StateSystem: Aesthetic changes to HT_IO.java

Make CACHE_SIZE have a pre-cached MASK.
Add comments.
Rename variables.
Remove "this" references.

Change-Id: Ib3b7c6421b614008f61312b6e22073d2c5c51d94
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63896
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>

Bug 483621: Move segment store statistics into timing analysis plug-ins

Also, o.e.t.analysis.timing.core.tests is introduced.

Keep packages internal for now since APIs not finalized.

Change-Id: I5a5d4abfd409ff8498d304f97f3c123e0b7a421b
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63891
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf: Make font depend on item height in time graph

The font used for drawing item names and state labels is now dependent
on each item's height.

The time graph combo no longer has to adjust the time graph control's
font when zooming vertically.

A map of fonts per height is used.

The item state rectangles are adjusted to use less margin height when
the item height is decreased.

Change-Id: I1838855d864e76646bc8e387682ca0cb3c28c418
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/62980
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

doc: Correct lttng session load description for conflicting profiles

Change-Id: Ic99b0e9503d00291072c5cac29feedd7fa3a1e13
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63760
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

Use pom-less plugins where we can

Tycho 0.24 introduced pom-less builds [1]. With this, we can
remove the pom.xml in any plugin, test or feature plugin that
uses the default configuration.

This bumps the minimum required version of Maven to 3.3.

Some plugins have extra specific configuration in their pom,
like ctf or pcap traces that call maven-antrun-plugin to
download test traces. These will still need their usual pom.xml
file.

For test plugins, the top-level pom defines the configuration
used for *core* tests. UI and SWTBot tests use slightly
different parameters. Instead of repeating this information in
every single plugin, we can define profiles that are only
activated in those plugins.

Using empty files to selectively enable these profiles looks a
bit clunky, but it is the simplest solution I have found.
Suggestions welcome!

[1] https://wiki.eclipse.org/Tycho/Release_Notes/0.24#POM-less_Tycho_builds

Change-Id: Ibd379905ca52cd1535b8f5a92e58e1f1fb311a19
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/63050
Reviewed-by: Hudson CI

doc: update user guide for kernel event filtering in Control view

Change-Id: I05da02b91b5b9bf1d73a4ae8c038f559d1744c87
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63648
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

control: add SWTBot test to test kernel filtering for events

Change-Id: I5de303ba182942b42cccb94db3995f0877cee5bd
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63592
Reviewed-by: Hudson CI
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

control: Support for filtering of kernel events

This feature is supported since LTTng Tools v2.7.0.

Change-Id: I1a17d6cc3807050bc15c8dd0eb6a763bbe4fb699
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63591
Reviewed-by: Hudson CI
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

analysis.test: add basic swtbot test for scatter charts

Change-Id: I35c19e240c66628a892646ae19ec6706324f2400
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63682
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

analysis.test: add basic tests for SWTChart in DensityViewerTest

Change-Id: I0efc9c1ebe7ef32fad7e4a455f75c64ce9e07447
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63680
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

analysis.test: add SWTBot test to Density view

Change-Id: I1f0932ebabb40ee15b45ed0b363b7883d069bcc6
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63596
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

TMF: Bug 454283: Initialize all analysis modules before executing them

Analysis modules are first initialized, then automatic ones are executed when
the traceOpened signal is received. This makes sure any dependent module
exists for a trace before the execution takes place.

Change-Id: Ica4d77ca5805b37ec73a42af2b9bc5f15c9e21cd
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Reviewed-on: https://git.eclipse.org/r/37698
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI

Remove API tooling from test plugins

We do not track the API of test plugins so it is not necessary to have
them configured for API tooling. This could lead to errors in the case
where the API baseline contained some test plugins. In Oomph, for
example, the API baseline is just a zip of the complete update site, so
it does contain the test plugins and this triggers errors.

Change-Id: I6bd735c2890ae5eaf329316d4f1bae43348c2281
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61839
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

doc: Add user guide for LTTng session profiles

Change-Id: I42f7de09aa21d66e9e137ee89f15d275f7d3b47f
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63665
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-by: Hudson CI

analsysis: Remove lingering alltest.java

Change-Id: I036ecf7c79a477cd75a149123af5d7e0a6fd1dbb
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63593
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Reviewed-by: Hudson CI

analysis: fix Loose Coupling and int division to double conversion

AbstractSegmentStoreDensityViewer is fixed and made more portable.

Change-Id: Ic7ca60aefb308eaa4a00c914cb7dff6b274c4bb7
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63646
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

analysis: fix modifier order for AbstractSegmemtStoreDensityView

Change-Id: Ibd1dc5a4a55f1063c5fda0624597cab06be49ed4
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63651
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf : add of a range selection from other sources for the tmfEventsTable

Only the events within the range will be selected. If there is no event
within the range of the selection, we select the next event outside of
the range. For a single selection, the event at or the next event after
the requested timestamp is selected.

Change-Id: I1294da7e5d653477502ca45f9430a5e5c8583187
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/57526
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

analysis: simplify perSyscallStats iteration

Change-Id: Ibac78c3912ac647ec90add64f2a2f16ce90539ef
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63496
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

Latency: introduce latency density view

This patch introduces a view to view segment densities. The
view is based on SWTChart's bar graph. Its x axis is NOT
time aligned as it is not on the main timeline. It takes
the liberty to display time in a more "human readable"
form showing a scale and up to 3 decimals max. (1.234 ms)

There is a second part to this view: a selection table.
This table shows the selected segments from a drag operation.

Follow up tasks for this can be to introduce other filters and
improve the tooltips.

An example implementation is provided for the SystemCall analysis.

Change-Id: I238a61dc8eeeefdca0da107424686a64fb63d8c3
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/57573
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

Some cleanups related to investigating redundant null checks

Change-Id: I6c71c90c2d61fe4c4c3d30c8b1fb1cb0c9eac2f2
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61197
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf: Avoid recursion in AbstractSegmentStoreTableViewer pack listener

In the SWT.SetData listener TableColumn.pack() is called. This requires
all visible items from being filled in order to pack the column. For
each visible item a SWT.SetData event is fired which triggers another
TableColumn.pack().

In Windows, after 12 such iterations in the stack, the table somehow
behaves badly and the OS incorrectly returns an item count of zero,
which causes ArrayIndexOutOfBoundsException and other problems.

Since the intention is to pack the columns only once, the SWT.SetData
listener is removed before the pack instead of after the pack, thus
avoiding the recursive calls.

Change-Id: Id44c54a474ed6e04aa2d79c051d863011f55a8aa
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/63513
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

analysis: fix @NonNull issues with SystemCallLatencyStatisticsViewer

Change-Id: I2d14203943774a9447f618a76d300e509a2e212d
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63495
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf: Fix remote.ui test on Windows due to line separators

This can happen when core.autocrlf is set to false in the git config.
The file containing an expected string is then not converted to crlf and
since the generated string in the test contains crlf, the comparison
fails.

To make tests consistent everywhere, I suggest to disable the autocrlf
of git directly in the Git repository using .gitattributes, see
http://git-scm.com/docs/gitattributes

And to fix the test on Windows, we can always convert the generated
String to lf line endings.

Change-Id: Ife8912ace7895ebca4d19adb53c6ba169ca2d003
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63080
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

tmf: Fix failure in StandardImportGzipTraceTest related to test ordering

If TmfAlignTimeAxisTest ran before, and the window size was small
enough, it was possible for the test to be running in an "Align 1"
perspective with insufficient height for the events table. Because of
that, the expected content in the table could not be found. The tests in
TmfAlignTimeAxisTest switch to generated perspectives without returning
to a "normal" perspective. It makes sense that the tests "cleanup" by
returning to a more normal perspective like the Tracing perspective.

Also switch to Tracing perspective explicitly in
StandardImportGzipTraceTest which is more consistent with the other
tests and is somewhat less error prone.

Change-Id: I0d28af39266bdcdc81e765b5522c92a4a8085602
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63081
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

Clean up top-level pom.xml

Re-arrange entries, fix indentation and add spacing to make the
whole file more readable.

Also specify <project.reporting.outputEncoding> (removes a
warning in the build output).

Change-Id: Ia508f2b24de0813fbaf8d9db1e79e755e351ac5f
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/62348
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>

rcp: Update RCP product plugin's ID

It was simply "tracing", typically our artifactIds include the
whole symbolic name of the plugin.

Change-Id: I46a0faf45f18c64bf8005ad90cf2744e8bcb53d9
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/63059
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>

ctf: make computeAttributeMap return a Map instead of ImmutableMap

Change-Id: Iba609c83c20f11b26bd7b4db7deb2242ad41cdf6
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/63058
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-by: Hudson CI

ctf: fix modifier order of @Nullable annotation

Change-Id: Ib5f40b243b6ecae25fda50dd09ef7c8df16e36f7
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62951
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>

Fix main shell not active using Xvfb

There seems to be problems on some system where the main shell is not in
focus initially. This was seen using Xvfb and Xephyr on some occasions.

Change-Id: Ib6dda93e856deb96cf865dce6253b48c99171184
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62979
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>

tmf.ctf: Accelerate ByteArrayDefintion parsing

Make BADs no longer need to generate definitions uselessly.

This patch yields a 15-50% acceleration in Synchronization benchmarks,
a 5% acceleration in the Kernel Analysis benchmark, and an average event
size reduction of 33%.

Change-Id: Ie581abaecf1d8b188de9f1acc3010fe3f7a65d7a
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/52024
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>

ctf: Collapse if statement in StructDefinition

Change-Id: I2736507e0ebbee70a172930cd4315c89872eb001
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/51857
Reviewed-by: Hudson CI
Reviewed-by: Francis Giraldeau <francis.giraldeau@gmail.com>

Remove all "AllTests" suites

tycho-surefire-plugin's default configuration is to look for @Tests
in classes that start or end with "Test", or end with "TestCase".

We can make use of this and avoid all the nested AllTests.java suites
found in every single test package in the tree. This also avoids
the need for most custom configuration in each test plugin's pom.xml.

To run a complete test plugin from within Eclipse, one can still
right-click on the plugin itself, and choose
Run As -> Junit Plug-in Test.
(Doing this before this patch would actually run all tests twice!)

Note: this breaks the suites in the "alltests" plugin, but those could
be brought back by using something like ClasspathSuite, if desired.

Change-Id: Ibf960df37c5d3c98dc0e0dd722d7c9fe5c0b9d4f
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/59442
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>

tmf: Fix GTK bug when setting top item after changing tree font

Change-Id: If2658c1bdcd98f6a34ce31a76af0dc57c959de2c
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/62953
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>

analysis: Make SystemCallLatencyTableAnalysisTest more stable

Even if the jobs are finished, the table is not necessarily filled yet.
The update could be pending as a event in the UI thread queue, etc. We
can wait until cells are filled instead.

Also export the package so that it can be used elsewhere, (alltests
stress test for example).

Change-Id: I8177c7e6b7e487d59b248e0279c6bcd8da43b515
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62959
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

tmf: Support vertical zoom in time graph with Shift+Ctrl+Mouse Wheel

Change-Id: Ie5486855cd29c95ee3b0d7caad0ceac6c184d725
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/62534
Reviewed-by: Francis Giraldeau <francis.giraldeau@gmail.com>
Reviewed-by: Hudson CI

tmf: Bug 484422: Fix inconsistent vertical zoom

In GTK3, after changing the tree font, the tree item bounds are not
updated until the tree has been painted. The alignment of tree items
should be done after a paint event has occurred.

Change-Id: I47617b881d3f0f981bdacf3ddda572c296dffaf7
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/62876
Reviewed-by: Hudson CI
Reviewed-by: Francis Giraldeau <francis.giraldeau@gmail.com>

doc: Update user guide for time graph markers and vertical zoom

Change-Id: Id4e16737c46492efebb643b75455855ee619e3e1
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Signed-off-by: Francis Giraldeau <francis.giraldeau@gmail.com>
Reviewed-on: https://git.eclipse.org/r/62531
Reviewed-by: Hudson CI

ctf: make CtfTmfEvent expose packet context attributes

This patch allows acces to the packet context of an event from that
event.

Change-Id: I8409c8fbcb7766a2d7be4104159a7d7de6aa188d
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61817
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

ctf: expose stream in CtfTmfEvent

This allows one to access the stream number. It may lead to improved UX later.
The stream ID can allow analyses to know which stream an event is coming
from. If a trace has a control channel and a data channel, this will
allow the events to be easily filtered into the appropriate views.

Change-Id: Iba9de10d9f1f633deed6df81c0d849acf5a3aceb
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62779
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

ctf: Make packet descriptor information available to event

This allows the event to know what the packet context is for
its parent packet. This is especially useful if one puts non-
LTTng information in the packet such as the source gpu.

Change-Id: I21ff283c1032e5a3157408b304b2611d1113d4f7
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61816
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

ctf: Expose map of attributes in Packet Descriptor

The motivation is to allow accessing custom fields in CTF
packet contexts. It decouples the CTF parser from LTTng a bit.

This breaks an unused api.

Change-Id: I2bead35a7e080d4e3ab325395629c44eea455cfc
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61815
Reviewed-by: Hudson CI
Reviewed-by: Francis Giraldeau <francis.giraldeau@gmail.com>
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

ctf: make SIPIE enums store both key and value

By adding the getStringValue and getIntegerValue,
instead of getValue, we have more information available.
This is useful for enums so we can look up the string
name but also sort it easier.

Change-Id: Icfa095a0a8c2b8924809c389dbfe53055c3b63be
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61814
Reviewed-by: Hudson CI
Reviewed-by: Francis Giraldeau <francis.giraldeau@gmail.com>
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

ctf: Introduce ICTFStream

This allows a CTFStream to be shared while not being modifyiable.

Change-Id: I4bf4985abd5221ebfa543d0a7ff0cfeda0874fce
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62767
Reviewed-by: Francis Giraldeau <francis.giraldeau@gmail.com>
Reviewed-by: Hudson CI

Disable overlay scrollbars when running tests to prevent crash

When running tests, the Eclipse launcher is not used which means
that the code that usually disables the overlay scrollbars is not
executed. The overlay scrollbars under GTK2 are unstable and can
cause crash. Specifically, a crash was frequently seen in
tmf.ui.swtbot.tests on Ubuntu.

Change-Id: I130d7fd95affa4f7eb68d2c2d1e0b0134e84ee07
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62780
Reviewed-by: Hudson CI
Reviewed-by: Francis Giraldeau <francis.giraldeau@gmail.com>

ctf: make HashMap child a Map and extract lookupStructScope

The map is self explanatory. The method extraction improves readibility
in my opinion

Change-Id: I9fb23c3223641d80e2f70a18142c3d4b968f0dab
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/51854
Reviewed-by: Hudson CI
Reviewed-by: Francis Giraldeau <francis.giraldeau@gmail.com>

tmf: Remove an annotation temporarily to work around Sonarqube bug

There is a Sonarqube bug with type annotations and arrays, see
https://groups.google.com/forum/#!topic/sonarqube/VTsAYxhGRiA
and
https://github.com/SonarSource/sonar-java/pull/537

By removing the @NonNull, it brings back a warning but makes the
Sonarqube build work again.

Change-Id: I6a03295432c7c452b4dca9018d4c17ba4b991980
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62684
Reviewed-by: Hudson CI
Reviewed-by: Francis Giraldeau <francis.giraldeau@gmail.com>

tmf: Keep usage of ViewerSorter in navigatorContent extension point

Reverting the change to TmfViewerSorter from commit e707a55.

Platform UI doesn't support ViewerComparator in the
org.eclipse.ui.navigator.navigatorContent extension point in the
navigatorContent/commonSorter element (refer to Bug 484248).

This causes a ClassCastException and prevents the TMF project model
elements from being sorted properly.

Change-Id: I3d63032763dc435b594c3f9e8df04d0cb6b38609
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/62640
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>

releng: Fix alltests failure on Mac

The alltests plugin was failing on Mac because it tried to run 0 tests
on the UI thread but it needs to start on the first thread
(-XstartOnFirstThread). We can just set it to not un on the UI thread.
There is no need for the UI thread when running 0 tests.

Change-Id: I1e260f96e1ec51e256e8280262bcbd3c83a92de3
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62574
Reviewed-by: Hudson CI

tmf: FilterColorEditorTest on Mac

On Mac, RGB values that are captured with ImageHelper are affected by
monitor color profiles. To account for this, we can draw the expected
color in a simple shell and use that color as expected value instead.

Also, switching to the Filter mode in the table is not instantaneous,
we need to wait until it's done before capturing the next image.

Change-Id: I930570027ba53f3dbdcecb5925a8e67c24c239b2
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62573
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf : Fix the xml condition multiple problem

This patch initiates the number of state values involved in a comparison
only when the xml element is <condition>. Without this fix, conditions
like OR or AND with more than two operands fail because we were
validating that the maximum number of state values was actually two.
Since the XSD validates that already the condition is removed in this
patch and the allocation of the state value list is done only when the
element is <condition>. In other cases, the list is instantiated to an
empty list.
This patch also add a test with a condition with three operands in
TmfXmlConditionTest to validate this case.

Change-Id: I6e99f770b35f3359a5617bc01b8a53bc77df637e
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62425
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>

tmf: Fix existing trace not being reopened when it as a short path (Win)

This fixes the TestTraceOffsetting test.

On Windows, the temporary path can look like this:
java.io.tmpdir=C:\Users\MARC-A~1\AppData\Local\Temp\ Which is a "short
path". When used in comparison with full, canonical paths, it fails.

In the context of this bug, because the comparison failed, it ended up
creating a second trace for the same location then the test could not
find the proper editor because the title contained a '(2)'.

Change-Id: I41def2180ff910e25489415c9de513212d2f2ea2
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62529
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

ctf: Fix InvalidPathException in CtfTmfTraceValidateTest on Windows

The path beeing passed to Paths.get was not a format supported (/C:/)

Change-Id: I80eb54ae3198381d91ab19c55c991004721d06dc
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62521
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

ctf: Fixed leaked file handle when calling CTFTraceReader.update

If an instance of CTFStreamInputReader already existed in a
CTFTraceReader, it was not getting closed after checking whether or not
it existed. This was causing a test failure on Windows in
CTFTraceGrowingTest.

Change-Id: Ia064d3270882515b32834456771a108970ce8f8e
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62515
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf: Bug 479675: Fix for index not being properly deleted on Windows

Change-Id: Idc97913cf34658452fef95f6e3ba502a57fc13ab
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/60858
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-by: Hudson CI

doc: Update developer guide for time graph markers

Change-Id: I9d8b97e397b94359b8bf2f881f253bf8b19bd1a6
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/62434
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>

tmf: Add vertical zooming support to time graph viewer and combo

Vertical zoom is triggered with Ctrl+"+" and Ctrl+"-". Reset is
triggered by Ctrl+"0".

Ctrl+"=" is equivalent to Ctrl+"+" for vertical zoom-in, and "=" is now
equivalent to "+" for horizontal zoom-in also.

The font of the time graph control is changed along with the zooming,
and the views with post-draw handling to draw labels on time graph items
are modified to use a new method which centers the text vertically.

Change-Id: Ic75c1ebe7e937c612b64346e2814894c500f964f
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/62359
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

Replace usage of ViewerSorter with ViewerComparator

The former has been deprecated in 4.6, it's now recommended to use
ViewerComparator and .setComparator() instead.

Change-Id: Ib7b447a74756b2b4524924f76b93ec0d64f71ed1
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/62262
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>

lttng: Add SWTBot tests for Resources view and time graph markers

Change-Id: Ida3d7c7bed355f8032797d8074db3a9d8551eed2
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/62138
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

ctf: Delete and regenerate trace for tests in LttngKernelTraceGenerator

Later, this can be improved to be conditionally done only if the trace
generator version has changed.

Change-Id: I4d43f388d89d6bafba6c492a93e3d89c9ed9ed22
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/62385
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>

Use maven-enforcer-plugin to enforce minimum versions

<prerequisites> tags are not inherited by children. We can use
maven-enforcer-plugin to really check the Maven version.

See http://stackoverflow.com/q/18055009/4227853 .

It's not really clear if <prerequisites> is really deprecated
or not, but "mvn versions" complains if it is not present, so
we can at least leave it in the top-level pom.

While at it we can also configure it to check for the minimum
Java version the project needs (1.8).

Change-Id: I607c33e3fecf39d932ae7a44d58b85ce20336dc7
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/62280
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>

Fix latest batch of null warnings

2f79cfb was about fixing the errors that cropped up with the new
version of JDT. This latest patch fixes the related warnings.

It introduces external @NonNull annotations to the following
methods in Guava's Multimap:

get()
keySet()
values()

The last two follow the same reasoning as the corresponding Map
methods. The first one, get(), may seem surprising. But Multimap's
get() is really @NonNull, it will at most return an empty Collection
if there are no values in the map matching the passed key.

Change-Id: Ie8dccff9491a8c0200b503ad76fd44092eb975e2
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/61929
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>

analysis: Fix IndexOutOfBoundsException in ResourcesView

This can happen if a new IRQ or SoftIRQ attribute is created
concurrently with a zoom thread that has already queried the state
system for a full state that does not include the new attribute.

The zoom thread can then provide the full state to the ResourcesView to
get the event list of an IRQ or SoftIRQ entry that did not exist at the
time that the full state was created.

Change-Id: I1c5f8a8332f44b44f1954b01c3503a844dedc22e
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/59616
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

lttng: Keep tracing project between tests in KernelTest

This avoids having to wait for the same synthetic-trace to finishing
indexing for every test method.

Change-Id: Ib4e75dde2391ab3609d13923d0ebd5628d8477e9
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/62137
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf: Fix duplicate trace check in TmfOpenTraceHelper

If the provided path was a convoluted path which resolved to the same
path as an existing trace location of the same name, it was incorrectly
considered a different trace and renamed with a numbered suffix.

Change-Id: Ic802cd5ffd83c5a29e422dfddbb0463aca1b372a
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/62136
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Reviewed-by: Hudson CI

analysis: test SystemCallLatencyAnalysis

This test opens a trace, then opens the system call view to ensure
that the table is properly populated.

Change-Id: Id38c61ab64703cbe0333d00f1980073f0f339bda
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62042
Reviewed-by: Hudson CI
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

control: add swtbot test for testing load and save feature

Change-Id: I26068880a63c9a69a03d4b923ff05a1cfb679587
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61921
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

control: Add initial SWTBot tests for control view

Change-Id: Ibc3fdf3eaa00474d4b54ca799860c18143a2ecc8
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61900
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

lttng: Don't call PlatformUI.getWorkbench() if the workbench is not running

Workaround for Bug 480948.

Change-Id: Ib461f802f496bbfe5ad1ecef7fdea85ffd674eee
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/59294
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

analysis: Test SystemCallLatencyTableView

This series of unit tests tests the underlying widget with no analysis
or trace.

Change-Id: If30a333117691fb42545bc16b70b4a9e20c26237
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62033
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

control: Bug 483560: Refresh UI once after adding all children events

This will redraw the tree only once after all children event components
are added for channels, ust and kernel provider. Before this change the
UI was redrawn after each event.

Change-Id: If48ac0740be8884bcf52bb7f3da4a15859a3639f
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62200
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

Add @since to doRun in AbstractTimeGraphView's ZoomThread

Change-Id: I214ae12c966bf6736c18cf66920158082accf1b7
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62324
Reviewed-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

analysis: Allow SegementStoreContentProvider to accept Collection<ISegment>

This will allow easier testing.  This also would also make the code easier to
re-use by removing part of the dependencies on segments.

Change-Id: I3d77a647306cae0c496ce894f5d326ddf1fd4e3b
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62031
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

analysis: add statistics core tests for segments

Change-Id: Ie2ac17401f7f1e155c694ad26fd8ea2cb9ce792e
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61835
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Reviewed-by: Hudson CI

analysis: add standard deviation to the segment store statistics

This patch will be useful for extracting more key metrics and flagging
outlyer segments.

Some reminders:

The standard deviation is the square root of the variance. The variance
is the sum of squared elements divided by the cardinality of a list. To
calculate the standard deviation incrementally, we keep an accumulator
of all the squares of a latency. This is then divided when needed by the
number of elements (segments) and square rooted.

Also, this patch calculates an online mean with less rounding errors.

Change-Id: Ia918f08f2351d7086bd05aac1ad645cfff13eb58
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61824
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

analysis: introduce os linux swtbot tests

Change-Id: Ib82f829852d75ebacf87e1bd60a91d788efff831
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61926
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

analysis: expose GetTableViewer in AbstractSegmentStoreTableView

This allows for easier testing.

Change-Id: I3ff2e08e2acd4887b25e805e080b74fc7509207a
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62032
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf: fix ctf validation tests due to missing traces

Certain test were not executed because trace files where not found.
This problem was introduced when the test traces repository was created

Change-Id: I81a88870615ef3188d7f50de41535e5dff76ff71
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62063
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI

lttng: Make ControlFlowViewTest more robust

Added a condition to wait on. Added the concept of the view being dirty
so that we know if the view is ready or not. This required some rework
of the zoom thread methods to make it less error prone in keeping the
dirty flag consistent.

Change-Id: Ib73d84363e90aae1ca046c101f2086e3840b5bc2
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62154
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

tmf : Add test suite for the XML conditions

Change-Id: I2a338f8b61cd920a7b3eb528be00aef696d6776b
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61673
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf : Add comparison between two state values in the XML core

Previously, it was only possible to compare a state value to a constant.
Now, we can compare directly two state values. With that, it is possible
to query and compare two state values already stored in the state
system. For now, only the "equal" operator is supported.

Change-Id: Iebbcb90c2c0603c5ec1191530e25dd3b0a1cf853
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61671
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

Fix some null warnings

Change-Id: Ib0290125def1e0ea58bcf81c2075ef99db3f1f1a
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/59428
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>

control: add service test cases for load and save commands

Change-Id: I3455c2c9df7cb43bca9fe0905d418bbe66ad0c38
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61554
Reviewed-by: Hudson CI
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

ctf: Add lost events to synthetic LTTng kernel trace generator

The trace generator now creates packets with lost events pseudo-randomly
approximately once out of 10 times the number of channels.

The trace generator also now sets the content size of the last packet of
each channel so that the total number of events (including lost events)
is exactly the number of requested trace events.

Change-Id: Ie804e5650d5a4186225b5dfcc75526b0fe9c2704
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/61599
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

ctf: Fix lost events timestamp in CTFPacketReader

The lost events should be inserted after the last event in the packet,
not before the first event in the packet.

This was a regression where the changes to CTFStreamInputPacketReader
from https://git.eclipse.org/r/50683 were not ported to CTFPacketReader.

Change-Id: I055887b02212bd43052d36fbbd0dde3ca36edb6e
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/61598
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI

tmf : use of StringBuilder.append() instead of String operator +

Change-Id: I062c411af568289f5c17a279953703d720109119
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62015
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf : remove occurrences of literals in FilterTreeLabelProvider.java

Change-Id: Ic30315e1bed6f360ec4760341ddee873891e43ee
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/60555
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

common: Annotate String.split

This can't return null or null elements. Also cannot accept null
parameter. Fixes a warning in LTTngControlService.

Change-Id: Idb605e95fd20abfe8ecf73037bcc2a7b2e17dd70
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62075
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Tested-by: Alexandre Montplaisir <alexmonthy@efficios.com>

common: Annotate Element.getAttribute

It cannot return null. It returns an empty String when the attribute is
not present. Fixes a warning in XmlViewInfo

Change-Id: Ia4b3ee9ccc2990597cbba16c2d0d1e3349a78bec
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62078
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>

common: Annotate some Class.get*Name methods

This fixes a warning in AlgorithmManager. getName and getSimpleName
cannot return null whereas getCanonicalName can (anonymous, local
classes).

Change-Id: I2ee6ff668a93b47617389d2a3a25157af7184349
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62077
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>

common: Annotate RandomAccessFile.readUTF

It can't return null. Fixes a warning in TmfUml2SDTestTrace

Change-Id: I7698d58639dc2099bed60c5f37c28a249d5fb22c
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/62076
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>

tmf: Use prioritized List instead of Set in XmlStateProvider

This allows to have a kind of priority between the event handlers. It
helps to improve dependency between the event handlers. Now, when
receiving an event E, it is possible that the result of a handler X
depends of the change made by a handler Y on the same event.

Change-Id: I33592e907d4c2f024a46cc365f0bbe796e940049
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61672
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

ss: Avoid concatenating nonliterals in TmfStateInterval#toString()

Change-Id: Ia5a18114a73418ad677b7d8b3a606c2f67af36b2
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/60557
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

remote: provide output listener for command execution

This listener interface will allow users of the ICommandShell to provide
2 callback methods to get intermediate results for the standard and
error output. Useful for long lasting command executions.

Change-Id: Ibc90bfd679226eee7ef34f170b63abd819bb75e5
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61586
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

Fix new errors due to automatic annotation of Class<T> types

The null analysis now automatically assumes that generic class types
(the "T" in Class<T>) are always @NonNull. See bug 477719.
This clears a bunch of warnings we had due to using Something.class
as parameters. Great!

However, this introduces a new problem. When a generic type parameter
is used both as parameter and as return value, and the client specifies
an annotation on the parameter during a call, the default behaviour is
to assume this annotation on the return type too.

In some cases this assumption is correct:

    List<@NonNull String>.get() returns a @NonNull String

But in some others it is not:

    Map<@NonNull K, @NonNull V>.get() should *NOT* return a @NonNull V.

Now some methods of the form:

    V something(Class<V> type);

also follow this pattern, and the "automatic" @NonNull applied to V
also gets applies to the return value. In some (most?) cases this is
incorrect, and we have to supply external annotations to change the
return value to @Nullable. Just like we did for Map.get().

See bug 483143 for more information.

The return values of the following methods are now annotated as
@Nullable:

Class.getAnnotation()
DsfServicesTracker.getService()
IRemoteConnection.getService()
IRemoteConnectionType.getService()

and related null-checks were added.

Change-Id: I2c60835160a46e88ff293a5fd68774021c2b00a9
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/61521
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>

ctf: preserve byte order if explicitly set in a typedef

Change-Id: I904cd931da955f17b7cafa285f48d60c0c0358a1
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61684
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Reviewed-by: Hudson CI

ctf: do not assume endianness of integers for singletons

Change-Id: I59ff6253564a05122da5b702e04b1f291f1b54fb
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61709
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

common: Annotate ByteOrder#getNativeOrder

Change-Id: I6c64addc87111f2974bc0b97a527ebd8fd3f6b1f
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61683
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

ctf: remove some null warnings in TmfTraceManagerTest

Change-Id: I403f87cfbaa018831f497d2374675ba224cc027d
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61682
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>

ust: Add initial SwtBot test for memory usage view

Change-Id: Ied9bb9b8539e3a15cc8facc6678f486c2e092558
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/61504
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>