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

Set the guest with single host cpu pinned, then start the guest, wait for 3 seconds then found the guest is crash with core dump.

Please provide the package NVR for which the bug is seen:

beaker host: fujitsu-fx700-01-n00.khw.eng.bos2.dc.redhat.com

host packages versions:

# rpm -q libvirt qemu-kvm kernel-64k
libvirt-10.9.0-1.el9.aarch64
qemu-kvm-9.1.0-1.el9.aarch64
kernel-64k-5.14.0-527.el9.aarch64

guest kernel: 5.14.0-524.el9.aarch64+64k

How reproducible is this bug?:  5%

Steps to reproduce

1. Config the guest like below libvirt xml:

<memory unit='KiB'>4194304</memory>
<currentMemory unit='KiB'>4194304</currentMemory>
<vcpu placement='static' cpuset='0'>4</vcpu>

...

2. Start the guest

# virsh start avocado-vt-vm1
Domain 'avocado-vt-vm1' started

3. Wait for 3 seconds then found the guest is shut off, and there is core dump for guest

# virsh list --all
 Id   Name             State
---------------------------------
 -    avocado-vt-vm1   shut off

core dump back trace is like below:

(gdb) bt
#0  0x0000ffff990c23c8 in __pthread_kill_implementation () from /lib64/libc.so.6
#1  0x0000ffff9907a6bc in raise () from /lib64/libc.so.6
#2  0x0000ffff99066fb4 in abort () from /lib64/libc.so.6
#3  0x0000ffff99074010 in __assert_fail_base () from /lib64/libc.so.6
#4  0x0000ffff99074080 in __assert_fail () from /lib64/libc.so.6
#5  0x0000aaaad827f89c in render_memory_region ()
#6  0x0000aaaad827f4dc in render_memory_region ()
#7  0x0000aaaad827f59c in render_memory_region ()
#8  0x0000aaaad827f004 in generate_memory_topology ()
#9  0x0000aaaad8278334 in memory_region_transaction_commit ()
#10 0x0000aaaad7ed1848 in pci_bridge_write_config ()
#11 0x0000aaaad7ed9d2c in rp_write_config ()
#12 0x0000aaaad8279828 in memory_region_write_accessor ()
#13 0x0000aaaad8279654 in access_with_adjusted_size ()
#14 0x0000aaaad82793e4 in memory_region_dispatch_write ()
#15 0x0000aaaad828cdfc in flatview_write_continue_step ()
#16 0x0000aaaad8286fbc in flatview_write ()
#17 0x0000aaaad8286e50 in address_space_write ()
#18 0x0000aaaad82dc398 in kvm_cpu_exec ()
#19 0x0000aaaad82e1c5c in kvm_vcpu_thread_fn ()
#20 0x0000aaaad848ff74 in qemu_thread_start ()
#21 0x0000ffff990c0778 in start_thread () from /lib64/libc.so.6
#22 0x0000ffff9912ad5c in thread_start () from /lib64/libc.so.6

Expected results

Guest should not crash.

Actual results

Guest crashed with core dump file.

Additional info

This issue is not reproducible if ping guest to multiple host cpus.

This issue is not reproducible on server "ampere-mtjade-altra.." or "nvidia-grace-grace.."