diff options
author | Eric Anholt <[email protected]> | 2010-09-22 11:47:03 -0700 |
---|---|---|
committer | Eric Anholt <[email protected]> | 2010-09-22 13:09:51 -0700 |
commit | b39e6f33b60ef9bbaf81f320aaca6a440d8a6a8f (patch) | |
tree | c3547819e402b729d84a4847aed95b81ec09fdbf /src/glsl/ir_constant_propagation.cpp | |
parent | 38da5c9cb636387539daaf5688c2a3badee32447 (diff) |
glsl: Rework assignments with write_masks to have LHS chan count match RHS.
It turns out that most people new to this IR are surprised when an
assignment to (say) 3 components on the LHS takes 4 components on the
RHS. It also makes for quite strange IR output:
(assign (constant bool (1)) (x) (var_ref color) (swiz x (var_ref v) ))
(assign (constant bool (1)) (y) (var_ref color) (swiz yy (var_ref v) ))
(assign (constant bool (1)) (z) (var_ref color) (swiz zzz (var_ref v) ))
But even worse, even we get it wrong, as shown by this line of our
current step(float, vec4):
(assign (constant bool (1)) (w)
(var_ref t)
(expression float b2f (expression bool >=
(swiz w (var_ref x))(var_ref edge))))
where we try to assign a float to the writemasked-out x channel and
don't supply anything for the actual w channel we're writing. Drivers
right now just get lucky since ir_to_mesa spams the float value across
all the source channels of a vec4.
Instead, the RHS will now have a number of components equal to the
number of components actually being written. Hopefully this confuses
everyone less, and it also makes codegen for a scalar target simpler.
Reviewed-by: Kenneth Graunke <[email protected]>
Reviewed-by: Ian Romanick <[email protected]>
Diffstat (limited to 'src/glsl/ir_constant_propagation.cpp')
-rw-r--r-- | src/glsl/ir_constant_propagation.cpp | 16 |
1 files changed, 12 insertions, 4 deletions
diff --git a/src/glsl/ir_constant_propagation.cpp b/src/glsl/ir_constant_propagation.cpp index f7a0599f425..390f295d6e9 100644 --- a/src/glsl/ir_constant_propagation.cpp +++ b/src/glsl/ir_constant_propagation.cpp @@ -168,18 +168,26 @@ ir_constant_propagation_visitor::handle_rvalue(ir_rvalue **rvalue) if (!found) return; + int rhs_channel = 0; + for (int j = 0; j < 4; j++) { + if (j == channel) + break; + if (found->write_mask & (1 << j)) + rhs_channel++; + } + switch (type->base_type) { case GLSL_TYPE_FLOAT: - data.f[i] = found->constant->value.f[channel]; + data.f[i] = found->constant->value.f[rhs_channel]; break; case GLSL_TYPE_INT: - data.i[i] = found->constant->value.i[channel]; + data.i[i] = found->constant->value.i[rhs_channel]; break; case GLSL_TYPE_UINT: - data.u[i] = found->constant->value.u[channel]; + data.u[i] = found->constant->value.u[rhs_channel]; break; case GLSL_TYPE_BOOL: - data.b[i] = found->constant->value.b[channel]; + data.b[i] = found->constant->value.b[rhs_channel]; break; default: assert(!"not reached"); |