41a80fce20
x86emitter: finally support 64 bits instructions
...
Note: a couple of instructions don't support REX prefix. Some
exception ought to be implemented later
2016-01-17 17:02:38 +01:00
40438cf2bd
x86emitter: port couple of simd instrusction to xOpWrite0F template
2016-01-17 16:59:44 +01:00
9ac22ee9aa
x86emitter: rewrite remaining function to use generic template
2016-01-17 16:42:39 +01:00
fc5e293ef6
x86emitter: rewrite move to use generic template
2016-01-17 13:44:29 +01:00
25cd35147e
x86emitter: rewrite groups to use generic template
2016-01-17 12:34:44 +01:00
a9864a5619
x86emitter: create xOpAccWrite handy template for special accumulator opcode
2016-01-17 12:34:28 +01:00
d2436fa86d
x86emitter: remove unused function
2016-01-17 12:07:41 +01:00
16057d8b04
x86emitter: inc/dec are illegal on x86_64
2016-01-17 12:07:03 +01:00
f2ecfc99ad
x86emitter: align EmitRex interface on EmitSibMagic
2016-01-17 12:06:08 +01:00
e50e9a5e65
x86emitter: sed/32orLess/64orLess/
2016-01-17 00:29:19 +01:00
395d43fd9c
x86emitter: jump: use base type directly
...
which fix 16 bits call (no prefix)
2016-01-17 00:21:05 +01:00
968af6148e
x86emitter: extend SSE instruction type
2016-01-17 00:21:05 +01:00
4bc9cbc24f
x86emitter: extend lea/bswap
2016-01-17 00:21:05 +01:00
eb4db2152c
x86emitter: add x64 to push/pop
2016-01-17 00:21:05 +01:00
ef21a8dbd0
x86emitter: extend group8 type
2016-01-17 00:21:05 +01:00
39a6bda52d
x86emitter: extend test & bitscan type
...
Factorize the code to support any types
2016-01-17 00:21:05 +01:00
e285036455
x86emitter: shift: extend it to 64 bits type
2016-01-17 00:21:05 +01:00
fdaf0e9979
x86emitter:group: uses base type
...
Add 64 bits support for free ;)
Note: 64b still requires the REX prefix
2016-01-17 00:21:05 +01:00
e2602b3967
x86emitter:move: factorize type and extend it to 64 bits
...
It's still missing the REX prefix support
2016-01-17 00:21:05 +01:00
41f509873d
x86emitter: extend xIndirect32orLess to support 64 bits
...
It would be renamed later to avoid conflict with pending updates
2016-01-17 00:12:31 +01:00
39dc8368fa
x86emitter: add GetPrefix16 helper
2016-01-17 00:12:21 +01:00
be0b8cf2f5
x86emitter: implement some function to emit the REX prefix
...
Same fashion as EmitSibMagic
2016-01-17 00:10:21 +01:00
53b5ecf34a
x86emitter: add the new x86_64 registers set
2016-01-17 00:10:21 +01:00
366f793cf0
core: use xRegisterLong instead of xRegister32
...
Code needs to work with xAddressReg however the x32 inheritance doesn't
exits anymore on 64 bits.
Note: it might be possible to uses some kind of autoconversion with
xRegister32or64. Could be a future improvement.
2016-01-17 00:10:21 +01:00
1328865279
x86emitter: xAddressReg size will depend on the arch
...
Note: displacement are never 8B. The max is 4B which is sign extended to 8B.
So we can't store a pointer into it anymore.
Add xRegisterLong that will be x64 on 64 bits and x32 on 32 bits
2016-01-17 00:10:21 +01:00
a865ba8cf5
x86emitter: add various type for 64 bits
2016-01-17 00:10:21 +01:00
2fea78a6c4
x86emitter: remove not virtual inherited function
...
IsSIMD isn't virtual
2016-01-17 00:10:21 +01:00
6ec7d2c340
x86emitter: ease template stuff
2016-01-16 14:34:00 +01:00
aafa333355
added the profiler to the visual studio solution. also added preliminary cop1 (fpu) tables, ready for implementation
2016-01-15 00:26:52 +01:00
a7a8c542f5
Merge pull request #1100 from PCSX2/recompiler-abi-wrapper
...
Recompiler abi wrapper
2016-01-14 19:21:27 +01:00
3c528f0fdd
common define _32kb and _4kb
2016-01-11 18:00:25 +01:00
6e66bea152
x86emitter: improve the various abi wrapper
...
V2:
* Add various option to handle the different frame
* Basic x86-64 port (untested so likely completely broken)
v3:
Create an templated xImpl_FastCall object
v4:
clang compilation fix
2016-01-11 09:21:32 +01:00
a6eb871b42
pcsx2: use a common general intrin include
...
Avoid issue with various compiler conversion
Fix build with GCC4.8
2016-01-11 09:13:52 +01:00
afdf5cdfe2
x86emitter: remove old unused macro
...
Add some notes for a GCC "bug"
2016-01-10 16:49:50 +01:00
92d7639267
profiler: add functions to plug linux Perf Event on JIT
...
Basically it creates a /tmp/perf-`pid`.map file which will contains
a mapping of the x86 code with the EE/IOP/VU code
* You need to enable the profiler with a define
* You can split the profiling by block (inside a recomp buffer)
v2: add new file to VS xml files
v3: remove useless include
2016-01-10 13:42:59 +01:00
50c7cb1717
x86emitter: implement some BMI instructions
...
Only a couple of one to do some memory profiling
2016-01-09 23:29:01 +01:00
9e13a7bdaa
x86emitter: add support for the VEX opcode
2016-01-09 22:59:31 +01:00
29b0b17f50
x86emitter: add BMI1 & BMI2 instruction detection
2016-01-09 22:59:31 +01:00
fd56195b2d
x86emitter: drop old register define
2016-01-09 21:28:10 +01:00
834cc3f2c5
core|x86emitter: port basic jump instruction
2016-01-09 21:26:25 +01:00
743e615224
core: overload _freeX86reg with xRegister32
2016-01-09 21:15:25 +01:00
60a9463e7a
x86emitter: remove unused legacy function
...
And mark the remaining one as deprecated
Clang only as it relies on C++14
2016-01-09 21:13:29 +01:00
de2b97c640
linux compilation fix
2016-01-08 22:35:55 +01:00
74db92bee4
Merge pull request #978 from juhalaukkanen/apple_osx_master_merge
...
OSX 32bit build
2016-01-08 20:09:37 +01:00
4f28403c03
Merge pull request #972 from PCSX2/clean-gcc-opt-cmake
...
cmake: reenable a couple of missing gcc optimization
2016-01-08 19:32:18 +01:00
fb1a19f157
x86emitter: clean unused legacy type
...
Note: no need to check -1 on register allocation failure (a exception will be fired)
2015-12-02 19:09:43 +01:00
21445fe0d0
x86emitter: Add several architecture dependant wrapper
...
* A nice object to create the stack frame
* Various xStdCall/xFastCall to emulate standard & fast call
2015-12-02 19:06:52 +01:00
7edf747e84
Darwin/OSX Mutex & MAP_ANON vs. MAP_ANONYMOUS.
...
OSX compilation fix: mutex: Windows
2015-12-02 05:00:23 +02:00
fc3ff48777
Darwin/OSX SysMessage stub & __POSIX__ define.
2015-11-30 12:18:44 +02:00
52be531b8e
utilities:windows: Use standard memset for memset_8
...
aka commit 2194685
2015-11-29 11:46:59 +00:00
57ad97d212
x86emitter: Remove bool_to_char
...
It was unused.
2015-11-29 10:58:32 +00:00
4d97089990
x86emitter: Fix Windows AVX2 detection
...
Technically it did work, but it was undefined behaviour. Use __cpuidex,
which allows setting ecx as well.
Also fix the _xgetbv intrinsic - I have no idea how PCSX2 even
compiled on Windows.
2015-11-29 10:57:51 +00:00
9aec4229d5
x86emitter: support AVX2 (linux only?)
...
Nah kidding, it is only the cpu detection for the log
x86 Features Detected:
SSE2.. SSE3.. SSSE3.. SSE4.1.. SSE4.2.. AVX.. AVX2.. FMA
2015-11-28 23:40:15 +01:00
02186d5a54
x86emitter: always use fxsave intrinsic/asm
2015-11-28 22:19:42 +01:00
d5efd6eacb
x86emitter: remove the never called SIMD_ExceptionTest function
2015-11-28 22:02:08 +01:00
a59a90c734
common:windows: Add unicode compatible functions
...
It's not great, but it will work.
2015-11-25 20:03:44 +00:00
f813b9e25b
Darwin/OSX semaphore & thread & os backend.
2015-11-17 19:30:20 +02:00
21857ec12d
Merge pull request #967 from PCSX2/remove-lazy-allocation
...
Reduce lazy allocation
2015-11-15 00:12:07 +01:00
d5830ef708
cmake: reenable a couple of missing gcc optimization
2015-11-14 09:54:09 +01:00
abb4cb4810
plugin API: use const char* insead of char*
...
I hope it doesn't change the ABI
v2: fix GSnull/zzogl/zerogs
v3: duplicated code on windows...
2015-11-12 12:11:42 +01:00
4bfb60865c
common: ifdef x64 specific code
...
gcc warning: set but unused
2015-11-12 12:10:49 +01:00
26774a2fe0
recompilers: always allocate the first block
...
Avoid several annoying SIGSEGV at startup
2015-11-12 10:35:10 +01:00
7565bcc789
ee: drop SpatialArrayReserve allocator
...
Let's the kernel manage the memory either with builtin lazy allocation or
swapped memory.
Avoid to handle SIGSEGV manually (nicer for debug) and removes 250 lines of code.
2015-11-12 10:35:10 +01:00
44d89308c9
Revert "Revert "gsdx|spu2-x: Use atlcomcli.h for CComPtr definition""
...
This reverts commit 9bb990e9ed
2015-11-09 20:55:07 +00:00
9bb990e9ed
Revert "gsdx|spu2-x: Use atlcomcli.h for CComPtr definition"
...
This reverts commit 0eb771c2a6
2015-11-08 20:46:41 +00:00
49ff324882
Merge pull request #955 from turtleli/replace-comptr-h
...
gsdx|spu2-x: Use atlcomcli.h for CComPtr definition
2015-11-08 13:27:28 +00:00
870e799baa
memcpy: remove AMD copyright
...
memcmp comes from zerofrog (him or external) but likely not AMD
https://www.cs.virginia.edu/stream/FTP/Contrib/AMD/memcpy_amd.asm
2015-11-07 15:57:34 +01:00
0eb771c2a6
gsdx|spu2-x: Use atlcomcli.h for CComPtr definition
...
VS2013/VS2015 community includes it, so everything will still compile
for everyone.
Also delete common/include/comptr.h.
2015-11-06 18:52:33 +00:00
08dfa20da1
utilities:linux: Add function to change stdout stream
...
stdout is not necessarily at stdout - at least not after I'm done with it.
2015-10-31 13:50:02 +00:00
618ef8b549
pcsx2:linux: Use narrow characters for stdout
...
Things seem to break when you mix wide and narrow characters.
Plugins messages sent to stdout are now viewable.
2015-10-29 19:09:49 +00:00
2194685d76
memset-linux: use std memset for memset_8
...
This memset is only used in the init (to clear the page that will
contains the dynarec dispatcher
2015-10-28 14:25:12 +01:00
cfd02bff53
Utilities: Initialize all class members
...
CID 147010 (#1 of 1): Uninitialized pointer field (UNINIT_CTOR)4. uninit_member: Non-static class member Enabled is not initialized in this constructor nor in any functions that it calls.
2015-10-25 14:11:34 +05:30
4f86cca306
common: don't return a boolean for an assert
...
It might help to fix those 2 coverity reports.
CID 151744 (#1 of 1): Useless call (USELESS_CALL)
side_effect_free: Calling EnumAssert(id) is only useful for its return value, which is ignored
CID 151745 (#1 of 1): Useless call (USELESS_CALL)
side_effect_free: Calling EnumAssert(id) is only useful for its return value, which is ignored.
2015-10-17 11:49:04 +02:00
75b2406997
macros: fix pxReleaseCode (was inverted), add pxNonReleaseCode
...
pxReleaseCode was not used at the code so nothing is affected.
pxNonReleaseCode will be used at the next commit.
2015-10-07 04:08:09 +03:00
927dd827ce
common: make DESTRUCTOR_CATCHALL macro really exception safe
...
Console.Error() can trigger some exceptions (like out of memory)
v2:
Add a default fallback catch(...) in case someone badly add a new
exception in the codebase
2015-09-25 18:41:19 +02:00
2eb73644e9
pcsx2|common: use DESTRUCTOR_CATCHALL to catch exception in various destructor
...
Allow to print a nice error message instead of a brutal abort in case of
massive failure
2015-09-23 09:47:27 +02:00
8083b2c754
common: init values in the constructors of SpatialArrayReserve and pxStaticText
2015-09-23 09:47:27 +02:00
2c914b9b8f
common:FastFormatString: remove unimplemented m_Length
2015-09-23 09:47:27 +02:00
a6c29a5061
common: always set a default color in trace log
2015-09-23 09:47:27 +02:00
3064d0c300
common: init m_handled in the SrcType_PageFault constructor
...
Coverity:
CID 147021 (#1 of 1): Uninitialized scalar field (UNINIT_CTOR)i
2. uninit_member: Non-static class member m_handled is not initialized in this constructor nor in any functions that it calls.
2015-09-23 09:47:27 +02:00
b60dd9d1fe
Merge pull request #814 from PCSX2/coverity-qa-high
...
Fix a bunches of coverity warning/errors.
2015-09-20 22:18:03 +02:00
d10bbb73f7
FixedInt - round instead of truncate on construction
...
This could have caused issues where e.g. Fixed100(59.94) differed from
Fixed100::fromString("59.94") due to precision compilation flags
(the former could, and did on Devel builds, end up with Raw == 5993, which
differs from the value constructed from the string at the ini file,
and then it would be incorrectly identified as a custom rate).
Rounding seems the more likely intention when effectively decreasing the
precision of a value.
Unlikely that we have code which depends on truncating behavior, though not
impossible.
2015-09-17 00:30:08 +03:00
67dc65ea91
linux: Fix _aligned_realloc implementation
...
Out-of-bounds memory is no longer accessed if the realloc size is larger.
If reallocation fails, the old memory will not be freed and a memcpy
will not take place.
This should match the Windows _aligned_realloc behaviour, except that an
extra parameter is used.
2015-09-13 21:05:33 +01:00
3df5c2835e
Remove useless postBuild stuff
2015-09-13 04:31:41 -04:00
f1cb2cfa35
Fix segmentation fault with the new libstdc++ ABI.
...
Local variable goes out of scope and everything goes boom.
2015-09-12 17:35:45 -04:00
53680d538d
Merge pull request #817 from micove/GIT_TZ
...
Fix git version detection.
2015-09-12 06:33:00 -04:00
5c6915f633
utilities: Simplify _aligned_malloc wrapper
...
Windows doesn't actually use it at all, so let's exclude it from the
VS build. Also include ScopedAlloc.h into the VS project file.
2015-09-12 10:01:04 +01:00
8312e21f00
utilities: Cleanup unused memcpy defines
...
All refer to memcpy, and only memcpy_fast is used, so there's no point
keeping them.
Also remove the _memset16_unaligned function prototype since there's no
function definition for it.
2015-09-12 10:00:49 +01:00
239306e9a2
pcsx2|gsdx:linux: Fix posix_memalign
...
The AddressSanitizer can be used on Linux again. Well, PCSX2 doesn't get
very far when it's used though.
2015-09-12 09:58:00 +01:00
49072d6e02
Fix git version detection
...
- It currently fails to detect a 64bit git.
- It currently breaks with a TZ left of the Greenwich meridian.
- The output of git show -s can cause compilation errors pipe to NUL.
- Pipe the two can't find the git command error messages to NUL too.
2015-09-11 19:20:23 -04:00
0260c9119e
common: suspicious missing comma
2015-09-11 08:54:29 +02:00
e2f1bafeef
gui:windows: Fix tooltip wrapping
...
wxWidgets 3.0 now wraps tooltips automatically on Windows, so the custom
tooltip wrapping functionality is no longer necessary.
2015-09-04 22:54:11 +01:00
cc1d50a06c
gui:windows: Show tooltips for as long as possible.
...
It'll show for 32.767s. Windows unfortunately doesn't have an infinite
value, and also uses a signed 16-bit integer for the delay.
2015-09-01 18:14:03 +01:00
f29dd8b927
Fix Windows XP Home vs Pro detection.
...
VER_SUITE_PERSONAL is set for XP Home therefore the check was backwards.
2015-08-26 16:02:26 -04:00
45fc9603e3
Recognize Win 8, 8.1 and 10.
...
https://msdn.microsoft.com/en-us/library/windows/desktop/dn481241(v=vs.85).aspx
Windows 8.1 and 10 require a manifest file or they will be reported as
6.2 which is Windows 8.
.
Also add the information for Win 8, 8.1 and 10 in the WinMisc.cpp file.
2015-08-25 21:30:27 -04:00
c97c4f878a
Update pthreads project.
...
- Add x64
- Use static library
2015-08-25 09:53:00 -04:00
09c8a41294
Update SoundTouch to 1.9.0.
...
It claimed to be 1.7.1 but it had a mixture from various
versions. It was hard to update as everything in the top directory
so I used upstream's way to organize files. I renamed include to
soundtouch since I did not want to #ifdef that for windows.
.
Wavfile.h is a private header so I used the private path instead of
moving the file over. This changed 3 files in the plugin folder.
2015-08-22 10:16:27 -04:00
a5a510656f
x86emitter requires the setup.h generated by wx.
...
Build wx30config before x86emitter.
2015-08-19 18:28:42 -04:00
a1de0614d7
Remove vs2012 files.
...
It does not support several things from C99/C++11 that are
used in the project.
2015-08-18 01:15:51 -04:00
51bd9fe40e
Fix compilation errors of the old plugins.
...
- zerospu2: include stdint.h in Windows. (VC2012+)
- CDVDolio: Remove hash_map (not used, VC2015+)
- zerogs: Fix extern and link to utilities. (VC2012+)
- zzogl: Port windows part to wx30. (VC2012+)
2015-08-18 00:45:03 -04:00
Gregory Hainaut