Commit Graph

1203 Commits

Author SHA1 Message Date
degasus eb310cbd1d VideoCommon: disable efb access + perf querys on cph thread
The usual way to handle this kind of request is to rise a flag which the gpu thread polls.
The gpu thread itself either generates the result or just write zeros if disabled.
After this, it rise another flag which says that this work is done.

So if disabled, we still have the cpu-gpu round trip time. This commit just returns 0 on the cpu thread
instead of playing ping pong...
2014-01-09 18:37:59 +01:00
Ryan Houdek cdf69adcb0 Fix for previous commit. I was supposed to use GL_VERSION, not GL_RENDERER. 2014-01-07 19:35:06 -06:00
Ryan Houdek d2e62796fc Probe the GL_RENDERER string on Nvidia to grab the driver version. 2014-01-07 18:29:04 -06:00
degasus c42f274e22 OpenGL: use shader 420pack if available to staticly bind ubo location
Bindung locations after compiling a shader stalls the driver. So if we manage not to bind anything after compilation, the lag would be reduced much.
2014-01-05 10:38:45 +01:00
degasus 4fff5ac90d OpenGL: drop UBO-workaround usage for efb2ram shaders
It's just brainfuck to use this workaroung there. Just fetch the uniform location like all other util shaders.
2014-01-05 09:52:26 +01:00
degasus 0f0a3cc509 ogl: clamp to edge for out of bound efb access
fixes issue 6898

OpenGL defaults are GL_REPEAT, which is even more unlikely than GL_CLAMP_TO_EDGE.
As I can't test the behavoir of the real hardware, I changed it to how it works before,
but I guess just clip the texture makes more sense.
2014-01-03 08:15:19 +01:00
Ryan Houdek 0faf696be7 Merge Fail. We don't disable BaseVertex on broken pinned memory. 2014-01-01 17:20:09 -06:00
Ryan Houdek 1118226f27 Merge branch 'master' into buffer_storage 2013-12-31 19:18:30 -06:00
Ryan Houdek 8d8b0fc884 Merge branch 'master' into buffer_storage
Conflicts:
	Source/Core/VideoBackends/OGL/Src/Render.cpp
	Source/Core/VideoCommon/Src/DriverDetails.cpp
	Source/Core/VideoCommon/Src/DriverDetails.h
2013-12-31 15:41:50 -06:00
Jasper St. Pierre 34692ab826 Remove unnecessary Src/ folders 2013-12-31 14:03:19 -05:00
Jasper St. Pierre 43e618682e Convert all vcxproj files to UNIX line endings 2013-12-31 14:03:18 -05:00
Ryan Houdek 6d63db96e9 Disable primitive restart on buggy OS X Intel HD 3000 drivers. 2013-12-30 18:26:55 -06:00
Tony Wasserka de16b7207c D3D/OGL: Add a TODO noting that we don't support GX_CULL_ALL, most notably required for accurate zfreeze emulation. 2013-12-30 20:37:59 +01:00
Ryan Houdek a5bfdbf6cd Free performance for AMD+Linux. Instead of disabling BaseVertex due to pinned_memory. Just disable pinned_memory for the ELEMENT_ARRAY_BUFFER type which has issues with baseVertex(ELEMENT_ARRAY_BUFFER)+pinned_memory. In my two tests with this, I went from 5FPS to 11FPS in one test, and 2FPS to 6FPS in another. 2013-12-28 08:00:44 -06:00
Ryan Houdek 0b31fc8f52 [buffer_storage] Temporary fix for Android/GLES3 just like how everything else is. Cleanup for this will be in a new branch. 2013-12-28 05:16:39 -06:00
Ryan Houdek 8fdcba0c06 [buffer_storage] Code formatting changes suggested by Bh44L 2013-12-28 05:15:43 -06:00
Ryan Houdek 935e1fdf99 [Android] Enable shader cache. 2013-12-28 04:55:39 -06:00
Ryan Houdek d8b7f4d73f [Android] Support grabbing OpenGL extensions and a function for checking for support. 2013-12-28 01:08:52 -06:00
Ryan Houdek 77ba051361 [buffer_storage] Add the CLIENT_STORAGE_BIT since we access the buffer more frequently on the client side than the server side. That is exactly what the hint is for. 2013-12-27 22:58:36 -06:00
Ryan Houdek d8ceb97a60 [buffer_storage] Require GLEW 1.10 which has the new OpenGL 4.4 methods. Fixes linux build. 2013-12-27 15:37:31 -06:00
Ryan Houdek ce99921c20 [buffer_storage] Implement ARB_buffer_storage. Disable it for GL_ARRAY_BUFFER due to a bug in Nvidia's drivers that causes black screen with it. 2013-12-27 10:56:03 -06:00
Ryan Houdek a87e0e7489 Remove our GLU dependency. Fix static zlib 2013-12-23 16:01:34 -07:00
Ryan Houdek e697d7a2dd [Android] Work around Qualcomm's broken garbage in their v53 drivers. This doesn't fix the issue, just a work around. This is the stupidest issue coming from Qualcomm. Now Dolphin Mobile won't crash immediately, but there are new SPS issues. 2013-12-19 17:30:39 -06:00
Ryan Houdek 945b903499 Work around AMD's broken Linux drivers when it comes to pinned memory and base_vertex usage. It seems that using pinned memory with base_vertex disabled is quicker than the other way around. 2013-12-19 09:40:13 -06:00
Ryan Houdek 8c264a3964 In the shader log message, say if the error came from a PS or VS. 2013-12-18 22:21:14 -06:00
Ryan Houdek f9ff0bc55d Have our OpenGL/ES context creation be less stupid. 2013-12-12 12:43:49 -06:00
Ryan Houdek eb3b933dd0 Remove all instances of OpenCL in the Dolphin Project. A brief history of OpenCL in Dolphin. OpenCL was originally added to the Dolphin codebase 1 month after it was released with OS X Snow Leopard in 2009. OpenCL was one of the largest group projects that Dolphin ever has had. The OpenCL texture decoder was originally aded with version 1.0 of the OpenCL spec; This version didn't have the capability of a OpenCL-OpenGL interop which would allow for uploading textures once and have it decoded directly to a OpenGL texure. This was to be worked out when the OpenCL 1.1 spec was released and allowed the interop. This work has never been done, and no one in the team is willing to work on it for various reasons. OpenCL has had the unreasonable expectation that it increases the performance of video games that require a large amount of EFB copies like NSMBW. In reality, enabling OpenCL just put the graphics card in a higher power mode which increased the game speed. This is due to the unfortunate effect of Dolphin tending to not push GPUs out of their lower frequency power savings modes. Thanks to everyone that had contributed to the OpenCL texture decoder. 2013-12-11 15:15:55 -06:00
degasus 134c89ef98 OpenGL: disable shader cache when ShaderDebugging is enabled
fixes 6859
2013-12-09 16:45:20 +01:00
degasus 42619c1d2d Merge branch 'ogl-tex2d'
Conflicts:
	Source/Core/VideoBackends/OGL/Src/TextureConverter.cpp
2013-12-09 13:04:14 +01:00
degasus 687097d4bc OGL: use integer uniforms for efb2ram texture converter 2013-12-09 12:33:50 +01:00
degasus edc879ce48 OpenGL: fix crash in revision 95aeedec19
Real xfb didn't provide any read_stride, so there is a division by zero.

This commit calculates the correct read_stride for real_xfb, so there is also no hack for texture vs xfb needed.
2013-12-03 00:09:37 +01:00
degasus 95aeedec19 OpenGL: readback efb2ram with different strides at once
This is done with a pixel buffer object. We still have to stall the GPU, but
we only do it once per efb2ram call.
As the cpu can't access the vram, it has to queue a memcpy for the gpu and
wait for the gpu to finish this copy. We did this for every cache line which
is just stupid. Now we copy the complete texture into a pbo and readback this
at once. So we don't have to wait for lots of round-trip-times.
2013-11-26 20:05:49 +01:00
degasus 1138c2e155 OpenGL: reset EFB after efb2ram FB initialization 2013-11-26 04:07:59 +01:00
degasus 0b4cb2e15f OpenGL: split real xfb + efb2ram framebuffers 2013-11-25 22:27:11 +01:00
Ryan Houdek 9dbb262aab Fix for OpenGL ES 3. 2013-11-25 15:11:06 +00:00
degasus 6750a81972 TextureConverter: Use integer math for swizzling
also move int(efb_coord) -> float(ogl_fb_coord) into WriteSampleColor
2013-11-25 15:49:13 +01:00
degasus bcb31b09d3 TextureConverter: Use gl_FragCoord instead of uv0 2013-11-25 15:01:18 +01:00
degasus 454e1dd9a2 OpenGL: attributeless rendering for efb2ram
This wasn't as easy as we now have to cache also the uniform locations.
2013-11-25 13:03:35 +01:00
degasus 6ed3f82aff OpenGL: attributeless rendering for postprocessing 2013-11-25 12:36:17 +01:00
degasus 38fe05b1df OpenGL: attributeless rendering in emulate format changes
only cleanup
2013-11-25 12:27:54 +01:00
degasus dd0b74ac15 OpenGL: cleanup efb2tex
Also use attributeless rendering. But we need the src rect, so set it by uniform.
If there is a slowdown here (I doubt as the driver likely has a fast path to update uniforms)
then we should check if this rect changes and only then update the uniform.
2013-11-25 12:19:34 +01:00
degasus b93756df87 OpenGL: drop texture_rect hack
Everything is moved to texture2d (but often in a hacky way), so we don't need this global hack any more.
2013-11-25 09:05:50 +01:00
degasus afcf0e65d1 OpenGL: fix emulate format changes for texture2d 2013-11-25 08:59:04 +01:00
degasus b904d56036 OpenGL: fix efb2tex for texture2d 2013-11-25 08:43:55 +01:00
degasus 1a3e790d9e OpenGL: fix xfb for texture2d 2013-11-25 08:38:00 +01:00
degasus e8f23af10b OpenGL: always use texture2d as efb 2013-11-25 08:32:41 +01:00
degasus 230e12ae8c OpenGL: also remove VAO from xfb convertion
We use attributeless rendering, so officially we have to bind _any_ VAO.
As the state of this VAO doesn't matter, we don't have to switch it.

Also fix an AMD issue as they don't like to render from an empty VAO.
2013-11-25 07:38:20 +01:00
Ryan Houdek da3eef1019 Fix the issue with COLOROUT not being defined anymore. Fix a issue where Mali shader compiler is idiotic in finding an overload for the mix function. 2013-11-25 00:06:29 +00:00
Ryan Houdek f292819ff5 [Android] Due to recent changes in code breaking Tegra 4 support, and also the upcoming code which will be breaking GLES2 support entirely. Taking the initiative to drop the remaining support code from the codebase in preparation for the upcoming changes. For a look at how Dolphin on Tegra 4 looked like prior and would not have been able to be fixed at all due to Tegra 4 not supporting the precision we need in our shaders; Look at this Youtube video http://youtu.be/Ga7Jc_Ote7U 2013-11-24 15:49:23 -06:00
degasus d410fe7c96 OpenGL: cleanup yuv2rgb (real xfb) workflow
We neither scale nor render from subimages, so we by using gl_Position, we don't have to generate _any_ vertices for this converting.

Also remove the glTexSubImage optimization as every driver does it when needed. But there are some workflows (eg on APU) where it's better to realloc this texture instead of a second memcpy or stall.
2013-11-24 15:56:50 +01:00
Scott Mansell 12741f6406 Add comments for anybody attempting accuracy improvements in the future. 2013-11-25 01:30:53 +13:00
Scott Mansell b5bd2ba847 OpenGL: Enable filtering for EFB to Real XFB copies.
This fixes Real XFB Jaggies in OpenGL on games which use yscaling, such
as most PAL games.

This fixes the last of the "Real XFB Macroblocking" issues for opengl,
see issue #6503
2013-11-25 01:11:42 +13:00
Scott Mansell e6b35642df Fix Desktop GLSL versions in the recent changes.
Seems OpenGL ES 3 Requires you must have an lod argument, while Desktop
versions require you must not have a lod argument if you are using a
Sampler2DRect (which doesn't do Mipmapping).
2013-11-24 22:48:10 +13:00
Ryan Houdek 531f840720 Fix OpenGL ES 3 in the recent changes. texelFetch doesn't require the lod argument in desktop GLSL versions, but in GLSL ES 3 it is a required argument. 2013-11-24 07:45:42 +00:00
degasus 994426b3dc Opengl: fix real XFB sample positions
(0,0) and (1,0) aren't accurate xfb sample positions. This fixes the image shift to the left and some blocking on higher IR.
2013-11-24 05:39:40 +01:00
Scott Mansell eef2cddfd7 Opengl: Fix opengl realxfb "macroblocking"/bluring issue.
YUYV textures should NEVER be interpolated/filtered in RGB colour space.
Use TexelFetch to always fetch an actual texture sample.

issue 6503
2013-11-24 17:15:11 +13:00
Matthew Parlane 0a31255943 Unused arguments removed from XFBSource::Draw
Thanks neo.
2013-11-24 11:33:43 +13:00
Tony Wasserka 672fa65ee7 OpenGL: Enable pinned memory even for index buffers (works for me). Big-ish speedup on AMD GPUs for streaming intensive games. 2013-11-22 18:09:52 +01:00
Matthew Parlane 33d8166620 Use IOFile for TextureToPng to support non-ascii
Changed save texture/screenshot uses to std::string
Removed unneeded new/delete calls when dealing with temp data.
2013-11-17 11:14:38 +13:00
Lioncash cce869ae01 [Core] Fix a mismatched dealloc in Render.cpp in the OGL backend. Should be delete[] 2013-11-16 17:05:51 -05:00
Matthew Parlane 71d70d896f Api was too confusing for people. 2013-11-17 10:34:34 +13:00
Ryan Houdek e8a4cc0f71 Screenshot capability of Software rasterizer for feature completness. 2013-11-15 22:08:08 -06:00
Matthew Parlane 3a13dfdd9b [OGL] Textures now save to PNG not TGA 2013-11-16 15:59:59 +13:00
Matthew Parlane 033ed9477e Reworked Screenshot saving.
Now OGL doesn't rely on WX for PNG saving.
FlipImageData supports (pixel data len > 3) now.
TextureToPng is now in ImageWrite.cpp/h
Video Common depends on zlib and png.
D3D no longer depends on zlib and png.
2013-11-15 13:00:38 +13:00
Jordan Cristiano 0fced651a5 More warning fixes, OSX build fix. 2013-11-14 03:11:40 -05:00
Jordan Cristiano 39a4d4329d more warning fixes 2013-11-14 02:19:15 -05:00
Tillmann Karras 268bdf19ce Fix format string warnings 2013-11-13 04:01:16 +01:00
Tony Wasserka 0b02880b76 Revert "Be less dumb."
Revert "Actually, filename really does need to be a parameter because of some random debug thing."
Revert "fix non-HAVE_WX case"
Revert "Handle screenshot saving in RenderBase.  Removes dependency on D3DX11 for screenshots (texture dumping is still broken)."

This reverts commits 00fe5057f1, 74b5fb3ab4, cd46138d29 and 5f72542e06 because taking screenshots in D3D still crashed for me so there was no point in the code changes (which I found ugly anyway).
2013-11-10 00:22:33 +01:00
degasus 274f6dd7ab ogl: check default framebuffer for msaa
default framebuffer must not be sampled, so throw a panicalert if it is
2013-11-05 17:38:48 +01:00
Matthew Parlane e15f628935 Fix {Read,Write}FileToString.
We should be using binary always.
2013-11-05 00:33:41 +13:00
comex 21610c2e54 Run code through clang-modernize -add-override to add 'override' decls. 2013-11-03 20:54:05 -05:00
comex c579637eaf Run code through the advanced tool 'sed' to remove trailing whitespace. 2013-11-03 20:54:05 -05:00
comex 965b32be9c Run code through clang-modernize -loop-convert to create range-based for loops, and manually fix some stuff up. 2013-11-03 20:54:01 -05:00
comex 00fe5057f1 Be less dumb. 2013-11-03 14:37:56 -05:00
comex 74b5fb3ab4 Actually, filename really does need to be a parameter because of some random debug thing. 2013-11-02 22:44:27 -04:00
comex cd46138d29 fix non-HAVE_WX case 2013-11-02 22:23:29 -04:00
comex 5f72542e06 Handle screenshot saving in RenderBase. Removes dependency on D3DX11 for screenshots (texture dumping is still broken). 2013-11-02 22:19:21 -04:00
comex 82729fcc8f Merge remote-tracking branch 'shuffle2/vc12'
Conflicts:
	Source/Core/Common/Common.vcxproj
	Source/Core/Common/Common.vcxproj.filters
2013-10-31 16:51:56 -04:00
degasus 498d491942 Merge branch 'viewportCorrection' 2013-10-29 22:33:51 +01:00
degasus 35824aa4d5 ogl: ceil viewport, rounding isn't supported on vs and this is more like the old behavior 2013-10-29 21:42:40 +01:00
degasus f4a9deefa8 ogl: remove some unneeded includes 2013-10-29 18:55:50 +01:00
degasus acc0b3b21c ogl: don't call SetViewportChanged from backend as it's not needed any more 2013-10-29 18:55:50 +01:00
degasus 3151d8709c VideoCommon: move ViewportCorrection into VideoCommon
D3D doesn't allow bigger viewports than rendertargets. But flipper does, so the viewport will be clipped and the transformation matrix will be changed.
This was done in the D3D backend itself. This is now moved into VideoCommon. This don't reduce code, but in this way, VideoCommon doesn't depend on the backends.
2013-10-29 18:55:50 +01:00
Ryan Houdek f18c461fd6 [viewport-float] Fix OpenGL ES 3 build. 2013-10-29 17:10:02 +00:00
degasus edf0511d4e OGL: Use floating points for viewport parameters.
As told by neobain, ARB_viewport_array introduced float viewports.
They must be supported on OGL4.1+, but it's usually also implemented on 3.3 GPUs.
2013-10-29 15:29:06 +01:00
Shawn Hoffman ccd30024b3 Update to VS2013 and a slew of build-related updates. Notes:
* Currently there is no DEBUGFAST configuration. Defining DEBUGFAST as a preprocessor definition in Base.props (or a global header) enables it for now, pending a better method. This was done to make managing the build harder to screw up. However it may not even be an issue anymore with the new .props usage.
* D3DX11SaveTextureToFile usage is dropped and not replaced.
* If you have $(DXSDK_DIR) in your global property sheets (Microsoft.Cpp.$(PlatformName).user), you need to remove it. The build will error out with a message if it's configured incorrectly.
* If you are on Windows 8 or above, you no longer need the June 2010 DirectX SDK installed to build dolphin. If you are in this situation, it is still required if you want your built binaries to be able to use XAudio2 and XInput on previous Windows versions.
* GLew updated to 1.10.0
* compiler switches added: /volatile:iso, /d2Zi+
* LTCG available via msbuild property: DolphinRelease
* SDL updated to 2.0.0
* All Externals (excl. OpenAL and SDL) are built from source.
* Now uses STL version of std::{mutex,condition_variable,thread}
* Now uses Build as root directory for *all* intermediate files
* Binary directory is populated as post-build msbuild action
* .gitignore is simplified
* UnitTests project is no longer compiled
2013-10-26 17:55:38 -07:00
Lioncash 97cd42f5c3 Remove some unnecessary includes as well as simplifying exisiting ones if possible. 2013-10-19 19:58:56 -04:00
comex de1773affb Basic precompiled header support for Linux/OS X. Shaves 20-30% off full rebuild time on my system. 2013-10-17 00:07:35 -04:00
degasus f850d283d1 ogl: only use bufferdata if buffersubdata is broken
Else nvidia fallback would be slower than needed
2013-10-16 15:46:58 +02:00
Ryan Houdek f0bf5aa113 Fix rendering on Adreno and Mali. This adds a new DriverDetails bug to know if the bufferstreaming is broken. Thanks degasus for fixing this. 2013-10-16 11:27:58 +00:00
degasus 5de7ac4791 OGL: update ubo workaround 2013-10-12 20:31:59 +02:00
degasus 6e2fe72b8f D3D: also uses VideoCommon constant buffer handling
As now both backends uses the VideoCommon one, the old setting API was removed.
2013-10-12 20:29:56 +02:00
degasus 7c14463d11 ogl: implement useful constant buffer upload
this will remove the additional memcpy introduced in my last commit
2013-10-12 20:29:56 +02:00
degasus 4377618438 VideoCommon: keep a copy of the const buffer in VideoCommon
The upload in the backend isn't done, it's just pushed by the mostly removed SetMulti*SConstant4fv.
Also no optimizations was done on VideoCommon side, but I can start now :-)

Sorry for the hacky way, but I think this is a nice (working) snapshot for a much bigger change.
2013-10-12 20:29:56 +02:00
degasus 0753ce5bda ogl: use GL_RGBA instead of GL_RGBA8
8 bit per pixel is _everywhere_ the default, so we don't need to define it explicitly.
Also we don't rely on exact 8 bit resolution.
2013-10-12 16:34:06 +02:00
Ryan Houdek 0556193442 Build fix for previous commit. 2013-10-11 21:45:44 -05:00
Ryan Houdek cf7521dc36 [Android] Make Tegra 4 output graphics that aren't visually appealing. 2013-10-11 21:42:07 -05:00
Matthew Parlane 414ed6ef63 C.K. should fix his website before throwing stones at obsolete xml. 2013-10-09 23:56:16 +13:00
Matthew Parlane fbcc41c18e Fix references for OGL, SW and D3D. 2013-10-09 23:49:00 +13:00
Matthew Parlane 4a0745799f Missed OGL includes apparently. 2013-10-08 18:03:33 +13:00
Matthew Parlane eb480a406c Fix Windows vcxproj files. 2013-10-08 17:52:18 +13:00
Jasper St. Pierre 9920362581 Fix MSVC project files
yay relative paths
2013-10-07 18:50:21 -04:00
Jasper St. Pierre a7c7208103 Put Plugins/ in Core/, rename to VideoBackends 2013-10-07 10:37:01 -04:00