From: Yifan Liu <yifan1.liu@...> When passing through PCI devices we have an option "no_reset", which, if specified, does not do device reset before passthrough. However this option does not tak

Starting from elementpath 4.0.0, the XPath2Parser class no longer has a SYMBOLS member and completeness checking based on that are removed as well. This patch updates the elementpath_overlay in the co

When assigning a physical interrupt to a Post-launched VM, if it has been assigned to ServiceVM, we should remove that mapping first to reset ioapic pin state and rte, and build new mapping for the Po

ptirq_remapping_info records which physical interrupt is mapped to the virtual interrupt in a VM. As we need to knonw whether a physical sid has been mapped to a VM and whether a virtual sid in a VM h

This series fixes hypervisor side PCIe INTx passthru issue occurred when passthru USB xDCI, SMbus controllers and IPU to Post-launched VM. The first patch fixes the hash table implementation used to l

When assigning a physical interrupt to a Post-launched VM, if it has been assigned to ServiceVM, we should remove that mapping first to reset ioapic pin state and rte, and build new mapping for the Po

ptirq_remapping_info records which physical interrupt is mapped to the virtual interrupt in a VM. As we need to knonw whether a physical sid has been mapped to a VM and whether a virtual sid in a VM h

This series fixes hypervisor side PCIe INTx passthru issue occurred when passthru USB xDCI, SMbus controllers and IPU to Post-launched VM. The first patch fixes the hash table implementation used to l

How about move the check/remove/add logic to hypercall.c ?

Will add. Will rename vm_virt_sid_htable to virt_sid_htable Besides, it seems better to remove the encapsulation struct and define hash tables in this common style: struct hlist_head virt_sid_htable[P

This means you could re-assign a ptirq from pre-launched/Service/post-launched VM to any other VM ?

whether a physical IRQ (INTx or MSI/MSI-X) has been assigned to a VM and Which a virtual IRQ been mapped to this physical IRQ in a VM ? Maybe it's better to show all the hlist heads in this picture. v

When assigning a physical interrupt to a Post-launched VM, if it has been assigned to ServiceVM, we should remove that mapping first to reset ioapic pin state and rte, and build new mapping for the Po

ptirq_remapping_info records which physical interrupt is mapped to the virtual interrupt in a VM. As we need to knonw whether a physical sid has been mapped to a VM and whether a virtual sid in a VM h

This series fixes hypervisor side PCIe INTx passthru issue occurred when passthru USB xDCI, SMbus controllers and IPU to Post-launched VM. The first patch fixes the hash table implementation used to l

I think this synchronization has already existed. SOS won't got launched until create_vm finishs setting ptirq_remapping_info for all Prelaunched VMs and loaded_pre_vm_nr reaches PRE_VM_NUM. ```c int3

INTx remappings are just added to Pre-launched VM according to static configurations. In `prepare_vm`, INTx remappings for Pre-launched VMs are handled before ServiceVM starts. So there is no reassign