From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-vk1-xa2f.google.com (mail-vk1-xa2f.google.com [IPv6:2607:f8b0:4864:20::a2f]) by sourceware.org (Postfix) with ESMTPS id 981903858C2A for ; Sat, 24 Jun 2023 10:07:55 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.2 sourceware.org 981903858C2A Authentication-Results: sourceware.org; dmarc=pass (p=none dis=none) header.from=cs.washington.edu Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=cs.washington.edu Received: by mail-vk1-xa2f.google.com with SMTP id 71dfb90a1353d-47179f18c76so534869e0c.3 for ; Sat, 24 Jun 2023 03:07:55 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=cs.washington.edu; s=goo201206; t=1687601274; x=1690193274; h=content-transfer-encoding:cc:to:subject:message-id:date:from :in-reply-to:references:mime-version:from:to:cc:subject:date :message-id:reply-to; bh=raUuSbP6E7MoD3ut3dnLXeWYLrKrqui+/yVsVaf7dTQ=; b=PsM1NMUMV0UewZquMzziN7pooCBPZB80aCo3zhxjVWybxlN1/8wJ/y0G3h0XSGrjz3 VXLg51UfnRXGAho1m5cg9XAySe1qCJW0XhC+mE07Y1TVx3ZikjH/3JLN8hFTOJpEnW33 P8r8LKMbadLQ18QnOfedgwUYlTDCLmigZYWys= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1687601274; x=1690193274; h=content-transfer-encoding:cc:to:subject:message-id:date:from :in-reply-to:references:mime-version:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=raUuSbP6E7MoD3ut3dnLXeWYLrKrqui+/yVsVaf7dTQ=; b=MeuDAt8YgTlujfePINAsXA3W2Zqlh7qCKY4zcB0GANcn4M6ao/enaolnmpwg0CCp6g Sg2TS2TQdE1m+9TfyNmC0jp511W81nngWCc84kmNZsmn8n8hZ+Tk+08aEo8auPLufQ2z moEiz8eSU+syaF6N4AM0b2ruQTLyCiLU3AnemOhk5m+z0gcB9nxy5WfZXYOYhDMePcGf +teUYTS35EeperUIazC42tAqkZ/MLs0o+JlGuKas8hpFvm+19fRP+oYlx/69bWMFOIJO iq2112GJodvrwuYq3vpdKymeX1x9M43f77AifZ3NmyTqv3MQmLRWRmPlLyKXWECJmUSA W7Zg== X-Gm-Message-State: AC+VfDz/Uc9esoFoHQuHJ4CuGTawzynryqxawsKt3dZcg3WJjTpV4ytV 6+b18mLaJGhsRBnOiuiQ50N/OpHNGcx7xdfP3h7cBjRFj1kmfbtgtIc= X-Google-Smtp-Source: ACHHUZ59IRmOjATHZ9Qv8riJHIXbUVBPBdKTpT9b36vi7eaHTZnybFVqeT48/9JcGt+z/OD0g0RK9NdsCuXrpE5glf8= X-Received: by 2002:a67:f783:0:b0:437:db1d:7edf with SMTP id j3-20020a67f783000000b00437db1d7edfmr11998268vso.26.1687601274410; Sat, 24 Jun 2023 03:07:54 -0700 (PDT) MIME-Version: 1.0 References: <20230615122129.20213-1-kmatsui@cs.washington.edu> In-Reply-To: From: Ken Matsui Date: Sat, 24 Jun 2023 03:07:38 -0700 Message-ID: Subject: Re: [PATCH 1/2] c++: implement __remove_pointer built-in trait To: Patrick Palka Cc: gcc-patches@gcc.gnu.org Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Spam-Status: No, score=-9.3 required=5.0 tests=BAYES_00,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,GIT_PATCH_0,RCVD_IN_DNSWL_NONE,SPF_HELO_NONE,SPF_PASS,TXREP,T_SCC_BODY_TEXT_LINE autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on server2.sourceware.org List-Id: On Tue, Jun 20, 2023 at 8:22=E2=80=AFAM Patrick Palka w= rote: > > On Sat, 17 Jun 2023, Ken Matsui via Gcc-patches wrote: > > > Hi, > > > > I conducted a benchmark for remove_pointer as well as is_object. Just > > like the is_object benchmark, here is the benchmark code: > > > > https://github.com/ken-matsui/gcc-benches/blob/main/remove_pointer_benc= hmark.cc > > > > On my computer, using the gcc HEAD of this patch for a release build, > > the patch with -DUSE_BUILTIN took 8.7% less time and used 4.3-4.9% > > less memory on average compared to not using it. Although the > > performance improvement was not as significant as with is_object, the > > benchmark demonstrated that the compilation was consistently more > > efficient. > > Thanks for the benchmark. The improvement is lesser than I expected, > but that might be because the benchmark is "biased": > > template > struct Instantiator : Instantiator { > static_assert(!std::is_pointer_v>); > }; > > This only invokes remove_pointer_t on the non-pointer type Instantiator, > and so the benchmark doesn't factor in the performance of the trait when > invoked on pointer types, and traits typically will have different > performance characteristics depending on the kind of type it's given. > > To more holistically assess the real-world performance of the trait the > benchmark should also consider pointer types and maybe also cv-qualified > types (given that the original implementation is in terms of > __remove_cv_t and thus performance of the original implementation may be > sensitive to cv-qualification). So we should probably uniformly > benchmark these classes of types, via doing e.g.: > > static_assert(!std::is_pointer_v>); > static_assert(!std::is_pointer_v>); > static_assert(!std::is_pointer_v>)= ; > static_assert(!std::is_pointer_v>= ); > > (We could consider other kinds of types too, e.g. reference types and > integral types, but it seems clear based on the implementations being > benchmarked that performance won't be sensitive to reference-ness > or integral-ness.) Thank you for your review! I totally agree with your opinion that the benchmark should have been exhaustive. I conducted a benchmark for this new change: https://github.com/ken-matsui/gcc-benches/blob/main/remove_pointer.md#tue-j= un-20-030313-pm-pdt-2023 Time: -27.918% Peak Memory Usage: -19.0755% Total Memory Usage: -20.0199% > > > > Sincerely, > > Ken Matsui > > > > On Thu, Jun 15, 2023 at 5:22=E2=80=AFAM Ken Matsui wrote: > > > > > > This patch implements built-in trait for std::remove_pointer. > > > > > > gcc/cp/ChangeLog: > > > > > > * cp-trait.def: Define __remove_pointer. > > > * semantics.cc (finish_trait_type): Handle CPTK_REMOVE_POINTE= R. > > > > > > gcc/testsuite/ChangeLog: > > > > > > * g++.dg/ext/has-builtin-1.C: Test existence of __remove_poin= ter. > > > * g++.dg/ext/remove_pointer.C: New test. > > > > > > Signed-off-by: Ken Matsui > > > --- > > > gcc/cp/cp-trait.def | 1 + > > > gcc/cp/semantics.cc | 4 ++ > > > gcc/testsuite/g++.dg/ext/has-builtin-1.C | 3 ++ > > > gcc/testsuite/g++.dg/ext/remove_pointer.C | 51 +++++++++++++++++++++= ++ > > > 4 files changed, 59 insertions(+) > > > create mode 100644 gcc/testsuite/g++.dg/ext/remove_pointer.C > > > > > > diff --git a/gcc/cp/cp-trait.def b/gcc/cp/cp-trait.def > > > index 8b7fece0cc8..07823e55579 100644 > > > --- a/gcc/cp/cp-trait.def > > > +++ b/gcc/cp/cp-trait.def > > > @@ -90,6 +90,7 @@ DEFTRAIT_EXPR (IS_DEDUCIBLE, "__is_deducible ", 2) > > > DEFTRAIT_TYPE (REMOVE_CV, "__remove_cv", 1) > > > DEFTRAIT_TYPE (REMOVE_REFERENCE, "__remove_reference", 1) > > > DEFTRAIT_TYPE (REMOVE_CVREF, "__remove_cvref", 1) > > > +DEFTRAIT_TYPE (REMOVE_POINTER, "__remove_pointer", 1) > > > DEFTRAIT_TYPE (UNDERLYING_TYPE, "__underlying_type", 1) > > > DEFTRAIT_TYPE (TYPE_PACK_ELEMENT, "__type_pack_element", -1) > > > > > > diff --git a/gcc/cp/semantics.cc b/gcc/cp/semantics.cc > > > index 8fb47fd179e..885c7a6fb64 100644 > > > --- a/gcc/cp/semantics.cc > > > +++ b/gcc/cp/semantics.cc > > > @@ -12373,6 +12373,10 @@ finish_trait_type (cp_trait_kind kind, tree = type1, tree type2, > > > if (TYPE_REF_P (type1)) > > > type1 =3D TREE_TYPE (type1); > > > return cv_unqualified (type1); > > > + case CPTK_REMOVE_POINTER: > > > + if (TYPE_PTR_P (type1)) > > > + type1 =3D TREE_TYPE (type1); > > > + return type1; > > Maybe add a newline before the 'case' to visually separate it from the > previous 'case'? LGTM otherwise, thanks! Will do! Thank you! > > > > > > case CPTK_TYPE_PACK_ELEMENT: > > > return finish_type_pack_element (type1, type2, complain); > > > diff --git a/gcc/testsuite/g++.dg/ext/has-builtin-1.C b/gcc/testsuite= /g++.dg/ext/has-builtin-1.C > > > index f343e153e56..e21e0a95509 100644 > > > --- a/gcc/testsuite/g++.dg/ext/has-builtin-1.C > > > +++ b/gcc/testsuite/g++.dg/ext/has-builtin-1.C > > > @@ -146,3 +146,6 @@ > > > #if !__has_builtin (__remove_cvref) > > > # error "__has_builtin (__remove_cvref) failed" > > > #endif > > > +#if !__has_builtin (__remove_pointer) > > > +# error "__has_builtin (__remove_pointer) failed" > > > +#endif > > > diff --git a/gcc/testsuite/g++.dg/ext/remove_pointer.C b/gcc/testsuit= e/g++.dg/ext/remove_pointer.C > > > new file mode 100644 > > > index 00000000000..7b13db93950 > > > --- /dev/null > > > +++ b/gcc/testsuite/g++.dg/ext/remove_pointer.C > > > @@ -0,0 +1,51 @@ > > > +// { dg-do compile { target c++11 } } > > > + > > > +#define SA(X) static_assert((X),#X) > > > + > > > +SA(__is_same(__remove_pointer(int), int)); > > > +SA(__is_same(__remove_pointer(int*), int)); > > > +SA(__is_same(__remove_pointer(int**), int*)); > > > + > > > +SA(__is_same(__remove_pointer(const int*), const int)); > > > +SA(__is_same(__remove_pointer(const int**), const int*)); > > > +SA(__is_same(__remove_pointer(int* const), int)); > > > +SA(__is_same(__remove_pointer(int** const), int*)); > > > +SA(__is_same(__remove_pointer(int* const* const), int* const)); > > > + > > > +SA(__is_same(__remove_pointer(volatile int*), volatile int)); > > > +SA(__is_same(__remove_pointer(volatile int**), volatile int*)); > > > +SA(__is_same(__remove_pointer(int* volatile), int)); > > > +SA(__is_same(__remove_pointer(int** volatile), int*)); > > > +SA(__is_same(__remove_pointer(int* volatile* volatile), int* volatil= e)); > > > + > > > +SA(__is_same(__remove_pointer(const volatile int*), const volatile i= nt)); > > > +SA(__is_same(__remove_pointer(const volatile int**), const volatile = int*)); > > > +SA(__is_same(__remove_pointer(const int* volatile), const int)); > > > +SA(__is_same(__remove_pointer(volatile int* const), volatile int)); > > > +SA(__is_same(__remove_pointer(int* const volatile), int)); > > > +SA(__is_same(__remove_pointer(const int** volatile), const int*)); > > > +SA(__is_same(__remove_pointer(volatile int** const), volatile int*))= ; > > > +SA(__is_same(__remove_pointer(int** const volatile), int*)); > > > +SA(__is_same(__remove_pointer(int* const* const volatile), int* cons= t)); > > > +SA(__is_same(__remove_pointer(int* volatile* const volatile), int* v= olatile)); > > > +SA(__is_same(__remove_pointer(int* const volatile* const volatile), = int* const volatile)); > > > + > > > +SA(__is_same(__remove_pointer(int&), int&)); > > > +SA(__is_same(__remove_pointer(const int&), const int&)); > > > +SA(__is_same(__remove_pointer(volatile int&), volatile int&)); > > > +SA(__is_same(__remove_pointer(const volatile int&), const volatile i= nt&)); > > > + > > > +SA(__is_same(__remove_pointer(int&&), int&&)); > > > +SA(__is_same(__remove_pointer(const int&&), const int&&)); > > > +SA(__is_same(__remove_pointer(volatile int&&), volatile int&&)); > > > +SA(__is_same(__remove_pointer(const volatile int&&), const volatile = int&&)); > > > + > > > +SA(__is_same(__remove_pointer(int[3]), int[3])); > > > +SA(__is_same(__remove_pointer(const int[3]), const int[3])); > > > +SA(__is_same(__remove_pointer(volatile int[3]), volatile int[3])); > > > +SA(__is_same(__remove_pointer(const volatile int[3]), const volatile= int[3])); > > > + > > > +SA(__is_same(__remove_pointer(int(int)), int(int))); > > > +SA(__is_same(__remove_pointer(int(*const)(int)), int(int))); > > > +SA(__is_same(__remove_pointer(int(*volatile)(int)), int(int))); > > > +SA(__is_same(__remove_pointer(int(*const volatile)(int)), int(int)))= ; > > > -- > > > 2.41.0 > > > > > > >