This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] Fix ICE in get_memory_rtx

> > -		  && TYPE_SIZE_UNIT (TREE_TYPE (inner))
> > -		  && host_integerp (TYPE_SIZE_UNIT (TREE_TYPE (inner)), 0))
> > +		  && host_integerp (DECL_SIZE_UNIT (field), 0))
> >  		{
> >  		  HOST_WIDE_INT size
> > -		    = tree_low_cst (TYPE_SIZE_UNIT (TREE_TYPE (inner)), 0);
> > +		    = TREE_INT_CST_LOW (DECL_SIZE_UNIT (field));
>
> Why change from tree_low_cst to TREE_INT_CST_LOW?  Shouldn't the only
> change here be from TYPE_SIZE_UNIT to DECL_SIZE_UNIT?

It's indeed not required, but I personally avoid writing

  if (host_integerp (t, 0))
    ... = tree_low_cst (t, 0);

since it's totally redundant.  Not sure what the policy wrt tree_low_cst vs 
TREE_INT_CST_LOW is though (if any).

-- 
Eric Botcazou
Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]