env.CodeGenerate(
target = 'my_source.c',
script = 'my_generator.py',
source = ['input.txt', 'another.txt'],
command = 'python $SCRIPT $SOURCE > $TARGET'
)
It will take care generating all appropriate dependencies, including any
module imported by the generator script, and the respective .pyc file
side effects.
It's possible to call update_samplers() between the time a fragment shader
is bound and when a texture image is defined (such as glClear). This
fixes the case where we don't have a complete texture object yet.
Need to translate VERT_RESULT_PSIZ, BFC0, BFC1 to TGSI shader output slots
after all other attributes have been handled. This fixes a bug where
generic vertex program outputs (varying vars) could get mapped to the
same slot at point size or back-face colors.
If the 'shader' parameter is wrong, need to either generate GL_INVALID_VALUE
or GL_INVALID_OPERATION. It depends on whether 'shader' actually names a
'program' or is a totally unknown ID.
There might be other cases to fix...
cherry-picked from master
This fixes a failure for cases like:
vec4 v;
v[1] *= 2.0;
The v[1] actually acts like a writemask, equivalent to v.y
The fix is a bit convoluted, but will do for now.
cherry-picked from master
Before, the presence of a 'return' statement always prevented inlining
a function. This was because we didn't want to accidentally return from
the _calling_ function. We still need the semantic of 'return' when inlining
but we can't always use unconditional branches/jumps (GPUs don't always
support arbitrary branching).
Now, we allow inlining functions w/ return if the return is the last
statement in the function. This fixes the common case of a function
that returns a value, such as:
vec4 square(const in vec4 x)
{
return x * x;
}
which effectively compiles into:
vec4 square(const in vec4 x)
{
__retVal = x * x;
return;
}
The 'return' can be no-op'd now and we can inline the function.
cherry-picked from master
