Re: [PATCH] nvdimm: virtio_pmem: serialize flush requests

[Ira Weiny]

Michael S. Tsirkin wrote:
> On Fri, Jan 30, 2026 at 02:52:12PM -0600, Ira Weiny wrote:
> > Li Chen wrote:

[snip]

> > > diff --git a/drivers/nvdimm/nd_virtio.c b/drivers/nvdimm/nd_virtio.c
> > > index c3f07be4aa22..827a17fe7c71 100644
> > > --- a/drivers/nvdimm/nd_virtio.c
> > > +++ b/drivers/nvdimm/nd_virtio.c
> > > @@ -44,19 +44,24 @@ static int virtio_pmem_flush(struct nd_region *nd_region)
> > >  	unsigned long flags;
> > >  	int err, err1;
> > >  
> > > +	might_sleep();
> > > +	mutex_lock(&vpmem->flush_lock);
> > 
> > Assuming this does fix a bug I'd rather use guard here.
> 
> Do you, from code review, agree with the logic that
> it's racy right now?

I do now.  I was hoping to understand the test being run.  The additional
detail that it takes multiple runs helps.

> Whether the bug is reproducible isn't really the question.
> 

True.  But we should still use guard().  I'll look for v2.

Ira