| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
| |
Currently this module is used by GENIVI and AGL
Change-Id: I9b1024d3b9546de418e5d235e3bd845c8402db09
|
|
|
|
| |
Change-Id: I56ac7f4a9ac4e2559a31cb66e0233a0c0707e1a6
|
|
|
|
| |
Change-Id: I73024dda7e13e64988aa909c1fa8cf89b12ec310
|
|
|
|
|
|
| |
And use new submodules funtionality
Change-Id: Iea2820fbb14aab182cd20fea8c9040fe1a03cc96
|
|
|
|
|
|
| |
This reverts commit aae6c313a831b8a5d876b89e351ef1fbe3b7741d.
Change-Id: I09bb426f2d8199d3edc45a4d707ff718b602e945
|
|
|
|
|
|
| |
Currently this module is used by GENIVI and AGL
Change-Id: I3a867e64cd650088342256dcda3974f04d5ef110
|
|
|
|
| |
Change-Id: I7be27624f25ba5d96cf032c21b9c53bd1a017f4e
|
|
|
|
|
|
|
|
|
|
|
|
| |
This commit partially reverts 19f8bfce95741de8ada632aab1177f8f844ebf0c.
Turns out we weren't fixing the .gitmodules, but creating them to fix
build issues.
These issues seem to be fixed in the master branch, but there aren't new
releases with the fixes in. Another solution would be to investigate if we
could build from master directly.
Change-Id: Ib234e02e2c9460442b1429dc3a03aff86361f19d
|
|
|
|
| |
Change-Id: I0374a55cdbc62f78115de14fbd12132383fd7bb3
|
|
|
|
| |
Change-Id: Id85a2412958b902f9081dce4cb78963271d41156
|
|
|
|
|
|
|
|
| |
Upgrades
- coreutils to 8.25
- findutils to 4.6.0
Change-Id: If63d96b6d34cbc1f69d67de80880ab30f3c25e11
|
|
|
|
| |
Change-Id: I44e125835bdf7c8168585808407f931ab0f8d12b
|
|
|
|
| |
Change-Id: Ib7a8e10f81c50b1d3a86cd9992269403a348ba7f
|
|
|
|
| |
Change-Id: Icca1708bd241e051b5529cc641bdbfb268c14bb2
|
|
|
|
| |
Change-Id: I58460cf6d0a52adccb3a89c9604bc9f1795e3068
|
|
|
|
|
|
|
| |
Some components still use 'baserock/*' branches because they weren't
using a branch just for fixing .gitmodules file
Change-Id: I89d0edd3d7fb0bea51107cc21f9e348dd6ddc140
|
|
|
|
| |
Change-Id: Ie6c11252bab86971ccc8e6217297c2a2b211b222
|
|
|
|
| |
Change-Id: I9e9c9cce8766e2778e8de9b11da9dce086ab9a50
|
|
|
|
|
|
|
| |
Don't upgrade for now, we might want to test that our Ansible scripts
still work with a newer version first.
Change-Id: Iee9107ba07957ae06fe95e835cd20c376d7c1337
|
|
|
|
| |
Change-Id: I68c12f9425bc2a3f0f6478d2a434bbc2b590d6d9
|
|
|
|
|
|
|
| |
Its a common dependency between gnome and genivi strata, also
a optional one for the ostree stratum
Change-Id: I19f5df84e1f10319d3a7268a9da729545d52fd55
|
|
|
|
| |
Change-Id: I46ddfbcdda64097cc816a2ce1da64651b4f9e392
|
|
|
|
| |
Change-Id: Ib6af59cfd4c4294ba33aa2ea22ac5101dd1f8d90
|
|
|
|
| |
Change-Id: I30162f7ec4561d2cead1ba53271f96825fcfb49c
|
|
|
|
|
|
|
| |
This is the new Qt web runtime, which replace the now deprecated
QtWebkit
Change-Id: If6a873b218ac4f8779f48e7b706b4e773c523b8f
|
|
|
|
| |
Change-Id: I566daebbd32a8924fa160f6b01eb18db0b73ca0d
|
|
|
|
|
|
|
| |
Note: Migration scripts live now in:
http://git.baserock.org/cgit/baserock/baserock/spec.git
Change-Id: Ica4fd12d96e563e5195172318d2d65847266146f
|
|
|
|
|
|
|
|
|
|
|
| |
Also, switch to using upstream v2.4.6 tag of GNU Libtool.
This is easy to do once definitions V8 is merged -- no more
.gitmodules patching!
Tested by building a minimal-system-x86_64
Change-Id: If2f01507d364245736727699a33225ce929655dc
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This removes support for definitions V6, and adds support for
definitions V8.
The last release of the Baserock reference system images was 15.47.1,
which used definitions V6.
Full list of changes:
a7f1247 Add support for definitions version 8
9404317 Remove support for definitions version 6
575c40c Adapt integration test to definitions version 7
0ee859a morphloader: simplify API
d58d8e8 Unify local and remote repo cache modules
014a029 Get rid of the CachedRepo class (almost)
e8a67a7 Hide get_repo() and cache_repo() functions, always use
get_updated_repo()
bb139ec Fix some test failures
Change-Id: Ied8aa209ab8eede2a160c4273f93c15b7ff41793
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The gnulib bootstrap script is idiotic, when it does get_version(),
it looks for a string v${major}.${minor}.${micro}... it starts this
by taking the last 'v' on the line and removing everything else.
On an arm arch, instead of getting 5.22.0 from the (v5.22.0) string
output from perl --version... it gets 5tejl-linux from the armv5tejl-linux
string on the same line... get_version() thus thinks we have "perl 5" and
breaks.
This is not a problem for armv7lhf builds, because the bootstrap script
conveniently interprets perl's version as perl version 7, which is newer
than the required 5.6.
Change-Id: I1445cfd84efd37c1828eae09def88b27f18af250
|
|
|
|
| |
Change-Id: Ie273f3ba1e9f3f665c9dbf64741f7c9427ddfdb9
|
|
|
|
| |
Change-Id: I86c485bb991e262c7272b2ca2e6140a3ffb7220d
|
|
|
|
|
|
|
|
| |
This patch has already been merged in master in commit
96adac823500be172424e94b5094c7d97e68175d of wayland-ivi-extension,
but it's not included in any release yet.
Change-Id: Ia7e281e9f68cf35ade618c3566f56fd659e769f3
|
|
|
|
|
|
|
|
|
| |
systemd uses libmount from util-linux now, it doesn't try to run the
system's `mount` program directly any more. So we no longer need to
hack it into working with Busybox's `mount`.
Change-Id: I798635cba1407f96cdd66ecc36bd5958364edda1
|
|
|
|
|
|
|
| |
Which includes a fix to a build break when using libical 2.0
in commit 70a2e0e1119eb042af7d3963fad6b1f8e5cdc9ee
Change-Id: Ied397dcbc94ad533af5c1f7629f154f0c19e5a5b
|
|
|
|
| |
Change-Id: I5cb2b62dc60defad54574da9050ca6aa220b2412
|
|
|
|
|
|
|
|
| |
Before we were only specifically casing armv7lhf, use the same
configuration for arm* instead, this causes the mesa build to
pass on armv5l
Change-Id: I9dda3c771270254e1c2c44857b6275af2cad3de2
|
|
|
|
|
|
|
| |
- xz is enabled by default
- Use the default install directories, ie, /usr/lib and /usr/bin
Change-Id: I038113fcbf0288703fbb06cc4b64b34d09d8fc72
|
|
|
|
|
|
|
| |
This doesnt seem to be needed: the service file is configured correctly to
use '/bin/sh' even without using this workaround
Change-Id: I8cb9ff567c5e1b3f75a3101a958f5421ed87d571
|
|
|
|
|
|
|
| |
So the location of the service files is not harcoded in
/lib/systemd/system
Change-Id: Ifa771678b11391a5759f370805b8b23b043042e4
|
|
|
|
|
|
|
|
| |
Cherry pick the upstream fix to use linux/types.h to define types (previously
fixed in Baserock as 0479b766923a01553366d8dbf197ac5dabf4b9d6). The upstream
patch also fixes on platforms other than Linux.
Change-Id: I09cfd055092b24f9259da4179a3d6862f032007a
|
|
|
|
| |
Change-Id: I8b6f699df0100a745d278d4b6cb3ab17566a113c
|
|
|
|
| |
Change-Id: Ie577c50e3e62c543bb48ab4cd72c8d2f0cfe8ce5
|
|
|
|
| |
Change-Id: Ifa0b08168ed9910a7c405f1fee076fc7d51cc205
|
|
|
|
|
|
|
|
|
|
| |
This stratum is included in systems like Trove, upgrading Git here is
needed too.
This commit completes 23f354034df7c6d2652bca285047d29f5abef560, which
only upgrades 'git-minimal' in strata/core.morph
Change-Id: I9e29190f9ef2fa2dbadc894853dca7a601695094
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This contains commit 9831e92bfa833ee9c0ce464bbc2f941ae6c2698d which
removes the path_name() function. That fixes a remote-code execution
security hole, described in CVE-2016-2315 and CVE-2016-2324.
I have read in some places that 2.7.1 and later are not vulnerable,
but I've not been able to prove that, nor find proof. At time of writing
the Debian advisory doesn't show that 2.7.1 and later are safe, only
2.8.0-rc2:
https://security-tracker.debian.org/tracker/CVE-2016-2324
See also:
https://ma.ttias.be/remote-code-execution-git-versions-client-server-2-7-1-cve-2016-2324-cve-2016-2315/
Change-Id: I8948b295030f2f498780777aa62a54f2337518b5
|
|
|
|
|
|
|
| |
Also remove max-jobs restriction as per the comments in commit
ef6b19c4fade2dda8b7f6e7beb73cef6c60eb3b3
Change-Id: I17e121bb9244aba0218559801c60a81a161de1cd
|
|
|
|
|
|
|
|
|
|
|
|
| |
Their build system is broken in version 1.0.1, and this shouldn't slow
down the build too much given that this chunk takes only around 40
seconds to build on a 2vCPU 2G RAM VM.
It looks like version 2.0.0 includes fixes to this problem. See commit
dcc40c7ae2d337a0d83c077bf9d8f283499c6717 in libical repo, and consider
removing this restriction when upgrading.
Change-Id: Ic55f13944094336addf9a54805cac67a6f3b9e26
|
|
|
|
|
|
|
|
|
|
|
|
| |
sandboxlib is a runtime dependency of ybd. While ybd is not currently a
part of Baserock I found it surprising I couldn't at least run ybd from git
using the build image from wiki.baserock.org.
Created a new stratum as I was unable to find a suitable home
amongst the existing strata.
Also moves linux-user-chroot into the new stratum.
Change-Id: If55c0ba4b9a48a0abc69ad5d39900c6c7807c72b
|
|
|
|
|
|
|
| |
Nothing should depend on them anymore
(As a reference, Arch systems dont include them anymore)
Change-Id: I0b74bff2968b388bbd797ca626ffa2ac65a09fff
|