Commit Graph
104 Commits
Author SHA1 Message Date
Henrik Rydgård 506a86300d More explicit invalidation of any cached state in Thin3D. Fixes #13307 2020-08-22 00:30:29 +02:00
Henrik Rydgård b43698a13d Remove most instances of base/logging.h from Common, Core, GPU, more 2020-08-15 19:08:44 +02:00
Unknown W. Brackets b8342fb8ec SaveState: Rename ChunkFile files to Serialize.
Makes more sense and less weird than ChunkFileDoMap, etc.
2020-08-10 08:04:05 +00:00
Henrik Rydgård eadd3c5785 Remove the LARGE_VIEWPORTS separate path. The fewer paths the better, all need to work anyway.
Required fixing a bug in ShaderUniforms.cpp (used by D3D11 and Vulkan,
the two backends that previously exposed LARGE_VIEWPORTS).
2020-08-05 19:03:08 +02:00
Henrik Rydgård d57edfbdac Rename FramebufferCommon.cpp/h to FramebufferManagerCommon.cpp/h for consistency 2020-08-03 23:17:22 +02:00
Unknown W. Brackets 9ef521d945 GPU: Avoid immediate postshader FBO recreate.
Before, we created things at the wrong size, then recreated at the right.
Now we just start at the correct size once.
2020-06-12 12:35:35 -07:00
Unknown W. Brackets 22e46b51c2 GPU: Centralize DestroyAllFBOs().
This cleans up the postshader update code.
2020-05-13 18:15:04 -07:00
Unknown W. Brackets 4ea5d165f5 D3D11: Always use accurate depth.
The Direct3D APIs use a "min" and "max" value, and it's required that min
must be less than or equal to max.  If this requirement is not met, the
viewport parameters are not updated.

We forced accurate depth on for NVIDIA and AMD, but this just meant Intel
was still broken and less people were reporting bugs about it.

Although the PSP GE has depth viewport transform, this is not used for
clipping and is not actually the depth range.  The minz and maxz GE
registers are the depth range, and minz must be less than or equal to maxz
for drawing to occur.
2020-04-18 22:34:33 -07:00
Unknown W. Brackets ad98609819 GPU: Use consistent buffered rendering state.
The setting can be changed in Qt and Windows between frames, so let's
track the current setting in most places.  This is everywhere fbman is
easily accessible.
2020-04-04 10:51:47 -07:00
Unknown W. Brackets cebcfb1bbd GPU: Use old frame when presenting a skip.
If we flip using a skipped frame, we may show an even older frame causing
weird flickering.
2020-03-01 13:55:28 -08:00
Unknown W. Brackets 3c1e8abcfe GPU: Centralize vsync interval logic.
We already do unthrottle checking here.
2020-02-29 22:27:00 -08:00
Henrik Rydgård 8e7da3fef4 Disable vertex range culling when the depthrange hack is enabled. Temporary workaround for #11576. 2019-02-26 16:07:11 +01:00
Unknown W. Brackets 1f594f3fb5 GPU: Track draw in shader manager.
With device restore etc. everywhere.
2018-12-23 12:55:37 -08:00
Henrik Rydgård fa40bcff8d Disable triangle range culling on D3D9 on Intel. Should fix #11477 2018-10-21 09:36:15 +02:00
Unknown W. Brackets cd0662c065 GPU: Cleanup outdated flushbefore code.
Backends didn't match.
2018-09-20 20:36:41 -07:00
Unknown W. Brackets b7d84926fa D3D11: Enable depth clamping.
Uses the same logic as Vulkan, improving #9545 on most D3D11 devices.
2018-09-18 21:49:17 -07:00
Unknown W. Brackets a4c0640f01 GE Debugger: Factor out host calls some.
Moving more of this to cross platform for the web debugger.
2018-09-01 08:32:03 -07:00
Henrik Rydgård 36fd2711d6 Revert "Remove further remains of hardware skinning."
This reverts commit 40db61a680.
2018-04-10 12:22:41 +02:00
Henrik Rydgård 40db61a680 Remove further remains of hardware skinning. Fixes #10661 2018-03-05 00:03:47 +01:00
Henrik Rydgård c7dcb7c365 D3D9/D3D11: Fix GPU feature detection. Enforce accurate depth on nVidia as well, they no longer seem to support reverse depth so it's needed as a workaround. Fixes #10655 2018-03-01 09:38:06 +01:00
Henrik Rydgård 64ec46e705 Unify FramebufferDirty()/FramebufferReallyDirty() for the hw backends 2018-02-26 11:58:17 +01:00
Henrik Rydgård 78467d6092 Unify FastRunLoop for the hardware backends. 2018-02-26 11:52:16 +01:00
Henrik Rydgård 6a2f45c2e1 Unify UpdateCmdInfo 2018-02-26 11:48:56 +01:00
Henrik Rydgård 8cef1f0f8d GPU: Unify command table (cmdInfo_) 2018-02-26 11:44:02 +01:00
Henrik Rydgård 967018b7f8 Unify Execute_LoadClut 2018-02-26 11:32:02 +01:00
Henrik Rydgård c7f8f4c5ca Unify Execute_Prim 2018-02-26 11:22:27 +01:00
Unknown W. Brackets 0151c5e649 GPU: Track stats on uploads.
A lot can lead to slowdown, so it's useful when logging frame drops.
2017-12-28 19:27:45 -08:00
Unknown W. Brackets f9750dd137 Display: Avoid limiting FPS without a clear.
A bit of a dirty heuristic to avoid the slowdown in #8538.
2017-12-26 19:37:40 -08:00
Henrik Rydgård fb74c9dfeb Clean up a small inconsistency 2017-12-07 21:09:09 +01:00
Henrik Rydgård e6be86ce74 ARM buildfix 2017-11-22 11:01:42 +01:00
Henrik Rydgård 2c86217552 Cleanup GPU vendor detection a bit, use it to avoid the issue mentioned in #9545 2017-11-21 16:19:28 +01:00
Henrik Rydgård 6a3d41892a Workarounds for issue with "accurate depth". See #9545, #10087 2017-11-21 12:11:05 +01:00
Henrik Rydgård 313223364c Buildfix, update comments slightly 2017-11-15 14:21:17 +01:00
Henrik Rydgård d4df0181db More duplicate code removal 2017-11-14 09:20:36 +01:00
Henrik Rydgård 04e313ecba Unify Execute_Bezier and Execute_Spline 2017-11-13 10:45:27 +01:00
Unknown W. Brackets 9cf17d9288 D3D11: Clear input layouts on shader clear.
When we do clear shaders, make sure to also clear input layouts so we
don't have corrupt keys.
2017-11-11 08:11:10 -08:00
Unknown W. Brackets f93ef92960 D3D11: Avoid clearing shaders on load state.
They don't need to be cleared (unlike FBOs), and clearing them requires
clearing even more things.  Fixes #9637.
2017-11-11 08:09:56 -08:00
Henrik Rydgård d618b3673b Count readbacks in a frame and show in GPU debug stats. 2017-11-08 11:57:53 +01:00
Henrik Rydgård 8c69be9bfa Get rid of GPUCommon's "Internal" functions. 2017-11-05 23:21:52 +01:00
Henrik Rydgård 26e52ca5e8 Get rid of a bunch of sync stuff that's no longer needed. 2017-11-05 23:21:49 +01:00
Henrik Rydgård 7312576239 More buildfixes, warning fix, memory leak fix 2017-10-26 11:43:22 +02:00
Henrik Rydgård 4a30aedc53 Unify and move around code to cleanup some debug accessors 2017-10-18 13:10:05 +02:00
Henrik Rydgård d53c88456a Unify GetFramebufferList() (trivial) 2017-10-18 12:49:15 +02:00
Henrik Rydgård 2f85e6516e Minor optimizations (use the new hashmap in a few more places) 2017-08-20 19:18:46 +02:00
Henrik Rydgård 5680332343 Minor cleanups 2017-08-17 15:20:21 +02:00
Henrik Rydgård 2c4e5e2303 Fastpath in fastrunloop when diff=0. Remove need for Execute for UV scale/offset. 2017-08-17 15:20:16 +02:00
Henrik Rydgård 71baecabd6 Make FLAG_FLUSHBEFORE a no-op, move flushes into the execution functions.
It would only be necessary if we needed the previous value, which we
don't do in any of these.
2017-08-17 15:20:10 +02:00
Henrik Rydgård 983bb3bd0f Can't dirty VERTEXSHADER_STATE before the flush in Execute_VertexTypeSkinning, causes obscure bugs when software skinning is enabled 2017-08-15 14:27:54 +02:00
Henrik Rydgard 078e1151f9 Dirty-flag the fragment shader ID. 2017-08-14 11:36:07 +02:00
Unknown W. BracketsandHenrik Rydgård c84efcfc97 Dirty vertexshader state on hw bezier/spline. 2017-08-14 11:36:06 +02:00