Background
==========
I have a test environment which runs QEMU 4.2 with a plugin that runs two
copies of a PCIE device simulator on Ubuntu 18.04/CentOS 7.5 host and with an
Ubuntu 18.04 guest.
When running with a single QEMU hw thread/CPU using:
-cpu kvm64,+lahf_lm -M q35,kernel-irqchip=off -device
intel-iommu,intremap=on
Our tests run fine.
But when running with multiple hw threads/cpu's:
2 cores 1 thread per core (2 hw threads/cpus):
-cpu kvm64,+lahf_lm -M q35,kernel-irqchip=off -device
intel-iommu,intremap=on -smp 2,sockets=1,cores=2
1 core, t threads per core (2 hw threads/cpus)
-cpu kvm64,+lahf_lm -M q35,kernel-irqchip=off -device
intel-iommu,intremap=on -smp 2,sockets=1,cores=1
2 cores, 2 threads per core (4 hw threads/cpus):
-cpu kvm64,+lahf_lm -M q35,kernel-irqchip=off -device
intel-iommu,intremap=on -smp 4,sockets=1,cores=2
The values retuned are correct all the way up the call stack and in
KVM_EXIT_MMIO in kvm_cpu_exec (qemu-4.2.0/accel/kvm/kvm-all.c:2365) but the
value returned to the device driver which initiated the read is 0.
I'm currently testing this issue on
Ubuntu 18.04.4 LTS
Kernel: 4.15.0-108-generic
KVM Version: 1:2.11+dfsg-1ubuntu7.28
And:
CentOS: 7.5.1804
Kernel 4.14.78-7.x86_64
KVM Version : 1.5.3
Seeing the same issues in both cases.
Questions
=========
I have the following questions:
Is anyone else running QEMU 4.2 in multi hw thread/cpu mode?
Is anyone getting incorrect reads from memory mapped device registers
when running in this mode?
Does anyone have any pointers on how best to debug the flow from
KVM_EXIT_MMIO back to the device driver running on the guest
