summaryrefslogtreecommitdiffstats
path: root/drivers
Commit message (Collapse)AuthorAgeFilesLines
...
* | | Further fixes to avoid memory corruption, closes #25336Juan Linietsky2019-01-264-18/+8
| | |
* | | Properly dispose of custom shaders, closes #19300Juan Linietsky2019-01-254-2/+58
| | |
* | | Do not use depth prepass if effects are disabled, fixes #25267Juan Linietsky2019-01-251-1/+1
| | |
* | | Revert "Fix GLES3 error 502 on iOS"Bastiaan Olij2019-01-261-1/+1
| | |
* | | Ensure contact shadows are only used when lights use them.Juan Linietsky2019-01-251-1/+11
| | |
* | | Fix GLES3 error 502 on iOSBastiaan Olij2019-01-261-1/+1
| | |
* | | Fix depth prepass in GLES2, closes #23321Juan Linietsky2019-01-251-0/+6
| | |
* | | Ensure WebGL uses highp by default, and fix some wrong mediump usages. Fixes ↵Juan Linietsky2019-01-245-7/+33
| | | | | | | | | | | | #22962.
* | | Fix crash with cube shadows in gles2, closes #22635Juan Linietsky2019-01-241-1/+1
|/ /
* | Revert "Fix errors on iOS"Juan Linietsky2019-01-242-12/+2
| |
* | Makes screen texture work in GLES2 (2D for now), fixes #23604Juan Linietsky2019-01-244-5/+85
| |
* | Merge pull request #25257 from karroffel/tex3d-get-data-fixRémi Verschelde2019-01-242-4/+144
|\ \ | | | | | | implemented texture_get_data() for TextureLayered
| * | implemented texture_get_data() for TextureLayeredthomas.herzog2019-01-232-4/+144
| | |
* | | Fix problem with texture2Dlod, closes #25263Juan Linietsky2019-01-233-3/+10
| | |
* | | Merge pull request #25242 from BastiaanOlij/fix_ios_issuesRémi Verschelde2019-01-232-2/+12
|\ \ \ | | | | | | | | Fix errors on iOS
| * | | Fix errors on iOSBastiaan Olij2019-01-232-2/+12
| |/ /
* / / Remove unused method in RasterizerStorageGLES2Rémi Verschelde2019-01-231-9/+0
|/ / | | | | | | Added in 4f4e46edd539b0c26a6b086aa19c303b10de66b1 but not used in the end.
* | Implement unpacking for compressed vertex formats on GLES2 when not ↵Juan Linietsky2019-01-225-6/+262
| | | | | | | | supported, fixes #22957
* | Do not use shadow cubemaps if depth write is not supported to avoid errors, ↵Juan Linietsky2019-01-221-1/+1
| | | | | | | | closes #25219
* | Changes to GLES2 renderer to not use cube shadows if not available, fixes #25132Juan Linietsky2019-01-225-2/+12
| |
* | WebGL1 some changessantouits2019-01-223-3/+29
| | | | | | | | | | | | | | | | | | | | | | | | glRenderBufferStorage doesn't accept the _DEPTH_COMPONENT24_OES so I changed it to GL_DEPTH_COMPONENT16 https://developer.mozilla.org/en-US/docs/Web/API/WebGLRenderingContext/renderbufferStorage https://www.khronos.org/registry/webgl/extensions/rejected/OES_depth24/ Enabled ext_shader_texture_lod extension. I guess it does the same like the GL_ARB_shader_texture_lod extension, but in WebGL1 https://www.khronos.org/registry/webgl/extensions/EXT_shader_texture_lod/ And last WebGL1 doesn't allow indexes not constant, so I changed it into a loop reading this: https://stackoverflow.com/questions/19529690/index-expression-must-be-constant-webgl-glsl-error
* | Merge pull request #25196 from santouits/webgl-backslashesRémi Verschelde2019-01-211-10/+5
|\ \ | | | | | | webgl1 in firefox doesn't like backslashes in #define directive
| * | webgl1 doesn't like backslashes in #definesantouits2019-01-211-10/+5
| |/ | | | | | | | | | | | | According to https://www.khronos.org/registry/OpenGL/specs/es/2.0/GLSL_ES_Specification_1.00.pdf , backslashes aren't necessary supported as line continuation characters in preprocessor directives
* | Add function to obtain filesystem type from DirAccess.Juan Linietsky2019-01-214-0/+39
| | | | | | | | Change EditorFileSystem to not use directory modification times on FAT32, fixes #20946
* | Merge pull request #25115 from marxin/fix-19633-properRémi Verschelde2019-01-212-17/+12
|\ \ | | | | | | Fix #19633 by proper store to &ubo_data.shadow_matrix[1234].
| * | Fix #19633 by proper store to &ubo_data.shadow_matrix[1234].marxin2019-01-192-17/+12
| |/ | | | | | | | | | | | | | | | | It is not valid in C++ to store into shadow_matrix1[16] with shadow_matrix1[16 * j] (for j > 0). Even though there's a valid space in a struct after shadow_matrix1. Knowing that GCC performs aggressive optimizations that eventually lead to a wrong code. Code has been changed into union where one can either use shadow_matrix[4 * 16], or individual shadow_matrix1, shadow_matrix2, etc. GCC pragma is not needed any longer.
* | Merge pull request #25117 from dragmz/patch-6Rémi Verschelde2019-01-211-7/+4
|\ \ | | | | | | Replace CreateThread with QueueUserWorkItem
| * | Replace CreateThread with QueueUserWorkItemMarcin Zawiejski2019-01-191-7/+4
| | | | | | | | | Fixes #24869 stuttering on Windows by reusing long running threadpool threads instead of creating a new thread on each call to Thread::start.
* | | Cleanup after @reduz :)Rémi Verschelde2019-01-211-4/+0
| |/ |/| | | | | Fixes #25172.
* | Fix excluding GCC pragmas from ClangRémi Verschelde2019-01-191-2/+2
| | | | | | | | Why the heck does Clang define __GNUC__...
* | Only define GCC pragmas for GCCRémi Verschelde2019-01-191-1/+5
|/
* Use 16 bit indices on phones that dont support 32, fixes #19797Juan Linietsky2019-01-183-3/+20
|
* Merge pull request #25101 from hpvb/fix-19633Juan Linietsky2019-01-181-0/+6
|\ | | | | Work around a GCC optimizer bug at -O3
| * Work around a GCC optimizer bug at -O3Hein-Pieter van Braam2019-01-181-0/+6
| | | | | | | | | | | | | | This bug has been reported upstream, once it gets fixed we'll add some version guards to only disable this for specific GCC versions. This fixes #19633
* | Merge pull request #25070 from RedMser/shader_errors_linefixRémi Verschelde2019-01-182-2/+2
|\ \ | |/ |/| Fix shader compile error line numbers starting at 0
| * Fix shader compile error line numbers starting at 0.RedMser2019-01-172-2/+2
| |
* | Fix enum scopeEmanuele Fornara2019-01-171-8/+8
|/
* Fixes to 2D lights, closes #24750Juan Linietsky2019-01-173-23/+20
|
* Ensure texture hints are obeyed, fixes #24875Juan Linietsky2019-01-162-0/+26
|
* Ensure canvas copy texscreen will not crash if render target is configured ↵Juan Linietsky2019-01-151-0/+5
| | | | without copy buffers. Closes #24749.
* GLES2: Fix typo in project settingRémi Verschelde2019-01-151-2/+2
| | | | It used a different name than the equivalent GLES3 parameter.
* Use GLES2 approach to vertex shading in GLES3, which has been more ↵Juan Linietsky2019-01-141-5/+55
| | | | developed. Fixes #21852
* Implement black margins in GLES2, which was missing, fixes #24556Juan Linietsky2019-01-141-0/+58
|
* GLES2: Make Nvidia flicker workaround opt-inRémi Verschelde2019-01-142-119/+130
| | | | | | | | | | | It has a big impact on 2D and text rendering performance (cf. #24466) so the solution seems worse than the bug it aims to work around. It's now opt-in via "rendering/quality/2d/gles2_use_nvidia_rect_flicker_workaround" for those who need it and have a simple enough game for the performance drop not to be an issue. Fixes #24466.
* Was not properly re-setting light uniforms, fixes #24976Juan Linietsky2019-01-142-29/+41
|
* Do not use the workaround for desktop nvidia on mobile and html5.Juan Linietsky2019-01-141-1/+2
|
* GLES2: Clarify why we exclude debug code on iOSRémi Verschelde2019-01-111-6/+11
| | | | Supersedes and closes #24873.
* Consistency in resource format saver/loader de-registrationRémi Verschelde2019-01-101-4/+2
| | | | | | | | | | Some used 'is_valid()' checks, others not. Validity is already checked in 'unref()', and 'remove_resource_format_*()' has an ERR_FAIL condition on 'is_null()' already (which shouldn't happen since we're only unregistering things that we previously registered. Also add missing GDCLASS statement in ResourceFormatLoaderVideoStreamGDNative, missed in #20552 which was last amended before #19501 was merged.
* Merge pull request #24301 from marcelofg55/default_device_pulseaudioRémi Verschelde2019-01-071-0/+42
|\ | | | | PulseAudio driver will now change the device if the default system device changes
| * PulseAudio driver will now change the device if the default system device ↵Marcelo Fernandez2018-12-121-0/+42
| | | | | | | | changes