What were you trying to do that didn't work?

When do zero copy migration through exec, the qemu core dump during migration

Please provide the package NVR for which bug is seen:

host info: kernel-5.14.0-411.el9.x86_64 && qemu-kvm-8.2.0-3.el9.x86_64
guest info: RHEL 9.4.0

How reproducible:

100%

Steps to reproduce
1. Boot a guest
2. Stop the guest: 

{"execute": "stop", "id": "li2cUdzQ"}

3. Enable multifd and zero copy:  

{"execute": "migrate-set-capabilities", "arguments": {"capabilities": [{"capability": "multifd", "state": true}]}, "id": "vfRx8k0c"}
{"execute": "migrate-set-capabilities", "arguments": {"capabilities": [{"capability": "zero-copy-send", "state": true}]}, "id": "IjL8TblI"}

4. Start zero copy migration through exec

{"execute": "migrate", "arguments":{"uri": "exec:gzip -c > //home/ipa/STATEFILE.gz"}, "id": "IqbEdS0U"}  

Expected results

After step 4, get migration failed when send the migrate command or when query migrate status

Actual results

After step 4, the qemu core dump during migration

Additional info:
1. Also hit this issue on RHEL 9.4.0 aarch64
2. As layered products don't support this scenario, so the user can't reproduce. But we need to fix the qemu core dump issue. so set the priority to be minor and add qe_no_customer_impact label.