Commit Graph

1396 Commits

Author SHA1 Message Date
Unknown W. Brackets
5094bd0ea1 Add include for Android NDK r12. 2016-07-24 12:52:24 -07:00
Unknown W. Brackets
0bf65150d6 Vulkan: Prefer 32-bit color formats if possible. 2016-07-01 14:05:15 -07:00
LunaMoo
24fd336e60 Create openIniFile function in FileUtil to avoid duplicates
+ correction
2016-06-22 05:55:45 +02:00
Unknown W. Brackets
9471107d0d If possible, use /dev/shm instead of /tmp.
For some variants, /tmp will be tmpfs, but not always.  In case our mmap
fails to prevent flushing to disk (as apparently happens on Raspberry Pi),
shm will work better for us.

Thanks go to klyoungblood.
2016-06-03 19:49:22 -07:00
Unknown W. Brackets
30e13c7c37 Clean up some unused result warnings. 2016-05-31 11:49:59 -07:00
Unknown W. Brackets
419b960403 x86jit: Fix params in Replace_fabsf. 2016-05-31 10:40:14 -07:00
Unknown W. Brackets
02db45982a Pass Build.BOARD from Java, avoid lookup.
Fixes arm64 build with r10, but is ugly coupling.
2016-05-27 20:11:25 -07:00
Henrik Rydgard
fd0af5b271 Vulkan: Mostly pass the validation layers in SDK 1.0.13 2016-05-24 21:41:35 +02:00
Unknown W. Brackets
4113fd940c Add ugly invalidation workaround for SGS7s.
Otherwise they just crash, and crash often.  Special thanks to Jaaan for
numerous trials to try to find the best way to solve the crashes.
2016-05-23 21:35:28 -07:00
Unknown W. Brackets
70c78effb0 Use Android system props for CPU info fallback.
Some devices don't provide any useful information at all, such as the
Galaxy S7.
2016-05-23 21:18:57 -07:00
Henrik Rydgard
0a58424e75 Vulkan: Remove validation layer that is now included in standard_validation 2016-05-18 23:53:17 +02:00
Unknown W. Brackets
9cda186f4b Remove __TIME__ from Log.h.
This seems to cause more ccache misses, and I don't think it's terribly
useful really.
2016-05-14 10:40:21 -07:00
Henrik Rydgard
52517ab609 Fix the fix 2016-05-08 22:32:08 +02:00
Henrik Rydgard
ed0a0378d7 Another buildfix 2016-05-08 22:27:20 +02:00
Unknown W. Brackets
9280bb3eda Merge pull request #8715 from unknownbrackets/tex-replace
Initial texture replacement support
2016-05-01 09:58:16 -07:00
Henrik Rydgard
ffe4c266ef Add CodeBlockCommon base class to remove further arch-specificity in JitBlockCache
Remove unused ArmThunk.
2016-05-01 11:40:00 +02:00
Henrik Rydgard
a5be0976bd Remove preprocessor hacks to choose JIT implementation.
Opens up for having multiple JIT implementations available at runtime,
which could be use for experimenting with new JIT compiler types or for
unit testing one JIT on another architecture.

Very few of the newly virtual calls are on any sort of critical path so
hopefully there will not be a performance loss.
2016-05-01 11:39:53 +02:00
Unknown W. Brackets
9ffc717de1 Properly save 16-bit textures for replacements. 2016-05-01 00:30:47 -07:00
Henrik Rydgård
0c1e7dcc18 Use the OS to clear the CPU cache instead of the gcc builtin. Experiment for the S7 issues. 2016-04-09 16:16:45 +02:00
Henrik Rydgard
617bcebcd6 Update to latest vulkan header (1.0.8) 2016-04-07 22:53:37 +02:00
Henrik Rydgard
5cd3956575 Reformat VulkanContext.cpp (fix tab indentation) 2016-04-03 00:01:56 +02:00
Henrik Rydgard
e1f660ebc1 Update vulkan structure init to the new style 2016-04-02 23:57:23 +02:00
Henrik Rydgard
32f5930e15 Move clear handling to FramebufferVulkan, center the clear properly in non-buffered. more cleanup. 2016-03-31 00:50:50 +02:00
Henrik Rydgard
731af0d0b4 Add some infrastructure for 2D drawing, to be used by the Vulkan framebuffer manager 2016-03-31 00:49:10 +02:00
Henrik Rydgard
40d05e292c Cleanup in FramebufferVulkan to reduce the size of future diffs 2016-03-28 20:23:29 +02:00
Henrik Rydgard
a8c386b94c Make VulkanImage a little more flexible (make it possible to create depth images)
Just preparation.
2016-03-28 18:25:05 +02:00
Unknown W. Brackets
8dd93576ec Vulkan: Ignore queued frees after destroy.
We flush slabs so this gives false assertions in some cases.
2016-03-27 12:38:15 -07:00
Unknown W. Brackets
048a12f589 Buildfix.
Some compilers didn't like that, oops.
2016-03-27 09:25:25 -07:00
Unknown W. Brackets
3352608027 Vulkan: Free texture slabs on exit. 2016-03-27 08:30:43 -07:00
Unknown W. Brackets
16570f10bd Vulkan: Handle texture allocation failure.
Users hit out of memory even using desktop GL devices, and it will
definitely be possible on mobile and desktop Vulkan.
2016-03-27 08:30:42 -07:00
Unknown W. Brackets
27a5697a96 Vulkan: Use the slab allocator for textures. 2016-03-27 08:30:42 -07:00
Unknown W. Brackets
62b6432121 Vulkan: Allocate forwards in device slabs. 2016-03-27 08:30:41 -07:00
Unknown W. Brackets
87d65e696c Vulkan: Correct alloc near slab end. 2016-03-27 08:30:40 -07:00
Unknown W. Brackets
9921fd2ea7 Vulkan: Don't bother with extra userdata, comments.
Was a failed attempt to avoid new/delete.
2016-03-27 08:30:40 -07:00
Unknown W. Brackets
bd7c431532 Vulkan: Increase new size for new slabs. 2016-03-27 08:30:39 -07:00
Unknown W. Brackets
973a9f6124 Vulkan: Delay frees for device memory slabs. 2016-03-27 08:30:38 -07:00
Unknown W. Brackets
7d5a8aa470 Vulkan: Protect better against memory alloc errors. 2016-03-27 08:30:38 -07:00
Unknown W. Brackets
f0dc921ed5 Vulkan: Setup a pool memory manager for textures. 2016-03-27 08:30:37 -07:00
Unknown W. Brackets
e7ea1da99f Vulkan: Swizzle instead of converting colors. 2016-03-26 15:02:06 -07:00
Unknown W. Brackets
adc9946641 Vulkan: Ask for COHERENT staged texture memory. 2016-03-26 12:18:49 -07:00
Henrik Rydgård
b0329e8106 Merge pull request #8657 from unknownbrackets/vulkan-resize
Support resizing the screen on Vulkan
2016-03-22 09:22:58 +01:00
Unknown W. Brackets
963b18e6c2 Vulkan: Prep for resize, destroy surface on exit.
I suppose this is a validation layer bug, it didn't report the leak.
2016-03-21 19:41:20 -07:00
Henrik Rydgard
b97e02febe Typo fixes. Delete irrelevant comments 2016-03-21 20:23:53 +01:00
Henrik Rydgard
dff585e440 VulkanContext: Detect the appropriate depth buffer format to use, expose it.
This should help AMD compatibility.
2016-03-21 20:11:49 +01:00
Henrik Rydgård
6831f6fc46 Merge pull request #8650 from unknownbrackets/vulkan-buf
Vulkan: Cache only device in push buffers
2016-03-21 08:27:50 +01:00
Henrik Rydgård
2a35d4b4c5 Merge pull request #8653 from unknownbrackets/warnings
Warning fixes from clang
2016-03-21 08:27:25 +01:00
Unknown W. Brackets
2d34c0a8bc Vulkan: Free anything pending before the device.
This includes, for example, device memory and etc.
2016-03-21 00:02:59 -07:00
Unknown W. Brackets
9c3e12ead0 Remove some unused things reported by clang. 2016-03-20 23:14:08 -07:00
Unknown W. Brackets
d90527e41e Correct some missing overrides. 2016-03-20 23:05:05 -07:00
Unknown W. Brackets
6d947a057e Vulkan: Handle oversized push buf allocations. 2016-03-20 22:49:38 -07:00
Unknown W. Brackets
9fd040eed0 Vulkan: Cache only device in push buffers. 2016-03-20 22:49:38 -07:00
Unknown W. Brackets
e3d9630b9a Vulkan: Handle oom slightly better. 2016-03-20 15:25:59 -07:00
Unknown W. Brackets
9e4389cbf5 Vulkan: Defrag buffers when growing. 2016-03-20 15:20:42 -07:00
Unknown W. Brackets
f2513ffcd6 Vulkan: Allocate more buffers on overflow. 2016-03-20 15:15:13 -07:00
Unknown W. Brackets
ef1dc583a2 Fix various minor warnings. 2016-03-20 14:17:51 -07:00
Unknown W. Brackets
87eadad326 Fix a failed lock release warning. 2016-03-20 14:16:54 -07:00
Henrik Rydgard
0b1cfaf751 Be more economical with UBO pushbuffer space by reusing the last data when possible. 2016-03-20 19:35:31 +01:00
Henrik Rydgard
5d19f3dfb8 Decode vertex data directly into the vertex pushbuffer, saving a memcpy. 2016-03-20 19:35:03 +01:00
Henrik Rydgard
5d45413dab Fix a resource leak. Request COHERENT memory for pushbuffers. 2016-03-20 19:35:03 +01:00
Henrik Rydgard
2c61b1e763 Change the VulkanPushBuffer API to allow for adding support for dynamic growth 2016-03-20 19:35:03 +01:00
Henrik Rydgard
5d56537a16 Move VulkanPushBuffer into VulkanMemory.h 2016-03-20 19:31:02 +01:00
Henrik Rydgard
966cb89d41 Vulkan: Fix our use of dual source blending, re-enable if available. 2016-03-20 19:31:02 +01:00
Unknown W. Brackets
82f3df1e06 Vulkan: Initial support for aniso filtering. 2016-03-20 19:31:02 +01:00
Henrik Rydgard
da50370328 Split out VulkanTexture from VulkanContext.cpp/h into VulkanImage.cpp/h 2016-03-20 19:31:02 +01:00
Henrik Rydgard
e45c24b75f Show how much pushbuffer space is used each frame.
Optimizing this number (by avoiding redundant UBO uploads etc) will probably help performance.
2016-03-20 19:31:02 +01:00
Henrik Rydgard
84e649f74b Vulkan: Enable mipmapping 2016-03-20 19:31:02 +01:00
Henrik Rydgard
8a62724316 Upload PSP textures through a push buffer instead of image copy. More flexible. 2016-03-20 19:31:02 +01:00
Henrik Rydgard
5ea01ffff6 Add a secondary way to upload textures - through buffers. 2016-03-20 19:31:02 +01:00
Unknown W. Brackets
82205a4660 Handle Vulkan init errors more gracefully. 2016-03-20 19:31:02 +01:00
Henrik Rydgard
fe270c592a Typo fix 2016-03-20 19:31:02 +01:00
Henrik Rydgard
b0098f7961 Replace ugly workaround with VK_LAYER_LUNARG_unique_objects which prevents handle reuse.
See https://github.com/KhronosGroup/Vulkan-LoaderAndValidationLayers/issues/121
2016-03-20 19:31:02 +01:00
Henrik Rydgard
575cc890b7 Turn off annoying "debug flicker" I used to see if we were swapping buffers properly early on
Also minor logging changes, refine the false-positive debug layer check
2016-03-20 19:31:02 +01:00
Henrik Rydgard
5ca012e83d Fix a bunch of new vulkan validation layer failures. 2016-03-20 19:31:02 +01:00
Henrik Rydgard
6e2e0bfa33 Should not specify the Vulkan patch version when requesting Vulkan 1.0. 2016-03-20 19:31:02 +01:00
Unknown W. Brackets
c92790c5a8 Specify a useful app version when loading Vulkan. 2016-03-20 19:31:02 +01:00
Henrik Rydgard
432fbe30d1 Vulkan: We're doing dual src blending wrong (GLSL part, not sure how to do it right) so disable it. 2016-03-20 19:31:02 +01:00
Henrik Rydgard
b06cf4e688 Re-fix depth buffer initialization 2016-03-20 19:30:39 +01:00
Henrik Rydgård
9c457da4c9 Gets the Vulkan backend running, but not drawing correctly, on Android.
Not sure what's wrong, I get a yellow flashing screen.
2016-03-20 19:30:39 +01:00
Scott Mansell
77420f799e VulkanLoader: Detect loader failure on windows. 2016-03-20 19:30:39 +01:00
Scott Mansell
e97a6c6f77 VulkanLoader: Load vkDestroySurfaceKHR 2016-03-20 19:30:39 +01:00
Henrik Rydgard
d325aa2b30 Hook up Vulkan on Android, no idea if it works. Move VulkanContext to Common. 2016-03-20 19:30:11 +01:00
Henrik Rydgard
2481714a8e Load Vulkan dynamically. Fix Android build (though - Vulkan is not working on Android yet) 2016-03-20 19:30:11 +01:00
Henrik Rydgard
4063f7e0db Fill out GPU_Vulkan.cpp with what's mostly a copy of GLES_GPU.cpp 2016-03-20 19:30:11 +01:00
Henrik Rydgard
2a71e164f5 Enable vulkan validation 2016-03-20 19:25:30 +01:00
Henrik Rydgard
2a449f78bb More vulkan fixing 2016-03-20 19:25:30 +01:00
Henrik Rydgard
eedd81988c Implement a trivial SPIR-V disassembler, just for fun 2016-03-20 19:18:35 +01:00
Henrik Rydgard
c8457d39ed Fix a bunch of compiler warnings, delete some unused code 2016-03-13 15:17:31 +01:00
Henrik Rydgard
ec6559537b Android EGL: Experiment: Only apply the 16-bit workaround on really old devices. 2016-03-13 15:17:31 +01:00
Unknown W. Brackets
f1f046d885 Remove now-unused mutex implementation. 2016-03-06 14:49:16 -08:00
Unknown W. Brackets
da03b80c97 Standardize on just one mutex implementation. 2016-03-06 14:49:15 -08:00
Henrik Rydgard
88f25fd50e x86-64: Fix L bit in VEX instruction emitter. Ported fix from Citra.
Currently unused in the emulator, though.
2016-02-28 13:07:24 +01:00
Henrik Rydgard
05d4caeeaf Buildfix 2016-02-21 12:14:44 +01:00
Henrik Rydgard
76e1d074ec Logging improvements in EGL init 2016-02-21 11:40:02 +01:00
Henrik Rydgard
35b6c7341b Fix a bunch of memory leaks, after enabling file+line-number memory leak tracking for the MS Debug CRT.
Few of these were serious, most were on-exit, but good to have a cleaner memory report after running.
2016-02-10 15:36:10 +01:00
Unknown W. Brackets
f1afb50fd1 Shutdown EGL properly when quitting render loop. 2016-02-07 16:30:40 -08:00
Henrik Rydgard
da528d15ad ARM32 emitter: Tighten up distance check for forward jumps. Was off by one. 2016-01-30 21:38:42 +01:00
Henrik Rydgard
e0cba1dd5a EGL: Try again with ES 2 if we fail to create an ES 3 context. 2016-01-24 11:33:02 +01:00
Henrik Rydgård
19b0858d5e Merge pull request #8490 from unknownbrackets/ui-tweaks
Show save state titles in the save state manager
2016-01-24 00:51:04 +01:00
Unknown W. Brackets
6a5096e546 Save the game title in savestates, if available.
So that the UI can show them.  Old savestates still load fine, just don't
have the title.
2016-01-23 12:56:36 -08:00
Henrik Rydgard
81fe914137 More EGL heuristics 2016-01-23 19:16:46 +01:00
Henrik Rydgard
1d6e48b817 Even more EGL logging 2016-01-23 18:42:05 +01:00
Henrik Rydgard
1a02274def A little bit more EGL logging 2016-01-23 18:31:34 +01:00
Henrik Rydgard
4155712113 Improve EGL logging, fallback to a really simple config if the normal one fails. May help #8469, if not may at least give us more info. 2016-01-23 13:17:53 +01:00
Henrik Rydgard
3ba911bf07 Replace some calls to fopen with File::OpenCFile (utf8). 2016-01-17 15:59:20 +01:00
Unknown W. Brackets
da78ec0b4d Use USING_WIN_UI for desktop Windows console. 2016-01-03 15:10:50 -08:00
Unknown W. Brackets
d165680ee7 Don't assume _MSC_VER / _WIN32 mean x86/x64. 2016-01-03 15:07:36 -08:00
Henrik Rydgård
b29f46f9ac Merge pull request #8363 from unknownbrackets/jit-alloc
Jit: Properly re-allocate when jit is freed a lot
2016-01-03 18:55:12 +01:00
Unknown W. Brackets
f3a616df24 Jit: Properly re-allocate when jit is freed a lot.
Fixes #8334.
2016-01-02 13:24:58 -08:00
Henrik Rydgard
ae3458ec6e Buildfixes linux 2016-01-01 14:40:17 +01:00
Henrik Rydgard
15de6e6b98 GraphicsContext: Abstract away things like swapbuffers etc before adding even more backends.
Needed to prevent clutter all over the codebase.

Does not go all the way yet, goal would be a common render loop between platforms but not there yet.
2016-01-01 14:40:16 +01:00
gmvbif
281f0e18e6 Fixed some errors from PVS-Studio in Core project
FixedSizeQueue
- Added private operator=
- Changed signature of EndPop
Some changed to printf like functions.
Fixed typo in syn-att.c
2015-12-31 10:25:11 +03:00
latot
c7b98b84c2 Add extended keys(latam) 2015-12-27 15:45:50 -03:00
gmvbif
dff476c4f4 Make field for HWND in ConsoleListener 2015-12-26 09:54:34 +03:00
gmvbif
ae1c1a0295 Fixed some errors from PVS-Studio in Common project
Color conversion assigned 'a' twice.
Console listener use local variable with the same name a class member.
Some additionals to nocopy for FileUtils and mutex
2015-12-25 09:49:26 +03:00
Unknown W. Brackets
103b959d50 Fix a type comparison warning. 2015-12-23 15:22:13 -08:00
Henrik Rydgard
ebf9bd74c6 Minor fixes 2015-12-17 19:38:23 +01:00
Henrik Rydgard
fd2557dbd7 EGL: Add autodetect mode that only chooses between ES2 and ES3 2015-12-16 21:48:19 +01:00
Henrik Rydgard
db633aa377 Lots more debug logging. Terrifying compatibility hack for devices that for some reason create 16-bit buffers.
Like Xperia Play.
2015-12-16 21:37:39 +01:00
Henrik Rydgard
13515bf235 Use the hardware scaler again on Android. Move all DPI handling code from java to app-android.cpp. 2015-12-16 14:52:44 +01:00
Henrik Rydgard
04f8bffa29 Use EGL from native code to initialize and run OpenGL on Android.
Should be more robust, allows initializing desktop GL if available, and lets us take control of the render loop.
2015-12-16 10:57:24 +01:00
Henrik Rydgard
c2eb6a2fcb Minimal port of Android EGL code from Dolphin. 2015-12-13 22:46:50 +01:00
Unknown W. Brackets
7bfe100b0f Fix some unused variable warnings.
The CheckAlpha one looks like it will matter.
2015-11-25 16:11:53 -08:00
Henrik Rydgard
44be9f2a50 Move Windows GPU init code into Windows/GPU 2015-11-03 23:22:09 +01:00
Henrik Rydgard
c84f0e6059 Show the id of unknown mapped keys in the UI. 2015-10-31 15:20:02 +01:00
latot
afeb012d04 Add missing include files to unittest 2015-10-24 13:24:05 -03:00
Henrik Rydgård
e0de346036 Merge pull request #8046 from unknownbrackets/jit-alloc
Reset memory alloc pointer when jit alloc fails
2015-10-10 18:13:49 +02:00
Unknown W. Brackets
935b802dad Reset memory alloc pointer when jit alloc fails.
On Windows x64, we keep track of our last allocation.  Once we run out of
our 2GB range, we need to start over or it'll just fail.

Fixes #8021.
2015-10-10 09:07:34 -07:00
Henrik Rydgard
8914cd9914 ARM64 disassembler improvements (show many kinds of branch targets properly) 2015-10-10 11:27:28 +02:00
Henrik Rydgard
6785a95384 Fix some memory leaks on corrupt savestates 2015-10-04 14:11:33 +02:00
Henrik Rydgård
5dc6e96168 Fix key mapping defaults on nVidia Shield TV 2015-10-04 13:25:57 +02:00
Henrik Rydgard
6d5bde9748 Fix more buggy string arithmetic in LogManager 2015-10-04 11:09:44 +02:00
Henrik Rydgard
8996f58693 Generate save state IDs for homebrew using the elf or directory name where applicable. 2015-09-25 19:27:40 +02:00
Henrik Rydgard
6fcbfabdd2 Rename fields to avoid some gross #define hack that linux is doing 2015-09-23 12:10:23 +02:00
Henrik Rydgard
77e9ea38aa Get rid of remaining uses of stat() on Windows 2015-09-23 11:34:22 +02:00
Henrik Rydgard
51a5adfff2 More buildfixes 2015-09-23 10:59:12 +02:00
Henrik Rydgard
58d3137b10 Get rid of a bunch of more uses of stat64 on Windows. 2015-09-23 09:48:55 +02:00
Henrik Rydgard
0084b2ac30 Simplify FileUtil::Exists. Minor cleanups.
Might possibly help #7967 (XP) by avoiding _wstat64? That would be a libc bug though.
2015-09-22 22:50:29 +02:00
Henrik Rydgard
351baa101c Some Android warning fixes 2015-09-22 18:30:57 +02:00
Henrik Rydgard
60bbe16c89 Make LogManager string handling code safer. 2015-09-19 15:55:49 +02:00
Henrik Rydgard
ecbef7a2ff Remove easy_file.cpp/h. Minor cleanups. 2015-09-19 14:29:02 +02:00
Henrik Rydgård
9636e4cef4 Fix assorted key mapping issues affecting Shield and probably pads in general on Android. See #7976 2015-09-17 22:46:59 +02:00
Henrik Rydgard
3232e46c36 Make debug build nearly warning free in MSVC 2015 2015-09-17 22:02:15 +02:00
Henrik Rydgård
a71fbccfbc Warning fixes for new LLVM on mac (warns on inconsistent usage of override) 2015-09-17 20:29:37 +02:00
sergiobenrocha2
5ec87f3459 Improve detection of for shared libs in both sdl and Qt (zlib, libzip, snappy and glew). 2015-09-16 00:59:31 -03:00
Unknown W. Brackets
0c1a5324fb Merge pull request #7956 from unknownbrackets/arm64-minor
Unknown's min/max unordered-float fix for ARM64
2015-09-12 11:59:05 -07:00
Henrik Rydgard
62b348452d Add ARM64 emitter bugfix by HdkR. 2015-09-10 18:52:31 +02:00
Unknown W. Brackets
1fba38fd41 arm64: Add TST alias to the emitter. 2015-09-08 20:06:53 -07:00
Unknown W. Brackets
85533dad6e Native merge part 2: update build files.
And a couple minor stragglers.
2015-09-06 13:06:57 -07:00
Henrik Rydgard
fdea85138c Update project files, fix minor issues for VS 2015 2015-09-06 19:32:15 +02:00
Henrik Rydgard
1c67de0c21 We don't need a third set of sized integer typedefs, get rid of it. 2015-09-06 19:10:08 +02:00
Henrik Rydgard
86a02fed61 Minor cleanup 2015-09-05 19:42:58 +02:00
Jonas Höglund
a78d71029a Use DEVICE_ID_ANY for the NK_BACK key in default mappings 2015-09-02 10:23:18 +02:00
Jonas Höglund
06fadd5cad Set DPad keys in native for UI navigation
Complements the update to the key repeat logic in native.  Also renames
UpdateConfirmCancelKeys to UpdateNativeMenuKeys since even before this commit,
the function also set the TabLeft and TabRight keys.
2015-09-02 10:23:18 +02:00
Jonas Höglund
5bb2f20493 Use KeyDef for Confirm/Cancel/TabLeft/TabRight keys 2015-09-02 10:23:18 +02:00
Jonas Höglund
0a6ab528ea Move KeyDef and AxisPos to native 2015-09-02 10:23:18 +02:00
aroulin
8a09dedf94 x64Emitter: add RCPPS and RCPSS SSE instructions 2015-08-23 16:43:07 +02:00
Lioncash
959e4e70c0 Arm64Emitter: Fix FCMEQ instruction encoding 2015-08-10 21:12:08 -04:00
Ryan Houdek
f67abeeb2e Arm64Emitter: Fix FCMGE instruction encoding 2015-08-10 21:10:31 -04:00
Henrik Rydgård
0763dbc641 Cleanup and comments in sceGe.cpp 2015-07-24 19:52:42 +02:00
Henrik Rydgard
e69a14ffff Don't set ARM "FastMode". Not worth the compat impact as is, could be made smarter. Should help issue #7234 2015-07-19 18:22:22 +02:00
Henrik Rydgard
f42f81a4da ARM64: Optimize small adjustments of pointerified registers 2015-07-11 16:59:11 +02:00
Henrik Rydgard
28456bef90 ARM64: Add NEON EOR encoding 2015-07-11 12:36:44 +02:00
Henrik Rydgard
72651835c3 A bunch more ARM64 encodings (*MIN,*MAX,XTN,SHRN, etc) 2015-07-11 12:16:56 +02:00
Henrik Rydgard
0be2b93ceb More ARM32 NEON encoding horrors (VMOVL, VSHLL) 2015-07-10 23:56:42 +02:00
Henrik Rydgard
59164cdbfb ARM32: Disasm and emit NEON shifts-by-immediate 2015-07-09 00:24:57 +02:00
Unknown W. Brackets
bf14ad4667 Add a shortcut key for the devmenu.
This is useful when you don't have touch or mouse, only keyboard/pad.
2015-07-05 16:09:29 -07:00
Unknown W. Brackets
2c6278501a arm64: Correct emitter for LDP/STP. 2015-07-04 07:18:58 -07:00
Unknown W. Brackets
7a7d8b6489 Buildfix for Windows.
Oops, forgot to recompile this file in Visual Studio after update.
2015-07-04 06:57:12 -07:00
Unknown W. Brackets
75c8ccd2c1 arm64: Update emitter from Dolphin.
From dolphin-emu/dolphin@d09d59007a.
2015-07-04 00:00:28 -07:00
Unknown W. Brackets
3780b6c0a5 arm64: Verify offset is aligned. 2015-07-02 20:31:36 -07:00
Unknown W. Brackets
31a5227076 arm64: Add umull alias. 2015-07-02 20:31:35 -07:00
Unknown W. Brackets
0a5b1c030b arm64: Implement ext and ins. 2015-06-28 08:45:17 -07:00
Unknown W. Brackets
f355600c20 arm64: Handle negative MOVI2F more correctly.
Both -0.0 and some negated constant values were handled wrong.
2015-06-27 20:13:41 -07:00
Unknown W. Brackets
3110d6f928 Show some cpu name if possible on Android. 2015-06-26 21:55:51 -07:00
Unknown W. Brackets
b3aa6d89e9 Fix UBFX encoding (thanks SonicAdvance1.) 2015-06-26 21:27:03 -07:00
Henrik Rydgard
2c05334d47 ARM64: Fix bug where we didn't save the FP registers correctly in the vertex decoder.
Also port a few ops from dolphin's ARM64 emitter.
2015-06-14 12:56:44 +02:00
Henrik Rydgård
2a1f6bca97 Fix NEON 4444 CheckAlpha, workaround for failed bNEON detection on ARM64 2015-05-31 12:29:08 +02:00
Jan Beich
d0c5c1e659 Common: also use sysctl(3) code on GNU/kFreeBSD 2015-05-26 14:11:23 +00:00
Jan Beich
9887244759 Common: implement GetExeDirectory on some BSDs 2015-05-26 15:20:12 +03:00
Jan Beich
2eb1d578dd Common: check against MAP_FAILED on every Unix after mmap()
This is required by POSIX but may matter only in MAX_FIXED case.
2015-05-26 15:20:11 +03:00
Jan Beich
b24b12f4a0 Common: DragonFly needs MAP_NOSYNC as well 2015-05-26 15:20:11 +03:00
Jan Beich
61c106c57f Common: avoid redefining bswap* via <sys/types.h> on BSDs
In file included from Common/CPUDetect.cpp:27:
In file included from Common/Common.h:139:
Common/Swap.h:76:23: error:
      C++ requires a type specifier for all declarations
inline unsigned short bswap16(unsigned short x) { return (x << 8) | (x >> 8); }
                      ^
/usr/include/sys/endian.h:59:20: note: expanded from macro 'bswap16'
                        ^
/usr/include/x86/endian.h:74:16: note: expanded from macro '__bswap16'
        ((__uint16_t)(__builtin_constant_p(x) ? \
                      ^
In file included from Common/CPUDetect.cpp:27:
In file included from Common/Common.h:139:
Common/Swap.h:76:23: error:
      expected ')'
/usr/include/sys/endian.h:59:20: note: expanded from macro 'bswap16'
                        ^
/usr/include/x86/endian.h:74:40: note: expanded from macro '__bswap16'
        ((__uint16_t)(__builtin_constant_p(x) ? \
                                              ^
Common/Swap.h:76:23: note:
      to match this '('
/usr/include/sys/endian.h:59:20: note: expanded from macro 'bswap16'
                        ^
/usr/include/x86/endian.h:74:15: note: expanded from macro '__bswap16'
        ((__uint16_t)(__builtin_constant_p(x) ? \
                     ^
In file included from Common/CPUDetect.cpp:27:
In file included from Common/Common.h:139:
Common/Swap.h:76:59: error:
      use of undeclared identifier 'x'
inline unsigned short bswap16(unsigned short x) { return (x << 8) | (x >> 8); }
                                                          ^
Common/Swap.h:76:70: error:
      use of undeclared identifier 'x'
inline unsigned short bswap16(unsigned short x) { return (x << 8) | (x >> 8); }
                                                                     ^
Common/Swap.h:77:21: error:
      expected ')'
inline unsigned int bswap32(unsigned int x) { return (x >> 24) | ((x & 0xFF0...
                    ^
/usr/include/sys/endian.h:60:20: note: expanded from macro 'bswap32'
                        ^
/usr/include/x86/endian.h:77:27: note: expanded from macro '__bswap32'
        (__builtin_constant_p(x) ?      \
                                 ^
Common/Swap.h:77:21: note:
      to match this '('
/usr/include/sys/endian.h:60:20: note: expanded from macro 'bswap32'
                        ^
/usr/include/x86/endian.h:77:2: note: expanded from macro '__bswap32'
        (__builtin_constant_p(x) ?      \
        ^
In file included from Common/CPUDetect.cpp:27:
In file included from Common/Common.h:139:
Common/Swap.h:77:21: error:
      functions that differ only in their return type cannot be overloaded
inline unsigned int bswap32(unsigned int x) { return (x >> 24) | ((x & 0xFF0...
                    ^
/usr/include/sys/endian.h:60:20: note: expanded from macro 'bswap32'
                        ^
/usr/include/x86/endian.h:77:3: note: expanded from macro '__bswap32'
        (__builtin_constant_p(x) ?      \
         ^
native/base/basictypes.h:92:44: note:
      previous implicit declaration is here
inline uint16 swap16(uint16 _data) {return bswap16(_data);}
                                           ^
/usr/include/sys/endian.h:59:20: note: expanded from macro 'bswap16'
                        ^
/usr/include/x86/endian.h:74:16: note: expanded from macro '__bswap16'
        ((__uint16_t)(__builtin_constant_p(x) ? \
                      ^
In file included from Common/CPUDetect.cpp:27:
In file included from Common/Common.h:139:
Common/Swap.h:78:27: error:
      expected ')'
inline unsigned long long bswap64(unsigned long long x) { return ((unsigned ...
                          ^
/usr/include/sys/endian.h:61:20: note: expanded from macro 'bswap64'
                        ^
/usr/include/x86/endian.h:80:27: note: expanded from macro '__bswap64'
        (__builtin_constant_p(x) ?      \
                                 ^
Common/Swap.h:78:27: note:
      to match this '('
/usr/include/sys/endian.h:61:20: note: expanded from macro 'bswap64'
                        ^
/usr/include/x86/endian.h:80:2: note: expanded from macro '__bswap64'
        (__builtin_constant_p(x) ?      \
        ^
In file included from Common/CPUDetect.cpp:27:
In file included from Common/Common.h:139:
Common/Swap.h:78:27: error:
      functions that differ only in their return type cannot be overloaded
inline unsigned long long bswap64(unsigned long long x) { return ((unsigned ...
                          ^
/usr/include/sys/endian.h:61:20: note: expanded from macro 'bswap64'
                        ^
/usr/include/x86/endian.h:80:3: note: expanded from macro '__bswap64'
        (__builtin_constant_p(x) ?      \
         ^
native/base/basictypes.h:92:44: note:
      previous implicit declaration is here
inline uint16 swap16(uint16 _data) {return bswap16(_data);}
                                           ^
/usr/include/sys/endian.h:59:20: note: expanded from macro 'bswap16'
                        ^
/usr/include/x86/endian.h:74:16: note: expanded from macro '__bswap16'
        ((__uint16_t)(__builtin_constant_p(x) ? \
                      ^
8 errors generated.
2015-05-26 15:20:11 +03:00
Jan Beich
05e5085f68 Common: drop bogus #ifdef for FreeBSD
Common/CPUDetect.cpp:70:2: error: use of undeclared identifier '__cpuidex'; did you mean 'do_cpuidex'?
        __cpuidex((int *)regs, cpuid_leaf, ecxval);
        ^~~~~~~~~
        do_cpuidex
Common/CPUDetect.cpp:69:6: note: 'do_cpuidex' declared here
void do_cpuidex(u32 regs[4], u32 cpuid_leaf, u32 ecxval) {
     ^
Common/CPUDetect.cpp:70:12: error: cannot initialize a parameter of type 'u32 *' (aka 'unsigned int *') with an rvalue of type 'int *'
        __cpuidex((int *)regs, cpuid_leaf, ecxval);
                  ^~~~~~~~~~~
Common/CPUDetect.cpp:69:21: note: passing argument to parameter 'regs' here
void do_cpuidex(u32 regs[4], u32 cpuid_leaf, u32 ecxval) {
                    ^
Common/CPUDetect.cpp:73:2: error: use of undeclared identifier '__cpuid'; did you mean 'do_cpuid'?
        __cpuid((int *)regs, cpuid_leaf);
        ^~~~~~~
        do_cpuid
Common/CPUDetect.cpp:72:6: note: 'do_cpuid' declared here
void do_cpuid(u32 regs[4], u32 cpuid_leaf) {
     ^
Common/CPUDetect.cpp:73:10: error: cannot initialize a parameter of type 'u32 *' (aka 'unsigned int *') with an rvalue of type 'int *'
        __cpuid((int *)regs, cpuid_leaf);
                ^~~~~~~~~~~
Common/CPUDetect.cpp:72:19: note: passing argument to parameter 'regs' here
void do_cpuid(u32 regs[4], u32 cpuid_leaf) {
                  ^
4 errors generated.
2015-05-26 15:20:11 +03:00
Jan Beich
59a21dbfcb Common: try to unbreak StdMutex.h on more C++11 compilers
Clang still pretends as GCC 4.2.1 even with libc++ or newer libstdc++.

In file included from Common/ConsoleListener.cpp:33:
In file included from Common/ConsoleListener.h:20:
In file included from Common/LogManager.h:26:
Common/StdMutex.h:197:28: error:
      reference to 'adopt_lock_t' is ambiguous
        lock_guard(mutex_type& m, adopt_lock_t)
                                  ^
Common/StdMutex.h:183:6: note:
      candidate found by name lookup is 'std::adopt_lock_t'
enum adopt_lock_t { adopt_lock };
     ^
/usr/include/c++/v1/__mutex_base:55:25: note: candidate found by name lookup is
      'std::__1::adopt_lock_t'
struct _LIBCPP_TYPE_VIS adopt_lock_t {};
                        ^
In file included from Common/ConsoleListener.cpp:33:
In file included from Common/ConsoleListener.h:20:
In file included from Common/LogManager.h:26:
Common/StdMutex.h:230:29: error:
      reference to 'defer_lock_t' is ambiguous
        unique_lock(mutex_type& m, defer_lock_t)
                                   ^
Common/StdMutex.h:181:6: note:
      candidate found by name lookup is 'std::defer_lock_t'
enum defer_lock_t { defer_lock };
     ^
/usr/include/c++/v1/__mutex_base:53:25: note: candidate found by name lookup is
      'std::__1::defer_lock_t'
struct _LIBCPP_TYPE_VIS defer_lock_t {};
                        ^
In file included from Common/ConsoleListener.cpp:33:
In file included from Common/ConsoleListener.h:20:
In file included from Common/LogManager.h:26:
Common/StdMutex.h:234:29: error:
      reference to 'try_to_lock_t' is ambiguous
        unique_lock(mutex_type& m, try_to_lock_t)
                                   ^
Common/StdMutex.h:182:6: note:
      candidate found by name lookup is 'std::try_to_lock_t'
enum try_to_lock_t { try_to_lock };
     ^
/usr/include/c++/v1/__mutex_base:54:25: note: candidate found by name lookup is
      'std::__1::try_to_lock_t'
struct _LIBCPP_TYPE_VIS try_to_lock_t {};
                        ^
In file included from Common/ConsoleListener.cpp:33:
In file included from Common/ConsoleListener.h:20:
In file included from Common/LogManager.h:26:
Common/StdMutex.h:238:29: error:
      reference to 'adopt_lock_t' is ambiguous
        unique_lock(mutex_type& m, adopt_lock_t)
                                   ^
Common/StdMutex.h:183:6: note:
      candidate found by name lookup is 'std::adopt_lock_t'
enum adopt_lock_t { adopt_lock };
     ^
/usr/include/c++/v1/__mutex_base:55:25: note: candidate found by name lookup is
      'std::__1::adopt_lock_t'
struct _LIBCPP_TYPE_VIS adopt_lock_t {};
                        ^
In file included from Common/ConsoleListener.cpp:33:
In file included from Common/ConsoleListener.h:20:
In file included from Common/LogManager.h:26:
Common/StdMutex.h:345:6: warning:
      variable templates are a C++14 extension [-Wc++14-extensions]
void swap(unique_lock<Mutex>& x, unique_lock<Mutex>& y)
     ^
Common/StdMutex.h:345:6: error:
      variable has incomplete type 'void'
Common/StdMutex.h:345:11: error:
      reference to 'unique_lock' is ambiguous
void swap(unique_lock<Mutex>& x, unique_lock<Mutex>& y)
          ^
Common/StdMutex.h:215:7: note:
      candidate found by name lookup is 'std::unique_lock'
class unique_lock
      ^
/usr/include/c++/v1/__mutex_base:96:29: note: candidate found by name lookup is
      'std::__1::unique_lock'
class _LIBCPP_TYPE_VIS_ONLY unique_lock
                            ^
In file included from Common/ConsoleListener.cpp:33:
In file included from Common/ConsoleListener.h:20:
In file included from Common/LogManager.h:26:
Common/StdMutex.h:345:23: error:
      'Mutex' does not refer to a value
void swap(unique_lock<Mutex>& x, unique_lock<Mutex>& y)
                      ^
Common/StdMutex.h:344:17: note:
      declared here
template <class Mutex>
                ^
Common/StdMutex.h:345:31: error:
      use of undeclared identifier 'x'
void swap(unique_lock<Mutex>& x, unique_lock<Mutex>& y)
                              ^
Common/StdMutex.h:345:34: error:
      reference to 'unique_lock' is ambiguous
void swap(unique_lock<Mutex>& x, unique_lock<Mutex>& y)
                                 ^
Common/StdMutex.h:215:7: note:
      candidate found by name lookup is 'std::unique_lock'
class unique_lock
      ^
/usr/include/c++/v1/__mutex_base:96:29: note: candidate found by name lookup is
      'std::__1::unique_lock'
class _LIBCPP_TYPE_VIS_ONLY unique_lock
                            ^
In file included from Common/ConsoleListener.cpp:33:
In file included from Common/ConsoleListener.h:20:
In file included from Common/LogManager.h:26:
Common/StdMutex.h:345:46: error:
      'Mutex' does not refer to a value
void swap(unique_lock<Mutex>& x, unique_lock<Mutex>& y)
                                             ^
Common/StdMutex.h:344:17: note:
      declared here
template <class Mutex>
                ^
Common/StdMutex.h:345:54: error:
      use of undeclared identifier 'y'
void swap(unique_lock<Mutex>& x, unique_lock<Mutex>& y)
                                                     ^
Common/StdMutex.h:345:56: error:
      expected ';' at end of declaration
void swap(unique_lock<Mutex>& x, unique_lock<Mutex>& y)
                                                       ^
                                                       ;
Common/StdMutex.h:346:1: error:
      expected unqualified-id
{
^
In file included from Common/ConsoleListener.cpp:33:
In file included from Common/ConsoleListener.h:20:
Common/LogManager.h:54:7: error:
      reference to 'mutex' is ambiguous
        std::mutex m_log_lock;
             ^
/usr/include/c++/v1/__mutex_base:27:24: note: candidate found by name lookup is
      'std::__1::mutex'
class _LIBCPP_TYPE_VIS mutex
                       ^
Common/StdMutex.h:107:7: note:
      candidate found by name lookup is 'std::mutex'
class mutex
      ^
In file included from Common/ConsoleListener.cpp:33:
In file included from Common/ConsoleListener.h:20:
Common/LogManager.h:116:7: error:
      reference to 'mutex' is ambiguous
        std::mutex m_listeners_lock;
             ^
/usr/include/c++/v1/__mutex_base:27:24: note: candidate found by name lookup is
      'std::__1::mutex'
class _LIBCPP_TYPE_VIS mutex
                       ^
Common/StdMutex.h:107:7: note:
      candidate found by name lookup is 'std::mutex'
class mutex
      ^
In file included from Common/ConsoleListener.cpp:33:
In file included from Common/ConsoleListener.h:20:
Common/LogManager.h:131:7: error:
      reference to 'mutex' is ambiguous
        std::mutex log_lock_;
             ^
/usr/include/c++/v1/__mutex_base:27:24: note: candidate found by name lookup is
      'std::__1::mutex'
class _LIBCPP_TYPE_VIS mutex
                       ^
Common/StdMutex.h:107:7: note:
      candidate found by name lookup is 'std::mutex'
class mutex
      ^
1 warning and 16 errors generated.
2015-05-26 15:20:11 +03:00
Jan Beich
5f278b2d94 Unbreak build with SSE 4.1 (clang -march=penryn)
Common/ColorConv.cpp:199:33: error: use of undeclared identifier '_mm_packus_epi32';
      did you mean '_mm_packs_epi32'?
                _mm_store_si128(&dstp[i / 2], _mm_packus_epi32(c1, c2));
                                              ^~~~~~~~~~~~~~~~
                                              _mm_packs_epi32
/usr/bin/../lib/clang/3.6.0/include/emmintrin.h:1254:1: note: '_mm_packs_epi32'
      declared here
_mm_packs_epi32(__m128i __a, __m128i __b)
^
Common/ColorConv.cpp:241:33: error: use of undeclared identifier '_mm_packus_epi32';
      did you mean '_mm_packs_epi32'?
                _mm_store_si128(&dstp[i / 2], _mm_packus_epi32(c1, c2));
                                              ^~~~~~~~~~~~~~~~
                                              _mm_packs_epi32
/usr/bin/../lib/clang/3.6.0/include/emmintrin.h:1254:1: note: '_mm_packs_epi32'
      declared here
_mm_packs_epi32(__m128i __a, __m128i __b)
^
2 errors generated.

GPU/Common/TextureScalerCommon.cpp:294:6: error: unknown type name '__m128'
                                        __m128 result = _mm_set1_ps(0.0f);
                                        ^
GPU/Common/TextureScalerCommon.cpp:294:22: error: use of undeclared identifier
      '_mm_set1_ps'
                                        __m128 result = _mm_set1_ps(0.0f);
                                                        ^
GPU/Common/TextureScalerCommon.cpp:305:9: error: unknown type name '__m128i'
                                                                __m128i samp...
                                                                ^
GPU/Common/TextureScalerCommon.cpp:305:26: error: use of undeclared identifier
      '_mm_cvtsi32_si128'
  ...__m128i sample = _mm_cvtsi32_si128(data[csy*w + csx]);
                      ^
GPU/Common/TextureScalerCommon.cpp:307:9: error: unknown type name '__m128'
                                                                __m128 col ...
                                                                ^
GPU/Common/TextureScalerCommon.cpp:308:31: error: use of undeclared identifier
      '_mm_set1_ps'
  ...col = _mm_mul_ps(col, _mm_set1_ps(weight));
                           ^
GPU/Common/TextureScalerCommon.cpp:314:6: error: unknown type name '__m128i'
                                        __m128i pixel = _mm_cvtps_epi32(_mm_...
                                        ^
GPU/Common/TextureScalerCommon.cpp:314:57: error: use of undeclared identifier
      '_mm_set1_ps'
  ...__m128i pixel = _mm_cvtps_epi32(_mm_mul_ps(result, _mm_set1_ps(bicubicInvSum...
                                                        ^
8 errors generated.
2015-05-26 15:20:10 +03:00
Unknown W. Brackets
4110a24d8f Fix Direct3D9 framebuffer upload conversion funcs.
All of the 16 bit formats were, afaict, wrong in different ways.
2015-05-25 10:47:37 -07:00
Unknown W. Brackets
539a7b1772 Correct order in ConvertRGBA4444ToRGBA8888().
I assume this is right, since it's pulling in a 4444 tex from PSP VRAM.
2015-05-25 10:32:23 -07:00
Unknown W. Brackets
b0a825ea9f Small optimization to ConvertRGBA5551ToRGBA8888(). 2015-05-25 10:31:27 -07:00
Unknown W. Brackets
3b96bf8712 Oops, typo. 2015-05-23 12:20:00 -07:00
Unknown W. Brackets
4900bc9082 Add a NEON 4444 flip color conversion func. 2015-05-23 11:30:07 -07:00
Unknown W. Brackets
df53bc2e7b Add a NEON 565 flip colorconv function. 2015-05-23 11:30:06 -07:00
Unknown W. Brackets
bbe15d2518 Add a NEON color conv for 5551 -> 1555. 2015-05-23 11:30:06 -07:00
Unknown W. Brackets
ccdaf41b53 Add some additional SSE for color conversion. 2015-05-17 16:04:42 -07:00
Unknown W. Brackets
1cb2302f5c Use SSE in ConvertRGBA5551ToRGBA8888().
This func is used when showing videos, some loading screens, etc.  Also
when uploading graphics to VRAM.

The implementation takes 85% less than without SSE, although it could
probably be optimized further.  Won't help game runtime much, but may help
some loading times or special effects.
2015-05-17 16:04:42 -07:00
Unknown W. Brackets
7269574a10 Add structure for NEON color conversion funcs. 2015-05-17 16:04:41 -07:00
Unknown W. Brackets
1767bd958c Move color conversion funcs to ColorConv.
This paves the way a bit for NEON conversion funcs.
2015-05-17 13:45:30 -07:00
Unknown W. Brackets
3f29329ed2 Actually use the SSE color conversion code. 2015-05-17 13:42:17 -07:00
Unknown W. Brackets
c10373ef06 ARM64: Fix some minor MSVC warnings. 2015-04-10 12:05:51 -07:00
Henrik Rydgard
db110d60a1 Rename some colorconv functions 2015-04-08 23:01:17 +02:00
Henrik Rydgard
4621586342 Move simple conversion functions out of the ColorConv header. 2015-04-08 22:52:49 +02:00
Henrik Rydgard
6660985210 Extract a bunch more color conversion loops. 2015-04-08 22:52:48 +02:00
Henrik Rydgard
b5acf15cc9 Extract some explicit loops into calls to color conversion functions. 2015-04-08 22:52:48 +02:00
Henrik Rydgard
c2cf2bd97e Move some more color conversion functions to ColorConv 2015-04-08 22:52:47 +02:00
Henrik Rydgard
31961eace9 Color conv attempt 2. Start consolidating simple color conversion functions in a single header+cpp. 2015-04-08 22:52:46 +02:00
Henrik Rydgård
39be916d8a ARM64: More emitter/disasm 2015-04-06 18:13:44 +02:00
Henrik Rydgård
459ba28655 ARM64: SW skinning runs without crashing but is broken. 2015-04-06 18:13:44 +02:00
Henrik Rydgård
597595f279 ARM64: Start implementing soft-skinning. Disabled for now, needs work. 2015-04-06 18:13:43 +02:00
Henrik Rydgård
f82b613371 ARM64 emitter/disasm: Implement ASIMD LDP/STP instructions 2015-04-06 18:13:42 +02:00
Henrik Rydgård
013bbc71af ARM64: Fix DUP disasm, INS disasm 2015-04-06 18:13:40 +02:00
Henrik Rydgård
a3db3ed5c1 ARM64 emitter: Fix UXTL/SXTL and friends. Add disasm. 2015-04-06 18:13:40 +02:00
Henrik Rydgård
853d6ea34b ARM64 emitter: Add FMLA/FMLS vector versions 2015-04-06 18:13:39 +02:00
Henrik Rydgård
065cd97dc2 ARM64: LLVM buildfix, some more disasm 2015-04-06 18:13:39 +02:00
Henrik Rydgård
13c9390c53 ARM64: Emitter fix, disable swl/swr/lwl/lwr again fully 2015-04-06 18:13:38 +02:00
Henrik Rydgård
9f24076b63 Make the unittests build on mac 2015-04-06 18:13:37 +02:00
Henrik Rydgard
0a70618f87 ARM64: Accurate floating point rounding. For some reason, FTZ doesn't seem to work though. 2015-04-06 18:13:36 +02:00
Henrik Rydgard
4618275f99 ARM64: Add a few aliases to emitter. Disasm fixes. 2015-04-06 18:13:33 +02:00
Henrik Rydgard
acf08eefa8 ARM64: Fix FCVTL, use it in v2hf 2015-04-06 18:13:33 +02:00
Henrik Rydgard
0849e270ee ARM64: fmla encoding, more disasm 2015-04-06 18:13:31 +02:00
Henrik Rydgard
ca58f322e5 ARM64: Port over some missing VFPU instructions from ARM. Not much left now. 2015-04-06 18:13:30 +02:00
Henrik Rydgard
9c79831237 ARM64 emitter: Fix FP immediates 2015-04-06 18:13:29 +02:00
Henrik Rydgard
9a5a093105 ARM64 emitter/disasm: More scalar FPU instructions 2015-04-06 18:13:27 +02:00
Henrik Rydgard
ff758f58ad ARM64: The Return of the Pointer Truncator - The Movie 2015-04-06 18:13:26 +02:00
Henrik Rydgard
6cb107d6fc ARM64: Fix LDP disassembly 2015-04-06 18:13:25 +02:00
Henrik Rydgard
5a5f3c94fd ARM64: Guess what? more emitter & disasm 2015-04-06 18:13:24 +02:00
Henrik Rydgard
0f430e6cd1 ARM64: More asimd scalar disasm, encoding fix 2015-04-06 18:13:23 +02:00
Henrik Rydgard
9f689b871a ARM64: Work on emitting and disassembling fp conversion ops 2015-04-06 18:13:22 +02:00
Henrik Rydgard
a8be978154 ARM64 emitter: Add fixed point versions of SCVTF and UCVTF 2015-04-06 18:13:19 +02:00
Henrik Rydgard
d5faf8b97a ARM64: Fix conditional branch instruction (!) 2015-04-06 18:13:18 +02:00
Henrik Rydgard
acd9502b44 ARM64: stp/ldp disasm improvements 2015-04-06 18:13:17 +02:00
Henrik Rydgard
5dff3f8c89 ARM64: Implement scalar FMOV. This makes the FPU2op ops work. 2015-04-06 18:13:16 +02:00
Henrik Rydgard
4233921ab7 ARM64: Some more instructions, func replacements 2015-04-06 18:13:16 +02:00
Henrik Rydgard
2bca05c4f2 ARM64: implement shifts, movz/movn. Corresponding fixes to emitter/disasm 2015-04-06 18:13:14 +02:00
Henrik Rydgard
70f44c3894 ARM64 emitter: Default to non-extend instruction forms (LSL 0), fix bugs. also fixes to disasm. 2015-04-06 18:13:12 +02:00
Henrik Rydgard
d3669daba4 ARM64: Fixes to emitter and disassembly for logical immediates 2015-04-06 18:13:12 +02:00
Henrik Rydgard
0922db6062 ARM64: Some FP work. 2015-04-06 18:13:11 +02:00
Henrik Rydgard
2ff608b72e ARM64: Add FDIV to emitter 2015-04-06 18:13:10 +02:00
Henrik Rydgard
742f48ad56 ARM64: More emitter and disasm work 2015-04-06 18:13:08 +02:00
Henrik Rydgard
c4cdca83fc ARM64: Fix some branch distance checks, add more stuff to disasm 2015-04-06 18:13:08 +02:00
Henrik Rydgard
73dd26fb75 More ARM64 disasm, warning fixes 2015-04-06 18:13:07 +02:00
Henrik Rydgard
80a813bcbd ARM64 Emitter: Optimize MOVI2R by always starting with a MOVZ, killing the register dependency in the out-of-order engine of the CPU. 2015-04-06 18:13:06 +02:00
Henrik Rydgard
58b059ca14 Some casting cleanup, misc 2015-04-06 18:13:06 +02:00
Henrik Rydgard
feec574ece Arm64Emitter: Add some helper functions to add / sub by immediate 2015-04-06 18:13:05 +02:00
Henrik Rydgard
f732fbb885 ARM64 emitter: Add MRS/MSR support (only for the flags register). Add some more to ARM64 Disasm 2015-04-06 18:13:04 +02:00
Henrik Rydgard
524583d53e ARM64 emitter work, dispatcher work, etc. More work needed. 2015-04-06 18:13:03 +02:00
Henrik Rydgard
3aebc06329 Better logic immediate support in ARM emitter. From V8. 2015-04-06 18:13:02 +02:00
Henrik Rydgard
b309c83973 Initial work on ARM64, based on the ARM jit. 2015-04-06 18:13:01 +02:00
Unknown W. Brackets
41631eb2b1 Try a different format of HLE logging. 2015-03-22 20:49:42 -07:00
Henrik Rydgard
c3f38f31f9 Update the (currently unused) ARM64 emitter from Dolphin's current one (by HdkR). 2015-03-06 00:46:34 +01:00
Henrik Rydgard
a8a7adc85f When using controller autoconf on PC, throw in a couple of useful keyboard mappings too.
See #7399
2015-02-04 11:52:13 +01:00
Unknown W. Brackets
8484af23ef Avoid min/max issues from Atomics.h. 2015-02-01 11:55:02 -08:00
Henrik Rydgard
55c5215881 Add error checking when getting the modification time of a file 2015-02-01 19:54:07 +01:00
Henrik Rydgard
4ebdf49fb6 Update ffmpeg to master, also switching to static linking of MSVCRT. Don't forget to update your submodules! 2015-01-24 13:30:37 +01:00
Henrik Rydgard
cae58cafee Revert "Merge pull request #7361 from hrydgard/color-conv-centralize"
This reverts commit f1b57dabf5, reversing
changes made to 41001637ce.
2015-01-22 19:53:32 +01:00
Henrik Rydgard
07933cad42 Revert "Minor optimizations, add a failsafe"
This reverts commit 48e4d1edae.
2015-01-22 19:52:58 +01:00
Henrik Rydgard
9c64351578 Revert "Fix issues with scaling ARGB1555 textures in OpenGL, introduced by the recent merge"
This reverts commit 08eaa6e1f7.
2015-01-22 19:52:49 +01:00
Henrik Rydgard
08eaa6e1f7 Fix issues with scaling ARGB1555 textures in OpenGL, introduced by the recent merge 2015-01-20 22:48:39 +01:00
Henrik Rydgard
48e4d1edae Minor optimizations, add a failsafe 2015-01-20 20:08:12 +01:00
Henrik Rydgard
7faddd6100 Consolidate the two TextureScaler copies into one.
Also lets us clean up ColorConv slightly.
2015-01-20 17:32:55 +01:00
Henrik Rydgard
4d1c1e41f3 Optimize and consolidate color conversion a bit 2015-01-20 17:32:55 +01:00
Henrik Rydgard
2623a48b4a Centralize color conversion functions in Common/ColorConv. 2015-01-20 17:32:54 +01:00
Unknown W. Brackets
df3f90d763 x86jit: Fix a copy/paste error. 2015-01-18 13:26:14 -08:00
Unknown W. Brackets
116ddaaaf4 Fix a bad nullcheck. 2015-01-17 19:49:47 -08:00
Unknown W. Brackets
de8ffc18b3 Match up two arrays for error messages. 2015-01-17 18:43:04 -08:00
Unknown W. Brackets
cf0e337b27 Fix a memleak on an error condition. 2015-01-17 18:21:03 -08:00
Unknown W. Brackets
f3654795a3 Minor warning fixes. 2015-01-11 14:23:42 -08:00
Henrik Rydgard
26e4cb4a21 More cleanup 2015-01-11 20:35:52 +01:00
Henrik Rydgard
2b50996a85 Symbian buildfix 2015-01-11 18:16:15 +01:00
Henrik Rydgard
e508004a4b Update Atomic*.h 2015-01-11 18:06:12 +01:00
Henrik Rydgard
cac7a2feed Add a sample rate parameter to DSound, clean it up a bit 2015-01-11 14:11:09 +01:00
Henrik Rydgard
604abe933e Update submodules, add x64Emitter bugfix from Dolphin (plus a few new instrs), misc 2015-01-11 00:12:32 +01:00
Henrik Rydgard
a9b010b0cd Add a basic debug console UI to the devmenu that shows some log scrollback and has a little command line.
Useful for debugging on non-Windows in the future. For now non-mobile only (although may lift that restriction later)
2015-01-05 01:25:23 +01:00
Unknown W. Brackets
1ec19310cd Fix a Symbian compile issue. 2014-12-28 20:54:29 -08:00
Unknown W. Brackets
a581720a59 Correct File::Rename() with Unicode paths.
Fixes savestates in e.g. Chinese folder paths, #7235.
2014-12-28 20:05:07 -08:00
Henrik Rydgård
6bf2c02908 x86 jit: Allow storing all imms directly without bouncing to a register, not just zero. 2014-12-23 22:25:53 +01:00
Henrik Rydgard
6c3ef819e0 ARM cpu detect: Only drop last character if it's actually a newline 2014-12-22 00:19:59 +01:00
Henrik Rydgard
c408928977 Arm64Emitter: Some warning fixes. Update native. 2014-12-19 00:34:41 +01:00
Gabriel
df3b407427 Detect azerty and qwertz
Addresses #2462.
2014-12-19 00:07:20 +01:00
Henrik Rydgård
cd573cf731 Merge pull request #7063 from Pulfer/axis-swap
Add a hotkey to swap D-pad and left analog stick keys
2014-12-18 23:15:28 +01:00
Henrik Rydgård
0c12cbab42 Merge pull request #7094 from chinhodado/patch-8
Change to pass by reference
2014-12-18 22:38:31 +01:00
Henrik Rydgård
a9be90229a Merge pull request #7140 from AbandonedCart/master
use libc++ for iOS to avoid additional directives
2014-12-18 22:38:14 +01:00
Henrik Rydgård
9c73272604 Merge pull request #7178 from unknownbrackets/mips-emitter
mips: Minor tweaks to the mips emitter
2014-12-18 22:22:36 +01:00
Unknown W. Brackets
f7f664c641 Remove a few stragglers from PPC.
This makes it so the build isn't dirty every time.
2014-12-16 23:00:25 -08:00
Henrik Rydgard
a54e0cf244 Goodbye PowerPC, you can always be resurrected when the time comes 2014-12-16 21:58:38 +01:00
Lioncash
f0f7e48ef4 Common: Move headers over to using pragma once
Just makes the headers more consistent.
2014-12-15 17:09:40 -05:00
Henrik Rydgard
4ec30d98e1 Port the x86 and ARM emitters over to use the generic CodeBlock class 2014-12-15 22:32:55 +01:00
Henrik Rydgard
b2951f0def Transplant Dolphin's ARM64 emitter over. Not hooked up to anything (yet...) 2014-12-15 22:09:26 +01:00
Unknown W. Brackets
231d4a3da2 mips: Improve MOVP2R() typesafety. 2014-12-13 19:38:43 -08:00
Unknown W. Brackets
74ef968503 mips: Prevent compilation of an easy mistake.
It's easy to write LW(T0, T1, T2) but that's not valid.
2014-12-13 19:38:43 -08:00
Andrey Bondrov
35d0149162 Less code for bool variable inversion 2014-12-14 00:29:58 +10:00
Andrey Bondrov
89c6f76456 Less confusing function name 2014-12-13 12:20:31 +10:00
Andrey Bondrov
c41e4ef905 Don't make Axis <-> D-pad swap persistent 2014-12-13 12:13:42 +10:00
Andrey Bondrov
38268a3d37 Use constants instead of hardcoded numbers for keys swap 2014-12-13 10:36:13 +10:00
Unknown W. Brackets
42cd751060 Prevent a few savestate cross-compat issues.
We don't need to save gpu stats at all, so don't, makes it easier to
change it and dunno if anyone defines double as other than 64 bits.

Also, apparently there was padding in the event state but not on all
platforms.
2014-12-08 21:18:56 -08:00
Henrik Rydgård
82d69fad8d Merge pull request #7146 from hrydgard/sha256
Quick implementation of sceSha256Digest from #7134
2014-12-08 10:40:56 +01:00
Henrik Rydgård
e46976f08e Merge pull request #7147 from lioncash/moarconst
Mark some functions as const
2014-12-08 00:25:42 +01:00
Henrik Rydgard
2bce7bc460 X64Emitter: Merge some AVX stuff from Dolphin 2014-12-07 23:09:38 +01:00
Lioncash
a962bc5a6c Mark some functions as const 2014-12-07 17:08:13 -05:00
Henrik Rydgard
78240198dc Oops 2014-12-07 21:31:44 +01:00
Henrik Rydgard
8102a67602 Implement sceSha256Digest (untested) 2014-12-07 19:25:45 +01:00
Henrik Rydgard
29dcc0a303 Minor cleanups, warning fixes 2014-12-06 12:25:28 +01:00
Henrik Rydgard
fb251c9a5c Merge Intel Atom detection from Dolphin's CPU detect 2014-12-06 11:35:01 +01:00
TwistedUmbrella
3f4d58b0b2 use libc++ for iOS to avoid additional directives 2014-12-05 19:47:02 -05:00
Henrik Rydgard
5290ffd929 Minor cleanup in vtfm. Re-enable vrot combination. Optimize vfad/vavg when dpps is available.
Also fixes bug in emitter of dpps.
2014-12-03 22:44:32 +01:00
Unknown W. Brackets
bafa7d0670 Add git-version to more than just Core.
This won't catch native or armips, but it will catch more before and still
won't make the build wrongly dirty all the time.

See #5575.
2014-12-02 22:55:19 -08:00
Henrik Rydgard
66d74981b5 Merge ARM emitter updates from the NEON branch 2014-11-29 10:49:22 +01:00
Henrik Rydgard
344f71b092 x86 jit: Commit commented-out haddps-based vdot.q as reminder not to use haddps... 2014-11-28 00:19:11 +01:00
Henrik Rydgard
5033babb10 x86 Jit: SIMD-ify vdot 2014-11-26 23:47:18 +01:00
Henrik Rydgard
804de50711 x86 jit: SIMD-ify VFPU register file writebacks where possible 2014-11-26 01:33:05 +01:00
Unknown W. Brackets
e2e97a8f4e mips: Add smarter delay slot handling to emitter. 2014-11-20 00:37:38 -08:00
Sacha
0d2c7846f5 MIPS: Add EXT/INS/BLTZ eo emitter. Fix LUI. 2014-11-19 13:31:08 +10:00
Sacha
10f6db27ca Add Mips CPU Detection. 2014-11-19 08:51:15 +10:00
Unknown W. Brackets
44fc7c6c71 mips: MIPSX -> MIPS. 2014-11-18 09:21:46 -08:00
Unknown W. Brackets
4de9b9692e mips: Add JALR/JR. QuickCallFunction should link. 2014-11-18 09:12:16 -08:00
Unknown W. Brackets
5b18353584 mips: Add helpers for constant loads. 2014-11-18 08:41:28 -08:00
Unknown W. Brackets
29365e6775 mips: Clean up MIPSXEmitter a bit.
Don't even want dangerous funcs like SUB().

Keeping things in opcode order at least locally so it doesn't get
confusing to maintain.  Also fixed a bunch of missing asserts.
2014-11-18 08:41:20 -08:00
xSacha
4167b66808 Add some more MIPS instructions 2014-11-18 08:41:09 -08:00
Sacha
65e008f053 Implement a MipsJit. 2014-11-18 08:40:57 -08:00
Unknown W. Brackets
8b6a5f2ecc mips: Duh, oops. LUI is one reg, of course. 2014-11-18 08:40:30 -08:00
Unknown W. Brackets
0df0ea7d85 mips: Add a basic MIPSXCodeBlock. 2014-11-18 08:40:30 -08:00
Unknown W. Brackets
2903e72daa mips: Add basic load/store for the MIPS emitter. 2014-11-18 08:40:30 -08:00
Unknown W. Brackets
5cadab0273 mips: Add some basic jumping and logic instructions. 2014-11-18 08:40:29 -08:00
Unknown W. Brackets
6407e4a32a mips: Initial structure for a MIPS emitter. 2014-11-18 08:40:29 -08:00
Henrik Rydgard
53b5d331b4 Assorted minor optimizations 2014-11-17 21:21:44 +01:00
Henrik Rydgard
28ca8d4818 x86 jit: Use LEA to emulate addu but only when it can save a few bytes 2014-11-16 17:39:47 +01:00
chinhodado
3a15da48ac Change to pass by reference 2014-11-14 22:26:28 -05:00
chinhodado
4bac356df6 Use const 2014-11-14 16:13:06 -05:00
xSacha
57e4088216 Introduce fake vertex decoder JIT as well.
Compiles and links on CI20 but gets unknown crash in GL driver.
2014-11-13 17:10:29 +10:00
Sacha
a0086f6412 Introduce a Fake JIT for generic builds. 2014-11-13 00:09:51 +10:00
Henrik Rydgard
9e0aa8eaa3 Drop silly way of doing CPUID on Android, the right way seems to work 2014-11-12 00:29:16 +01:00
Henrik Rydgard
784bf82b58 Improve AVX check in CPUDetect. Warning fix.
Keeping the ifdef for zenfone - still doesn't work without it
2014-11-11 23:48:58 +01:00
Henrik Rydgard
7409263eb0 Make CPUDetect.cpp not blow up on Zenfone devices.
Don't really know what's going on here, but this seems like the easiest way out. Appears that CPUID really is buggy on this hardware, which seems weird. Alternate explanations would be that we are checking the wrong bits for xgetbv support or something...
2014-11-11 23:08:28 +01:00
Sacha
ee0c593258 Fix Android. 2014-11-12 03:19:46 +10:00
Sacha
1ba9103cef Better support for new archs. 2014-11-11 22:55:49 +10:00
Unknown W. Brackets
bc7497857a x86jit: Micro optimize vi2x a bit with ssse3/sse4.
Both are small wins.
2014-11-08 12:13:26 -08:00
Unknown W. Brackets
0e646f748a x86jit: Implement vi2x instructions.
Also, my opcodes were wrong in the test (shifted the pair bit the wrong
way, oops.)

AFAICT, there's no reason PSRAD/etc. were not encoding REX...
2014-11-08 12:13:26 -08:00
Andrey Bondrov
5201f03720 Add a hotkey to swap D-pad and left analog stick keys 2014-11-07 22:40:53 +10:00
Unknown W. Brackets
d7bdded6f8 x86jit: fix rip addressing on PEXTRW/PINSRW.
I think this is right anyway, not 100% sure.
2014-11-03 23:18:32 -08:00
Unknown W. Brackets
844c7e73d3 x86jit: Add SSE 4.1 rounding ops to emitter. 2014-11-03 23:18:09 -08:00
Henrik Rydgård
f1d43e7e90 Merge pull request #7046 from unknownbrackets/perf
Improve load and shutdown performance a bit
2014-11-03 21:50:10 +01:00
Unknown W. Brackets
ad6b176e11 Naturally, modern C++ would not build on Symbian. 2014-11-03 08:56:45 -08:00
Unknown W. Brackets
9d86d3ca9b Use std::unordered_multimaps in a few places. 2014-11-03 08:31:52 -08:00
Unknown W. Brackets
dd497e5a02 Show an error when savedata can't be written.
Normally we'll log when the disk is full for files, but if we can't even
create the folder, we'll get errors trying to write into a non-existing
directory.
2014-11-02 13:30:00 -08:00
Henrik Rydgård
28efc0aa13 Merge pull request #6694 from ToadKing/master
Try to allocate executable memory close to PPSSPP memory if not below 4GB
2014-10-18 11:14:20 +02:00
Henrik Rydgård
259c6c94c4 Merge pull request #6952 from fcooper/master-bbb-support
Support ARM Linux device and add Beaglebone as a support device
2014-10-18 11:11:49 +02:00
Henrik Rydgård
6cb2c9c97d Merge pull request #6989 from hrydgard/x86-emitter-merge
Merge from Dolphin's x86-64 emitter
2014-10-12 19:52:59 +02:00
Henrik Rydgård
7bde976069 Merge x64 emitter from a newer Dolphin version.
This one can generate slightly smaller code by exploiting some EAX-only
encoding and various other short forms, and adds support for many newer
CPU instructions.
2014-10-12 19:46:58 +02:00
Henrik Rydgård
281ab5f9cb Sync x64 emitter to Dolphin's. 2014-10-12 19:45:26 +02:00
Henrik Rydgard
8177b4c43b Avoid an ifdef using PTRBITS 2014-10-12 19:35:55 +02:00
Henrik Rydgård
f99c2cd010 x86 Jit: Generate nicer code for some cases of addiu 2014-10-12 17:47:53 +02:00
Unknown W. Brackets
42fe8ee32e Convert FormatMessage() to utf-8 to fix locale.
Otherwise we get non-utf-8 garbage if the user isn't in a Latin codepage.
2014-09-27 09:04:24 -07:00