14, wherein the synchronization value is provided through assertion of a synchronization bit in an entry of the second queue.

16. A system comprising:

a memory;

a graphics controller; and

a chipset coupled between the memory and the graphics controller, the chipset including a memory controller to ensure data coherency through a flush command and to prevent reordering of a first memory access command ahead of a second memory access command previously issued by the graphics controller, the memory controller having a command interface including;

a first queue to enqueue the flush command,

a second queue to enqueue a synchronization value, and

a queue advance circuit to delay the flush command from being dequeued from the first queue until the synchronization value is advanced to a head of the second queue and the flush command is advanced to a head of the first queue.

17. The system of claim 16, wherein the synchronization value enqueued in the second queue is a no-operation command processed in coordination with the flush command.

18. The system of claim 17, wherein the first queue includes a read command queue.

19. The system of claim 18, wherein the second queue includes a write command queue.

20. The system of claim 17, wherein the memory controller further includes a command reordering logic coupled to the command interface, the command reordering logic to reorder the flush command for execution and conversion to a control signal according to a selected optimization strategy.

21. The system of claim 20, wherein the memory controller further comprises a memory access logic coupled to the command reordering logic, the memory access logic to produce an acknowledge signal upon receiving the control signal in lieu of accessing the memory and to return the acknowledge signal to the graphics controller.

22. The system of claim 21, wherein the memory controller further comprising a data buffer coupled to the memory access logic.