Cannot print field in kernel

[lucifer1004]

import taichi as ti


@ti.data_oriented
class B():
    def __init__(self):
        self.c = ti.field(dtype=float, shape=(5, 5))


@ti.data_oriented
class A():
    def __init__(self):
        self.b = B()

    @ti.kernel
    def print(self):
        print(self.b.c)


ti.init()
a = A()
a.print()

will cause the following error:

print(self.b.c)
        ^^^^^^^^^^^^^^^
Invalid constant scalar data type: <class 'taichi.lang.field.ScalarField'>

However, both this

import taichi as ti


@ti.data_oriented
class B():
    def __init__(self):
        self.c = ti.field(dtype=float, shape=(5, 5))


@ti.data_oriented
class A():
    def __init__(self):
        self.b = B()

    def print(self):
        print(self.b.c)


ti.init()
a = A()
a.print()

and this

@ti.data_oriented
class B():
    def __init__(self):
        self.c = ti.field(dtype=float, shape=(5, 5))


@ti.data_oriented
class A():
    def __init__(self):
        self.b = B()

    @ti.kernel
    def print(self):
        for i in ti.grouped(self.b.c):
            print(i)


ti.init()
a = A()
a.print()

can work.

[lin-hitonami]

If I remember correctly we don't support printing a field in a kernel...

[lucifer1004]

So what is the obstacle here? I think it should be equivalent to printing each element of a field, which is valid.

[lucifer1004]

I have confirmed that this is not related to data_oriented. The following cannot work, either.

import taichi as ti
ti.init()

a = ti.field(ti.f32, shape=(4, 4))

@ti.kernel
def f():
    print(a)

f()

[jim19930609]

Directly printing a field in Taichi scope hasn't been supported yet. There are some performance concerns especially in terms of sparse fields such as Pointer SNode, Dynamic SNode, etc.

However, we're considering supporting this feature in the future.

[kurt-rhee]

This also does not work for ndarrays

ti.init()
test = ti.ndarray(dtype=ti.int32, shape=4)
test[2] = 5
test_type = ti.types.ndarray(dtype=ti.int32, ndim=1)

@ti.kernel
def do_test(
    t: test_type
):
    do_test_internal(t)

@ti.func
def do_test_internal(t:test_type):
    print(t)


do_test(t=test)

[Taichi] version 1.6.0, llvm 15.0.1, commit f1c6fbbd, win, python 3.10.4
[Taichi] Starting on arch=x64
Traceback (most recent call last):
  File "C:\Program Files\JetBrains\PyCharm Community Edition 2023.1.3\plugins\python-ce\helpers\pydev\pydevconsole.py", line 364, in runcode
    coro = func()
  File "<input>", line 1, in <module>
  File "C:\Program Files\JetBrains\PyCharm Community Edition 2023.1.3\plugins\python-ce\helpers\pydev\_pydev_bundle\pydev_umd.py", line 197, in runfile
    pydev_imports.execfile(filename, global_vars, local_vars)  # execute the script
  File "C:\Program Files\JetBrains\PyCharm Community Edition 2023.1.3\plugins\python-ce\helpers\pydev\_pydev_imps\_pydev_execfile.py", line 18, in execfile
    exec(compile(contents+"\n", file, 'exec'), glob, loc)
  File "C:\Users\krhee\PycharmProjects\shade-engine\test_garbage.py", line 25, in <module>
    do_test(t=test)
  File "C:\Users\krhee\PycharmProjects\shade-engine\venv\lib\site-packages\taichi\lang\kernel_impl.py", line 976, in wrapped
    raise type(e)("\n" + str(e)) from None
taichi.lang.exception.TaichiTypeError: 
File "C:\Users\krhee\PycharmProjects\shade-engine\test_garbage.py", line 18, in do_test:
    do_test_internal(t)
    ^^^^^^^^^^^^^^^^^^^
File "C:\Users\krhee\PycharmProjects\shade-engine\test_garbage.py", line 22, in do_test_internal:
    print(t)
    ^^^^^^^^
Invalid constant scalar data type: <class 'taichi.lang.any_array.AnyArray'>