[Bug rtl-optimization/110237] gcc.dg/torture/pr58955-2.c is miscompiled by RTL scheduling after reload

["rguenther at suse dot de" <gcc-bugzilla@gcc.gnu.org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110237

--- Comment #5 from rguenther at suse dot de <rguenther at suse dot de> ---
On Tue, 20 Jun 2023, crazylht at gmail dot com wrote:

> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110237
> 
> --- Comment #4 from Hongtao.liu <crazylht at gmail dot com> ---
> (In reply to Richard Biener from comment #3)
> > This looks like the same issue as PR110309.  We have
> > 
> > (insn 38 35 39 3 (set (mem:V16SI (plus:DI (reg:DI 40 r12 [orig:90 _22 ] [90])
> >                 (const:DI (plus:DI (symbol_ref:DI ("b") [flags 0x2] 
> > <var_decl 0x7ffff6e28d80 b>)
> >                         (const_int -4 [0xfffffffffffffffc])))) [1 MEM
> > <vector(16) int> [(int *)vectp_b.12_28]+0 S64 A32])
> >         (vec_merge:V16SI (reg:V16SI 20 xmm0 [118])
> >             (mem:V16SI (plus:DI (reg:DI 40 r12 [orig:90 _22 ] [90]) 
> >                     (const:DI (plus:DI (symbol_ref:DI ("b") [flags 0x2] 
> > <var_decl 0x7ffff6e28d80 b>)
> >                             (const_int -4 [0xfffffffffffffffc])))) [1 MEM
> > <vector(16) int> [(int *)vectp_b.12_28]+0 S64 A32])
> > 
> > so instead of a masked load we see a vec_merge with a (mem:V16SI ...)
> > based on the vectp_b.12_28 pointer that has full size but the load of b[1]
> > we try disambiguate against refers to int b[10] which is too small for
> > a load of 64 bytes so we disambiguate based on that.
> 
> 
>   /* If the pointer based access is bigger than the variable they cannot
>      alias.  This is similar to the check below where we use TBAA to
>      increase the size of the pointer based access based on the dynamic
>      type of a containing object we can infer from it.  */
>   poly_int64 dsize2;
>   if (known_size_p (size1) --- should be unknown??
>       && poly_int_tree_p (DECL_SIZE (base2), &dsize2)
>       && known_lt (dsize2, size1))
>     return false;
> 
> Should we set MEM_SIZE_KNOWN_P to false for maskstore/maskload?
> It seems to me maxsize should be 64bytes, but real size should be unknown.

Yes, you shouldn't have MEM_ATTRs that indicate the size is known.