mth/ebiten
1
0
Fork 0
mirror of https://github.com/hajimehoshi/ebiten.git synced 2024-11-10 13:07:26 +01:00
Code Issues Packages Projects Releases Wiki Activity
6,384 Commits 43 Branches 315 Tags 148 MiB
d5ce0dbd90
Commit Graph

2880 Commits

Author SHA1 Message Date
Hajime Hoshi
f192971080 ebiten: Rename HasGamepadStandardLayoutMapping -> IsStandardGamepadLayoutAvailable 
Updates #1557
 2021-07-20 02:46:12 +09:00
Hajime Hoshi
aa694be6f6 ebiten: Add the standard gamepad layout 
This change introduces the standard gamepad layout. This changes adds
these APIs:

  * func HasGamepadStandardLayoutMapping
  * func IsGamepadStandardButtonPressed
  * func GamepadStandardAxisValue
  * type StandardGamepadButton
  * type StandardGamepadAxis

The standard gamepad layout is based on the web standard. See
https://www.w3.org/TR/gamepad/#remapping.

On desktops, the SDL's gamecontrllerdb.txt is used. If the gamepad is
listed in the text file, the mapping works. GLFW's mapping featrue is
not used.

On browsers, the property of a gamepad 'mapping' is used. When the
mapping value is 'standard', the gamepad is recognized to have the
standard mapping.

On mobiles, the implementation is still WIP.

Updates #1557
 2021-07-20 01:32:28 +09:00
Hajime Hoshi
fcfdf9b2ba Update URLs 2021-07-19 23:09:27 +09:00
Hajime Hoshi
a6fe7801ff internal/uidriver/glfw: Rename gamePad -> gamepad 2021-07-19 22:25:43 +09:00
Hajime Hoshi
dbcb220742 ebiten: Add GamepadAxisValue replacing GamepadAxis 
Updates #1718
Closes #1719
 2021-07-19 01:22:35 +09:00
Hajime Hoshi
342bf6ae7e Update GLFW: suppress joystick issue at PollEvents 
Updates #1229
 2021-07-15 11:19:39 +09:00
Hajime Hoshi
ec5b034cbf Update GLFW: suppress joystick issue (again) 
Updates #1229
 2021-07-15 01:02:57 +09:00
Hajime Hoshi
66dbca7fdd Update GLFW: suppress joystick issue 
Updates #1229
 2021-07-14 23:54:48 +09:00
Hajime Hoshi
50d2d7ed61 internal/graphicsdriver/opengl: Remove unused code using PBO 
Updates #1678
 2021-07-14 21:51:27 +09:00
Hajime Hoshi
99a6b1b03e internal/driver: Change some functions to Append* 
Updates #1692
 2021-07-10 01:52:21 +09:00
Hajime Hoshi
5d2c8ad9be internal/uidriver/mobile: Bug fix: Use mutex at RuneBuffers 2021-07-10 01:39:32 +09:00
Hajime Hoshi
853c1f2b92 internal/shaderir/glsl: Bug fix: Remove uncalled functions 
Some built-in functions like dFdx is not available in a vertex shader,
then a function that calls such built-in function should not be in
a vertex shader.

Closes #1701
 2021-07-09 20:22:40 +09:00
Hajime Hoshi
b0106e95b9 internal/restorable: Bug fix: Test failed on macOS (Metal) 2021-07-09 20:21:04 +09:00
Hajime Hoshi
4573883b03 internal/graphicsdriver/metal: Make FPS stable by 'presentsWithTransaction' 
Closes #1196
 2021-07-08 18:11:19 +09:00
Hajime Hoshi
ae0e30196b internal/graphicsdriver/metal: Bug fix: The store action must always be 'store' 
Closes #1700
 2021-07-08 17:39:37 +09:00
Hajime Hoshi
337d8d5113 internal/graphicsdriver/metal: Separate GCing the buffers 
Updates #1196
 2021-07-08 02:10:26 +09:00
Hajime Hoshi
edc2f8b961 internal/graphicsdriver/metal: Delay to get a texture for nextDrawable 
See https://developer.apple.com/documentation/quartzcore/cametallayer

> To avoid stalls in your app, request a new drawable only when you
> need it, and release any references to it as quickly as possible
> after you’re done with it.

Updates #1196
 2021-07-08 01:56:14 +09:00
Hajime Hoshi
b314b6b9b6 Revert "internal/graphicsdriver/metal: Do not retain MTLCommandBuffer for MTLBuffer" 
This reverts commit e0fbfc2bb0.

Reason: Rendering issue

Closes #1699
 2021-07-08 00:27:33 +09:00
Hajime Hoshi
6213c17abc internal/driver: Add Graphics.Initialize 2021-07-07 13:58:42 +09:00
Hajime Hoshi
519363930a internal/uidriver/glfw: Bug fix: Memory leak in Objective-C code 
Use an autorelease pool block.

Closes #1698
 2021-07-07 04:34:21 +09:00
Hajime Hoshi
e0fbfc2bb0 internal/graphicsdriver/metal: Do not retain MTLCommandBuffer for MTLBuffer 
Before this change, a command buffer is retained indirectly by
a buffer, and this might extend the life of drawable unexpectedly.

This change stops using command buffers as a key of the buffers pool,
and use a counter increated by nextDrawable calls.

Updates #1196
 2021-07-07 02:52:04 +09:00
Hajime Hoshi
80ac0646d5 internal/graphicsdriver/metal: Optimize the screen writing 2021-07-07 02:07:09 +09:00
Hajime Hoshi
5e83f409e6 internal/graphicscommand: Add a missing colon to a command string 2021-07-07 01:54:22 +09:00
Hajime Hoshi
38ce325958 internal/graphicsdriver/metal: Skip clearing the screen on Metal 2021-07-07 01:45:51 +09:00
Hajime Hoshi
9cb1ff9cea internal/graphicsdriver/metal: Set framebufferOnly true 
Updates #1196
 2021-07-07 00:38:55 +09:00
Hajime Hoshi
bd3f16dbba internal/graphicsdriver/metal: Change the order of GCed buffers 
As a big buffer is likely reused, we should remove smaller buffers
first.

Updates #1196
 2021-07-06 21:56:27 +09:00
Hajime Hoshi
09bd8b6f4a internal/graphicsdriver/metal: Remove println 
Updates #1196
 2021-07-06 21:51:57 +09:00
Hajime Hoshi
ee2f891fcc internal/graphicsdriver/metal: Reuse MTLBuffer objects 
In Metal, MTLBuffer objects are not 'transient' and are expensive
to create. Reuse them whenever possible.

See also: https://developer.apple.com/library/archive/documentation/Miscellaneous/Conceptual/MetalProgrammingGuide/Cmd-Submiss/Cmd-Submiss.html

Updates #1196
 2021-07-06 21:48:54 +09:00
Hajime Hoshi
be1a0e90e7 internal/graphicsdriver/metal: Bug fix: TestImageEvenOdd failed 
RenderCommandEncoder must be reset whenever the stencil mode is
'prepareStencil' in order to clear the stencil buffer.
 2021-07-06 15:11:23 +09:00
Hajime Hoshi
3670b7dd62 internal/graphicsdriver/metal: Reuse DepthStencilState objects 
In Metal, *State objects are not 'transient' and are expensive to
create. Reuse them whenever possible.

See also: https://developer.apple.com/library/archive/documentation/Miscellaneous/Conceptual/MetalProgrammingGuide/Cmd-Submiss/Cmd-Submiss.html
 2021-07-06 15:10:05 +09:00
Hajime Hoshi
ab26312108 internal/graphicsdriver/metal: Release objects appropriately 2021-07-06 14:20:19 +09:00
Hajime Hoshi
17d8cb5311 internal/graphicsdriver/metal: Bug fix: Crashed with METAL_DEVICE_WRAPPER_TYPE=1 
Closes #1697
 2021-07-06 14:06:06 +09:00
Hajime Hoshi
59fa689f22 internal/graphicscommand: Misspelling 2021-07-05 23:32:19 +09:00
Hajime Hoshi
5e1d6c06f1 internal/graphicscommand: Merge even-odd draw-triangles commands when possible 
Closes #1685
 2021-07-05 23:28:33 +09:00
Hajime Hoshi
63a00f6171 internal/graphicsdriver/metal/mtl: Bug fix: Test failure 
Closes #1696
 2021-07-05 21:28:25 +09:00
Hajime Hoshi
92b3dab33c internal/graphicscommand: Refactoring: Use type assesion instead of interface 
This was an optimization in GopherJS era. This change removes this
to simplify the implementation.
 2021-07-05 21:16:04 +09:00
Hajime Hoshi
b2d4521e22 internal/graphicscommand: Log 'even-odd' property 2021-07-05 19:35:34 +09:00
Hajime Hoshi
674802d2f5 ebiten: Bug fix: Draw commands with EvenOdd should not be merged 
Updates #1684
 2021-07-05 18:08:55 +09:00
Hajime Hoshi
daa883d799 ebiten: Bug fix: Stencil buffers should not be cleared until all the vertices are rendered 
Updates #1684
 2021-07-05 17:41:06 +09:00
Hajime Hoshi
b466a0cbd7 ebiten: Add EvenOdd to DrawTrianglesOptions and DrawShaderTrianglesOptions 
Updates #844
Closes #1684
 2021-07-05 03:35:55 +09:00
Hajime Hoshi
5c4d3325f6 internal/graphicsdriver/opengl: Add APIs for stencil buffers 
Updates #1684
 2021-07-05 00:24:44 +09:00
Hajime Hoshi
b8bf277825 internal/graphicsdriver/metal/mtl: Add APIs for stencil buffers 
Updates #1684
 2021-07-05 00:11:25 +09:00
Hajime Hoshi
19f034e818 internal/graphicsdriver/metal: Refactoring 2021-07-04 20:35:56 +09:00
Hajime Hoshi
3ec02f767b internal/graphicsdriver/metal: Refactoring 2021-07-04 18:37:14 +09:00
Hajime Hoshi
7f86761dde internal/graphicsdriver/metal: Reuse RenderCommandEncoder when possible 2021-07-04 18:13:18 +09:00
Hajime Hoshi
7668052a6b internal/graphicsdriver/metal: Update the view only when necessary 2021-07-04 17:24:39 +09:00
Hajime Hoshi
f6e210e086 internal/graphicsdriver/opengl: Refactoring: Remove redundant conversions 2021-07-03 22:39:48 +09:00
Hajime Hoshi
b0b7a24f89 internal/uidriver: Remove invalid comments 2021-07-03 19:49:12 +09:00
Hajime Hoshi
56f4e6b2d0 internal/graphicsdriver/opengl: Refactoring: Rename a variable 2021-07-03 02:41:37 +09:00
Hajime Hoshi
269ea7c489 internal/graphicsdriver/opengl: Refactoring: Reorder functions 
glPixelStorei affects the result of glTexImage2D. This change makes
this fact more explicit.
 2021-07-03 02:09:47 +09:00
Hajime Hoshi
6b95dd0d4d internal/graphicsdriver/metal: Refactoring: Unify two drawTriangles* functions 2021-07-02 00:25:43 +09:00
Hajime Hoshi
ff2dd6fe7e internal/graphicsdriver/opengl: Refactoring: Unify two drawTriangles* functions 2021-07-02 00:02:21 +09:00
Hajime Hoshi
e9b6237f61 internal/driver: Refactoring: Merge two Draw* functions 2021-07-01 15:55:48 +09:00
Hajime Hoshi
b9d52c0267 internal/driver: Change InvalidImageID/InvalidShaderID to 0 2021-07-01 14:58:55 +09:00
Hajime Hoshi
73bf1f36e5 internal/driver: Define InvalidImageID / InvalidShaderID 2021-07-01 13:56:42 +09:00
Enrico
3ef1d04935
internal/uidriver/glfw: Bug fix: do not execute loop function if init failed (#1689) 
In `internal/uidriver/glfw/run_notsinglethread.go`, if the `UserInterface.init()` function returns an error, the `loop` is
executed regardless and the error is discarded. This behavior will hide the error returned by `init()` and might trigger
some crashes (see #1688).

A partial fix was implemented in 6c4edf8 , however that commit alone is not enough: the code now is correctly returning
the error via the `ch` channel, but it still executes the `loop()` function. This merge request skips `loop()` call if `init()`
had an error.

Updates #1688
 2021-06-29 23:51:25 +09:00
Hajime Hoshi
6c4edf8605 internal/uidriver/glfw: Bug fix: Abort the game immediately on error 
Updates #1688
 2021-06-29 02:23:17 +09:00
Hajime Hoshi
9c5b7b56b2 internal/graphics: Reafctoring 2021-06-28 23:28:43 +09:00
Hajime Hoshi
817d176994 internal/graphicscommand: Misspelling at newShaderCommand.String 2021-06-27 23:20:27 +09:00
Hajime Hoshi
c725937cc6 internal/graphics: Reuse the vertices backend 
This is basically a reland of 9cb631e30f.
This change locks the vertices backend at the end-frame phase to
protect from vertices usages by other goroutines.

Updates #1546
Closes #1681
 2021-06-27 02:31:01 +09:00
Hajime Hoshi
699bb095c3 internal/atlas: Better algorithm to release the temporary pixels 2021-06-26 23:29:45 +09:00
Hajime Hoshi
067e3c004d Remove internal/web 2021-06-26 17:04:00 +09:00
Hajime Hoshi
8ff42f55a1 internal/graphicsdriver/opengl: Give up restorign on browsers 
Using restorable images is expensive as this sometimes requires to
load pixels from GPU. On browsers, just reloading the application
should be fine when the context lost happens.

Closes #1603
 2021-06-26 16:54:21 +09:00
Hajime Hoshi
38ce46328a internal/atlas: Reuse byte array for ReplacePixels 
This reduces memory allocation at ReplacePixels.

Updates #1681
 2021-06-26 03:48:25 +09:00
Hajime Hoshi
414a64ce78 internal/atlas: Allow extending a texture atlas 
As PBO is no longer used (#1678), using a big atlas with
ReplacePixels is no longer problematic.

Closes #1674
Updates #1675
 2021-06-24 21:35:36 +09:00
Hajime Hoshi
2a8e8d0eb6 internal/graphicsdriver/opengl: Stop using PBO 
PBO was introduced to improve the performance at ReplacePixels,
but we found that PBO can degrades the performance. Also, now
multiple glTexImage2Ds are called successively like a batch, so
the situation is now different from that time when PBO was
introduced.

Let's remove PBO usages and wait and see.

Closes #1678
 2021-06-24 21:20:58 +09:00
Hajime Hoshi
7f60cd41eb internal/graphicsdriver/opengl: Refactoring: Remove unused arguments from texSubImage2D 2021-06-24 12:06:28 +09:00
Hajime Hoshi
fa8e271f2a internal/atlas: Bug fix: Set the minimum texture size to 8192px 
Closes #1675
Updates #1674
 2021-06-19 18:52:08 +09:00
Hajime Hoshi
43c964a3b8 internal/thread: Refactoring 2021-06-15 00:43:48 +09:00
Hajime Hoshi
49c3c30c79 ebiten: Add IsWindowBeingClosed / SetWindowClosingHandled / IsWindowClosingHandled 
IsWindowBeingClosed reports whether the window is being closed by
the user.

SetWindowClosingHandled sets whether the window closing is handled
or not. If the state is true, the window is not closed immediately
by the user and the game can handle the closing state. In this case,
the Update function should return an error in order to end the game.

This change also adds examples/windowclosing.

Closes #1574
 2021-06-14 02:29:11 +09:00
Hajime Hoshi
f989ce4e64 internal/uidriver/glfw: Rename variables 2021-06-13 23:25:24 +09:00
Hajime Hoshi
bceb512e78 internal/glfw: Bug fix: build failure with GOOS=js 
Updates #1672
 2021-06-13 22:59:57 +09:00
Hajime Hoshi
ed6756662a internal/glfw: Clean up the code 
Updates #1672
 2021-06-13 22:48:27 +09:00
Hajime Hoshi
3c2d562967 internal/glfw: Bug fix: NewCallbackCDecl objects were leaked 
Callbacks created by NewCallbackCDecl were never released and then
they are leaked, especially when the window size was changed by
SetWindowSize on Windows.

This change defines new callback ID types with uintptr, and reuse
the callbacks.

Closes #1672
 2021-06-13 22:23:01 +09:00
Hajime Hoshi
0cea5f2f1a internal/glfw: Remove one of go-vet warnings 2021-06-13 14:41:39 +09:00
Hajime Hoshi
584c6524ee internal/graphicsdriver/opengl: Refactring 2021-06-13 14:19:33 +09:00
Hajime Hoshi
b54ad73a2b Add go:build comments with go1.17beta1 fmt 2021-06-11 01:11:46 +09:00
Hajime Hoshi
a627c41217 Remove support of Go 1.13 and Go 1.14 
Updates #1258
Updates #1415
Updates #1462
 2021-06-07 21:44:22 +09:00
Hajime Hoshi
040fcd930f Revert "Revert 2 commits using ColorM cache" 
This reverts commit dd7a8bc497.

Reason: DrawTriangles worked fine for the reporter's purpose.

Updates #1655
 2021-05-30 15:17:11 +09:00
Hajime Hoshi
dd7a8bc497 Revert 2 commits using ColorM cache 
This reverts commit 87c8625692.
This reverts commit a6dd6196b4.

Reason: Performance regression (#1655)

Closes #1655
 2021-05-30 03:10:00 +09:00
Hajime Hoshi
58843b68f9 internal/hooks: Enable to return error at suspend/resume audio 2021-05-04 23:09:01 +09:00
Tom Lister
bea5ab3335
internal/uidriver/glfw: Native macOS implementation for setting cursor shape (#1624) 
Updates #1624
 2021-05-02 14:50:50 +09:00
Hajime Hoshi
b509ce523d internal/uidriver/js: Bug fix: SetFullscreen didn't work on Safari 
Updates #1611
 2021-04-28 23:28:58 +09:00
Tom Lister
5d3a76cbe6
internal/uidriver/js: Implement {Set,}Fullscreen for browsers (#1623) 
Closes #1611
 2021-04-27 23:58:32 +09:00
Hajime Hoshi
091e51ae4c internal/uidriver/glfw: Avoid allocating a new slice 2021-04-25 00:57:06 +09:00
Hajime Hoshi
e2ceb7b3a6 internal/uidriver/glfw: Call fromGLFWPixels from the main thread 
fromGLFWPixels can invoke driverScaleFactor, and then using the main
thread is safer.
 2021-04-24 23:30:51 +09:00
Hajime Hoshi
b6364c46fe internal/clock: Misspelling 2021-04-24 23:24:11 +09:00
Hajime Hoshi
2534b78755 internal/clock: Simplify the logic to calculate tick per frame 
Updates #1444
Updates #1597
 2021-04-24 23:06:27 +09:00
Hajime Hoshi
2e6ecc766d internal/clock: Integrate 'now' implementations 
The Go runtime already uses QueryPerformanceCounter at runtime.nanotime.
 2021-04-24 22:32:47 +09:00
Hajime Hoshi
011a524ac7 internal/clock: Update comments 2021-04-24 22:10:00 +09:00
Hajime Hoshi
91f8347500 internal/uidriver/glfw: Avoid busy loop by sleeping 
Updates #1580
 2021-04-24 05:19:23 +09:00
Hajime Hoshi
3df31c0fce internal/uidriver/glfw: Bug fix: Deadlock at FramebufferSize callback 
glfw.PollEvents might invoke multiple FramebufferSize callbacks in
theory, this is very rare though. In this case, the sending an object
to the channel never ends.

This change fixes this deadlock by using 'select'.

Closes #1618
 2021-04-24 02:22:51 +09:00
Hajime Hoshi
ac9a18931c internal/uidriver/glfw: Fix comments 
Updates #1618
 2021-04-23 03:28:14 +09:00
Hajime Hoshi
d446aa39a9 internal/uidriver/glfw: Bug fix: Always wait for SetFramebuffer 
In case when the callback is not called, let's use timeout.

Updates #1618
 2021-04-23 03:02:12 +09:00
Hajime Hoshi
c4c331b0d5 Fix misspellings 2021-04-21 22:50:00 +09:00
Hajime Hoshi
4cb6cb9d6a internal/uidriver/glfw: Misspelling 2021-04-21 22:29:54 +09:00
Hajime Hoshi
1013ca9c66 internal/uidriver/glfw: Bug fix: A window can never be 'maximized' on fullscreen 
When the window is fullscreen, calling MaximizeWindow never returned.
Apparently the attribute glfw.Maximized can never be true on the
fullscreen mode.

This change fixes the issue by checking the fullscreen state.
 2021-04-21 22:25:14 +09:00
corfe83
d24313e236
internal/devicescale: Fallback to simpler logic when trying to get DPI from monitor and failing (#1600) 
Closes #1612
 2021-04-21 11:43:23 +09:00
Hajime Hoshi
c7dd9a4107 internal/uidriver/glfw: Bug fix: Wait for Restore/Maximize/Iconify finishes explicitly 
On Linux (and probably other UNIX OSes), glfwRestore, glfwMaximize,
and glfwIconify returns even though the window state has not changed.
Wait for its finish by a for loop and glfw.PollEvents.

Closes #1608
 2021-04-21 00:52:06 +09:00
Hajime Hoshi
58572fb000 internal/uidriver/glfw: Bug fix: Unexpected small window size on Linux 
At the initial phase, GLFW window's GetSize seems unreliable. Do not
call it on Linux. On the other hand, this is still necessary on macOS.

Updates #1606
Closes #1609
 2021-04-20 23:38:34 +09:00