Re: 5.13-rt1 + KVM = WARNING: at fs/eventfd.c:74 eventfd_signal()

[Michael S. Tsirkin]

On Wed, Jul 14, 2021 at 12:35:27PM +0200, Paolo Bonzini wrote:
> On 14/07/21 11:23, Jason Wang wrote:
> > > This was added in 2020, so it's unlikely to be the direct cause of the
> > > change.  What is a known-good version for the host?
> > > 
> > > Since it is not KVM stuff, I'm CCing Michael and Jason.
> > 
> > I think this can be probably fixed here:
> > 
> > https://lore.kernel.org/lkml/20210618084412.18257-1-zhe.he@xxxxxxxxxxxxx/
> 
> That seems wrong; in particular it wouldn't protect against AB/BA deadlocks.
> In fact, the bug is with the locking; the code assumes that
> spin_lock_irqsave/spin_unlock_irqrestore is non-preemptable and therefore
> increments and decrements the percpu variable inside the critical section.
> 
> This obviously does not fly with PREEMPT_RT; the right fix should be
> using a local_lock.  Something like this (untested!!):
> 
> --------------- 8< ---------------
> From: Paolo Bonzini <pbonzini@xxxxxxxxxx>
> Subject: [PATCH] eventfd: protect eventfd_wake_count with a local_lock
> 
> eventfd_signal assumes that spin_lock_irqsave/spin_unlock_irqrestore is
> non-preemptable and therefore increments and decrements the percpu
> variable inside the critical section.
> 
> This obviously does not fly with PREEMPT_RT.  If eventfd_signal is
> preempted and an unrelated thread calls eventfd_signal, the result is
> a spurious WARN.  To avoid this, protect the percpu variable with a
> local_lock.
> 
> Reported-by: Daniel Bristot de Oliveira <bristot@xxxxxxxxxx>
> Fixes: b5e683d5cab8 ("eventfd: track eventfd_signal() recursion depth")
> Cc: stable@xxxxxxxxxxxxxxx
> Cc: He Zhe <zhe.he@xxxxxxxxxxxxx>
> Cc: Jens Axboe <axboe@xxxxxxxxx>
> Signed-off-by: Paolo Bonzini <pbonzini@xxxxxxxxxx>


Makes sense ... 

Acked-by: Michael S. Tsirkin <mst@xxxxxxxxxx>

want to send this to the windriver guys so they can test?
Here's the list from that thread:

To: xieyongji@xxxxxxxxxxxxx, mst@xxxxxxxxxx, jasowang@xxxxxxxxxx,
	stefanha@xxxxxxxxxx, sgarzare@xxxxxxxxxx, parav@xxxxxxxxxx,
	hch@xxxxxxxxxxxxx, christian.brauner@xxxxxxxxxxxxx,
	rdunlap@xxxxxxxxxxxxx, willy@xxxxxxxxxxxxx,
	viro@xxxxxxxxxxxxxxxxxx, axboe@xxxxxxxxx, bcrl@xxxxxxxxx,
	corbet@xxxxxxx, mika.penttila@xxxxxxxxxxxx,
	dan.carpenter@xxxxxxxxxx, gregkh@xxxxxxxxxxxxxxxxxxx,
	songmuchun@xxxxxxxxxxxxx,
	virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx, kvm@xxxxxxxxxxxxxxx,
	linux-fsdevel@xxxxxxxxxxxxxxx, iommu@xxxxxxxxxxxxxxxxxxxxxxxxxx,
	linux-kernel@xxxxxxxxxxxxxxx, qiang.zhang@xxxxxxxxxxxxx,
	zhe.he@xxxxxxxxxxxxx


> 
> diff --git a/fs/eventfd.c b/fs/eventfd.c
> index e265b6dd4f34..7d27b6e080ea 100644
> --- a/fs/eventfd.c
> +++ b/fs/eventfd.c
> @@ -12,6 +12,7 @@
>  #include <linux/fs.h>
>  #include <linux/sched/signal.h>
>  #include <linux/kernel.h>
> +#include <linux/local_lock.h>
>  #include <linux/slab.h>
>  #include <linux/list.h>
>  #include <linux/spinlock.h>
> @@ -25,6 +26,7 @@
>  #include <linux/idr.h>
>  #include <linux/uio.h>
> +static local_lock_t eventfd_wake_lock = INIT_LOCAL_LOCK(eventfd_wake_lock);
>  DEFINE_PER_CPU(int, eventfd_wake_count);
>  static DEFINE_IDA(eventfd_ida);
> @@ -71,8 +73,11 @@ __u64 eventfd_signal(struct eventfd_ctx *ctx, __u64 n)
>  	 * it returns true, the eventfd_signal() call should be deferred to a
>  	 * safe context.
>  	 */
> -	if (WARN_ON_ONCE(this_cpu_read(eventfd_wake_count)))
> +	local_lock(&eventfd_wake_lock);
> +	if (WARN_ON_ONCE(this_cpu_read(eventfd_wake_count))) {
> +		local_unlock(&eventfd_wake_lock);
>  		return 0;
> +	}
>  	spin_lock_irqsave(&ctx->wqh.lock, flags);
>  	this_cpu_inc(eventfd_wake_count);
> @@ -83,6 +88,7 @@ __u64 eventfd_signal(struct eventfd_ctx *ctx, __u64 n)
>  		wake_up_locked_poll(&ctx->wqh, EPOLLIN);
>  	this_cpu_dec(eventfd_wake_count);
>  	spin_unlock_irqrestore(&ctx->wqh.lock, flags);
> +	local_unlock(&eventfd_wake_lock);
>  	return n;
>  }