On Tue, 28 Feb 2023, Jakub Jelinek wrote:
> On Tue, Feb 28, 2023 at 08:58:20AM +0000, Richard Biener wrote:
> > > Without LTO, TYPE_POINTER_TO/TYPE_REFERENCE_TO chains are only maintained
> > > inside of build_{pointer,reference}_type_for_mode and those routines
> > > ensure that the pointer/reference type added to the chain is really
> > > unqualified (including address space), without extra user alignment
> > > and has just one entry for each of the TYPE_MODE/TYPE_REF_CAN_ALIAS_ALL
> > > pair (unless something would modify the types in place, but that would
> > > be wrong).
> >
> > Is that so? I can't find any code verifying that (verify_type?).
> > Of course build_{pointer,reference}_type_for_mode will always build
> > unqualified pointers, but then the LTO code does
> >
> > /* The following reconstructs the pointer chains
> > of the new pointed-to type if we are a main variant. We do
> > not stream those so they are broken before fixup. */
> > if (TREE_CODE (t) == POINTER_TYPE
> > && TYPE_MAIN_VARIANT (t) == t)
> > {
> > TYPE_NEXT_PTR_TO (t) = TYPE_POINTER_TO (TREE_TYPE (t));
> > TYPE_POINTER_TO (TREE_TYPE (t)) = t;
> > }
> > else if (TREE_CODE (t) == REFERENCE_TYPE
> > && TYPE_MAIN_VARIANT (t) == t)
> > {
> > TYPE_NEXT_REF_TO (t) = TYPE_REFERENCE_TO (TREE_TYPE (t));
> > TYPE_REFERENCE_TO (TREE_TYPE (t)) = t;
> > }
> >
> > which was supposed to ensure only putting unqualified pointers
> > (not pointed to types!) to the chain. So to me the question is
> > rather why a type with TYPE_USER_ALIGN is a the main variant - that's
> > what looks wrong here?
>
> First of all, it is unclear how the above ensures that type isn't
> added to those chains even when it already is there (or some similar type).
> Say lto1 during initialization needs build_pointer_type (float_type_node)
> and later on lto_fixup_prevailing_type is called on some float *.
We try to make sure to put all built types into the type merging
machinery, so I think it shouldn't happen - but then I cannot rule
it out. I'm also not sure whether duplicates would really pose
a problem here (other than waste). If we want to make sure we should
probably enhance verify_types to verify the TYPE_POINTER_TO chain ...
> I'll try to debug why those user aligned types become TYPE_MAIN_VARIANT
> though...
>
> Jakub
>
>
--
Richard Biener <rguent...@suse.de>
SUSE Software Solutions Germany GmbH, Frankenstrasse 146, 90461 Nuernberg,
Germany; GF: Ivo Totev, Andrew Myers, Andrew McDonald, Boudien Moerman;
HRB 36809 (AG Nuernberg)
