Commit Graph

2145 Commits

Author SHA1 Message Date
ca3b3adb50 Merge pull request #1849 from Sonicadvance1/AArch64_Emitter_Improvements
[AArch64] Emitter Improvements and additions.
2015-01-11 07:08:02 -06:00
5b825e9091 StringUtils: use C locale instead of .1252
Fixes regression on Linux.
2015-01-11 00:15:57 -08:00
394257f33b Merge pull request #1798 from FioraAeterna/overclock
Support overclocking and underclocking the Gamecube CPU
2015-01-11 06:12:27 +01:00
8237004448 JIT: optimize for the common case of unquantized psq_l/st
Optimistically assume used GQRs are 0 in blocks that only use one GQR, and
bail at the start of the block and recompile if that assumption fails.

Many games use almost entirely unquantized stores (e.g. Rebel Strike, Sonic
Colors), so this will likely be a big performance improvement across the board
for games with heavy use of paired singles.
2015-01-10 14:14:43 -08:00
c2c950108d Support overclocking and underclocking the Gamecube CPU
Won't work with all games, but provides a nice way to spend extra CPU to make
a variable framerate game faster (e.g. Spyro or The Last Story), or to make
a game use less CPU at the cost of a lower framerate (e.g. Rogue Leader).
2015-01-10 10:46:38 -08:00
c511ee763a [AArch64] Add the float emitter. 2015-01-08 19:55:31 -06:00
6bc274a968 Merge pull request #1844 from Tilka/sse_mov
x64Emitter: fix MOVLPD/MOVHPD
2015-01-08 19:04:14 -06:00
f0f8384639 Merge pull request #1785 from degasus/custom_texture
VideoCommon: Custom texture handling
2015-01-08 13:29:45 +01:00
5a0133c478 [AArch64] Add a few more VFP register helpers.
Renames Is128Bit to IsQuad to line up more with the other helpers.
2015-01-07 13:05:55 -06:00
2b4f1aed40 [AArch64] Minor shifted register adjustment.
If we have a shift amount that is the full length of the source register then we have an invalid instruction.
This can happen when dealing with a couple of PowerPC instructions.
This same adjustment is already in the ARMv7 emitter.
2015-01-07 13:03:51 -06:00
0a23ca9461 [AArch64] Add MUL/MNEG instruction aliases to the emitter. 2015-01-07 13:03:01 -06:00
d2eaba5cb7 [AArch64] Minor MOVI2R improvement.
Use the ZR for both input arguments in the case the immediate is the maximum immediate value.
This allows it to be aliased to MVN when disassembling.
2015-01-07 13:00:39 -06:00
8dba84dd7c [AArch64] Fix 8 & 16 bit loadstore indexes.
I wasn't bit shifting correctly for 8 and 16bit loadstores.
2015-01-07 12:58:37 -06:00
f4f59ea71e [AArch64] Fix ADDS/SUBS emitter functions.
These weren't emitting the flag bit. So they were regular ADD and SUB emitters.
2015-01-07 12:56:45 -06:00
1efa9b8b72 x64Emitter: add MOVLPS/MOVHPS 2015-01-07 11:23:52 +01:00
3daa6ab259 x64Emitter: fix MOVLPD/MOVHPD
These instructions were using the wrong prefix which turned
MOVLPD(reg, mem) into MOVDDUP(reg, mem) and made the rest of them
invalid.
2015-01-07 11:23:52 +01:00
89b7f1057f Merge pull request #1804 from FioraAeterna/fastermmu2_master
MMU: various improvements, bugfixes, optimizations
2015-01-07 00:49:58 +01:00
d8d4c94e07 Merge pull request #1819 from Tilka/cmake
CMake: simplify some expressions
2015-01-06 09:27:59 +01:00
53b44ccb3a x64ABI: enhance MOVTwo to take an offset
This lets us merge displacements into MOVTwo in trampolines.
2015-01-05 10:45:58 -08:00
8a6ea918b6 GekkoDisassembler: fix display of CRs in branch operations 2015-01-03 19:53:55 -08:00
6bcdb10eee CMake: simplify some expressions 2015-01-03 13:17:57 +01:00
479d1e56c3 Merge pull request #1783 from degasus/disablelogs
OGL: disable driver warnings fetch
2015-01-01 14:12:02 -06:00
c6dd5044d6 VideoCommon: make hash independet from hires textures 2014-12-29 10:24:27 +01:00
c5a0b6bf50 Merge pull request #1432 from randomstuff/linux-perf
Add Linux perf JIT support (/tmp/perf-$pid.map)
2014-12-28 23:12:52 +01:00
1ed41672f5 OGL: disable driver warnings fetch
This did give a decent slowdown on some drivers.
2014-12-28 22:31:24 +01:00
47b67fe527 Merge pull request #1744 from Buddybenj/google-code
Update Outdated Google Code References
2014-12-21 20:55:56 -06:00
fba3c48ec4 Update Outdated Google Code References 2014-12-20 21:17:51 -06:00
d3c2e8fb0a [AArch64] Improvements to the AArch64 emitter.
Fixes issues with negative offsets in loadstore instructions.
Adds ADRP/ADR instructions.
Optimizes MOVI2R function to take advantage of ADRP on pointers, can change a 3 instruction operation down to one.
Adds GPR push/pop operations for ABI related things.
2014-12-20 19:35:52 -06:00
dad7911214 Merge pull request #1486 from rohit-n/goto
Remove some gotos.
2014-12-20 16:47:55 +01:00
6a2973ac1d Fix hires texture path. 2014-12-11 19:24:16 -05:00
6e928d08a0 Jit64: clean up casts in memory operands 2014-12-07 20:23:27 +01:00
4e98078daf Merge pull request #1658 from lioncash/locale
Common: Remove locale based functions from CommonFuncs.
2014-12-07 01:12:00 -06:00
986d19b9d5 x64Emitter: fix bitwise AVX opcodes
The reason this didn't break is that bitwise instructions like VPAND,
VANDPS, and VANDPD do the exact same thing. The only difference is the
data type they are intended for.
2014-12-06 22:28:03 +01:00
9bcadc8029 Common: Remove locale based functions from CommonFuncs.
Since %f isn't used anymore in the shader generators, these can go.
2014-12-05 20:55:29 -05:00
e7ddaf86f5 Fix building with PCH disabled. 2014-12-03 22:01:12 -06:00
7bc78827ed Merge pull request #1574 from degasus/profiler
Common: Add a built-in profiler
2014-12-04 13:22:31 +11:00
2c39d4044d [AArch64] Add loadstore paired emitter instructions. 2014-12-02 21:08:34 -06:00
94d9d138d9 Common: Add a built-in profiler 2014-12-03 00:50:41 +01:00
a53c5217bb Common: Remove goto in MemArena. 2014-12-01 11:47:37 -06:00
e3abad234c Common: Remove gotos in FileUtil, and replace FILE*s with IOFile/ifstreams. 2014-12-01 11:47:36 -06:00
414e36d8c9 Merge pull request #1258 from FioraAeterna/avoidfmulround
JIT: optimize single-precision ops based on knowledge of their inputs
2014-11-30 15:47:34 -06:00
c27ee21dd9 [AArch64] Workaround builtin byteswap bug.
The builtin byteswap routines cause critical failure on AArch64 when built with the Android toolchain.
I didn't experience this issue when building for Linux using a local qemu chroot.
Seems to be only an issue with the Android toolchain when building AArch64.
Use our generic version instead.
2014-11-30 01:23:03 -06:00
72c96c20d3 JIT: more optimizing of float ops based on known input characteristics
If the inputs are both float singles, and the top half is known to be identical
to the bottom half, we can use packed arithmetic instead of scalar to skip
the movddup.

This is slower on a few rather old CPUs, plus the Atom+Silvermont, so detect
Atom and disable it in that case.

Also avoid PPC_FP on stores if we know that the output came from a float op.
2014-11-29 11:33:11 -08:00
ce059769f6 Merge pull request #1439 from Armada651/ogl-stereo-3d
OGL: Stereoscopic 3D Support
2014-11-28 11:45:38 -06:00
e9b09a4c9f Arm(64)Emitter: Make some variables static 2014-11-25 23:27:48 -05:00
245ff601b7 Merge pull request #1552 from waddlesplash/dolphin-qt
DolphinQt: GameList!
2014-11-25 00:24:32 -05:00
5b9aeaa686 Use CLI argument for Linux perf JIT support 2014-11-24 23:30:29 +01:00
9722ae2a5d Move the JIT registration logic in its own file
Move the JITed function/basic-block registration logic out of the CPU
subsystem in order to add JIT registration to JITed DSP and
Video/VertexLoader code.

This necessary in order to add /tmp/perf-$pid.map support to other
JITed code as they need to write to the same file.
2014-11-24 23:18:18 +01:00
4d931c16ad Silence a few "zero as null pointer" warnings. 2014-11-24 12:36:46 -06:00
f6ea293027 VertexShaderManager: Compute stereoscopy projection matrices. 2014-11-23 14:23:41 +01:00