nir: Insert b2b1s around booleans in nir_lower_to

By inserting a b2b1 around the load_ubo, load_input, etc. intrinsics
generated by nir_lower_io, we can ensure that the intrinsic has the
correct destination bit size.  Not having the right size can mess up
passes which try to optimize access.  In particular, it was causing
brw_nir_analyze_ubo_ranges to ignore load_ubo of booleans which meant
that booleans uniforms weren't getting pushed as push constants.  I
don't think this is an actual functional bug anywhere hence no CC to
stable but it may improve perf somewhere.

Shader-db results on ICL with iris:

    total instructions in shared programs: 16076707 -> 16075246 (<.01%)
    instructions in affected programs: 129034 -> 127573 (-1.13%)
    helped: 487
    HURT: 0
    helped stats (abs) min: 3 max: 3 x̄: 3.00 x̃: 3
    helped stats (rel) min: 0.45% max: 3.00% x̄: 1.33% x̃: 1.36%
    95% mean confidence interval for instructions value: -3.00 -3.00
    95% mean confidence interval for instructions %-change: -1.37% -1.29%
    Instructions are helped.

    total cycles in shared programs: 338015639 -> 337983311 (<.01%)
    cycles in affected programs: 971986 -> 939658 (-3.33%)
    helped: 362
    HURT: 110
    helped stats (abs) min: 1 max: 1664 x̄: 97.37 x̃: 43
    helped stats (rel) min: 0.03% max: 36.22% x̄: 5.58% x̃: 2.60%
    HURT stats (abs)   min: 1 max: 554 x̄: 26.55 x̃: 18
    HURT stats (rel)   min: 0.03% max: 10.99% x̄: 1.04% x̃: 0.96%
    95% mean confidence interval for cycles value: -79.97 -57.01
    95% mean confidence interval for cycles %-change: -4.60% -3.47%
    Cycles are helped.

    total sends in shared programs: 815037 -> 814550 (-0.06%)
    sends in affected programs: 5701 -> 5214 (-8.54%)
    helped: 487
    HURT: 0

    LOST:   2
    GAINED: 0

The two lost programs were SIMD16 shaders in CS:GO.  However, CS:GO was
also one of the most helped programs where it shaves sends off of 134
programs.  This seems to reduce GPU core clocks by about 4% on the first
1000 frames of the PTS benchmark.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4338>

diff --git a/src/compiler/nir/nir_lower_io.c b/src/compiler/nir/nir_lower_io.c
index 92d2a1f8ba0f89853947c918d7c496d0cc761249..c127802a59d89e9c2c97d377879d9a2a828b1d69 100644 (file)
--- a/src/compiler/nir/nir_lower_io.c
+++ b/src/compiler/nir/nir_lower_io.c
@@ -352,6 +352,13 @@ lower_load(nir_intrinsic_instr *intrin, struct lower_io_state *state,
       }
 
       return nir_vec(b, comp64, intrin->dest.ssa.num_components);
+   } else if (intrin->dest.ssa.bit_size == 1) {
+      /* Booleans are 32-bit */
+      assert(glsl_type_is_boolean(type));
+      return nir_b2b1(&state->builder,
+                      emit_load(state, vertex_index, var, offset, component,
+                                intrin->dest.ssa.num_components, 32,
+                                nir_type_bool32));
    } else {
       return emit_load(state, vertex_index, var, offset, component,
                        intrin->dest.ssa.num_components,
@@ -445,6 +452,14 @@ lower_store(nir_intrinsic_instr *intrin, struct lower_io_state *state,
          write_mask >>= num_comps;
          offset = nir_iadd_imm(b, offset, slot_size);
       }
+   } else if (intrin->dest.ssa.bit_size == 1) {
+      /* Booleans are 32-bit */
+      assert(glsl_type_is_boolean(type));
+      nir_ssa_def *b32_val = nir_b2b32(&state->builder, intrin->src[1].ssa);
+      emit_store(state, b32_val, vertex_index, var, offset,
+                 component, intrin->num_components,
+                 nir_intrinsic_write_mask(intrin),
+                 nir_type_bool32);
    } else {
       emit_store(state, intrin->src[1].ssa, vertex_index, var, offset,
                  component, intrin->num_components,