lkml.org 
[lkml]   [2009]   [Aug]   [21]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    Date
    From
    SubjectRe: [PATCH 00/14] kconfig: streamline distro configs for testers

    * Steven Rostedt <rostedt@goodmis.org> wrote:

    >
    > On Fri, 21 Aug 2009, Ingo Molnar wrote:
    >
    > >
    > > * Steven Rostedt <rostedt@goodmis.org> wrote:
    > >
    > > >
    > > > On Fri, 21 Aug 2009, Steven Rostedt wrote:
    > > > >
    > > > > OK, I found the bug. There's a dependency missing.
    > > > >
    > > > > $ make localyesconfig
    > > > > HOSTCC scripts/basic/fixdep
    > > > > HOSTCC scripts/basic/docproc
    > > > > HOSTCC scripts/basic/hash
    > > > > using config: '/boot/config-2.6.27.25-78.2.56.fc9.x86_64'
    > > > > /bin/sh: line 8: scripts/kconfig/conf: No such file or directory
    > > > > make[1]: *** [localyesconfig] Error 127
    > > > > make: *** [localyesconfig] Error 2
    > > > >
    > > > > Thus the script failed to run. But the sed command that converts the '=m'
    > > > > to '=y' still ran. This gives us a distro config with all modules
    > > > > converted to built in!
    > > >
    > > > Funny that localmodconfig does not have this bug.
    > >
    > > ... add to that i dont use modules much so i probably hit that
    > > localyesconfig bug ;-)
    > >
    > > btw: mind adding localrandconfig too? It would be a perfect tool to
    > > randomize all configs except a 'must have' core set of options.
    > >
    > > ( Or is this already possible via some existing kconfig feature? I
    > > remember having seen something related, did that ever get
    > > upstream? A quick grep in scripts/ shows nothing but i might have
    > > missed it.)
    >
    > OK, I'll add that to my todo list. Note, it the local config is
    > only as good as the starting .config. That is, it never touches =y
    > settings, because it only uses loaded modules as a base. Thus if
    > something is set to =y it will always be set to =y.
    >
    > What would you have in mind. Just randomly enabling modules? If we
    > enable a "y" config then the next local*config will still keep
    > that set.

    Why - isnt local*config picking up environmental (currently running
    kernel) properties? I.e. it would always set those things that are
    needed - and randomize the rest. Have i misunderstood how it works?

    Ingo


    \
     
     \ /
      Last update: 2009-08-21 21:55    [W:0.032 / U:1.228 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site