Gentoo Archives: gentoo-musl

From: "Anthony G. Basile" <blueness@g.o>
To: Lei Zhang <zhanglei.april@×××××.com>, Eric Fiselier <eric@××××.ca>
Cc: Clang Dev <cfe-dev@××××××××××.org>, gentoo-musl@l.g.o
Subject: Re: [gentoo-musl] Re: [cfe-dev] Clang++ always defines _GNU_SOURCE
Date: Tue, 14 Jun 2016 11:42:47
Message-Id: 20875379-b555-8f81-fd8b-5deb16b692ca@gentoo.org
In Reply to: [gentoo-musl] Re: [cfe-dev] Clang++ always defines _GNU_SOURCE by Lei Zhang
1 On 6/14/16 5:54 AM, Lei Zhang wrote:
2 > 2016-06-14 16:54 GMT+08:00 Eric Fiselier <eric@××××.ca>:
3 >>> Does libc++ also rely on this macro to work on Linux?
4 >>
5 >> Yes. Adding -U_GNU_SOURCE during the libc++ build results in a litany of
6 >> errors.
7 >> The libc++ headers depend on a number of C library symbols that only get
8 >> defined when -D_GNU_SOURCE=1 is present.
9 >
10 > Is it feasible to use some finer-grain control like _ISOC99_SOURCE,
11 > instead of resorting to the too versatile _GNU_SOURCE?
12 >
13 > I'd be willing to work out such a patch to libc++, if it makes sense.
14 >
15 >
16 > Lei
17 >
18
19 That is the better approach. The problem is that we tend to lump groups
20 of functions under macros like _GNU_SOURCE or __UCLIBC__ only to run
21 into problems later.
22
23 --
24 Anthony G. Basile, Ph.D.
25 Gentoo Linux Developer [Hardened]
26 E-Mail : blueness@g.o
27 GnuPG FP : 1FED FAD9 D82C 52A5 3BAB DC79 9384 FA6E F52D 4BBA
28 GnuPG ID : F52D4BBA