[OE-core] [PATCH 00/13] [jethro] Consolidated pull

Hi RP, There are two PULLs, the one on openembedded-core-contrib is for oe-core, the other one on poky-contrib is for poky, the later one has 2 more commits: poky.conf: Bump version for 2.0.3 jethro release build-appliance-image: Update to jethro head revision I can't send them in oe-core's

[OE-core] what is the closest alternative to red hat's ABRT in OE?

as in: https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/6/html/Deployment_Guide/ch-abrt.html specifically, to manage core files? i would have thought judicious use of "ulimit" would be the solution. others? rday --

Re: [OE-core] [PATCH 3/6] keyutils: new recipe (version 1.5.9)

On 30 November 2016 at 22:43, Andreas Oberritter wrote: > Used by nfs-utils for nfsidmap, if available. > This breaks under musl: | key.dns_resolver.c: In function 'afsdb_hosts_to_addrs': | key.dns_resolver.c:374:21: error: 'UINT_MAX' undeclared (first use in this

Re: [OE-core] [PATCH 00/33] Accumulated patches for deb packaging

On 06.12.2016 15:36, Burton, Ross wrote: > > On 6 December 2016 at 11:49, Andreas Oberritter > wrote: > > These are most of my patches which accumulated since our distro > switched from opkg to dpkg and apt two years ago. They were

[OE-core] [PATCH] dhcp: 4.3.4 -> 4.3.5

Upgrade dhcp from 4.3.4 to 4.3.5. Signed-off-by: Huang Qiyu --- meta/recipes-connectivity/dhcp/binutils-gdb | 1 + meta/recipes-connectivity/dhcp/{dhcp_4.3.4.bb => dhcp_4.3.5.bb} | 4 ++-- 2 files changed, 3 insertions(+), 2 deletions(-) create

Re: [OE-core] [PATCH] less: 481 -> 487

On 6 December 2016 at 23:05, Huang Qiyu wrote: > LIC_FILES_CHKSUM = "file://COPYING;md5=d32239bcb673463ab874e80d47fae504 \ > -file://LICENSE;md5=48c26a307f91af700e1f00585f215aaf" > +

Re: [OE-core] [PATCH 03/33] dpkg-native: hardcode SYSCONFDIR to /etc in update-alternatives

On 6 December 2016 at 11:49, Andreas Oberritter wrote: > +sed -i -e 's|SYSCONFDIR|"/etc"|' ${S}/utils/update-alternatives.c > Hardcoding /etc is bad, can't you just remove the native sysroot prefix from $sysconfdir? Ross --

Re: [OE-core] [oe-commits] [openembedded-core] 67/76: gnutls: update to 3.5.6

On Tue, Dec 06, 2016 at 02:49:04PM +0100, Martin Jansa wrote: > On Wed, Nov 30, 2016 at 03:49:52PM +, g...@git.openembedded.org wrote: > > rpurdie pushed a commit to branch master > > in repository openembedded-core. > > > > commit 27f306a752d15ec62d2821d0146be4ffa10b7013 > > Author:

Re: [OE-core] [PATCH 00/33] Accumulated patches for deb packaging

On 6 December 2016 at 11:49, Andreas Oberritter wrote: > These are most of my patches which accumulated since our distro > switched from opkg to dpkg and apt two years ago. They were tested > thoroughly on dora and krogoth and just a little bit on master. > Is this on a

[OE-core] [PATCH] insane: Add SH4 musl mapping to the QA arch tests

This change allows to pass QA for packages built with sh4-oe-linux-musl toolchain, the problem is reproted while building core-image-minimal target: ERROR: readline-7.0-r0 do_package_qa: Error executing a python function in exec_python_func() autogenerated Signed-off-by: Vladimir Zapolskiy

Re: [OE-core] what is the closest alternative to red hat's ABRT in OE?

On Tue, Dec 6, 2016 at 3:49 PM, Robert P. J. Day wrote: > > as in: > > https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/6/html/Deployment_Guide/ch-abrt.html > > specifically, to manage core files? i would have thought judicious use > of "ulimit"

Re: [OE-core] [PATCH 3/6] keyutils: new recipe (version 1.5.9)

On 06.12.2016 16:49, Burton, Ross wrote: > > On 30 November 2016 at 22:43, Andreas Oberritter > wrote: > > Used by nfs-utils for nfsidmap, if available. > > > This breaks under musl: > > | key.dns_resolver.c: In function

Re: [OE-core] [PATCH] dhcp: 4.3.4 -> 4.3.5

On 6 December 2016 at 22:38, Huang Qiyu wrote: > +++ b/meta/recipes-connectivity/dhcp/binutils-gdb > @@ -0,0 +1 @@ > +Subproject commit 39eeab253474493bc9477dbb2bd9c8041f05764b > Can you resend without this bit? Ross --

[OE-core] [PATCH V6] package_manager: remove strings and migrate to direct arrays

When using subprocess call and check_output, it is better to use arrays rather than strings when possible to avoid whitespace and quoting problems. [ YOCTO #9342 ] Signed-off-by: Stephano Cetola --- meta/lib/oe/package.py | 13 +--

[OE-core] [PATCH V6] cleanup subprocess

Changed since V5: Smart needs to dump the installed packages to a file, as this file is used to detect changes in the install. This was breaking: test_incremental_image_generation Stephano Cetola (1): package_manager: remove strings and migrate to direct arrays meta/lib/oe/package.py

[OE-core] [morty][PATCH 1/2] python-3.5-manifest: Add imp to importlib

The imp python module is the forerunner of importlib. Include imp in the importlib subpackage instead of the misc subpackage so that it can be depended on without bringing in a bunch of unrelated, unused modules. Signed-off-by: George McCollister ---

[OE-core] [morty][PATCH 0/2] Resolve gdb-cross-canadian dependency on imp

Add imp to python3-importlib. Make gdb-cross-canadian depend on nativesdk-python3-importlib when python is enabled. The following changes since commit c8d96b10ee3bc2eae0fd269d2564286fd0bc82ed: rm_work: Ensure we don't remove sigbasedata files (2016-11-16 10:34:34 +) are available in the

Re: [OE-core] [PATCH 01/33] dpkg: implement offline mode for update-alternatives

On 6 December 2016 at 11:49, Andreas Oberritter wrote: > +++ b/meta/recipes-devtools/dpkg/dpkg/0003-update-alternatives- > Implement-offline-mode.patch > @@ -0,0 +1,399 @@ > +From b4f0f6ced469095a2b21b01b59c7aded057e Mon Sep 17 00:00:00 2001 > +From: Andreas Oberritter

Re: [OE-core] [oe-commits] [openembedded-core] 67/76: gnutls: update to 3.5.6

On 6 December 2016 at 14:44, Martin Jansa wrote: > Probably caused by nettle upgrade which no longer enables gmp with older > gmp 4.2.1 used in non-GPLv3 builds. > I was going to say I don't see this. Clearly we need to extend the non-GPL3 testing. As you can reproduce

Re: [OE-core] [oe-commits] [openembedded-core] 67/76: gnutls: update to 3.5.6

On Tue, Dec 06, 2016 at 02:47:42PM +, Burton, Ross wrote: > On 6 December 2016 at 14:44, Martin Jansa wrote: > > > Probably caused by nettle upgrade which no longer enables gmp with older > > gmp 4.2.1 used in non-GPLv3 builds. > > > > I was going to say I don't see

[OE-core] [PATCH] less: 481 -> 487

Upgrade less from 481 to 487. Signed-off-by: Huang Qiyu --- meta/recipes-extended/less/{less_481.bb => less_487.bb} | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) rename meta/recipes-extended/less/{less_481.bb => less_487.bb} (83%) diff --git

[OE-core] [PATCH] postinst: Add a test case to verify postinst scripts behavior

From: Jose Perez Carranza Add test case that verify behavior of postinst scripts at roofts time and when is delayed to the first boot directly on the target. Signed-off-by: Jose Perez Carranza ---

Re: [OE-core] [PATCH] gstreamer1.0-plugins-bad: Add PKG_CONFIG_SYSROOT_DIR to output of pkg-config

On 5 December 2016 at 19:08, Khem Raj wrote: > having said that, there could be a multilib issue if pkgdatadir thats in > .pc file from wayland-protocols package is using variables like ${libdir} > which then are > multilib dependent, if this is the case then we need to fix

[OE-core] [master][PATCH 2/2] gdb-cross-canadian: Depend on nativesdk-python3-importlib

Add missing dependency on nativesdk-python3-importlib so the imp Python module is installed. Before this patch, running gdb from the sdk would give the following error: Python Exception No module named 'imp': Signed-off-by: George McCollister ---

[OE-core] [master][PATCH 1/2] python-3.5-manifest: Add imp to importlib

The imp python module is the forerunner of importlib. Include imp in the importlib subpackage instead of the misc subpackage so that it can be depended on without bringing in a bunch of unrelated, unused modules. Signed-off-by: George McCollister ---

[OE-core] [master][PATCH 0/2] Resolve gdb-cross-canadian dependency on imp

Add imp to python3-importlib. Make gdb-cross-canadian depend on nativesdk-python3-importlib when python is enabled. The following changes since commit 11063a01d4511b2688ea7ba2d7359e4e07328c66: ruby: upgrade to 2.3.1 (2016-11-30 15:47:17 +) are available in the git repository at:

Re: [OE-core] [PATCH 20/33] package_manager/deb: let apt-get handle postinst scripts

On 06.12.2016 18:26, Burton, Ross wrote: > Can you elaborate on why the old code can be deleted? With all required environment variables and configuration options in apt.conf in place, apt-get is able to install packages offline, i.e. when creating the rootfs, including the execution of postinst

Re: [OE-core] [PATCH 21/33] package_manager/deb: create Packages.xz

On 6 December 2016 at 18:41, Andreas Oberritter wrote: > That's a good question. I guess I didn't add anything, because there's > no direct dependency in place for gzip-native either. > > I don't know if it counts, but there's a dependency on apt-native and > apt's source

[OE-core] [morty][PATCH 2/2] gdb-cross-canadian: Depend on nativesdk-python3-importlib

Add missing dependency on nativesdk-python3-importlib so the imp Python module is installed. Before this patch, running gdb from the sdk would give the following error: Python Exception No module named 'imp': Signed-off-by: George McCollister ---

Re: [OE-core] [PATCH 20/33] package_manager/deb: let apt-get handle postinst scripts

On 6 December 2016 at 18:20, Andreas Oberritter wrote: > With all required environment variables and configuration options in > apt.conf in place, apt-get is able to install packages offline, i.e. > when creating the rootfs, including the execution of postinst scripts >

Re: [OE-core] [PATCH 21/33] package_manager/deb: create Packages.xz

On 06.12.2016 18:22, Burton, Ross wrote: > > On 6 December 2016 at 11:49, Andreas Oberritter > wrote: > > +xz = bb.utils.which(os.getenv('PATH'), "xz") > > > Are the relevant dependencies present to ensure that xz-native has

Re: [OE-core] [PATCH 02/33] dpkg: update-alternatives-dpkg should conflict with other providers

On 06.12.2016 20:52, Burton, Ross wrote: > This needs more than just RCONFLICTS: > > ERROR: dpkg-native-1.18.7-r0 do_populate_sysroot: The recipe dpkg-native > is trying to install files into a shared area when those files already > exist. Those files and their manifest location are: > >

Re: [OE-core] [PATCH 28/33] dpkg: update packages and files to match Debian more closely

On 06.12.2016 21:13, Burton, Ross wrote: > > On 6 December 2016 at 11:49, Andreas Oberritter > wrote: > > +RRECOMMENDS_dpkg-perl = "gnupg gpgv" > > > With plain master this causes a build failure: > > ERROR: Multiple versions of gnupg

[OE-core] [PATCH 0/1] libarchive: fix ALTERNATIVE_PRIORITY to avoid conflict

The following changes since commit 11063a01d4511b2688ea7ba2d7359e4e07328c66: ruby: upgrade to 2.3.1 (2016-11-30 15:47:17 +) are available in the git repository at: git://git.openembedded.org/openembedded-core-contrib ChenQi/libarchive-priority

[OE-core] [PATCH 1/1] libarchive: fix ALTERNATIVE_PRIORITY to avoid conflict

'tar' utility from tar and bsdtar has the same alternative priority. 'cpio' utility from cpio and bsdcpio has the same alternative priority. Lower the ALTERNATIVE_PRIORITY to avoid conflict. Signed-off-by: Chen Qi --- meta/recipes-extended/libarchive/libarchive_3.2.2.bb

Re: [OE-core] [PATCH 3/4] kern-tools: fix processing for no branch meta-data

On Tue, 06 Dec 2016 22:14:41 +0100 Patrick Ohly wrote: > On Fri, 2016-12-02 at 16:09 -0500, Bruce Ashfield wrote: > > Lernel meta-data that has patches, but no branches, can trigger an > > error due to no branch specific patch queue. > > > > This error then cascades to

Re: [OE-core] [PATCH 21/33] package_manager/deb: create Packages.xz

On 06.12.2016 19:46, Burton, Ross wrote: > > On 6 December 2016 at 18:41, Andreas Oberritter > wrote: > > That's a good question. I guess I didn't add anything, because there's > no direct dependency in place for gzip-native either.

[OE-core] [PATCH 1/2] archiver.bbclass: fix do_ar_original error for matchbox-desktop

Error: ~~~ ERROR: matchbox-desktop-2.1-r0 do_ar_original: Can not determine archive names for original source because 'name' URL parameter is unset in more than one URL. Add it to at least one of these:

[OE-core] [PATCH 0/2] archiver.bbclass: fix some errors for do_ar_original and do_ar_recipe

The following changes since commit 9e63f81c78e284c9b325fe04a1b59e61c7ad8a1a: bitbake: ast: remove BBVERSIONS support (2016-11-30 15:48:10 +) are available in the git repository at: git://git.openembedded.org/openembedded-core-contrib ddk/fix-some-error-for-archiver-bbclass

[OE-core] [PATCH 2/2] archiver.bbclass: fix do_ar_recipe error for bonnie++ and libsigc++-2.0

When recipes name contains regular expression special characters, such as "++", in this case, the re.compile function in do_ar_recipe can't recognize it, so we should associate with re.escape to recognize the special characters in pattern. Signed-off-by: Dengke Du ---

Re: [OE-core] [PATCH 03/33] dpkg-native: hardcode SYSCONFDIR to /etc in update-alternatives

On 06.12.2016 17:12, Burton, Ross wrote: > > On 6 December 2016 at 16:08, Andreas Oberritter > wrote: > > I guess ${sysconfdir_native} should work, unless someone overrides > STAGING_ETCDIR_NATIVE. Would this be ok? > > > If the

Re: [OE-core] [PATCH 03/33] dpkg-native: hardcode SYSCONFDIR to /etc in update-alternatives

On 6 December 2016 at 16:37, Andreas Oberritter wrote: > I was asking because the result after stripping would essentially equal > ${sysconfdir_native}: > > STAGING_ETCDIR_NATIVE = "${STAGING_DIR_NATIVE}${sysconfdir_native}" > > It wouldn't work if someone set it to

Re: [OE-core] [PATCH 3/6] keyutils: new recipe (version 1.5.9)

On 6 December 2016 at 16:03, Andreas Oberritter wrote: > Apparently there's a recipe in meta-ivi which already has a fix. I'm > going to merge both recipes and resubmit. > Thanks Andreas, feel free to just send v2 for just this patch, as the rest are sitting in my staging

[OE-core] [PATCH v2] keyutils: new recipe (version 1.5.9)

Used by nfs-utils for nfsidmap, if available. Includes a backported patch for musl and a description text from meta-ivi. Signed-off-by: Andreas Oberritter --- .../0001-Include-limits.h-for-UINT_MAX.patch | 30 ++

Re: [OE-core] [PATCH 03/33] dpkg-native: hardcode SYSCONFDIR to /etc in update-alternatives

On 06.12.2016 16:39, Burton, Ross wrote: > > On 6 December 2016 at 11:49, Andreas Oberritter > wrote: > > +sed -i -e 's|SYSCONFDIR|"/etc"|' ${S}/utils/update-alternatives.c > > > Hardcoding /etc is bad, can't you just remove the

Re: [OE-core] [PATCH 21/33] package_manager/deb: create Packages.xz

On 6 December 2016 at 11:49, Andreas Oberritter wrote: > +xz = bb.utils.which(os.getenv('PATH'), "xz") > Are the relevant dependencies present to ensure that xz-native has been staged? Is it always present because apt links to a library in xz so the binaries get

Re: [OE-core] [PATCH 20/33] package_manager/deb: let apt-get handle postinst scripts

Can you elaborate on why the old code can be deleted? Ross On 6 December 2016 at 11:49, Andreas Oberritter wrote: > Signed-off-by: Andreas Oberritter > --- > meta/lib/oe/package_manager.py | 86 -- > >

Re: [OE-core] [PATCH 03/33] dpkg-native: hardcode SYSCONFDIR to /etc in update-alternatives

On 6 December 2016 at 16:08, Andreas Oberritter wrote: > I guess ${sysconfdir_native} should work, unless someone overrides > STAGING_ETCDIR_NATIVE. Would this be ok? > If the recipe inherits native then the class reassigns: sysconfdir = "${STAGING_ETCDIR_NATIVE}" So

[OE-core] [PATCH] sysstat: fixup pkg_postinst to allow SYSTEMD_AUTO_ENABLE to work

The logic added to the pkg_postinst in commit 6bf82c26f953 has the side effect of rendering SYSTEMD_AUTO_ENABLE ineffective. The systemd service will not be configured as 'enabled' either offline(do_rootfs) or during first boot. Since the volatiles, as used, in the pkg_postinst are unused with

Re: [OE-core] [PATCH 02/33] dpkg: update-alternatives-dpkg should conflict with other providers

This needs more than just RCONFLICTS: ERROR: dpkg-native-1.18.7-r0 do_populate_sysroot: The recipe dpkg-native is trying to install files into a shared area when those files already exist. Those files and their manifest location are:

Re: [OE-core] [PATCH 0/4] kernel-yocto: consolidated pull request

On Fri 2016-12-02 @ 04:09:21 PM, Bruce Ashfield wrote: > This pull request is mainly to fix a couple of bugs that were reported > on the mailing list recently, but it also includes some kernel version > updates that I *think* I sent previously. Thanks Bruce, I've tested these and they look good.

[OE-core] [PATCH v2] run-postinsts: Print message before running deferred postinst scripts

Package managers can defer running postinst scripts to first boot, which can take a while on some systems. E.g. The output of `opkg configure` (or whatever pm is used) is redirected to a file when logging is enabled ($POSTINST_LOGGING == 1), making the machine appear hung during this process. This

Re: [OE-core] oetest.py construction of pkgmanifest is probably wrong

Hm, I see this whole thing is being re-written. Maybe that will fix it. Joe From: Burton, Ross [mailto:ross.bur...@intel.com] Sent: Tuesday, December 06, 2016 1:52 AM To: Slater, Joseph Cc: openembedded-core@lists.openembedded.org Subject: Re: [OE-core] oetest.py construction of

Re: [OE-core] [PATCH 3/5] libsdl2: add EXTRA_OECONF[vardepsexclude] = "MACHINE"

On Tue, Dec 6, 2016 at 1:24 AM, Khem Raj wrote: > There probably is no need to use this additional variable. We can just use > PKG_CONFIG_SYSROOT Problem is not finding the wrong .pc files. Problem is that the output of pkg-config must be prefixed to find the protocols.

[OE-core] [PATCH V2 5/6] runqemu: fixes for slirp, network device and hostfwd

Fixed: - Add QB_NETWORK_DEVICE to set network device, it will be used by both slirp and tap. - Set QB_NETWORK_DEVICE to "-device virtio-net-pci" in qemuboot.bbclass but runqemu will default to "-device e1000" when QB_NETWORK_DEVICE is not set, this is because oe-core's qemu targets support

[OE-core] [PATCH V2 2/6] qemuboot.bbclass: use IMGDEPLOYDIR

So that "bitbake -ccleansstate" can remove qemuboot.conf Signed-off-by: Robert Yang --- meta/classes/qemuboot.bbclass | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/meta/classes/qemuboot.bbclass b/meta/classes/qemuboot.bbclass index

[OE-core] [PATCH V2 0/6] runqemu: fix for slirp, network device and hostfwd

* V2 - Add QB_NETWORK_DEVICE to set network device for both slirp and tap, the idea is from Randy and Nathan. - Use different mac sections for slirp and tap to fix conflicts when running both of them on the same host. * V1 - Initial version The following changes since commit

Re: [OE-core] [PATCH] gdb-cross-canadian: Depend on nativesdk-python3-misc

On 6 December 2016 at 00:16, George McCollister < george.mccollis...@gmail.com> wrote: > importlib is the replacement for imp (it's deprecated since version > 3.4) so this sounds reasonable to me. Shall I send a patch to add > imp.py to python3-importlib and another to add python3-importlib to >

Re: [OE-core] oetest.py construction of pkgmanifest is probably wrong

On 6 December 2016 at 02:24, Slater, Joseph wrote: > This variable is a dictionary with entries based on package, but for > multillib a package might exist in only some variants. Perhaps > > each entry should be a dictionary of multilib variants? > > Yeah, probably.

Re: [OE-core] [PATCH 2/2] gstreamer1.0-vaapi: Import from meta-intel

Hi, On Fri, Dec 2, 2016 at 9:32 PM, Khem Raj wrote: > Update to 1.10.1 at the same time > This suggests the versions should be kept in sync with the gstreamer versions: http://git.yoctoproject.org/cgit/cgit.cgi/meta-intel/commit/?id=3f51f61efe93c104ba7996f54f381c6c1a5e6546

[OE-core] [PATCH V2 6/6] qemuboot.bbclass: add blank lines in comments

Add blank lines in comments to make it easy for readind and updating. Signed-off-by: Robert Yang --- meta/classes/qemuboot.bbclass | 18 ++ 1 file changed, 18 insertions(+) diff --git a/meta/classes/qemuboot.bbclass b/meta/classes/qemuboot.bbclass

[OE-core] [PATCH V2 1/6] scripts/runqemu: fix checking for .cpio.gz

When "runqemu /path/to/.cpio.gz", it used the last suffix "gz" as the fstype which was wrong. Check filename against self.fstypes firstly can fix the problem. Signed-off-by: Robert Yang --- scripts/runqemu | 22 -- 1 file changed, 16 insertions(+),

[OE-core] [PATCH V2 4/6] runqemu: support mutiple qemus running when nfs

Fixed: * In build1: $ runqemu nfs qemux86-64 In build2: $ runqemu nfs qemux86-64 It would fail before since the port numerbs and conf files are conflicted, now make runqemu-export-rootfs work together with runqemu to fix the problem. * And we don't need export PSEUDO_LOCALSTATEDIR in

[OE-core] [PATCH V2 3/6] runqemu-export-rootfs: fix inconsistent var names

Fixed: $ runqemu nfs qemux86-64 [snip] On your target please remember to add the following options for NFS nfsroot=IP_ADDRESS:/path/to/nfsroot,nfsvers=3,port=,mountprog=,nfsprog=,udp,mountport= [snip] Note that the values are null, this is because their var names are inconsistent. [YOCTO #10519]

Re: [OE-core] [PATCH V2 0/6] runqemu: fix for slirp, network device and hostfwd

On 12/06/2016 04:55 PM, Robert Yang wrote: * V2 - Add QB_NETWORK_DEVICE to set network device for both slirp and tap, the idea is from Randy and Nathan. Add Randy and Nathan in the loop, I had added them in git send-email, but there were not in the CC list, look strange. // Robert

[OE-core] [PATCH] liburi-perl: update to 1.71

Old version was broken by perl update. It was found by investigating configure errors in meta-qt5-extra/kf5: | [superandy@mueller-a-nb-linux 5.28.0-r0]$ /home/superandy/tmp/oe-core-glibc/sysroots/x86_64-linux/usr/bin/perl-native/perl -e "use URI::Escape" | \C no longer supported in regex;

Re: [OE-core] [meta-intel] [PATCH 2/2] gstreamer1.0-vaapi: Import from meta-intel

On 6 December 2016 at 09:57, Ylinen, Mikko wrote: > This suggests the versions should be kept in sync with the gstreamer > versions: > http://git.yoctoproject.org/cgit/cgit.cgi/meta-intel/commit/?id= > 3f51f61efe93c104ba7996f54f381c6c1a5e6546 > Khem sent a GStreamer

[OE-core] [PATCH 08/33] apt: merge apt-package.inc into .bb

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/apt/apt-package.inc | 78 -- meta/recipes-devtools/apt/apt_1.2.12.bb | 79 ++- 2 files changed, 78 insertions(+), 79 deletions(-) delete mode 100644

[OE-core] [PATCH 21/33] package_manager/deb: create Packages.xz

apt-get prefers it over Packages.gz. Signed-off-by: Andreas Oberritter --- meta/lib/oe/package_manager.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/meta/lib/oe/package_manager.py b/meta/lib/oe/package_manager.py index 12dff20..b18148b 100644 ---

[OE-core] [PATCH 15/33] apt: merge with apt-native

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/apt/apt-native_1.2.12.bb | 50 -- meta/recipes-devtools/apt/apt.inc | 3 ++ meta/recipes-devtools/apt/apt_1.2.12.bb| 11 ++ 3 files changed, 14 insertions(+), 50

[OE-core] [PATCH 10/33] apt: use default packaging rules for manpages

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/apt/apt_1.2.12.bb | 35 + 1 file changed, 1 insertion(+), 34 deletions(-) diff --git a/meta/recipes-devtools/apt/apt_1.2.12.bb b/meta/recipes-devtools/apt/apt_1.2.12.bb index

[OE-core] [PATCH 14/33] apt-native: drop obsolete patches

0001-Revert-always-run-dpkg-configure-a-at-the-end-of-our.patch and noconfigure.patch are obsolete due to dpkg's new --force-script-chrootless option. db_linking_hack.patch and environment.patch are marked as backports from 2007. no-curl.patch: curl-native is already a prerequisite.

[OE-core] [PATCH 13/33] apt: rewrite do_install

Install missing files and directories (apt, apt-ftparchive, apt-helper, apt-mark, trusted.gpg.d). Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/apt/apt_1.2.12.bb | 73 ++--- 1 file changed, 41 insertions(+), 32 deletions(-) diff

[OE-core] [PATCH 12/33] apt: use default packaging rules for -dev package

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/apt/apt_1.2.12.bb | 1 - 1 file changed, 1 deletion(-) diff --git a/meta/recipes-devtools/apt/apt_1.2.12.bb b/meta/recipes-devtools/apt/apt_1.2.12.bb index ec93eb8..c531a49 100644 ---

[OE-core] [PATCH 11/33] apt: use default packaging rules for base package, prepend apt-utils

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/apt/apt_1.2.12.bb | 10 ++ 1 file changed, 2 insertions(+), 8 deletions(-) diff --git a/meta/recipes-devtools/apt/apt_1.2.12.bb b/meta/recipes-devtools/apt/apt_1.2.12.bb index ba49571..ec93eb8 100644 ---

[OE-core] [PATCH 32/33] apt: fix rpath error during configure

| checking for shared library run path origin... /bin/sh: ../apt-1.0.10.1/buildlib/config.rpath: No such file or directory Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/apt/apt_1.2.12.bb | 3 +++ 1 file changed, 3 insertions(+) diff --git

[OE-core] [PATCH 25/33] dpkg: clean-up packaging of start-stop-daemon

Dpkg-start-stop is not an intuitive name. Just call the package start-stop-daemon. Don't rdepend on it, because it might be empty. Install start-stop-daemon to /sbin to match Debian and busybox. Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/dpkg/dpkg.inc |

[OE-core] [PATCH 27/33] dpkg: simplify replacing perl interpreter in scripts

Target scripts already contain the correct value. Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/dpkg/dpkg.inc | 10 -- 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/meta/recipes-devtools/dpkg/dpkg.inc

[OE-core] [PATCH 24/33] dpkg: set license field to more common "GPLv2+"

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/dpkg/dpkg.inc | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/meta/recipes-devtools/dpkg/dpkg.inc b/meta/recipes-devtools/dpkg/dpkg.inc index 5b5ea01..2eabc30 100644 ---

[OE-core] [PATCH 22/33] package_manager/deb: implement BAD_RECOMMENDATIONS

Signed-off-by: Andreas Oberritter --- meta/classes/rootfs_deb.bbclass | 6 -- meta/lib/oe/package_manager.py | 27 +++ meta/lib/oe/rootfs.py | 2 ++ 3 files changed, 29 insertions(+), 6 deletions(-) diff --git

[OE-core] [PATCH 29/33] dpkg: always install dpkg-configure.service

systemd.bbclass handles everything based on DISTRO_FEATURES. Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/dpkg/dpkg.inc | 22 -- 1 file changed, 8 insertions(+), 14 deletions(-) diff --git a/meta/recipes-devtools/dpkg/dpkg.inc

[OE-core] [PATCH 30/33] dpkg: use correct path for perl libraries

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/dpkg/dpkg.inc | 6 ++ 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/meta/recipes-devtools/dpkg/dpkg.inc b/meta/recipes-devtools/dpkg/dpkg.inc index 414ee5b..1c404fc 100644 ---

[OE-core] [PATCH 20/33] package_manager/deb: let apt-get handle postinst scripts

Signed-off-by: Andreas Oberritter --- meta/lib/oe/package_manager.py | 86 -- meta/lib/oe/rootfs.py | 6 +-- 2 files changed, 9 insertions(+), 83 deletions(-) diff --git a/meta/lib/oe/package_manager.py

[OE-core] [PATCH 23/33] dpkg: use PACKAGECONFIG

Keep default dependencies as before. Set default compressor to xz if selected, gzip otherwise. Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/dpkg/dpkg.inc | 31 ++- 1 file changed, 18 insertions(+), 13 deletions(-) diff --git

[OE-core] [PATCH 17/33] apt: group variables (license, configuration, packaging)

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/apt/apt_1.2.12.bb | 32 +--- 1 file changed, 17 insertions(+), 15 deletions(-) diff --git a/meta/recipes-devtools/apt/apt_1.2.12.bb b/meta/recipes-devtools/apt/apt_1.2.12.bb index

[OE-core] [PATCH 16/33] apt: merge apt.inc into .bb

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/apt/apt.inc | 38 meta/recipes-devtools/apt/apt_1.2.12.bb | 39 - 2 files changed, 38 insertions(+), 39 deletions(-) delete mode 100644

[OE-core] [PATCH 18/33] apt: more fine-grained packaging

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/apt/apt_1.2.12.bb | 9 +++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/meta/recipes-devtools/apt/apt_1.2.12.bb b/meta/recipes-devtools/apt/apt_1.2.12.bb index a350b36..2c4f11d 100644 ---

[OE-core] [PATCH 33/33] apt: dselect depends on perl

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/apt/apt_1.2.12.bb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/meta/recipes-devtools/apt/apt_1.2.12.bb b/meta/recipes-devtools/apt/apt_1.2.12.bb index 5dc50a4..0b640f9 100644 ---

[OE-core] [PATCH 31/33] package_deb.bbclass: compress control.tar with the same algorithm as data.tar

Yields better compression with xz and fixes a problem invoking gzip. Signed-off-by: Andreas Oberritter --- meta/classes/package_deb.bbclass | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/meta/classes/package_deb.bbclass

[OE-core] [PATCH 19/33] apt: recommend gnupg

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/apt/apt_1.2.12.bb | 3 +++ 1 file changed, 3 insertions(+) diff --git a/meta/recipes-devtools/apt/apt_1.2.12.bb b/meta/recipes-devtools/apt/apt_1.2.12.bb index 2c4f11d..4d781a2 100644 ---

[OE-core] [PATCH 26/33] dpkg: remove unneeded do_configure override

Exporting PERL_LIBDIR is enough, compiler.m4 isn't needed. Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/dpkg/dpkg.inc | 6 -- 1 file changed, 6 deletions(-) diff --git a/meta/recipes-devtools/dpkg/dpkg.inc b/meta/recipes-devtools/dpkg/dpkg.inc index

[OE-core] [PATCH 28/33] dpkg: update packages and files to match Debian more closely

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/dpkg/dpkg.inc | 53 + 1 file changed, 48 insertions(+), 5 deletions(-) diff --git a/meta/recipes-devtools/dpkg/dpkg.inc b/meta/recipes-devtools/dpkg/dpkg.inc index

Re: [OE-core] [meta-intel] [PATCH 2/2] gstreamer1.0-vaapi: Import from meta-intel

Hi, On Tue, Dec 6, 2016 at 12:48 PM, Burton, Ross wrote: > > On 6 December 2016 at 09:57, Ylinen, Mikko wrote: > >> This suggests the versions should be kept in sync with the gstreamer >> versions: >>

[OE-core] [PATCH 05/33] package_manager/deb: pass option --force-script-chrootless to dpkg

Fixes removal of packages with new dpkg version. Signed-off-by: Andreas Oberritter --- meta/lib/oe/package_manager.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/meta/lib/oe/package_manager.py b/meta/lib/oe/package_manager.py index 7ba2e4d..4ef4f6d

[OE-core] [PATCH 04/33] apt-native: Use option --force-script-chrootless for rootfs

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/apt/files/apt.conf | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/meta/recipes-devtools/apt/files/apt.conf b/meta/recipes-devtools/apt/files/apt.conf index 0335135..9ad61cc 100644 ---

[OE-core] [PATCH 07/33] apt-native: merge apt-native.inc into .bb

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/apt/apt-native.inc | 70 meta/recipes-devtools/apt/apt-native_1.2.12.bb | 74 -- 2 files changed, 71 insertions(+), 73 deletions(-) delete mode 100644

[OE-core] [PATCH 03/33] dpkg-native: hardcode SYSCONFDIR to /etc in update-alternatives

It needs to be relative to DPKG_INSTDIR when installing. Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/dpkg/dpkg.inc | 4 1 file changed, 4 insertions(+) diff --git a/meta/recipes-devtools/dpkg/dpkg.inc b/meta/recipes-devtools/dpkg/dpkg.inc index

[OE-core] [PATCH 00/33] Accumulated patches for deb packaging

These are most of my patches which accumulated since our distro switched from opkg to dpkg and apt two years ago. They were tested thoroughly on dora and krogoth and just a little bit on master. Most importantly, this adds support for Debian's update-alternatives on rootfs creation time. For this

[OE-core] [PATCH 02/33] dpkg: update-alternatives-dpkg should conflict with other providers

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/dpkg/dpkg.inc | 1 + 1 file changed, 1 insertion(+) diff --git a/meta/recipes-devtools/dpkg/dpkg.inc b/meta/recipes-devtools/dpkg/dpkg.inc index ec0117b..f7d9e77 100644 --- a/meta/recipes-devtools/dpkg/dpkg.inc

[OE-core] [PATCH 01/33] dpkg: implement offline mode for update-alternatives

Signed-off-by: Andreas Oberritter --- meta/recipes-devtools/dpkg/dpkg.inc| 12 +- ...tinsts-expect-D-to-be-set-when-running-in.patch | 70 ...pdate-alternatives-Implement-offline-mode.patch | 399 +

  1   2   >