8b84ddce9a
VideoCommon: rewrite frame skipping code
2014-09-04 18:07:39 +02:00
ef6f6a7fa9
VideoCommon: remove XFReg copy optimization
...
This code is just ugly and I doubt there is a way that copying twice is faster.
2014-09-04 17:56:17 +02:00
86db0bf8c3
Added the EmuRunningState check to the GPU thread's FIFO loop so that the GPU thread services any waiting save states. This is needed for games that have the Sync GPU option enabled.
2014-09-04 22:02:21 +10:00
dd5be7c0dc
Merge pull request #924 from comex/fifo-command-runnable
...
Refactor opcode decoding a bit to kill FifoCommandRunnable.
2014-09-02 23:27:30 -04:00
266992684d
msvc: remove some remnants of SDL and DSound from projects and general cleanup.
2014-09-01 21:27:44 -07:00
b51aa4fa89
Rename Log2 and add IsPow2 to MathUtils for future use
...
Also remove unused pow2/pow2f functions.
2014-09-01 20:41:07 -07:00
608f9bcd67
Refactor opcode decoding a bit to kill FifoCommandRunnable.
...
Separated out from my gpu-determinism branch by request. It's not a big
commit; I just like to write long commit messages.
The main reason to kill it is hopefully a slight performance improvement
from avoiding the double switch (especially in single core mode);
however, this also improves cycle calculation, as described below.
- FifoCommandRunnable is removed; in its stead, Decode returns the
number of cycles (which only matters for "sync" GPU mode), or 0 if there
was not enough data, and is also responsible for unknown opcode alerts.
Decode and DecodeSemiNop are almost identical, so the latter is replaced
with a skipped_frame parameter to Decode. Doesn't mean we can't improve
skipped_frame mode to do less work; if, at such a point, branching on it
has too much overhead (it certainly won't now), it can always be changed
to a template parameter.
- FifoCommandRunnable used a fixed, large cycle count for display lists,
regardless of the contents. Presumably the actual hardware's processing
time is mostly the processing time of whatever commands are in the list,
and with this change InterpretDisplayList can just return the list's
cycle count to be added to the total. (Since the calculation for this
is part of Decode, it didn't seem easy to split this change up.)
To facilitate this, Decode also gains an explicit 'end' parameter in
lieu of FifoCommandRunnable's call to GetVideoBufferEndPtr, which can
point to there or to the end of a display list (or elsewhere in
gpu-determinism, but that's another story). Also, as a small
optimization, InterpretDisplayList now calls OpcodeDecoder_Run rather
than having its own Decode loop, to allow Decode to be inlined (haven't
checked whether this actually happens though).
skipped_frame mode still does not traverse display lists and uses the
old fake value of 45 cycles. degasus has suggested that this hack is
not essential for performance and can be removed, but I want to separate
any potential performance impact of that from this commit.
2014-09-01 14:35:23 -04:00
494a60e41b
VertexLoader: Change VtxDesc to use u64 instead of u32
...
This is required to make packing consistent between compilers: with u32, MSVC
would not allocate a bitfield that spans two u32s (it would leave a "hole").
2014-09-01 11:18:02 +02:00
4af8d9d248
VideoCommon: Clean up brace placements
2014-08-30 18:06:45 -04:00
1d706b2311
Get rid of C-style empty function parameter indicators
2014-08-30 15:23:48 -04:00
a4a533e39f
Re-enable the vertex loader JIT on OS X.
...
Why was it ever disabled?
2014-08-27 23:50:59 -04:00
e31d6feaa2
Unify three types of non-FIFO requests to the GPU thread around Common::Event and Common::Flag.
...
The only possible functionality change is that s_efbAccessRequested and
s_swapRequested are no longer reset at init and shutdown of the OGL
backend (only; this is the only interaction any files other than
MainBase.cpp have with them). I am fairly certain this was entirely
vestigial.
Possible performance implications: efbAccessReady now uses an Event
rather than spinning, which might be slightly slower, but considering
the slow loop the flags are being checked in from the GPU thread, I
doubt it's noticeable.
Also, this uses sequentially consistent rather than release/acquire
memory order, which might be slightly slower, especially on ARM...
something to improve in Event/Flag, really.
2014-08-26 12:43:39 -04:00
45a4236283
A tiny restructuring to allow inlining of FifoCommandRunnable. Probably useless.
2014-08-26 12:43:39 -04:00
14125cf951
Refactor SetCpStatus into two functions for from-GPU and from-CPU mode rather than a boolean parameter.
...
This shouldn't affect functionality. I'm not sure if the breakpoint
distinction is actually necessary (my commit messages from the old
dc-netplay last year claim that breakpoints are broken anyway, but I
don't remember why), but I don't actually need to change this part of
the code (yet), so I'll stick with the trimmings change for now.
2014-08-26 12:43:39 -04:00
07c7e6f35e
CommandProcessor: mark some functions as static
2014-08-25 21:09:42 +02:00
bf93920c05
Revert "Catch broken configurations inside of the Post Processing shaders."
2014-08-25 14:33:41 +02:00
2f2f992bc7
Merge pull request #828 from Sonicadvance1/pp-shader-catch-broken-config
...
Catch broken configurations inside of the Post Processing shaders.
2014-08-25 09:17:30 +02:00
327d35377d
windows: remove now-extraneous NOMINMAX and WIN32_LEAN_AND_MEAN #defines from dolphin code.
...
Wrap dinput.h in a header defining DIRECTINPUT_VERSION instead of repeating it multiple places.
2014-08-23 10:48:48 -07:00
f17dcd2019
Merge pull request #764 from magcius/new-nogui-2
...
Rewrite GLInterface
2014-08-21 14:14:54 -04:00
4bf031c064
msvc: resolve all warnings in VideoCommon.
2014-08-19 22:33:46 -07:00
63f1a16969
Core: Remove UpdateFPSDisplay
...
This is effectively unused, as the window handles that we pass to the
GLInterface are window handles for the frame which isn't ever a real
toplevel window. Host_UpdateTitle is what actually sets the proper title
on the render window.
2014-08-19 10:05:58 -04:00
7ca8d8dfc7
Core: Don't pass through a reference to the window handle
...
Now that MainNoGUI is properly architected and GLX doesn't need to
sometimes craft its own windows sometimes which we have to thread back
into MainNoGUI, we don't need to thread the window handle that GLX
creates at all.
This removes the reference to pass back here, and the g_pWindowHandle
always be the same as the window returned by Host_GetRenderHandle().
A future cleanup could remove g_pWindowHandle entirely.
2014-08-19 10:05:58 -04:00
7d90a00bbe
Qualcomm fixed screen rotation in their latest v66 development drivers.
...
The framebuffer is no longer rotated the wrong way around in Qualcomm's latest development drivers.
They did something right, only took them over a year.
2014-08-18 00:17:09 -05:00
2d624780c0
Catch broken configurations inside of the Post Processing shaders.
...
This catches most instances of configuration failures that can happen in a post processing shader.
Gives a user a helpful error message that lets them know what they have failed to set up correctly
2014-08-17 23:59:21 -05:00
2270c3e90a
Merge pull request #797 from shuffle2/msvc-pch
...
Windows: Use a shared precompiled header for dolphin code under Source/
2014-08-16 14:58:28 -07:00
6f34a8ac47
Removed warnings by assigning to bool
2014-08-16 14:16:10 -05:00
a64b0bf499
VertexLoader: cache NativeVertexFormat
...
This fix a performance regression of PR #672 .
2014-08-16 12:58:52 +02:00
f1b82a34b2
Windows: Use a shared precompiled header for dolphin code under Source/
2014-08-14 23:51:13 -07:00
b8a21b3744
Add the PostProcessing class object to RenderBase in VideoCommon.
...
Backends will initialize this variable with their own inherited PostProcessing class object.
2014-08-13 01:05:14 -05:00
6bdc32c54a
Add the VideoCommon PostProcessing class.
...
This class loads all the common PP shader configuration options and passes those options through to a inherited class that OpenGL or D3D will have.
Makes it so all the common code for PP shaders is in VideoCommon instead of duplicating the code across each backend.
2014-08-13 01:05:10 -05:00
c54fef5496
VideoBackendBase: Remove unused stub Initialize implementation
...
Both D3D and OGL have their own overrides, so this isn't used.
2014-08-06 21:35:52 -04:00
16f180524c
VertexLoader: do not prepare for vertices if we need to skip them
2014-08-04 20:47:02 -07:00
15920d0f10
Merge pull request #394 from degasus/d3d_lighting_fix
...
VideoCommon: normalize light direction
2014-08-03 21:21:23 -07:00
4c42b38de1
Merge pull request #428 from Sonicadvance1/x86_32-removal
...
Remove x86_32 support from Dolphin.
2014-08-03 21:17:28 -07:00
d9b5482840
Remove x86_32 from VertexLoader.
2014-08-03 13:44:37 -05:00
6f715a1fbe
VertexLoader: Remove more global state dependencies (this time IndexGenerator and VertexManager)
2014-08-02 09:34:39 -07:00
83838a645f
Merge pull request #690 from Armada651/d3dfullscreen_fixes
...
Exclusive fullscreen fixes
2014-07-30 16:28:56 -07:00
3b5625c76b
VideoConfig: Ignore Borderless Fullscreen setting when the backend does not support exclusive fullscreen.
...
This was expected to be handled by VerifyValidity(), but that only verifies the validity of the INI files.
2014-07-30 12:15:58 +02:00
4501aeefbe
CFrame: Check borderless fullscreen setting before enabling exclusive fullscreen in the video config.
...
Fixes a bug where "Use Fullscreen" would initialize into exclusive fullscreen regardless of the borderless fullscreen setting.
Also relieves the need for the video renderer to check the borderless fullscreen setting each time.
2014-07-30 12:15:26 +02:00
522a5c35ad
Convert some more header inclusions into forward declarations
2014-07-29 20:55:07 -04:00
ec402a0d5f
FPSCounter: Initialize members.
2014-07-26 14:37:18 +02:00
8e865f3848
Merge pull request #506 from Armada651/d3dfullscreen
...
D3D: Add exclusive fullscreen support.
2014-07-26 13:22:11 +02:00
6724ce6275
Cosmetic changes based on feedback on PR #506 .
2014-07-26 13:04:39 +02:00
bd9953d97e
Remove the 3D Vision hack.
...
The hack was needed because the Nvidia 3D Vision heuristics are documented to only support surfaces that are the same size as the backbuffer. This would be the case if you enabled the hack and selected the "Auto (Window Size)" internal resolution.
However, on recent drivers the same effect is achieved by selecting the "Auto (Multiple)" internal resolution. Therefore the hack is no longer required.
2014-07-26 12:45:10 +02:00
73f9a22e2e
VertexLoader: Remove global state dependency on g_nativeVertexFmt
2014-07-26 01:35:09 +02:00
78c3a22060
VertexLoader: take the VAT object directly for RunVertices
2014-07-24 01:51:37 +02:00
069801a7d1
VertexLoader: Simplify SetVAT
2014-07-24 01:25:23 +02:00
20369743a4
VertexLoaderUID: remove global state dependency
2014-07-24 01:12:12 +02:00
009b4dd376
Exit exclusive fullscreen when the stop confirmation is shown.
...
Also have the renderer remember its own fullscreen state. This is done to prevent a case where we exit exclusive fullscreen through the configuration and a focus shift at the same time. In this case the renderer would fail to detect that the fullscreen state was changed.
2014-07-21 20:50:48 +02:00
cd94ff1966
VideoConfig: Add "Borderless Fullscreen" option.
...
This option will disable exclusive fullscreen for users who prefer the old behaviour.
2014-07-20 22:02:57 +02:00
77bc879384
D3D: Add exclusive fullscreen support.
2014-07-19 21:14:44 +02:00
bc9ef95643
Support Sampler binding in the shader.
...
In the cases where we support the binding layout keyword, use it for more than binding UBO location.
This changes it so it is supported for samplers as well.
Instances when this is enabled is if a device supports GL_ARB_shading_language_420pack, or if it supports GLES 3.10.
2014-07-18 17:04:03 -05:00
b9dc69105d
Merge pull request #595 from Armada651/pref_log
...
FPSCounter: Flush the logs every second and close them when the renderer is shut down.
2014-07-18 12:59:04 +02:00
eaa7460636
FPSCounter: Remove redundant destructor.
2014-07-18 12:49:40 +02:00
3b978f7c27
Turn the FPSCounter namespace into a class.
2014-07-16 20:40:40 +02:00
01fd96ab31
PixelShaderGen: fix indentation
2014-07-16 17:24:43 +02:00
4063694d20
VideoCommon: fix ifdef expression
2014-07-15 04:15:49 +02:00
0c6eeaff05
Merge pull request #617 from Tilka/clang_bug
...
VideoCommon: fix clang version check
2014-07-14 01:55:31 -07:00
dbc30c6c76
VideoCommon: make version check easier to read
2014-07-14 03:05:56 +02:00
0be03252cc
VideoCommon: fix clang version check
...
That was... er... a typo!
2014-07-14 02:59:31 +02:00
3f67ec0d50
Merge pull request #611 from Tilka/clang_bug
...
VideoCommon: version-check clang for workaround
2014-07-13 17:52:54 -07:00
b6f3ae23bc
VideoCommon: version-check clang for workaround
...
The bug was fixed in clang 3.4.
2014-07-14 02:12:48 +02:00
8876ee120a
Change libav* autodetection to support framedumping on Ubuntu 14.04
...
Add an "ugly" workaround in the AVIDump code, but looking at other project this
seems to be the most common way to handle this API change.
2014-07-13 23:06:20 +02:00
0ccee6c87b
Fix warnings unearthed by #579
2014-07-13 02:16:51 +02:00
7e79806efc
remove unused globals
...
Also change globals into statics which are only used in one file
2014-07-11 16:10:20 +02:00
81ed17be53
avoid the extern keyword in .cpp files
2014-07-11 16:10:20 +02:00
6d3f249dcc
mark all local variables as static
2014-07-11 16:10:20 +02:00
22e1aa5bb4
mark all local functions as static
2014-07-11 16:07:23 +02:00
6def4ead01
FPSCounter: Flush the logs every second and close them when the renderer is shut down.
2014-07-10 23:11:28 +02:00
1754cbda9d
Move FPSCounter calls to RenderBase.
2014-07-10 23:11:09 +02:00
15c1250d9d
Merge pull request #596 from delroth/master
...
AVIDump: fix FFV1 encoding
2014-07-09 18:02:40 -07:00
da697df6ee
AVIDump: fix FFV1 encoding
...
ffmpeg 2.0 changed requirements for the FFV1 encoder and made them more strict,
requiring more fields of the input frame to be initialized. Explicitly setting
pixfmt, width and height solve the EINVAL issues with FFV1 encoding.
Original fix from http://ffmpeg.org/pipermail/libav-user/2013-October/005759.html
2014-07-10 02:53:12 +02:00
09304cab57
FPSCounter: Change format string to match value.
2014-07-09 19:45:56 +02:00
95b579746f
Replace "Log FPS to file" by the "Log render time to file" feature.
2014-07-09 17:56:11 +02:00
61d44cf73f
FPSCounter: Use a Timer for the FPS update time.
2014-07-09 17:53:41 +02:00
efeadb7fe9
FPSCounter: Add "Log render time to file" feature.
...
Allows for a more accurate performance measurement.
2014-07-09 17:53:31 +02:00
ec1e52de53
VideoCommon: Get rid of an snprintf call in VideoConfig.cpp
2014-07-06 15:33:08 -04:00
cc3dda5b22
Merge pull request #362 from Tilka/ffmpeg_libav_new
...
AVIDump: use new ffmpeg/libav API
2014-07-06 19:33:27 +02:00
a798548c30
Merge pull request #546 from workhorsy/header_guard_to_pragma_once
...
Changed lingering header include guards to pragma once.
2014-07-06 14:19:32 +02:00
4483b64bcb
Merge pull request #463 from degasus/vertex_format_cache
...
VideoCommon: Cache native vertex formats
2014-07-06 05:26:42 -05:00
48ff45b8a8
VideoCommon: Remove some unused constants from VertexShaderGen.h.
2014-07-05 23:46:07 -04:00
bb2fc8ecbb
VideoCommon: Cache native vertex formats
...
We are used to have a 1:1 mapping of GX vertex formats and the native (OGL + D3D) ones, but there are by far more GX ones.
This new cache maps them directly so that we don't flush on GX vertex format changes as long as the native one doesn't change.
The idea is stolen from galop1n.
2014-07-04 14:39:27 +02:00
02ac5e95c8
VideoCommon: normalize lighting direction.
...
It seems that the lighting direction must be normalized. This fixes lots of lighting issues mostly shown in the d3d backend.
2014-07-03 21:08:19 +02:00
00efaedb02
FPS counter cleanup
...
- Isolate it into it's own namespace
- Shorten function names, the namespace self-documents.
- Just use the std I/O, we can just write directly to the stream for
logging.
2014-07-02 20:23:09 -04:00
124210c50f
Changed lingering header include guards to pragma once.
...
Some headers where using #ifndef to guard being including multiple times. But most were using pragma once. So for consistency I changed them all to use pragma once.
2014-07-01 22:17:33 -07:00
6b3e6e6ffb
AVIDump: rename frame variables
2014-06-27 19:48:36 +02:00
c2c46d7573
AVIDump: update ffmpeg/libav API usage
...
libav 10 was released on May 10th, 2014 and it drops support for some
long-deprecated stuff like avcodec_encode_video().
2014-06-27 19:48:36 +02:00
e3fef8c990
AVIDump: cleanup
2014-06-27 19:48:35 +02:00
7db5a4b22d
Statistics: Reformat stats string
2014-06-27 09:36:50 +02:00
f1ddd3c66a
VideoCommon: remove unused stats
2014-06-27 09:35:26 +02:00
ca5340ebde
Centralize the logging code into its own folder in Common.
2014-06-25 22:11:42 -04:00
8b13afbb8e
Remove the 32-bit config platform from the VS solution and projects
2014-06-24 22:07:26 -04:00
5dff577339
Merge pull request #500 from lioncash/ini
...
Use only section-based ini reading.
2014-06-22 17:21:45 +02:00
924ad1ee9f
LightingShader: hard code const variable
2014-06-19 16:46:53 +02:00
e456a5e64f
PixelShader: remove the duplicated ppl constants
2014-06-19 16:33:33 +02:00
d93f2973f7
PixelShader: use the vertex const buffer for ppl
2014-06-19 16:33:33 +02:00
027baad73b
VideoCommon: use the Light struct in XF memory
2014-06-19 16:33:29 +02:00
ce54c1e571
Kill off replaceable usages of s[n]printf.
2014-06-18 19:53:38 -04:00
9d87818853
VideoCommon: small cleanups. No functional change.
2014-06-16 14:03:29 -07:00
f05d3f6e5d
Use only section-based ini reading.
2014-06-16 01:31:23 -04:00
020b4fde1e
Merge pull request #492 from Armada651/master
...
Remove EmuWindow
2014-06-15 14:16:53 +02:00
f2759ffe65
Remove EmuWindow.
...
All it did was raise complexity.
2014-06-15 00:49:49 +02:00
d7736ac714
Merge pull request #445 from magumagu/video-dead-code
...
Video backends: remove dead code.
2014-06-13 22:38:22 +02:00
a8f32adcab
Merge pull request #470 from pauldacheez/s-Gamecube-GameCube
...
Fix the capitalization of "GameCube" throughout the project.
2014-06-13 21:52:08 +02:00
0bc6b49c07
Merge pull request #483 from neobrain/bitfield_fixes
...
BitField fixes
2014-06-13 20:43:02 +02:00
f87b913f0e
Fix the vertexloader on non-x86 targets.
...
When I dropped ARM from a generic target, this caused the vertexloader to try using the JIT path.
Instead of !_M_GENERIC, check for _M_X86 instead. Since it is only for the x86 target
2014-06-13 18:36:54 +00:00
78fbf2ecaa
Fix a few warnings caused by using BitField with non-typesafe functions.
2014-06-11 20:58:40 +02:00
c38e8e25dd
Remove unused video config variable.
2014-06-09 00:02:38 -04:00
5d793881b0
Fix the capitalization of "GameCube" throughout the project.
2014-06-08 11:24:49 +09:00
6e1d312091
Make it so ARMv7 isn't a generic target.
...
Rearranges a bit of code so that ARM isn't a generic build anymore. Because it obviously isn't
2014-06-07 20:26:31 -05:00
b6db0d0ab8
Merge pull request #457 from Tilka/jcc
...
x64Emitter: J_CC: use 32 bit offset automatically
2014-06-06 20:53:50 -05:00
834bc85e1f
Merge pull request #458 from Tilka/warnings
...
Fix some warnings
2014-06-06 00:58:37 +02:00
0eb341e542
Fix some warnings.
...
The 'offset' variable needs to stay because the disabled code below uses it.
2014-06-04 01:04:52 +02:00
f8280401f6
x64Emitter: J_CC: use 32 bit offset automatically
2014-06-03 23:08:58 +02:00
776e36b10a
Fix a typo in a BP register name (BPMEM_TX_SETLUT_4 -> BPMEM_TX_SETTLUT_4).
...
Also fixed the alignment of the register values.
2014-06-02 02:26:30 -04:00
12db989098
Add missing registers in GetBPRegInfo
2014-06-02 02:19:53 -04:00
0aecd9504e
Video backends: remove dead code.
2014-06-01 01:56:09 -07:00
49b0eef393
Remove the min/max functions in CommonFuncs.
...
The algorithm header has the same functions.
2014-05-29 21:44:41 -04:00
3097345929
Merge pull request #411 from lioncash/fifo-cleanup
...
Make GetBPRegInfo just take two strings as parameters
2014-05-30 01:51:11 +02:00
c96407bd2a
Make GetBPRegInfo just take two strings as parameters
...
Gets rid of the size parameters.
2014-05-29 19:44:14 -04:00
1583ce9363
Use strings instead of arbitrary buffers for video statistics
2014-05-25 21:11:29 -04:00
b58753bd69
Merge pull request #370 from Sonicadvance1/remove_specialized_memcmp
...
Removes ZeroFrog's "optimized" memcpy and memcmp functions.
2014-05-22 13:02:11 -07:00
9d161b4170
BPStructs: Consistently put the two shared copy args first
...
And rename them so they make a bit more sense.
2014-05-20 11:28:15 -04:00
1ae8edc1d0
BPStructs: Remove another function wrapper
2014-05-20 11:28:15 -04:00
b1d3c5937a
BPStructs: Move LoadBPReg here
2014-05-20 11:28:14 -04:00
763ad77a1c
BPStructs: Flatten out BPWritten
2014-05-20 11:28:14 -04:00
07ab77d31c
BPStructs: Reindent BPWritten
2014-05-20 11:28:08 -04:00
c33a1b4b28
BPStructs: Document BPMEM_BP_MASK better
2014-05-20 11:26:31 -04:00
2f122ea63c
BPMemory: Fix "DISPLAYCOPYFILER" typo
2014-05-20 11:15:10 -04:00
4e8e51b278
BPStructs: Remove calls to SetInterlacedMode when reloading state
...
SetInterlacedMode is a dummy no-op that does nothing.
2014-05-20 11:15:10 -04:00
833b7ee584
BPFunctions: Remove the rest of GetConfig
2014-05-20 11:15:09 -04:00
08611c3f36
PixelShaderManager: Fizzle out fog changes when disabled here
...
This lets us remove a use of GetConfig.
2014-05-20 11:15:09 -04:00
fe645b888b
BPFunctions: Remove use of a dumb method
...
GetPointer serves no purpose.
2014-05-20 11:15:08 -04:00
56228cea5d
BPFunctions: Remove an unused method
2014-05-20 11:15:08 -04:00
4f4e3fad85
BPMemory: Clean up and rename some variables and comments
...
This is about the register number, not the opcode.
2014-05-20 11:15:08 -04:00
e81b62aad1
BPFunctions: Clean up math slightly
...
No need for magic 342 numbers.
2014-05-20 11:14:00 -04:00
6950f533ae
Merge pull request #355 from magumagu/gx-missing-opcode
...
Opcode decoding: handle missing opcodes 0x88 etc.
2014-05-18 11:12:06 +02:00
a4bb0dafb4
Removes ZeroFrog's "optimized" memcpy and memcmp functions.
...
These were only compiled in on Windows and x86_32.
They provided "optimized" copies and compares based on blocksizes for the AMD Athlon and Duron CPU families.
The code was taken from something that AMD provides with a as-is license.
Just get rid of this crap.
2014-05-17 18:03:31 -05:00
fc34d5a130
Merge pull request #360 from magumagu/lighting-rounding
...
Video backends: fix rounding in lighting computation.
2014-05-17 21:06:31 +02:00
9b82d72070
Video backends: warn on usage of GL_DRAW_QUADS_2.
...
It's not normally used, so if it shows up, it could indicate a CPU emulation
bug.
2014-05-17 11:55:32 -07:00
0fac17da33
Merge pull request #269 from magumagu/swbackend-xfregisters
...
SW backend: use VideoCommon XFRegisters struct.
2014-05-17 10:40:57 +02:00
6fbf6fad3b
VideoCommon: cleanup Renderer::RecordVideoMemory.
2014-05-16 18:58:08 -07:00
1357277f40
Video backends: mass-replace "xfregs" with "xfmem".
2014-05-16 18:58:07 -07:00
8f5342c442
Video backend: merge global var xfmem into xfregs.
...
There isn't really any reason to keep them separate.
2014-05-16 18:55:31 -07:00
818c89313e
Video backends: unify xfregs/xfmem structures.
...
Removes the duplicate swxfregs global variable/struct from the software
backend in favor of the ones from VideoCommon.
2014-05-16 18:55:30 -07:00
8b82cea704
Video backends: fix behavior of "konst" TEV inputs.
...
Also fixes TEVCOLORARG_HALF.
Values verified to match hardwarere.
2014-05-16 12:19:35 -07:00
fee6efc4dc
Merge pull request #375 from degasus/remove_SkipVertices
...
VertexLoaderManager: Remove unused function.
2014-05-16 16:20:29 +02:00
25e8dc0aed
Merge pull request #354 from magumagu/gx-primitive-mask
...
Opcode decoding: 0xC0 isn't a valid command.
2014-05-16 15:59:59 +02:00
7bb44199fd
remove unused and unexported function
2014-05-16 14:33:00 +02:00
9e4eeb3b9b
Video backends: fix rounding in lighting computation.
...
For whatever reason, the hardware doesn't do a full divide by 255, but
instead uses an approximation with shifting, similar to the way it is done
in TEV.
2014-05-11 12:53:02 -07:00
15ce648b0d
Merge pull request #341 from lioncash/pragma
...
Move the header guard before the includes in VideoCommon's Statistics.h
2014-05-10 23:21:53 -05:00
39d439fc48
Opcode decoding: handle missing opcodes 0x88 etc.
...
Hardware testing shows that they do the same thing as the 0x80 family of
opcodes: they draw quads.
2014-05-10 20:33:28 -07:00
369c0c4ce2
Opcode decoding: 0xC0 isn't a valid command.
...
Fix our opcode decoders to handle this appropriately.
2014-05-08 15:49:19 -07:00
b4e1ac5f08
Merge pull request #185 from degasus/shader-cleanup
...
Shader cleanup
2014-05-06 11:37:11 +02:00
716b3fefd4
VideoCommon: recreate XFB texture when the XFB size changes.
...
We need to do this to correctly deal with games which dynamically change the
XFB width and height.
2014-05-05 11:01:14 -07:00
2f92b82b29
Merge pull request #345 from Sonicadvance1/Qualcomm-textureSize
...
Enables usage of GLSL textureSize on Qualcomm v66.
2014-05-05 11:41:40 -05:00
813e30ec2a
ShaderGen: make uv readonly
2014-05-05 17:06:38 +02:00
6109958e68
ShaderGen: OGL: don't make a copy of "Normal"
2014-05-05 17:06:37 +02:00
0e0da8c8fd
ShaderGen: make clipPos readonly
2014-05-05 17:06:37 +02:00
c80717ac2e
PixelShaderGen: extract iround
2014-05-05 17:06:37 +02:00
2bc2b73e03
PixelShaderGen: drop fmod as it isn't used any more
2014-05-05 17:06:37 +02:00
c82991df5b
ShaderGen: inline constant shaders
2014-05-05 17:06:37 +02:00
b0878c54b2
ShaderGen: use cbuffers for D3D
2014-05-05 17:06:37 +02:00
fe9fcfdd07
ShaderGen: merge early-z blocks of ogl+d3d
2014-05-05 17:06:37 +02:00
3e14bf511f
ShaderGen/D3D: inline centroid
2014-05-05 17:06:36 +02:00
8bb8446e34
Enables usage of GLSL textureSize on Qualcomm v66.
...
Qualcomm's v66 drivers have a working textureSize() function now.
Enable usage of it.
2014-05-05 08:50:13 -05:00
2e464800bc
VideoCommon: correctly compute whether an indirect texture stage is active.
...
This is consistent with the implementation in PixelShaderGen.
2014-05-04 22:44:10 -07:00
52c42a2946
VideoCommon: fix indirect texture coordinate computation in TEV.
...
Even if an indirect texture stage doesn't actually sample a texture, we
still need to emit the relevant math.
Fixes Issue 7092.
2014-05-04 22:12:57 -07:00
f4d80bd8a5
Move the #pragma once header guard before the includes in Statistics.h in VideoCommon.
2014-05-04 00:24:21 -04:00
557c3db462
Merge pull request #81 from degasus/skip_framelimit_hotkey
...
Add a hotkey for disabling the framelimit.
2014-05-01 12:40:01 +02:00
2d8cfb89d7
Changes posmtx vertex attribute to integer.
...
This makes it so we don't need to do some dumb casting from float to integer in our shaders.
Only tested in OpenGL, needs to be tested in D3D.
2014-04-30 19:11:06 -05:00
30586f4d29
Add set/get functions for IsFramelimiterTempDisabled
2014-04-30 12:51:13 +02:00
8483811b39
Add a hotkey for disabling the framelimit.
...
Atm this is hardcoded to '\t'.
2014-04-30 12:50:53 +02:00
25f5598e31
Merge pull request #306 from neobrain/pixel_center_correction
...
VertexShaderGen: Correct vertex shader output to consider shifted pixel centers.
2014-04-25 09:32:36 +02:00
ca2a79d0bd
Workaround dumb custom texture loading logic so that D3D11, GL core (used on OS X) and GLES code paths have less broken custom textures.
2014-04-24 00:33:48 +02:00
c47c32d4a8
VertexShaderGen: Correct vertex shader output to consider shifted pixel centers.
...
Fixes issue 267.
2014-04-24 00:21:17 +02:00
fd37a768a6
Fix texture conversion shaders for GLSL ES.
...
Noticed this while messing with EFB to RAM.
We were having an implicit conversion from integer to float, GLSL ES doesn't allow implicit conversion.
Changes it to a explicit conversion to float.
2014-04-22 15:48:26 -05:00
762572a08c
BPMemory: Fix GenMode using an incorrect number of bits for the number of color chans.
2014-04-21 22:47:08 +02:00
16d3dbc5ea
BPMemory: Use BitField for the GenMode fields.
2014-04-21 22:34:23 +02:00
92ec49ac9f
Change to ARM's naming convention in DriverDetails.
...
This matches how ARM handles their naming in their drivers for different models.
Really it's that way because both Mali-T6xx and Mali-T7xx fall under Midgard.
While everything else (except Mali-55) fall under Utgard.
2014-04-18 21:06:32 -05:00
b3a67821e4
VideoCommon: fix rounding in TEV.
...
The hardware backends pass the TEV tests in gxtest with this change.
2014-04-15 16:47:01 -07:00
068f26d2aa
PixelShaderGen: Fix an issue with color combiner compare mode and simplify the affected line a bit.
...
This was a mistake of mine when translating floating point values to integer values.
Also, the max() part of that line was just completely redundant because the sign of an absolute value is always greater than or equal to zero.
Fixes issue 7178.
2014-04-15 23:41:07 +02:00
4f3227b4a9
Merge pull request #261 from magumagu/pixelshadergen-extra-paren
...
PixelShaderGen: delete extra parenthesis
2014-04-14 09:48:02 +02:00
a2150ef1f8
PixelShaderGen: delete extra parenthesis.
...
Without this patch, we can generate a pixel shader which doesn't compile
in some cases.
2014-04-14 00:41:27 -07:00
a9fa49f34d
Support checking for the Mali-T7xx line of GPUs.
...
They are similar enough that they will share bugs with their drivers, so make them fall under the same Mali-Txxx umbrella of bug issues.
If there is ever a need in the future for having separate bugs depending on family, we can support that then.
2014-04-11 23:46:44 -05:00
fd9c1fa746
VideoBackend: remove unused config vars.
...
No point to keeping around variables which are always "true".
2014-04-11 14:53:12 -07:00
306b9afd1e
Merge pull request #224 from magumagu/swrender-deletepixelfork
...
Software backend: Delete forked PixelEngine.
2014-04-11 20:01:23 +02:00
cdf6172348
Merge pull request #213 from Jezze/vertexloader-cleanups
...
Vertexloader cleanups
2014-04-10 08:52:36 +02:00
87d106d65c
Remove dumb CodeBlock duplication in the emitters.
...
Fixes issue 6990.
This uses a bit of templating to remove the duplicate code that is the CodeBlocks in each emitter headers.
No actual functionality change in this.
2014-04-09 13:53:43 -05:00
3251d78f89
Add initial support for GLSL ES 3.10.
...
GLSL ES 3.10 adds implicit support for the binding layout qualifier that we use.
Changes our GLSL version enums to bit values so we can check for both ES versions easily.
2014-04-03 00:46:09 -05:00
9b03178673
Merge pull request #222 from comex/more-clang-fixes
...
More clang fixes
2014-03-30 13:24:08 +02:00
b5654a2464
Don't cast -1 to enum to represent a missing value.
...
This is undefined behavior in C++, and a clang warning suggests it is
actually producing bad code as a result:
../Source/Core/VideoCommon/BPFunctions.cpp:164:45: warning: comparison of constant 4294967295 with expression of type 'PEControl::PixelFormat' is always false [-Wtautological-constant-out-of-range-compare]
if (new_format == old_format || old_format == (unsigned int)-1)
2014-03-30 01:40:06 -04:00
0661efea84
Software backend: Delete forked PixelEngine.
...
Mostly just zapping a bunch of duplicated code; the only interesting thing
going on here is the changes to the performance counter implementation.
2014-03-29 12:07:20 -07:00
664c8d30a0
Remove all trailing whitespaces from our codebase.
2014-03-29 11:05:44 +01:00
73176d0333
VideoCommon/VertexLoader: Add more use of std::min and std::max
2014-03-27 00:33:41 +01:00
478a27e052
VideoCommon/VertexLoader: Remove duplicate point min and max calculation
2014-03-27 00:24:48 +01:00
0c62ae9c1a
VideoCommon/VertexLoader: Remove NRM enum
2014-03-26 23:56:57 +01:00
ea6b37cb75
Merge pull request #193 from neobrain/tev_combiner_fixes
...
PixelShaderGen: Cleanups and fixes for tev combiners.
2014-03-26 10:05:46 +01:00
c6070b94ce
BPMemory: Fix an enum to use u32 as its underlying type.
2014-03-26 10:02:57 +01:00
eb0f547a17
PixelShaderGen: Cleanups.
2014-03-26 00:23:36 +01:00
4f82d6f7af
PixelShaderGen: Implement tev combiner lerping in a faster way which also reproduces hardware behavior perfectly.
...
The new behavior has been verified to be correct by hardware tests. This is an improvement over the old code, which was just a guess.
2014-03-26 00:17:04 +01:00
fdad95c055
PixelShaderGen: Cleanups and fixes for tev combiners.
...
Fixes issue 4674.
2014-03-26 00:17:03 +01:00
16105db709
BPMemory: Make use of BitField in a number of structures.
2014-03-25 23:57:58 +01:00
8941f19cdb
BPMemory: Expose the pixel_format and zformat fields in PE_CONTROL as enumerations.
2014-03-25 23:57:58 +01:00
77a7bab5ae
BPMemory: Use the new BitField class in two selected structures.
2014-03-25 23:57:57 +01:00
4a68550d01
Remove superfluous bit shift
2014-03-18 04:07:45 +01:00
484fb46390
Clang 3.4 exposes some warnings on Android.
...
Fixes all warnings on Android build except for what is in externals.
Removes a function from TextureDecoder_Generic since it is unused and generates a warning.
2014-03-17 18:17:12 -05:00
6445e02d53
Merge pull request #159 from Tilka/misc
...
Cleanup stuff
2014-03-17 17:43:20 +13:00
2fcaca0603
More range-based loops and overrides
2014-03-17 02:55:55 +01:00
a9adaa5601
Prevent VideoConfig inheritance.
2014-03-16 17:00:29 -04:00
82f772fdcc
Remove the extra parameters field from the PH.
...
Metroid: Other M was the only game which required this field, but the
issue in that game can be fixed properly by enabling format change
emulation. Hence, there's no point in having this around anymore.
Fixes issue 6644.
2014-03-15 15:04:39 +02:00
8d679e76d2
Merge pull request #164 from lioncash/cstr-cull
...
Kill off some usages of c_str.
2014-03-15 00:57:56 +01:00
0f81cbd6be
PixelShaderGen: More code alignment fixes.
2014-03-14 22:33:26 +01:00
c1016205d3
Pixel/LightingShaderGen: Fix code alignment issues.
...
Most of these weren't even introduced by me, but hey - I'm nice and love wasting my time :p
2014-03-14 22:33:26 +01:00
2067f88e0f
PixelShaderGen: Don't make local lookup tables "static".
2014-03-14 22:33:26 +01:00
0ce92e0162
PixelShaderGen: Remove the "i" prefix for integer variables.
...
The prefix was just required in the development stage to reduce the risk of regressions.
2014-03-14 22:33:26 +01:00
bdd629c598
PixelShaderGen: Use spaces for alignment where appropriate.
2014-03-14 22:33:26 +01:00
6e65e02c9e
ConstantManager: Do not use single-element arrays.
2014-03-14 22:33:25 +01:00
6c2971eaf6
PixelShaderGen: Write 16777215 in hex (0xFFFFFF) so that it's easier to understand.
2014-03-14 22:33:25 +01:00
8ebb65ebf2
PixelShaderGen: Prettify generated shader source.
2014-03-14 22:33:24 +01:00
e2e1c5c905
PixelShaderGen: Add a note about a random idea which should be checked with hardware tests.
2014-03-14 22:33:12 +01:00
6fcbda6752
PixelShaderGen: Cleanup and clarify bump alpha combiner inputs.
2014-03-14 22:33:12 +01:00
9a96a1d525
PixelShaderGen: Remove old, mostly useless comments.
...
A few vague lines of comments cannot replace an afternoon reading of how TEV works.
2014-03-14 22:33:12 +01:00
8b8bb04fd3
PixelShaderGen: Use bit shifts instead of multiplications as a small optimization.
2014-03-14 22:33:12 +01:00
036a8c6951
PixelShaderGen: Clean up tev compare functionality.
2014-03-14 22:33:10 +01:00
fa7173d099
PixelShaderGen: Store tex scale as an integer.
2014-03-14 22:32:30 +01:00
16109fb453
PixelShaderGen: Treat UV coordinates as actual integers.
2014-03-14 22:32:29 +01:00
50526ae50a
PixelShaderGen: Remove some dead code.
2014-03-14 22:32:18 +01:00
065919f599
PixelShaderGen: Perform some of the fog calculations with integers.
2014-03-14 22:31:32 +01:00
605b687af8
PixelShaderGen: Use integer math for z textures.
2014-03-14 22:31:31 +01:00
3a6389992e
PixelShaderGen: Treat UV coordinates like S17.7 integers (they're still stored as float, though).
2014-03-14 22:31:30 +01:00
3e6efdb53e
LightingShaderGen: Perform more lighting calculations with integers.
2014-03-14 22:31:19 +01:00
387b9bf3c2
LightingShaderGen: Perform some lighting calculations with integers.
2014-03-14 22:31:19 +01:00
78623871f9
ShaderGen: Store material uniforms as integers.
2014-03-14 22:31:19 +01:00
4bf57565e8
ShaderGen: Store light color uniforms as integers.
2014-03-14 22:31:18 +01:00
68e91f0d55
PixelShader: Store fog color as an integer.
2014-03-14 22:31:18 +01:00
0238a56816
PixelShaderGen: Change indirect texture matrix uniforms to use integers.
2014-03-14 22:31:18 +01:00
c13a5c38e9
PixelShaderGen: Change the "alpha" uniform to use integers.
2014-03-14 22:31:18 +01:00
ec60acac3a
PixelShaderGen: Change the "colors" and "kcolors" uniforms to be integers.
2014-03-14 22:31:18 +01:00
df94e62350
PixelShaderGen: Process fog calculations with integer math.
2014-03-14 22:31:18 +01:00
21eb482a6e
PixelShaderGen: Write constants in decimal instead of hexadecimal where appropriate.
2014-03-14 22:31:11 +01:00
cff952c397
PixelShaderGen: Use integer math for indirect tev stage texcoord calculation.
2014-03-14 22:30:27 +01:00
e7a42d884f
PixelShaderGen: prev should be initialized to the proper value; tev output needs to be clamped between -1024 and 1023.
2014-03-14 22:30:16 +01:00
cb1514e082
PixelShaderGen: Remove superfluous registerstate stuff. Also, made alphabump an integer.
2014-03-14 22:30:15 +01:00
a11ae69cb0
PixelShaderGen: Use integer math for TEV combiners.
2014-03-14 22:28:32 +01:00
aaa8e74a68
PixelShaderGen: Use integer math for tev outputs.
2014-03-14 22:28:14 +01:00
654442feb7
PixelShaderGen: Use integer math for tev konst value.
2014-03-14 22:28:12 +01:00
e69ee6ae0a
PixelShaderGen: Remove remaining floating point bits for texture color.
2014-03-14 22:26:50 +01:00
ac1c77c392
PixelShaderGen: Use integer math for rasterizer color.
2014-03-14 22:26:50 +01:00
3ea97f7730
PixelShaderGen: Use integer math for indirect texture coords.
2014-03-14 22:26:50 +01:00
degasus