On Tue, 5 Jan 2021, Jakub Jelinek wrote:
> Hi!
>
> The following patch improves the A / (1 << B) -> A >> B simplification,
> as seen in the testcase, if there is unnecessary widening for the division,
> we just optimize it into a shift on the widened type, but if the lshift
> is widened too, there is no reason to do that, we can just shift it in the
> original type and convert after. The tree_nonzero_bits & wi::mask check
> already ensures it is fine even for signed values.
>
> I've split the vr-values optimization into a separate patch as it causes
> a small regression on two testcases, but this patch fixes what has been
> reported in the PR alone.
>
> Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?
OK.
Thanks,
Richard.
> 2021-01-05 Jakub Jelinek <ja...@redhat.com>
>
> PR tree-optimization/96930
> * match.pd ((A / (1 << B)) -> (A >> B)): If A is extended
> from narrower value which has the same type as 1 << B, perform
> the right shift on the narrower value followed by extension.
>
> * g++.dg/tree-ssa/pr96930.C: New test.
>
> --- gcc/match.pd.jj 2021-01-04 10:37:06.000000000 +0100
> +++ gcc/match.pd 2021-01-05 10:27:56.653791400 +0100
> @@ -321,7 +321,7 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT)
> (unsigned long long) (1 << 31) is -2147483648ULL, not 2147483648ULL,
> so it is valid only if A >> 31 is zero. */
> (simplify
> - (trunc_div @0 (convert? (lshift integer_onep@1 @2)))
> + (trunc_div (convert?@0 @3) (convert2? (lshift integer_onep@1 @2)))
> (if ((TYPE_UNSIGNED (type) || tree_expr_nonnegative_p (@0))
> && (!VECTOR_TYPE_P (type)
> || target_supports_op_p (type, RSHIFT_EXPR, optab_vector)
> @@ -336,7 +336,11 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT)
> & wi::mask (element_precision (TREE_TYPE (@1)) - 1,
> true,
> element_precision (type))) == 0)))))
> - (rshift @0 @2)))
> + (if (!VECTOR_TYPE_P (type)
> + && useless_type_conversion_p (TREE_TYPE (@3), TREE_TYPE (@1))
> + && element_precision (TREE_TYPE (@3)) < element_precision (type))
> + (convert (rshift @3 @2))
> + (rshift @0 @2))))
>
> /* Preserve explicit divisions by 0: the C++ front-end wants to detect
> undefined behavior in constexpr evaluation, and assuming that the division
> --- gcc/testsuite/g++.dg/tree-ssa/pr96930.C.jj 2021-01-04
> 14:18:15.513100038 +0100
> +++ gcc/testsuite/g++.dg/tree-ssa/pr96930.C 2021-01-04 14:25:35.512148709
> +0100
> @@ -0,0 +1,10 @@
> +// PR tree-optimization/96930
> +// { dg-do compile }
> +// { dg-options "-O2 -fdump-tree-optimized" }
> +// { dg-final { scan-tree-dump " = a_\[0-9]\\\(D\\\) >> b_\[0-9]\\\(D\\\);"
> "optimized" } }
> +
> +unsigned
> +foo (unsigned a, unsigned b)
> +{
> + return a / (unsigned long long) (1U << b);
> +}
>
> Jakub
>
>
--
Richard Biener <rguent...@suse.de>
SUSE Software Solutions Germany GmbH, Maxfeldstrasse 5, 90409 Nuernberg,
Germany; GF: Felix Imendörffer; HRB 36809 (AG Nuernberg)
