Hi All, Is it possible to setup a large Windows VM (say 512 vcpus) without adding viommu (EIM=on, IR=on)? When I try to power such VM, the qemu process crashes with error- ``` qemu-kvm: ../accel/kvm/kvm-all.c:1837: kvm_irqchip_commit_routes: Assertion `ret == 0’ failed
``` Stack trace- ``` #1 0x00007f484bc21ea5 abort (libc.so.6) #2 0x00007f484bc21d79 __assert_fail_base.cold.0 (libc.so.6) #3 0x00007f484bc47426 __assert_fail (libc.so.6) #4 0x000055b7215634d3 kvm_irqchip_commit_routes (qemu-kvm) #5 0x000055b7213bfc7e kvm_virtio_pci_vector_use_one (qemu-kvm) #6 0x000055b7213c02cf virtio_pci_set_guest_notifiers (qemu-kvm) #7 0x000055b7214dd848 vhost_scsi_common_start (qemu-kvm) #8 0x000055b72139b936 vhost_user_scsi_start (qemu-kvm) #9 0x000055b72139ba64 vhost_user_scsi_set_status (qemu-kvm) #10 0x000055b7214f865a virtio_set_status (qemu-kvm) #11 0x000055b7213bdc3f virtio_pci_common_write (qemu-kvm) #12 0x000055b721514e68 memory_region_write_accessor (qemu-kvm) #13 0x000055b72151489e access_with_adjusted_size (qemu-kvm) #14 0x000055b721514b89 memory_region_dispatch_write (qemu-kvm) #15 0x000055b72151e3fc flatview_write_continue (qemu-kvm) #16 0x000055b72151e553 flatview_write (qemu-kvm) #17 0x000055b72151ee76 address_space_write (qemu-kvm) #18 0x000055b721565526 kvm_cpu_exec (qemu-kvm) #19 0x000055b72156634d kvm_vcpu_thread_fn (qemu-kvm) #20 0x000055b721750224 qemu_thread_start (qemu-kvm) #21 0x00007f484c0081ca start_thread (libpthread.so.0) #22 0x00007f484bc39e73 ``` The error is due to invalid MSIX routing entry passed to KVM. The VM boots fine if we attach a vIOMMU but adding a vIOMMU can potentially result in IO performance loss in guest. I was interested to know if someone could boot a large Windows VM by some other means like kvm-msi-ext-dest-id. Overheads of viommu have been shown for example in - https://static.sched.com/hosted_files/kvmforum2021/da/vIOMMU%20KVM%20Forum%202021%20-%20v4.pdf Thanks and regards, Sandesh
