Debugging with Intel® Distribution for GDB*¶
Numba-dpex allows you to debug SYCL* kernels with Intel® Distribution for GDB*.
To enable the emission of debug information, set the debug environment variable
NUMBA_DPEX_DEBUGINFO, for example:
export NUMBA_DPEX_DEBUGINFO=1
To disable debugging, unset the variable:
unset NUMBA_DPEX_DEBUGINFO
Note
Enabling debug information significantly increases the memory consumption for each compiled kernel. For a large application, this may cause out-of-memory error.
Not all debugging features supported by Numba on CPUs are yet supported by numba-dpex. See Features and Limitations.
Requirements¶
Intel® Distribution for GDB* is required for numba-dpex debugging features to work. Intel® Distribution for GDB* is part of Intel oneAPI. For relevant documentation, refer to the Intel® Distribution for GDB* product page.
Example of Intel® Distribution for GDB* usage¶
You can use a sample kernel code, simple_sum.py, for basic debugging:
5import dpnp as np
6
7import numba_dpex as ndpx
8
9
10@ndpx.kernel(debug=True)
11def data_parallel_sum(item, a, b, c):
12 i = item.get_id(0)
13 c[i] = a[i] + b[i] # Condition breakpoint location
14
15
16global_size = 10
17N = global_size
18
19a = np.array(np.random.random(N), dtype=np.float32)
20b = np.array(np.random.random(N), dtype=np.float32)
21c = np.ones_like(a)
22
23ndpx.call_kernel(data_parallel_sum, ndpx.Range(global_size), a, b, c)
24
25print("Done...")
Use the following commands to create a breakpoint inside the kernel and run the debugger:
$ NUMBA_OPT=0 gdb-oneapi -q python
(gdb) set breakpoint pending on
(gdb) break simple_sum.py:22
(gdb) run simple_sum.py
...
Thread 2.2 hit Breakpoint 1, with SIMD lanes [0-7], __main__::data_parallel_sum () at simple_sum.py:22
22 i = dpex.get_global_id(0)
(gdb) next
Thread 2.3 hit Breakpoint 1, with SIMD lanes [0-1], __main__::data_parallel_sum () at simple_sum.py:22
22 i = dpex.get_global_id(0)
(gdb) next
23 c[i] = a[i] + b[i]
(gdb) continue
...
Done...