Commit Graph

6054 Commits

Author SHA1 Message Date
Markus Wick
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
Ryan Houdek
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
Markus Wick
c2f38f1d16 Merge pull request #2892 from Sonicadvance1/aarch64_frsp
[AArch64] Implement frspx
2015-08-22 09:44:14 +02:00
Ryan Houdek
ce32b76be3 [AArch64] Implement frspx
Improves performance in povray bench by 2%
2015-08-22 00:35:30 -05:00
flacs
95d958c03d Merge pull request #2889 from lioncash/interp
Interpreter: Use std::isnan instead of IsNAN
2015-08-21 21:43:08 +02:00
flacs
bb7f3d1822 Merge pull request #2867 from Tilka/mtspr_hid0
Jit64: implement HID0 case of mtspr
2015-08-21 21:04:35 +02:00
flacs
01aea965ba Merge pull request #2864 from Tilka/fpscr
Jit64: implement FPSCR related instructions
2015-08-21 21:04:20 +02:00
Lioncash
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
Ryan Houdek
5f628749ff Merge pull request #2886 from Sonicadvance1/aarch64_faster_lfd
[AArch64] Optimize lfd instructions if possible.
2015-08-21 05:38:53 -05:00
Ryan Houdek
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
Markus Wick
4f45d71840 Merge pull request #2760 from Sonicadvance1/aarch64_fcmp
[AArch64] Implement fcmp{u,o}
2015-08-21 11:03:20 +02:00
Markus Wick
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
Ryan Houdek
7ce4c3138e [AArch64] Optimize cases when an FPR is only used for non-paired ops. 2015-08-20 23:36:29 -05:00
Lioncash
95c57fcec1 Jit: Remove unnecessary namespace prefixes 2015-08-20 05:20:19 -04:00
degasus
896a02b3a8 DSP HLE: Remove timing informations from ucodes
On HLE, we don't emulate the timings on HLE, so there is also no need
to setup periods callbacks.
2015-08-19 16:20:17 +02:00
degasus
7277eb0e6c AX-HLE: Call HLE on mailbox write
It was done on Update() which was called exactly every 5ms.
But the game is allowed to use the DSP more often, eg to generate 48kHz audio.
2015-08-19 16:19:06 +02:00
Lioncash
291f857b0e Merge pull request #2859 from lioncash/netplay
NetPlay: Minor cleanup
2015-08-19 04:51:28 -04:00
Tillmann Karras
0f2c656687 Jit64: implement FPSCR related instructions 2015-08-18 18:12:32 +02:00
Tillmann Karras
415ad94dbe Interpreter: fix undefined bits of mffs
Lioncash tested this on hardware.
2015-08-18 15:07:14 +02:00
Lioncash
11f3ded296 NetPlay: Return mapping arrays and player list vectors directly
Simplifies pad map dialog initialization
2015-08-17 21:56:25 -04:00
Ryan Houdek
4baaa3755e Merge pull request #2840 from zeroZshadow/master
Ignore all writes to BBA_TXFIFOCNT
2015-08-17 10:38:00 -05:00
Tillmann Karras
1c78c7b864 Jit64: implement HID0 case of mtspr 2015-08-17 13:31:57 +02:00
Pierre Bourdon
58b4b7edea Merge pull request #2863 from Tilka/fallback
Jit64: load PC only once after fallbacks
2015-08-17 10:29:54 +02:00
Tillmann Karras
fbdc20ac37 Jit64: load PC only once after fallbacks 2015-08-17 06:05:10 +02:00
Tillmann Karras
c4a6612b01 PowerPC: fix typo in FPSCR comment 2015-08-17 06:03:11 +02:00
Tillmann Karras
a9447a8606 PowerPC: drop instructions not supported by GC/Wii 2015-08-17 06:03:11 +02:00
Lioncash
b1af2a6bbc NetPlay: Use std::array for the pad mappings 2015-08-16 00:08:09 -04:00
Lioncash
e3ffb2dd16 NetPlayProto: Minor cleanup 2015-08-16 00:00:59 -04:00
Markus Wick
b8a35f6996 Merge pull request #2856 from mathieui/netplay-free
[netplay] Fix a crash
2015-08-15 23:13:18 +02:00
Markus Wick
b3045f9664 Merge pull request #2852 from aserna3/master
Changed GC adapter "Direct Connect" to false by default
2015-08-15 21:25:23 +02:00
Markus Wick
e60018abd4 Merge pull request #2854 from Tilka/valgrind
Fix some small stuff found with Valgrind
2015-08-15 20:52:12 +02:00
Ryan Houdek
909ab23df5 Merge pull request #2853 from degasus/arm
JitArm64: Fix jit clearing
2015-08-15 13:50:36 -05:00
Markus Wick
b5a5883792 Merge pull request #2851 from JosJuice/unknown-instruction
Restore old assert message for unknown instruction
2015-08-15 20:10:16 +02:00
Markus Wick
3469694b46 Merge pull request #2676 from Stevoisiak/SpellingCorrections
Minor spelling corrections
2015-08-15 18:36:56 +02:00
zeroZshadow
84764db3b2 Rewrite RecvStart error checking as suggested by shuffle2 2015-08-15 16:46:37 +02:00
mathieui
2420004af8 [Netplay] Always set a playerid
And don’t blindly dereference a pointer.
2015-08-15 15:12:20 +02:00
Tillmann Karras
b0cc02658b CachedInterpreter: avoid uninitialized value
(The CachedInterpreter backend does not support block linking yet.)
2015-08-15 13:01:26 +02:00
degasus
9bfff0d461 JitArm64: Fix jit clearing
We have to reset m_lastCacheFlushEnd on clearing.
2015-08-15 11:41:01 +02:00
Anthony Serna
b6da5490c1 Changed GC adapter "Direct Connect" to false by default 2015-08-15 02:09:47 -07:00
JosJuice
c49074742c Restore old assert message for unknown instruction
The assert(0) that was introduced in PR #2811 is not user friendly
since it has no explanation at all about what happened. Regular users
probably won't think of looking at the log to get more information.
2015-08-15 10:00:49 +02:00
Tillmann Karras
dd5cc34951 Jit64: clean up GetAllocationOrder() 2015-08-15 07:25:14 +02:00
flacs
d8d62336b5 Merge pull request #2849 from lioncash/cond
JitInterface: Fix null checking in GetProfileResults
2015-08-15 00:59:19 +02:00
Lioncash
8db43501d5 JitInterface: Fix null checking in GetProfileResults
Technically a null pointer dereference can occur here.
2015-08-14 18:51:54 -04:00
Tillmann Karras
e5a2334744 Jit64: make use of ANDN again
This time, make sure not to use it with immediates.
2015-08-14 21:25:52 +02:00
Ryan Houdek
3bc5505272 Merge pull request #2814 from Sonicadvance1/aarch64_faster_fastmem
[AArch64] Banish slowmem operations to farcode.
2015-08-14 12:03:46 -05:00
Ryan Houdek
f76a30c9fd Merge pull request #2845 from degasus/arm
JitArm64: Implement dcbt
2015-08-14 12:03:25 -05:00
degasus
715802d5aa JitArm64: Move all pointers into the map
And cleanup all of the old code for sharing and generating this pointers.
2015-08-14 11:24:11 -05:00
degasus
878f919f63 JitArm64: Fastmem: fixup map & lookup 2015-08-14 11:24:03 -05:00
flacs
9ef349181e Revert "Jit64: use BMI1's ANDN for andcx" 2015-08-14 18:10:07 +02:00
Pierre Bourdon
86c28bfac6 Bump save state version. 2015-08-14 16:17:28 +02:00