Method and system for dynamically adding and removing display modes coordinated across multiple graphics processing units

• US 8,766,989 B2
• Filed: 07/29/2009
• Issued: 07/01/2014
• Est. Priority Date: 07/29/2009
• Status: Active Grant

• Current Assignee: NVIDIA Corporation
• Original Assignee: NVIDIA Corporation
• Inventors: Wyatt, David, Glanville, Linda
• Primary Examiner(s): NGUYEN, HAU H
• Application Number:
  US12/511,434
• Publication Number:
  US 20110025696A1
• Time in Patent Office: 1,798 Days
• Field of Search: 345502-505, 345/520
• US Class Current: 345/520
• CPC Class Codes: G09G 2360/02 Graphics controller able to handle multiple formats, e.g. input or output formats
• CPC Class Codes: G09G 2360/08 Power processing, i.e. workload management for processors involved in display operations, such as CPUs or GPUs
• CPC Class Codes: G09G 2360/125 using unified memory architecture [UMA]
• CPC Class Codes: G09G 2370/042 for monitor identification
• CPC Class Codes: G09G 2370/047 using display data channel standard [DDC] communication
• CPC Class Codes: G09G 2370/12 Use of DVI or HDMI protocol in interfaces along the display data pipeline
• CPC Class Codes: G09G 5/363 Graphics controllers

Abstract

The present invention provides a method and system for coordinating graphics processing units in a single computing system. A method is disclosed which allows for the construction of a list of shared display modes that may be employed by both of the graphics processing units to render an output in a display device. By creating the list of shared commonly supportable display modes, the output displayed in the display device may advantageously provide a consistent graphical experience persisting through the use of alternate graphics processing units in the system. One method builds a list of shared display modes by compiling a list from a GPU specific base mode list and dynamic display modes acquired from an attached display device. Another method provides the ability to generate graphical output configurations according to a user-selected display mode that persists when alternate graphics processing units in the system are used to generate graphical output.

Claims

1. A method for generating a mode list, the method comprising:

building a shared mode list comprising a plurality of display modes compatible with both a driver of a first GPU, and a driver of a second GPU;

determining a plurality of display modes available to a first display communicatively coupled to said first GPU;

constructing a dynamic mode list, the dynamic mode list comprising a plurality of display modes available to said first display not already comprised in the shared mode list;

compiling a final mode list comprising a combination of the dynamic mode list and the shared mode list; and

validating the final mode list with the second GPU.

2. The method according to claim 1, wherein, said sending, said building, said determining, said constructing and said compiling are performed by said driver of said first GPU.

3. The method according to claim 1 wherein, said sending, said building, said determining, said constructing, said compiling, and said validating are performed automatically by said driver of said first GPU when said first GPU is loaded.

4. The method according to claim 1, wherein said building a shared mode list comprises:

sending a request from said first GPU to said second GPU for a second base mode list comprising a plurality of display modes corresponding to a driver of a second GPU;

receiving said second base mode list from said second GPU;

comparing said second base mode list to a first base mode list comprising a plurality of display modes compatible with both a driver of a first GPU; and

collecting said display modes comprised in both said first base mode list and said second base mode list.

5. The method according to claim 1, wherein said validating the final mode list comprises:

verifying said plurality of display modes comprised in said final mode list is supported by said second GPU;

receiving a suggested display mode if a display mode of said plurality of display modes comprised in said final mode list is not supported by said second GPU;

evaluating if said suggested display mode is supported by said first GPU; and

adding said suggested display mode to said third mode list if said suggested display mode is supported by said first GPU and said second GPU.

6. The method according to claim 1, further comprising:

sending said final mode list to an operating system executing on said computing system.

7. The method according to claim 1, wherein said determining a plurality of display modes available to said first display comprises:

obtaining display identity information of said first display; and

parsing said display identity information of said first display for timings corresponding to a plurality of display modes specific to said first display.

8. The method according to claim 1, wherein said constructing said second mode list comprises:

comparing said plurality of display modes available to a first display to said shared mode list;

flagging any of said plurality of display modes available to a first display not comprised in said shared mode list; and

compiling said dynamic mode list from said plurality of display modes specific to a first display flagged.

9. The method according to claim 1, wherein constructing a dynamic mode list comprises:

detecting an addition of a second display to said computing system;

determining whether said second display has been communicatively coupled to said first GPU or said second GPU during a previous session; and

producing a display in said second display comprising a display mode consistent with a last display mode used during said previous session if said second display has been communicatively coupled to said first GPU or said second GPU.

10. The method according to claim 9, wherein said determining whether said second display has been communicatively coupled to said first GPU or said second GPU during a previous session further comprises:

determining said second display comprises a new display;

determining the second display comprises a new EDID;

obtaining said new EDID from said second display communicatively coupled to said first GPU or said second GPU;

parsing said new EDID for timings corresponding to a plurality of display modes specific to said second display;

adding said plurality of display modes specific to said second display to said final mode list;

flagging said new EDID;

sending said new EDID to said second GPU; and

validating said final mode list.

11. The method according to claim 9, wherein said determining whether said second display has been communicatively coupled to said first GPU or said second GPU during a previous session further comprises:

determining said second display comprises a new display;

determining the second display does not comprise a new EDID;

creating a simulated EDID corresponding to said new display;

sending said simulated EDID to said second GPU; and

validating said final mode list.

12. The method according to claim 1, the method further comprising:

detecting a detachment of a second display to said computing system;

flagging an EDID of said second display; and

sending an invalid EDID to said second GPU to signal said detachment.

13. The method according to claim 1, wherein said validating said final mode list with said second GPU comprises validating only a portion of said final mode list comprising said dynamic mode list with said second GPU.

14. The method according to claim 1, wherein said validating said final mode list with said second GPU comprises:

referencing a plurality of capabilities of said second GPU;

clamping said final mode list at said plurality of capabilities of said second GPU; and

validating a plurality of display modes comprised in said final mode list that do not exceed said plurality of capabilities of said second GPU.