I recently came across a number of PMU emulation bugs, all which can result in unexpected behaviours in an unsuspecting guest. The first two patches already have been discussed on the list, but I'm including them here as part of a slightly longer series. The third patch is new as of v2, and fixes a bug preventing chained events from ever being used. The last patch fixes an issue that has been here from day one, where we confuse architectural overflow of a counter and perf sampling period, and uses a terrible hack^W^W creative way to interact with the underlying PMU driver so that we can reload the period when handling the overflow. * From v2 [1] - Dropped PMUv3 patch and moved the logic into the KVM code - Properly use UPDATE/RELOAD - Collected Andrew's RB [1] https://lore.kernel.org/kvmarm/20191008160128.8872-1-maz@xxxxxxxxxx/ Marc Zyngier (4): KVM: arm64: pmu: Fix cycle counter truncation arm64: KVM: Handle PMCR_EL0.LC as RES1 on pure AArch64 systems KVM: arm64: pmu: Set the CHAINED attribute before creating the in-kernel event KVM: arm64: pmu: Reset sample period on overflow handling arch/arm64/kvm/sys_regs.c | 4 ++++ virt/kvm/arm/pmu.c | 48 ++++++++++++++++++++++++++++----------- 2 files changed, 39 insertions(+), 13 deletions(-) -- 2.20.1 _______________________________________________ kvmarm mailing list kvmarm@xxxxxxxxxxxxxxxxxxxxx https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
