[BUG] Jetson thor use run-iket output code -6

Which component has the problem?

CuTe DSL

Bug Report

Describe the bug
I want to use iket to profile examples/python/CuTeDSL/cute/blackwell/kernel/dense_gemm/dense_gemm_persistent.py,
I run with code

CUTE_DSL_COMPILER_OPT=iket run-iket \
 --output-dir ./persistent_iket profile \
--postprocess html -- \
python dense_gemm_persistent.py \
--mnkl 1543,2560,2048,1 \
--ab_dtype Int8 \
--acc_dtype Int32 \
--c_dtype Float16 \
--a_major k \
--b_major k \
--c_major n \
--mma_tiler_mn 256,256 \
--cluster_shape_mn 2,1 \
--use_2cta_instrs \
--use_tma_store \
--benchmart default \
--skip_ref_check

And I got Error:

[run-iket] ============= Dry run application first to collect some info... =============
[run-iket] Launch GPU workload exits abnormally with return code: -6 => Tool: tracker, Job: tracker
[run-iket] Please run "/root/w8a8_pergroup/persistent_iket/tracker/rerun.sh" to reproduce the issue!
[run-iket] Tracker failed with return code: -6

Steps/Code to reproduce bug
Use upper code can reproduce this bug

Environment details (please complete the following information):
my env is
Jetson thor T5000 sm_110
Jetson 7.2
driver version: 595.78
docker image: nvidia/pytorch:26.05-py3
cuda: 13.2
cutedsl: 4.6.1

Additional context
Add any other context about the problem here.

Start by running the provided command and then the generated tracker/rerun.sh for examples/python/CuTeDSL/cute/blackwell/kernel/dense_gemm/dense_gemm_persistent.py in the stated Jetson Thor environment. Inspect why the tracker exits with code -6; done means the profiling run completes and produces the requested HTML output.