This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Solaris 10 patch status
On Tue, 23 Nov 2004, Andrew Pinski wrote:
> PR 7544 was marked as a regression, then after rth commented on it,
> it was decided this was not a bug (why I did not close it at that point
> is because I had read the email wrong saying it should be rejected, it
> was only the asm which gcc 2.95.3 should be rejected).
>
> http://gcc.gnu.org/ml/gcc/2004-01/msg00660.html
The Sun manual referenced in that thread is a C++ manual. That bug report
is a C bug report. Please be more careful about distinguishing the two,
in both directions; only recently you closed a C++ bug report on the basis
of a C patch. The current Solaris pragma documentation
<http://docs.sun.com/source/817-5064/sun.specific.html#48658> is quite
clear that in
#pragma weak symbol1 = symbol2
symbol2 must be defined, but symbol1 need not be. There is a clear
regression on, for example,
#pragma weak bar = foo
void foo(void) {}
A more recent C++ manual than the one referenced in the thread
<http://docs.sun.com/source/817-0926/Pragmas_App.html> suggests that the
same is true for C++.
--
Joseph S. Myers http://www.srcf.ucam.org/~jsm28/gcc/
jsm@polyomino.org.uk (personal mail)
joseph@codesourcery.com (CodeSourcery mail)
jsm28@gcc.gnu.org (Bugzilla assignments and CCs)