nir/builder: Add nir_umod_imm helper

Like nir_udiv_imm, we can do a similar power-of-two trick. It's also really
convenient.

v2: Assert reasonable bounds on the modulus (Faith).

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl> [v1]
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> [v1]
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22010>

src/compiler/nir/nir_builder.h

@@ -882,6 +882,18 @@ nir_udiv_imm(nir_builder *build, nir_ssa_def *x, uint64_t y)
    }
 }
 
+static inline nir_ssa_def *
+nir_umod_imm(nir_builder *build, nir_ssa_def *x, uint64_t y)
+{
+   assert(y > 0 && y <= u_uintN_max(x->bit_size));
+
+   if (util_is_power_of_two_nonzero(y)) {
+      return nir_iand_imm(build, x, y - 1);
+   } else {
+      return nir_umod(build, x, nir_imm_intN_t(build, y, x->bit_size));
+   }
+}
+
 static inline nir_ssa_def *
 nir_ibfe_imm(nir_builder *build, nir_ssa_def *x, uint32_t offset, uint32_t size)
 {