On 11/23/19 11:56 AM, Marc Zyngier wrote: > HCR_EL2.TID3 mandates that access from EL1 to a long list of id > registers traps to EL2, and QEMU has so far ignored this requirement. > > This breaks (among other things) KVM guests that have PtrAuth enabled, > while the hypervisor doesn't want to expose the feature to its guest. > To achieve this, KVM traps the ID registers (ID_AA64ISAR1_EL1 in this > case), and masks out the unsupported feature. > > QEMU not honoring the trap request means that the guest observes > that the feature is present in the HW, starts using it, and dies > a horrible death when KVM injects an UNDEF, because the feature > *really* isn't supported. > > Do the right thing by trapping to EL2 if HCR_EL2.TID3 is set. > > Reported-by: Will Deacon <will@xxxxxxxxxx> > Signed-off-by: Marc Zyngier <maz@xxxxxxxxxx> > --- > There is a number of other trap bits missing (TID[0-2], for example), > but this at least gets a mainline Linux going with cpu=max. BTW, Peter, this appears to have been the bug that was causing me so many problems on my VHE branch. Probably *exactly* this bug wrt ptrauth, since that would also be included with -cpu max. I am now able to boot a kvm guest kernel to the point of the no rootfs panic, which I wasn't before. I can only think that I mis-identified the true cause in Lyon. Anyway, thanks Marc! r~ _______________________________________________ kvmarm mailing list kvmarm@xxxxxxxxxxxxxxxxxxxxx https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
