5af656ea85
wined3d: Allow texture updates in some cases even when mip levels don't match.
2015-04-30 20:07:08 +09:00
15720f09de
wined3d: Report ATI1N and ATI2N volume textures as not supported.
2015-04-30 19:58:01 +09:00
ca46914d07
wined3d: Use string_buffer_sprintf() in shader_glsl_generate_ffp_fragment_shader().
2015-04-29 17:02:30 +09:00
12e59c5278
wined3d: Use string_buffer_sprintf() in handle_ps3_input().
2015-04-29 17:02:27 +09:00
da24b51139
wined3d: Don't try to apply format fixups for SM4+ shaders.
2015-04-29 12:24:36 +09:00
28343db208
wined3d: Don't use the builtin FFP uniforms for the lights.
...
While at it, use D3D-like uniforms for the light parameters and
(consequently) implement range and proper D3D spotlights support.
2015-04-28 20:02:23 +09:00
e226e7593a
wined3d: Use struct wined3d_vec4 to store the light position and direction.
2015-04-28 20:02:21 +09:00
540130bfa9
wined3d: Store the directional light direction in the lightDirn field.
...
It's only an implementation detail that the direction needs to be passed
to the fixed function GL pipeline as GL_POSITION.
2015-04-28 20:02:18 +09:00
b5affa6d3c
wined3d: Set spotlight direction .w component to 0.0f.
...
It's a direction so we don't want to get translations applied to it.
The variable is only used internally.
2015-04-28 20:02:12 +09:00
f09288cfc4
wined3d: Cleanup the sampler map in shader_cleanup() (Valgrind).
2015-04-28 15:34:12 +09:00
4b86f13ade
wined3d: Also free the input signature in shader_cleanup() (Valgrind).
2015-04-27 18:05:24 +09:00
79f4ca9ec4
wined3d: Release the context before potentially destroying the swapchain in wined3d_device_uninit_3d() (Valgrind).
2015-04-27 18:05:22 +09:00
0152b0dad3
wined3d: Move the ARB_texture_rectangle check to init_format_texture_info.
2015-04-27 15:07:48 +09:00
0d6d64b358
wined3d: Move the ARB_texture_cube_map check to init_format_texture_info.
2015-04-27 15:07:46 +09:00
7109bebd02
wined3d: Move volume DXTn handling to apply_format_fixups.
2015-04-27 15:07:42 +09:00
978fe23218
wined3d: Don't use the builtin FFP uniforms for the material.
2015-04-27 15:07:11 +09:00
6984ca7f46
wined3d: Don't use the builtin FFP uniforms for texture transformation matrices.
2015-04-27 15:07:06 +09:00
84ca0203bf
wined3d: Introduce a scratch string buffers framework.
2015-04-27 15:07:03 +09:00
54b4df6731
wined3d: Reduce initial wined3d_string_buffer buffer size to 32.
...
Also try to avoid unnecessary HeapReAlloc() + vsnprintf() loops when we
know the required size of the buffer.
2015-04-27 15:06:58 +09:00
64463b81df
wined3d: Rename struct wined3d_shader_buffer to wined3d_string_buffer.
...
Also rename the shader_buffer_(clear|init|free) functions.
2015-04-27 15:06:55 +09:00
072e1a3fb2
wined3d: Move the EXT_texture3D check to init_format_texture_info.
2015-04-24 19:08:45 +09:00
9def5ff409
wined3d: Query info for each resource type from ARB_internalformat_query2.
2015-04-24 19:08:41 +09:00
e4e1e7d8d3
wined3d: Read flags of the right resource type in check_device_format.
2015-04-24 19:08:39 +09:00
3c0ed9ce0a
wined3d: Introduce resource-type specific format flags.
2015-04-24 19:08:35 +09:00
53ff2a9d22
wined3d: Drop the ARB_SHADER_OBJECTS extension flag.
...
The flag is currently unused and also unnecessary since both
ARB_vertex_shader and ARB_fragment_shader specs require
ARB_shader_objects support.
2015-04-24 19:08:15 +09:00
7928ffaca5
wined3d: Drop glPrioritizeTextures() call.
...
It's not supported on core profiles and I doubt it matters anyway.
For the records, the default value is 1 so this was actually lowering
the priority for DEFAULT pool textures.
2015-04-24 19:08:10 +09:00
d6db6c7295
wined3d: Fix specular lighting for non-local viewer.
...
The viewer is in the (0.0, 0.0, -1.0) direction in the D3D coordinate
system.
2015-04-24 19:08:01 +09:00
b963c73abb
wined3d: Introduce D3D7+ normal matrix computation.
2015-04-24 19:07:52 +09:00
cfd280fd66
wined3d: Get rid of the tracing code in shader_vaddline().
2015-04-24 14:29:39 +09:00
dd2398297c
wined3d: Simplify shader_arb_dump_program_source().
2015-04-24 14:29:32 +09:00
4fbaab2020
wined3d: Shadow format flags in wined3d_rendertarget_view.
2015-04-23 20:43:20 +09:00
6f6dfe91d6
wined3d: Make glsl_vertex_pipe_view() static.
2015-04-23 20:43:19 +09:00
9076612e2d
wined3d: Shadow format flags in the resource.
2015-04-22 23:24:41 +09:00
6e4ad78ff4
wined3d: Slightly simplify get_info_log_line().
2015-04-21 23:44:59 +09:00
de1450788b
wined3d: Completely initialize "correction_params" in shader_glsl_load_constants() (Valgrind).
...
Leaving the last two elements uninitialized is mostly harmless since the
shader doesn't actually use them, but the driver might try to do some kind of
analysis on them.
2015-04-21 22:22:00 +09:00
c7780791ea
wined3d: Introduce a function to compile ARB vertex/fragment programs.
2015-04-21 22:21:56 +09:00
8063d4cfcc
wined3d: Initialize "type.padding" in arbfp_blit_set() (Valgrind).
2015-04-21 22:21:52 +09:00
8b6cddbe25
wined3d: Get rid of context_invalidate_active_texture().
...
Calling wined3d_texture_bind() + context_invalidate_active_texture() is
equivalent to just calling wined3d_texture_bind_and_dirtify().
2015-04-16 20:44:43 +09:00
dcaaf915f8
wined3d: Use wined3d_texture_bind_and_dirtify() in wined3d_texture_upload_data().
2015-04-16 20:44:41 +09:00
5c9da580a8
wined3d: Use wined3d_texture_bind_and_dirtify() in device_update_volume().
...
I missed this in 7c4d512fee
2015-04-16 20:44:37 +09:00
52478e3809
wined3d: Bind the texture before calling wined3d_volume_upload_data() in wined3d_volume_load_location().
2015-04-16 20:44:31 +09:00
aad0ab8101
wined3d: Properly handle WINED3DSIH_DP2 in the GLSL shader backend.
2015-04-15 22:21:59 +09:00
a37099e132
wined3d: Bind sampler objects to the correct texture stage.
...
On current hardware this generally only makes a difference for vertex
textures. This fixes a regression introduced by commit
c6232e1d11
2015-04-15 22:21:58 +09:00
dc53f6c2e6
wined3d: Use the display format if WINED3DFMT_UNKNOWN is passed as backbuffer format for windowed mode resets.
2015-04-15 22:21:58 +09:00
27209c69a9
wined3d: Call arbfp_blit_surface and ffp_blit_blit_surface through the blit interfaces.
2015-04-15 22:21:58 +09:00
6b455e74a0
wined3d: Move fixed function draw blits out of surface_blt_special.
2015-04-15 22:21:58 +09:00
f84d741cb3
wined3d: Implement color keying in the ARB fixed function pipeline.
2015-04-15 22:21:58 +09:00
483d9550c4
wined3d: Use finer grained colorkey state invalidation.
2015-04-15 22:21:58 +09:00
382fa588a9
wined3d: Handle color key dirtification in wined3d_cs_exec_set_texture.
2015-04-15 22:21:57 +09:00
8fff427bc0
wined3d: Load ModelView matrix for glLightfv calls.
...
Analogous to the light() function in state.c, the correct ModelView
matrix has to be loaded when glLightfv is called.
2015-04-13 16:41:48 +09:00
c5c7137212
wined3d: Simplify alpha test and color keying in GLSL.
2015-04-10 21:44:13 +09:00
5c7707dc77
wined3d: Move alpha test states to the fragment pipeline.
...
This will allow some color keying cleanups. For Core Contexts we'll have
to emulate alpha testing in shaders anyway.
2015-04-10 21:44:01 +09:00
f4b6874491
wined3d: Match blitter and fragment processing color keying capabilities.
...
This is necessary if e.g. the nvts or ffp fragment processing pipeline
is selected for debugging purposes on a GPU that otherwise supports the
ARB program blitter.
2015-04-10 21:43:51 +09:00
5d51c97d07
wined3d: Implement color keying in the glsl fragment pipeline.
2015-04-10 21:43:31 +09:00
2d56694445
wined3d: Implement color keying in arbfp_blit.
2015-04-10 21:43:12 +09:00
439076d591
wined3d: Allow ARB blits from converted surfaces.
2015-04-10 21:42:47 +09:00
82db5954c4
wined3d: Set color keys through the command stream.
2015-04-10 21:42:41 +09:00
086d2af629
wined3d: Add support for Nvidia GTX 970M.
2015-04-10 21:42:27 +09:00
14219b0c3b
wined3d: Fix a typo in vertexshader_init.
...
Should fix the graphical corruptions recently introduced by
002713de6d
2015-04-08 23:14:24 +09:00
3a14df3b9a
wined3d: Fix typo in init_output_registers.
2015-04-08 16:22:41 +09:00
b413a4e202
wined3d: Add more formats to wined3d_format_convert_color_to_float.
2015-04-07 23:10:33 +09:00
fa0a9232e0
wined3d: Move surface_convert_color_to_float to utils.c.
2015-04-07 23:10:30 +09:00
8a101bc1f5
wined3d: Add a function for wined3d_gl_resource_type to texture target conversion.
2015-04-07 23:10:20 +09:00
f260c17b34
wined3d: Use a shader for unconverted blits.
2015-04-07 23:10:14 +09:00
e8b9f8134d
wined3d: Define a constant for the blit texture size parameter.
2015-04-03 00:33:29 +09:00
ad50f872ae
wined3d: Pass a struct arbfp_blit_type to the blit shader generation functions.
2015-04-03 00:33:28 +09:00
6cf421399b
wined3d: Use bitfields in struct arbfp_blit_type.
2015-04-03 00:33:28 +09:00
da7abd29b4
wined3d: Pass a wined3d_ffp_texture_type enum to blit shader functions.
2015-04-03 00:33:28 +09:00
844c4efb47
wined3d: Give tex_type and its values a better name.
2015-04-03 00:33:28 +09:00
94b0e895db
wined3d: Use GLSL 1.30 for SM4 shaders when available.
...
This is required for bitwise operators in particular if no other extension
provides them.
2015-03-31 18:49:28 +09:00
d0d8a7827b
wined3d: Take WINED3D_PIXEL_CENTER_INTEGER into account when calculating vpos.
2015-03-31 18:49:28 +09:00
94967c28b9
wined3d: Add support for half-integer pixel centers.
...
Like in OpenGL and Direct3D 10+.
2015-03-30 19:10:29 +09:00
05757794a0
wined3d: Don't use the builtin FFP uniform for the projection matrix.
2015-03-27 20:40:38 +09:00
7a8ac888de
wined3d: Don't use the builtin FFP uniform for the normal matrix.
2015-03-27 20:40:38 +09:00
be4c7a0a84
wined3d: Implement per stage constants in atifs.
2015-03-27 20:40:37 +09:00
3c2d0b983a
wined3d: Avoid constant collision in atifs.
2015-03-24 20:06:18 +09:00
239e8cad7c
wined3d: Add per-context private data for fragment pipelines.
2015-03-24 20:06:15 +09:00
1b0c063632
wined3d: Check for conversion changes in the atifs fragment pipeline.
...
Mapping colorop to pixelshader is necessary because misc_state_template
defines a pixel shader handler. We have to use the same representative
in all pipeline parts.
Wined3d_cs_exec_set_texture dirtifies the pixel shader state when the
fixup of a texture changes. Atifs no longer needs a pixel shader handler
on its own since vertex shader updates aren't delayed any more when the
pixel shader state is dirty.
2015-03-24 20:06:05 +09:00
92fee8c04f
wined3d: Improve color fixups in atifs shaders.
...
The abilities of this hardware is too limited to support generic sign /
swizzle fixups. A generic handler would consume 4 of the 8 available
color instruction slots and 2 alpha instruction slots.
The bump mapping handler code has its own way of handling the color
fixups. It merges the fixup into the perturbation calculation without
requiring extra shader instructions. In theory this is possible for the
majority of d3d texture ops as well, but I don't think this is worth the
effort. I expect that this code will only be used for the ddraw signed
format test in practice.
2015-03-24 20:05:46 +09:00
010ff106dd
d3d10core: Implement d3d10_device_DrawInstanced().
2015-03-24 17:41:17 +09:00
a33d3ad41b
wined3d: Add support for SV_InstanceID.
2015-03-24 17:41:12 +09:00
f9fafb3686
wined3d: Use the "sysval_semantic" field instead of "semantic_name" to recognize SV_POSITION.
2015-03-24 17:41:09 +09:00
1b08196c5a
wined3d: Implement d3d10 style instance data step rates.
2015-03-24 17:41:06 +09:00
74cda79ac3
wined3d: Introduce a get_texture_matrix() function.
2015-03-23 22:59:20 +09:00
2d270f31c6
wined3d: Introduce a get_projection_matrix() function.
...
Notice that I'm using floats instead of doubles in the new function,
mostly to be able to use struct wined3d_matrix and multiply_matrix().
At a rough estimate the precision should still be good enough.
2015-03-23 22:59:19 +09:00
65c8c40b18
wined3d: Use a separate STATE_VIEWPORT state handler in the GLSL pipeline.
2015-03-23 22:59:18 +09:00
88f230b56e
d3d10core: Also pass unused input layout elements to wined3d.
...
While these won't be used by the shader, they potentially affect the
calculated offset for WINED3D_APPEND_ALIGNED_ELEMENT elements.
2015-03-23 22:59:12 +09:00
cbe717b8f3
wined3d: Take the input slot into account when handling WINED3D_APPEND_ALIGNED_ELEMENT.
2015-03-23 22:59:11 +09:00
f3765e9f77
wined3d: Merge some common shader initialization code into shader_init().
2015-03-23 22:59:10 +09:00
002713de6d
wined3d: Store shader input signatures as a wined3d_shader_signature structure.
2015-03-23 22:59:08 +09:00
2e5ad1b6e3
wined3d: Store shader output signatures as a wined3d_shader_signature structure.
...
Instead of a fixed array of wined3d_shader_signature_element structures.
Shader model 4 shaders can have different semantics in a single
register, e.g. v1.xy TEXCOORD0 and v1.zw TEXCOORD1, so having a single
wined3d_shader_signature_element structure per register isn't
necessarily sufficient.
2015-03-23 22:59:07 +09:00
ec78d50193
wined3d: Fix a TRACE in wined3d_device_draw_indexed_primitive_instanced().
2015-03-20 21:32:50 +09:00
ba396e4212
wined3d: Simplify the wined3d_matrix structure.
2015-03-20 21:32:46 +09:00
f505e6fb4a
Assorted spelling fixes.
2015-03-19 21:54:14 +09:00
9811d85141
wined3d: Don't use the builtin FFP uniform for the modelview matrix.
2015-03-19 20:56:59 +09:00
425dc69fef
wined3d: Use a separate STATE_TRANSFORM(WINED3D_TS_VIEW) state handler in the GLSL pipeline.
2015-03-19 20:56:56 +09:00
99f3e835ac
wined3d: Use a separate STATE_VDECL state handler in the GLSL pipeline.
...
It's mostly a copy of vertexdeclaration() from state.c, with a few
differences due to the new function being private to the GLSL backend
(e.g. d3d_info.vs_clipping is known to be TRUE) and the fragment fog
update part being split out.
2015-03-19 20:56:50 +09:00
50861156e2
wined3d: Introduce a get_modelview_matrix() function.
2015-03-19 20:56:45 +09:00
fe4a226bfa
wined3d: Introduce a get_identity_matrix() function.
2015-03-19 20:56:42 +09:00
df829de88d
ddraw: Handle DDBLT_ROP in ddraw.
2015-03-18 21:03:22 +09:00
38e4cb29e7
wined3d: Depend on the vertex shader state instead of the vertex declaration state where appropriate.
2015-03-18 21:02:55 +09:00
d8c5e417c2
wined3d: Force stream info update on vertex shader change.
2015-03-18 21:02:41 +09:00
406cfefc2f
wined3d: Initialize WINED3DTA_TEMP.
2015-03-11 22:09:42 +09:00
f298d30b39
wined3d: Fix a typo in a trace in context_create().
2015-03-11 22:08:34 +09:00
d4db09d0e9
wined3d: Support WINED3DFMT_R5G5_SNORM_L6_UNORM with EXT_texture_snorm.
...
From: Stefan Dösinger <stefandoesinger@gmx.at>
2015-03-10 21:43:34 +09:00
6df52ca7e6
wined3d: Add support for GL_EXT_texture_snorm.
2015-03-10 21:43:32 +09:00
6f2f73a456
wined3d: Improve the unsigned emulation of WINED3DFMT_R5G5_SNORM_L6_UNORM.
2015-03-10 21:43:11 +09:00
ce836bf651
wined3d: Properly up-scale WINED3DFMT_R5G5_SNORM_L6_UNORM.
2015-03-10 21:43:07 +09:00
44ddcac6c3
wined3d: Recognize Nvidia GeForce 8100 with Nouveau driver (bug).
2015-03-10 21:42:50 +09:00
db80aba5ad
wined3d: Drop redundant check for instanced rendering from drawStridedFast.
...
We need GL_ARB_instanced_arrays to be able to take the fastest path and
that's already checked in draw_primitive. Note that
GL_ARB_instanced_arrays provides glDrawElementsInstanced if
GL_ARB_draw_instanced is not supported so we don't need to explicitly
check for the latter.
2015-03-02 13:48:00 +09:00
8ed9c2a11b
wined3d: Add a couple checkGLcall() in surface_prepare_rb().
2015-03-02 13:47:56 +09:00
2f804307bd
wined3d: Drop a few unnecessary &x[0].
2015-03-02 13:47:52 +09:00
7f618695c6
wined3d: Remove an obsolete comment.
...
Wined3d doesn't actually require the GLX extension for multisampling
at this point (while it does require FBOs).
2015-03-02 13:47:46 +09:00
67d2b52c27
wined3d: Remove ARB_/EXT_TEXTURE_ENV_ADD.
2015-02-26 20:08:57 +09:00
aca8ab91eb
wined3d: Use the core version of the INCR_WRAP and DECR_WRAP tokens.
2015-02-25 20:35:28 +09:00
78ae4f66d7
wined3d: Set GL_UNPACK_ALIGNMENT to 1.
...
Alignment should be taken care by the surface pitch and
GL_UNPACK_ROW_LENGTH alone just fine, as long as the pitch is a
multiple of the pixel size. Client APIs respect that,
internally we need to do the same and the changes to
surface_load_texture and wined3d_volume_upload_data are there
for this reason (e.g. WINED3DFMT_R16G16_SNORM is emulated with
GL_RGB16 when NV_texture_shader isn't supported).
2015-02-20 22:53:09 +09:00
7c4d512fee
wined3d: Make sure the correct texture is bound in device_update_volume.
2015-02-19 17:52:03 +09:00
b47695bf3a
wined3d: Reject unsupported pitches in wined3d_surface_update_desc().
2015-02-19 17:51:47 +09:00
d56c168da1
wined3d: Set the surface pitch for user-memory surfaces.
2015-02-19 17:51:36 +09:00
3fa8f4ece3
wined3d: Apply modifiers after floatBitsToInt()/floatBitsToUint() in shader_glsl_add_src_param().
2015-02-18 23:21:49 +09:00
52ab9e382e
wined3d: Read WINED3D_TEXTURE_COLOR_KEY and gl_color_key to find format conversions.
...
The idea is to make it clear that wined3d_texture_load is responsible
for checking the color key for changes. Once we can do color keying in
the shader we also have one place where we can disable the surface
conversion.
2015-02-18 22:02:39 +09:00
e53d22b0c0
wined3d: Manage color key changes in the texture.
...
This fixes a regression with color keyed mipmaps that was introduced by
ee8a5b7dd1
2015-02-18 22:02:33 +09:00
849149d5d3
wined3d: Add support for appending vertex declaration elements.
2015-02-13 18:53:17 +09:00
9b3ab8a6f4
wined3d: Set z = 0.0 via the projection matrix instead of depth clamping.
...
This works on cards that don't implement ARB_depth_clamp like r500
cards. Note that texturing is influenced by position.w, not position.z.
2015-02-12 22:20:05 +09:00
471e626052
wined3d: Fix the .spec.
2015-02-12 22:12:11 +09:00
e4cb3b5081
wined3d: Recognize the SM4 or opcode.
2015-02-11 23:05:17 +09:00
3225fea30c
wined3d: Recognize the SM4 ne opcode.
2015-02-11 23:05:13 +09:00
8506da62e0
wined3d: Implement WINED3DSIH_SQRT in the GLSL shader backend.
2015-02-11 23:05:08 +09:00
1dfea3a4ba
wined3d: Do not validate format restrictions for buffers.
...
Buffers don't have a format as such, and are created with either
WINED3DFMT_UNKNOWN or WINED3DFMT_VERTEXDATA.
2015-02-11 23:05:04 +09:00
5045ca639d
wined3d: Add some warnings when resource_init() fails due to format restrictions.
2015-02-11 23:04:59 +09:00
ac3df5e1f1
wined3d: Use the proper pitch in read_from_framebuffer().
...
This function is only used in the non-default "backbuffer" ORM.
2015-02-11 22:21:36 +09:00
20e61b396e
wined3d: Use unsigned short for the bitfields in struct color_fixup_desc.
2015-02-11 22:21:28 +09:00
b875a4a44a
wined3d: Use GL_CLAMP_TO_EDGE instead of GL_CLAMP.
...
GL_CLAMP was removed from core profile and probably GL_CLAMP_TO_EDGE was
intended to be used here anyway.
2015-02-10 16:21:55 +09:00
f4da723151
wined3d: Call glReadBuffer() through the appropriate function pointer.
2015-02-10 16:21:41 +09:00
1612dc4dc9
wined3d: Refuse pixel shaders declaring POSITION0 inputs.
2015-02-09 22:05:43 +09:00
74b4516bfa
wined3d: Only consider POSITION0 PS inputs as vpos.
2015-02-09 22:05:41 +09:00
96b48191c0
wined3d: Only copy POSITION0 output semantics into gl_Position.
2015-02-09 22:05:35 +09:00
1aa5015467
wined3d: Directly use the core version of glTex[Sub]Image3D.
2015-02-09 22:05:32 +09:00
da5041eb4b
wined3d: Check the d3d buffer to find converted attributes.
...
This fixes some tests with ARB shaders that were broken since
1360af6adc
2015-02-06 22:20:26 +09:00
e178b7ec9a
wined3d: Don't use TXP for rectangle P8 textures.
...
This was probably taken from the YUV shaders, incorrectly and in reverse
logic. It did happen to work though because fragment.texcoord[0].w is 1.0.
2015-02-04 23:15:03 +09:00
3d49060fdf
wined3d: Fix the color keyed P8 alpha ref value.
2015-02-04 23:14:59 +09:00
9defaa53a1
wined3d: Use the core version of the asynchronous queries functions.
2015-02-02 22:39:05 +09:00
3ef1d6bcd9
wined3d: Pass WM_SYSCOMMAND(SC_RESTORE) to DefWindowProc.
2015-01-28 11:06:48 +01:00
7b45103c4a
wined3d: Fix spec file entry for wined3d_texture_create.
2015-01-23 17:28:40 +01:00
d360a51f3c
wined3d: Use the core version of the shader object functions.
2015-01-23 12:02:37 +01:00
e192939652
wined3d: Use the core version of GL vertex attribute functions where possible.
2015-01-23 12:02:33 +01:00
584892b692
wined3d: Properly initialize the last two components of R16G16_SINT attributes.
2015-01-23 12:02:28 +01:00
270431c220
wined3d: Don't normalize R8G8B8A8_UINT attributes.
2015-01-22 13:37:55 +01:00
8ddb7e32f8
wined3d: Add separate stencil state setup via GL 2.0+ core functions.
2015-01-22 13:37:53 +01:00
915e0ef0cf
wined3d: Always use the core version of the glPointParameteri function.
...
The state function depends on GL 2.0+ so it's unnecessary to
check for NV_POINT_SPRITE.
2015-01-20 16:09:01 +01:00
Matteo Bruni