On Fri, 18 Sept 2026 at 10:25, Yize Wang <[email protected]> wrote: > > This series adds batch register access support to QEMU's KVM GICv3 > register synchronization path, reducing syscall overhead during > ARM64 live migration. > > Currently, QEMU issues one ioctl per register when saving/restoring VGIC > state. On large VM configurations this means tens of thousands of syscalls, > where lock acquisition and context switch overhead dominates migration > downtime. Thus, we provide a batch register method to allow userspace > read/write multiple distributor and redistributor registers in a single call. > In this way, we can significantly reduce syscalls and migration downtime. > > All batch paths probe kernel capability at runtime and fall back to the > original one-by-one path when the kernel does not support batch access, > ensuring backward compatibility. > > This series depends on the corresponding kernel patches that add > KVM_DEV_ARM_VGIC_GRP_BATCH_REGS support. > > Test the VM migration time under pressure conditions. > The VM specifications for migration are as follows: > - VM use 4-K page; > - the number of VCPU is 160; > - the total memory is 320Gigabit; > - use 'Redis SET-benchmark' to pressurize VM; > > Performance results (3-run average, ms): > > | Metric | Without patch | With patch | Improvement | > |---------------------|---------------|------------|-------------| > | Migration downtime | 536 | 321 | 40% | > | Source (total) | 344 | 230 | 33% | > | - VGIC put | 158 | 40 | 75% | > | - VGIC get | 120 | 19 | 84% | > | Destination (total) | 192 | 91 | 53% | > | - VGIC put | 132 | 27 | 80% | > > Yize Wang (4): > Sync KVM ARM VGIC batch register UAPI to QEMU headers > hw/intc/arm_gicv3_kvm: Add batch helper structures and functions > hw/intc/arm_gicv3_kvm: Refactor kvm_arm_gicv3_put for batch register access > hw/intc/arm_gicv3_kvm: Refactor kvm_arm_gicv3_get for batch register access
Just FYI for next time, patchsets like these that are the userspace half of a kernel KVM API change/enhancement that hasn't yet been accepted into the upstream kernel should be marked as "RFC" for the QEMU patchset. thanks -- PMM
