| Age | Commit message (Collapse) | Author |
|---|
|
If SkipParse is raised from something which isn't anonymous python, it wasn't
being handled correctly. This improves the handling for example from within inline
python.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The method this older code was accessing doesn't exist. A library shouldn't
be changing behaviour based on the environment anyway when sepcific function
calls exist to correctly setup the logging. Fix the crash if BBDEBUG was used
by moving the code to bin/bitbake.
[YOCTO #3319]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Executing "bitbake" doesn't get a sane message since the None return value
wasn't being handled correctly. Also fix msg -> cmd_action['msg'] as
otherwise an invalid variable is accessed which then crashes the server
due to the previous bug.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Without this, if an exception occurs the server will silently crash
with no feedback to the user about why (since traceback isn't imported).
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The inherits cache contains duplicate entries, some with the full patch, some
just starting classes/xxx. This is a waste of parse time and potentially
error prone. This patch fixes various pieces of code so the absolute paths are
always preferred and work correctly. The inherits_class function did not work
with full paths so the patch fixes this.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The current usage of EXPORT_FUNCTIONS is rather problematic since a class
list (classes) is passed into the ast statement and cached as it was
when first parsed. This class list may be different in other cases but
is locked once in the cache.
Worse, the construction of classes can be broken by exceptions during parsing
at the wrong moments since the state of the parser is not always reset
correctly. This can lead to leakage of other classes into the classes list.
The current EXPORT_FUNCTIONS implementation looks at the last two currently
inherited classes and sets up an indirect function call view the second last
class inherited, e.g.:
do_configure calls gnomebase_do_configure
gnomebase_do_configure calls autotools_do_configure
This intermediary doesn't seem to serve a useful purpose.
This patch therefore makes builds deterministic and fixes various cache problems
and indirection by removing the intermediaries and simply performing
directly mapping for the cases where its needed.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
parsing"
This reverts commit 3dc5e896a97154914cee6c47900e3bb2a2627cdb.
|
|
The classes variable in particular could be in an odd state if parsing of a
previous recipe had abruptly ended.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
I appreciate a fixed value is suboptimal but this is better than not
working at all!
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
There is a race where the registerEventHandler() server call may happen in
a different thread to the main UI, causing the main UI to see
httplib.CannotSendRequest exceptions.
This moves the call into the main thread, removing the init race and fixing
errors I was seeing with the xmlrpc backend.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
An issue was uncovered where changing:
IMAGE_INSTALL_append = "X"
to
IMAGE_INSTALL_append = "X Y"
in local.conf would not get noticed by bitbake. The issue is that
the configuration hash doesn't account for overrides or key expansion.
This patch improves get_hash to account for these. This means the hash
does account for changes like the above.
[YOCTO #3503]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Leaving the sigdata files around can aid debugging and doesn't harm
anything. This is the easiest way to allow this to happen.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The "*" part of the mask is to ensure we clean both any stamp, and any
setscene varient. It turns out we would also trample other tasks,
e.g. do_package_write could trample do_package_write_rpm. do_package also
tramples do_package_write_* but this is less of an issue since the other
tasks depend on it.
Rather than use the wildcard, we can just use a list instead.
[YOCTO #3484]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
wget commands for check and resume were
mixed-up, leading to the following issues:
1. long running "NOTE: Preparing runqueue"
reason: objects were downloaded, not spidered on the mirror
2. Failing network test in Build Appliance, because wget 1.14
(in use in BA) will fail if a file already exists.
During the network connectivity test, index.php file was
actually downloaded, not spidered (checked for existence on
yoctoproject.org website), leading to wget failure.
Signed-off-by: Cristian Iorga <cristian.iorga@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The last changes introduced an error in some of the logic. Add brackets
to clarify the meaning of the expression and fix certain build failures.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
- there are some layers which cannot be removed; so ,I have used
a new variable called BBLAYERS_NON_REMOVABLE located in bblayers.conf,
which contains those layers which cannot be deleted
- "meta-hob" layer is added to this variable in hob code, like it's
added to BBLAYERS var
[YOCTO #3176]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Currently there is code which uses FILESDIR in unpack to ensure
parent directories are created, leading to differing behaviour depending on
which search path is used to locate the directory.
This change standardises the code and takes the data from the fetcher in
question meaning we can standardise the code and deprecate FILESDIR.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Currently wildcard matches end up working by FILESDIR being defined
in the metadata to a default of "." in FILESPATH which is hacky at best.
This patch adds the behaviour into the fetcher so its at least slightly
more explicit.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Using curses initscr/endwin causes screen corruption if for example you
suspend bitbake and resume it. This changes the code to use a less
invasive approach to determining colour availability on the terminal.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Run the sanity check only once
[YOCTO #3377]
Signed-off-by: Constantin Musca <constantinx.musca@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
-to enable this in hob, first you have to enable this in bitbake using
BB_DISKMON_DIRS and/or BB_DISKMON_WARNINTERVAL vars
-created "restart the build" action on the build_fail_top_bar
[YOCTO #3357]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
In normal usage, we never hit the timeout issue. If we do, it becomes obvious
that the current error handling is not good enough. The request may have made it
to the server and the answer will get queued. This means the next command may get
the return value from the previous command with suitably puzzling results.
Without rewriting large sections of code, its not possible to avoid this problem.
It is better to increase the timeout to several seconds giving the server a chance
to respond and if it does timeout, hard exit since recovery is not possible with the
code base today.
I'd be happy to see the structure of this code improved but this quick fix at least
stops corrupted builds from happening which has to be a good thing.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Add bold color output to log level name and standard color output to log msg
when bitbake is run from an iteractive console. Color output is only
enabled
if the terminal supports color.
Used Jason Wessel's recommendation for transparency on verbose, note and
plain.
Signed-off-by: Seth Bollinger <seth.boll@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Any issues encountered during the build (fatal or not) is displayed
in the Issues tab, and the total number of issues is changed.
[YOCTO #3376]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
When the setscene code was originally written it was thought that we'd
allow "partial" coverage. For example, if we just want to build the target
"bash:do_populate_sysroot" and its available from sstate, it makes no sense
to install gcc-cross's sstate package as its simply not needed.
Due to various other issues in the codebase, this functionality was
disabled/removed to allow the setscene code and sstate to stabilise and allow
us to concentrate on other problems.
The time has now come to enable "partial" coverage. There are two major changes
in this patch:
a) Creation of an unskippable list. This lists direct dependencies of
build targets and hence things that cannot be skipped.
b) Addition of a handler which looks at a given setscene target and what depends
on it and then decides whether its necessary.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The data returned by get_file_depends() may me used in contexts like
checksums where order is important. The current usage of sets means
that some of the checksums can change in circumstances they should not.
This patch changes to use lists, thereby removing the problem.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The duplicated import could result in an UnboundLocalError.
Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
"bitbake -e ASSUME_PROVIDED" should fail, but the error message wasn't
clear enough in the past:
$ bitbake -e bzip2-native
[snip]
ERROR: Command execution failed: Traceback (most recent call last):
File "/buildarea/lyang1/poky/bitbake/lib/bb/command.py", line 94, in
runAsyncCommand
commandmethod(self.cmds_async, self, options)
File "/buildarea/lyang1/poky/bitbake/lib/bb/command.py", line 323, in
showEnvironmentTarget
command.cooker.showEnvironment(None, pkg)
File "/buildarea/lyang1/poky/bitbake/lib/bb/cooker.py", line 325, in
showEnvironment
fnid = taskdata.build_targets[targetid][0]
KeyError: 0
[snip]
With this patch, the massage will be:
[snip]
ERROR: bzip2-native is in ASSUME_PROVIDED
ERROR: Command execution failed: Exited with 1
Summary: There were 2 ERROR messages shown, returning a non-zero exit code.
[YOCTO #3392]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
-after pressing "build image" button, hob was freezing because it didn't
receive well the log file
[YOCTO #3398]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Currently, command.py can return an error message from runCommand, due to
being unable to run the command, yet few of our UIs (just hob) can handle it
today. This can result in seeing a TypeError with traceback in certain rare
circumstances.
To resolve this, we need a clean way to get errors back from runCommand,
without having to isinstance() the return value. This implements such a thing
by making runCommand also return an error (or None if no error occurred).
As runCommand now has a method of returning errors, we can also alter the
getCmdLineAction bits such that the returned value is just the action, not an
additional message. If a sync command wants to return an error, it raises
CommandError(message), and the message will be passed to the caller
appropriately.
Example Usage:
result, error = server.runCommand(...)
if error:
log.error('Unable to run command: %s' % error)
return 1
Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
-small changes to the text of the progress bar when parsing recipes
[YOCTO #3282]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This code is dead and doesn't do anything so lets remove it.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Dumping the environment data into the console log files directory is
invariably not what the user wants or expects and leads to confusion
when looking at the log directory.
This change forces the logs to be disabled by default when using
the -e option.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Its a minor correctness detail but the update flag should be set
when Start events are encountered.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
There was an issue where the last tasks run by bitbake would not
correctly get displayed in the knotty footer. This was due to the
total count including active tasks. This change ensures the footer
is displayed if the are any running tasks.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
-since the order of the layers can potentially impact
the build outcome, users should be able to reorder
the layers within the layers dialog;
-used TreeView Drag and Drop
[YOCTO #3270]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
User introduced proxy settings were not saved
when a proxy details input dialog was opened.
The proxy settings were lost upon return, and
restored from the previously stored one.
Also:
Code cleanup:
details_cb() function duplicate definition
removed
Fixes [YOCTO #3240]
Signed-off-by: Cristian Iorga <cristian.iorga@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
process state
- indicating on progress bar that hob has gone from parsing
recipes to "Generating dependency tree"; this will provide
some visibility of what has caused the "stop" button state
to change
[YOCTO #3282]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
- created a new file named "hobthreads.py", defining a thread
for opening the log file in a subprocess using subprocess module;
in the future I think we will add some other threads here, to
implement some other performance issues
- on "builddetailspage", "packageselectionpage" and "imagedetailspage"
I have changed the manner for opening the log file; it uses the thread
to open the file, and on main thread it creates a dialog to show a
progress bar, which pulses till the file is open
- this was added because when the log file is big, it takes time to
be opened; on the dialog you can use "Cancel" button to terminate the
process initiated to open the file
[YOCTO #2997]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Image selection is not properly initialized to none,
and it used before having a chance to have a value.
Due to dynamic nature of Python, variable is used before
it exists, in this case. This causes a crash.
Bug introduced during the fix of [YOCTO #3228]
Fixes [YOCTO #3334]
Signed-off-by: Cristian Iorga <cristian.iorga@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
- added a scroll bar in the shared state tab from settings
- added a signal on it, so when you add a new mirror, it will
auto-scroll to the end of the list
[YOCTO #3229]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Image selection is remembered correctly even after
advanced settings have been changed.
Selected image was reset even when it was not the case.
Fixes [YOCTO #3228]
Signed-off-by: Cristian Iorga <cristian.iorga@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
- when you do a build you should see "Log" page, not
"Build configuration" page
[YOCTO #2569]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
-when the recipe parsing process is stopped, the progress bar shows
"Stopping recipe parsing"
[YOCTO #3259]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
In "Edit Recipes" and "Edit packages" pages, the "Brought in by"
column is displayed correctly, with the right number of additional
packages and a proper title.
Fixes [YOCTO #2195].
Signed-off-by: Cristian Iorga <cristian.iorga@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
added a method in progressbar.py that doesn't update the
percentage in the progress bar shown in hob.
the call of the method is done in builder.py.
Signed-off-by: Andrei Dinu <andrei.adrianx.dinu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
-when returned to the Image configuration page, after canceling on the
Recipe selection page, the image selected previously is now shown corectly
[YOCTO #3205}
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
Richard Purdie
Cristiana Voicu
Cristian Iorga
Constantin Musca
Seth Bollinger
Christopher Larson
Robert Yang
Andrei Dinu