Yet another AMD reset bug thread (RX470)

Hi all, this is my first post here after a lot of lurking.

I have an RX470 card passed through to a Linux guest, and I am experience the reset issue (unknown pci header 127). I have tried several solutions from here, Reddit, and SE sites, but so far I have not been able to use the card after shutting down the VM it belongs to. Previously I was using Ubuntu Server 18.04.2, but because some threads led me to think that an updated kernel and/or qemu would help, I am now using Arch with kernel 4.19 and qemu 3.1. Unfortunately that did not work for me.
I have tried the eject – rescan method, and that did not work either.

echo "1" > /sys/bus/pci/devices/09:00.0/remove
echo "1" > /sys/bus/pci/devices/09:00.1/remove
echo "1" > /sys/bus/pci/rescan

This does remove the device (it no longer shows with lspci), but rescan does not work.

Also, I’d just like to give a big thanks to these forums in helping me get the GTX1070 I have passed to my Windows VM. I read numerous threads here in the course of getting that sorted out.

I also tried the less desriable workaround that turns off idling, but no luck with that either:
I created /etc/modprobe.d/vfio.conf and then put in the line

options vfio_pci ids=1002:67df,1002:aaf0 disable_idle_d3=1

The result of that was that main screen for my host was moved over to a monitor connected to the GTX1070 (the one I use for Windows) and no longer the budget NVIDIA card (Zotac GT710). That card (which is in the first slot) then only displayed command line stuff from the boot process.

So, are there any other possible solutions I am missing?