Postgres Email Lists

pgsql-hackers

❮

warning under gcc 16.1.0

Jump to comment-1
Erik Rijkers<er@xs4all.nl>
Apr 30, 2026, 8:52 PM UTC
Compiling with gcc 16.1.0 gives the following warning and notes:
(horizontally compressed a bit te avoid excessive wrapping)
clauses.c: In function ‘recheckcastfunction_args.isra’:
clauses.c:5139:19: warning: ‘actualargtypes’ may be used uninitialized [-Wmaybe-uninitialized]
```
 5139 |   rettype = enforce_generic_type_consistency(actual_arg_types,
```
```
  |             ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
```
```
 5140 |                declared_arg_types,
```
```
  |                ~~~~~~~~~~~~~~~~~~~
```
```
 5141 |                nargs,
```
```
  |                ~~~~~~
```
```
 5142 |                funcform->prorettype,
```
```
  |                ~~~~~~~~~~~~~~~~~~~~~
```
```
 5143 |                false);
```
```
  |                ~~~~~~
```
In file included from clauses.c:46:
../../../../src/include/parser/parsecoerce.h:85:17: note: by argument 1 of type ‘const Oid *’ {aka ‘const unsigned int *’} to ‘enforcegenerictypeconsistency’ declared here
```
   85 | extern Oid      enforce_generic_type_consistency(const Oid *actual_arg_types,
```
```
  |                 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
```
clauses.c:5125:33: note: ‘actualargtypes’ declared here
```
 5125 |         Oid                     actual_arg_types[FUNC_MAX_ARGS];
```
```
  |                                 ^~~~~~~~~~~~~~~~
```
HTH,
Erik
View in PostgreSQL Archives →
- Jump to comment-1
  Tom Lane<tgl@sss.pgh.pa.us>
  Apr 30, 2026, 9:21 PM UTC
  Erik Rijkers <er@xs4all.nl> writes:
  Compiling with gcc 16.1.0 gives the following warning and notes:
  (horizontally compressed a bit te avoid excessive wrapping)
  clauses.c: In function ‘recheckcastfunction_args.isra’:
  clauses.c:5139:19: warning: ‘actualargtypes’ may be used uninitialized
  [-Wmaybe-uninitialized]
  I'm inclined to write that off as "buggy early-version compiler".
  That code is visibly okay, and no other compiler version has warned
  about it.
  regards, tom lane
  View in PostgreSQL Archives →