Messages in this thread | | | Date | Sun, 25 May 2008 17:27:48 -0700 | From | "H. Peter Anvin" <> | Subject | Re: kernel coding style for if ... else which cross #ifdef |
| |
Adrian Bunk wrote: > > We do not have any serious problems where we actually depend on > randconfig. > > randconfig is nice, but even if it would suddenly become no longer > available we wouldn't have significantely more build breakages in > stable kernels (perhaps a few more in the pathological cornercases > only randconfig hits). > > If this was really "the whole point" it wasn't not worth it. > > Real value would come from getting errors for mistyped kconfig variable > names. >
Indeed. However, getting at least compiler coverage across all branches of code is worth something, at least.
-hpa
| |