c++: PR c++/95999: Improved error recovery in enumeration lists.
Commit Message
This patch resolves PR c++/95999 which is an ICE-after-error regression
in the g++ front-end. When parsing an enumerator list, the C++ parser
assumes that cp_parser_constant_expression always returns either an
INTEGER_CST or error_mark_node, but in testcase reported in the PR
actually returns a VAR_DECL.
The usual (but perhaps controversial) design philosophy is that the
routine that reports the error normally has a duty to indicate this to
the rest of the compiler (via error_mark_node), but here the return
value from calling require_rvalue_constant_expression (parser.cc:10666)
is ignored. I initially experimented with setting EXPRESSION to
error_mark_node here in cp_parser_constant_expression but (perhaps
conveniently) that's insufficient to resolve the problem. The simple
fix in this patch is to tweak the two places that require INTEGER_CST
to treat all other tree types as though they are error_mark_node.
This patch has been tested on x86_64-pc-linunx-gnu with make bootstrap
and make -k check with no new (unexpected) failures. Ok for mainline?
2022-02-21 Roger Sayle <roger@nextmovesoftware.com>
gcc/cp/ChangeLog
PR c++/95999
* decl.cc (finish_enum_value_list): If VALUE isn't an INTEGER_CST
consider it to be zero (i.e. treat it like error_mark_node).
(build_enumerator): Likewise, if PREV_VALUE isn't an INTEGER_CST,
set VALUE to error_mark_node.
gcc/testsuite/ChangeLog
PR c++/95999
* g++.dg/pr95999.c: New test case.
Thanks in advance,
Roger
--
Comments
On 2/21/22 08:17, Roger Sayle wrote:
>
> This patch resolves PR c++/95999 which is an ICE-after-error regression
> in the g++ front-end. When parsing an enumerator list, the C++ parser
> assumes that cp_parser_constant_expression always returns either an
> INTEGER_CST or error_mark_node, but in testcase reported in the PR
> actually returns a VAR_DECL.
>
> The usual (but perhaps controversial) design philosophy is that the
> routine that reports the error normally has a duty to indicate this to
> the rest of the compiler (via error_mark_node), but here the return
> value from calling require_rvalue_constant_expression (parser.cc:10666)
> is ignored. I initially experimented with setting EXPRESSION to
> error_mark_node here in cp_parser_constant_expression but (perhaps
> conveniently) that's insufficient to resolve the problem. The simple
> fix in this patch is to tweak the two places that require INTEGER_CST
> to treat all other tree types as though they are error_mark_node.
>
> This patch has been tested on x86_64-pc-linunx-gnu with make bootstrap
> and make -k check with no new (unexpected) failures. Ok for mainline?
OK.
> 2022-02-21 Roger Sayle <roger@nextmovesoftware.com>
>
> gcc/cp/ChangeLog
> PR c++/95999
> * decl.cc (finish_enum_value_list): If VALUE isn't an INTEGER_CST
> consider it to be zero (i.e. treat it like error_mark_node).
> (build_enumerator): Likewise, if PREV_VALUE isn't an INTEGER_CST,
> set VALUE to error_mark_node.
>
> gcc/testsuite/ChangeLog
> PR c++/95999
> * g++.dg/pr95999.c: New test case.
>
>
> Thanks in advance,
> Roger
> --
>
@@ -16202,7 +16202,7 @@ finish_enum_value_list (tree enumtype)
/* Update the minimum and maximum values, if appropriate. */
value = DECL_INITIAL (decl);
- if (value == error_mark_node)
+ if (TREE_CODE (value) != INTEGER_CST)
value = integer_zero_node;
/* Figure out what the minimum and maximum values of the
enumerators are. */
@@ -16491,7 +16491,7 @@ build_enumerator (tree name, tree value, tree enumtype, tree attributes,
which case the type is an unspecified integral type
sufficient to contain the incremented value. */
prev_value = DECL_INITIAL (TREE_VALUE (TYPE_VALUES (enumtype)));
- if (error_operand_p (prev_value))
+ if (TREE_CODE (prev_value) != INTEGER_CST)
value = error_mark_node;
else
{
new file mode 100644
@@ -0,0 +1,9 @@
+/* { dg-do compile } */
+/* { dg-options "-O2" } */
+int a;
+enum struct b;
+template <typename = enum struct b { c = a d }
+template <> enum struct b { e }; // { dg-error "explicit specialization" }
+// { dg-excess-errors "note" }
+// { dg-excess-errors "5:" }
+