piwalker/dolphin
1
0
Fork 0
mirror of https://github.com/dolphin-emu/dolphin.git synced 2024-11-15 13:57:57 -07:00
Code Issues Actions Packages Projects Releases Wiki Activity
6,902 Commits 5 Branches 11 Tags 1.3 GiB
864d24a86e
Commit Graph

22 Commits

Author SHA1 Message Date
xsacha
f9e4e73e42 Use SSSE3 shuffle for DataReader's DataReadU32xN in VideoCommon. The function is used for reading up to 16 u32's at a time (512-bits) and then converting endianness. 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@6802 8ced0084-cf51-0410-be5f-012b33b47a6e
 2011-01-10 13:14:56 +00:00
Rodolfo Osvaldo Bogado
c21f804c48 more work on dlist caching now should be a little faster 
a little bugfix in vertex loading and some fixes.
not much time to work these days but  at least i can spend a little time fixing thing.
please test for regressions.

git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@6409 8ced0084-cf51-0410-be5f-012b33b47a6e
 2010-11-14 14:42:11 +00:00
Jordan Woyak
841583b8f3 Merged identical VertexManager code from DX9/DX11/OGL plugins into VideoCommon. Still need to merge VertexManager::Flush (will be easier after TextureCache is merged). Purposely using a class/virtualfuncs rather than a namespace so multiple VertexManager can be in one plugin (VideoMergeNew? :p). 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@6249 8ced0084-cf51-0410-be5f-012b33b47a6e
 2010-10-03 00:41:06 +00:00
Soren Jorvang
453f7c67cd Newer versions of GCC's <tmmintrin.h> check for __SSSE3__ (-mssse3). 
No matter. We don't actually need it for our purposes.


git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@6016 8ced0084-cf51-0410-be5f-012b33b47a6e
 2010-07-31 15:26:46 +00:00
Soren Jorvang
b0041f00a3 I missed some cases in r5901: 
Because we only ever call Pos_ReadDirect (and through that, DataRead<T>)
from JIT generated code, the compiler may not get the heads-up to properly
prepare for run-time instantiation of those template functions.

Explicitly instantiating Pos_ReadDirect gets around that issue.

Also force DataRead* inline as gcc didn't always do that itself when the
DataRead functions in turn were called from (other) template functions.


git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@5902 8ced0084-cf51-0410-be5f-012b33b47a6e
 2010-07-18 08:30:40 +00:00
Soren Jorvang
faf586e8f1 Because we only ever call Pos_ReadDirect (and through that, DataRead<T>) 
from JIT generated code, the compiler may not get the heads-up to properly
prepare for run-time instantiation of those template functions.

Explicitly instantiating Pos_ReadDirect gets around that issue.

Also force DataRead* inline as gcc didn't always do that itself when the
DataRead functions in turn were called from (other) template functions.

I am far from a C++ language lawyer, so I cannot speak learnedly about the
correctness of this solution, but it works.

Dolphin.app built on OS X 10.6 now actually works on 10.5 as well.


git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@5901 8ced0084-cf51-0410-be5f-012b33b47a6e
 2010-07-18 05:31:51 +00:00
nodchip
956b8eb54d VideoCommon: Added automatic selection routines for SSSE3/SSE4.1 codes. It selects SSSE3/SSE4.1 codes only if a proper preprocessor definition is defined and the target cpu supports SSSE3/SSE4.1. The selection routines in VertexLoader_* use function pointers. TextureDecoder uses a combination of "#if" and "if" statements. 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@5302 8ced0084-cf51-0410-be5f-012b33b47a6e
 2010-04-09 15:13:42 +00:00
nodchip
6136c94de5 VideoCommon: merged SSSE3/SSE4.1 codes. Added some additional SSSE3/SSE4.1 codes which will be used in "The Legend of Zelda: Twilight Princess". 
These codes don't work unless "_M_SSE=0x301", for SSSE3, or "_M_SSE=0x401", for SSE4.1, is defined as a preprocessor definition.


git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@5300 8ced0084-cf51-0410-be5f-012b33b47a6e
 2010-04-09 03:02:12 +00:00
nodchip
1c34052266 Refactored VertexLoader::CompileVertexTranslator(). Now texture coordinates loaders are also selected from a function table. I will add a hack to increase the speed in the next commit. 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@5140 8ced0084-cf51-0410-be5f-012b33b47a6e
 2010-02-28 11:36:00 +00:00
nodchip
d348c33bf2 Refactored VertexLoader::CompileVertexTranslator(). Now the vertex position loader is selected from a function table. I will apply the same kind of refactoring to texture coordinates loader. This is a pre-preparation to optimize texture coordinates loaders. 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@5139 8ced0084-cf51-0410-be5f-012b33b47a6e
 2010-02-28 08:41:02 +00:00
nodchip
609151c6e8 Reverted because of some processor and performance issue. I will develop in a branch about SSSE3/SSE4.1. 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@5123 8ced0084-cf51-0410-be5f-012b33b47a6e
 2010-02-24 23:58:48 +00:00
nodchip
47fb73b71a Added SSSE3/SSE4.1 code for speed up. The code does not work in this revision because cpu_info is not initialized properly. I will fix the issue in another commit. 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@5119 8ced0084-cf51-0410-be5f-012b33b47a6e
 2010-02-24 09:56:41 +00:00
XTra.KrazzY
c86d2e5129 Now ALL copyright dates are fixed to the legally correct year: The year in which the project began. Now we don't have to change it every year :) 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@3898 8ced0084-cf51-0410-be5f-012b33b47a6e
 2009-07-28 21:32:10 +00:00
death2droid
73a7686044 Fix up all copyright dates. 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@3886 8ced0084-cf51-0410-be5f-012b33b47a6e
 2009-07-26 11:51:40 +00:00
donkopunchstania
fcd7fa997b Possible fix for projection hacks 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@3427 8ced0084-cf51-0410-be5f-012b33b47a6e
 2009-06-13 17:44:14 +00:00
hrydgard
c49f969563 vertexloader_pos cleanup - remove the branch per vertex. not much of a speedup though. 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@2256 8ced0084-cf51-0410-be5f-012b33b47a6e
 2009-02-15 14:46:17 +00:00
hrydgard
6deb87c176 VertexLoader: Change some pointer arithmetic to array syntax. should have no effect on performance. 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@2255 8ced0084-cf51-0410-be5f-012b33b47a6e
 2009-02-15 13:45:03 +00:00
hrydgard
e979db862c Convert some ugly macros to templates. 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@2189 8ced0084-cf51-0410-be5f-012b33b47a6e
 2009-02-09 21:24:32 +00:00
hrydgard
d209d4a577 Cache vertex array pointers, a la magumagu's idea. Misc other stuff. Should give very minor speed boost. 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@2185 8ced0084-cf51-0410-be5f-012b33b47a6e
 2009-02-09 20:35:30 +00:00
hyperiris
c450e44b08 optimize memory access. reduce Memory_Read_U##_type call(it uses Memory::GetPointer, which is complex), and gain speed, may be 1 or more fps faster, I've tested on zelda tww,tp, starfox assault only. 
Please review and may be the code need clean.

git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@2144 8ced0084-cf51-0410-be5f-012b33b47a6e
 2009-02-08 13:08:58 +00:00
nakeee
ad88b27c76 attempt to get rid of another __LOG crash 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@2110 8ced0084-cf51-0410-be5f-012b33b47a6e
 2009-02-04 17:29:12 +00:00
hrydgard
95a341a4e9 GL Vertex loader moved to VideoCommon. 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@1692 8ced0084-cf51-0410-be5f-012b33b47a6e
 2008-12-26 13:09:16 +00:00