]> git.ipfire.org Git - thirdparty/kernel/stable.git/commit
KVM: arm64: Wake-up from WFI when iqrchip is in userspace
authorMarc Zyngier <maz@kernel.org>
Thu, 23 Apr 2026 16:36:07 +0000 (17:36 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Thu, 14 May 2026 13:31:18 +0000 (15:31 +0200)
commitcd512834f7ce7487bc6cd6476d3f5f171c6e806b
tree7c17211650278ad9a46f2134f4ff135fc20c29fa
parente7c6d30169b03307d27c4479563df79c08f3a746
KVM: arm64: Wake-up from WFI when iqrchip is in userspace

commit 4ce98bf0865c349e7026ad9c14f48da264920953 upstream.

It appears that there is nothing in the wake-up path that
evaluates whether the in-kernel interrupts are pending unless
we have a vgic.

This means that the userspace irqchip support has been broken for
about four years, and nobody noticed. It was also broken before
as we wouldn't wake-up on a PMU interrupt, but hey, who cares...

It is probably time to remove the feature altogether, because it
was a terrible idea 10 years ago, and it still is.

Fixes: b57de4ffd7c6d ("KVM: arm64: Simplify kvm_cpu_has_pending_timer()")
Link: https://patch.msgid.link/20260423163607.486345-1-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
arch/arm64/kvm/arm.c