[PATCH] c++: Fix ICE with non-constant satisfaction [PR98644]

[PATCH] c++: Fix ICE with non-constant satisfaction [PR98644]

[Patrick Palka]

In the below testcase, the expression of the atomic constraint after
substitution is (int *) NON_LVALUE_EXPR <1> != 0B which is not a C++
constant expression, but its TREE_CONSTANT flag is set (from build2),
so satisfy_atom fails to notice that it's non-constant (and we end
up tripping over the assert in satisfaction_value).

Since TREE_CONSTANT doesn't necessarily correspond to C++ constantness,
this patch makes satisfy_atom instead check is_rvalue_constant_expression.

Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look OK for
trunk/10?

gcc/cp/ChangeLog:

	PR c++/98644
	* constraint.cc (satisfy_atom): Check is_rvalue_constant_expression
	instead of TREE_CONSTANT.

gcc/testsuite/ChangeLog:

	PR c++/98644
	* g++.dg/cpp2a/concepts-pr98644.C: New test.
---
 gcc/cp/constraint.cc                          | 2 +-
 gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C | 7 +++++++
 2 files changed, 8 insertions(+), 1 deletion(-)
 create mode 100644 gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C

diff --git a/gcc/cp/constraint.cc b/gcc/cp/constraint.cc
index 9049d087859..f99a25dc8a4 100644
--- a/gcc/cp/constraint.cc
+++ b/gcc/cp/constraint.cc
@@ -2969,7 +2969,7 @@ satisfy_atom (tree t, tree args, sat_info info)
     {
       result = maybe_constant_value (result, NULL_TREE,
 				     /*manifestly_const_eval=*/true);
-      if (!TREE_CONSTANT (result))
+      if (!is_rvalue_constant_expression (result))
 	result = error_mark_node;
     }
   result = satisfaction_value (result);
diff --git a/gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C b/gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
new file mode 100644
index 00000000000..6772f72a3ce
--- /dev/null
+++ b/gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
@@ -0,0 +1,7 @@
+// PR c++/98644
+// { dg-do compile { target c++20 } }
+
+template<class T> concept Signed = bool(T(1)); // { dg-error "reinterpret_cast" }
+static_assert(Signed<int*>); // { dg-error "non-constant" }
+
+constexpr bool B = requires { requires bool((char *)1); }; // { dg-error "reinterpret_cast" }
-- 
2.30.0

Re: [PATCH] c++: Fix ICE with non-constant satisfaction [PR98644]

[Jason Merrill]

On 1/13/21 12:05 PM, Patrick Palka wrote:
> In the below testcase, the expression of the atomic constraint after
> substitution is (int *) NON_LVALUE_EXPR <1> != 0B which is not a C++
> constant expression, but its TREE_CONSTANT flag is set (from build2),
> so satisfy_atom fails to notice that it's non-constant (and we end
> up tripping over the assert in satisfaction_value).
> 
> Since TREE_CONSTANT doesn't necessarily correspond to C++ constantness,
> this patch makes satisfy_atom instead check is_rvalue_constant_expression.
> 
> Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look OK for
> trunk/10?
> 
> gcc/cp/ChangeLog:
> 
> 	PR c++/98644
> 	* constraint.cc (satisfy_atom): Check is_rvalue_constant_expression
> 	instead of TREE_CONSTANT.
> 
> gcc/testsuite/ChangeLog:
> 
> 	PR c++/98644
> 	* g++.dg/cpp2a/concepts-pr98644.C: New test.
> ---
>   gcc/cp/constraint.cc                          | 2 +-
>   gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C | 7 +++++++
>   2 files changed, 8 insertions(+), 1 deletion(-)
>   create mode 100644 gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
> 
> diff --git a/gcc/cp/constraint.cc b/gcc/cp/constraint.cc
> index 9049d087859..f99a25dc8a4 100644
> --- a/gcc/cp/constraint.cc
> +++ b/gcc/cp/constraint.cc
> @@ -2969,7 +2969,7 @@ satisfy_atom (tree t, tree args, sat_info info)
>       {
>         result = maybe_constant_value (result, NULL_TREE,
>   				     /*manifestly_const_eval=*/true);
> -      if (!TREE_CONSTANT (result))

This should be sufficient.  If the result isn't constant, 
maybe_constant_value shouldn't return it with TREE_CONSTANT set.  See

>       /* This isn't actually constant, so unset TREE_CONSTANT.                                                 

in cxx_eval_outermost_constant_expr.

Jason

Re: [PATCH] c++: Fix ICE with non-constant satisfaction [PR98644]

[Patrick Palka]

On Tue, 19 Jan 2021, Jason Merrill wrote:

> On 1/13/21 12:05 PM, Patrick Palka wrote:
> > In the below testcase, the expression of the atomic constraint after
> > substitution is (int *) NON_LVALUE_EXPR <1> != 0B which is not a C++
> > constant expression, but its TREE_CONSTANT flag is set (from build2),
> > so satisfy_atom fails to notice that it's non-constant (and we end
> > up tripping over the assert in satisfaction_value).
> > 
> > Since TREE_CONSTANT doesn't necessarily correspond to C++ constantness,
> > this patch makes satisfy_atom instead check is_rvalue_constant_expression.
> > 
> > Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look OK for
> > trunk/10?
> > 
> > gcc/cp/ChangeLog:
> > 
> > 	PR c++/98644
> > 	* constraint.cc (satisfy_atom): Check is_rvalue_constant_expression
> > 	instead of TREE_CONSTANT.
> > 
> > gcc/testsuite/ChangeLog:
> > 
> > 	PR c++/98644
> > 	* g++.dg/cpp2a/concepts-pr98644.C: New test.
> > ---
> >   gcc/cp/constraint.cc                          | 2 +-
> >   gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C | 7 +++++++
> >   2 files changed, 8 insertions(+), 1 deletion(-)
> >   create mode 100644 gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
> > 
> > diff --git a/gcc/cp/constraint.cc b/gcc/cp/constraint.cc
> > index 9049d087859..f99a25dc8a4 100644
> > --- a/gcc/cp/constraint.cc
> > +++ b/gcc/cp/constraint.cc
> > @@ -2969,7 +2969,7 @@ satisfy_atom (tree t, tree args, sat_info info)
> >       {
> >         result = maybe_constant_value (result, NULL_TREE,
> >   				     /*manifestly_const_eval=*/true);
> > -      if (!TREE_CONSTANT (result))
> 
> This should be sufficient.  If the result isn't constant, maybe_constant_value
> shouldn't return it with TREE_CONSTANT set.  See
> 
> >       /* This isn't actually constant, so unset TREE_CONSTANT.                                                 
> 
> in cxx_eval_outermost_constant_expr.

I see, so the problem seems to be that the fail-fast path of
maybe_constant_value isn't clearing TREE_CONSTANT sufficiently.  Would
it make sense to fix this like so?

-- >8 --

Subject: [PATCH] c++: ICE with non-constant satisfaction value [PR98644]

Here during satisfaction the expression of the atomic constraint after
substitution is (int *) NON_LVALUE_EXPR <1> != 0B, which is not a C++
constant expression due to the reinterpret_cast, but TREE_CONSTANT is
set since its value is otherwise effectively constant.  We then call
maybe_constant_value on it, which proceeds via its fail-fast path to
exit early without clearing TREE_CONSTANT.  But satisfy_atom relies
on checking TREE_CONSTANT of the result of maybe_constant_value in order
to detect non-constant satisfaction.

This patch fixes this by making the fail-fast path of maybe_constant_value
clear TREE_CONSTANT in this case, like cxx_eval_outermost_constant_expr
in the normal path would have done.

Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look OK for
trunk?

	PR c++/98644

gcc/cp/ChangeLog:

	* constexpr.cc (maybe_constant_value): In the fail-fast path,
	clear TREE_CONSTANT on the result if it's set on the input.

gcc/testsuite/ChangeLog:

	* g++.dg/cpp2a/concepts-pr98644.C: New test.
	* g++.dg/parse/array-size2.C: Remove expected diagnostic about a
	narrowing conversion.
---
 gcc/cp/constexpr.cc                           | 4 +++-
 gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C | 7 +++++++
 gcc/testsuite/g++.dg/parse/array-size2.C      | 2 --
 3 files changed, 10 insertions(+), 3 deletions(-)
 create mode 100644 gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C

diff --git a/gcc/cp/constexpr.cc b/gcc/cp/constexpr.cc
index 4716694cb71..234cf0acc26 100644
--- a/gcc/cp/constexpr.cc
+++ b/gcc/cp/constexpr.cc
@@ -7965,8 +7965,10 @@ maybe_constant_value (tree t, tree decl, bool manifestly_const_eval)
 
   if (!is_nondependent_constant_expression (t))
     {
-      if (TREE_OVERFLOW_P (t))
+      if (TREE_OVERFLOW_P (t)
+	  || (!processing_template_decl && TREE_CONSTANT (t)))
 	{
+	  /* This isn't actually constant, so unset TREE_CONSTANT.  */
 	  t = build_nop (TREE_TYPE (t), t);
 	  TREE_CONSTANT (t) = false;
 	}
diff --git a/gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C b/gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
new file mode 100644
index 00000000000..6772f72a3ce
--- /dev/null
+++ b/gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
@@ -0,0 +1,7 @@
+// PR c++/98644
+// { dg-do compile { target c++20 } }
+
+template<class T> concept Signed = bool(T(1)); // { dg-error "reinterpret_cast" }
+static_assert(Signed<int*>); // { dg-error "non-constant" }
+
+constexpr bool B = requires { requires bool((char *)1); }; // { dg-error "reinterpret_cast" }
diff --git a/gcc/testsuite/g++.dg/parse/array-size2.C b/gcc/testsuite/g++.dg/parse/array-size2.C
index c4a69df3b01..e58fe266e77 100644
--- a/gcc/testsuite/g++.dg/parse/array-size2.C
+++ b/gcc/testsuite/g++.dg/parse/array-size2.C
@@ -15,8 +15,6 @@ void
 foo (void)
 {
   char g[(char *) &((struct S *) 0)->b - (char *) 0]; // { dg-error "40:size of array .g. is not an integral constant-expression" }
-						      // { dg-error "narrowing conversion" "" { target c++11 } .-1 }
-						      // { dg-message "expression has a constant value but is not a C.. constant-expression" "" { target c++11 } .-2 }
   char h[(__SIZE_TYPE__) &((struct S *) 8)->b];	      // { dg-error "10:size of array .h. is not an integral constant-expression" }
   bar (g, h);
 }
-- 
2.35.1.354.g715d08a9e5

Re: [PATCH] c++: Fix ICE with non-constant satisfaction [PR98644]

[Jason Merrill]

On 3/1/22 00:10, Patrick Palka wrote:
> On Tue, 19 Jan 2021, Jason Merrill wrote:
> 
>> On 1/13/21 12:05 PM, Patrick Palka wrote:
>>> In the below testcase, the expression of the atomic constraint after
>>> substitution is (int *) NON_LVALUE_EXPR <1> != 0B which is not a C++
>>> constant expression, but its TREE_CONSTANT flag is set (from build2),
>>> so satisfy_atom fails to notice that it's non-constant (and we end
>>> up tripping over the assert in satisfaction_value).
>>>
>>> Since TREE_CONSTANT doesn't necessarily correspond to C++ constantness,
>>> this patch makes satisfy_atom instead check is_rvalue_constant_expression.
>>>
>>> Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look OK for
>>> trunk/10?
>>>
>>> gcc/cp/ChangeLog:
>>>
>>> 	PR c++/98644
>>> 	* constraint.cc (satisfy_atom): Check is_rvalue_constant_expression
>>> 	instead of TREE_CONSTANT.
>>>
>>> gcc/testsuite/ChangeLog:
>>>
>>> 	PR c++/98644
>>> 	* g++.dg/cpp2a/concepts-pr98644.C: New test.
>>> ---
>>>    gcc/cp/constraint.cc                          | 2 +-
>>>    gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C | 7 +++++++
>>>    2 files changed, 8 insertions(+), 1 deletion(-)
>>>    create mode 100644 gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
>>>
>>> diff --git a/gcc/cp/constraint.cc b/gcc/cp/constraint.cc
>>> index 9049d087859..f99a25dc8a4 100644
>>> --- a/gcc/cp/constraint.cc
>>> +++ b/gcc/cp/constraint.cc
>>> @@ -2969,7 +2969,7 @@ satisfy_atom (tree t, tree args, sat_info info)
>>>        {
>>>          result = maybe_constant_value (result, NULL_TREE,
>>>    				     /*manifestly_const_eval=*/true);
>>> -      if (!TREE_CONSTANT (result))
>>
>> This should be sufficient.  If the result isn't constant, maybe_constant_value
>> shouldn't return it with TREE_CONSTANT set.  See
>>
>>>        /* This isn't actually constant, so unset TREE_CONSTANT.
>>
>> in cxx_eval_outermost_constant_expr.
> 
> I see, so the problem seems to be that the fail-fast path of
> maybe_constant_value isn't clearing TREE_CONSTANT sufficiently.  Would
> it make sense to fix this like so?
> 
> -- >8 --
> 
> Subject: [PATCH] c++: ICE with non-constant satisfaction value [PR98644]
> 
> Here during satisfaction the expression of the atomic constraint after
> substitution is (int *) NON_LVALUE_EXPR <1> != 0B, which is not a C++
> constant expression due to the reinterpret_cast, but TREE_CONSTANT is
> set since its value is otherwise effectively constant.  We then call
> maybe_constant_value on it, which proceeds via its fail-fast path to
> exit early without clearing TREE_CONSTANT.  But satisfy_atom relies
> on checking TREE_CONSTANT of the result of maybe_constant_value in order
> to detect non-constant satisfaction.
> 
> This patch fixes this by making the fail-fast path of maybe_constant_value
> clear TREE_CONSTANT in this case, like cxx_eval_outermost_constant_expr
> in the normal path would have done.
> 
> Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look OK for
> trunk?
> 
> 	PR c++/98644
> 
> gcc/cp/ChangeLog:
> 
> 	* constexpr.cc (maybe_constant_value): In the fail-fast path,
> 	clear TREE_CONSTANT on the result if it's set on the input.
> 
> gcc/testsuite/ChangeLog:
> 
> 	* g++.dg/cpp2a/concepts-pr98644.C: New test.
> 	* g++.dg/parse/array-size2.C: Remove expected diagnostic about a
> 	narrowing conversion.
> ---
>   gcc/cp/constexpr.cc                           | 4 +++-
>   gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C | 7 +++++++
>   gcc/testsuite/g++.dg/parse/array-size2.C      | 2 --
>   3 files changed, 10 insertions(+), 3 deletions(-)
>   create mode 100644 gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
> 
> diff --git a/gcc/cp/constexpr.cc b/gcc/cp/constexpr.cc
> index 4716694cb71..234cf0acc26 100644
> --- a/gcc/cp/constexpr.cc
> +++ b/gcc/cp/constexpr.cc
> @@ -7965,8 +7965,10 @@ maybe_constant_value (tree t, tree decl, bool manifestly_const_eval)
>   
>     if (!is_nondependent_constant_expression (t))
>       {
> -      if (TREE_OVERFLOW_P (t))
> +      if (TREE_OVERFLOW_P (t)
> +	  || (!processing_template_decl && TREE_CONSTANT (t)))
>   	{
> +	  /* This isn't actually constant, so unset TREE_CONSTANT.  */
>   	  t = build_nop (TREE_TYPE (t), t);

build_nop isn't appropriate for arbitrary expressions (classes, in 
particular).  We probably want to factor out the code in 
cxx_eval_outermost_constant_expr under the "this isn't actually 
constant" comment.

>   	  TREE_CONSTANT (t) = false;
>   	}
> diff --git a/gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C b/gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
> new file mode 100644
> index 00000000000..6772f72a3ce
> --- /dev/null
> +++ b/gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
> @@ -0,0 +1,7 @@
> +// PR c++/98644
> +// { dg-do compile { target c++20 } }
> +
> +template<class T> concept Signed = bool(T(1)); // { dg-error "reinterpret_cast" }
> +static_assert(Signed<int*>); // { dg-error "non-constant" }
> +
> +constexpr bool B = requires { requires bool((char *)1); }; // { dg-error "reinterpret_cast" }
> diff --git a/gcc/testsuite/g++.dg/parse/array-size2.C b/gcc/testsuite/g++.dg/parse/array-size2.C
> index c4a69df3b01..e58fe266e77 100644
> --- a/gcc/testsuite/g++.dg/parse/array-size2.C
> +++ b/gcc/testsuite/g++.dg/parse/array-size2.C
> @@ -15,8 +15,6 @@ void
>   foo (void)
>   {
>     char g[(char *) &((struct S *) 0)->b - (char *) 0]; // { dg-error "40:size of array .g. is not an integral constant-expression" }
> -						      // { dg-error "narrowing conversion" "" { target c++11 } .-1 }
> -						      // { dg-message "expression has a constant value but is not a C.. constant-expression" "" { target c++11 } .-2 }
>     char h[(__SIZE_TYPE__) &((struct S *) 8)->b];	      // { dg-error "10:size of array .h. is not an integral constant-expression" }
>     bar (g, h);
>   }

Re: [PATCH] c++: Fix ICE with non-constant satisfaction [PR98644]

[Patrick Palka]

On Thu, 10 Mar 2022, Jason Merrill wrote:

> On 3/1/22 00:10, Patrick Palka wrote:
> > On Tue, 19 Jan 2021, Jason Merrill wrote:
> > 
> > > On 1/13/21 12:05 PM, Patrick Palka wrote:
> > > > In the below testcase, the expression of the atomic constraint after
> > > > substitution is (int *) NON_LVALUE_EXPR <1> != 0B which is not a C++
> > > > constant expression, but its TREE_CONSTANT flag is set (from build2),
> > > > so satisfy_atom fails to notice that it's non-constant (and we end
> > > > up tripping over the assert in satisfaction_value).
> > > > 
> > > > Since TREE_CONSTANT doesn't necessarily correspond to C++ constantness,
> > > > this patch makes satisfy_atom instead check
> > > > is_rvalue_constant_expression.
> > > > 
> > > > Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look OK for
> > > > trunk/10?
> > > > 
> > > > gcc/cp/ChangeLog:
> > > > 
> > > > 	PR c++/98644
> > > > 	* constraint.cc (satisfy_atom): Check is_rvalue_constant_expression
> > > > 	instead of TREE_CONSTANT.
> > > > 
> > > > gcc/testsuite/ChangeLog:
> > > > 
> > > > 	PR c++/98644
> > > > 	* g++.dg/cpp2a/concepts-pr98644.C: New test.
> > > > ---
> > > >    gcc/cp/constraint.cc                          | 2 +-
> > > >    gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C | 7 +++++++
> > > >    2 files changed, 8 insertions(+), 1 deletion(-)
> > > >    create mode 100644 gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
> > > > 
> > > > diff --git a/gcc/cp/constraint.cc b/gcc/cp/constraint.cc
> > > > index 9049d087859..f99a25dc8a4 100644
> > > > --- a/gcc/cp/constraint.cc
> > > > +++ b/gcc/cp/constraint.cc
> > > > @@ -2969,7 +2969,7 @@ satisfy_atom (tree t, tree args, sat_info info)
> > > >        {
> > > >          result = maybe_constant_value (result, NULL_TREE,
> > > >    				     /*manifestly_const_eval=*/true);
> > > > -      if (!TREE_CONSTANT (result))
> > > 
> > > This should be sufficient.  If the result isn't constant,
> > > maybe_constant_value
> > > shouldn't return it with TREE_CONSTANT set.  See
> > > 
> > > >        /* This isn't actually constant, so unset TREE_CONSTANT.
> > > 
> > > in cxx_eval_outermost_constant_expr.
> > 
> > I see, so the problem seems to be that the fail-fast path of
> > maybe_constant_value isn't clearing TREE_CONSTANT sufficiently.  Would
> > it make sense to fix this like so?
> > 
> > -- >8 --
> > 
> > Subject: [PATCH] c++: ICE with non-constant satisfaction value [PR98644]
> > 
> > Here during satisfaction the expression of the atomic constraint after
> > substitution is (int *) NON_LVALUE_EXPR <1> != 0B, which is not a C++
> > constant expression due to the reinterpret_cast, but TREE_CONSTANT is
> > set since its value is otherwise effectively constant.  We then call
> > maybe_constant_value on it, which proceeds via its fail-fast path to
> > exit early without clearing TREE_CONSTANT.  But satisfy_atom relies
> > on checking TREE_CONSTANT of the result of maybe_constant_value in order
> > to detect non-constant satisfaction.
> > 
> > This patch fixes this by making the fail-fast path of maybe_constant_value
> > clear TREE_CONSTANT in this case, like cxx_eval_outermost_constant_expr
> > in the normal path would have done.
> > 
> > Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look OK for
> > trunk?
> > 
> > 	PR c++/98644
> > 
> > gcc/cp/ChangeLog:
> > 
> > 	* constexpr.cc (maybe_constant_value): In the fail-fast path,
> > 	clear TREE_CONSTANT on the result if it's set on the input.
> > 
> > gcc/testsuite/ChangeLog:
> > 
> > 	* g++.dg/cpp2a/concepts-pr98644.C: New test.
> > 	* g++.dg/parse/array-size2.C: Remove expected diagnostic about a
> > 	narrowing conversion.
> > ---
> >   gcc/cp/constexpr.cc                           | 4 +++-
> >   gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C | 7 +++++++
> >   gcc/testsuite/g++.dg/parse/array-size2.C      | 2 --
> >   3 files changed, 10 insertions(+), 3 deletions(-)
> >   create mode 100644 gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
> > 
> > diff --git a/gcc/cp/constexpr.cc b/gcc/cp/constexpr.cc
> > index 4716694cb71..234cf0acc26 100644
> > --- a/gcc/cp/constexpr.cc
> > +++ b/gcc/cp/constexpr.cc
> > @@ -7965,8 +7965,10 @@ maybe_constant_value (tree t, tree decl, bool
> > manifestly_const_eval)
> >       if (!is_nondependent_constant_expression (t))
> >       {
> > -      if (TREE_OVERFLOW_P (t))
> > +      if (TREE_OVERFLOW_P (t)
> > +	  || (!processing_template_decl && TREE_CONSTANT (t)))
> >   	{
> > +	  /* This isn't actually constant, so unset TREE_CONSTANT.  */
> >   	  t = build_nop (TREE_TYPE (t), t);
> 
> build_nop isn't appropriate for arbitrary expressions (classes, in
> particular).  We probably want to factor out the code in
> cxx_eval_outermost_constant_expr under the "this isn't actually constant"
> comment.

Gotcha, makes sense.  Like the following?  Bootstrapped and regtested on
x86_64-pc-linux-gnu.

-- >8 --

Subject: [PATCH] c++: ICE with non-constant satisfaction value [PR98644]

Here during satisfaction, the expression of the atomic constraint after
substitution is (int *) NON_LVALUE_EXPR <1> != 0B, which is not a C++
constant expression due to the reinterpret_cast, but TREE_CONSTANT is
set since its value is otherwise effectively constant.  We then call
maybe_constant_value on it, which proceeds via its fail-fast path to
exit early without clearing TREE_CONSTANT.  But satisfy_atom relies
on checking TREE_CONSTANT of the result of maybe_constant_value in order
to detect non-constant satisfaction.

This patch fixes this by making the fail-fast path of maybe_constant_value
clear TREE_CONSTANT in this case, like cxx_eval_outermost_constant_expr
in the normal path would have done.

	PR c++/98644

gcc/cp/ChangeLog:

	* constexpr.cc (mark_non_constant): Define, split out from ...
	(cxx_eval_outermost_constant_expr): ... here.
	(maybe_constant_value): Use it.

gcc/testsuite/ChangeLog:

	* g++.dg/cpp2a/concepts-pr98644.C: New test.
	* g++.dg/parse/array-size2.C: Remove expected diagnostic about a
	narrowing conversion.

Co-authored-by: Jason Merrill <jason@redhat.com>
---
 gcc/cp/constexpr.cc                           | 46 +++++++++++--------
 gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C |  7 +++
 gcc/testsuite/g++.dg/parse/array-size2.C      |  2 -
 3 files changed, 34 insertions(+), 21 deletions(-)
 create mode 100644 gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C

diff --git a/gcc/cp/constexpr.cc b/gcc/cp/constexpr.cc
index 4716694cb71..22d86f023d0 100644
--- a/gcc/cp/constexpr.cc
+++ b/gcc/cp/constexpr.cc
@@ -7579,6 +7579,29 @@ find_immediate_fndecl (tree *tp, int */*walk_subtrees*/, void */*data*/)
   return NULL_TREE;
 }
 
+/* T has TREE_CONSTANT set but has been deemed not a valid C++ constant
+   expression.  Return a version of T that has TREE_CONSTANT cleared.  */
+
+static tree
+mark_non_constant (tree t)
+{
+  gcc_checking_assert (TREE_CONSTANT (t));
+
+  /* This isn't actually constant, so unset TREE_CONSTANT.
+     Don't clear TREE_CONSTANT on ADDR_EXPR, as the middle-end requires
+     it to be set if it is invariant address, even when it is not
+     a valid C++ constant expression.  Wrap it with a NOP_EXPR
+     instead.  */
+  if (EXPR_P (t) && TREE_CODE (t) != ADDR_EXPR)
+    t = copy_node (t);
+  else if (TREE_CODE (t) == CONSTRUCTOR)
+    t = build1 (VIEW_CONVERT_EXPR, TREE_TYPE (t), t);
+  else
+    t = build_nop (TREE_TYPE (t), t);
+  TREE_CONSTANT (t) = false;
+  return t;
+}
+
 /* ALLOW_NON_CONSTANT is false if T is required to be a constant expression.
    STRICT has the same sense as for constant_value_1: true if we only allow
    conforming C++ constant expressions, or false if we want a constant value
@@ -7801,20 +7824,7 @@ cxx_eval_outermost_constant_expr (tree t, bool allow_non_constant,
   else if (constexpr_dtor)
     return r;
   else if (non_constant_p && TREE_CONSTANT (r))
-    {
-      /* This isn't actually constant, so unset TREE_CONSTANT.
-	 Don't clear TREE_CONSTANT on ADDR_EXPR, as the middle-end requires
-	 it to be set if it is invariant address, even when it is not
-	 a valid C++ constant expression.  Wrap it with a NOP_EXPR
-	 instead.  */
-      if (EXPR_P (r) && TREE_CODE (r) != ADDR_EXPR)
-	r = copy_node (r);
-      else if (TREE_CODE (r) == CONSTRUCTOR)
-	r = build1 (VIEW_CONVERT_EXPR, TREE_TYPE (r), r);
-      else
-	r = build_nop (TREE_TYPE (r), r);
-      TREE_CONSTANT (r) = false;
-    }
+    r = mark_non_constant (r);
   else if (non_constant_p)
     return t;
 
@@ -7965,11 +7975,9 @@ maybe_constant_value (tree t, tree decl, bool manifestly_const_eval)
 
   if (!is_nondependent_constant_expression (t))
     {
-      if (TREE_OVERFLOW_P (t))
-	{
-	  t = build_nop (TREE_TYPE (t), t);
-	  TREE_CONSTANT (t) = false;
-	}
+      if (TREE_OVERFLOW_P (t)
+	  || (!processing_template_decl && TREE_CONSTANT (t)))
+	t = mark_non_constant (t);
       return t;
     }
   else if (CONSTANT_CLASS_P (t))
diff --git a/gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C b/gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
new file mode 100644
index 00000000000..6772f72a3ce
--- /dev/null
+++ b/gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
@@ -0,0 +1,7 @@
+// PR c++/98644
+// { dg-do compile { target c++20 } }
+
+template<class T> concept Signed = bool(T(1)); // { dg-error "reinterpret_cast" }
+static_assert(Signed<int*>); // { dg-error "non-constant" }
+
+constexpr bool B = requires { requires bool((char *)1); }; // { dg-error "reinterpret_cast" }
diff --git a/gcc/testsuite/g++.dg/parse/array-size2.C b/gcc/testsuite/g++.dg/parse/array-size2.C
index c4a69df3b01..e58fe266e77 100644
--- a/gcc/testsuite/g++.dg/parse/array-size2.C
+++ b/gcc/testsuite/g++.dg/parse/array-size2.C
@@ -15,8 +15,6 @@ void
 foo (void)
 {
   char g[(char *) &((struct S *) 0)->b - (char *) 0]; // { dg-error "40:size of array .g. is not an integral constant-expression" }
-						      // { dg-error "narrowing conversion" "" { target c++11 } .-1 }
-						      // { dg-message "expression has a constant value but is not a C.. constant-expression" "" { target c++11 } .-2 }
   char h[(__SIZE_TYPE__) &((struct S *) 8)->b];	      // { dg-error "10:size of array .h. is not an integral constant-expression" }
   bar (g, h);
 }
-- 
2.35.1.455.g1a4874565f

Re: [PATCH] c++: Fix ICE with non-constant satisfaction [PR98644]

[Jason Merrill]

On 3/11/22 11:46, Patrick Palka wrote:
> On Thu, 10 Mar 2022, Jason Merrill wrote:
> 
>> On 3/1/22 00:10, Patrick Palka wrote:
>>> On Tue, 19 Jan 2021, Jason Merrill wrote:
>>>
>>>> On 1/13/21 12:05 PM, Patrick Palka wrote:
>>>>> In the below testcase, the expression of the atomic constraint after
>>>>> substitution is (int *) NON_LVALUE_EXPR <1> != 0B which is not a C++
>>>>> constant expression, but its TREE_CONSTANT flag is set (from build2),
>>>>> so satisfy_atom fails to notice that it's non-constant (and we end
>>>>> up tripping over the assert in satisfaction_value).
>>>>>
>>>>> Since TREE_CONSTANT doesn't necessarily correspond to C++ constantness,
>>>>> this patch makes satisfy_atom instead check
>>>>> is_rvalue_constant_expression.
>>>>>
>>>>> Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look OK for
>>>>> trunk/10?
>>>>>
>>>>> gcc/cp/ChangeLog:
>>>>>
>>>>> 	PR c++/98644
>>>>> 	* constraint.cc (satisfy_atom): Check is_rvalue_constant_expression
>>>>> 	instead of TREE_CONSTANT.
>>>>>
>>>>> gcc/testsuite/ChangeLog:
>>>>>
>>>>> 	PR c++/98644
>>>>> 	* g++.dg/cpp2a/concepts-pr98644.C: New test.
>>>>> ---
>>>>>     gcc/cp/constraint.cc                          | 2 +-
>>>>>     gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C | 7 +++++++
>>>>>     2 files changed, 8 insertions(+), 1 deletion(-)
>>>>>     create mode 100644 gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
>>>>>
>>>>> diff --git a/gcc/cp/constraint.cc b/gcc/cp/constraint.cc
>>>>> index 9049d087859..f99a25dc8a4 100644
>>>>> --- a/gcc/cp/constraint.cc
>>>>> +++ b/gcc/cp/constraint.cc
>>>>> @@ -2969,7 +2969,7 @@ satisfy_atom (tree t, tree args, sat_info info)
>>>>>         {
>>>>>           result = maybe_constant_value (result, NULL_TREE,
>>>>>     				     /*manifestly_const_eval=*/true);
>>>>> -      if (!TREE_CONSTANT (result))
>>>>
>>>> This should be sufficient.  If the result isn't constant,
>>>> maybe_constant_value
>>>> shouldn't return it with TREE_CONSTANT set.  See
>>>>
>>>>>         /* This isn't actually constant, so unset TREE_CONSTANT.
>>>>
>>>> in cxx_eval_outermost_constant_expr.
>>>
>>> I see, so the problem seems to be that the fail-fast path of
>>> maybe_constant_value isn't clearing TREE_CONSTANT sufficiently.  Would
>>> it make sense to fix this like so?
>>>
>>> -- >8 --
>>>
>>> Subject: [PATCH] c++: ICE with non-constant satisfaction value [PR98644]
>>>
>>> Here during satisfaction the expression of the atomic constraint after
>>> substitution is (int *) NON_LVALUE_EXPR <1> != 0B, which is not a C++
>>> constant expression due to the reinterpret_cast, but TREE_CONSTANT is
>>> set since its value is otherwise effectively constant.  We then call
>>> maybe_constant_value on it, which proceeds via its fail-fast path to
>>> exit early without clearing TREE_CONSTANT.  But satisfy_atom relies
>>> on checking TREE_CONSTANT of the result of maybe_constant_value in order
>>> to detect non-constant satisfaction.
>>>
>>> This patch fixes this by making the fail-fast path of maybe_constant_value
>>> clear TREE_CONSTANT in this case, like cxx_eval_outermost_constant_expr
>>> in the normal path would have done.
>>>
>>> Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look OK for
>>> trunk?
>>>
>>> 	PR c++/98644
>>>
>>> gcc/cp/ChangeLog:
>>>
>>> 	* constexpr.cc (maybe_constant_value): In the fail-fast path,
>>> 	clear TREE_CONSTANT on the result if it's set on the input.
>>>
>>> gcc/testsuite/ChangeLog:
>>>
>>> 	* g++.dg/cpp2a/concepts-pr98644.C: New test.
>>> 	* g++.dg/parse/array-size2.C: Remove expected diagnostic about a
>>> 	narrowing conversion.
>>> ---
>>>    gcc/cp/constexpr.cc                           | 4 +++-
>>>    gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C | 7 +++++++
>>>    gcc/testsuite/g++.dg/parse/array-size2.C      | 2 --
>>>    3 files changed, 10 insertions(+), 3 deletions(-)
>>>    create mode 100644 gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
>>>
>>> diff --git a/gcc/cp/constexpr.cc b/gcc/cp/constexpr.cc
>>> index 4716694cb71..234cf0acc26 100644
>>> --- a/gcc/cp/constexpr.cc
>>> +++ b/gcc/cp/constexpr.cc
>>> @@ -7965,8 +7965,10 @@ maybe_constant_value (tree t, tree decl, bool
>>> manifestly_const_eval)
>>>        if (!is_nondependent_constant_expression (t))
>>>        {
>>> -      if (TREE_OVERFLOW_P (t))
>>> +      if (TREE_OVERFLOW_P (t)
>>> +	  || (!processing_template_decl && TREE_CONSTANT (t)))
>>>    	{
>>> +	  /* This isn't actually constant, so unset TREE_CONSTANT.  */
>>>    	  t = build_nop (TREE_TYPE (t), t);
>>
>> build_nop isn't appropriate for arbitrary expressions (classes, in
>> particular).  We probably want to factor out the code in
>> cxx_eval_outermost_constant_expr under the "this isn't actually constant"
>> comment.
> 
> Gotcha, makes sense.  Like the following?  Bootstrapped and regtested on
> x86_64-pc-linux-gnu.

OK.

> -- >8 --
> 
> Subject: [PATCH] c++: ICE with non-constant satisfaction value [PR98644]
> 
> Here during satisfaction, the expression of the atomic constraint after
> substitution is (int *) NON_LVALUE_EXPR <1> != 0B, which is not a C++
> constant expression due to the reinterpret_cast, but TREE_CONSTANT is
> set since its value is otherwise effectively constant.  We then call
> maybe_constant_value on it, which proceeds via its fail-fast path to
> exit early without clearing TREE_CONSTANT.  But satisfy_atom relies
> on checking TREE_CONSTANT of the result of maybe_constant_value in order
> to detect non-constant satisfaction.
> 
> This patch fixes this by making the fail-fast path of maybe_constant_value
> clear TREE_CONSTANT in this case, like cxx_eval_outermost_constant_expr
> in the normal path would have done.
> 
> 	PR c++/98644
> 
> gcc/cp/ChangeLog:
> 
> 	* constexpr.cc (mark_non_constant): Define, split out from ...
> 	(cxx_eval_outermost_constant_expr): ... here.
> 	(maybe_constant_value): Use it.
> 
> gcc/testsuite/ChangeLog:
> 
> 	* g++.dg/cpp2a/concepts-pr98644.C: New test.
> 	* g++.dg/parse/array-size2.C: Remove expected diagnostic about a
> 	narrowing conversion.
> 
> Co-authored-by: Jason Merrill <jason@redhat.com>
> ---
>   gcc/cp/constexpr.cc                           | 46 +++++++++++--------
>   gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C |  7 +++
>   gcc/testsuite/g++.dg/parse/array-size2.C      |  2 -
>   3 files changed, 34 insertions(+), 21 deletions(-)
>   create mode 100644 gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
> 
> diff --git a/gcc/cp/constexpr.cc b/gcc/cp/constexpr.cc
> index 4716694cb71..22d86f023d0 100644
> --- a/gcc/cp/constexpr.cc
> +++ b/gcc/cp/constexpr.cc
> @@ -7579,6 +7579,29 @@ find_immediate_fndecl (tree *tp, int */*walk_subtrees*/, void */*data*/)
>     return NULL_TREE;
>   }
>   
> +/* T has TREE_CONSTANT set but has been deemed not a valid C++ constant
> +   expression.  Return a version of T that has TREE_CONSTANT cleared.  */
> +
> +static tree
> +mark_non_constant (tree t)
> +{
> +  gcc_checking_assert (TREE_CONSTANT (t));
> +
> +  /* This isn't actually constant, so unset TREE_CONSTANT.
> +     Don't clear TREE_CONSTANT on ADDR_EXPR, as the middle-end requires
> +     it to be set if it is invariant address, even when it is not
> +     a valid C++ constant expression.  Wrap it with a NOP_EXPR
> +     instead.  */
> +  if (EXPR_P (t) && TREE_CODE (t) != ADDR_EXPR)
> +    t = copy_node (t);
> +  else if (TREE_CODE (t) == CONSTRUCTOR)
> +    t = build1 (VIEW_CONVERT_EXPR, TREE_TYPE (t), t);
> +  else
> +    t = build_nop (TREE_TYPE (t), t);
> +  TREE_CONSTANT (t) = false;
> +  return t;
> +}
> +
>   /* ALLOW_NON_CONSTANT is false if T is required to be a constant expression.
>      STRICT has the same sense as for constant_value_1: true if we only allow
>      conforming C++ constant expressions, or false if we want a constant value
> @@ -7801,20 +7824,7 @@ cxx_eval_outermost_constant_expr (tree t, bool allow_non_constant,
>     else if (constexpr_dtor)
>       return r;
>     else if (non_constant_p && TREE_CONSTANT (r))
> -    {
> -      /* This isn't actually constant, so unset TREE_CONSTANT.
> -	 Don't clear TREE_CONSTANT on ADDR_EXPR, as the middle-end requires
> -	 it to be set if it is invariant address, even when it is not
> -	 a valid C++ constant expression.  Wrap it with a NOP_EXPR
> -	 instead.  */
> -      if (EXPR_P (r) && TREE_CODE (r) != ADDR_EXPR)
> -	r = copy_node (r);
> -      else if (TREE_CODE (r) == CONSTRUCTOR)
> -	r = build1 (VIEW_CONVERT_EXPR, TREE_TYPE (r), r);
> -      else
> -	r = build_nop (TREE_TYPE (r), r);
> -      TREE_CONSTANT (r) = false;
> -    }
> +    r = mark_non_constant (r);
>     else if (non_constant_p)
>       return t;
>   
> @@ -7965,11 +7975,9 @@ maybe_constant_value (tree t, tree decl, bool manifestly_const_eval)
>   
>     if (!is_nondependent_constant_expression (t))
>       {
> -      if (TREE_OVERFLOW_P (t))
> -	{
> -	  t = build_nop (TREE_TYPE (t), t);
> -	  TREE_CONSTANT (t) = false;
> -	}
> +      if (TREE_OVERFLOW_P (t)
> +	  || (!processing_template_decl && TREE_CONSTANT (t)))
> +	t = mark_non_constant (t);
>         return t;
>       }
>     else if (CONSTANT_CLASS_P (t))
> diff --git a/gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C b/gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
> new file mode 100644
> index 00000000000..6772f72a3ce
> --- /dev/null
> +++ b/gcc/testsuite/g++.dg/cpp2a/concepts-pr98644.C
> @@ -0,0 +1,7 @@
> +// PR c++/98644
> +// { dg-do compile { target c++20 } }
> +
> +template<class T> concept Signed = bool(T(1)); // { dg-error "reinterpret_cast" }
> +static_assert(Signed<int*>); // { dg-error "non-constant" }
> +
> +constexpr bool B = requires { requires bool((char *)1); }; // { dg-error "reinterpret_cast" }
> diff --git a/gcc/testsuite/g++.dg/parse/array-size2.C b/gcc/testsuite/g++.dg/parse/array-size2.C
> index c4a69df3b01..e58fe266e77 100644
> --- a/gcc/testsuite/g++.dg/parse/array-size2.C
> +++ b/gcc/testsuite/g++.dg/parse/array-size2.C
> @@ -15,8 +15,6 @@ void
>   foo (void)
>   {
>     char g[(char *) &((struct S *) 0)->b - (char *) 0]; // { dg-error "40:size of array .g. is not an integral constant-expression" }
> -						      // { dg-error "narrowing conversion" "" { target c++11 } .-1 }
> -						      // { dg-message "expression has a constant value but is not a C.. constant-expression" "" { target c++11 } .-2 }
>     char h[(__SIZE_TYPE__) &((struct S *) 8)->b];	      // { dg-error "10:size of array .h. is not an integral constant-expression" }
>     bar (g, h);
>   }