Debugging a crash when trying to run taichi 1.7.4 on local kernel

[lhphanto]

I am trying to use taichi running on my local kernel using notebook, I used taichi 1.7.4.
I tried to run

import taichi as ti
import taichi.math as tm

ti.init(arch=ti.cpu)

n = 32
pixels = ti.field(dtype=float, shape=(n * 2, n))

But then the notebook hangs and give the following error

[platform.cpp:79] register_pal() called multiple times. Subsequent calls will override the previous implementation. Previous implementation was registered from ../executorch/extension/pybindings/pybindings.cpp.
0.00s - Debugger warning: It seems that frozen modules are being used, which may
0.00s - make the debugger miss breakpoints. Please pass -Xfrozen_modules=off
0.00s - to python to disable frozen modules.
0.00s - Note: Debugging will proceed. Set PYDEVD_DISABLE_FILE_VALIDATION=1 to disable this validation.
kernel lxh_ml_xx on Python 3.12.12+ (3.12:4a5632f, Oct 09 2025, 11:06:46) [Clang 19.1.2 
I0109 13:36:45.875609 3728130 GojiraLibrary.cpp:218] 0D4506E7266AF4943FFC309E2BE14391 :: AnchorClient.whoami() found no active context.
I0109 13:36:45.876331 3728130 AnchorClient.cpp:796] local context couldn't be bootstrapped independently
...
[...] Starting the kernel at pid: 3728130
[...] To connect another client to this kernel, use:
[...]     --existing kernel-4b2189f2-ef01-43ce-bd26-27e663d54092.json
[...] Loading IPython extension: storemagic
...
munmap_chunk(): invalid pointer

It was running on a x86_64 CPU machine , CentOS
Seeking some help to figure out what could go wrong...

Thanks!