On 9/17/2018 4:20 PM, Bruce Richardson wrote: > On Mon, Sep 17, 2018 at 03:53:40PM +0100, Ferruh Yigit wrote: >> On 9/13/2018 3:44 PM, Bruce Richardson wrote: >>> Compiling with all warnings turned on causes errors about unused variables >>> and an unused label. Remove these to allow building without having to >>> disable those warnings. >> >> If this is fixing some build errors should it be backported? >> >> And why we didn't get those warnings until now, aren't we already compiling >> with >> all warnings turned on? >> >>> >>> Signed-off-by: Bruce Richardson <bruce.richard...@intel.com> >> > It does not appear so. When we add support for building this driver with > meson, the compiler warnings then appeared [See > http://patches.dpdk.org/patch/44698/]. > > While you can backport I'm not sure it's worthwhile, unless you plan on > backporting the new meson support too. The warnings don't come from > external header files that apps would be using directly with unknown > compiler flags, but are internal to DPDK and don't seem to cause any issues > with "make" builds.
Argh, it is missing warning flags in Makefile: +CFLAGS += $(WERROR_FLAGS) When above added "make" also gives "unused variable" warnings. Even we don't observe the issue with current makefile, if someone builds with: 'make EXTRA_CFLAGS="-Wunused-variable"' will observe the warnings. That is why I am for backporting this patch, does it make sense? And would you mind adding above CFLAGS update into this patch?
