On 03/03/14 11:27, Stefan Schmidt wrote: > Hello. > > On Mon, 2014-03-03 at 11:01, Tom Hacohen wrote: >> On 03/03/14 10:03, Stefan Schmidt wrote: >>> Hello. >>> >>> On Mon, 2014-03-03 at 09:35, Tom Hacohen wrote: >>>> On 28/02/14 15:12, Stefan Schmidt wrote: >>>>> Hello. >>>>> >>>>> I compiled the example today and got sad to see how many warnings we >>>>> have there. I fixed up a bunch which have been only noisec (unused >>>>> parameter) and now we can see some more interesting ones. Waiting for >>>>> people to look at them and fix. >>>>> >>>>> Even the main src di got tons of new warnings recently. Biggest >>>>> offenders right now are: >>>>> >>>>> evas_map_image_loop : tons of shadow local var warnings. One local >>>>> shadowing >>>>> a nother local. bad. >>>>> >>>>> gl_x11/evas_engine.c: signedness problems and more >>>>> >>>>> Lost of smaller warnings. >>>>> >>>>> It would be great if the related folks could have a look and make sure >>>>> they are not sneaking in new warnings all the time. >>>>> >>>>> I don't want to have a situation where we have an additional got fail: >>>>> and don't sport the warning for it. ;) (There is none, I know) >>>>> >>>> >>>> There used to be a warning for this (-Wunreachable-code), but it got >>>> removed. :( Fortunately, clang-analyzer still does it. >>> >>> Did not help a big company to find the double goto fail; :) >>> >>>> My cflags: >>>> -Wall -Wextra -Wshadow -Wno-type-limits -Wpointer-arith >>>> >>>> The biggest problem is that people don't use the right cflags. >>> >>> EFL_CHECK_COMPILER_FLAGS([EFL], [-Wall -Wextra -Wpointer-arith >>> -Wno-missing-field-initializers -fvisibility=hidden -fdata-sections >>> -ffunction-sections]) >>> >>> That is what we set by default. -Wshadow depends on if you have a >>> compiler that complains about shadowing globals. That might be an >>> excuse for people not having -Wshadow, but all other should not mess >>> with their cflags in a way that they get way less warning reports. >> >> Odd, cause Cedric didn't have -Wshadow. I guess that those flags are not >> appended, and are only used if CFLAGS is not set? > > I have none set and get the defaults. Having them not append kind of > fails the goals to have a sane set for efl but having them set will > make people scream on how much better they know what to set. :) > > The missing -WShadow could also be from the case that the compiler > would emit warnings for shadow global and thus not set the flag. We > have an explicit test for this in configure.ac >
We should improve our checks then. :) -- Tom. ------------------------------------------------------------------------------ Subversion Kills Productivity. Get off Subversion & Make the Move to Perforce. With Perforce, you get hassle-free workflows. Merge that actually works. Faster operations. Version large binaries. Built-in WAN optimization and the freedom to use Git, Perforce or both. Make the move to Perforce. http://pubads.g.doubleclick.net/gampad/clk?id=122218951&iu=/4140/ostg.clktrk _______________________________________________ enlightenment-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
