Re: [PATCH v2] mm/writeback: fix dereferencing NULL mapping->host

[Andrew Yang]

On Tue, 2022-11-29 at 09:56 -0500, Steven Rostedt wrote:
> On Tue, 29 Nov 2022 11:32:59 +0800
> Andrew Yang <andrew.yang@xxxxxxxxxxxx> wrote:
> 
> > From: "andrew.yang" <andrew.yang@xxxxxxxxxxxx>
> > 
> > Check before dereferencing mapping->host
> > 
> > Signed-off-by: andrew.yang <andrew.yang@xxxxxxxxxxxx>
> > ---
> >  include/trace/events/writeback.h | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/include/trace/events/writeback.h
> > b/include/trace/events/writeback.h
> > index 86b2a82da546..56f6e114d3ed 100644
> > --- a/include/trace/events/writeback.h
> > +++ b/include/trace/events/writeback.h
> > @@ -68,7 +68,7 @@ DECLARE_EVENT_CLASS(writeback_folio_template,
> >  		strscpy_pad(__entry->name,
> >  			    bdi_dev_name(mapping ?
> > inode_to_bdi(mapping->host) :
> >  					 NULL), 32);
> > -		__entry->ino = mapping ? mapping->host->i_ino : 0;
> > +		__entry->ino = mapping && mapping->host ? mapping-
> > >host->i_ino : 0;
> 
> I hate remembering precedence. Can we add parenthesis around this to
> be
> clear?
> 
> 		__entry->ino = (mapping && mapping->host) ? mapping-
> >host->i_ino : 0;
> 
> Thanks,
> 
> -- Steve
> 
> 
> >  		__entry->index = folio->index;
> >  	),
> >  
> 
> 
Sure, that's a good suggestion