diff options
| author | Marc Zyngier <maz@kernel.org> | 2021-10-10 17:09:08 +0200 |
|---|---|---|
| committer | Marc Zyngier <maz@kernel.org> | 2021-10-17 12:06:36 +0200 |
| commit | 0924729b21bffdd0e13f29ea6256d299fc807cff (patch) | |
| tree | 9b25d70b11e36ae6167b2cacd9b51bf2ee744a69 /arch/arm64/kvm/vgic | |
| parent | KVM: arm64: vgic-v3: Work around GICv3 locally generated SErrors (diff) | |
| download | linux-0924729b21bffdd0e13f29ea6256d299fc807cff.tar.xz linux-0924729b21bffdd0e13f29ea6256d299fc807cff.zip | |
KVM: arm64: vgic-v3: Reduce common group trapping to ICV_DIR_EL1 when possible
On systems that advertise ICH_VTR_EL2.SEIS, we trap all GICv3 sysreg
accesses from the guest. From a performance perspective, this is OK
as long as the guest doesn't hammer the GICv3 CPU interface.
In most cases, this is fine, unless the guest actively uses
priorities and switches PMR_EL1 very often. Which is exactly what
happens when a Linux guest runs with irqchip.gicv3_pseudo_nmi=1.
In these condition, the performance plumets as we hit PMR each time
we mask/unmask interrupts. Not good.
There is however an opportunity for improvement. Careful reading
of the architecture specification indicates that the only GICv3
sysreg belonging to the common group (which contains the SGI
registers, PMR, DIR, CTLR and RPR) that is allowed to generate
a SError is DIR. Everything else is safe.
It is thus possible to substitute the trapping of all the common
group with just that of DIR if it supported by the implementation.
Yes, that's yet another optional bit of the architecture.
So let's just do that, as it leads to some impressive result on
the M1:
Without this change:
bash-5.1# /host/home/maz/hackbench 100 process 1000
Running with 100*40 (== 4000) tasks.
Time: 56.596
With this change:
bash-5.1# /host/home/maz/hackbench 100 process 1000
Running with 100*40 (== 4000) tasks.
Time: 8.649
which is a pretty convincing result.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Alexandru Elisei <alexandru.elisei@arm.com>
Link: https://lore.kernel.org/r/20211010150910.2911495-4-maz@kernel.org
Diffstat (limited to 'arch/arm64/kvm/vgic')
| -rw-r--r-- | arch/arm64/kvm/vgic/vgic-v3.c | 15 |
1 files changed, 11 insertions, 4 deletions
diff --git a/arch/arm64/kvm/vgic/vgic-v3.c b/arch/arm64/kvm/vgic/vgic-v3.c index ae59e2580bf5..467c22bbade6 100644 --- a/arch/arm64/kvm/vgic/vgic-v3.c +++ b/arch/arm64/kvm/vgic/vgic-v3.c @@ -15,6 +15,7 @@ static bool group0_trap; static bool group1_trap; static bool common_trap; +static bool dir_trap; static bool gicv4_enable; void vgic_v3_set_underflow(struct kvm_vcpu *vcpu) @@ -296,6 +297,8 @@ void vgic_v3_enable(struct kvm_vcpu *vcpu) vgic_v3->vgic_hcr |= ICH_HCR_TALL1; if (common_trap) vgic_v3->vgic_hcr |= ICH_HCR_TC; + if (dir_trap) + vgic_v3->vgic_hcr |= ICH_HCR_TDIR; } int vgic_v3_lpi_sync_pending_status(struct kvm *kvm, struct vgic_irq *irq) @@ -676,14 +679,18 @@ int vgic_v3_probe(const struct gic_kvm_info *info) group0_trap = true; group1_trap = true; - common_trap = true; + if (ich_vtr_el2 & ICH_VTR_TDS_MASK) + dir_trap = true; + else + common_trap = true; } - if (group0_trap || group1_trap || common_trap) { - kvm_info("GICv3 sysreg trapping enabled ([%s%s%s], reduced performance)\n", + if (group0_trap || group1_trap || common_trap | dir_trap) { + kvm_info("GICv3 sysreg trapping enabled ([%s%s%s%s], reduced performance)\n", group0_trap ? "G0" : "", group1_trap ? "G1" : "", - common_trap ? "C" : ""); + common_trap ? "C" : "", + dir_trap ? "D" : ""); static_branch_enable(&vgic_v3_cpuif_trap); } |