Received-SPF: pass (google.com: domain of linux-kernel+bounces-39708-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Date: Fri, 26 Jan 2024 15:36:20 +0800
From: Xu Yilun <yilun.xu@linux.intel.com>
To: Sean Christopherson <seanjc@google.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>, kvm@vger.kernel.org,
	linux-kernel@vger.kernel.org, David Matlack <dmatlack@google.com>
Subject: Re: [PATCH 1/4] KVM: Always flush async #PF workqueue when vCPU is
 being destroyed
Message-ID: <ZbNg9BFT2o6NbgRX@yilunxu-OptiPlex-7050>
References: <20240110011533.503302-1-seanjc@google.com>
 <20240110011533.503302-2-seanjc@google.com>
 <Zau/VQ0B5MCwoqZT@yilunxu-OptiPlex-7050>
 <ZbFfVxp76qoBstul@google.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <ZbFfVxp76qoBstul@google.com>

On Wed, Jan 24, 2024 at 11:04:55AM -0800, Sean Christopherson wrote:
> On Sat, Jan 20, 2024, Xu Yilun wrote:
> > On Tue, Jan 09, 2024 at 05:15:30PM -0800, Sean Christopherson wrote:
> > > Always flush the per-vCPU async #PF workqueue when a vCPU is clearing its
> > > completion queue, e.g. when a VM and all its vCPUs is being destroyed.
> > > KVM must ensure that none of its workqueue callbacks is running when the
> > > last reference to the KVM _module_ is put.  Gifting a reference to the
> > > associated VM prevents the workqueue callback from dereferencing freed
> > > vCPU/VM memory, but does not prevent the KVM module from being unloaded
> > > before the callback completes.
> > > 
> > > Drop the misguided VM refcount gifting, as calling kvm_put_kvm() from
> > > async_pf_execute() if kvm_put_kvm() flushes the async #PF workqueue will
> > > result in deadlock.  async_pf_execute() can't return until kvm_put_kvm()
> > > finishes, and kvm_put_kvm() can't return until async_pf_execute() finishes:
> > > 
> > >  WARNING: CPU: 8 PID: 251 at virt/kvm/kvm_main.c:1435 kvm_put_kvm+0x2d/0x320 [kvm]
> > >  Modules linked in: vhost_net vhost vhost_iotlb tap kvm_intel kvm irqbypass
> > >  CPU: 8 PID: 251 Comm: kworker/8:1 Tainted: G        W          6.6.0-rc1-e7af8d17224a-x86/gmem-vm #119
> > >  Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/2015
> > >  Workqueue: events async_pf_execute [kvm]
> > >  RIP: 0010:kvm_put_kvm+0x2d/0x320 [kvm]
> > >  Call Trace:
> > >   <TASK>
> > >   async_pf_execute+0x198/0x260 [kvm]
> > >   process_one_work+0x145/0x2d0
> > >   worker_thread+0x27e/0x3a0
> > >   kthread+0xba/0xe0
> > >   ret_from_fork+0x2d/0x50
> > >   ret_from_fork_asm+0x11/0x20
> > >   </TASK>
> > >  ---[ end trace 0000000000000000 ]---
> > >  INFO: task kworker/8:1:251 blocked for more than 120 seconds.
> > >        Tainted: G        W          6.6.0-rc1-e7af8d17224a-x86/gmem-vm #119
> > >  "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> > >  task:kworker/8:1     state:D stack:0     pid:251   ppid:2      flags:0x00004000
> > >  Workqueue: events async_pf_execute [kvm]
> > >  Call Trace:
> > >   <TASK>
> > >   __schedule+0x33f/0xa40
> > >   schedule+0x53/0xc0
> > >   schedule_timeout+0x12a/0x140
> > >   __wait_for_common+0x8d/0x1d0
> > >   __flush_work.isra.0+0x19f/0x2c0
> > >   kvm_clear_async_pf_completion_queue+0x129/0x190 [kvm]
> > >   kvm_arch_destroy_vm+0x78/0x1b0 [kvm]
> > >   kvm_put_kvm+0x1c1/0x320 [kvm]
> > >   async_pf_execute+0x198/0x260 [kvm]
> > >   process_one_work+0x145/0x2d0
> > >   worker_thread+0x27e/0x3a0
> > >   kthread+0xba/0xe0
> > >   ret_from_fork+0x2d/0x50
> > >   ret_from_fork_asm+0x11/0x20
> > >   </TASK>
> > > 
> > > If kvm_clear_async_pf_completion_queue() actually flushes the workqueue,
> > > then there's no need to gift async_pf_execute() a reference because all
> > > invocations of async_pf_execute() will be forced to complete before the
> > > vCPU and its VM are destroyed/freed.  And that in turn fixes the module
> > > unloading bug as __fput() won't do module_put() on the last vCPU reference
> > > until the vCPU has been freed, e.g. if closing the vCPU file also puts the
> > 
> > I'm not sure why __fput() of vCPU fd should be mentioned here. I assume
> > we just need to say that vCPUs are freed before module_put(KVM the module)
> > in kvm_destroy_vm(), then the whole logic for module unloading fix is:
> > 
> >   1. All workqueue callbacks complete when kvm_clear_async_pf_completion_queue(vcpu)
> >   2. kvm_clear_async_pf_completion_queue(vcpu) must be executed before vCPU free.
> >   3. vCPUs must be freed before module_put(KVM the module).
> > 
> >   So all workqueue callbacks complete before module_put(KVM the module).
> > 
> > 
> > __fput() of vCPU fd is not the only trigger of kvm_destroy_vm(), that
> > makes me distracted from reason of the fix.
> 
> My goal was to call out that (a) the vCPU file descriptor is what ensures kvm.ko
> is alive at this point and (b) that __fput() very deliberately ensures module_put()
> is called after all module function callbacks/hooks complete, as there was quite

Ah, I understood. These are ensured by your previous fix which grants
kvm_vcpu_fops the module owner. LGTM now.

Thanks,
Yilun