Re: FFmepg rebuild Schedule ?

Hi, On 06/24/2012 05:41 PM, Sérgio Basto wrote: > On Dom, 2012-06-24 at 16:22 +0200, Hans de Goede wrote: >> Hi, >> >> Just a quick note, while doing a local build of all >> deps to work on my ffmpeg dependent packages, I >> noticed the following in x264's config.log file: >> >> checking for gf_malloc(1); gf_free(NULL); in gpac/isomedia.h... no >> Failed commandline was: >> -------------------------------------------------- >> gcc conftest.c -m64 -Wall -I. -I$(SRCPATH) -O2 -g -pipe -Wall -Wp,-D_FORTIFY_S >> conftest.c: In function 'main': >> conftest.c:2:24: error: ignoring return value of 'malloc', declared with attrib >> cc1: all warnings being treated as errors >> -------------------------------------------------- >> Failed program was: >> -------------------------------------------------- >> #include <gpac/isomedia.h> >> int main () { gf_malloc(1); gf_free(NULL); return 0; } >> -------------------------------------------------- >> >> I think we should fix this ... > > checking for gpac/isomedia.h... yes > checking for gf_isom_set_pixel_aspect_ratio(0,0,0,0,0); in > gpac/isomedia.h... yes > checking for gf_malloc(1); gf_free(NULL); in gpac/isomedia.h... no > > not have gf_malloc and gf_free , is problematic ? No, x264 thinking it does not have them, while they are actually available is problematic (I don't know if it is causing any real issues, but it is wrong). Look at the error closely: >> conftest.c: In function 'main': >> conftest.c:2:24: error: ignoring return value of 'malloc', declared with attrib >> cc1: all warnings being treated as errors So the problem is that -Wall -Werror somehow is in the CFLAGS when the tests run, causing the test to fail (where it should succeed) > what you suggest to fix ? Fix the test code, changing it to something like: void *p; p = gf_malloc(1); gf_free(p);