Unknown W. Brackets
50e9e45d65
Check version in each DoState() func.
...
They bail on PointerWrap error or bad version.
2013-09-14 20:23:03 -07:00
Unknown W. Brackets
2ac5bcff1f
Process savestates during single stepping.
...
This way we can handle them on the same thread, without weird delays.
2013-09-14 18:43:23 -07:00
Henrik Rydgård
3bc05a9941
Merge pull request #3764 from unknownbrackets/scheduling
...
Improve dispatch/interrupt handling in start/create thread
2013-09-14 02:57:09 -07:00
Henrik Rydgård
04b49dc42f
Merge pull request #3760 from unknownbrackets/mips-minor
...
Fix fpu non-finite rounding on x86
2013-09-14 02:56:24 -07:00
Unknown W. Brackets
cff9a7b819
Don't allow thread start/create inside interrupts.
...
This fixes the volatile power test, and returns the right message even for
bad arguments.
2013-09-14 02:48:10 -07:00
Unknown W. Brackets
cbf1df9b01
Check for nan/inf in trunc.w.s in x86 jit.
...
Now x86 jit passes the fpu test too.
2013-09-13 22:32:25 -07:00
Unknown W. Brackets
7b71dc54dd
Don't say nan is equal to anything.
...
I guess we were passing this because of fastmath? Fixes interpreter
cpu/fpu/fpu test on Windows anyway.
2013-09-13 21:28:59 -07:00
Unknown W. Brackets
455431ba13
Integer round inf/nan to INT_MAX/INT_MIN.
...
Instead of 0x80000000 (the default "invalid" value.)
Improves the fpu test.
2013-09-13 21:27:31 -07:00
The Dax
9dbf2b9d90
Add proper fullscreen and windowed launch arguments. Remove bFullScreenOnLaunch.
2013-09-12 16:56:18 -04:00
Unknown W. Brackets
ca6491ea22
Auto resume dispatch at thread start/create.
2013-09-12 08:02:43 -07:00
Ced2911
17b8d5d813
More recompiled instruction (fpu/alu)
2013-09-12 10:40:14 +02:00
Ced2911
23637db54c
more recompiled fpu (not tested)
2013-09-12 10:34:10 +02:00
Ced2911
e90d9c035a
more vfpu speed
2013-09-12 10:25:15 +02:00
Ced2911
bfe9018eb1
enable,fixe more vfpu
2013-09-12 10:25:15 +02:00
Ced2911
b90fde6c06
some fixes
2013-09-12 10:25:15 +02:00
Ced2911
e5024135a6
more on vfpu
2013-09-12 10:25:15 +02:00
Ced2911
a91d8bebe6
update ppc jit
2013-09-12 10:22:14 +02:00
Ced2911
7bf623d339
oops missed fpu reg cache
2013-09-12 10:22:01 +02:00
Ced2911
f8b5496bc8
some new recompiled instructions
2013-09-12 10:21:50 +02:00
Ced2911
dc41d0f0eb
recompile Fpu !
2013-09-12 10:21:39 +02:00
Ced2911
2ad39d3999
Almost all alu recompiled
...
Current issues:
Comp_RType3(min/max): Can't select start in disgaea
Comp_ShiftType(srl/srlv?): Crash ridge racer 2
2013-09-12 10:21:26 +02:00
Ced2911
fb24e0ed28
Add some badly tested instruction
2013-09-12 10:20:34 +02:00
Ced2911
47280f007f
implement sltiu, slti is bad
2013-09-12 10:20:17 +02:00
Ced2911
d4acf0edd7
more jitted instructions
2013-09-12 10:20:02 +02:00
Ced2911
24576fd72f
recompile shift left/right immediate
2013-09-12 10:19:47 +02:00
Ced2911
52f4cccfa4
fix Jit
2013-09-12 10:19:35 +02:00
Ced2911
3350061197
use _alloca
2013-09-12 10:19:23 +02:00
Ced2911
c98090e095
jit fixes
2013-09-12 10:19:11 +02:00
Ced2911
c94d9b62d0
Update ppc jit
2013-09-12 10:18:59 +02:00
Unknown W. Brackets
2c763365b4
Check size is within bounds in AllocAt().
2013-09-11 22:52:22 -07:00
Henrik Rydgård
e8b8ed7d65
Merge pull request #3733 from thedax/simplifyOSKBypass
...
Win32: Simplify NativeKeyboard(OSK bypass), and speed up the char copy loop.
2013-09-11 13:23:49 -07:00
Henrik Rydgard
91f9161d76
Fix some bugs found in reports in the Android crash reporting console
2013-09-11 22:21:32 +02:00
The Dax
58aab2eeb8
Simplify NativeKeyboard, and only copy the amount of bytes of the output plus null terminator.
2013-09-11 15:42:32 -04:00
Unknown W. Brackets
db7496a37c
Write only changed characters from OSK.
2013-09-11 08:04:01 -07:00
Unknown W. Brackets
a8c7421a0b
Add some additional and mostly pointless accuracy.
2013-09-11 00:15:49 -07:00
Unknown W. Brackets
8e69b70bf5
Start sceRtc at now, not 1970.
2013-09-11 00:12:33 -07:00
Henrik Rydgård
e875e8597a
Merge pull request #3728 from unknownbrackets/savedata-minor
...
Don't copy a negative size in savedata
2013-09-11 00:02:48 -07:00
Unknown W. Brackets
5352432d8f
Don't copy a negative size in savedata.
...
Not sure what's correct here, Street Fighter Alpha 3 passes -1.
It seems to call it twice, so probably the first time is to get the size?
Fixes #3722 , thanks @thedax.
2013-09-10 22:23:47 -07:00
Henrik Rydgard
2745744d99
Split the resolution setting into Window Size and Rendering Resolution (internal resolution).
...
Gets rid of our "antialiasing" (simply crank up internal resolution to 2x your window size).
Proper MSAA is coming later.
2013-09-11 00:19:34 +02:00
Henrik Rydgård
fb48defa75
Experiment: Disable VROT on x86 32-bit except on Windows where I know it works.
2013-09-10 18:17:00 +02:00
Henrik Rydgård
5a9bed50bf
Merge pull request #3715 from unknownbrackets/savedata-minor
...
Fix custom save dialog dates
2013-09-10 02:12:48 -07:00
Unknown W. Brackets
e75ea49a66
Fix #3706 , custom save dialog dates.
2013-09-10 02:09:18 -07:00
Henrik Rydgård
d20e7a93aa
Merge pull request #3713 from unknownbrackets/thread-stack
...
Correct sceKernelGetThreadStackFreeSize()
2013-09-10 02:04:26 -07:00
Unknown W. Brackets
9fb64183e6
Don't write more than the savedata buffer.
...
Fixes Numblast when there's savedata, probably others.
Could use more testing.
2013-09-10 01:46:49 -07:00
Unknown W. Brackets
b898e397aa
Correct sceKernelGetThreadStackFreeSize().
...
It was always returning 0.
2013-09-10 01:31:03 -07:00
Unknown W. Brackets
0f7ac377cd
Offset the root thread's stack by arg size.
...
Otherwise it'll just overwrite, right?
2013-09-10 01:12:51 -07:00
Henrik Rydgård
ff0eb6eb62
Merge pull request #3712 from unknownbrackets/volatile-mem
...
Improve volatile mem locking w/ suspended intr
2013-09-09 23:57:26 -07:00
Henrik Rydgård
164d0972a7
Merge pull request #3711 from unknownbrackets/thread-minor
...
Fixes to various thread functions + heap corruption fix
2013-09-09 23:56:07 -07:00
Unknown W. Brackets
6d949f4d0b
Improve volatile mem locking w/ suspended intr.
2013-09-09 23:49:06 -07:00
Unknown W. Brackets
15f66f2ff2
Log when syscalls fail due to intr/dispatch.
2013-09-09 23:14:40 -07:00
