linux/kernel/entry/kvm.c
Seth Forshee 3e684903a8 entry/kvm: Exit to user mode when TIF_NOTIFY_SIGNAL is set
A livepatch transition may stall indefinitely when a kvm vCPU is heavily
loaded. To the host, the vCPU task is a user thread which is spending a
very long time in the ioctl(KVM_RUN) syscall. During livepatch
transition, set_notify_signal() will be called on such tasks to
interrupt the syscall so that the task can be transitioned. This
interrupts guest execution, but when xfer_to_guest_mode_work() sees that
TIF_NOTIFY_SIGNAL is set but not TIF_SIGPENDING it concludes that an
exit to user mode is unnecessary, and guest execution is resumed without
transitioning the task for the livepatch.

This handling of TIF_NOTIFY_SIGNAL is incorrect, as set_notify_signal()
is expected to break tasks out of interruptible kernel loops and cause
them to return to userspace. Change xfer_to_guest_mode_work() to handle
TIF_NOTIFY_SIGNAL the same as TIF_SIGPENDING, signaling to the vCPU run
loop that an exit to userpsace is needed. Any pending task_work will be
run when get_signal() is called from exit_to_user_mode_loop(), so there
is no longer any need to run task work from xfer_to_guest_mode_work().

Suggested-by: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Petr Mladek <pmladek@suse.com>
Signed-off-by: Seth Forshee <sforshee@digitalocean.com>
Message-Id: <20220504180840.2907296-1-sforshee@digitalocean.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2022-06-07 11:19:00 -04:00

50 lines
1.1 KiB
C

// SPDX-License-Identifier: GPL-2.0
#include <linux/entry-kvm.h>
#include <linux/kvm_host.h>
static int xfer_to_guest_mode_work(struct kvm_vcpu *vcpu, unsigned long ti_work)
{
do {
int ret;
if (ti_work & (_TIF_SIGPENDING | _TIF_NOTIFY_SIGNAL)) {
kvm_handle_signal_exit(vcpu);
return -EINTR;
}
if (ti_work & _TIF_NEED_RESCHED)
schedule();
if (ti_work & _TIF_NOTIFY_RESUME)
resume_user_mode_work(NULL);
ret = arch_xfer_to_guest_mode_handle_work(vcpu, ti_work);
if (ret)
return ret;
ti_work = read_thread_flags();
} while (ti_work & XFER_TO_GUEST_MODE_WORK || need_resched());
return 0;
}
int xfer_to_guest_mode_handle_work(struct kvm_vcpu *vcpu)
{
unsigned long ti_work;
/*
* This is invoked from the outer guest loop with interrupts and
* preemption enabled.
*
* KVM invokes xfer_to_guest_mode_work_pending() with interrupts
* disabled in the inner loop before going into guest mode. No need
* to disable interrupts here.
*/
ti_work = read_thread_flags();
if (!(ti_work & XFER_TO_GUEST_MODE_WORK))
return 0;
return xfer_to_guest_mode_work(vcpu, ti_work);
}
EXPORT_SYMBOL_GPL(xfer_to_guest_mode_handle_work);