Upstream patch:
https://listman.redhat.com/archives/libvir-list/2023-March/238702.html

Pushed upstream as:
commit d5c7b7870e45575f81fffcb611c2546d0e02e778
Author:     Ján Tomko <jtomko>
CommitDate: 2023-03-14 17:10:01 +0100

    qemu: relax shared memory check for vhostuser daemons
    
    For some vhostuser daemons, we validate that the guest memory is shared
    with the host.
    
    With earlier versions of QEMU, it was only possible to mark memory
    as shared by defining an explicit NUMA topology.  Later, QEMU exposed
    the name of the default memory backend (defaultRAMid) so we can mark
    that memory as shared.
    
    Since libvirt commit:
      commit bff2ad5d6b1f25da02802273934d2a519159fec7
        qemu: Relax validation for mem->access if guest has no NUMA
    we already check for the case when user requests shared memory,
    but QEMU did not expose defaultRAMid.
    
    Drop the duplicit check from vhostuser device validation, to make
    it pass on hotplug even after libvirtd restart.
    
    This avoids the need to store the defaultRAMid, since we don't really
    need it for anything after the VM has been already started.
    
    https://bugzilla.redhat.com/show_bug.cgi?id=2078693
    https://bugzilla.redhat.com/show_bug.cgi?id=2177701
    
    Signed-off-by: Ján Tomko <jtomko>
    Reviewed-by: Michal Privoznik <mprivozn>

git describe: v9.1.0-200-gd5c7b7870e

Preverified on upstream build: libvirt v9.1.0-248-g27d8bcc337

Test steps:
1.Start a guest with shared memory:
...
<memoryBacking>
    <access mode='shared'/>
  </memoryBacking>
...

2.Prepare a virtiofs device xml:
#cat fs.xml
<filesystem accessmode="passthrough" type="mount">
	<driver queue="512" type="virtiofs"/>
	<source dir="/var/tmp/mount_tag0"/>
	<target dir="mount_tag0"/>
	<alias name="ua-05e63df0-c4ab-11ec-ba3d-fa163ed3db7d"/>
	<binary path="/usr/libexec/virtiofsd" xattr="on">
		<cache mode="none"/>
               <lock flock="off" posix="off"/>
</binary>
</filesystem>


3.Create the monut dir:
#mkdir /var/tmp/mount_tag0

4.Attach the device to the vm:
# virsh attach-device vm1 fs.xml 
Device attached successfully

5.Detach virtiofs device from the vm:
# virsh detach-device vm1 fs.xml 
Device detached successfully

6.Restart virtqemud:
#systemctl restart virtqemud

7.Attach virtiofs devcie to the vm again:
# virsh attach-device vm1 fs.xml 
Device attached successfully

8. Detach virtiofs device from the vm again:
# virsh detach-device vm1 fs.xml 
Device detached successfully

A similar ticket for RHEL8 can be found here: https://bugzilla.redhat.com/show_bug.cgi?id=2177701

Verified on upstream build:
# rpm -q libvirt qemu-kvm
libvirt-9.3.0-2.el9.x86_64
qemu-kvm-8.0.0-3.el9.x86_64

Test steps:
1.Start a guest with shared memory:
...
<memoryBacking>
    <access mode='shared'/>
  </memoryBacking>
...

2.Prepare a virtiofs device xml:
#cat fs.xml
<filesystem accessmode="passthrough" type="mount">
	<driver queue="512" type="virtiofs"/>
	<source dir="/var/tmp/mount_tag0"/>
	<target dir="mount_tag0"/>
	<alias name="ua-05e63df0-c4ab-11ec-ba3d-fa163ed3db7d"/>
	<binary path="/usr/libexec/virtiofsd" xattr="on">
		<cache mode="none"/>
               <lock flock="off" posix="off"/>
</binary>
</filesystem>


3.Create the monut dir:
#mkdir /var/tmp/mount_tag0

4.Attach the device to the vm:
# virsh attach-device vm1 fs.xml 
Device attached successfully

5.Detach virtiofs device from the vm:
# virsh detach-device vm1 fs.xml 
Device detached successfully

6.Restart virtqemud:
#systemctl restart virtqemud

7.Attach virtiofs devcie to the vm again:
# virsh attach-device vm1 fs.xml 
Device attached successfully

8. Detach virtiofs device from the vm again:
# virsh detach-device vm1 fs.xml 
Device detached successfully

It's a per-VM issue. When starting the VM, libvirt probes the QEMU capabilities, including whether it exposes the alias of the default memory device. Libvirt uses the alias to tell QEMU to mark the guest memory as shared. For older QEMU/libvirt, other methods of creating shared memory are used:
https://libvirt.org/kbase/virtiofs.html#other-options-for-vhost-user-memory-setup

Libvirt does not save the memory alias in the VM's status XML. After daemon restart, the QEMU capabilities are not probed again, but read from the status XML. The bug here is that the validation required either a memory alias, or one of the alternate ways of having shared memory. The fix removed the requirement for a memory alias, because this is already validated at VM startup.

The "workarounds" here are
a) do not restart the daemon
b) use a VM with one of the alternate setups of shared memory
c) restart the VM
to be able to hotplug a virtiofs <filesystem>

However none of them are applicable if you already have a running VM and restarted the daemon.

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (Moderate: libvirt security, bug fix, and enhancement update), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHSA-2023:6409