public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Jason Merrill <jason@redhat.com>
To: Marek Polacek <polacek@redhat.com>
Cc: GCC Patches <gcc-patches@gcc.gnu.org>
Subject: Re: [PATCH v2] c++: fix parsing with auto(x) [PR112410]
Date: Mon, 13 Nov 2023 21:26:41 -0500	[thread overview]
Message-ID: <5b947ae5-b4bb-4bf2-91b5-6dcace7b537d@redhat.com> (raw)
In-Reply-To: <ZU7VOhuVUvCPnqqG@redhat.com>

On 11/10/23 20:13, Marek Polacek wrote:
> On Thu, Nov 09, 2023 at 07:07:03PM -0500, Jason Merrill wrote:
>> On 11/9/23 14:58, Marek Polacek wrote:
>>> Bootstrapped/regtested on x86_64-pc-linux-gnu, ok for trunk?
>>>
>>> -- >8 --
>>> Here we are wrongly parsing
>>>
>>>     int y(auto(42));
>>>
>>> which uses the C++23 cast-to-prvalue feature, and initializes y to 42.
>>> However, we were treating the auto as an implicit template parameter.
>>>
>>> Fixing the auto{42} case is easy, but when auto is followed by a (,
>>> I found the fix to be much more involved.  For instance, we cannot
>>> use cp_parser_expression, because that can give hard errors.  It's
>>> also necessary to disambiguate 'auto(i)' as 'auto i', not a cast.
>>> auto(), auto(int), auto(f)(int), auto(*), auto(i[]), auto(...), etc.
>>> are all function declarations.  We have to look at more than one
>>> token to decide.
>>
>> Yeah, this is a most vexing parse problem.  The code is synthesizing
>> template parameters before we've resolved whether the auto is a
>> decl-specifier or not.
>>
>>> In this fix, I'm (ab)using cp_parser_declarator, with member_p=false
>>> so that it doesn't commit.  But it handles even more complicated
>>> cases as
>>>
>>>     int fn (auto (*const **&f)(int) -> char);
>>
>> But it doesn't seem to handle the extremely vexing
>>
>> struct A {
>>    A(int,int);
>> };
>>
>> int main()
>> {
>>    int a;
>>    A b(auto(a), 42);
>> }
> 
> Argh.  This test should indeed be accepted and is currently rejected,
> but it's a different problem: 'b' is at block scope and you can't
> have a template there.  But when I put it into a namespace scope,
> it shows that my patch doesn't work correctly.  I've added auto-fncast14.C
> for the latter and opened c++/112482 for the block-scope problem.
>   
>> I think we need to stop synthesizing immediately when we see RID_AUTO, and
>> instead go back after we successfully parse a declaration and synthesize for
>> any autos we saw along the way.  :/
> 
> That seems very complicated :(.  I had a different idea though; how
> about the following patch?  The idea is that if we see that parsing
> the parameter-declaration-list didn't work, we undo what synthesize_
> did, and let cp_parser_initializer parse "(auto(42))", which should
> succeed.  I checked that after cp_finish_decl y is initialized to 42.

Nice, that's much simpler.  Do you also still need the changes to 
cp_parser_simple_type_specifier?

> Bootstrapped/regtested on x86_64-pc-linux-gnu, ok for trunk?
> 
> -- >8 --
> Here we are wrongly parsing
> 
>    int y(auto(42));
> 
> which uses the C++23 cast-to-prvalue feature, and initializes y to 42.
> However, we were treating the auto as an implicit template parameter.
> 
> Fixing the auto{42} case is easy, but when auto is followed by a (,
> I found the fix to be much more involved.  For instance, we cannot
> use cp_parser_expression, because that can give hard errors.  It's
> also necessary to disambiguate 'auto(i)' as 'auto i', not a cast.
> auto(), auto(int), auto(f)(int), auto(*), auto(i[]), auto(...), etc.
> are all function declarations.
> 
> This patch rectifies that by undoing the implicit function template
> modification.  In the test above, we should notice that the parameter
> list is ill-formed, and since we've synthesized an implicit template
> parameter, we undo it by calling abort_fully_implicit_template.  Then,
> we'll parse the "(auto(42))" as an initializer.
> 
> 	PR c++/112410
> 
> gcc/cp/ChangeLog:
> 
> 	* parser.cc (cp_parser_simple_type_specifier): Disambiguate
> 	between a variable and function declaration with auto.
> 	(cp_parser_parameter_declaration_clause): Maybe call
> 	abort_fully_implicit_template if it turned out the parameter list was
> 	ill-formed.
> 
> gcc/testsuite/ChangeLog:
> 
> 	* g++.dg/cpp23/auto-fncast13.C: New test.
> 	* g++.dg/cpp23/auto-fncast14.C: New test.
> ---
>   gcc/cp/parser.cc                           | 27 +++++++++-
>   gcc/testsuite/g++.dg/cpp23/auto-fncast13.C | 61 ++++++++++++++++++++++
>   gcc/testsuite/g++.dg/cpp23/auto-fncast14.C |  9 ++++
>   3 files changed, 96 insertions(+), 1 deletion(-)
>   create mode 100644 gcc/testsuite/g++.dg/cpp23/auto-fncast13.C
>   create mode 100644 gcc/testsuite/g++.dg/cpp23/auto-fncast14.C
> 
> diff --git a/gcc/cp/parser.cc b/gcc/cp/parser.cc
> index 5116bcb78f6..947351b09b8 100644
> --- a/gcc/cp/parser.cc
> +++ b/gcc/cp/parser.cc
> @@ -19991,6 +19991,8 @@ cp_parser_simple_type_specifier (cp_parser* parser,
>   	  /* The 'auto' might be the placeholder return type for a function decl
>   	     with trailing return type.  */
>   	  bool have_trailing_return_fn_decl = false;
> +	  /* Or it might be auto(x) or auto {x}.  */
> +	  bool decay_copy = false;
>   
>   	  cp_parser_parse_tentatively (parser);
>   	  cp_lexer_consume_token (parser->lexer);
> @@ -20008,6 +20010,11 @@ cp_parser_simple_type_specifier (cp_parser* parser,
>   							 /*consume_paren*/true);
>   		  continue;
>   		}
> +	      else if (cp_lexer_next_token_is (parser->lexer, CPP_OPEN_BRACE))
> +		{
> +		  decay_copy = true;
> +		  break;
> +		}
>   
>   	      if (cp_lexer_next_token_is (parser->lexer, CPP_DEREF))
>   		{
> @@ -20019,6 +20026,11 @@ cp_parser_simple_type_specifier (cp_parser* parser,
>   	    }
>   	  cp_parser_abort_tentative_parse (parser);
>   
> +	  if (decay_copy)
> +	    {
> +	      type = error_mark_node;
> +	      break;
> +	    }
>   	  if (have_trailing_return_fn_decl)
>   	    {
>   	      type = make_auto ();
> @@ -24973,7 +24985,20 @@ cp_parser_parameter_declaration_clause (cp_parser* parser,
>        parameter-declaration-list, then the entire
>        parameter-declaration-clause is erroneous.  */
>     if (parameters == error_mark_node)
> -    return NULL_TREE;
> +    {
> +      /* For code like
> +	  int x(auto(42));
> +	  A a(auto(i), 42);
> +	 we have synthesized an implicit template parameter and marked
> +	 what we thought was a function as an implicit function template.
> +	 But now, having seen the whole parameter list, we know it's not
> +	 a function declaration, so undo that.  */
> +      if (parser->fully_implicit_function_template_p
> +	  /* Don't do this for the inner ().  */
> +	  && parser->default_arg_ok_p)
> +	abort_fully_implicit_template (parser);
> +      return NULL_TREE;
> +    }
>   
>     /* Peek at the next token.  */
>     token = cp_lexer_peek_token (parser->lexer);
> diff --git a/gcc/testsuite/g++.dg/cpp23/auto-fncast13.C b/gcc/testsuite/g++.dg/cpp23/auto-fncast13.C
> new file mode 100644
> index 00000000000..1bceffb70cf
> --- /dev/null
> +++ b/gcc/testsuite/g++.dg/cpp23/auto-fncast13.C
> @@ -0,0 +1,61 @@
> +// PR c++/112410
> +// { dg-do compile { target c++23 } }
> +
> +int f1 (auto(int) -> char);
> +int f2 (auto x);
> +int f3 (auto);
> +int f4 (auto(i));
> +
> +int v1 (auto(42));
> +int v2 (auto{42});
> +int e1 (auto{i}); // { dg-error "not declared" }
> +int i;
> +int v3 (auto{i});
> +int v4 (auto(i + 1));
> +int v5 (auto(+i));
> +int v6 (auto(i = 4));
> +
> +int f5 (auto(i));
> +int f6 (auto());
> +int f7 (auto(int));
> +int f8 (auto(f)(int));
> +int f9 (auto(...) -> char);
> +// FIXME: ICEs (PR c++/89867)
> +//int f10 (auto(__attribute__((unused)) i));
> +int f11 (auto((i)));
> +int f12 (auto(i[]));
> +int f13 (auto(*i));
> +int f14 (auto(*));
> +
> +int e2 (auto{}); // { dg-error "invalid use of .auto." }
> +int e3 (auto(i, i)); // { dg-error "invalid use of .auto." }
> +
> +char bar (int);
> +char baz ();
> +char qux (...);
> +
> +void
> +g (int i)
> +{
> +  f1 (bar);
> +  f2 (42);
> +  f3 (42);
> +  f4 (42);
> +  f5 (42);
> +  f6 (baz);
> +  f7 (bar);
> +  f8 (bar);
> +  f9 (qux);
> +//  f10 (42);
> +  f11 (42);
> +  f12 (&i);
> +  f13 (&i);
> +  f14 (&i);
> +
> +  v1 = 1;
> +  v2 = 2;
> +  v3 = 3;
> +  v4 = 4;
> +  v5 = 5;
> +  v6 = 6;
> +}
> diff --git a/gcc/testsuite/g++.dg/cpp23/auto-fncast14.C b/gcc/testsuite/g++.dg/cpp23/auto-fncast14.C
> new file mode 100644
> index 00000000000..9e7a06c87d5
> --- /dev/null
> +++ b/gcc/testsuite/g++.dg/cpp23/auto-fncast14.C
> @@ -0,0 +1,9 @@
> +// PR c++/112410
> +// { dg-do compile { target c++23 } }
> +
> +struct A {
> +   A(int,int);
> +};
> +
> +int a;
> +A b1(auto(a), 42);
> 
> base-commit: e0c1476d5d7c450b1b16a40364cea4e91237ea93


  reply	other threads:[~2023-11-14  2:26 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-11-09 19:58 [PATCH] " Marek Polacek
2023-11-10  0:07 ` Jason Merrill
2023-11-11  1:13   ` [PATCH v2] " Marek Polacek
2023-11-14  2:26     ` Jason Merrill [this message]
2023-11-14 15:58       ` Marek Polacek
2023-11-14 22:27         ` Jason Merrill
2023-11-15 22:24           ` [PATCH v3] " Marek Polacek
2023-11-15 23:18             ` 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=5b947ae5-b4bb-4bf2-91b5-6dcace7b537d@redhat.com \
    --to=jason@redhat.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=polacek@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).