V-Shuttle-S not progressing after seedpool setups

[candymate]

I followed instructions in V-Shuttle-S's readme, with QEMU 5.1.0 and hcd-ohci. I was successfully able to collect seeds with collect_seed.sh script (with Ubuntu 20.04 guest), and used them for fuzz.sh. However, the fuzzer doesn't progress its job after [+] Seedpool 0 setup OK!.

What I tried is...

1. Changed hcd_ohci from ehci, using the ehci patch. However, the same thing happens.
2. Attached gdb each to afl-fuzz threads and QEMU. All afl-fuzz threads but one thread stop at pthread_cond_wait in run_target, and one thread stops at if ((res = read(fsrv_st_fd, &tmp, 4)) != 4) { in main. (Stopped at read) At the same time, QEMU is stopped at read(FORKSRV_ID, &was_killed, 4); in _afl_start of fuzz-seedpool.h, so I suspect this is a deadlock.

I had no problem with setting and running V-Shuttle-M. Are there any solutions for this? Or, am I the only one seeing this phenomenon?

[cyruscyliu]

I don't have this problem. How many seed pools are there in ./in. It should be five IIRC.

[CrowGao]

I hvae same problem. Have you solved it?