> -----Original Message----- > From: Marc Zyngier [mailto:maz@xxxxxxxxxx] > Sent: 05 March 2021 18:53 > To: Paolo Bonzini <pbonzini@xxxxxxxxxx> > Cc: Alexandru Elisei <alexandru.elisei@xxxxxxx>; Andre Przywara > <andre.przywara@xxxxxxx>; Andrew Scull <ascull@xxxxxxxxxx>; Catalin > Marinas <catalin.marinas@xxxxxxx>; Christoffer Dall > <christoffer.dall@xxxxxxx>; Howard Zhang <Howard.Zhang@xxxxxxx>; Jia > He <justin.he@xxxxxxx>; Mark Rutland <mark.rutland@xxxxxxx>; Quentin > Perret <qperret@xxxxxxxxxx>; Shameerali Kolothum Thodi > <shameerali.kolothum.thodi@xxxxxxxxxx>; Suzuki K Poulose > <suzuki.poulose@xxxxxxx>; Will Deacon <will@xxxxxxxxxx>; James Morse > <james.morse@xxxxxxx>; Julien Thierry <julien.thierry.kdev@xxxxxxxxx>; > kernel-team@xxxxxxxxxxx; linux-arm-kernel@xxxxxxxxxxxxxxxxxxx; > kvmarm@xxxxxxxxxxxxxxxxxxxxx; kvm@xxxxxxxxxxxxxxx > Subject: [PATCH 7/8] KVM: arm64: Workaround firmware wrongly advertising > GICv2-on-v3 compatibility > > It looks like we have broken firmware out there that wrongly advertises > a GICv2 compatibility interface, despite the CPUs not being able to deal > with it. > > To work around this, check that the CPU initialising KVM is actually able > to switch to MMIO instead of system registers, and use that as a > precondition to enable GICv2 compatibility in KVM. > > Note that the detection happens on a single CPU. If the firmware is > lying *and* that the CPUs are asymetric, all hope is lost anyway. > > Reported-by: Shameerali Kolothum Thodi > <shameerali.kolothum.thodi@xxxxxxxxxx> > Tested-by: Shameer Kolothum <shameerali.kolothum.thodi@xxxxxxxxxx> > Signed-off-by: Marc Zyngier <maz@xxxxxxxxxx> Is it possible to add stable tag for this? Looks like we do have systems out there and reports issues. Thanks, Shameer > --- > arch/arm64/kvm/hyp/vgic-v3-sr.c | 35 +++++++++++++++++++++++++++++++-- > arch/arm64/kvm/vgic/vgic-v3.c | 8 ++++++-- > 2 files changed, 39 insertions(+), 4 deletions(-) > > diff --git a/arch/arm64/kvm/hyp/vgic-v3-sr.c > b/arch/arm64/kvm/hyp/vgic-v3-sr.c > index 005daa0c9dd7..ee3682b9873c 100644 > --- a/arch/arm64/kvm/hyp/vgic-v3-sr.c > +++ b/arch/arm64/kvm/hyp/vgic-v3-sr.c > @@ -408,11 +408,42 @@ void __vgic_v3_init_lrs(void) > /* > * Return the GIC CPU configuration: > * - [31:0] ICH_VTR_EL2 > - * - [63:32] RES0 > + * - [62:32] RES0 > + * - [63] MMIO (GICv2) capable > */ > u64 __vgic_v3_get_gic_config(void) > { > - return read_gicreg(ICH_VTR_EL2); > + u64 val, sre = read_gicreg(ICC_SRE_EL1); > + unsigned long flags = 0; > + > + /* > + * To check whether we have a MMIO-based (GICv2 compatible) > + * CPU interface, we need to disable the system register > + * view. To do that safely, we have to prevent any interrupt > + * from firing (which would be deadly). > + * > + * Note that this only makes sense on VHE, as interrupts are > + * already masked for nVHE as part of the exception entry to > + * EL2. > + */ > + if (has_vhe()) > + flags = local_daif_save(); > + > + write_gicreg(0, ICC_SRE_EL1); > + isb(); > + > + val = read_gicreg(ICC_SRE_EL1); > + > + write_gicreg(sre, ICC_SRE_EL1); > + isb(); > + > + if (has_vhe()) > + local_daif_restore(flags); > + > + val = (val & ICC_SRE_EL1_SRE) ? 0 : (1ULL << 63); > + val |= read_gicreg(ICH_VTR_EL2); > + > + return val; > } > > u64 __vgic_v3_read_vmcr(void) > diff --git a/arch/arm64/kvm/vgic/vgic-v3.c b/arch/arm64/kvm/vgic/vgic-v3.c > index c3e6c3fd333b..6f530925a231 100644 > --- a/arch/arm64/kvm/vgic/vgic-v3.c > +++ b/arch/arm64/kvm/vgic/vgic-v3.c > @@ -575,8 +575,10 @@ early_param("kvm-arm.vgic_v4_enable", > early_gicv4_enable); > int vgic_v3_probe(const struct gic_kvm_info *info) > { > u64 ich_vtr_el2 = kvm_call_hyp_ret(__vgic_v3_get_gic_config); > + bool has_v2; > int ret; > > + has_v2 = ich_vtr_el2 >> 63; > ich_vtr_el2 = (u32)ich_vtr_el2; > > /* > @@ -596,13 +598,15 @@ int vgic_v3_probe(const struct gic_kvm_info *info) > gicv4_enable ? "en" : "dis"); > } > > + kvm_vgic_global_state.vcpu_base = 0; > + > if (!info->vcpu.start) { > kvm_info("GICv3: no GICV resource entry\n"); > - kvm_vgic_global_state.vcpu_base = 0; > + } else if (!has_v2) { > + pr_warn(FW_BUG "CPU interface incapable of MMIO access\n"); > } else if (!PAGE_ALIGNED(info->vcpu.start)) { > pr_warn("GICV physical address 0x%llx not page aligned\n", > (unsigned long long)info->vcpu.start); > - kvm_vgic_global_state.vcpu_base = 0; > } else { > kvm_vgic_global_state.vcpu_base = info->vcpu.start; > kvm_vgic_global_state.can_emulate_gicv2 = true; > -- > 2.29.2
