public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Jason Merrill <jason@redhat.com>
To: Patrick Palka <ppalka@redhat.com>, gcc-patches@gcc.gnu.org
Subject: Re: [PATCH] c++: some missing-SFINAE fixes
Date: Tue, 13 Sep 2022 08:53:14 -0400	[thread overview]
Message-ID: <30bb32f6-3f08-5a45-e39a-ae8f49bc4af0@redhat.com> (raw)
In-Reply-To: <20220913114538.2741902-1-ppalka@redhat.com>

On 9/13/22 07:45, Patrick Palka wrote:
> It looks like we aren't respecting SFINAE for:
> 
>    * an invalid/non-constant conditional explicit-specifier
>    * a non-constant conditional noexcept-specifier
>    * a non-constant argument to __integer_pack
> 
> This patch fixes these issues in the usual way, by passing complain
> and propagating error_mark_node appropriately.
> 
> Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look OK for
> trunk?
> 
> gcc/cp/ChangeLog:
> 
> 	* decl.cc (build_explicit_specifier): Pass complain to
> 	cxx_constant_value.
> 	* except.cc (build_noexcept_spec): Likewise.
> 	* pt.cc (expand_integer_pack): Likewise.
> 	(tsubst_function_decl): Propagate error_mark_node returned
> 	from build_explicit_specifier.
> 
> gcc/testsuite/ChangeLog:
> 
> 	* g++.dg/cpp1z/noexcept-type26.C: New test.
> 	* g++.dg/cpp2a/explicit19.C: New test.
> 	* g++.dg/ext/integer-pack6.C: New test.
> ---
>   gcc/cp/decl.cc                               |  2 +-
>   gcc/cp/except.cc                             |  2 +-
>   gcc/cp/pt.cc                                 |  6 ++++--
>   gcc/testsuite/g++.dg/cpp1z/noexcept-type26.C | 12 ++++++++++++
>   gcc/testsuite/g++.dg/cpp2a/explicit19.C      | 12 ++++++++++++
>   gcc/testsuite/g++.dg/ext/integer-pack6.C     | 13 +++++++++++++
>   6 files changed, 43 insertions(+), 4 deletions(-)
>   create mode 100644 gcc/testsuite/g++.dg/cpp1z/noexcept-type26.C
>   create mode 100644 gcc/testsuite/g++.dg/cpp2a/explicit19.C
>   create mode 100644 gcc/testsuite/g++.dg/ext/integer-pack6.C
> 
> diff --git a/gcc/cp/decl.cc b/gcc/cp/decl.cc
> index 936f1cf0197..5404d7e084c 100644
> --- a/gcc/cp/decl.cc
> +++ b/gcc/cp/decl.cc
> @@ -18557,7 +18557,7 @@ build_explicit_specifier (tree expr, tsubst_flags_t complain)
>   
>     expr = build_converted_constant_bool_expr (expr, complain);
>     expr = instantiate_non_dependent_expr (expr, complain);
> -  expr = cxx_constant_value (expr);
> +  expr = cxx_constant_value (expr, NULL_TREE, complain);

The patch is OK, but perhaps we want another overload that omits the 
object parameter?

>     return expr;
>   }
>   
> diff --git a/gcc/cp/except.cc b/gcc/cp/except.cc
> index 7fdbc747c22..4d7f0ce102d 100644
> --- a/gcc/cp/except.cc
> +++ b/gcc/cp/except.cc
> @@ -1257,7 +1257,7 @@ build_noexcept_spec (tree expr, tsubst_flags_t complain)
>       {
>         expr = build_converted_constant_bool_expr (expr, complain);
>         expr = instantiate_non_dependent_expr (expr, complain);
> -      expr = cxx_constant_value (expr);
> +      expr = cxx_constant_value (expr, NULL_TREE, complain);
>       }
>     if (TREE_CODE (expr) == INTEGER_CST)
>       {
> diff --git a/gcc/cp/pt.cc b/gcc/cp/pt.cc
> index 1c6f4b84612..f16f5ffca20 100644
> --- a/gcc/cp/pt.cc
> +++ b/gcc/cp/pt.cc
> @@ -3869,7 +3869,7 @@ expand_integer_pack (tree call, tree args, tsubst_flags_t complain,
>     else
>       {
>         hi = instantiate_non_dependent_expr (hi, complain);
> -      hi = cxx_constant_value (hi);
> +      hi = cxx_constant_value (hi, NULL_TREE, complain);
>         int len = valid_constant_size_p (hi) ? tree_to_shwi (hi) : -1;
>   
>         /* Calculate the largest value of len that won't make the size of the vec
> @@ -14312,7 +14312,9 @@ tsubst_function_decl (tree t, tree args, tsubst_flags_t complain,
>   				    /*function_p=*/false,
>   				    /*i_c_e_p=*/true);
>         spec = build_explicit_specifier (spec, complain);
> -      if (instantiation_dependent_expression_p (spec))
> +      if (spec == error_mark_node)
> +	return error_mark_node;
> +      else if (instantiation_dependent_expression_p (spec))
>   	store_explicit_specifier (r, spec);
>         else
>   	{
> diff --git a/gcc/testsuite/g++.dg/cpp1z/noexcept-type26.C b/gcc/testsuite/g++.dg/cpp1z/noexcept-type26.C
> new file mode 100644
> index 00000000000..73058395f0a
> --- /dev/null
> +++ b/gcc/testsuite/g++.dg/cpp1z/noexcept-type26.C
> @@ -0,0 +1,12 @@
> +// Verify a non-constant conditional noexcept-specifier of a function type
> +// is a SFINAE error.
> +// { dg-do compile { target c++17 } }
> +
> +template<class T> void f(void() noexcept(T::value)) = delete;
> +template<class T> void f(...);
> +
> +struct B { static bool value; };
> +
> +int main() {
> +  f<B>(nullptr);
> +}
> diff --git a/gcc/testsuite/g++.dg/cpp2a/explicit19.C b/gcc/testsuite/g++.dg/cpp2a/explicit19.C
> new file mode 100644
> index 00000000000..47903813680
> --- /dev/null
> +++ b/gcc/testsuite/g++.dg/cpp2a/explicit19.C
> @@ -0,0 +1,12 @@
> +// Verify a conditional explicit-specifier is a SFINAE context.
> +// { dg-do compile { target c++20 } }
> +
> +struct A {
> +  template<class T> explicit(T::value) A(T) = delete;
> +  A(...);
> +};
> +
> +struct B { static bool value; };
> +
> +A x(0);
> +A y(B{});
> diff --git a/gcc/testsuite/g++.dg/ext/integer-pack6.C b/gcc/testsuite/g++.dg/ext/integer-pack6.C
> new file mode 100644
> index 00000000000..7e16bc44b40
> --- /dev/null
> +++ b/gcc/testsuite/g++.dg/ext/integer-pack6.C
> @@ -0,0 +1,13 @@
> +// Verify a non-constant argument to __integer_pack is a SFINAE error.
> +// { dg-do compile { target c++11 } }
> +
> +template<int...> struct A { };
> +
> +template<class T> auto f(int) -> A<__integer_pack(T::value)...> = delete;
> +template<class T> void f(...);
> +
> +struct B { static int value; };
> +
> +int main() {
> +  f<B>(0);
> +}


  reply	other threads:[~2022-09-13 12:53 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-09-13 11:45 Patrick Palka
2022-09-13 12:53 ` Jason Merrill [this message]
2022-09-13 13:46   ` Patrick Palka
2022-09-13 13:49     ` Jason Merrill

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=30bb32f6-3f08-5a45-e39a-ae8f49bc4af0@redhat.com \
    --to=jason@redhat.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=ppalka@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).