Here is what I came up with for combining the two:
/* For (x << c) >> c, optimize into x & ((unsigned)-1 >> c) for
unsigned x OR truncate into the precision(type) - c lowest bits
of signed x (if they have mode precision or a precision of 1) */
(simplify
(rshift (nop_convert? (lshift @0 INTEGER_CST@1)) @@1)
(if (wi::ltu_p (wi::to_wide (@1), element_precision (type)))
(if (TYPE_UNSIGNED (type))
(bit_and @0 (rshift { build_minus_one_cst (type); } @1))
(if (INTEGRAL_TYPE_P (type))
(with {
int width = element_precision (type) - tree_to_uhwi (@1);
tree stype = build_nonstandard_integer_type (width, 0);
}
(if (TYPE_PRECISION (stype) == 1 || type_has_mode_precision_p (stype))
(convert (convert:stype @0))))))))
Let me know what you think.
> Btw, I wonder whether we can handle
> some cases of widening/truncating converts between the shifts?
I will look into this.
Drew
On Wed, Jul 26, 2023 at 4:40 AM Richard Biener <richard.guent...@gmail.com>
wrote:
> On Tue, Jul 25, 2023 at 9:26 PM Drew Ross <drr...@redhat.com> wrote:
> >
> > > With that fixed I think for non-vector integrals the above is the most
> suitable
> > > canonical form of a sign-extension. Note it should also work for any
> other
> > > constant shift amount - just use the appropriate intermediate
> precision for
> > > the truncating type.
> > > We _might_ want
> > > to consider to only use the converts when the intermediate type has
> > > mode precision (and as a special case allow one bit as in your above
> case)
> > > so it can expand to (sign_extend:<outer> (subreg:<inner> reg)).
> >
> > Here is a pattern that that only matches to truncations that result in
> mode precision (or precision of 1):
> >
> > (simplify
> > (rshift (nop_convert? (lshift @0 INTEGER_CST@1)) @@1)
> > (if (INTEGRAL_TYPE_P (type)
> > && !TYPE_UNSIGNED (type)
> > && wi::gt_p (element_precision (type), wi::to_wide (@1), TYPE_SIGN
> (TREE_TYPE (@1))))
> > (with {
> > int width = element_precision (type) - tree_to_uhwi (@1);
> > tree stype = build_nonstandard_integer_type (width, 0);
> > }
> > (if (TYPE_PRECISION (stype) == 1 || type_has_mode_precision_p (stype))
> > (convert (convert:stype @0))))))
> >
> > Look ok?
>
> I suppose so. Can you see to amend the existing
>
> /* Optimize (x << c) >> c into x & ((unsigned)-1 >> c) for unsigned
> types. */
> (simplify
> (rshift (lshift @0 INTEGER_CST@1) @1)
> (if (TYPE_UNSIGNED (type)
> && (wi::ltu_p (wi::to_wide (@1), element_precision (type))))
> (bit_and @0 (rshift { build_minus_one_cst (type); } @1))))
>
> pattern? You will get a duplicate pattern diagnostic otherwise. It
> also looks like this
> one has the (nop_convert? ..) missing. Btw, I wonder whether we can handle
> some cases of widening/truncating converts between the shifts?
>
> Richard.
>
> > > You might also want to verify what RTL expansion
> > > produces before/after - it at least shouldn't be worse.
> >
> > The RTL is slightly better for the mode precision cases and slightly
> worse for the precision 1 case.
> >
> > > That said - do you have any testcase where the canonicalization is an
> enabler
> > > for further transforms or was this requested stand-alone?
> >
> > No, I don't have any specific test cases. This patch is just in response
> to pr101955.
> >
> > On Tue, Jul 25, 2023 at 2:55 AM Richard Biener <
> richard.guent...@gmail.com> wrote:
> >>
> >> On Mon, Jul 24, 2023 at 9:42 PM Jakub Jelinek <ja...@redhat.com> wrote:
> >> >
> >> > On Mon, Jul 24, 2023 at 03:29:54PM -0400, Drew Ross via Gcc-patches
> wrote:
> >> > > So would something like
> >> > >
> >> > > (simplify
> >> > > (rshift (nop_convert? (lshift @0 INTEGER_CST@1)) @@1)
> >> > > (with { tree stype = build_nonstandard_integer_type (1, 0); }
> >> > > (if (INTEGRAL_TYPE_P (type)
> >> > > && !TYPE_UNSIGNED (type)
> >> > > && wi::eq_p (wi::to_wide (@1), element_precision (type) - 1))
> >> > > (convert (convert:stype @0)))))
> >> > >
> >> > > work?
> >> >
> >> > Certainly swap the if and with and the (with then should be indented
> by 1
> >> > column to the right of (if and (convert one further (the reason for
> the
> >> > swapping is not to call build_nonstandard_integer_type when it will
> not be
> >> > needed, which will be probably far more often then an actual match).
> >>
> >> With that fixed I think for non-vector integrals the above is the most
> suitable
> >> canonical form of a sign-extension. Note it should also work for any
> other
> >> constant shift amount - just use the appropriate intermediate precision
> for
> >> the truncating type. You might also want to verify what RTL expansion
> >> produces before/after - it at least shouldn't be worse. We _might_ want
> >> to consider to only use the converts when the intermediate type has
> >> mode precision (and as a special case allow one bit as in your above
> case)
> >> so it can expand to (sign_extend:<outer> (subreg:<inner> reg)).
> >>
> >> > As discussed privately, the above isn't what we want for vectors and
> the 2
> >> > shifts are probably best on most arches because even when using -(x &
> 1) the
> >> > { 1, 1, 1, ... } vector would often needed to be loaded from memory.
> >>
> >> I think for vectors a vpcmpgt {0,0,0,..}, %xmm is the cheapest way of
> >> producing the result. Note that to reflect this on GIMPLE you'd need
> >>
> >> _2 = _1 < { 0,0...};
> >> res = _2 ? { -1, -1, ...} : { 0, 0,...};
> >>
> >> because whether the ISA has a way to produce all-ones masks isn't known.
> >>
> >> For scalars using -(T)(_1 < 0) would also be possible.
> >>
> >> That said - do you have any testcase where the canonicalization is an
> enabler
> >> for further transforms or was this requested stand-alone?
> >>
> >> Thanks,
> >> Richard.
> >>
> >> > Jakub
> >> >
> >>
>
>
