JPH0365987A - Window control system at the time of using double frame buffer - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] [Summary] Regarding the double frame buffer 1 use/window management method for quickly reading out image data stored in an external storage device to the frame buffer memory, the double frame buffer 1 is used even when a window manager is used. The purpose is to provide a window management method when using a double frame buffer that allows high-speed display switching using buffer memory.
In a workstation having double frameback memory, a display buffer flag storage means for storing a flag indicating that the contents of one of the frame buffers is being displayed, and management regarding the window in which the contents of each frame buffer are displayed. The present invention is configured to have a window information management storage means for storing information, and a control means for controlling processing for displaying each window, processing for switching frame buffers, and processing for receiving a mouse click event using the above two means. .

[Industrial application field]

The present invention relates to a screen display method in a workstation, and more particularly to a window management method when using a double frame buffer for rapidly reading image data stored in an external storage device into a frame buffer memory.

Workstation screen display methods have been diversified, and in particular, a multi-window method in which a plurality of windows are displayed on one screen and processing is performed independently within each window has come to be effectively used. Workstation displays generally have a resolution of, for example, 1024 x 1024 dots, and if a color bitmap display uses 1 byte per image, the total screen capacity is 1 byte x IM dot. This becomes image data for IM bytes. Image data for one screen, that is, one frame of image, is read out and displayed on a frame buffer memory in which one screen image corresponds to one address. Each frame is stored in an external storage device such as a hard disk, and is written to the requested frame buffer memory and displayed. When rewriting from one frame to the next frame, there is a problem in that it takes about 5 to 10 seconds to read the new frame from the external storage device. Therefore, there has been an increasing need for a technique that shortens the time required to replace frames.

[Conventional technology]

Conventionally, there is a double frame buffer method as a method for exchanging frames at high speed.

FIG. 8 is a conceptual diagram of the double frame buffer system.

In the diagram, 1 is the display, 2 is the hard disk, 3 and 4
Both are frame buffers. In the double frame buffer method, for example, while frame buffer #1 is being displayed, image data to be displayed next is read from the hard disk into frame buffer #2. After that, if you switch the buffer displayed on the display using hardware, the screen stored in frame buffer #2 will immediately switch from the screen corresponding to frame buffer #1 to be displayed on the display. . Similarly, by sequentially repeating reading image data into frame buffer #1 while displaying frame buffer #2 and then switching buffers, high-speed display switching becomes possible.

FIG. 9 is a conceptual diagram of overlapping windows of a wind manager in a conventional double frame buffer system.

In the figure, A, B, C, A', B', and C' are windows, and windows A, B, and C are stored in frame buffer #1, and A', B'', and C' are windows stored in frame buffer #2. In the conventional method, the windows in the two frame buffers are stored in, for example, C, B, A, C',
Assuming that B' and A' are created in that order, the window manager is not aware that it is a double frame buffer, so it is as if those six windows exist on one frame buffer and their overlap is not created. They were managed by stacking them up from the bottom in the order they were completed.

(Problem to be solved by the invention) In other words, windows A and B with low overlap on the screen
, C are currently displayed via frame buffer #1, but if they are stacked from the bottom in the order of creation, A'B', C' will be displayed from A, B, and C as shown in the figure. It's at the top. At this time, the user is currently viewing windows A and B.

Even if you click the mouse at point P on the upper window A in C, the window manager will notify the mouse click to the highest window among the windows displaying at point P. It will pass the event. That is, in this example, the event is passed to window A' instead of window A. For example, in the case of erasing processing, the result is that window A' is erased when window A should be erased. In other words, a window manager that performs integrated management of window interactions and reception of signals from devices such as keyboards and mice, for example, may perform deletion processing that stops displaying when a mouse click event is received. A major problem has arisen in that windows that should not be erased are erased.

SUMMARY OF THE INVENTION An object of the present invention is to provide a window management method when using a double frame buffer that allows high-speed display switching using a double frame buffer memory even when using a window manager.

[Means to solve the problem]

FIG. 1 is a diagram showing the principle of the present invention.

The present invention relates to a workstation run with double frame buffer memory.

The display buffer flag storage means 12 stores a flag indicating that the content of one of the frame buffers is being displayed.

The window information management storage means 13 stores management information regarding the window in which the contents of each frame buffer are displayed.

The control means 14 controls the process of displaying each window, the process of switching frame buffers, and the process of receiving a mouse click event using the above two means.

[For production]

The present invention uses a window information table that separately manages information on windows displayed in each frame buffer and sequentially stores entries corresponding to overlapping windows in each frame buffer.

Then, the window manager manages the window group to be displayed when switching frame buffers so that it is positioned above the window group displayed in the other frame buffer.

〔Example〕

Embodiments of the present invention will be described with reference to the drawings.

FIG. 2 is an embodiment of a screen display on a workstation for explaining the present invention in detail.

In the figure, A is a display of frame buffer #1, and B is an example diagram of a display of frame buffer #2.

On these two frame buffers, the six windows C, B, A, C'', B', A'' are placed in this order: C, B.
, A are in frame buffer #1, and C', B', A
' is created in frame buffer #2. Frame buffer #1 is currently displayed on the display screen on the workstation, and the window is configured such that B is above C and A is above B. On the other hand, frame buffer #2 is displayed next on the display, and B' and A' exist above C' without overlapping. In the double frame buffer method, while frame buffer #1 is being displayed, the next image data to be displayed is loaded into frame buffer #2, and then by switching the buffer to be displayed on the display using hardware, the display screen can be changed instantly. Changes can be made.

On the other hand, in Works Steelan, there is a wind manager in the boat, which manages the wind by managing a collection of winds, receiving signals from devices such as keyboards, microphones, etc.0 For example, frame buffer # In display 1, when a mouse click event is received at point P of window A, an event is generated to stop displaying window A, that is, to close it, and in this case, the wind manager
Among the windows displaying at the position, the event is passed to the highest one, that is, window A. Also, when the next frame buffer #2 is displayed, if the user clicks the mouse at point P', it will be displayed in the topmost window among the windows displayed at point P', that is, A'. Previously, when using such a window manager, it was necessary to pass an event indicating a mouse click, and the problem was that it was not possible to use a double frame buffer. Therefore, in the present invention, when switching frame buffers, a request is passed to the window manager so that the group of windows displayed in that buffer is positioned above the group of windows displayed in the other frame buffer. This allows the event to be passed to the window requested by the user. Below, the method of the present invention will be explained step by step.

FIG. 3 is a configuration diagram of software and hardware related to the wind manager used in the present invention.

The software 8 has an application program 5 at the top, a window manager 6 below it, and a device driver 7 below it.

The hardware 9 includes a double frame buffer memory 10 and a keyboard and mouse 11. The application program 5 makes various requests such as window display to the window manager 6, and issues a frame buffer switching request to the device driver 7. On the other hand, window manager 6 is application 5
Provides events related to window overlap and events from various devices. When there is a frame buffer switching request from application 5,
Device driver is double frame buffer memory 10
That is, the buffers are switched and displayed for the contents of frame buffers #l and #2. The window manager 6 responds to various requests such as window display from the application 5, manages overlapping windows, etc., and controls the windows by rewriting the contents of the double frame buffer memory 10 via the device driver 7. Further, when a window or the like is specified using the keyboard or mouse 11, the request is passed to the window manager 6 via the device driver 7.

The window manager 6 notifies applications of requests from various devices as events, and manages windows in response to events. Further, the window manager 6 immediately notifies the application 5 of an event regarding a collection of windows.

In the present invention, in such a system, when using the window manager 6, the double frame buffer 1
0 can be used, and in this invention, in particular,
It manages the information of the windows displayed in each frame buffer separately.

FIG. 4 is an example diagram of a display buffer flag and a window information table used in the present invention.

In the figure, (a) is the display buffer flag, (b) and (
C) is a window information table that stores window information displayed in frame buffers #1 and #2, respectively. When the display buffer flag is O, it means that frame buffer #1 is currently being displayed, and when the flag is 1, it means that frame buffer #2 is currently being displayed.

In the window information shown in FIG. 2(b), entries are stored corresponding to the positional relationship of windows to be displayed in frame buffer #1 of FIG. 2(a). As shown in FIG. 2(a), C, B, and A overlap in this order from the bottom. Correspondingly, on the table, the window ID corresponding to C is 1, the window ID corresponding to B is 2, and the window ID corresponding to window A is 3, which are stored in order from the bottom. Window C has upper left coordinates (0
, O), the upper left coordinates of the window are stored as (0, 0), and the height and width of the window are stored as 1024 and 1024, respectively, assuming that the height and width are the entire screen. Furthermore, the window B at the level above it has a window ID of 2, and the upper left coordinates of the window are X=15.
0, y is at position 150, and the window height is 5
001 width is 550. Also, Wind A is Wind 1
0 is shown in the 3 row and the top left coordinate of the window is x = 50
0, y = 300 and the height and width of the window are 400.550, respectively.

In FIG. 4(C), window information for frame buffer #2 is stored. As shown in Figure 2), window C' corresponds to window ID 4, the upper left coordinates of the window are (0, 0), and the width of the height of the window is the entire screen, so each window is 1024. and 1024. Window ID 5 corresponds to window 10, and the upper left coordinate of the window is X50. y is 50. The height and width of the window are 400 and 800. Window C' is a row where window [D is 6, and the window position is
is 500. y is 300. The height and width of the window are 400 and 550. In this way, in the present invention, a display buffer flag, that is, a flag indicating which frame buffer is currently being displayed;
It is assumed that the information of the windows displayed in each frame buffer is managed separately, and the window information table is used from the bottom of the entry upward to match the overlap of the windows.

Next, using these display buffer flags and the window information table, display processing for FIG. 2(a), Φ),
The processing when switching frame buffers and the processing sequence of an application that receives a mouse click event will be described.

FIGS. 5(a) and 5(b) are functional block diagrams of the processing sequencer up to displaying the screen of FIG. 2.

When the process starts, the application issues a request to open window C shown in FIG. 2(a) on the SPI. Arguments for the request are the X and Y coordinates of the upper left corner of the window, and the vertical and horizontal sizes. This information is shown in the window IDI in FIG. 4(b). When this open request for Window C is passed from the application to the Wind Manager, the Wind Manager requests the device to
Performs management for window display processing. To do this, the window manager first registers the upper left coordinates and vertical and horizontal sizes of the requested window in the management table, and then notifies the application of the assigned window ID, that is, the ID 1 (S2). Diagram (a)
In order to create a screen like this, it is necessary to further register windows B and A, and in S3, register window A, that is, the last unused entry in the table with nothing above it. The entry closest to the end of the number, that is, the entry with window ID 3 is registered, and the registration ends.

Next, before issuing a display request for window C in S4, the application reads display data from the hard disk to the main memory. The argument for the request in this case is the window ID, that is, the window I that corresponds to window C.
It is a DI and is read as an argument for the start address of the main memory where display data is stored. In other words, the target of the request is an argument as to which window should be stored where in the main memory. When these requests are made to the window manager, the window manager performs display processing on the device and notifies the application of the displayed results in S5.

Window C will then be displayed.

When the above operations are similarly performed for windows B and A, frame buffer #1 is displayed. In this case, the display buffer flag is 0.

Next, when the contents of frame buffer #2 are to be displayed on the display, the display buffer flag is set to I to enable frame buffer #2. Then, the display is switched from frame buffer #1 to frame buffer #2 (S7).

The following routine is almost the same as the screen display processing for frame buffer #1, and in SP7, a request is made to open window C' in FIG. , the horizontal size.

When a request is made to the window manager, the window manager performs window display processing on the device. To this end, first in S9, the window manager registers the contents shown in FIG. 4(C) in the management table it is using.

That is, first, information on C' of window ID 4 is created in a table. Then, when the assigned window ID is notified to the application, the information for window D4, 5.6 is stored until the end of the table in Figure 4 (C), and it is stored as the last unused entry. Registration of the entry corresponding to window 106 is completed (310
). Then, in the 811 application, Fig. 2 (b
), the display data is first read from the hard disk to the main memory before issuing a display request for window C'. The arguments in this case are the window ID and the start address of the main memory where the display data is stored. Using the argument, the window manager performs display processing on the device and notifies the application result (312). In other words, first C
′ is displayed.

The above operations are similarly performed for the window ID and A' (S13), and the contents of frame buffer #2, that is, the display as the overlapping windows of FIG. 2(b) appear on the display (314).

Therefore, first we begin with Figure 2 of the contents of frame buffer #1 (
The display contents of a) are displayed, and then the contents of frame buffer #2, that is, the contents of Figure 2 (b) are displayed in place of it.
) screen will be displayed.

FIGS. 6) and 6(C) are embodiment diagrams showing the relationship between the overlapping windows of frame buffers #1 and #2 displayed according to the functional block diagrams of FIGS. 5(a) and 5(b).

Figure 6(C) shows the overlapping state of windows managed by the wind manager when frame buffer #l is displayed, and the area above the dotted line corresponds to the display in Figure 2(a). The windows C, B, and A are stacked in this order from the bottom. And this is what is currently displayed. Although not displayed on the screen, in frame buffer #2 there are windows A' and B' above C'. Overlapping windows are formed as an indication of an overlapping shape (in FIG. 2). Figure 6 (b
) is the opposite, and is the overlapping relationship of windows managed by the window manager when frame buffer #2 is displaying. In this case, what is currently displayed is frame buffer #2, so the display buffer flag is 1, and the one where this flag is 1 is on the upper side, that is, above the dotted line. , Figure 2 (b
) is displayed. Therefore, A' above the window C'
and B' overlap. On the other hand, the lower display buffer flag is currently 1, so the frame buffer #
2 is displayed, frame buffer #1 is not displayed but is at a lower level and is below the dotted line. That is, the display content of FIG. 2(a) is C.
It is shown that they overlap in the order of A.

FIG. 6(a) is a functional block diagram of the processing sequencer when switching frame buffers.

First, in S15, the application program issues a frame buffer switching request in order to instantly switch the display from the state shown in FIG. 2(b) to the state shown in FIG. 2(a). The device driver for the frame buffer receives the request, and in accordance with the request returns a response to the application in step 516 indicating that a buffer switching request has been issued to the frame buffer device. This concludes the frame buffer device driver routine. on the other hand,
When the application receives a notification that a buffer switching request has been issued to the frame buffer device, it performs the process in 517. That is, when the screen in FIG. If the display buffer flag is 1 during display, the window manager manages the windows as if they are overlapping, as shown in FIG. 6(b). Therefore, on the screen it is shown as (in Figure 1), but in order to display as shown in Figure 2 (a), the window manager should have the windows overlap as shown in Figure 6 (C). I have to make a decision, so I issue a request to bring Window C to the top. In this case, the only argument is the window ID, and as shown in the table, the argument for the window ID of window C is 1. This request is made to the wind manager. Then the wind manager
In S18, the management table of the wind manager is changed and managed so that C is placed at the top. The management of Window C is now complete, so when the application is notified of this, similar processing is performed for Window B and A as well. That is, in S19, the window manager determines that the windows B and A overlap as shown in FIG. 6(C). The management table is configured in this way. At this time, it is assumed that an event such as a mouse click will not be transmitted to a window that is against the window's will. Next, the application moves to S20 and changes the display buffer flag in FIG. 4(a) to 0, which means that buffer #1 is being displayed. At this time, all entries have already been stored in the management table, so by switching the display from Figure 2 (b) to (a), the contents of the management table, that is, Figure 4 (
FIG. 2(a) is displayed according to the window information in c) (S21).

FIG. 7 is a functional block diagram of a processing sequence when receiving a mouse click event according to the present invention. A routine for erasing window A as an event and rewriting a portion of windows B and C hidden below it after the erasure will be described as an embodiment. First, at 322, the window manager selects the currently displayed second window.
While displaying frame buffer #l on the screen in Figure (a), that is, window C with B and A overlapping, the user points to the point shown in Figure 2(a). P
When you click the mouse, the window manager notifies the application.

The arguments in this case are the window ID, that is, the window ID 3, and its coordinates, that is, the coordinates of the point P.

In S23, the application receives this argument, issues a request to erase window A in FIG. 2(a), and clears the information in the corresponding entry in FIG. 4(b). That is, the row of window ID 3 is cleared to make it a free area. In order to fill up the free space, the data table is first sorted, that is, the row with the window ID 3 is found and set as the free space. In this case, the only argument is the window ID. Then, in S24, the window manager executes management for the device to erase the display portion of window A. That is,
Notify the application that the relevant data has been deleted from the management table of the wind manager. Then, S
At 25, Window A disappears from the screen. After Wind A disappears, there is no part of it at all, so
Since parts of B and C hidden underneath are missing, it is necessary to show the missing parts of the windows corresponding to B and C. Therefore, at 326, the window manager notifies the application of an event indicating that after window A disappears from the screen, parts of B and C that were hidden beneath it have appeared.

The arguments at this time are the window IDs, 2 and 1 corresponding to B and C, and then the upper left coordinates and vertical and horizontal sizes of the rectangular area that appears are passed to the application. Then,
In S27, the application program issues a request to rewrite part of B in FIG. 2(a). In this case, the argument is the window ID, so 2 and 1 correspond to B and C.
The start address of the main memory where the display data is stored and the upper left coordinates and vertical and horizontal sizes of the rectangular area to be rewritten are passed to the window manager as arguments.

Then, in step 328, the window manager performs display processing for the corresponding area on the device, and notifies the application of the result of displaying parts of windows B and C that should be displayed there after window A disappears from the screen. This process is performed for window B, but the above operation is also performed for window C (S29). Then the process ends.

〔Effect of the invention〕

To provide a workstation user with a window manager environment capable of high-speed display switching without changing software such as O3, a window manager, a device driver, etc., and hardware such as a double frame batza memory, keyboard, mouse, etc.

[Brief explanation of drawings]

Figure 1 is a block diagram of the principle of the present invention. Figures 2 (a) and (b) are examples of screen displays on a workstation according to the present invention. Figure 3 is software related to the wind manager used in the present invention. and hardware component countries, Figure 4 (a)
is an example diagram of a display buffer flag used in the present invention,
FIGS. 4(b) and (C) are examples of the window information table, and FIGS. 5(a) and (b) are flowcharts for the processing sequencer up to displaying the screens in FIGS. 2(a) and Cb). , Figure 6 (a) is a diagram of the processing sequence when switching frame buffers, Figure 6 (middle) and (C) are overlapping frames of frame buffer windows displayed according to the flowcharts of Figures 5 (a) and (b). Figure 7 is a diagram showing the processing sequence when receiving a mouse click event, Figure 8 is a conceptual diagram of the conventional double frame buffer method, Figure 9 is a diagram showing the overlapping of windows in a conventional window manager. It is a conceptual diagram. 12...Display buffer flag storage means, 13.
... window information storage means, 14 ... control means.

Claims (1)

[Claims] 1) In a workstation having a double frame buffer memory, display buffer flag storage means (12) for storing a flag indicating that the content of one of the frame buffers is being displayed; Window information management storage means (
13), and a control means (14) for controlling processing for displaying each window, processing for switching frame buffers, and processing for receiving a mouse click event using the two means described above. Wind management method in use. 2) The window information management storage means (13) is characterized in that it has information on a window ID, window position, and window size for each window as window information in a table existing for each frame buffer. A window management method when using a double frame buffer according to claim 1. 3) The window management method when using a double frame buffer according to claim 1, wherein the window information in each table is stored in the table corresponding to the order of overlapping screens.