On 12/14/2012 08:08 AM, Viktor Mihajlovski wrote:
Some older versions of GCC report a false positive on code like char * haystack, needle; strchr(haystack, needle);
Added an extra check in configure.ac which will #define BROKEN_GCC_WLOGICALOP 1 in this case, allowing to special handle "offending" code.
Signed-off-by: Viktor Mihajlovski <mihajlov@linux.vnet.ibm.com> --- configure.ac | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+)
I know this is already pushed, but I think it would fit better under m4/virt-compile-warnings.m4 rather than at the top-level configure.ac.
+ +if test "x$gcc_false_strchr_warning" = xyes; then + AC_DEFINE_UNQUOTED([BROKEN_GCC_WLOGICALOP], 1, + [GCC -Wlogical-op is reporting false positive on strchr]) +fi
Potential ouch. This sets BROKEN_GCC_WLOGICALOP to 1 on all non-gcc compilers. Then in patch 2/3, you blindly use this condition to request a pragma: +/* + we need to ignore warnings about strchr caused by -Wlogical-op + for some GCC versions. + Doing it via a local pragma keeps the damage smaller than + disabling it on the package level. + Unfortunately, the affected GCCs don't allow diagnostic push/pop + which would have further reduced the impact. + */ +# if BROKEN_GCC_WLOGICALOP +# pragma GCC diagnostic ignored "-Wlogical-op" +# endif We are not guaranteed that other compilers behave nicely when presented with unknown #pragma GCC. So far, our use of such pragmas has been guarded by knowing that we are using gcc; so I'd feel better if the configure-time check for setting this flag were limited to just gcc in the first place (which it would be easier to do by moving this test to m4/virt-compile-warnings.m4). -- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org