broadcom: implement nir_op_uclz

This enables some algebraic optimizations.

No changes in shader-db, but it does cause some CTS tests to
produce less instructions.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30614>

src/broadcom/compiler/nir_to_vir.c

@@ -1615,6 +1615,10 @@ ntq_emit_alu(struct v3d_compile *c, nir_alu_instr *instr)
                 result = vir_NOT(c, src[0]);
                 break;
 
+        case nir_op_uclz:
+                result = vir_CLZ(c, src[0]);
+                break;
+
         case nir_op_ufind_msb:
                 result = vir_SUB(c, vir_uniform_ui(c, 31), vir_CLZ(c, src[0]));
                 break;

src/broadcom/vulkan/v3dv_pipeline.c

@@ -220,6 +220,7 @@ v3dv_pipeline_get_nir_options(const struct v3d_device_info *devinfo)
       .lower_fquantize2f16 = true,
       .has_fsub = true,
       .has_isub = true,
+      .has_uclz = true,
       .vertex_id_zero_based = false, /* FIXME: to set this to true, the intrinsic
                                       * needs to be supported */
       .lower_interpolate_at = true,