[YOCTO #1761]
Reconnect to the backend Sqlite DB in 'database is locked' exception so
the timeout can be leveraged in each time retry.
(Bitbake rev: b310382764367b573c84f33d847c6eb821266f9e)
Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Joshua Lock <josh@linux.intel.com>
This is being added to resolve setscene race issues where we do have
particular dependencies required between setscene tasks. This allows
specific dependencies to be specified. This allows us to fix the races
in sstate with the useradd class in OE-Core.
Any tasks being depended upon have their reverse dependencies cleared to
ensure we don't have circular references.
(Bitbake rev: e1b157d26374a70e6274edcb4c0b9f3bc48f765c)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Joshua Lock <josh@linux.intel.com>
When constructing the setscene inter-dependencies, we need to account for all task,
not just the last one found. This patch corrects this oversight and ensures all
dependencies are added, not just the first one found.
(Bitbake rev: b9b5b5129d066e1ff7d3effda116afc3c6657beb)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Joshua Lock <josh@linux.intel.com>
To quote my report of this to the git mailing list:
"""
I have a problem with git clone commands using alternates failing by
mixing up different repositories. I have a situation where I could end
up with both:
/srv/mirrors/repo
/srv/mirrors/repo.git
as bare clones.
I then try cloning "repo" with alternates with the command:
$ git clone -s -n /srv/mirrors/repo /tmp/foo
Cloning into /tmp/foo...
done.
$ cat /tmp/foo/.git/objects/info/alternates
/srv/mirrors/repo.git/objects
Note how I'm now referencing repo.git, not repo. This doesn't work as
expected giving some very bizarre results when actually using the
repository.
I appreciate this is a rather bizarre corner case but its one that is
breaking the build system I work with. Ideally people would use a
consistent URL for the same repository but we have an example where they
haven't and this really shouldn't break like this.
Looking at the code, the cause seems to be
clone.c:get_repo_path():
static char *suffix[] = { "/.git", ".git", "" };
since its looking in order for:
repo/.git (fails)
repo.git (suceeds, incorrect)
repo (never looked at)
I'm not sure what would break if that order were to change, swapping the
last two options.
I can "force" the issue by running:
git clone -s -n /srv/mirrors/repo/ /tmp/foo
but this results in the slightly odd looking:
$ cat /tmp/foo/.git/objects/info/alternates
/srv/mirrors/repo//objects
which does at least work.
"""
This patch adds the trailing slash to ensure the correct repository is
referenced at the expense of some ugliness in the alternates file.
(Bitbake rev: d978e7b35550e3785c7c567ffe4c40a3c3947450)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
As reported on the mailing list[1] when changing IMAGE_FSTYPES through the
hob preferences a reparse is required before the changes will be picked up
by the system. This patch sets the reload_required property of the class to
true when the image types have been modified to ensure the reparse is
triggered.
1. https://lists.yoctoproject.org/pipermail/yocto/2011-December/006002.html
(Bitbake rev: 6c0babf08909307ab69a66ed06e77e8818b2a8c5)
Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The knotty UI just ignores these and so should RunningBuild, if these events
aren't handled the UI appears to hang.
Fixes [YOCTO #1665]
(Bitbake rev: 540ba78075bd525776aa23bf38bee66350c66534)
Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Some BB_HASHBASE_WHITELIST variables are in the lists of variable
dependencies for signatures. Ignore those differences in lists
since this difference does not matter
(Bitbake rev: 71b53a3f0766ca464560a1f6a449f9424fbdf7ae)
Signed-off-by: Matthew McClintock <msm@freescale.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Since bitbake switched back to the fork instead of the exec model,
it no longer used bitbake-runtask and the code has suffered some bitrot.
bitbake-runtask is a useful tool for excuting the task without
the scheduler of bitbake, so that the external tool can invoke it
easily. It also provides a useful example of how to invoke exec_task()
with low overhead without a lot of the bitbake threading/UI overhead.
Significant changes:
* This patch changes the argument order so that the commonly used
and mandatory arguments come first.
* The taskhash file and dryrun options are now optional
* It now uses the bitbake logging mechanisms to provide processed
logging output to the console.
* The process handling to do with stdout/stderr redirection
are removed since they're no longer required.
[YOCTO #1229]
RP: Logging updates to the patch based on Roberts original patch
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This allows the code to safely fall back to dumping the basehash data
if the taskhash data isn't present for some reason. We could effecitvely
obsolete the runtime option and use this approach instead exclusively.
(Bitbake rev: 5ace320ccc01f4e326f90b7ba060dcbff3380dca)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
There is a subtle but nasty problem that a function's cwd can vary
depending on whether ${B} (often ${S}) exists before the funciton is
called or not. Most functions in the system can cope with this but
its bad practise and I've just witnessed build failures resulting
from this during image generation from bootimg.bbclass. I also
suspect this could explain some odd fetcher behaviour witnessed in
the past.
This change ensures we always call funcitons with a specific build
directory making things deterministic.
(Bitbake rev: ef0888f83fa4408eb768257d7e03700202faad18)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Using bitbake -e -b virtual:xxxx:/path/to/the.bb would result in
zero matches since the virtual:xxxx piece wasn't being processed.
This adds in the necessary functionality to handle it correctly.
[YOCTO #1793]
(Bitbake rev: bd5a727c8447bcb747c1d2463b7de2ab6d21a7de)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
USER is the correct variable to use, also this can affect sstate
cache as well.
(Bitbake rev: d7f9edda65dae2e046871afa275c5a51dff48fc4)
Signed-off-by: Matthew McClintock <msm@freescale.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If the vardeps flag is not None, we now silence the warnings about
non-literal usage for that variable.
(Bitbake rev: e724b9f417d1baf898f5afc6376c73c1a2ad8db9)
Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
There are two usual cases involving bb.data.expand:
- Calling it with a string literal -- "bb.data.expand('${FOO}/${BAZ}/bleh', d)".
- Calling it on getVar results (legacy) -- "bb.data.expand(bb.data.getVar('FOO', d), d)"
Nothing in any of the usual layers uses it in any other way, and I'm
having trouble coming up with any real use cases beyond this. The first
of the above cases is already tracked, via the expandWithRefs called
on the python code string. The second didn't emit a warning anyway,
since the getVar was already handled.
Given this, I see no reason for us to maintain explicit expansion
tracking. Further, we weren't using its results anyway (the var_expands
member).
(Bitbake rev: 405dfe69e6a608826e599ebf2f83ef8cf5083b96)
Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
- If a name is passed to the parser, prepend the messages with "while
parsing <name>:". This gives a bit more context.
- Tweak the warning messages slightly (they had to be altered anyway to
inject the variable being parsed).
Before:
DEBUG: Warning: in call to 'bb.data.getVar': argument ''%s' % var' is \
not a literal
After:
DEBUG: while parsing emit_pkgdata, in call of bb.data.getVar, argument \
''%s' % var' is not a string literal
(Bitbake rev: 1060193ae4d54e667735dbff5d1d2be49a3f95c9)
Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
With the previous method, using the compare_name methods, we split the
requested match name by '.', reversed it, then compared them piecemeal
during the node traversal. The new method walks the nodes and hands back
the name of what's being called, and then we check that. This also
consolidates the two different implementations of traversal of the
attribute/name nodes (one in compare_name, one for the execs).
(Bitbake rev: 84e535b5165c7e936c5b1486bdf4626ed3649f5f)
Signed-off-by: Christopher Larson <kergoth@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The split is even less necessary now that we use ast.walk rather than an
actual NodeVisitor subclass.
(Bitbake rev: d6c44fac184abae8395bfa7078f06675218aa534)
Signed-off-by: Christopher Larson <kergoth@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We have a problem if we want to inject specific information into the variable
dependency code. There are cases for example where we want a dependency
on the value of X but it doesn't matter how X was constructed or what
dependencies it might have had, we only care about the absolute value.
With the current code, its near enough impossible to do this.
This patch adds such a mechanism so the user can trigger this with code like:
baselib[vardepvalue] = "${baselib}"
It also refactors some of the code so we do variable lookups once
instead of doing this in two different functions.
[YOCTO #1583]
(Bitbake rev: 6c879b44ccf42dc73fe4467076e114700d7ba81b)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Currently if you specify a mirror like:
file://.* http://linux.freescale.net/yocto/sstate-cache
it won't work as you expect whilst:
file://.* http://linux.freescale.net/yocto/sstate-cache/
will since it has the trailing slash.
This patch handles both cases correctly. It also adds some debug to
the uri_replace function since its near impossible to debug it without
some kind of output.
[YOCTO #1578]
(Bitbake rev: a0246bf09c93bb657eaf6ba61d090b247ed33640)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Defaulting to any file in DL_DIR as the first match for a file:// url
doesn't make much sense and can lead to unexpected results.
This patch changes the logic so this is the last fallback location
instead. Whether it should be using DL_DIR at all for this is a
good question but something for another patch.
[YOCTO #1710]
(Bitbake rev: 5597a68fac0954c682b67471722c2643e2415f99)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Without this patch the tash hash dependencies can be in a order
that is dependent upon directory/filesystem layout. With this
change the data is sorted the same regardless.
Without this the dependent hashes could be in different orders
on different systems and consequently final md5 hash would differ
as well even though nothing else changed.
(Bitbake rev: 9a2029899c946ce9aa8adbc85f2cfe7a85b92182)
Signed-off-by: Matthew McClintock <msm@freescale.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This is useful for really longs lists to pinpoint what has
actually changed
(Bitbake rev: f1eb6d3dcc10c42bb09383a87bde3afa69bc6ed9)
Signed-off-by: Matthew McClintock <msm@freescale.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The existing looping code can mask an existing "found = True"
by forcing it to False each time. This can lead to dependency
lists not being fully searched and results in dependency errors.
An exmaple of this was the autobuilder building linux-yocto from
sstate but then rebuilding some of the recipe's tasks for no
apparent reason. Separating the logic into two variables solves this
problem since any "found = True" value is now always preserved.
(Bitbake rev: 61017fc5d30b7a13308d038872ec92efc1a84cef)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If A depends upon B which depends upon C and the setscene for B
succeeds but C is going to get rebuilt, we should wait for C to
try and build A but currently we don't.
This is due to the timing of when we run the task_skip() as this
triggers other tasks to become buildable. This patch moves the timing
of that call to a more appropriate place allowing dependencies to
behave as expected.
(Bitbake rev: b7114d8e5d9b0720339bd5d24d243c0f2a7c1f3b)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Its possible for relative paths to creep into FILE. These confuse the
build system no end as its not clear where they might be releative to.
This patch ensures we always use resolved absolute paths for FILE
so that things behave in a deterministic way.
(Bitbake rev: 658d7daa70e46c2b20973b90ee53f0bbadc8bf5d)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ensure that the list of dependencies is included in the hash
as well as their contents
Prior to this, adding or removing dependencies with values
of "None" would not change the hash, despite diffsigs reporting
this difference.
(Bitbake rev: 727ca945177ce9bd44515cf611e3e95a09466d98)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When comparing sig files, if the recipe locations had changed, the
dependent tasks list would show as changed even if the actual hash
had not changed. This updates the code to only compare the base part
of the pathnames.
It also tweaks some of the output to add newlines to aid comparing
two lists of variables as it makes the location of the difference
clearer.
(Bitbake rev: 165a22ddcc647b945707fb5c483146bb336d5f66)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Previously, it was calling var_expands.update() rather than add(), with
a string argument, resulting in adding each character of that string to
the var_expands set, rather than the string itself.
(Bitbake rev: 8e4e75383e43d6da2c16ec5286186a0d0569b0f8)
Signed-off-by: Christopher Larson <kergoth@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The code which populates setscene_covered list was adding a task to the
covered list if all of the tasks that depend upon it were also covered;
however, this means that tasks that would have installed "runtime"
dependencies were being marked as covered also, e.g. gmp-native and
mpfr-native are needed by gcc-cross at runtime since they are shared
libraries that gcc links to, but their do_populate_sysroot tasks were
being marked as covered, resulting in failures later on if gcc-cross was
available from sstate but mpfr-native and gmp-native weren't.
Since we currently have no real way to handle runtime dependencies for
native packages, add a workaround which avoids marking tasks as covered
if one or more of their revdeps are from a different recipe.
Fixes [YOCTO #1536].
(Bitbake rev: e492eb4dc9016cd0bed194377c6f2b85cf0ad113)
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The lockfile retry parameter is expected to return immediately after
attempting to take the lock. There was a bug in the logic which this
patch fixed to ensure it does that.
(Bitbake rev: f421ef819f00ac659504d9af41bcc8323422ff8c)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Clearly a logic/indentation error - we should only try and load the recipe
should the file-chooser return OK.
Fixes [YOCTO #1668]
(Bitbake rev: db59297aa1861614ffaea4295b9b054baa8a12b9)
Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
git could need these environment variables when working behind
a proxy
(Bitbake rev: dca46cc2e1c75b6add2c4801e2994a4812745f5b)
Signed-off-by: Matthew McClintock <msm@freescale.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
git fetches can fail (or at least return failed) when trying to
fetch and prune rebased branches. This patch simply adds a -f
to the git fetch command so these failure are ignore
Generally, if some SHA was rebased away it's not coming back so
there is no point in not doing this force
(Bitbake rev: a7b75e4db52445b30ec0fc0053dcf454f5f7d2db)
Signed-off-by: Matthew McClintock <msm@freescale.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The debug and verbosity levels (as set by the -D and -v command line
options respectively) were not being passed through within msg.py since
bitbake revision 45aad2f9647df14bcfa5e755b57e1ddab377939a due to
incorrect variable names.
Fixes [YOCTO #1513].
(Bitbake rev: c6e88b7c0e61f9586a275df53f48b90687c5f92f)
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This fixes the internal dirtiness tracking such that if the Save menu item
is selected after loading a recipe the existing file is updated rather than
the user being prompted for the path to create a recipe at.
(Bitbake rev: 00fc1d7249b5e217cc7c36ac71b63ddad1c5b769)
Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
After the reparse we were setting the model to reflect the values before
the reparse was triggered but clearing the internal variables used to test
whether these values are set, leading to the UI erroneously reporting that
selections had not been made.
(Bitbake rev: 656eafe0f2c9ec7730d33e15705b8c720f787c49)
Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Instead of calling getVariable commands each time the BBPATH and BBFILES
entries need testing cache the results as a member variable at object
instantiation.
Fixes [YOCTO #1521]
(Bitbake rev: 109e1597671dfb7222672e268190aabc727960ca)
Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The _build_revision method in Hg class gets called with the wrong number
of arguments. This tiny patch adds a 5th argument to the method
declaration to prevent python from throwing an exception.
(Bitbake rev: 623e9c7f7a9cf12b8c81c26cc608990682a601dd)
Signed-off-by: Julian Pidancet <julian.pidancet@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Firstly, rather than polling the DEPLOY_DIR_IMAGE directory each time an
image is built store the variable as a member of the hobeventhandler.
Secondly emit the generic "build-complete" signal *after* the specialised
"build-failed" or "build-succeeded" signals such that the appropriate
state variables are set before we try and use them.
(Bitbake rev: 043914a8b478fd4a7799acd1b44bdb3b0af2165a)
Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Pass the correct callback name to the connect method.
(Bitbake rev: d2b9167a7b0eb3fcb31d64f915beef41aac659a5)
Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Once an image build has been performed the selected_image variable would
prevent the package only build choice being triggered - by setting the
variable to None on reset we are able to perform package only builds after
an image build.
(Bitbake rev: 705156e2812afb288632f01416fcbbf9add26bee)
Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The different code paths for package vs image builds have lead to some bit
rot in the package only build and an incompatability between package only
builds and the build again mechanism.
This patch unifies the code paths and fixes build again for package only
builds.
Fixes [YOCTO #1480]
(Bitbake rev: ab6c63f35739c30df0e8a9d0f1ae982b96e538ed)
Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When disabling the toolchain w/ headers checkbox in an early commit I
accidentally disabled too much code. This re-enables adding the custom
response handler to the Preferences dialog which triggeres a reparse of
the metadata when appropriate preference changes are made.
Fixes [YOCTO #1479]
(Bitbake rev: bdbd724e806c9ca69cc1508bac7ec0a2dce95543)
Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Make the git fetcher's subpath (path within the git repo to fetch)
option set the destsuffix (destination directory) option by default.
This reverts the behaviour of subpath to the same as when it was
introduced.
Based on a patch by Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
(Bitbake rev: 3e7f8afeacf7c8c8de3e87778a3907e33d4a06b3)
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* in some cases there could be output like this
ERROR: ld.so: object 'libpseudo.so' from LD_PRELOAD cannot be preloaded: ignored.
before wc -l output and returned 'output.split()[0] != 0' is always True
(Bitbake rev: 3a54dcc09a12406ec6cf22b4b1a2cc4fc203822c)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* logging/logger typo was fixed in 38a598731b49c8a0ba0ede570adc33eb1e848235
but debug level is still missing
(Bitbake rev: 9de432fe2348cdbc93037bb49abb508d1fd38336)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Lianhao Lu
Richard Purdie
Joshua Lock
Matthew McClintock
Robert Yang
Christopher Larson
Christopher Larson
Paul Eggleton
Julian Pidancet
Martin Jansa