tmf: Add the possibility to choose the root node for synchronization

The root node in the spanning tree used for synchronization was
always the first hostId in lexicographic order.
Now the user can choose the trace that will be used as reference
for the synchronization.

When the user wants to synchronize an experiment a window similar
as the one used to add a trace to an experiment is shown.

The first trace is selected by default.

Change-Id: Iaaeb4c7905dc6171d9f8a40c4dbf4e8ba65d5ebd
Signed-off-by: Cédric Biancheri <cdc.biancheri@gmail.com>
Reviewed-on: https://git.eclipse.org/r/56769
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

analysis: add icons for analyses

Change-Id: I494ca354404aa85e9af240567b7981e2921bb069
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/58121
Reviewed-by: Hudson CI

analysis: Fix TreeMapStore's serialization

Change-Id: Ia6c3299d8e0e940cae52118c14ca04c16aedd049
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/58118
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-by: Hudson CI

analysis: add latency table icon

Change-Id: I2565d266495f6db2a555cfa1840ff66d3fa8e924
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/58122
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

ust: add memory analysis icon

Change-Id: I3affe2d06c1ac80cf99408d37b55d6525c4e4294
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/58124
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

Add null-checks for Map.get()

Map.get() can return null, and this can mean two different things:

1) The value associated with this key is 'null'.
2) There is no such key in this map.

Java 8 type annotations (Map<K, @NonNull V> for example) can cover
for the first case, but not the second! Map.get() can *always*
return null, no matter the map declaration. The enhanced null analysis
reveals that.

Many places in the code call Map.get() without null-checking, we can
fix these cases in advance. In many cases, this also exposed places
where iterating on map entries should have been done instead of
iterating on keys and doing a lookup.

Change-Id: I8b94a7ed7193f41479b3b0bc9b5990213b9718a3
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/52015
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

Add a script to check for outdated Maven dependencies

We already have a Hudson job that runs this, we could commit
the script in the git tree so that it can be run locally.

Change-Id: Iefcae17dad4639e917e3440bce6929811773bd9c
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/51682
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-by: Hudson CI

tmf: Allow refreshing manually all ITmfTraceEditors and traces under experiment

This is a follow-up to the comments in https://git.eclipse.org/r/#/c/57465

Change-Id: Ia6e8ecc056a5c662255c35f029f72bbe1cc2785a
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/57559
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

Update Oomph setup file with Trace Compass 1.1.0 baseline

Change-Id: Ibf9ccc8474731010dc59e5a3a3d2d1d562c088b2
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/57562
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

tmf: Fix minor typo in test

Change-Id: I50db090d9eeaccbbad5fc57d753bbc33770f4bf2
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/57560
Reviewed-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-by: Hudson CI

tmf: Fix time-axis alignment for cases where the series is not sorted

The time-axis alignment code for XY chart viewers (SWTCharts) assumes
that the series in the chart will be sorted and that the last item in
the series will be the one with the highest value. Since SWTChart
allows specifying series values in no specific order, this should be
handled properly in the time-axis alignment code.

We do not have any views right now that shows this behavior but if
you want to see this bug, you can try this patch in combination with:
https://git.eclipse.org/r/#/c/57011/

Change-Id: I8a850f9a12f945fe5616b0b82176d98ea45c33ae
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/57552
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI

Remove tracecompass-baseline-1.0

Change-Id: Ia04f2a5b1a3cbe3576579712e458f39efc9bbd35
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/57557
Reviewed-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-by: Hudson CI

releng: Upgrade jboss repository-utils to 0.23.2

This is the latest stable version available for the plugin.

We need to use skipBuildInfo otherwise some extra json files
are generated.

Change-Id: I437afd1483705311cdf36961f23d6c8df105c054
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56317
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

Change some @since to 1.1 for things that were really added in 1.1

Change-Id: I57e900622e1e31984233391beebe3ce5a0cb6e07
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/57456
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

Add Trace Compass 1.1 baseline, fix some API errors

Change-Id: I356706eae6fd2afb85f0f32fae91291c7e11dcad
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/57447
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

tmf.core: make ITmfCallsite getFileName @NonNull

Change-Id: Ie93ca879cb2954bb86eaf68f3dea40fb056b6c2e
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/42916
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>

tracecompass: update pom sonar plugin

This should add support for sonar 5.1+ if eclipse.org upgrades to it.

Change-Id: I02d8887d1f8b95a0b5696564d447ee341f643b21
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/55990
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: Link bookmarks with resource markers in abstract time graph view

Change-Id: I5a0c8f03df5606c2bb42517bae10b7f52e539fcc
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/57254
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf: Support user bookmarks in time graph

Adds the Toggle Bookmark, Select Previous Bookmark and Select Next
Bookmark actions to the time graph viewer and the abstract time graph
view.

Adds an Add Bookmark dialog.

Adds an ITimeGraphBookmarkListener interface.

Uses the old Previous/Next Event icons for the Previous/Next Bookmark
actions, and modifies the Previous/Next Event icons.

Change-Id: Ic7e51c4cf7aab4f78b49ec7bc9e2b87faff9f325
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/57135
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf: Add SWTBot tests for manually refresh of traces

Change-Id: I6e52066382b958e2c50ebdd9982e8651bdc87889
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/57138
Reviewed-by: Hudson CI

swtbot: Add some failure messages to test conditions

Change-Id: I725b75e7f66447532b3eacd5bd86518ebfee8c8c
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/56017
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf: Add command to manually refresh traces

At the moment, only the events table will be refreshed.

Change-Id: I7b9933d8e52e7e7fc821a3b41f1e688bf4c81d09
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/57137
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf: Delete unused code in GraphNode

Change-Id: Iab9a0769235f3b05ed25100e3c821d283a6e0422
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/57087
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

lttng: Add utility methods to get CTF tracer versions

It's not uncommon to check for tracer names and versions of CTF
traces. We can offer convenience methods to do so, instead of
having to parse the environment metadata every time.

Change-Id: I9c4cbc7e36d24d972889f6c4bbc16d2db852f741
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/51661
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI

lttng: Use SWTBotJunit4ClassRunner in CallStackViewTest for screenshot on fail

Change-Id: I078f3c1ca4c18fd0f9d5a247f42ea71ae1f3823d
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/57145
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf : Redefine CompareTo() in EventOccurrenceObject

Change-Id: I6251531e51ff398afdcc892d915092203f011dec
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/57108
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf: Support markers in abstract time graph views

Change-Id: Iadbe8cd42681fec05ca295b095b60e333edb6865
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/56753
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf: Avoid duplicate literals in GraphNode

The string literals have been extracted into a constant variable

Change-Id: I4f48ca60faac42186bba26f9f9513993409d2bbe
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/57085
Reviewed-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf.ui : Iterate in OffsetTraceHandler using the entryset instead of the keyset

Change-Id: I2c7c510feec20793570faa6723a7c8e1e0ed1ab6
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56992
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

Use pde.junit runtime from target to allow e4.5 tests to run on Eclipse 4.6

Fixes trying to run tests with Eclipse 4.6 targetting Eclipse 4.5 using Java 7.
Whitout this change, the runtime from Eclipse 4.6 gets picked up and it
requires Java 8.

Change-Id: I8597ee30fd094bad5b004e21d3fa5b939ca49b7a
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56563
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Tested-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>

ctf: Remove the callsite support

The existing support for CTF callsites was based on an out-of-tree
branch of LTTng-UST. This branch was since then abandonned, and
callsites (now rather called "debug info") will be implemented
differently.

This makes CtfTmfEvent not extend ITmfSourceLookup, because for
all intents and purposes it should never have. Correct source
lookup will be added with the new support.

Change-Id: Ic8b9c133965a1230f7754bbdcb7bc933414f38b2
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/50310
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI

lttng: Move the UST Callstack analysis to use ILttngUstEventLayout

The UST callstack implementation was defining the event and field
names it's looking for on its own. This is bad and prone to future
breakage.

Instead, integrate these event definitions in the LttngUstEventLayout,
and move the analysis to use these instead.

Change-Id: I098d8569ab2c9c0d2ffd0975aba528a177a02af2
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/50686
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf.ui : use an entryset instead of a keyset in the GraphNode

Change-Id: I1175efa2bae9b933d1abcbb041c0fbc9ce198e9d
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56991
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>

tmf: Bug 478578: Add weak binary validation to text trace types

Change-Id: I0e693ff8f03907b7b3fc4c2f2dafcf547c2db698
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56936
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI

lttng: Support for cancelling of latency request

Change-Id: I86a113c3ed322e673cb0f21c1fbdff472c0b3958
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/56897
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

lttng: Improve and speed-up control tests handling

For each test a 3 seconds delay is done waiting that the ControlView
is open and initialized properly. This patch optimizes this.

Change-Id: I5308973613c23c481757fd26c84f87c0a9edf94a
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/52971
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf: Bug 467751: Re-introduce fix for concurrency bug

The problem was brought back accidentally when using the
TmfUiRefreshHandler instead of display.asyncExec(). It had been fixed
in Trace Compass 1.0. It's not part of Trace Compass 1.1 anymore.

Change-Id: I29096b35c02d8404b9a2ad8deb0bb62fcf0dc33a
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56951
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

lttng: Fix error in virtual machine analysis

The analysis was not checking if the virtual cpu was idle
before setting the current state to preempted.
Now it is ignored in this situation.

Change-Id: Iddef5c2ec1c916ea0b591d0106b4e6473febb819
Signed-off-by: Cédric Biancheri <cdc.biancheri@gmail.com>
Reviewed-on: https://git.eclipse.org/r/56728
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-by: Francis Giraldeau <francis.giraldeau@gmail.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

ss: Speed up TreeMapStore#add()

The add() does not need to call contains(), a O(n) operation, to check
if the element was already present. It can simply re-use the return
value of TreeMultimap.put().

Change-Id: Iab1971f6310163099906df0a1702d106e3640316
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/56904
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI

swtbot: add callstack view tests

This test reads a trace and tests the callstack at different places.

Change-Id: I5d66f68aa5c65eaa325b1db10ea3df9aa1be475a
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/41787
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 UST swtbot plug-in

Change-Id: I5a6db70e104734c8158a57b3c635bdbbbf4837dd
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/44641
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: Move the UST memory analysis to use ILttngUstEventLayout

This will allow the analysis to work with different versions
of the UST.

Removed the getRequirements() test, because testing (what used
to be) statically-defined values is not particularly useful.

Fixes Bug: 470186.

Change-Id: Id3c80467a55c9c3d47c8073d49e8f4ea3ead4f5b
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/54272
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: Support to select start and end event in LatencyAnalysisViewer

Change-Id: I325f51a687ff82b08f6216912ec03d4adb5e407d
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56709
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf: API to add commands to context sensitive menu of TmfSimpleTable

Change-Id: Ibb3b105d0f7c39404fb4e3bb087ab2c4f65d7407
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56708
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

ss: Make ISegmentStore implement Collection

It used to implement Iterable, not many changes are required to
support the entirety of Collection.

Change-Id: I1622c222d3eb72afdde47ca0a462003dda85d0cd
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/56680
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf: Support markers in time graph control

The time graph can be provided with a list of markers.

Each marker has a time, duration, and color. The color can have an alpha
value for transparency. A marker can be specific to a particular time
graph entry, otherwise it is drawn for the full height of the control. A
marker can have an optional text label. A marker can be drawn in the
foreground (above the entry's time events) or in the background (below
the entry's time events).

The drawing of markers can be turned on or off globally.

Change-Id: I15d19925ab5371f21fbaf81d51182d17868e92a6
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/56472
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf: Update time graph control drawing order

Draw the item's name and middle line as part of the background layer.
This allows grid lines, and later markers, to be drawn on top of the
middle line.

Change-Id: Icfe9fcdaa7c39bfe2bc9c055929a86f157b21773
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/56372
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

Change "Discouraged Reference" to warning

In alltests plug-in, this allows easier reuse of SWTBotStressTests
to run an individual test that is outside the alltests plug-in
(and is internal).

Change-Id: Ie570d5df176f5eb7d25c72070f296cd6d4d1b324
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56675
Reviewed-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-by: Hudson CI

tmf.ui: fix UpdateThread's illegal thread access on updateData

The method updateData should only be called from the UI thread. This
patch makes the update thread call a ui runnable to update the data.

Change-Id: If1336322a0be20213dd53e151d2e4707481610b7
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56383
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-by: Hudson CI

treemapstore: make the iterator copy on read after write.

This may appear to be an odd synchronization scheme, but
there are normally 1 million writes per read, and then at
the end there are no more writes, so this will minimize copies
of the data much more than a traditional copy on write.

Change-Id: I9d78bc2a1de788ca41742e57d0df45d42ac4642a
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56468
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Tested-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>

segments: make TreeMapStore thread safe

This change will allow querying/iterating over the TreeMapStore is being built.
The data is not always accurate, but it is true to the time of the query.

Change-Id: I8721002fda68f7e88cb9ec742d76568a8c53e9ae
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56381
Reviewed-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Tested-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-by: Hudson CI

Add new custom test suite to run the SWTBot tests repeatedly

Change-Id: Ida815384e4aed172194b830d196f3a595a1dc8d4
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/51779
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 a ILttngUstEventLayout

The event names exposed by LTTng-UST for the memory analysis have
changed in LTTng 2.7. They are now prefixed by "lttng_ust_*",
instead of the previous "ust_*".

To support various versions in parallel, we can use a "event
layout" interface, similarly to what was done for kernel traces
with IKernelEventLayout.

This patch adds a new ILttngUstEventLayout interface, and
implementations for 2.6- and 2.7+ versions. The memory
analysis itself will be moved to use this layout in the
following patch.

Change-Id: Id1f4bde4f5fc56bb3f6d003e51716944ce2c3f93
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/50684
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: Bug 478216: Incorrect text highlighting in event table due to tabs

Use GC.textExtent() instead of GC.stringExtent() to calculate the
matching text background highlight x-coordinates, as this method
supports tab expansion.

Change-Id: I98cfdc7e46e7f10086882650f941fc099544f898
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/56554
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

rcp: Fix error "Could not resolve module: org.eclipse.equinox.ds"

org.eclipse.equinox.event needs to be auto-started.
It was also necessary to tweak the start levels for this to work.
This was inspired by the platform's sdk.product file. See
http://git.eclipse.org/c/platform/eclipse.platform.releng.aggregator.git/tree/eclipse.platform.releng.tychoeclipsebuilder/sdk/sdk.product

Change-Id: Ie7e6cc7f5ba5d4fb35a7ae82d02d244bcfdba20e
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56457
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf: Bug 476390: Support SHIFT+left click to update the selection range

Change-Id: Ie11006e88c325c77b16d5d7bdb3765944b54e0f6
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/55050
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf: Bug 477794: Persist trace group root during remote fetching

This will improve the import feature of LTTng Control the performance
because it doesn't scan the parent folder for the relevant traces of
a session.

Change-Id: I07739c7e229c2026cb8393a19061152a51678dde
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56224
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf: Fix for visible middle scrollbar on Mac

On Windows we need to use a workaround to hide the vertical
scrollbar. By using that workaround on Mac, it actually makes the
vertical scrollbar visible all the time. Instead, we can use the
SWT.H_SCROLL | SWT.NO_SCROLL combination to only get the horizontal
scrollbar. This doesn't work on Windows but can still be used on Mac
and Linux (GTK).

Bug: 466870
Change-Id: If199708b41ed1b9467c6950d5fcf7ce7e461f1e1
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/50115
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

control: Fix dead store sonar warning in destroy session handler

Change-Id: I6fdc75a373cae524f45cd5d63fb1f724286f08fc
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56431
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

control: Don't hold a lock when in dialogs and add null annotations

Change-Id: I45d2e1d45eceeefd4c39ffed06a8f674c91d3e78
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56368
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

control: Bug 477630: Fix concurrency issue in destroy session handler

Use a lock to protect the list of sessions from concurrent access and
modification.

Change-Id: Ie671910b31b226b202e3ac70e02e32ca8fb9fd01
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56333
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf: add import/export package swtbot tests

Change-Id: Ie4da7c0cd4b9dce58b770f776192d42306f3a291
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/37772
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>

GDB: No longer synchronize on a string.

This is dangerous for two reasons:
1- if the string value changes, the object changes too, so we can
corrupt data.
2- if the string is internalized, it can cause unwanted delays with
other objects synchronizing coincidentally on the same string.

Change-Id: Ibfeb3702ef82424b2fca86956f8227b52ae8b945
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56133
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

analysis: Correctly track the state of scheduled-out processes

On a sched_switch, the state of a the process going out should not
always be WAIT_BLOCKED (the yellow state in the CFV). If the prev_state
field indicates 0 (TASK_RUNNING), it means the process was scheduled
out due to preemption, not to being blocked on something. In those
cases, we should put the process into the WAIT_FOR_CPU state.

Bug: 477052
Change-Id: Ic181811c11029d6374a5e65149a103febd25b5ea
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/55770
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

ctf: fix output writer segment intersection algorithm

Let's say you have two segments, A and B, they have the points
A0, A1, B0, B1. to see if there is intersection between the two
segments in one dimension, here are the exhaustive possibilities.

     A----A
B--B              1
   B--B           2
      B--B        3
         B--B     4
            B--B  5
  B----------B    6

we need to only reject 1 and 5

In 1, B0 && B1 < A0
In 5, B0 && B1 > A1

Now B0 < B1 so we can simplify to
B1 < A0 || B0 > A1
Applying demorgan we get
!(B1 >= A0 && B0 <= A1)
If B is the entry, B0 = entry.timestampStart() and B1 = entry.timestampEnd()

The intersection is the not(!) of the formula above.
entry.timestampEnd >= startTime && entry.timestampStart <= endTime

The patch changes the intersection algorith to that.

Change-Id: Ie3205a3c7da970daf94c0cb93cb336fe0851f039
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/53001
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

ctf: Make CtfStreamInputReader NonNullByDefault

Change-Id: Idb221d898e7b734fddcacc31e3868f5f43a0c01a
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/54647
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf: Fix rounding error in seekEvent by ratio for TextTrace class

A similar fix was merged for the CustomTxtTrace (see 391277) but has
been never transferred to the TextTrace class.

Change-Id: I94267e47e4cf2b080b5a396ea3a98313c01a43bb
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56112
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

analysis: Enable filter dialog in Resources view

Change-Id: I5c50ed03a072c4ea24aed839d765890c3773a6bb
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/56002
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf: Split time graph filter dialog from TimeGraphCombo

This allows the filter dialog to be used with a TimeGraphViewer.

Change-Id: I78323a0848ced5139fc96b49973a19998e564cbf
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/56001
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf: Fix packing of columns in TimeGraphFilterDialog

The packing should be done after the elements have been expanded.

Change-Id: I7e4351056391ffb76734ea3bae034dbb0982df8f
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/56000
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

ctf: Remove useless alignment check

Now that default alignment is 1, statements like "alignment =
((size % DEFAULT_ALIGNMENT) == 0) ? 1 : DEFAULT_ALIGNMENT;" make no sense.
A- the % 1 will always return 0 and B- the return values are either 1 or 1.

Change-Id: I30613fba9cb586b884675c55544727fea0efea89
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/51853
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>

swtbot: fix npe in import tests

The trace helper getEvent(rank) no longer disposes the trace.

Change-Id: I6ee7753c4909c1eba231f0402eda4737d8c167ea
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/53340
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

ss: update some classes to current code standard

* Name fields fSomething
* Remove this. before field names

Change-Id: I49028e34a496e16aa5fa89be01fad7ef906d4fc0
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Reviewed-on: https://git.eclipse.org/r/52754
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf: Fix incorrect TimeGraphFilterDialog result when dialog shell closed

The result is set to null when the Cancel button is pressed. But when
the dialog shell is closed by pressing the [x] widget, the result
remains set to the value set the last time that the dialog was closed
with a button. The first time that the dialog is opened, the result is
initialized to an empty array, so closing the shell filters all items.

Change-Id: I82225608e54a233edc9fa7ee66dd826c55d50c52
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/55999
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

ctf.core.test: add trace to .gitignore

Change-Id: I0f52a1a266f03f15331e9dc9c8ec5ab04e485d73
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/56020
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

ctf: clean up EventDeclaration#equals

Use NonNullUtils.equalsNullable() to make code more legible.

Change-Id: I36ccfce40e41c71907483db357906fc96f3d4d42
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/48038
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-by: Hudson CI

tmf: Fix SWTException in FilteredCheckboxTree

The tree can be disposed by the time the refresh job is done.

Change-Id: Ic1f4761940dc2866762601945c1b93b3836db330
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/55998
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Reviewed-by: Hudson CI

tmf: Persist filters of opened traces in AbstractTimeGraphView

The filters of each opened trace are stored in a map and restored when
switching from one trace to another.

Change-Id: I101961009eaf9123aca0995cc1e7ea862f13d262
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/55749
Reviewed-by: Hudson CI

tmf: Fix TimeGraphCombo item height synchronization issues

- Remove the caching of item height data in the tree item widget that is
meant to prevent repeatedly setting the time graph item height.

This only saved a hash map lookup in the time graph, but required two
linear searches in the TreeItem widget keyed data. It also could prevent
the height alignment to be performed if the time graph changed its item
heights without the time graph combo knowing about it. For example, all
time graph item heights are reset on refresh.

- Set the input on the time graph before setting the global item height.

Getting the global item height involves refreshing the list of visible
expanded elements, which also synchronizes their expanded state. If the
time graph input is not set yet, this causes all tree items to be
unnecessary (but temporarily) collapsed. As as counterpart to changing
the order, setting the global item height on the time graph will update
the item height of every existing time graph item. This only happens
once (when the input is set).

Change-Id: Id3519cbf8f9ce283583ac75af9c4f3ec7174b24a
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/55786
Reviewed-by: Hudson CI

Analysis: Add unit tests for dependency graph builder module

Change-Id: I214e1539b93705789bbb04e007ad893383d5cbbd
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Signed-off-by: Francis Giraldeau <francis.giraldeau@gmail.com>
Reviewed-on: https://git.eclipse.org/r/51080
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

Analysis: Add the dependency graph builder module

Change-Id: Icca0b33d07b14eee7009a0abd8fd064da061eed0
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Signed-off-by: Francis Giraldeau <francis.giraldeau@gmail.com>
Reviewed-on: https://git.eclipse.org/r/51079
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

analysis: improve graph statistics scalability

The TmfGraphStatistics was keeping all duration values. Individual
values are not required to display summary statistics. Instead, this
patch updates the sum per IGraphWorker. Also, we introduce the fTotal
field and remove the special placeholder String key in the map, allowing
to rectify the keys type from Object to IGraphWorker.

* update v2:
  * synchronize only on fWorkerStats
  * make only one map lookup to update worker sum

Change-Id: I202945bdad6a0cb759b2fc4cbb791b3afef7c3cc
Signed-off-by: Francis Giraldeau <francis.giraldeau@gmail.com>
Reviewed-on: https://git.eclipse.org/r/55780
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>

tmf: Update the initial item height calculation in Linux

In Linux the item height is calculated by the difference of the y
positions of 2 first consecutive tree items. When the 2 tree items
are available a paint listener is added to the tree and in the paint
method the height is calculated using the 2 stored items. The
paint listener is needed to make sure that the items are painted
and the y values are correct. However, between the time the items are
retrieved and the paint listener is executed the first item could have
been expanded and the y position second item is changed so that the
calculated height is much bigger. This patch fixes it by retrieving
the items in the paint listener.

Up to now this condition happens very, very rarely. However, with
patch https://git.eclipse.org/r/#/c/55749 this will happen consistently

Change-Id: I528c9baa5602cf8897ff84dc98cf71be2fb1fb91
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/55779
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>

tmf: Clamp single time selection to time graph bounds

When the selection is updated to a single time, it should be clamped to
the current bounds. This will make it consistent with the handling of a
selection range update.

Additionally, when the selection range is updated with possible
notification, the check to see if the selection range has changed is now
made by comparing the clamped values, not the received values. Although,
in practice, since this update is triggered internally, it should not be
received with out-of-bounds values.

Change-Id: Iee93f7ef401fa6efc087aa03687de51f7f994590
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/55671
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf: Avoid zoom on selection range update in AbstractTimeGraphView

When the selection range is updated externally, the selection is set on
the time graph widget with the 'ensure visible' flag set to true. This
may or may not cause the window range to be updated.

The AbstractTimeGraphView unconditionally starts a zoom thread in case
that the window range has changed. It should start it only if the range
has changed.

However since the fix to Bug 476148, if the selection update causes the
window range to change, the AbstractTimeGraphView will always be
notified through the ITimeGraphRangeListener callback, and this will
start a zoom thread. So it is now unnecessary to worry about starting a
zoom thread on handling of the TmfSelectionRangeUpdatedSignal.

Additionally, in the handling of this signal, the call to method
synchingToTime() is changed to use the time graph's selection time, not
the one received in the signal, as the time graph may clamp the
selection time according to the current time bounds.

Change-Id: I9e5d1f69107f56d4940fb2a952b41cb150832635
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/55670
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf: Merge TmfEventProvider#getType() and ITmfTrace#getEventType()

TmfTrace has access to both getType() and getEventType(), which
come from two separate super class/interface, but do the exact
same thing.

We can simplify this by only declaring it once, in ITmfEventProvider
(which ITmfTrace extends).

Change-Id: I8a605da5bfd28548d36cd4ce18a1e94d48698176
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/55344
Reviewed-by: Hudson CI

analysis: Fix the state of processes in the LTTng Kernel Analysis Module

A process with prev_state=1024 was shown as blocked, while it is actually
ready to be run on CPU, so that state must be ignored.

Change-Id: I0f2e7acd05e1592e5a91d898869a9a82ad66ead1
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/35492
Reviewed-by: Hudson CI

ctf: Fix unreliable handling of invalid location

This could lead to bad seeks when opening experiments (bug 476816).

In experiments, it is often the case (expected) that a seek to a
location will result to an invalid location for one of the traces in
the experiment.

There are issues with the handling of seeking to an invalid
location (CtfLocation.INVALID_LOCATION)
1. In one spot, == is used instead of .equals which means that a
location read from the index is not considered invalid, but rather,
it seeks to an offset of 0.
2. Even if the location is properly considered invalid (== fixed),
the code relies on the "last packet end time" to seek past the end of
the trace which then make the context location become invalid. This
works correctly when the trace was fully read the first time but the
"last packet end time" is not necessarily the end of the trace when
it is opened a second time.

Instead, we can explicitly check for CtfLocation.INVALID_LOCATION
in CtfTmfContext.setLocation and CtfIterator.seek to make sure no
bad seeks occur.

Bug: 476816
Change-Id: I4c74259ae3b67fb22cae302f2d5f09f1adda51f0
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/55429
Reviewed-by: Hudson CI

tmf: Bug 477012: Time graph fails to send window range notification

Change-Id: I29b57b77db71b78deb0161e57c24135f3af60191
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/55596
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf: Bug 477009: AbstractTimeGraphView sets wrong bounds

When no supported trace is selected, set the defaults bounds on the time
graph widget to [SWT.DEFAULT, SWT.DEFAULT] instead of [+INF, -INF].

Change-Id: I814fb72886a25be5d2fe935f5f0e37dc207686f5
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/55593
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

btf: Remove AutoCloseable from BtfTrace

Despite many great advantages, AutoCloseable does not work well
with subclasses of ITmfTrace, because it would imply that such
objects are not exposed publicly by other objects tracking them.

This is not the case in TMF, as many objects expose ITmfTrace's
through getters. Using instead TmfTrace's standard dispose()
method is more appropriate in subclasses.

Change-Id: I1d342d15dba7f8f6e75985f8aee6ed159838d424
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/51540
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf: Bug 476551: Omit empty traces in standard import and remote fetch

Change-Id: I1a0307f98f479c28300ca4c42f4c47d4f1fdfc45
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/55188
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

ctf: Remove AutoCloseable from CtfTmfTrace

Implementations of TmfTrace should not be AutoCloseable,
since they are exposed by other objects through getters.

Change-Id: Ibdb9b056802262cb3e0d67b6f5d3434f79c7face
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/51542
Reviewed-by: Hudson CI

pcap: Remove AutoCloseable from PcapTrace

Change-Id: Ide58410feef5b85bbe7c2c2a274b200bf614045b
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/51543
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

lttng: Open and close trace only once in ImportAndReadKernelSmokeTest

Change-Id: Ieaae2b0871fcbb5ff7009b84cca678d83806d4ac
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/51541
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

doc: Update user guide for split aspect combo in filter viewer

Change-Id: I09158bbce40217e0211bbd4326836c52785243c0
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/55345
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

tmf: Bug 476481: NullTimeEvent should not affect time graph entry bounds

This fixes the process entry's Birth time in the Control Flow view.

Change-Id: If0cb2aa0c644d930720fd4e95b074e3edb748cd8
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/55232
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

ctf: Add error logging when a closed iterator is selected for reattribution

In replaceRandomElement, a check has been added to create a new
iterator when encountering a closed one. This situation should
never happen, and so means that something is broken somewhere
else, e.g. that an ITmfContext hasn't been disposed properly.
This patch adds an error message log when that case is met to
allow further investigation and correction of the origin of
the problem.

Change-Id: Ib35f9906045f21042fbc9fea216a0e5da81221d3
Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Reviewed-on: https://git.eclipse.org/r/55242
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>

ss: Add proper locking to the AttributeTree

Instead of using a Collections.synchronizedList, extend the locking
the whole object. This makes sure that data stays consistent in the
longer methods like getQuarkAndAdd()/getQuarkDontAdd().

Bug: 476474.

Change-Id: Ia57f77ad109dd8c0b90e1824ba04d3a57140c1b6
Signed-off-by: Alexandre Montplaisir <alexmonthy@voxpopuli.im>
Reviewed-on: https://git.eclipse.org/r/55156
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI

tmf: Split aspect combo in filter viewer

Instead of a single aspect combo containing all trace types and their
respective event aspects, use a first combo for the trace type and a
second combo for the event aspects corresponding to the selected trace
type in the first combo.

Change-Id: I9e6189590d701ac8fea048a59e16d064491ccd01
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/54941
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>