Commit Graph

2069 Commits

Author SHA1 Message Date
aroulin
0a0e012fab x64Emitter: add RCPPS and RCPSS SSE instructions 2015-08-23 16:59:27 +02:00
Ryan Houdek
cc3fb7e7b4 Merge pull request #2883 from degasus/master
Profiler: Sort output by total time
2015-08-22 17:52:54 -05:00
Lioncash
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
Lioncash
caec42135d MathUtil: Remove IsNAN and IsINF
These aren't necessary, since the stdlib provides equivalents.
2015-08-21 15:05:43 -04:00
Lioncash
a59f00a5e4 x64Emitter: Remove unused code 2015-08-20 23:05:20 -04:00
Lioncash
b903921b14 x64Emitter: Make WriteModRM and WriteSIB private
These shouldn't be public.
2015-08-20 19:29:40 -04:00
degasus
17932935d9 Profiler: Sort output by total time 2015-08-20 11:50:43 +02:00
Lioncash
822cf2bcbf IniFile: Mark getter functions as const 2015-08-19 22:27:18 -04:00
flacs
b0c9d73465 Merge pull request #2842 from Tilka/bmi2_flags
x64Emitter: don't check flags for most BMI2 ops
2015-08-19 21:38:09 +02:00
degasus
9bfff0d461 JitArm64: Fix jit clearing
We have to reset m_lastCacheFlushEnd on clearing.
2015-08-15 11:41:01 +02:00
Tillmann Karras
022286fb90 x64Emitter: don't check flags for most BMI2 ops
With the exception of BZHI, BMI2 instructions don't affect flags, so
don't check if they're locked.
2015-08-14 23:39:17 +02:00
Tillmann Karras
5e9fe4cd13 x64Emitter: check for immediates in BMI ops 2015-08-14 21:25:41 +02:00
degasus
878f919f63 JitArm64: Fastmem: fixup map & lookup 2015-08-14 11:24:03 -05:00
Ryan Houdek
d5c99a5b48 Add support for a CodeBlock holding a child.
This is required to make sure two code spaces are relatively close to one another.
In this case I need the AArch64 JIT codespace and its farcode space to be within 128MB of one another for branches.
2015-08-12 12:57:06 -05:00
Lioncash
144ea9f4aa Arm64Emitter: Fix encoding of '2-reg misc' variant of FCMEQ 2015-08-10 19:48:36 -04:00
Ryan Houdek
922d476dab [AArch64] Fix FCMGE instruction encoding.
Fixes a crash when ps_sel is used (PSO 1&2 intro movies).
2015-08-09 14:54:55 -05:00
Tillmann Karras
439fb26b9b x64Emitter: add MOVSLDUP/MOVSHDUP 2015-08-06 10:39:43 +02:00
flacs
0aad44c980 Merge pull request #2753 from Tilka/fix_warnings
Fix some warnings
2015-08-04 18:54:31 +02:00
Tillmann Karras
5a05187b3b Fix some warnings
reorder, sign-compare, pessimizing-move
2015-08-04 10:58:24 +02:00
Lioncash
3a450f72f8 CPUDetect: Remove a memset call on the this pointer 2015-07-28 23:44:18 -04:00
skidau
ad68de59bc Merge pull request #2665 from AdmiralCurtiss/relative-memory-card-paths
GameCube Config: Store paths relatively when selected file is within Dolphin's directory. (Windows)
2015-07-20 14:09:36 +10:00
degasus
6f38d1baa1 CodeBlock: Add a shared IsAlmostFull function
This function shall keep care about the low watermark of code space.
If we ran out of space, the JITs shall clear their block cache.
2015-07-15 08:53:05 +02:00
degasus
b8dd68beef JitArm64: Far Code Cache 2015-07-12 09:41:32 +02:00
Rohit Nirmal
c38ae5236e Fix building with PCH disabled. 2015-07-08 13:52:43 -05:00
Jeffrey Pfau
c9a25f9484 Common: CallLambdaTrampoline can return a value
As it is currently written, CallLambdaTrampoline does not return a
value. However, some of the functions that are being wrapped may
return a value that the JIT is expected to understand. A compiler
*cough cough clang* may opt to alter %rax after the wrapped lambda
returns, e.g. popping a previous value, which can clobber the
return value. If we actually have a return value, then the compiler
must not clobber it.
2015-07-04 00:57:41 -07:00
Lioncash
d09d59007a Arm64Emitter: Add a missing const specifier for an array table 2015-07-02 11:09:44 -04:00
Ryan Houdek
b174f99b17 Merge pull request #2662 from Tilka/interpreter
Interpreter: turn SNaNs into QNaNs
2015-07-01 01:09:43 -05:00
Ryan Houdek
afc3d30f5c [AArch64] Implement BFI & UBFIZ in the emitter.
Also fixes a bug in the UBFX instruction emitter. Naughty Naughty PPSSPP, not testing emitter functions you add.
2015-06-29 19:00:22 -05:00
Lioncash
daa205990f Use emplace() instead of insert() where applicable for maps. 2015-06-28 19:52:40 -04:00
Admiral H. Curtiss
df70f50fdf GetExeDirectory() shouldn't return paths with /../ in the middle. 2015-06-26 23:51:40 +02:00
Tillmann Karras
b78310bbe5 Interpreter: simplify fres 2015-06-26 09:29:18 +02:00
Jules Blok
da7ec75350 Merge pull request #2643 from comex/fix-do-file-search
Fix DoFileSearch returning the passed-in directories themselves.
2015-06-25 15:20:16 +02:00
Tillmann Karras
a725c06425 Logging: trigger _assert_() in release builds 2015-06-24 15:16:55 +02:00
Tillmann Karras
5b23182793 Common: drop GEKKO define
Afaict, it was used to build parts of Dolphin with DSPSpy, but that's
been broken forever, so lets clean up a little bit.
2015-06-24 12:13:41 +02:00
Tillmann Karras
de3c01dc2a x64FPURoundMode: fix comments 2015-06-23 02:17:36 +02:00
comex
acd6cb79a4 Fix DoFileSearch returning the passed-in directories themselves.
Fixes https://code.google.com/p/dolphin-emu/issues/detail?id=8697&can=3
2015-06-21 16:07:56 -04:00
Lioncash
fa7d93690e Hash: Remove the HashFNV prototype
It has no function definition.
2015-06-20 21:23:25 -04:00
Ryan Houdek
87b45de634 Merge pull request #2606 from Sonicadvance1/aarch64_optimize_psq_st
[AArch64] Optimize paired quantized stores.
2015-06-14 19:03:50 -05:00
Matthew Parlane
6d9711c02c Merge pull request #2599 from Sonicadvance1/aarch64_clean_cpudetect
[AArch64] Clean up our CPUDetect.
2015-06-14 15:19:03 +10:00
Jeffrey Pfau
b6d1512587 Common: Use more portable invocation of shm_open 2015-06-13 21:52:47 -07:00
Jeffrey Pfau
7085fcc8d6 Fix FreeBSD build 2015-06-13 21:52:47 -07:00
Ryan Houdek
5dc148159f [AArch64] Implement {U, S}QXTN{,2}
Also split out XTN to XTN and XTN2.
2015-06-13 23:16:17 -05:00
Matthew Parlane
f32cede086 Revert "x64: build a Position-Independent Executable (PIE)" 2015-06-14 16:06:26 +12:00
Lioncash
74b359e390 Arm64Emitter: Remove unused variable from EncodeLoadStoreRegisterOffset 2015-06-13 14:27:15 -04:00
Ryan Houdek
0af5bdcf1d [AArch64] Clean up our CPUDetect.
Drops ARMv7 feature detection.
2015-06-13 08:04:23 -05:00
Ryan Houdek
3d2b116323 [AArch64] Implement a couple instructions in the emitter.
Implements LD2R.
Implements LD1R/LD2R with post-indexing support.
Implements vector min/max instructions.
2015-06-09 18:10:56 -05:00
Tony Wasserka
6833c203a6 Make make_scmrev.h.js recognize "stable" as a stable branch. 2015-06-09 20:03:25 +02:00
Jules Blok
d5788f75a3 Merge pull request #2533 from degasus/syncgpu
Fifo: Rewrite SyncGPU
2015-06-09 09:43:26 +02:00
Fog
519d2549dd Fix File Extension Search Being Case Sensitive 2015-06-08 21:39:00 -04:00
Ryan Houdek
ffe085f5ea [AArch64] Implement CRC32 texture hashing.
In a particular hashing heavy scene in Crazy Taxi the Murmur3 hash used 3.11% CPU time.
The new CRC32 hash in the same scene used 1.86%
This was tested on a Nvidia SHIELD Android TV with Cortex-A57s.

This will be a bit slower on the Nexus 9, the Denver CPU core is a bit slower with CRC32 texture hashing than Murmur3 texture hashing.
2015-06-08 19:16:50 -05:00