Re: [PATCH] mm/hmm: potential deadlock in nonblocking code

[Dan Carpenter <dan.carpenter@xxxxxxxxxx>]

On Mon, Feb 04, 2019 at 05:42:03AM -0800, Matthew Wilcox wrote:
> On Mon, Feb 04, 2019 at 04:20:44PM +0300, Dan Carpenter wrote:
> >  
> > -	if (!nrange->blockable && !mutex_trylock(&hmm->lock)) {
> > -		ret = -EAGAIN;
> > -		goto out;
> > +	if (!nrange->blockable) {
> > +		if (!mutex_trylock(&hmm->lock)) {
> > +			ret = -EAGAIN;
> > +			goto out;
> > +		}
> >  	} else
> >  		mutex_lock(&hmm->lock);
> 
> I think this would be more readable written as:
> 
> 	ret = -EAGAIN;
> 	if (nrange->blockable)
> 		mutex_lock(&hmm->lock);
> 	else if (!mutex_trylock(&hmm->lock))
> 		goto out;

I agree, that does look nicer.  I will resend.

regards,
dan carpenter