On Tue, 15 Feb 2022, Jason Merrill wrote:
> On 2/15/22 15:13, Patrick Palka wrote:
> > On Tue, 15 Feb 2022, Patrick Palka wrote:
> >
> > > Here we're crashing from potential_constant_expression because it tries
> > > to perform trial evaluation of the first operand '(bool)__r' of the
> > > conjunction (which is overall wrapped in a NON_DEPENDENT_EXPR), but
> > > cxx_eval_constant_expression ICEs on unhandled trees (of which CAST_EXPR
> > > is one).
> > >
> > > Since cxx_eval_constant_expression always treats NON_DEPENDENT_EXPR
> > > as non-constant, and since NON_DEPENDENT_EXPR is also opaque to
> > > instantiate_non_dependent_expr, it seems futile to have p_c_e_1 ever
> > > return true for NON_DEPENDENT_EXPR, so let's just instead return false
> > > and avoid recursing.
>
> Well, in a template we use pce1 to decide whether to complain about something
> that needs to be constant but can't be. We aren't trying to get a value yet.
Makes sense.. though for NON_DEPENDENT_EXPR in particular, ISTM this
tree is always used in a context where a constant expression isn't
required, e.g. in the build_x_* functions.
And if something is required to be a constant expression and we're
inside a template, then it seems at that point we're dealing with the
final templated form of that thing (which doesn't contain
NON_DEPENDENT_EXPR), which I suppose explains why the patch can get away
with asserting !(flags & tf_error) inside the NON_DEPENDENT_EXPR case of
p_c_e_1.
So I guess I don't fully understand the purpose of NON_DEPENDENT_EXPR
or how it should interact with fold_non_dependent_expr and constant
evaluation...
>
> Actually, why are we seeing a NON_DEPENDENT_EXPR here? Did it leak into the
> AST somehow? They should all be temporary within build_x_whatever functions.
Hmm yes, kind of. The unusual thing about this testcase is that
build_non_dependent_expr (called from grok_array_decl) for the
COND_EXPR
(bool)__r && __s ? 1 : 2
wraps only the condition operand, yielding
NON_DEPENDENT_EXPR<<<(bool)__r && __s>>> ? 1 : 2 // #1
rather than wrapping the whole thing i.e.
NON_DEPENDENT_EXPR<<<(bool)__r && __s ? 1 : 2>>> // #2
cp_build_array_ref then tries to speculatively fold the non-dependent #1
as a whole, during which the COND_EXPR case of tsubst_copy_and_build
tries to speculative fold #1's condition NON_DEPENDENT_EXPR<<<(bool)__r &&
__s>>>
on its own, which ends in a crash from p_c_e_1 because at this point
processing_template_decl is cleared so p_c_e_1 attempts trial evaluation
of the CAST_EXPR (bool)__r.
If instead build_non_dependent_expr yielded #2, speculative folding of
#2 as a whole just yield #2 since tsubst doesn't look through
NON_DEPENDENT_EXPR.
>
> > > Alternatively p_c_e_1 could continue to recurse into NON_DEPENDENT_EXPR,
> > > but with trial evaluation disabled by setting processing_template_decl,
> > > but as mentioned it seems pointless for p_c_e_1 to ever return true for
> > > NON_DEPENDENT_EXPR.
> > ... Since we're not issuing a diagnostic in this case, I suppose we should
> > also assert that tf_error isn't set. Bootstrapped and regtested on
> > x86_64-pc-linux-gnu.
> >
> > -- >8 --
> >
> > PR c++/104507
> >
> > gcc/cp/ChangeLog:
> >
> > * constexpr.cc (potential_constant_expression_1)
> > <case NON_DEPENDENT_EXPR>: Return false instead of recursing.
> > Assert tf_error isn't set.
> >
> > gcc/testsuite/ChangeLog:
> >
> > * g++.dg/template/non-dependent21.C: New test.
> > ---
> > gcc/cp/constexpr.cc | 5 ++++-
> > gcc/testsuite/g++.dg/template/non-dependent21.C | 9 +++++++++
> > 2 files changed, 13 insertions(+), 1 deletion(-)
> > create mode 100644 gcc/testsuite/g++.dg/template/non-dependent21.C
> >
> > diff --git a/gcc/cp/constexpr.cc b/gcc/cp/constexpr.cc
> > index 7274c3b760e..b363ef08411 100644
> > --- a/gcc/cp/constexpr.cc
> > +++ b/gcc/cp/constexpr.cc
> > @@ -9065,6 +9065,10 @@ potential_constant_expression_1 (tree t, bool
> > want_rval, bool strict, bool now,
> > case BIND_EXPR:
> > return RECUR (BIND_EXPR_BODY (t), want_rval);
> > + case NON_DEPENDENT_EXPR:
> > + gcc_checking_assert (!(flags & tf_error));
> > + return false;
> > +
> > case CLEANUP_POINT_EXPR:
> > case MUST_NOT_THROW_EXPR:
> > case TRY_CATCH_EXPR:
> > @@ -9072,7 +9076,6 @@ potential_constant_expression_1 (tree t, bool
> > want_rval, bool strict, bool now,
> > case EH_SPEC_BLOCK:
> > case EXPR_STMT:
> > case PAREN_EXPR:
> > - case NON_DEPENDENT_EXPR:
> > /* For convenience. */
> > case LOOP_EXPR:
> > case EXIT_EXPR:
> > diff --git a/gcc/testsuite/g++.dg/template/non-dependent21.C
> > b/gcc/testsuite/g++.dg/template/non-dependent21.C
> > new file mode 100644
> > index 00000000000..89900837b8b
> > --- /dev/null
> > +++ b/gcc/testsuite/g++.dg/template/non-dependent21.C
> > @@ -0,0 +1,9 @@
> > +// PR c++/104507
> > +
> > +extern const char *_k_errmsg[];
> > +
> > +template<class>
> > +const char* DoFoo(int __r, int __s) {
> > + const char* n = _k_errmsg[(bool)__r && __s ? 1 : 2];
> > + return n;
> > +}
>
>
