Like we do for GLSL; there's no reason to abort compilation here. Note
that this also avoids leaking "gen->buffer" and "gen->string_buffers" on
the error path.
Every call to shader_instruction_array_insert_at() means a possible
reallocation of all vsir instructions in the program. This means that all
previous pointers are potentially no longer valid.
We are currently using these potentially invalid pointers in some cases,
usually in the form of "ins->location". This commit fixes these.
I moved all pointer changes to right after the call to
shader_instruction_array_insert_at() to make this more evident.
The following pixel shader currently triggers an infinite loop during
copy propagation, which is fixed by this commit:
sampler s;
Texture2D t1, t2;
float4 main() : sv_target
{
Texture2D t = t1;
t1 = t2;
t2 = t;
return t1.Sample(s, float2(0, 0)) + t2.Sample(s, float2(0, 0));
}
The infinite loop occurs because copy_propagation_transform_object_load()
replaces t1 in the resource_load(t1, ...) instruction
with t2, t1, t2, ... repeatedly.
Note that we still have to preempt the propagation to SM1 pixel shader
uniforms. Otherwise this will turn the many constant derefs that appear
from the <index-val> copy generated in lower_index_loads() into a single
non-constant deref, causing it to allocate all the registers instead of
up until the last one used.
Some SM1 src registers have idx_count = 0, in which case we have to
respect that instead of always reading reg->reg.idx[0].offset even when
it is invalid.
