openembedded-core.git/meta/classes, branch master-next2

classes/image_live: allow different filesystems to be used for live images

2017-04-28T10:32:21+00:00

The idea is copied from commit df0b217f3df2c36a32e5c4afaec36a28bfc77bbb:
[ classes/image_vm: allow different filesystems to be used for VM images ]

The same logic should apply to image_live.bbclass, to allow other
filesystems to be used vs just ext4. The default value of ext4 is kept
so there is no functional change unless LIVE_ROOTFS_TYPE is set in the
inherting recipe.

Signed-off-by: Ming Liu 
Signed-off-by: Ross Burton

cmake.bbclass: use weakest ??= assignment for default OECMAKE_SOURCEPATH

2017-04-28T10:26:06+00:00

Make it slightly easier to support situations where the default path
needs to be over-ridden more than once.

Signed-off-by: Andre McCurdy 
Signed-off-by: Richard Purdie

cmake.bbclass: Do not use bitbake variable syntax for shell variables

2017-04-28T10:26:06+00:00

Using bitbake variable syntax (i.e., ${FOO}) for shell variables is
bad practice. First of all it is confusing, but more importantly it
can lead to weird problems if someone actually defines a bitbake
variable with the same name as the shell variable.

Also use lower case for local shell variables.

Signed-off-by: Peter Kjellerstedt 
Signed-off-by: Ross Burton

kernel-yocto/kern-tools: fix do_validate_branches clean stage

2017-04-28T10:26:05+00:00

It was reported that do_validate_branches was failing with the following
error:

  Log data follows:
  | DEBUG: Executing shell function do_validate_branches
  | HEAD is now at fe0fb8d Merge tag 'v4.10.9' into standard/base
  | mkdir: cannot create directory .: File exists
  |
  | [ERROR] Can't find patch dir at ./patches/standard/base
  | usage: kgit s2q
  | WARNING: exit code 1 from a shell command.
  | ERROR: Function failed: do_validate_branches

This was triggered by the execution of 'kgit-s2q --clean' after forcing
the SRCREV to something other than the tip of the branch. --clean is
being run to remove any sentinel files from previous kernel builds to
ensure that the tree is in a consistent state.

There were two bugs, --clean was being executed and not exiting the
script as it was supposed to. Hence validation for applying patches
was done, and threw the error that eventually makes it to the console.

And the second bug is that since do_validate_branches actually calls
kgit-s2q --clean, the dependency on kern-tools-native needs to be on
that function (versus do_kernel_metadata which runs later).

With the tweaked kern-tool + the dependency fix, we no longer see this
error.

Signed-off-by: Bruce Ashfield 
Signed-off-by: Richard Purdie

useradd-statids.bbclass: Add support for -P / --clear-password option

2017-04-28T08:16:02+00:00

The commit 31dee7946340bf0f1e94e4e714191d3d6ca3bf6a added a new useradd and
groupadd option to specify a clear text password.  The parsing logic in the
useradd-staticid class did not understand this new option.  If the
meta-skeleton examples were run with the class enabled an error would be
generated, as an example uses the -P option.

Note, the code has a check that we do not attempt to set both a crypt and
clear text password.  It is not allowed that these two options are set
at the same time, so we prefer the crypt option if they happen to be.

Signed-off-by: Mark Hatle 
Signed-off-by: Richard Purdie

testimage.bbclass: add additional dependency

2017-04-21T07:22:03+00:00

qemu-native-helper has an additional task that needs to be run in order
for testimage to work. This task is usually run by default in a full
build but there are use cases where it might be skipped. This commit
adds the dependency explicitly.

Also, this commit adds a try/catch error message to make it clearer what
you need to do if you try to run testimage before you have built or
downloaded the image artifacts.

[YOCTO #11375]

Signed-off-by: brian avery 
Signed-off-by: Ross Burton

qemuboot.bbclass: save relative paths in conf file

2017-04-20T06:55:20+00:00

This saves relative paths in the qemuboot.conf file instead of absolute
paths. This is to allow the images and kernels to be relocated and still
have the testimage and runqemu work.

[YOCTO #11375]

Signed-off-by: brian avery 
Signed-off-by: Ross Burton

rootfs-postcommands.bbclass: save relative paths

2017-04-20T06:55:10+00:00

We pass the TOPDIR to do a search/replace in export2json so that we save
relative paths in the testdata.json file rather than absolute paths.
This is to allow the images and kernels to be relocated yet still allow
testimage to work.

[YOCTO #11375]

Signed-off-by: brian avery 
Signed-off-by: Ross Burton

kernel-arch.bbclass: conditional error message

2017-04-20T06:54:27+00:00

The single purpose of "map_kernel_arch" is to set

   export ARCH = "some-arch"

The case when "some-arch" is not a valid Linux architecture results in an error.
This makes sense if the TARGET_OS is Linux, but that is not always the case.
kernel-arch is also inherited by toolchain-script, which may be used to build
toolchains for architectures not supported by Linux.

Rather than modifying toolchain-script to provide its own version of "map_arch"
this patch bypasses the error if the TARGET_OS is not linux.

Signed-off-by: Juro Bystricky 
Signed-off-by: Ross Burton

package_rpm.bbclass: Dosen't filter Conflicts if found in Provides

2017-04-20T06:54:10+00:00

This filter was add to make compilence with debian packaging but in
package_deb.bbclass is allowed to have the same values in Conflicts and
Provides.

With this filtering errors in recipe meta-data are hidden and could end
on install two packages that conflicts [2].

Reviewing the RPM spec from Fedora doesn't have anything that denies to
use the both Conflicts and Provides with the same value [3], also in
debian manual section 7.6.2 of [4] this behaviour is allowed to force
the removal of the conflicted package and RPM is compilence with this
behaviour after remove the filtering this is seen [5].

[1]
http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=4b611b66743a5ec220aef34d796af63029bb5fd9
[2] https://bugzilla.yoctoproject.org/show_bug.cgi?id=9349#c9
[3]
https://docs.fedoraproject.org/en-US/Fedora_Draft_Documentation/0.1/html/RPM_Guide/ch-advanced-packaging.html
[4] https://www.debian.org/doc/debian-policy/ch-relationships.html
[5] https://bugzilla.yoctoproject.org/show_bug.cgi?id=9349#c12

Signed-off-by: Aníbal Limón 
Signed-off-by: Ross Burton