Messages in this thread | | | Subject | Re: 5.13-rt1 + KVM = WARNING: at fs/eventfd.c:74 eventfd_signal() | From | Paolo Bonzini <> | Date | Fri, 16 Jul 2021 09:59:15 +0200 |
| |
On 16/07/21 09:55, Hillf Danton wrote: > On Fri, 16 Jul 2021 08:54:58 +0200 Paolo Bonzini wrote: >> On 16/07/21 04:06, Hillf Danton wrote: >>>> With the patch: >>>> - no warn >>>> - continue using the VM normally... >>> Well with the patch applied, the VM works fine without the stuff protected >>> by the spin_lock_irqsave(), then without the patch why simply printing a >>> warning makes the VM dumb, given the warning is there actually also >>> preventing you from touching the lock. >> >> If the warning is triggered, eventfd_signal will not do the wakeup. > > [I am assuming we are not talking about the deadlock in the comment.] > > No preemption occured without the warning printed. > Why will the wakeup behavior change without peemption?
Sorry, I don't follow. What I'm saying is that without the patch:
* the warning only occurs if preemption occurs during the spin_lock_irqsave critical section (and therefore it can only occur in PREEMPT_RT kernels)
* the warning causes an early return 0 that messes up the VM's networking
Paolo
| |