[Buildroot] [PATCH 1/2] nilfs-utils: need NPTL threads

[Kurt Van Dijck]

Hey,

> On Thu, Jan 04, 2018 at 11:10:51AM +0100, Kurt Van Dijck wrote:
> > nilfs-utils use clock_nanosleep(), which comes with NPTL threads
> 
> What is your indication that this is the case? In the uClibc-ng config 
> UCLIBC_HAS_REALTIME is selected by both linuxthreads and NPTL. When 
> UCLIBC_HAS_REALTIME is enabled, UCLIBC_HAS_ADVANCED_REALTIME is also enabled 
> by default. UCLIBC_HAS_ADVANCED_REALTIME provides clock_nanosleep().
> 
> > Signed-off-by: Kurt Van Dijck <dev.kurt at vandijck-laurijssen.be>
> > ---
> >  package/nilfs-utils/Config.in | 1 +
> >  1 file changed, 1 insertion(+)
> > 
> > diff --git a/package/nilfs-utils/Config.in b/package/nilfs-utils/Config.in
> > index 0c6b7a0..9dda27c 100644
> > --- a/package/nilfs-utils/Config.in
> > +++ b/package/nilfs-utils/Config.in
> > @@ -1,6 +1,7 @@
> >  config BR2_PACKAGE_NILFS_UTILS
> >  	bool "nilfs-utils"
> >  	depends on BR2_TOOLCHAIN_HAS_THREADS # sem_open()
> > +	depends on BR2_TOOLCHAIN_HAS_THREADS_NPTL # clock_nanosleep()
> 
> NPTL implies threads support, so the NPTL dependency alone is enough.
> 
> You also need to update the comment.
> 

IMHO, appending the comment 'sem_open()' to BR2_..._NPTL would only
cause confusion, since sem_open() does not depend on BR2_..._NPTL.

Given the recent discussion about clock_nanosleep not really depending
on BR2_..._NPTL, I suspect a change in the near future, so I hesitate to
remove the BR2_..._THREADS dependency.

How exactly is the current patch wrong?
KConfig resolves this, better than any of us would, so I see no
advantage in modifying the patch.

> >  	depends on BR2_USE_MMU # util-linux libmount, libblkid
> >  	select BR2_PACKAGE_UTIL_LINUX_LIBUUID
> >  	select BR2_PACKAGE_UTIL_LINUX_LIBBLKID

Kind regards,
Kurt