On Thu, Jan 15, 2009 at 05:16:34PM +0000, Richard W.M. Jones wrote:
On Tue, Jan 13, 2009 at 05:42:41PM +0000, Daniel P. Berrange wrote:
The strerror() method is not guarenteed to be re-entrant, which is rather a pain because the strerror_r() method is rather unpleasant to use. In addition our code is quite inconsistent about using VIR_ERR_SYSTEM_ERROR vs VIR_ERR_INTERNAL_ERROR for problems which have an 'errno' avilable. Likewise we're not very consistent about OOM reporting error codes
This patch thus introduces two convenient functions for reporting a system error and OOM error.
virReportSystemError(conn, theerrno, fmt,...) virReportOOMError(conn)
Yes, a very welcome change. I've looked through the patch briefly and it looks good, and hoping gcc would pick up any gross argument mismatches. So +1.
diff --git a/autobuild.sh b/autobuild.sh --- a/autobuild.sh +++ b/autobuild.sh @@ -65,6 +65,7 @@ if [ -x /usr/bin/i686-pc-mingw32-gcc ]; --build=$(uname -m)-pc-linux \ --host=i686-pc-mingw32 \ --prefix="$AUTOBUILD_INSTALL_ROOT/i686-pc-mingw32/sys-root/mingw" \ + --enable-compile-warnings=error \ --without-sasl \ --without-avahi \ --without-polkit \
Intended?
Yes, I noticed the MinGW build is now completely free of warnings so figured I'd add that to stop us ever re-introducing warnings on MinGW :-) Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|