Specified a vector size of 3 but function signature generated a vector size of 4

[nkoskelo]

import numpy as np
import loopy as lp
import pyopencl as cl

ctx = cl.create_some_context()
queue = cl.CommandQueue(ctx)

n = 3**4
a = 2
b = 5

knl = lp.make_kernel(
     "{ [i]: 0<=i<n }",
     """
     out[i] = a if i == 0 else b
     """
)
vec_size = 3
knl = lp.split_array_axis(knl, "out", 0, vec_size)
knl = lp.split_iname(knl, "i", vec_size)
knl = lp.tag_inames(knl, {"i_inner": "vec"})
knl = lp.tag_array_axes(knl, "out", "c,vec")
knl = lp.set_options(knl, write_code=True)
knl = lp.assume(knl, f"n % {vec_size} = 0 and n>0")

knl = lp.add_and_infer_dtypes(knl, {"a": np.dtype(np.float32), "b": np.dtype(np.float32)})

device_code = lp.generate_code_v2(knl).device_code()

assert "float3" in device_code

Instead of float3 you will get float4