Eric N. Vander Weele added the comment: I am able to replicate what Michael has provided (i.e., xlc does not support signed short). Sorry for the confusion: I overlooked that the compiler is emitting the error with may patch and assuming 'unsigned'.
So it seems like there are two things to address: (1) The bit fields should be explicitly marked as 'signed', since that appears to be the desired intent. (2) What to do about the test case. It seems like we all agree on (1). For (2), is this something that should be stubbed out on AIX/xlc, resolved by xlc for supporting implementation-defined (short) bit-fields, or remove the short members in the struct since C99 (6.7.2.1) allows "a qualified or unqualified version of _Bool, signed int, unsigned int, or some other implementation-defined type"; thus removing the ambiguity for implementation-defined behavior? ---------- _______________________________________ Python tracker <rep...@bugs.python.org> <http://bugs.python.org/issue27643> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
