Commit Graph

4564 Commits

Author SHA1 Message Date
Nekotekina eb377f425c PPU LLVM: Analyser fix
ppu_branch_target removed
2016-07-14 16:27:14 +03:00
Ivan 4c4e4fc772 Merge pull request #1902 from kd-11/master
gl: bug fixes
2016-07-14 00:01:22 +03:00
kd-11 6401eefb26 gl: reset texture scaling factor during bind
reset vertex textures as well
2016-07-13 21:01:16 +03:00
kd-11 ea6f3ff5ed gl: Properly determine front face winding 2016-07-13 21:01:16 +03:00
kd-11 97581d8635 gl: Fix primitive type checks 2016-07-13 21:01:16 +03:00
Nekotekina 6bac4a535b Vulkan: glslang workaround 2016-07-13 18:10:35 +03:00
Nekotekina 0a5b518aea PPU Analyser fix 2016-07-13 02:54:34 +03:00
Nekotekina 94da0dbd4d PPU LLVM improvements
Removed interpreter dependencies
Reciprocal instructions improved
2016-07-12 17:33:13 +03:00
raven02 87e38118a5 Ignore clear if surface target is set to CELL_GCM_SURFACE_TARGET_NONE (#1898) 2016-07-12 21:52:11 +08:00
Nekotekina ba8fd825ec fs::get_data_dir implemented 2016-07-11 22:12:39 +03:00
kd-11 2c981cf940 rsx: mark register access with divider op enabled and frequency 1 (#1892) 2016-07-12 02:53:52 +08:00
Nekotekina 11f7b167ae Update Vulkan submodules
Temporarily disabled Vulkan build for Linux
2016-07-11 19:27:50 +03:00
Nekotekina 06c6c8212d Minor fixes
elf_object
Loader fix
Analyser fix
2016-07-11 16:30:11 +03:00
raven02 b89961f01d rsx: set frame limiter before rsx->flip() (#1885) 2016-07-11 21:00:04 +08:00
kd-11 59df7e7104 glsl: not all shaders write to col0 (#1884)
fix typo
2016-07-11 09:05:13 +08:00
kd-11 bbf7e6332c glsl; dump shaderlog (#1883)
fix typo
2016-07-11 01:16:31 +08:00
Lena 8c3b3f7ab4 CMake: improve compiler checks (#1847)
Moves the compiler check out of rpcs3/CMakeLists.txt and adds them
into rpcs3/cmake_modules/ConfigureCompiler.cmake
Add better compiler checks and eliminate a TODO
2016-07-10 16:15:56 +08:00
raven02 47987efb75 rsx: add default case for to_front_face() (#1879) 2016-07-10 08:33:20 +08:00
raven02 ed8ebda804 vk: add missing compare func not_equal (#1876) 2016-07-10 04:32:09 +08:00
raven02 d4da0e052d sys_net: add few socket options (#1874) 2016-07-10 02:30:56 +08:00
kd-11 6705a6ecc9 rsx: properly check for inlined array before discarding draw (#1877) 2016-07-09 22:54:44 +08:00
raven02 dfa968863a DX12: get_scissor() fix (#1868) 2016-07-09 01:00:15 +08:00
raven02 70885ada67 RSX: throw exception removal for user_command() (#1872) 2016-07-09 00:22:10 +08:00
vlj 71b21fe793 Merge pull request #1869 from raven02/vertex
RSX: check vertex_draw_count
2016-07-08 15:54:33 +02:00
raven02 1f4ebf1501 RSX: check vertex_draw_count 2016-07-08 16:56:12 +08:00
vlj dcb3169f56 Merge pull request #1865 from vlj/rsx-refactor
Factorize rsx state
2016-07-07 22:49:23 +02:00
Vincent Lejeune 772706ca4c Factorize rsx state 2016-07-07 21:38:57 +02:00
Ivan 77594dc66c PPU LLVM: New analyser (#1858)
Minor fixes
VEX prefix support
2016-07-07 21:42:39 +03:00
Ivan 0dc00b8104 Merge pull request #1866 from GhettoGirl/gui-fixes
GUI: improve prx library selection
2016-07-06 23:22:21 +03:00
GhettoGirl 758897a929 GUI: add prx library search functionality in settings dialog
Add the possibility too lookup specific libraries in the settings dialog.
2016-07-05 20:46:51 +02:00
GhettoGirl 5c4af45bf5 GUI: improve prx library selection
Sorts the libraries in the settings dialog alphabetically, so they are easier to find.
Sorts selected and unselected modules alphabetically seperatly.
2016-07-05 00:00:46 +02:00
raven02 effd379c25 Vulkan/DX12 : refactor cull face code (#1849)
* Vulkan/DX12 : refactor cull face code

1

* Add optimal build options to CMakeLists (#1841)

Provides two options when building RPCS3

    USE_SYSTEM_FFMPEG BOOL (default: OFF)
    USE_SYSTEM_LIBPNG BOOL (default: OFF)

These options lets the user select between the system provided and builtin libraries of ffmpeg and libpng to overcome possible system issues.

Also adds support for older libpng releases if the user doesn't have libpng 1.5 or higher.

* DX12: depth buffer compare should perform only if depth testing is enabled (#1848)

* GL: front face regression fix (#1854)

* GUI: Add bulk decryption of LLE modules (#1845)

Adds a new menu to "Tools" called "&Decrypt SPRX libraries" which opens a dialog to select
multiple *.sprx files, which are than decrypted all at once.
This speeds up the LLE module decryption and saves users a lot of time.

The output path is the same where the input module resides and the extension is changed
to *.prx

* vk: Avoid double-copy for vertex attributes (#1852)

* vk: Avoid double-copy for vertex attributes

fix buffer overflow

vk: Fix vertex attrib offset_in_dst for batched draw calls

* whitespace fix only
2016-07-03 10:35:51 +08:00
kd-11 a6b7c9c309 vk: Avoid double-copy for vertex attributes (#1852)
* vk: Avoid double-copy for vertex attributes

fix buffer overflow

vk: Fix vertex attrib offset_in_dst for batched draw calls

* whitespace fix only
2016-07-02 22:27:53 +08:00
Lena 85c4392b34 GUI: Add bulk decryption of LLE modules (#1845)
Adds a new menu to "Tools" called "&Decrypt SPRX libraries" which opens a dialog to select
multiple *.sprx files, which are than decrypted all at once.
This speeds up the LLE module decryption and saves users a lot of time.

The output path is the same where the input module resides and the extension is changed
to *.prx
2016-07-02 19:19:32 +08:00
raven02 c3bac7c92b GL: front face regression fix (#1854) 2016-07-01 20:47:56 +08:00
raven02 45d41267f5 DX12: depth buffer compare should perform only if depth testing is enabled (#1848) 2016-07-01 15:02:40 +08:00
Lena 869e10027d Add optimal build options to CMakeLists (#1841)
Provides two options when building RPCS3

    USE_SYSTEM_FFMPEG BOOL (default: OFF)
    USE_SYSTEM_LIBPNG BOOL (default: OFF)

These options lets the user select between the system provided and builtin libraries of ffmpeg and libpng to overcome possible system issues.

Also adds support for older libpng releases if the user doesn't have libpng 1.5 or higher.
2016-07-01 12:07:16 +08:00
raven02 246b593783 Vulkan/GL: add default case for front face in case of garbage value (#1844) 2016-06-30 12:46:25 +08:00
raven02 d22017bfca DX12: using better approximation for constant color/alpha blending (#1839)
* DX12: using better approximation for constant color/alpha blending

* typo fix

* Comment updated.
2016-06-28 20:19:04 +08:00
DHrpcs3 c7965405d3 Merge pull request #1840 from RPCS3/WIP
RSX improvements
2016-06-28 14:22:12 +03:00
DH d7358848e1 Updated rsx_program_decompiler 2016-06-28 13:24:48 +03:00
DH 989f954432 Added WIP vertex textures support 2016-06-28 12:58:44 +03:00
DH 4d88296485 RSX shaders cache: group programs by indexes 2016-06-28 12:54:08 +03:00
DH 086e1c34dd RSX: added hack to the nv3089::image_in
Remove it when textures cache will be implemented properly
2016-06-28 12:52:10 +03:00
DH e6c6a7fb8b RSX: added proper nv0039::buffer_notify implementation from rsx_cache branch 2016-06-28 12:48:45 +03:00
DH 32830d45ff Improved shaders cache 2016-06-27 21:53:56 +03:00
Ivan 223979c088 Merge pull request #1813 from Nekotekina/llvm1
PPU LLVM: Improvements
2016-06-27 20:52:38 +03:00
Nekotekina 0ea0c21fed LLVM: Indirect call map 2016-06-27 19:34:14 +03:00
Nekotekina 7a921cbdf9 cpu_thread compressed 2016-06-27 16:43:57 +03:00
Nekotekina 9db7de29fb Syscall analysis implemented 2016-06-27 16:43:56 +03:00