On Thu, Apr 18, 2024 at 12:25:21PM -0700, Guenter Roeck wrote:
> On Thu, Apr 11, 2024 at 11:53:29AM +0200, Greg Kroah-Hartman wrote:
> > This is the start of the stable review cycle for the 5.4.274 release.
> > There are 215 patches in this series, all will be posted as a response
> > to this one. If anyone has any issues with these being applied, please
> > let me know.
> >
> > Responses should be made by Sat, 13 Apr 2024 09:53:55 +0000.
> > Anything received after that time might be too late.
> >
> [ ... ]
> >
> > Sean Christopherson <seanjc@xxxxxxxxxx>
> > KVM: Always flush async #PF workqueue when vCPU is being destroyed
> >
>
> This backport is bad. In kvm_setup_async_pf(), it removes a call to
> kvm_get_kvm(). However, it does not remove the call to kvm_put_kvm()
> in its error handler. Also see upstream commit 7863e346e108 ("KVM:
> async_pf: Cleanup kvm_setup_async_pf()") which explains that one of
> the error paths in kvm_setup_async_pf() which is not supposed to be
> observed can be observed after all.
>
Reverting the above commit from v5.4.y fixes the problem. Alternatively,
applying commit 7863e346e108 on top of v5.4.274 fixes it as well.
I added Phillip Pearson to Cc:; he did all the testing and can provide
additional information if needed.
Thanks,
Guenter
