Commit Graph

8419 Commits

Author SHA1 Message Date
Yuri Kunde Schlesner
8369ee5803 Rasterizer: Pre-divide vertex attributes by W
Execute the division-by-W for perspective-correct interpolation of
values in the clipper, moving them out of the rasterization inner loop.
2014-12-29 02:08:11 -02:00
Yuri Kunde Schlesner
fe186d3a59 GPU: Bitwise texture swizzling
Replace the loop-based texture address swizzling code by a bit-twiddling
implementation, providing a very small speed up. Also simplify
addressing code.
2014-12-29 02:08:11 -02:00
Yuri Kunde Schlesner
2012e1420f Rasterizer: Common sub-expression elimination
Move the computation of some values out of loops so that they're not
constantly recalculated even when they don't change.
2014-12-29 02:08:10 -02:00
Yuri Kunde Schlesner
7e9bc85cc8 Clipper: Compact buffers on each clipping pass
Use a new buffer management scheme in the clipper that allows using a
bounded minimal amount of buffer space. Even though it copies more data
it is still slightly faster likely due to using less cache.
2014-12-29 02:08:10 -02:00
Yuri Kunde Schlesner
da04976437 CMake: Require Boost 1.57.0 (fixes Travis OS X) 2014-12-29 02:08:10 -02:00
Yuri Kunde Schlesner
a320d1a5b4 Clipper: Avoid dynamic allocations
The triangle clipper was allocating its temporary input, output and work
buffers using a std::vector. Since this is a hot path, it's desirable to
use stack allocation instead.
2014-12-29 02:08:09 -02:00
Yuri Kunde Schlesner
d151d797b1 Vertex Shader: Zero OutputVertex to avoid denormals
Unused OutputVertex attributes were being left un-initialized. The
leftover garbage sometimes decoded as floating-point denormalized
values, causing fallbacks to microcode and massive slowdowns in the rest
of the rasterization pipeline even though the results were unused. By
zeroing the structure we ensure these attributes only contain harmless
zeros.
2014-12-29 02:08:09 -02:00
Lioncash
5c198686ce dyncom: Implement UXTB16/UXTAB16 2014-12-28 22:45:04 -05:00
bunnei
487a80f9f7 Merge pull request #347 from bunnei/frameskip
Frameskip
2014-12-28 22:39:17 -05:00
bunnei
df728cb4c2 Merge pull request #355 from lioncash/simp
armemu: Simplify some instructions.
2014-12-28 22:20:49 -05:00
bunnei
3b9d181b8e GPU: Implement frameskip and remove forced framebuffer swap hack. 2014-12-28 22:14:05 -05:00
bunnei
5763d40319 Merge pull request #352 from xdec/PR_XDEC_172
Qt: Citra crashes if disassembler pause button is pressed while no game is running. #172
2014-12-28 22:00:47 -05:00
bunnei
9a403d7dfb Merge pull request #359 from lioncash/vfp
vfp: Actually make the code somewhat readable
2014-12-28 21:54:53 -05:00
bunnei
77363d9590 Merge pull request #331 from yuriks/handle-reform
New Handle manager
2014-12-28 21:11:55 -05:00
Lioncash
9c7f2570f7 vfp: Actually make the code somewhat readable 2014-12-28 18:55:01 -05:00
bunnei
4bf803579f Merge pull request #357 from bunnei/dyncom-pkhbt-pkhtb
Implement PKHBT and PKHTB on dyncom, fix on armemu
2014-12-28 16:57:16 -05:00
bunnei
bf9b33aa9f dyncom: Implement PKHBT and PKHTB. 2014-12-28 16:50:08 -05:00
bunnei
58cb62fe7b armemu: Fix PKHTB to do an arithmetic shift and correctly decode immediate field. 2014-12-28 16:18:52 -05:00
bunnei
082f5b2311 Merge pull request #356 from lioncash/dynusad
dyncom: Implement USAD8/USADA8
2014-12-28 13:40:22 -05:00
xdec
e6162ed91e Qt: we don't need to check if model is valid. 2014-12-28 10:11:51 -08:00
Lioncash
7d322b5c6f dyncom: Implement USAD8/USADA8 2014-12-28 12:40:51 -05:00
Lioncash
6ce2a38ec4 armemu: Simplify SSAT/SSAT16/SXTB/SXTAB 2014-12-28 12:19:31 -05:00
Lioncash
9f5b53f9ff armemu: Simplify REV/REV16/SXTH/SXTAH 2014-12-28 12:13:13 -05:00
bunnei
762f16c4ad Merge pull request #354 from lioncash/usaduflow
armemu: Fix underflows in USAD8/USADA8
2014-12-28 11:57:25 -05:00
Lioncash
914ecfe04f armemu: Simplify USAT16/UXTB/UXTAB 2014-12-28 11:57:14 -05:00
Lioncash
5e16216afb armemu: Simplify REVSH/UXTH/UXTAH 2014-12-28 11:57:09 -05:00
Yuri Kunde Schlesner
7e2903cb74 Kernel: New handle manager
This handle manager more closely mirrors the behaviour of the CTR-OS
one. In addition object ref-counts and support for DuplicateHandle have
been added.

Note that support for DuplicateHandle is still experimental, since parts
of the kernel still use Handles internally, which will likely cause
troubles if two different handles to the same object are used to e.g.
wait on a synchronization primitive.
2014-12-28 11:52:55 -02:00
Yuri Kunde Schlesner
23f2142009 Kernel: Replace GetStaticHandleType by HANDLE_TYPE constants 2014-12-28 11:52:53 -02:00
Yuri Kunde Schlesner
73fba22c01 Rename ObjectPool to HandleTable 2014-12-28 11:52:52 -02:00
Lioncash
059c65a27a armemu: Fix underflows in USAD8/USADA8
Initially reported by xdec.
2014-12-28 06:09:43 -05:00
xdec
8de09d87ab Fix crash when the disassembler pause button is pressed while no game is running. 2014-12-28 01:56:07 -08:00
bunnei
3422d81f05 Merge pull request #350 from lioncash/qops
Implement the rest of the UQ* ops.
2014-12-27 21:15:13 -05:00
Lioncash
af69b0840b dyncom: Implement UQADD8, UQADD16, UQSUB8, UQSUB16, UQASX, and UQSAX. 2014-12-27 17:24:34 -05:00
Lioncash
60523113a9 armemu: Implement UQADD8, UQADD16, UQSUB16, UQASX, and UQSAX 2014-12-27 17:06:19 -05:00
bunnei
28e64806cd Merge pull request #349 from lioncash/uhdync
dyncom: Implement UHADD8, UHADD16, UHSUB8, UHSUB16, UHASX, and UHSAX
2014-12-27 01:04:18 -05:00
Lioncash
52d889d85d dyncom: Implement UHADD8, UHADD16, UHSUB8, UHSUB16, UHASX, and UHSAX 2014-12-27 00:57:32 -05:00
bunnei
b48ca7878b Merge pull request #348 from lioncash/uhops
armemu: Implement UHADD8, UHADD16, UHSUB8, UHSUB16, UHASX, and UHSAX
2014-12-27 00:34:39 -05:00
Lioncash
84a0438cf5 armemu: Implement UHADD8, UHADD16, UHSUB8, UHSUB16, UHASX, and UHSAX 2014-12-26 23:55:39 -05:00
bunnei
a2005d0657 GPU: Change internal framerate to 30fps. 2014-12-26 21:48:32 -05:00
bunnei
01c675685e Merge pull request #327 from Apology11/master
Fix visual studio ambiguous symbol error
2014-12-26 21:43:59 -05:00
Daniel Lundqvist
7e3f62a367 Remove duplicate work 2014-12-26 20:12:11 +01:00
Daniel Lundqvist
ba4ca041f4 Allow focus only when in popout mode
Only allow manually setting focus to the rendering widget when in Single Window mode. Apply this behavior to when changing the mode while an app is running.
2014-12-26 19:42:27 +01:00
bunnei
e5ddbfee02 Merge pull request #339 from bunnei/fixup-gsp-synch
Fixup gsp synch
2014-12-25 22:52:40 -05:00
bunnei
5d2ee78d8f Merge pull request #330 from purpasmart96/new_srv
More services & small clean ups
2014-12-25 22:51:38 -05:00
bunnei
5799025ac4 GPU: Further improve synchronization. 2014-12-25 22:46:55 -05:00
bunnei
4783133bbd ARM: Add a mechanism for faking CPU time elapsed during HLE.
- Also a few cleanups.
2014-12-25 22:46:44 -05:00
purpasmart96
9796bc1fa2 More services & small clean ups 2014-12-25 19:08:33 -08:00
bunnei
9c8ec675d9 Merge pull request #343 from lioncash/smmla
armemu: Implement SMMUL, SMMLA, and SMMLS.
2014-12-25 21:57:31 -05:00
bunnei
254567056f Merge pull request #341 from lioncash/moresmops
armemu: Implement SMLALD/SMLSLD
2014-12-25 21:46:54 -05:00
bunnei
529640cfca Merge pull request #340 from lioncash/whoops
armemu: Fix GE/Q flag setting semantics
2014-12-25 21:46:00 -05:00