On Mon, Oct 24, 2022 at 10:29:55AM -0700, Nathan Chancellor wrote: > On Fri, Oct 21, 2022 at 08:05:19PM -0700, Xin Li wrote: > > Andrew Cooper suggested upgrading the orphan section warning to a hard link > > error. However Nathan Chancellor said outright turning the warning into an > > error with no escape hatch might be too aggressive, as we have had these > > warnings triggered by new compiler generated sections, and suggested turning > > orphan sections into an error only if CONFIG_WERROR is set. Kees Cook echoed > > and emphasized that the mandate from Linus is that we should avoid breaking > > builds. It wrecks bisection, it causes problems across compiler versions, etc. > > > > Thus upgrade the orphan section warning to a hard link error only if > > CONFIG_WERROR is set. > > > > Suggested-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx> > > Suggested-by: Nathan Chancellor <nathan@xxxxxxxxxx> > > Signed-off-by: Xin Li <xin3.li@xxxxxxxxx> > > Thanks for the patch! > > Reviewed-by: Nathan Chancellor <nathan@xxxxxxxxxx> > Tested-by: Nathan Chancellor <nathan@xxxxxxxxxx> > > We could deduplicate the '$(if $(CONFIG_WERROR),error,warn)' logic if we > hoisted it into Kconfig by having something like > > config LD_ORPHAN_WARN_LEVEL > string > depends on LD_ORPHAN_WARN > default "error" if WERROR > default "warn" > > in init/Kconfig then using it everywhere like > > --orphan-handling=$(CONFIG_LD_ORPHAN_WARN_LEVEL) > > but I will let others decide if they would prefer that over the > direction we went here. I think this makes it look cleaner, yeah. -- Kees Cook
