This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug c/12213] [3.4 regression] warning: conflicting types for builtin-function 'gamma'
- From: "mec at shout dot net" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 8 Sep 2003 21:24:54 -0000
- Subject: [Bug c/12213] [3.4 regression] warning: conflicting types for builtin-function 'gamma'
- References: <20030908190213.12213.mec@shout.net>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12213
------- Additional Comments From mec at shout dot net 2003-09-08 21:24 -------
Subject: Re: [3.4 regression] warning: conflicting types for builtin-function 'gamma'
Okay, I found the language in a draft of C9X. Section 7.1.3 says that
external identifiers are reserved unconditionally, without saying that
the header file must be included. I don't like it but that's the
specification.
But are you sure that 'gamma' is defined in C89? I don't have a C89
standard, but I do have a draft C9X standard, and there is no 'gamma'
in math.h in C9X. There is only:
lgamma lgammaf lgammal
tgamma tgammaf tgammal
I wonder if 'gamma' is defined in C89?
So 'gamma' should not be a reserved identifier with "gcc -std=c99". And
indeed, "gcc -std=c99" compiles the test program just fine. So does
"gcc -std=c89".
I guess that means 'gamma' is a gnu extension, which is okay.
Michael C