Re: [PATCH v4] x86/sgx: Fix deadlock and race conditions between fork() and EPC reclaim

[Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>]

On Thu, Apr 16, 2020 at 08:25:01PM +0300, Jarkko Sakkinen wrote:
> On Wed, Apr 15, 2020 at 09:28:10PM -0700, Sean Christopherson wrote:
> > On Tue, Apr 14, 2020 at 09:45:28PM +0300, Jarkko Sakkinen wrote:
> > > On Tue, Apr 14, 2020 at 10:17:49AM +0300, Jarkko Sakkinen wrote:
> > > > On Mon, Apr 13, 2020 at 09:32:34PM -0700, Sean Christopherson wrote:
> > > > > On Mon, Apr 06, 2020 at 11:56:26PM +0300, Jarkko Sakkinen wrote:
> > > > > > From: Sean Christopherson <sean.j.christopherson@intel.com>
> > > > >   	spin_lock(&encl->mm_lock);
> > > > > > +
> > > > > >  	list_add_rcu(&encl_mm->list, &encl->mm_list);
> > > > > > -	spin_unlock(&encl->mm_lock);
> > > > > >  
> > > > > > -	synchronize_srcu(&encl->srcu);
> > > > > > +	/* Even if the CPU does not reorder writes, a compiler might. */
> > > > > 
> > > > > The preferred (by maintainers) style of comment for smp_wmb()/smp_rmb()
> > > > > comments is to explicitly call out the associated reader/writer.  If you
> > > > > want to go with a minimal comment, my vote is for something like:
> > > > > 
> > > > > 	/*
> > > > > 	 * Add to list before updating version.  Pairs the with smp_rmb() in
> > > > > 	 * sgx_reclaimer_block().
> > > > > 	 */
> > > > > 
> > > > > And if you want to go really spartan, I'd take:
> > > > > 
> > > > > 	/* Pairs with smp_rmb() in sgx_reclaimer_block(). */
> > > > > 
> > > > > over a generic comment about the compiler reordering instructions.
> > > > 
> > > > Thaks Sean, makes sense, I'll go with your "spartan" suggestion.
> > > 
> > > Updated, ready to squash?
> > 
> > Any objection to using the spartan comment for the smb_rmb() in
> > sgx_reclaimer_block() as well?
> 
> For sure. I think here the role of the comment is to help with
> the navigation.
> 
> /Jarkko

Finally squashed.

/Jarkko