More of a call from David. If you could take a look at my CMakeFile.txt changes that would be appreciated too :-) They add detection of available optimisations under OSX systems. The PR passes travis tests, so it doesn't appear to impact builds on Ubuntu systems. Worth checking on other distros, but I doubt it will cause issues (unless somehow your distro identifies itself as 'Darwin'!!)
73 Mark On Mon, Apr 22, 2019 at 11:38 PM Richard Shaw <[email protected]> wrote: > Renaming to "unsigned int" fixed building for me on MINGW... > > diff --git a/src/idct.c b/src/idct.c > index 08aa41c..b467b65 100644 > --- a/src/idct.c > +++ b/src/idct.c > @@ -28,8 +28,8 @@ float std[] = { > int main(int argc, char *argv[]) { > FILE *fin, *fout; > fin = stdin; fout = stdout; > - uint ret; > - uint stride = NB_BANDS; > + unsigned int ret; > + unsigned int stride = NB_BANDS; > int measure = 0; > int scaling = 0; > > @@ -63,7 +63,7 @@ int main(int argc, char *argv[]) { > float sum[NB_BANDS] = {0.0}; > float sumsq[NB_BANDS] = {0.0}; > float dctLy[stride], Ly[stride]; > - uint i; for(i=0; i<stride; i++) Ly[stride] = 0.0; > + unsigned int i; for(i=0; i<stride; i++) Ly[stride] = 0.0; > long n = 0; > while(fread(dctLy, sizeof(float), stride, fin) == stride) { > idct(Ly, dctLy); > > > Any objection to committing? > > Thanks, > Richard > KF5OIM > _______________________________________________ > Freetel-codec2 mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/freetel-codec2 >
_______________________________________________ Freetel-codec2 mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/freetel-codec2
