Re: [Xen-devel] [PATCH] tools: link executables with libtinfo explicitly

[Daniel Kiper <daniel.kiper@xxxxxxxxxx>]

On Mon, Jun 01, 2015 at 02:06:44PM +0100, Wei Liu wrote:
> On Sat, May 30, 2015 at 12:07:28AM +0200, Daniel Kiper wrote:
> > binutils 2.22 changed ld default from --copy-dt-needed-entries
> > to -no-copy-dt-needed-entries. This revealed that some objects
> > are linked implicitly with libtinfo and newer ld fails to build
> > relevant executables.
> >
> > Below is short explanation why we should not do that...
> >
> > http://fedoraproject.org/wiki/UnderstandingDSOLinkChange says:
> >
> > The default behaviour for ld (my note: before version 2.22) allows
> > users to 'indirectly' link to required objects/libraries through
> > intermediate objects/libraries. While this is convenient, it can
> > also be dangerous because it makes your program's dependencies tied
> > to the dependencies of other objects. If those objects ever change
> > their linkages, they can break your program without any changes
> > to your own code!
> >
> > Signed-off-by: Daniel Kiper <daniel.kiper@xxxxxxxxxx>
> > ---
> >  config/Tools.mk.in            |    1 +
> >  tools/configure               |   46 
> > +++++++++++++++++++++++++++++++++++++++++
>
> Since your autoconf version might be different from the one used by
> committers, it might be better you don't include the changes to
> configure in your patch and add a line to prompt committer to rerun
> autogen.sh.

OK.

> >  tools/configure.ac            |    4 ++++
> >  tools/misc/Makefile           |    2 +-
> >  tools/xenstat/xentop/Makefile |    2 +-
> >  5 files changed, 53 insertions(+), 2 deletions(-)
> >
> > diff --git a/config/Tools.mk.in b/config/Tools.mk.in
> > index d67352e..aef9ed6 100644
> > --- a/config/Tools.mk.in
> > +++ b/config/Tools.mk.in
> > @@ -77,5 +77,6 @@ CONFIG_LIBICONV     := @libiconv@
> >  CONFIG_GCRYPT       := @libgcrypt@
> >  EXTFS_LIBS          := @EXTFS_LIBS@
> >  CURSES_LIBS         := @CURSES_LIBS@
> > +TINFO_LIBS          := @TINFO_LIBS@
> >
> >  FILE_OFFSET_BITS    := @FILE_OFFSET_BITS@
> [...]
> >
> >
> > diff --git a/tools/configure.ac b/tools/configure.ac
> > index 9bf6450..1a06ddf 100644
> > --- a/tools/configure.ac
> > +++ b/tools/configure.ac
> > @@ -318,6 +318,10 @@ i[[3456]]86|x86_64)
> >  esac
> >   AX_CHECK_UUID
> >   AX_CHECK_CURSES
> > +AS_IF([test "$ncurses" = "y"], [
> > +AC_CHECK_LIB([tinfo], [define_key], [TINFO_LIBS=-ltinfo])
>
> "define_key"?

Why not? It is used by gtraceview. If you wish we could test for
halfdelay() which is used by xentop.

> > +])
> > +AC_SUBST(TINFO_LIBS)
> >
>
> This doesn't look right. Tinfo is needed by libxenstat. It should not
> depend on presence of curses library.

libtinfo is a part of ncurses.

Daniel

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel