[OE-core] [morty][PATCH] Check libcc1 directory before disabling hardcoding of RPATHs

Daniel Gomez daniel.gomez at silicon-gears.com
Fri Aug 10 14:39:51 UTC 2018


On vie, 2018-07-27 at 12:59 -0700, Andre McCurdy wrote:
> On Fri, Jul 27, 2018 at 6:07 AM, Daniel Gomez
> <daniel.gomez at silicon-gears.com> wrote:
> > 
> > Check if libcc1 directory exits before disabling hardcode_into_libs
> > configuration because of broken libtool in gcc recipes.
> A patch for morty should either be a backport which has already been
> merged into all applicable later branches (ie if the same fix is
> needed in sumo, rocko or pyro it needs to be merged to those branches
> first) or it should explain clearly why it's only applicable to morty.
Well, I have realized about the real problem and it is only affecting to 
the gcc_4.9.bb recipe in the meta-linaro repo and not in the poky one.
Basically, libcc1 support is included in gcc versions >= 5.x but not in
older versions like 4.9.x and poky does not check the 'hardcode_into_libs'
flags in versions where libcc1 is not included.
> 
> > 
> > Signed-off-by: Daniel Gomez <daniel.gomez at silicon-gears.com>
> > ---
> >  meta/recipes-devtools/gcc/gcc_5.4.bb | 3 ++-
> >  meta/recipes-devtools/gcc/gcc_6.4.bb | 3 ++-
> >  2 files changed, 4 insertions(+), 2 deletions(-)
> > 
> > diff --git a/meta/recipes-devtools/gcc/gcc_5.4.bb b/meta/recipes-devtools/gcc/gcc_5.4.bb
> > index b0a523c..9da8ece 100644
> > --- a/meta/recipes-devtools/gcc/gcc_5.4.bb
> > +++ b/meta/recipes-devtools/gcc/gcc_5.4.bb
> > @@ -9,7 +9,8 @@ ARM_INSTRUCTION_SET_armv4 = "arm"
> >  do_configure_prepend() {
> >         # Easiest way to stop bad RPATHs getting into the library since we have a
> >         # broken libtool here
> > -       sed -i -e 's/hardcode_into_libs=yes/hardcode_into_libs=no/' ${S}/libcc1/configure
> > +       if [ -d "${S}/libcc1" ]; then
> > +               sed -i -e 's/hardcode_into_libs=yes/hardcode_into_libs=no/' ${S}/libcc1/configure
> Missing fi ?
Yes! I forgot to copy it from the original patch:
https://lists.linaro.org/pipermail/openembedded/2018-March/000137.html
> > 
> >  }
> > 
> >  BBCLASSEXTEND = "nativesdk"
> > diff --git a/meta/recipes-devtools/gcc/gcc_6.4.bb b/meta/recipes-devtools/gcc/gcc_6.4.bb
> > index b0a523c..9da8ece 100644
> > --- a/meta/recipes-devtools/gcc/gcc_6.4.bb
> > +++ b/meta/recipes-devtools/gcc/gcc_6.4.bb
> > @@ -9,7 +9,8 @@ ARM_INSTRUCTION_SET_armv4 = "arm"
> >  do_configure_prepend() {
> >         # Easiest way to stop bad RPATHs getting into the library since we have a
> >         # broken libtool here
> > -       sed -i -e 's/hardcode_into_libs=yes/hardcode_into_libs=no/' ${S}/libcc1/configure
> > +       if [ -d "${S}/libcc1" ]; then
> > +               sed -i -e 's/hardcode_into_libs=yes/hardcode_into_libs=no/' ${S}/libcc1/configure
> >  }
> > 
> >  BBCLASSEXTEND = "nativesdk"
> > --
> > 2.7.4
> > 
> > --
> > _______________________________________________
> > Openembedded-core mailing list
> > Openembedded-core at lists.openembedded.org
> > http://lists.openembedded.org/mailman/listinfo/openembedded-core



More information about the Openembedded-core mailing list