Date: Tue, 8 Oct 2002 14:31:00 -0400 (EDT) From: Garrett Wollman <wollman@lcs.mit.edu> To: Tim Robbins <tjr@FreeBSD.ORG> Cc: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/usr.bin Makefile src/usr.bin/c99 Makefile c99.1 c99.c Message-ID: <200210081831.g98IV0rU047347@khavrinen.lcs.mit.edu> In-Reply-To: <20021008122652.A63354@dilbert.robbins.dropbear.id.au> References: <200210070937.g979bt5T039003@freefall.freebsd.org> <200210071549.g97Fnxwe039156@khavrinen.lcs.mit.edu> <20021008122652.A63354@dilbert.robbins.dropbear.id.au>
next in thread | previous in thread | raw e-mail | index | archive | help
<<On Tue, 8 Oct 2002 12:26:52 +1000, Tim Robbins <tjr@FreeBSD.ORG> said:
> I've added back the library name remapping code that I removed before I
> committed it the first time. It now does this:
> -lpthread -> -pthread
I don't think that this is sufficient. I believe that `c99' must
invoke `cc -pthread' (or whatever is necessary to get the right
header definitions for threaded programs) always, since there is no
requirement that `-l pthread' be specified on the command line for
separate compilation, only for linkage. Should probably also ignore
any explicit requests for `-l c'.
> Is this ok? I didn't bother with -ltrace since we don't claim to support
> those interfaces.
I'd like to see all of them at least stubbed out, so that if we
implement the interface later, we don't have to come back and touch
`c99' again. For example:
#if _POSIX_TRACE < 1
if (strcmp(lib, "trace")
#if _POSIX_TRACE == 0
&& sysconf(_SC_TRACE) > 0
#endif
)
warnx("-l trace is not supported");
else
#else
errx(APPROPRIATE_VALUE, "-l trace needs to be implemented");
#endif
-GAWollman
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200210081831.g98IV0rU047347>