Re: [meta-networking][PATCH] autofs: update to current patchlevel

[Joe MacDonald <Joe.MacDonald@windriver.com>]

[-- Attachment #1: Type: text/plain, Size: 3503 bytes --]

[Re: [oe] [meta-networking][PATCH] autofs: update to current patchlevel] On 13.03.19 (Tue 16:40) Andreas Oberritter wrote:

> On 19.03.2013 13:49, Joe MacDonald wrote:
> > [[oe] [meta-networking][PATCH] autofs: update to current patchlevel] On 13.03.19 (Tue 11:24) Andreas Oberritter wrote:
> > 
> >> * Apply all patches from ftp.kernel.org/pub/linux/daemons/autofs/v5/patches-5.0.8/
> >>
> >> Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
> >> ---
> >>  .../autofs-5.0.7-add-symlink-pseudo-option.patch   |  217 +++++++++
> >>  ...witch_conf-to-not-contain-automount-lines.patch |   50 +++
> >>  ...s-5.0.7-depricate-nosymlink-pseudo-option.patch |   40 ++
> >>  ...-5.0.7-document-browse-option-in-man-page.patch |   46 ++
> >>  ....0.7-dont-fail-on-master-map-self-include.patch |   59 +++
> >>  ....7-dont-schedule-new-alarms-after-readmap.patch |   69 +++
> >>  ...ent-d_type-to-filter-out-files-in-scandir.patch |   41 ++
> >>  ...s-5.0.7-fix-automounter-support-on-parisc.patch |   24 +
> >>  ...descriptor-leak-when-reloading-the-daemon.patch |  179 ++++++++
> >>  ...utofs-5.0.7-fix-init-script-status-return.patch |   39 ++
> >>  .../autofs-5.0.7-fix-libtirpc-build-option.patch   |   66 +++
> >>  ...-fix-map-entry-duplicate-offset-detection.patch |   37 ++
> >>  ...ofs-5.0.7-fix-nobind-man-page-description.patch |   46 ++
> >>  .../autofs-5.0.7-fix-requires-in-spec-file.patch   |   37 ++
> >>  .../autofs-5.0.7-fix-submount-offset-delete.patch  |   45 ++
> >>  ...ofs-5.0.7-fix-systemd-unidir-in-spec-file.patch |   72 +++
> >>  ....7-fix-use-get_proximity-without-libtirpc.patch |   54 +++
> >>  ...s-5.0.7-fix-wildcard-multi-map-regression.patch |  225 ++++++++++
> >>  ...e-WITH_LDAP-conditional-around-LDAP-types.patch |  448 +++++++++++++++++++
> >>  ...-5.0.7-make-yellow-pages-support-optional.patch |   15 +-
> >>  ...odules-replicated-use-sin6.addr-s6_addr32.patch |   39 ++
> >>  .../autofs-5.0.7-update-kernel-include-files.patch |   92 ++++
> >>  ...-protocol-ids-instead-of-protoent-structs.patch |  471 ++++++++++++++++++++
> >>  ...karound-missing-GNU-versionsort-extension.patch |  149 +++++++
> >>  .../recipes-daemons/autofs/autofs_5.0.7.bb         |   25 +-
> >>  25 files changed, 2582 insertions(+), 3 deletions(-)
> > 
> > I had started to ask why we wouldn't just update to 5.0.8 directly
> > instead of patching our way there.  Then I thought I'd look myself and
> > now I know.  :-)
> > 
> > This is just a simple (and wholesale) integration of the patches already
> > upstream?  That seems reasonable to me.
> 
> Yes.
> 
> In the future, maybe we should switch to Git. A repository for autofs is
> available at
> git://git.kernel.org/pub/scm/linux/storage/autofs/autofs.git. For now, I
> didn't want to change the recipe too much.
> 
> One of these patches fixed a problem, where the automount daemon would
> hang after receiving SIGHUP, but I don't know which one exactly.

Woo.  Nice.  Well, on the whole I don't have any objection to
5.0.8-ifying this anyway, so importing the batch is reasonable even if
it's just to resolve a single problem.

I'd be a little concerned about going to a straight git source based
solely on some hassle I've had with meta-selinux in the past, but
there's a history for that, too, so maybe it's not a bad idea either.

As it stands, though, I'm happy to merge this.  We can visit the git
tree question pretty much any time.

-- 
-Joe MacDonald.
:wq

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]