<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux/poky.git/bitbake/lib/bb/runqueue.py, branch 1.2_M4.rc3.1</title>
<subtitle>Mirror of git.yoctoproject.org/poky</subtitle>
<id>https://git.enea.com/cgit/linux/poky.git/atom?h=1.2_M4.rc3.1</id>
<link rel='self' href='https://git.enea.com/cgit/linux/poky.git/atom?h=1.2_M4.rc3.1'/>
<link rel='alternate' type='text/html' href='https://git.enea.com/cgit/linux/poky.git/'/>
<updated>2012-03-30T16:20:21+00:00</updated>
<entry>
<title>runqueue: wait and deal with those &lt;defunct&gt; sub-processes as soon as they are os.killed() when "Force stop"</title>
<updated>2012-03-30T16:20:21+00:00</updated>
<author>
<name>Shane Wang</name>
<email>shane.wang@intel.com</email>
</author>
<published>2012-03-30T09:12:41+00:00</published>
<link rel='alternate' type='text/html' href='https://git.enea.com/cgit/linux/poky.git/commit/?id=984c23f7d7d18464a060aa66be7c02abb8266a08'/>
<id>urn:sha1:984c23f7d7d18464a060aa66be7c02abb8266a08</id>
<content type='text'>
When "Force stop" is performed during the build stage, after os.kill() kills the build sub-processes, there are many &lt;defunct&gt; python processes in the system. In Hob, when the user initiates a new build, os.waitpid() in runqueue_process_waitpid() will be called, and the pids of those &lt;defunct&gt; processes will be returned as result[0], then self.build_pids[result[0]] will throw KeyError exception because now for the new build self.build_pids is empty.

This patch is to address the above issue to collect the results and handle the sub-processes as soon as they are killed.

[Yocto #2186]

(Bitbake rev: e9f4ca467e795bbc520d12b0e7a5985b6ff0a20e)

Signed-off-by: Shane Wang &lt;shane.wang@intel.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</content>
</entry>
<entry>
<title>Typo: "nonexistant" -&gt; "non-existent"</title>
<updated>2012-03-20T15:21:32+00:00</updated>
<author>
<name>Robert P. J. Day</name>
<email>rpjday@crashcourse.ca</email>
</author>
<published>2012-03-19T13:58:14+00:00</published>
<link rel='alternate' type='text/html' href='https://git.enea.com/cgit/linux/poky.git/commit/?id=8b206d38c7fb172022e98a0879963e5f44bc2092'/>
<id>urn:sha1:8b206d38c7fb172022e98a0879963e5f44bc2092</id>
<content type='text'>
(Bitbake rev: c4c8df0e1e77f99f3ee0b07e1748116fc362819f)

Signed-off-by: Robert P. J. Day &lt;rpjday@crashcourse.ca&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</content>
</entry>
<entry>
<title>runqueue.py: When checking whether stamps are valid for setscene, recurse</title>
<updated>2012-03-16T10:51:17+00:00</updated>
<author>
<name>Richard Purdie</name>
<email>richard.purdie@linuxfoundation.org</email>
</author>
<published>2012-03-16T10:46:05+00:00</published>
<link rel='alternate' type='text/html' href='https://git.enea.com/cgit/linux/poky.git/commit/?id=4ca7075b0972163d95e3c3f5a40c602bf3bb0245'/>
<id>urn:sha1:4ca7075b0972163d95e3c3f5a40c602bf3bb0245</id>
<content type='text'>
Currently the code checking whether stamps are valid only traverses one step
of the dependency graph. This works fine in the normal cases where we've already
validated dependencies but for the setscene code, it doesn't work well. A typical
problem usecase is something like:

bitbake gcc-cross -c unpack -f
bitbake gcc-cross

which will ignore any sstate files already cached which could be used to speed
up the gcc-cross build. This becomes particularly problematic with multiple gcc
builds where only one should rebuild yet they all do.

This patch teaches the stamp code to be able to recurse within a given fn
which gives the behaviour people would expect from the code and allows
bitbake to make better use of prebuild sstate objects.

(Bitbake rev: e68814cb2e8da523d4ddf05e8ceddfaa19244851)

Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</content>
</entry>
<entry>
<title>meta/classes: Convert to use appendVar and appendVarFlags</title>
<updated>2012-03-05T18:36:53+00:00</updated>
<author>
<name>Richard Purdie</name>
<email>richard.purdie@linuxfoundation.org</email>
</author>
<published>2012-03-03T10:41:41+00:00</published>
<link rel='alternate' type='text/html' href='https://git.enea.com/cgit/linux/poky.git/commit/?id=ff73b02a7203623e5edfaaa10df17eaa63cf4ed3'/>
<id>urn:sha1:ff73b02a7203623e5edfaaa10df17eaa63cf4ed3</id>
<content type='text'>
(From OE-Core rev: 3b57de68e70e77dbc03c0616a83a29a2e99e40b4)

Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</content>
</entry>
<entry>
<title>bitbake/runqueue: fix python error on scenequeue task failure</title>
<updated>2012-03-01T15:51:36+00:00</updated>
<author>
<name>Paul Eggleton</name>
<email>paul.eggleton@linux.intel.com</email>
</author>
<published>2012-03-01T14:57:35+00:00</published>
<link rel='alternate' type='text/html' href='https://git.enea.com/cgit/linux/poky.git/commit/?id=8078553d7be408af9327eafc3d5e68108fad08de'/>
<id>urn:sha1:8078553d7be408af9327eafc3d5e68108fad08de</id>
<content type='text'>
Fixes a regression introduced in commit
e8a3499c95a6d4f2b8fed002fb9504733c5be3c6 which resulted in a
backtrace on setscene task failure due to trying to dereference
the setscene task ID twice.

(Bitbake rev: 8b846a92a58b5c20d7cfd2efd32b763e95c3c2fd)

Signed-off-by: Paul Eggleton &lt;paul.eggleton@linux.intel.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</content>
</entry>
<entry>
<title>runqueue.py: initialize rqexe at RunQueue's init function</title>
<updated>2012-03-01T15:51:36+00:00</updated>
<author>
<name>Dongxiao Xu</name>
<email>dongxiao.xu@intel.com</email>
</author>
<published>2012-02-29T14:15:28+00:00</published>
<link rel='alternate' type='text/html' href='https://git.enea.com/cgit/linux/poky.git/commit/?id=ba8538b3a9ecdd8d77eecd277b913c9cf96facd1'/>
<id>urn:sha1:ba8538b3a9ecdd8d77eecd277b913c9cf96facd1</id>
<content type='text'>
Sometimes user stops the build before runqueue is established,
for example, at the stage of running add_unresolved() function. This
will cause RunQueue to use rqexe field in finish_runqeue() before
initialized. This will cause endless print of "Running idle function"
if use process server.

This commit initialize rqexe variable in RunQueue's init function,
and add a judgement in finish_runqueue().

(Bitbake rev: 59f817723172092a87738c79f555e605f55ea375)

Signed-off-by: Dongxiao Xu &lt;dongxiao.xu@intel.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</content>
</entry>
<entry>
<title>bitbake: fix setscene task start/failure handling</title>
<updated>2012-02-28T12:27:22+00:00</updated>
<author>
<name>Paul Eggleton</name>
<email>paul.eggleton@linux.intel.com</email>
</author>
<published>2012-02-27T18:54:11+00:00</published>
<link rel='alternate' type='text/html' href='https://git.enea.com/cgit/linux/poky.git/commit/?id=ce0ecf5d847b168ad05117cdd1f5fdcc561d770d'/>
<id>urn:sha1:ce0ecf5d847b168ad05117cdd1f5fdcc561d770d</id>
<content type='text'>
* When a setscene task starts, print out that it's starting in the UI
  (ensuring we get the correct task name)
* When a setscene task fails, ensure we remove it from the list of
  running tasks so that if you break out any time afterwards it
  is not still listed.

(Bitbake rev: e8a3499c95a6d4f2b8fed002fb9504733c5be3c6)

Signed-off-by: Paul Eggleton &lt;paul.eggleton@linux.intel.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</content>
</entry>
<entry>
<title>bitbake/runqueue: make dry-run do everything except execute</title>
<updated>2012-02-28T12:27:21+00:00</updated>
<author>
<name>Paul Eggleton</name>
<email>paul.eggleton@linux.intel.com</email>
</author>
<published>2012-02-27T18:54:12+00:00</published>
<link rel='alternate' type='text/html' href='https://git.enea.com/cgit/linux/poky.git/commit/?id=e7952c188374508bcf65903d85ae27a593de7465'/>
<id>urn:sha1:e7952c188374508bcf65903d85ae27a593de7465</id>
<content type='text'>
Make dry-run do everything except executing the task, instead of
cutting it off earlier. This fully tests the code path for running the
child task (parsing and fakeroot), as well as enabling future
functionality such as using dry-run to produce signature files.

(Bitbake rev: bf1d7739618dabf3872a868230c0112b9ad2a2c0)

Signed-off-by: Paul Eggleton &lt;paul.eggleton@linux.intel.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</content>
</entry>
<entry>
<title>V5 Disk space monitoring</title>
<updated>2012-02-26T11:05:27+00:00</updated>
<author>
<name>Robert Yang</name>
<email>liezhi.yang@windriver.com</email>
</author>
<published>2012-02-26T08:48:15+00:00</published>
<link rel='alternate' type='text/html' href='https://git.enea.com/cgit/linux/poky.git/commit/?id=9674ea28ffebd75f117531238e517df41ed7dafb'/>
<id>urn:sha1:9674ea28ffebd75f117531238e517df41ed7dafb</id>
<content type='text'>
Monitor disk availability and take action when the free disk space or
amount of free inode is running low, it is enabled when BB_DISKMON_DIRS
is set.

* Variable meanings(from meta-yocto/conf/local.conf.sample):

  # Set the directories to monitor for disk usage, if more than one
  # directories are mounted in the same device, then only one directory
  # would be monitored since the monitor is based on the device.
  # The format is:
  # "action,directory,minimum_space,minimum_free_inode"
  #
  # The "action" must be set and should be one of:
  # ABORT: Immediately abort
  # STOPTASKS: The new tasks can't be executed any more, will stop the build
  #           when the running tasks have been done.
  # WARN: show warnings (see BB_DISKMON_WARNINTERVAL for more information)
  #
  # The "directory" must be set, any directory is OK.
  #
  # Either "minimum_space" or "minimum_free_inode" (or both of them)
  # should be set, otherwise the monitor would not be enabled,
  # the unit can be G, M, K or none, but do NOT use GB, MB or KB
  # (B is not needed).
  #BB_DISKMON_DIRS = "STOPTASKS,${TMPDIR},1G,100K WARN,${SSTATE_DIR},1G,100K"
  #
  # Set disk space and inode interval (only works when the action is "WARN",
  # the unit can be G, M, or K, but do NOT use the GB, MB or KB
  # (B is not needed), the format is:
  # "disk_space_interval, disk_inode_interval",  the default value is
  # "50M,5K" which means that it would warn when the free space is
  # lower than the minimum space(or inode), and would repeat the action
  # when the disk space reduces 50M (or the amount of inode reduces 5k)
  # again.
  #BB_DISKMON_WARNINTERVAL = "50M,5K"

[YOCTO #1589]
(Bitbake rev: 4d173d441d2beb8e6492b6b1842682f8cf32e6cc)

Signed-off-by: Robert Yang &lt;liezhi.yang@windriver.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</content>
</entry>
<entry>
<title>bitbake/runqueue: Ensure finish_now() sets the runqueue state consistently</title>
<updated>2012-02-26T11:04:38+00:00</updated>
<author>
<name>Richard Purdie</name>
<email>richard.purdie@linuxfoundation.org</email>
</author>
<published>2012-02-25T16:02:29+00:00</published>
<link rel='alternate' type='text/html' href='https://git.enea.com/cgit/linux/poky.git/commit/?id=aac45679fb17c720f511a30261d34eb64e6a4c70'/>
<id>urn:sha1:aac45679fb17c720f511a30261d34eb64e6a4c70</id>
<content type='text'>
If we call finish_now(True), rq.state is not updated to match. This
makes the behaviour of finish_now(False) and finish_now(True) consistent
so both leave rq.state consistently.

(Bitbake rev: 9079ae0ab74f9232b7e9853b2013b051d4fcf623)

Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</content>
</entry>
</feed>
