The buffer upload size appears to have been incorrect for quite some time, which causes uploading from undefined memory.
(cherry picked from commit 3dc0e97d05)
This was comparing arrays, GCC 12 raises a warning for it:
```
drivers/gles2/shader_gles2.cpp: In member function 'bool ShaderGLES2::bind()':
drivers/gles2/shader_gles2.cpp:80:71: error: comparison between two arrays [-Werror=array-compare]
80 | if (active != this || !version || new_conditional_version.key != conditional_version.key) {
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gles2/shader_gles2.cpp:80:71: note: use unary '+' which decays operands to pointers or '&'component_ref' not supported by dump_decl<declaration error>[0] != &'component_ref' not supported by dump_decl<declaration error>[0]' to compare the addresses
```
(cherry picked from commit 76df26b110)
When compressed vertex positions are used in a blend shapes mesh, we
need to make sure we set the w-component of the position vector to 1.0
When octahedral compression is used on normals/tangents, they need to be
converted to cartesian floats to be used for blend shapes
This conversion also changes the number of components of that vertex
attribute, which caused issues because previously there was an
assumption that you had the same number of components in the blend shape
buffer as you did in the original mesh's buffer (which is not true for
oct norm/tang)
(cherry picked from commit 733a84f7a4)
Wrapper functions for uploading buffers to OpenGL take all sizes and offsets in bytes. Some buffer sizes are specified as units (e.g. float) so require conversion to bytes when calling the buffer upload functions.
Two such bugs have been fixed in blendshapes, and parameter names and comments have been changed to emphasize that sizes should be in bytes.
In addition DEV_ASSERTS in the upload wrappers have been changed to ERR_FAIL.
(cherry picked from commit 614dc363ab)
On some platforms, exporters are prevented from exporting S3TC textures. This causes problems if the .import file contains a reference to such a texture - the exported project will attempt to load the S3TC, fail, and probably crash.
This PR prevents this problem by faking lack of hardware support for S3TC on the affected platforms. This prevents the engine attempting to load the S3TC and avoids the problem.
(cherry picked from commit b15ad8e786)
A previous change missed setting a flag that specified whether half
floats were being used on vertex positions when in the GLES2 driver
This caused errors with the vertex buffer when platforms (specifically
iOS) which do not properly support half float vertex attributes on
GLES2, try to remapt the vertex buffer to stop using half floats
(in this case that remapping never happened and caused artifacts)
Re-enable setting that flag to fix rendering issues on these platforms
(cherry picked from commit bf5b2f48e6)
the 'trunc' function was comparing integers with floats using the '>' operator, which is not supported in Android. This issue is now gone.
(cherry picked from commit 9f68b4fea3)
This PR adds the code necessary to correctly free environments in GLES2, which was previously missing causing them to leak.
(cherry picked from commit 7134ef90bb)
The RID Handles PR introduced a couple of bug fixes, they are added here without the RID handles functionality in order to keep 3.4 and 3.x roughly in sync, allowing further fixes to RID lifetimes to be backported.
For octahedral compressed normals/tangents, we use vec4 in the shader
regardless of whether a normal/tangent does/doesn't exist
For the case where we only have a normal vector, we need to specify that
there are only two components being used when calling glVertexAttrib
Before we would always specify that there were 4 components, and used
offsets to determine where in the vertex buffer to read data from but
this doesn't work on all platforms
(cherry picked from commit 8a43b222c7)
Sets `AlignOperands` to `DontAlign`.
`clang-format` developers seem to mostly care about space-based indentation and
every other version of clang-format breaks the bad mismatch of tabs and spaces
that it seems to use for operand alignment. So it's better without, so that it
respects our two-tabs `ContinuationIndentWidth`.
Immediate meshes do not have geometry of type Surface so we check
to see the mesh isn't immediate before trying to cast to surface
to check for octahedral compression
This patch fixes rendering of multiple reflection probes for a single
mesh in the GLES2 renderer. If there were two reflection probes, one of
them would always have the `interior` option flipped, resulting in
broken blending between probes and flickering of ambient reflection.
Also make note of GLES2 reflection probe limit
This provides more realistic lighting with a very small performance cost.
The option is available in both GLES3 and GLES2, and can be enabled in
the Project Settings. This goes well with the ACES Fitted tonemapping mode
that was recently added.
When enabled, this also makes upgrading Godot 3.x projects to Godot 4.0 easier,
since lighting in 3.x will better match how it'll look in Godot 4.0.
Change the existing DEV_ASSERT function to be switched on and off by the DEV_ENABLED define. DEV_ASSERT breaks into the debugger as soon as hit.
Add error macros DEV_CHECK and DEV_CHECK_ONCE to add an alternative check that ERR_PRINT when a condition fails, again only enabled in DEV_ENABLED builds.
Update mesh_surface_get_format_stride and
mesh_surface_make_offsets_from_format to return an array of offsets and
an array of strides in order to support vertex stream splitting
Update _get_array_from_surface to also support vertex stream splitting
Add a condition on split stream usage to ensure it does not get used on
dynamic meshes
Handle case when Tangent is compressed but Normal is not compressed
Make stream splitting option require a restart in the settings
Update SoftBody and Sprite3D to support and use strides and offsets
returned by updated visual_server functions
Update Sprite3D to use the dynamic mesh flag
This backports the high quality glow mode from the `master` branch.
Previously, during downsample, every second row was ignored.
Now, when high-quality is used, we sample two rows at once to ensure
that no pixel is missed. It is slower, but looks much better and has
a much high stability while moving.
High quality also takes an additional horizontal sample the width of the
horizontal blur matches the height of the vertical blur.
With the octahedral compression, we had attributes of a size of 2 bytes
which potentially caused performance regressions on iOS/Mac
Now add padding to the normal/tangent buffer
For octahedral, normal will always be oct32 encoded
UNLESS tangent exists and is also compressed
then both will be oct16 encoded and packed into a vec4<GL_BYTE>
attribute
Implement Octahedral Compression for normal/tangent vectors
*Oct32 for uncompressed vectors
*Oct16 for compressed vectors
Reduces vertex size for each attribute by
*Uncompressed: 12 bytes, vec4<float32> -> vec2<unorm16>
*Compressed: 2 bytes, vec4<unorm8> -> vec2<unorm8>
Binormal sign is encoded in the y coordinate of the encoded tangent
Added conversion functions to go from octahedral mapping to cartesian
for normal and tangent vectors
sprite_3d and soft_body meshes write to their vertex buffer memory
directly and need to convert their normals and tangents to the new oct
format before writing
Created a new mesh flag to specify whether a mesh is using octahedral
compression or not
Updated documentation to discuss new flag/defaults
Created shader flags to specify whether octahedral or cartesian vectors
are being used
Updated importers to use octahedral representation as the default format
for importing meshes
Updated ShaderGLES2 to support 64 bit version codes as we hit the limit
of the 32-bit integer that was previously used as a bitset to store
enabled/disabled flags
Ansraer
lawnjelly
Rémi Verschelde
Omar El Sheikh
Bastiaan Olij
Lucas Ângelo
Hugo Locurcio
Timur Celik
Rémi Verschelde
Haoyu Qiu
20kdc
clayjohn
paru
Yuri Roubinsky