0d92c8fb89
Jit64: Optimize dcbx
2015-08-24 18:33:23 +02:00
ac84d6d0fa
Jit64: some cache flush changes
...
- dynamically allocate third scratch register instead of forcing ECX
- use LEA as 3 operand add if possible
- use BT,JC instead of SHR,TEST,JNZ
- merge MOV,TEST
- use appropriate ABI function (no asm change)
2015-08-24 18:33:23 +02:00
6f34b27323
Jit64: implement dcbf + dcbi
2015-08-24 18:33:19 +02:00
0ad6fa8f62
Merge pull request #2903 from lioncash/cast
...
Memmap: Remove pointer casts
2015-08-24 15:42:56 +02:00
abd3b124be
Memmap: Remove pointer casts
2015-08-24 09:07:09 -04:00
4baf3e10c6
Merge pull request #2902 from Tilka/fpscr
...
Jit64: quickfix for mtfsfx
2015-08-24 13:19:26 +02:00
33eefc2d86
Jit64: quickfix for mtfsfx
2015-08-24 12:12:31 +02:00
d3176fe22a
[AArch64] Implement stfiwx
...
Improves povray performance by ~4%
2015-08-24 01:10:55 -05:00
80fa9af9b1
Merge pull request #2898 from degasus/linking
...
JitArm64: Faster linking of continuous blocks
2015-08-23 18:09:02 -05:00
8bc311ab3c
Merge pull request #2897 from degasus/arm
...
JitArm64: Implement subfex, subfcx, addex, subfic, divwux, srwx
2015-08-23 23:52:35 +02:00
7320d519b4
JitArm64: Implement srwx
2015-08-23 23:29:48 +02:00
4722a69fd0
JitArm64: Implement divwux
2015-08-23 23:29:18 +02:00
9e4366963c
JitArm64: Implement subfic
2015-08-23 23:29:07 +02:00
95be17772f
JitArm64: Implement addex
2015-08-23 23:29:02 +02:00
025e7c835a
JitArm64: Implement subfcx
2015-08-23 23:28:28 +02:00
550a90e691
JitArm64: Implement subfex
2015-08-23 23:28:24 +02:00
561744819e
[AArch64] Implement fctiwzx
...
Improves the povray benchmark time by 5.6%
2015-08-23 15:35:18 -05:00
4fa23abbe1
[AArch64] Implement MOVI and ORR(imm) in the NEON emitter.
2015-08-23 15:34:53 -05:00
0a0e012fab
x64Emitter: add RCPPS and RCPSS SSE instructions
2015-08-23 16:59:27 +02:00
77a6798094
JitArm64: Faster linking of continuous blocks
2015-08-23 14:44:23 +02:00
73067b1ef1
Merge pull request #2888 from degasus/jit64
...
Jit64: Faster linking of continuous blocks
2015-08-23 13:24:15 +02:00
2a1abf8dd6
Merge pull request #2896 from lioncash/using
...
Core: Minor CPU core typedef cleanup
2015-08-22 19:00:23 -04:00
cc3fb7e7b4
Merge pull request #2883 from degasus/master
...
Profiler: Sort output by total time
2015-08-22 17:52:54 -05:00
8b881a6c34
Merge pull request #2891 from Sonicadvance1/aarch64_implement_crxxx
...
[AArch64] Implement the cr instructions
2015-08-23 00:44:47 +02:00
fdafa5d063
Core: Move includes out of instruction table headers
...
These aren't necessary (and cause unnecessary indirect inclusions).
2015-08-22 14:15:02 -04:00
a248a4d2ce
Jit64/JitIL: Relocate instruction typedefs
2015-08-22 14:15:00 -04:00
c56717e058
Core: Shorten the _interpreterInstruction typedef
...
The class itself already acts as a namespace trailer, so '_interpreter'
isn't necessary. This also gets rid of a duplicate typedef in the
Interpreter_Tables.
2015-08-22 14:14:49 -04:00
b4e4a4cef4
Disable OpenGL ES 3.1 on all Qualcomm Adreno devices.
...
Their new driver that supports GLES3.1 + AEP has issues with it.
At the very least they don't implement all of the geometry shader features fully which causes shader linker issues when we attempt to use them.
I don't have a device so I can't fully test, so until I do I'm going to blanket disable the whole thing.
2015-08-22 09:12:19 -05:00
a39c0910c4
Merge pull request #2893 from Sonicadvance1/aarch64_memory_base_register
...
[AArch64] Use a register as a constant for the memory base.
2015-08-22 15:41:57 +02:00
dba579c52f
[AArch64] Use a register as a constant for the memory base.
...
Removes a /lot/ of redundant movk operations in fastmem loadstores.
Improves performance of the povray bench by ~5%
2015-08-22 08:36:34 -05:00
3f5ff98c1b
Merge pull request #2890 from lioncash/ptr
...
x64Emitter: Remove pointer casts from Write{8,16,32,64} functions
2015-08-22 10:09:28 +02:00
2d505bc2a6
Merge pull request #2894 from Sonicadvance1/no_more_eaten_canary
...
Fix the shader overrunning our max shader size.
2015-08-22 10:08:14 +02:00
c2f38f1d16
Merge pull request #2892 from Sonicadvance1/aarch64_frsp
...
[AArch64] Implement frspx
2015-08-22 09:44:14 +02:00
3242e1a617
Fix the shader overrunning our max shader size.
...
The Star Wars games really push the hardware to its limits, which can cause the shaders that are produced to be 18kb or more.
Double our maximum shader size to compensate.
Fixes issue #8860
2015-08-22 01:01:03 -05:00
ce32b76be3
[AArch64] Implement frspx
...
Improves performance in povray bench by 2%
2015-08-22 00:35:30 -05:00
d74eb0ea58
[AArch64] Fix the bugs in the cr instructions
...
Makes it a bit more efficient in the process.
2015-08-21 23:24:29 -05:00
e9ade0abe1
JitArm64: implement crXXX
2015-08-21 20:49:08 -05:00
a69755d9ee
x64Emitter: Remove pointer casts from Write{8,16,32,64} functions
...
This also silences quite a few ubsan asserts from firing when the emitter is being used.
2015-08-21 18:09:48 -04:00
95d958c03d
Merge pull request #2889 from lioncash/interp
...
Interpreter: Use std::isnan instead of IsNAN
2015-08-21 21:43:08 +02:00
caec42135d
MathUtil: Remove IsNAN and IsINF
...
These aren't necessary, since the stdlib provides equivalents.
2015-08-21 15:05:43 -04:00
bb7f3d1822
Merge pull request #2867 from Tilka/mtspr_hid0
...
Jit64: implement HID0 case of mtspr
2015-08-21 21:04:35 +02:00
01aea965ba
Merge pull request #2864 from Tilka/fpscr
...
Jit64: implement FPSCR related instructions
2015-08-21 21:04:20 +02:00
18d658df1f
Interpreter_FloatingPoint: Use std::isnan instead of IsNAN
...
Same thing, except one is part of the stdlib.
2015-08-21 15:04:03 -04:00
78aa01e06e
Jit64: Faster linking of continuous blocks
...
We compile the blocks as they are executed, so it's common
to link them continuously. We end with calling JMP after every
block, but often just with a distance of 0.
So just emitting NOPs instead also "calls" the next block, but
easier for the CPU.
2015-08-21 17:41:53 +02:00
c325c310d6
Merge pull request #2884 from lioncash/emitter
...
x64Emitter: Minor cleanup
2015-08-21 13:03:51 +02:00
5f628749ff
Merge pull request #2886 from Sonicadvance1/aarch64_faster_lfd
...
[AArch64] Optimize lfd instructions if possible.
2015-08-21 05:38:53 -05:00
df53b37253
[AArch64] Optimize lfd instructions if possible.
...
If we are going to be using lfd, then chances are it is going to be used in double heavy areas of code.
If we only need to load the lower register, then we should also not worry about having to insert in to the low 64bits of the guest register.
So add a new flag to the backpatching to handle lfd to directly to the destination register.
This gives ~3% performance improvement to Povray.
2015-08-21 04:31:54 -05:00
4f45d71840
Merge pull request #2760 from Sonicadvance1/aarch64_fcmp
...
[AArch64] Implement fcmp{u,o}
2015-08-21 11:03:20 +02:00
39ced2a2d7
AVIDump: fix -Wsign-compare warning
...
Cast the other side of the comparison to avoid a warning with newer
ffmpeg/libav versions (cb3591e69738c808d26ba15eb02414fedfcd91cc).
2015-08-21 10:26:35 +02:00
6cb87a9227
Merge pull request #2837 from Sonicadvance1/aarch64_faster_nonpaired
...
[AArch64] Optimize cases when an FPR is only used for non-paired ops.
2015-08-21 09:51:45 +02:00
degasus