Matthew Khouzam [Fri, 8 Apr 2016 14:25:19 +0000 (10:25 -0400)]
tmf.core: simplify timestamp implementations
This patch makes BIG_BANG and BIG_CRUNCH and ZERO special singleton
objects. They now normalize to themselves. Big bang and big crunch are
always the largest or smallest values.
It also internalizes TmfRealTimestamp inside TmfTimestamp as
TmfTimestamp$Impl.
The timestamp comparison is very streamlined. The performance
should be similar though.
One bug found that will never occur in the real world:
additions with -Long.MIN_VALUE gave errors as -Long.MIN_VALUE ==
LONG.MIN_VALUE.
Change-Id: I284ed1b8f0dd1f6d5299e172dd07128d53ac5945
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70245
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Bernd Hufmann [Thu, 14 Apr 2016 15:45:27 +0000 (11:45 -0400)]
tmf: Bug 491714: Fix population of XmlTimeGraphView for experiments
With this patch the XmlTimeGraphView is populated with the content of
all traces part of an experiment.
When rebuilding the event list the view cache (event list, filters,
view context) is cleared. This is important because right now
there is only one instance of XmlTimeGraphView per trace even if there
are more than one XML analyses and views defined.
Change-Id: I535cace0089147bea4584c5af8b7a0c57157bd36
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70681
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Marc-Andre Laperle [Mon, 18 Apr 2016 17:43:41 +0000 (13:43 -0400)]
lttng: Fix not being able to get available context with LTTng 2.8
'lttng add-context -h' was replaced with 'lttng add-context --list' in LTTng 2.8
Bug: 491933
Change-Id: If277b86410a5d19cfa53531b03dfb5b7350c4c5b
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70894
Reviewed-by: Hudson CI
Bernd Hufmann [Wed, 13 Apr 2016 01:40:10 +0000 (21:40 -0400)]
tmf: Add/update '-', '+' and '*' key handling in TimeGraphViewer
The handling of '-'and '+' key have been updated when the mouse is over
the namespace. Before they were used for time zooming regardless of the
mouse position. The handling of '*' key has been added when the mouse
is over the namespace.
When the mouse is over the namespace, pressing the
- key '-' will collapse recursively selected entry
- key '+' will expand selected entry
- key '*' will expand selected entry to the level with at least one
collapsed entry
This patch is to align the keystroke handling to the TimeGraphCombo.
Change-Id: I964168bd1aa388430775055a39f625beaffb5985
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70583
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Bernd Hufmann [Mon, 18 Apr 2016 17:59:52 +0000 (13:59 -0400)]
tmf: fix conflicting find handler message in time graph views
Change-Id: Icf6ff1772acad9b86278e507b0cf987fba859857
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70896
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Marc-Andre Laperle [Mon, 18 Apr 2016 19:37:06 +0000 (15:37 -0400)]
lttng: Fix filters not shown with LTTng 2.8
In LTTng 2.8, a new MI field was added, filter_expression.
This needs to be handled by Trace Compass.
Bug: 491941
Change-Id: If1e796ca24d5542422f61f63c91cb4c37d702edd
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70905
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Marc-Andre Laperle [Fri, 8 Apr 2016 03:21:29 +0000 (23:21 -0400)]
lttng: Make schema validation optional
With this change, the schema validation is disabled by default except
for tests. The validation can be enabled easily as a Tracing option
in the launch configuration. To be able to enable this by default for
tests, it is also exposed in a system variable which could also be
used on the command-line.
Change-Id: I492f2877a7e066e14176a209159da17cea83c3f1
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70199
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Bernd Hufmann [Thu, 14 Apr 2016 01:46:05 +0000 (21:46 -0400)]
tmf: Update setting of minimum size of dialog box
Change-Id: I6a5e722c5a435ea8d34eafe8312a429c710f0503
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70606
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-by: Hudson CI
Alexandre Montplaisir [Tue, 5 Apr 2016 22:19:51 +0000 (18:19 -0400)]
lttng.ust: Retrieve the function name when calling addr2line
Add the -f flag to also query function/symbol names. This will
allow us to populate the "function name" field of the generated
callsite objects, which in turn will populate the Callstack View
appropriately.
Also pass the -C flag, which demangles C++ function names, without
apparantly affecting native names.
Change-Id: Ife6e8581347ab2f94558e258dc350a4c1b04c3fa
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/69971
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Alexandre Montplaisir [Sat, 9 Apr 2016 22:36:32 +0000 (18:36 -0400)]
tmf: Update BasicSymbolPrefPage dialog description
"provider" is an internal implementation concept the user
won't know about.
Change-Id: I3be1167f32183f4ad78d4e2135fda50466d3a024
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/70315
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Alexandre Montplaisir [Mon, 4 Apr 2016 19:49:43 +0000 (15:49 -0400)]
tmf: Rework TmfTraceUtils#resolveIntEventAspectOfClassForEvent
The generic type bound now ensures the aspect class passsed to
resolveIntEventAspectOfClassForEvent() really resolves to an
Integer, which means we can skip the check.
Also rework both aspect-finding methods to use flat Stream operations.
Change-Id: I8a341021699a69ac21b7a147c9ea91d58619fa99
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/69856
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Marc-Andre Laperle [Fri, 8 Apr 2016 20:40:04 +0000 (16:40 -0400)]
Upgrade to Tycho 0.25.0
See https://wiki.eclipse.org/Tycho/Release_Notes/0.25
Change-Id: I309392787e2f1c8b041e4c83648b4a096e8032e4
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70296
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Matthew Khouzam [Sun, 28 Feb 2016 14:59:03 +0000 (09:59 -0500)]
tmf.core: Introduce TmfTimestamp factory methods
Often times, a TmfTimestamp will be created using a scale of -9 or 0.
The result is a less efficient timestamp that takes more memory than
the nano or simple timestamp and is slower to compare. This patch makes
timestamps created always return the more efficient type.
At the moment, the factory will either create a new TmfNanoTimestamp if the
scale is NANOSECOND_SCALE, TmfSecondTimestamp if the scale is SECOND_SCALE,
or a plain old TmfTimestamp (TmfRealTimestamp).
This patch adds the following helper methods:
* create(value, scale) - create a timestamp with an arbitrary scale and value
* fromNanos(value) - create a nanosecond based timestamp (more efficient)
* fromMicros(value) - create a microsecond based timestamp
* fromMillis(value) - create a millisecond based timestamp
* fromSeconds(value) - create a second based timestamp (more efficient)
Also, TmfTimestamp is converted to an abstract class with "getScale()" and
"getValue()" needing to be implemented. This allows a timestamp like
TmfNanoTimestamp or TmfSecondTimestamp to not store an extra integer and
save 33% of their sizes.
Change-Id: I0996c9b4f626a70436b58e82446ed69d2e50af98
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/65312
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-by: Hudson CI
Jean-Christian Kouame [Fri, 15 Apr 2016 17:45:57 +0000 (13:45 -0400)]
analysis: Add count column to latency statistics view
Change-Id: Id2db373022a6c1b78ddac2a6c7b580459c189f3a
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70895
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Loïc Prieur-Drevon [Mon, 18 Apr 2016 12:10:33 +0000 (08:10 -0400)]
tmf.ui: callstack - replaced multiple single SS queries by full query
CallStackView would do a single query for every thread,
at the start of the statesystem, or at the end of the
statesystem.
This patch replaces all single queries in buildEntryList
by two full queries and reads to the result of those queries
Change-Id: Id59b355fd84bfc0b8e30ef3f3f375f8ca0dbe634
Signed-off-by: Loïc Prieur-Drevon <loic.prieurdrevon@gmail.com>
Reviewed-on: https://git.eclipse.org/r/70854
Reviewed-by: Hudson CI
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Jean-Christian Kouame [Wed, 9 Mar 2016 19:25:39 +0000 (14:25 -0500)]
tmf : Add pattern analysis behavior
This patch defines the behavior of the pattern analysis state provider
and defines all the models needed.
Change-Id: I6561feb02b06627ef5059d777a25a8bde56a70be
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/65751
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Bernd Hufmann [Tue, 22 Mar 2016 11:12:48 +0000 (07:12 -0400)]
analysis: Bug 489573: Internalize Control Flow view implementation
Change-Id: I92e81e1b158cb5dbf1910f9d25b880482fdb3355
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/69007
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Bernd Hufmann [Fri, 11 Mar 2016 18:31:07 +0000 (13:31 -0500)]
lttng: Add analysis requirements for LTTng UST Call Stack analysis
With this the user gets notified when the call stack analysis cannot be
executed. The help text will give some details about this.
Change-Id: Ib3628606a74ae5e83b60fde81720f7035d7eae86
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/68235
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Reviewed-by: Hudson CI
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Bernd Hufmann [Fri, 11 Mar 2016 18:27:04 +0000 (13:27 -0500)]
tmf: add events field analysis requirement class and test
This class allows to specify required events fields for one given event
type.
This class is useful for CTF, where optional event contexts are provided
as event fields.
Moreover, this will be useful for custom txt and xml trace types, where
for each event the event type is the same but the event fields might
be different.
Change-Id: I445976ea7b93ee95be887504bc223912c2cab298
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/68233
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Alexandre Montplaisir [Wed, 30 Mar 2016 23:31:45 +0000 (19:31 -0400)]
tmf: Add generics to ITmfEventAspect
The generic type can allow us to define the type of
the resolve() method's return value.
Change-Id: Id9bb9ddaf143fa6b3ed09bf32eab98e1718e7ee3
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/69569
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Geneviève Bastien [Tue, 8 Mar 2016 20:41:40 +0000 (15:41 -0500)]
analysis.os: Add layouted requirements to the CPU analysis
That patch attempts to bring back the concept of requirements to linux kernel
analyses. The requirements were disabled since the event layouts appeared.
Change-Id: I12b4befa69c94ca7205ea0db7670ee126788dc42
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Reviewed-on: https://git.eclipse.org/r/68015
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Matthew Khouzam [Fri, 15 Apr 2016 17:27:23 +0000 (13:27 -0400)]
analysis.linux.ui: Make disk io display read before write
This is more inline with the typical views of hard disk usage.
An added bonus is that reads are blue and writes are red. It is better
to avoid red/green for accessibility reasons, and typically red == write.
Change-Id: I7397fdd72539646ba2fbe79f8258580bb45908d0
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70779
Reviewed-by: Hudson CI
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Matthew Khouzam [Fri, 15 Apr 2016 17:18:57 +0000 (13:18 -0400)]
lttng.ui: move UST memory legend to left
This maximizes readability.
Change-Id: I36c08a5f16477e1c60752a2efd7b2c4698dc5589
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70778
Reviewed-by: Hudson CI
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Matthew Khouzam [Fri, 15 Apr 2016 17:06:23 +0000 (13:06 -0400)]
tmf.ui: make CommonXLineChartViewer hide the X axis label by default.
The X axis label is always "Time" for the common line chart viewer, this
patch saves some vertical space by not displaying it.
The legend is moved by default to the bottom, this saves a lot of space
due to time alignment.
Change-Id: I965290de363d2196140e03e60f3abe69fe35efb0
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70776
Reviewed-by: Hudson CI
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Reviewed-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Tested-by: Bernd Hufmann <bernd.hufmann@ericsson.com>
Bernd Hufmann [Wed, 9 Mar 2016 16:18:16 +0000 (11:18 -0500)]
tmf: Fix NPE in XmlPresentationProvider for undefined states
A NullPointerException was caused when creating a state tooltip for
state values that are not defined in the XML definition.
This patch avoids the NPE in this case. The state tooltip won't be
shown when this happens.
Change-Id: I07ca75d7bb39d3479c225b4b144e1acc119eae3b
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/68072
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Loïc Prieur-Drevon [Fri, 15 Apr 2016 14:41:05 +0000 (10:41 -0400)]
os.linux.ui: Remove redundant query in ResourcesPresentationProvider
getEventHoverToolTipInfo would do a full state query then
a single query for each CPU/IRQ.
This patch replaces the single query by reading the full
state results.
Change-Id: I10959152e094674b8276608dddb2fe03368f1d96
Signed-off-by: Loïc Prieur-Drevon <loic.prieurdrevon@gmail.com>
Reviewed-on: https://git.eclipse.org/r/70765
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-by: Hudson CI
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Marc-Andre Laperle [Mon, 18 Apr 2016 03:56:58 +0000 (23:56 -0400)]
releng: Fix Mylyn update site in targets
Mylyn 3.19 was release and no new milestones has been produced so
we can use that release until (if) there is a new one for Neon.
Change-Id: I150eb050bcbb576ac2deda102451e02575c78e66
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70830
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-by: Hudson CI
Matthew Khouzam [Wed, 6 Apr 2016 18:05:07 +0000 (14:05 -0400)]
tmf: Make Analysis Requirements implement Predicates
This allows us to use requirements as filters.
Change-Id: Ib103e7fec613455e140d5c63c24e7d2f53248d9c
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70049
Reviewed-by: Hudson CI
Michael Jeanson [Thu, 14 Apr 2016 19:11:37 +0000 (15:11 -0400)]
tmf.common: Fix boundaries in data size formatter
This patch fixes the display of units on boundaries,for example
"1024" will be displayed as "1 KB" instead of "1024 B".
Change-Id: I3464d1d54e05254f3133d76faf2b75890690b766
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Reviewed-on: https://git.eclipse.org/r/70698
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Reviewed-by: Hudson CI
Geneviève Bastien [Tue, 8 Mar 2016 20:40:10 +0000 (15:40 -0500)]
tmf: Add an event names requirement
It is preferable to use inheritance instead of String-typed values.
Change-Id: I62eaa0f7a01ac459ed06c471379f8f20bcd9b4ef
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Reviewed-on: https://git.eclipse.org/r/68014
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Matthew Khouzam [Wed, 13 Apr 2016 18:25:18 +0000 (14:25 -0400)]
enable new performance benchmarks.
Change-Id: Iae2f6ec13f60fd5a203c08c1eeed63866a0ac74f
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70584
Reviewed-by: Hudson CI
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Matthew Khouzam [Wed, 13 Apr 2016 16:36:36 +0000 (12:36 -0400)]
linux: tid analysis benchmark
This benchmark is nearly identical to the kernel usage benchmark, but it uses
the tid analysis. It should be interesting to compare and contrast the two.
Change-Id: I6db6ce5142c518e2d79959a4724db56ee1401348
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70578
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Reviewed-by: Hudson CI
Geneviève Bastien [Thu, 25 Feb 2016 03:48:44 +0000 (22:48 -0500)]
analysis.os: Benchmark some typical usages of Kernel Analysis
This adds benchmark of getting the thread running on a CPU from a kernel
analysis and making full queries on that analysis.
It paves the way to adding some more of those benchmarks.
Change-Id: I853a350a7647973b87e49dabaf2ac14b09cd0136
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Reviewed-on: https://git.eclipse.org/r/67444
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Alexandre Montplaisir [Thu, 14 Apr 2016 21:24:57 +0000 (17:24 -0400)]
Revert "ss: Add serialization logic to state values"
This reverts commit
ce148788289534e20b56d0790a8ad93ce2e6473d.
Change-Id: I81e3a1f3557ef8edceaf930ba9b8d8cb3b23f200
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/70711
Reviewed-by: Hudson CI
Michael Jeanson [Thu, 14 Apr 2016 19:29:44 +0000 (15:29 -0400)]
Remove MaxPermSize from all launchers
The Permanent Generation space was completely removed from Java 8,
using the MaxPermSize parameter only prints a warning and has no
effect on the JVM.
Change-Id: I1d378c0c09a9d8a2861e0d39bfe3bbdf37df0bad
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Reviewed-on: https://git.eclipse.org/r/70700
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Jean-Christian Kouame [Fri, 8 Apr 2016 19:20:45 +0000 (15:20 -0400)]
tmf : Move preconditions at the same level as states in the XML file
Change-Id: Ib4a6c59d841831d4b91298970cda039272149e04
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70292
Reviewed-by: Hudson CI
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Raphaël Beamonte [Thu, 14 Apr 2016 16:27:23 +0000 (12:27 -0400)]
os.linux: Bump KernelStateProvider version number
Previous commits changed the content generated by the
KernelStateProvider for process' priorities and exec
names. Bumping the version number allows to take advantage
of those changes without manually removing the history
tree.
Change-Id: Iad483b0f3c51fb53242264948f52e98709ebb1bc
Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Reviewed-on: https://git.eclipse.org/r/70688
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Tested-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-by: Hudson CI
Jean-Christian Kouame [Mon, 11 Apr 2016 17:37:11 +0000 (13:37 -0400)]
doc : Add user guide for time graph Find dialog
Change-Id: I27f17a27373270ccbe7af302dd85308d14a8f8f5
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70409
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>
Raphaël Beamonte [Tue, 12 Apr 2016 20:34:44 +0000 (16:34 -0400)]
SchedSwitchHandler: rename private methods
The setNewProcessPrio and setNewProcessExecName methods
were respectively renamed setProcessPrio and
setProcessExecName as the 'New' part of the name can be
confusing with the 'next process' when handling a
sched_switch event. The parameters of those methods have
been renamed accordingly.
Change-Id: Id1c0f00f165e2d906cb7650d511b1bebc20d4b1e
Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Reviewed-on: https://git.eclipse.org/r/70514
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Tested-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Raphaël Beamonte [Tue, 12 Apr 2016 20:32:53 +0000 (16:32 -0400)]
SchedSwitchHandler: update the exec name of the former process
Update the SchedSwitchHandler class to update the exec name of
the former process as well as the one of the new one. This
allows to keep a 'sort of' up to date exec name in the absence
of other events updating it.
Change-Id: Ie0a1ebc5b64214150edb0b41b62ccaab4447b041
Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Reviewed-on: https://git.eclipse.org/r/70513
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Tested-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Raphaël Beamonte [Tue, 12 Apr 2016 17:48:33 +0000 (13:48 -0400)]
SchedSwitchHandler: update the priority of the former process
Update the SchedSwitchHandler class to update the priority of
the former process as well as the one of the new one. This
allows to keep a 'sort of' up to date priority in the absence
of other events updating it.
Change-Id: Ie0da99274b9c932a0784dbd737d3279f2cfa6165
Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Reviewed-on: https://git.eclipse.org/r/70504
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Tested-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Bernd Hufmann [Tue, 22 Mar 2016 11:29:09 +0000 (07:29 -0400)]
analysis: Bug 489573: Internalize Resources view implementation
Change-Id: I35fc6e359b2727b19ad77e5788bb9e7c0f1dfb51
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/69008
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Bernd Hufmann [Tue, 22 Mar 2016 11:37:39 +0000 (07:37 -0400)]
analysis: Bug 489573: Internalize CPU Usage view implementation
Change-Id: I45e930b51b718f0fd469096f968b8484965b0971
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/69009
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Patrick Tasse [Tue, 12 Apr 2016 21:42:22 +0000 (17:42 -0400)]
tmf: Bug 491548: Do not incrementally build full time graph event list
The full event list should be recomputed for the full range at every
iteration of the build thread. If built incrementally it can create too
many events in the full event list.
The build thread should not insert NullTimeEvent in the full event list.
Change-Id: I2523291197b2d93267b2aa799e6126ea74eea0f9
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/70519
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Patrick Tasse [Thu, 7 Apr 2016 22:13:55 +0000 (18:13 -0400)]
tmf: Update AbstractTimeGraphView and TimeGraphEntry API
AbstractTimeGraphView.buildEventList() is renamed to buildEntryList()
which is more representative of its implementation. The Javadoc of this
method and getEventList() is improved to better explain their use.
TimeGraphEntry methods setParent(), getParent(), getChildren() and
addChild(...) are changed to restrict the entries to TimeGraphEntry.
This avoids instanceof checks in AbstractTimeGraphView implementations,
and avoids unexpected behavior in case a child or parent entry is not of
type TimeGraphEntry.
Change-Id: Ibf1c3f3fad467ab7ff215fd2552bb3350f00ef7e
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/70185
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Marc-Andre Laperle [Wed, 13 Apr 2016 16:58:27 +0000 (12:58 -0400)]
tmf.test: Use SWTBotJunit4ClassRunner in every SWTBot test
Using the runner will produce screenshots if a test fails.
Change-Id: I6460bf1eb230ea2b3fe8cfd86ee8bc01399cb24f
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70579
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Marc-Andre Laperle [Tue, 12 Apr 2016 19:38:02 +0000 (15:38 -0400)]
lttng: Make ZoomThread more thread-safe (also helps ResourcesViewTest)
In AbstractTimeGraphView and extenders, when a ZoomThread applies its
results (links, markers), there is no guarantee of the order in the
presence of multiple ZoomThreads.
For example, in the case where a new ZoomThread is started and an old
one is still executing, the old one might be canceled so late that it
will still apply its results and do so *after* the new ZoomThread
completes. This change introduces some synchronization to make sure the
results are applied only in the case of the last ZoomThread spawned.
To reproduce this issue, go to AbstractTimeGraphView.getTraceMarkerList
and add a Thread.sleep(5000); before and after the isCanceled() check.
Because the test takes much longer, you need to increase the timeout in
ResourcesViewTest.timeGraphIsReadyCondition. Then you can run
ResourcesViewTest in a loop using SWTBotStressTests and after a few
iterations, it should fail.
Change-Id: If41f7b264676c04538cdc39c5c9bb40507b7e521
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70510
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Marc-Andre Laperle [Wed, 13 Apr 2016 17:31:02 +0000 (13:31 -0400)]
tmf: Increase timeout for indexing in TestRefreshTextTrace
The additional output added recently to this test confirmed that
during the test, the trace is being slowly indexed without being
stuck and ends up timing out. This change significantly increases
the timeout (20 secs to 5 mins) in order to allow more indexing
time when the machine running the test is under heavy load.
Change-Id: I8e76ced07a5a09b6bd14a7b11d12c925c1ce3df2
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70581
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Jean-Christian Kouame [Tue, 22 Mar 2016 19:15:40 +0000 (15:15 -0400)]
tmf : Add search dialog to timegraph views
The search dialog is based on the Eclipse FindReplaceDialog
implementation.
Change-Id: Id0758fc7e709d1d66778e0fc33d9f848b626cc70
Signed-off-by: Jean-Christian Kouame <jean-christian.kouame@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/69484
Tested-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>
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Raphaël Beamonte [Tue, 12 Apr 2016 17:46:32 +0000 (13:46 -0400)]
IKernelAnalysisEventLayout: add fieldPrevPrio() method
Adds a fieldPrevPrio() method in the layout to access the
priority of the previous process. Up to now, we could access
the previous process' name, but its priority can be useful.
Change-Id: I52c9ae1f3ba377139702e8cd4a6aad75cac7757c
Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Reviewed-on: https://git.eclipse.org/r/70503
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@efficios.com>
Patrick Tasse [Thu, 17 Mar 2016 15:18:57 +0000 (11:18 -0400)]
swtbot: Add SWTBot tests for time graph marker axis
Change-Id: I885d3b880bd875012b2b4f2a3b13b9f7e3a4553a
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/68682
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Marc-Andre Laperle [Mon, 11 Apr 2016 14:11:31 +0000 (10:11 -0400)]
Upgrade jboss repository-utils to 0.23.3
This is the latest release.
Change-Id: Iad05716cb3a4c3da07cba1a8022c54cbf3828e10
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70367
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Bernd Hufmann [Fri, 8 Apr 2016 17:44:20 +0000 (13:44 -0400)]
tmf: Bug 491346: fix time pinning of CallStackView for window ranges
Change-Id: If7668535ca2cf26f0bf1a499ca3a606f31527eaa
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70281
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Matthew Khouzam [Mon, 11 Apr 2016 15:07:26 +0000 (11:07 -0400)]
tmf.ui: Add Lambdas to TimeGraphCombo
Change-Id: Ic8e5ee349256bbc6a523b3453dddb49128323d2d
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70389
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Alexandre Montplaisir [Fri, 8 Apr 2016 21:41:23 +0000 (17:41 -0400)]
os.linux: Move buildThreadAttributeName() methods to Attributes interface
The method buildThreadAttributeName() and its counterpart
parseThreadAttributeName() are related to attributes tree
walking, and not to the event handlers themselves. Move them
to the Attributes interface.
Change-Id: Ide7b7154cfff0e12f74eda7a64f3545114b93e80
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/70298
Alexandre Montplaisir [Mon, 11 Apr 2016 21:06:24 +0000 (17:06 -0400)]
os.linux: Move Attributes class to internal package
The os.linux.kernel analysis should eventually encapsulate all
its state system accesses, which means clients should not call
the Attributes class directly. Move Attributes class to the
analysis's internal package so it gets removed from the API.
We can make some known users x-friends for now, but eventually
they should be moved to using the real API.
Change-Id: I0b78c8ff88b039311d89d6313d40c2e9381c29af
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/70424
Raphaël Beamonte [Tue, 12 Apr 2016 17:37:45 +0000 (13:37 -0400)]
Fix typo in method name: setNewProcessPio -> setNewProcessPrio
An obvious typo was introduced in that method whose function is to
update the new process priority.
Change-Id: Id0600f339426c147fb75f0f157769298c3b6dfe7
Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Reviewed-on: https://git.eclipse.org/r/70502
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Tested-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Bernd Hufmann [Fri, 8 Apr 2016 15:13:24 +0000 (11:13 -0400)]
tmf: add support for enter to expand/collapse tree TimeGraphCombo
This aligns the behavior of the TimeGraphViewer.
Change-Id: I211d69d72353622b380bbef88802d2745407b10f
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70257
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Marc-Andre Laperle [Tue, 12 Apr 2016 14:13:36 +0000 (10:13 -0400)]
tmf: Fix CopyToClipboardTest
The contextMenu method in SWTBot now throws (properly) a
WidgetNotFoundException instead of a TimeoutException
Change-Id: I9e7b8e469134ed7836923bbdd2d59ae9cf3eff59
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70478
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-by: Hudson CI
Bernd Hufmann [Thu, 7 Apr 2016 13:06:55 +0000 (09:06 -0400)]
tmf: Support for context-sensitive menu in Time Graph viewers
Up to now, there was a specific context-sensitive menu in the
Resources Flow view using the right mouse button on the namespace.
This commit adds a generic support for context-sensitive menus to the
AbstractTimeGraph views when using Time Graph viewers alone. It allows
extending classes to add both programmatically and declaratively
(using the plug-in.xml) menu entries to the context-sensitive menu.
The implementation of the Resources view have been adjusted to use this
new API
Change-Id: Ic9f3ff08fda27ffabc91d6054fe460b8fc764835
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70140
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Bernd Hufmann [Thu, 7 Apr 2016 12:34:21 +0000 (08:34 -0400)]
tmf: Fix ISelectionProvider implementation of TimeGraphControl
The TimeGraphControl class implements ISelectionProvider interface
put never notifies the registered as ISelectionChangedListener.
Change-Id: I71ebc17c215a5e1e487f9488bb0727ebef898d31
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70139
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Bernd Hufmann [Thu, 7 Apr 2016 12:42:38 +0000 (08:42 -0400)]
tmf: Fix mouse detect handling in TimeGraphControl
Prevent the menu be shown before mouse up event is received. Otherwise
the zoom won't work. This is because the mouse up event will be received
by the menu pop-up and not by the TimeGraphControl instance.
Change-Id: Iaa905d39205a4a45abb8fd5cfeaac81664561820
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70138
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>
Bernd Hufmann [Thu, 7 Apr 2016 12:25:23 +0000 (08:25 -0400)]
tmf: Support right mouse button selection of time graph entries
when clicking on the entry in the namespace.
Change-Id: I4561c4d5398300a8206348c4939f49cecd9c1c32
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70137
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Alexandre Montplaisir [Sun, 10 Apr 2016 22:06:37 +0000 (18:06 -0400)]
ss: Add serialization logic to state values
Right now the HTInterval object takes care of state value
serialization. This is a bad separation of responsibilities,
the state value itself should be the master of its destiny.
Implement state value serialization to and from byte arrays.
In a second step, the HT can be moved to use that logic instead
of its own.
Change-Id: I1d3379e0a4d7d82816164ae214838462ab28c5fd
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/70337
Reviewed-by: Hudson CI
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Matthew Khouzam [Mon, 11 Apr 2016 20:24:55 +0000 (16:24 -0400)]
lttng: Bump Bundle-Version in lttng.ust.ui
This fixes api errors introduced in https://git.eclipse.org/r/#/c/68013/
Change-Id: Ib254fff53c06a6bee38b0a72274c551a68b3e81a
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70419
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Reviewed-by: Hudson CI
Bernd Hufmann [Thu, 7 Apr 2016 12:10:47 +0000 (08:10 -0400)]
tmf: Support for context-sensitive menu in Time Graph combos
Up to now, there were specific context-sensitive menus in Call Stack
and Control Flow view using the right mouse button on the tree.
This commit adds a generic support for context-sensitive menus to the
AbstractTimeGaph views when using Time Graph combos. It allows
extending classes to add both programmatically and declaratively
(using the plug-in.xml) menu entries to the context-sensitive menu.
The implementation of the Call Stack view and Control Flow view have
been adjusted to use this new API.
Change-Id: Ic0f0971ac32767446d0f555315f6ecbf0aeba09a
Signed-off-by: Bernd Hufmann <Bernd.Hufmann@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70136
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Geneviève Bastien [Fri, 8 Apr 2016 18:14:42 +0000 (14:14 -0400)]
tmf.ui: Use the Y formatter to format tooltip values of XY charts
The XY viewers with a formatter defined for the Y axis will now use this
formatter for the tooltip.
Change-Id: Ibd2987199f8af50297b1a3c00b6f41e4c333316f
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Reviewed-on: https://git.eclipse.org/r/70286
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Geneviève Bastien [Fri, 8 Apr 2016 19:00:03 +0000 (15:00 -0400)]
lttng.ust: Use the data size formatter for UST memory view
Change-Id: Ie904826b7261b65671bff054b57f0764569f5087
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Reviewed-on: https://git.eclipse.org/r/70289
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Patrick Tasse [Mon, 4 Apr 2016 22:38:02 +0000 (18:38 -0400)]
tmf: Bug 490400: Leaking widgets due to incorrect cleanup in dispose()
Subclasses of Widget now perform their cleanup in a DisposeListener.
Disposing any child Control of a Composite is handled by the platform
and is therefore removed from the cleanup handling.
Subclasses of TmfComponent that are associated with a Control now
dispose their Control when TmfComponent.dispose() is called, and perform
their cleanup in a DisposeListener, including calling super.dispose().
The DisposeListener is attached to the Control.
Change-Id: Ie6bf7b18b8fad6a0fc0a683209a92fa39bbc8907
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/70057
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Matthew Khouzam [Fri, 8 Apr 2016 21:30:05 +0000 (17:30 -0400)]
tmf.ui: Lambda some handlers in CustomXmlParserInputWizardPage
This should make the code lighterweight.
Change-Id: I1f8034b31365c5c6e87341b7ba2186e30107992e
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70368
Reviewed-by: Hudson CI
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Tested-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Alexandre Montplaisir [Sat, 9 Apr 2016 22:27:11 +0000 (18:27 -0400)]
tmf: Fix opening symbol config dialog for experiments
An ArrayIndexOutOfBounds exception could be thrown when opening
the symbol mapping configuration dialog in the context of a
multi-trace experiment. Index "-1" would be passed to
updateMessage(int).
When the window is first opened, the selection index is at -1,
simply skip the updateMessage() in this case.
Change-Id: I43b384016ec7dfff1b16b272b34155579bc101cc
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/70314
Reviewed-by: Hudson CI
Reviewed-by: Robert Kiss <robert.kiss@gmail.com>
Alexandre Montplaisir [Sun, 10 Apr 2016 22:20:16 +0000 (18:20 -0400)]
common: Annotate ByteBuffer.array()
to make it return a "byte @NonNull []", which means that
the array reference cannot be null.
It returns the buffer's internal array, and the implementation
first checks if it is null, and throws a
UnsupportedOperationException if it is.
Change-Id: I91d8a90d11dbe30ba0895103fadba78ed4b35ecf
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/70336
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Matthew Khouzam [Mon, 11 Apr 2016 14:11:32 +0000 (10:11 -0400)]
ctf: Add comments to CTFClock
I was asked, is it deliberate that CTFClock's constructor is empty and
what is a clock, how does it work. This patch attempts to answer some of
the frequently asked questions.
Change-Id: I42dd8254f73df2d92d5e2497600b0b6ffc3682c4
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70369
Reviewed-by: Hudson CI
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Houssem Daoud [Tue, 29 Mar 2016 18:39:43 +0000 (14:39 -0400)]
analysis.io: Introduce an I/O Activity view
This view is linked to the Input/Output Analysis Module.
Change-Id: I4647540aee7ee9e78f949da9a1de0e01385e21cc
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Signed-off-by: Houssem Daoud <houssemmh@gmail.com>
Reviewed-on: https://git.eclipse.org/r/69670
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Patrick Tasse [Wed, 6 Apr 2016 15:08:54 +0000 (11:08 -0400)]
tmf: Workaround for leaking views
As a workaround for Bug 490400, clear the actions bars of views that
contribute to the tool bar manager so that their actions don't retain
references to the view when it is closed.
Change-Id: I3af262aad4416e40edf587f1313c8480517541b0
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/70059
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Marc-Andre Laperle [Sat, 9 Apr 2016 02:58:08 +0000 (22:58 -0400)]
Add some debug output to TestRefreshTextTrace
This test has been failing intermittently but my suspicion is that
it fails when the server is under heavy load and the trace fails to
index in 20 seconds. To verify this, this change adds some output
as the wait condition is evaluated. This test didn't already log the
SWTBot output so this means that even with this change, it's not too
verbose.
Change-Id: Ic6da7b19c45faa5345b2118c902101db6e18276b
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70329
Reviewed-by: Hudson CI
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Geneviève Bastien [Fri, 8 Apr 2016 17:55:56 +0000 (13:55 -0400)]
tmf.common: Add a data size and speed formatter
This allows to format numbers representing sizes and speeds in bytes[/s] to
the closest thousand with 3 decimals. Format also includes the units.
This patch updates the kernel memory usage view to use this formatter.
Change-Id: I637f9e408d75f6a3948169d6fde5e8da596b17ce
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Reviewed-on: https://git.eclipse.org/r/70287
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Geneviève Bastien [Wed, 30 Mar 2016 15:59:56 +0000 (11:59 -0400)]
tmf.ui: Return minimal number of X values in XY viewer
When the number of time steps is lower than the requested number of X values,
only those steps will be returned.
Change-Id: I94c877906083584546d4921e2bd90d82b5e72c49
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Reviewed-on: https://git.eclipse.org/r/69546
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Matthew Khouzam [Wed, 16 Mar 2016 18:56:58 +0000 (14:56 -0400)]
analysis.ui: Add follow cpu context menu to resources view
This generates a TmfCpuSelectectedSignal that can be used by views listening in.
Change-Id: Ib7a19b600dfd6be8510b07b4b48cc60e05722833
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/66651
Reviewed-by: Hudson CI
Matthew Khouzam [Tue, 8 Mar 2016 15:37:25 +0000 (10:37 -0500)]
linux.ui: Add support to show CPU usage per-cpu
This adds support to the CpuUsageView to make it update when a
TmfCpuSelectedSignal is received.
The title updates to show which cpus are selected.
Change-Id: Ia315159b4746b8f33f747da6b2f6327065a32995
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/65423
Reviewed-by: Hudson CI
Geneviève Bastien [Tue, 8 Mar 2016 20:39:24 +0000 (15:39 -0500)]
tmf: Move the analysis requirements to their own package
The o.e.t.tmf.core.analysis package is getting crowded.
Change-Id: I2b8cc1cd9719733b260117787fa2865e03633c60
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Reviewed-on: https://git.eclipse.org/r/68013
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Tested-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Marc-Andre Laperle [Thu, 7 Apr 2016 20:50:09 +0000 (16:50 -0400)]
rcptt: Hard-code RCPTT runner to specific milestone release
So that it doesn't redownload all the time (every day).
This version (2.1.0-M7_rc1) seems to support Neon correctly so this
should be good enough for a while until we need to consume other
fixes.
Change-Id: I0900b12706390510931d9caf62e7a3f095e9bf60
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70175
Reviewed-by: Hudson CI
Matthew Khouzam [Thu, 7 Apr 2016 20:52:42 +0000 (16:52 -0400)]
linux: make priority aspect work with tid == 0
Change-Id: I57353c2bef4e1ec0911e8d9d5c97cc954262ff1a
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70176
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Marc-Andre Laperle [Sun, 10 Apr 2016 18:22:29 +0000 (14:22 -0400)]
Increase timeouts for tests that have low values
Some tests have short timeouts (20 secs) that would usually not be
a problem on an performant machine but when build servers are under
load, especially in I/O, this can lead to intermittent failures.
This changes increases those 20 secs timeouts to 1 minute to help
in those situations.
Change-Id: I38b56c65d3981c4cf6490724f97aa26de7b0f736
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70330
Reviewed-by: Hudson CI
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Tested-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Alexandre Montplaisir [Tue, 5 Apr 2016 19:44:37 +0000 (15:44 -0400)]
tmf: Add the notions of timestamp and PID to ISymbolProvider
To support completely generic symbol resolution, the methods in
ISymbolProvider needs to consider two new concepts: the timestamp
of the query, and the process ID.
Process ID:
One trace can contain events from multiple processes. Even if
they all execute the same executable, the memory addresses, being
virtual, are specific to each process. The provider may need to know
the query is for which process in particular.
Timestamp:
Since libraries can be loaded and unloaded at runtime, with calls to
dlopen() and dlclose() for example, one address may refer to
different libraries at different moments in a trace. The provider
then needs to know the timestamp of the query, so it can refer to the
correct library if needed.
To ease the transition, we can introduce the new method as default
methods that will ignore the new parameters by default.
Change-Id: I6ec8aa5d97c690d84a9864af15f99a3ec9f4aa3d
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/69969
Reviewed-by: Hudson CI
Patrick Tasse [Fri, 8 Apr 2016 17:04:59 +0000 (13:04 -0400)]
releng: Add SWTBot Generator feature to target definition files
This allows developers to use the SWTBot Generator by adding its
plug-ins from the target platform in their launch configuration.
The SWTBot Generator can then be launched with File -> New -> Other...
-> Run Test Recorder.
Change-Id: I845377851f442babd9cb108a0bebac61d6b06eca
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/70279
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Patrick Tasse [Thu, 7 Apr 2016 21:15:23 +0000 (17:15 -0400)]
analysis: Remove zoomed event list handling from Critical Path View
The Critical Path View does not use a zoomed event list so there is no
need to fake it. Its event list is built by the HorizontalLinksVisitor,
and contains all time events regardless of the resolution.
The zoomed event list was built only once for the full entry range and
was therefore strictly equal to the full event list. It was not more
detailed (since the full event list is already at the most detailed
level) and it was not a subset of the event list for the current time
range, since the zoom thread never updated it, due to the
CriticalPathBaseEntry having no child entries, preventing any recursion
into the deeper entries.
Change-Id: I469ef9314dcbcc835a5ef03c93d9ca11c5f7eb70
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/70184
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Reviewed-by: Francis Giraldeau <francis.giraldeau@gmail.com>
Patrick Tasse [Tue, 5 Apr 2016 22:24:24 +0000 (18:24 -0400)]
tmf: Fix leaking static Histogram instances
The two Histogram instances in HistogramView are static and therefore
are retained by the class in the ClassLoader with references to the last
histogram data model, even after the view is closed.
Change-Id: I7853628e99a7bd8f3815253060bf891eed217cae
Signed-off-by: Patrick Tasse <patrick.tasse@gmail.com>
Reviewed-on: https://git.eclipse.org/r/70058
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Matthew Khouzam [Thu, 7 Apr 2016 20:19:12 +0000 (16:19 -0400)]
linux: make TID analysis handle no TID more gracefully - Bug 491275
This stops AttributNotFoundExceptions from being thrown in the beginning
of a trace.
Change-Id: I917fa81b3882ce3c98f23f782a87a6f7f8947b10
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70170
Reviewed-by: Hudson CI
Reviewed-by: Patrick Tasse <patrick.tasse@gmail.com>
Tested-by: Patrick Tasse <patrick.tasse@gmail.com>
Geneviève Bastien [Thu, 31 Mar 2016 18:45:47 +0000 (14:45 -0400)]
analysis.io: Provide information from the IO model
This extracts a base Disk class from the DiskWriteModel, that can be used
by outputs and views to get data from the analysis.
It also adds an Information Provider for the input output analysis module
It comes with its units tests
Change-Id: Iffb8428eb161bbd5cfcb834deebfa55d812f4d00
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Reviewed-on: https://git.eclipse.org/r/69669
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Matthew Khouzam [Thu, 7 Apr 2016 18:54:36 +0000 (14:54 -0400)]
linux: use DefaultEventLayout.getInstance() in TidAnalysis
Make the newly merged TidAnalsysis use the layout in the proper way.
Change-Id: Iff7d882946de3783dc23586c0aa1a468c61c17f5
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/70166
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-by: Hudson CI
Matthew Khouzam [Wed, 6 Apr 2016 00:20:11 +0000 (20:20 -0400)]
linux.core: Introduce TID analysis
This analysis calculates a subset of what the kernel
state system has: which tid is running on which cpu.
This will allow these aspects to be available much faster
during state construction and in the reading after.
The new state system is typically 1% of the size of the kernel
state system. Therefore the seeks, even though they are log(n)
will accelerate. As getTid is something MANY analyses do often
this patch should improve performance accross the board.
In one corner case example, (System call analysis), where each state
required a TID, performance passed from 3000 to 140000 events per seconds.
On average, searches on the TID column should reduce in time by 5-10%
depending on the trace configuration (how many contexts etc...).
Change-Id: Icf7921c49232481129b9c021170556c9d06d5ca4
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/68878
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Reviewed-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
Reviewed-by: Hudson CI
Alexandre Montplaisir [Mon, 4 Apr 2016 22:14:29 +0000 (18:14 -0400)]
os.linux: Do not define the default kernel layout in the interface
Instead point users directry to DefaultEventLayout.
This avoids a circular reference from IKernelEventLayout layout
to one of its own implementations.
Change-Id: I3c87767df4285c18b9216209942b069fef4d34cf
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/70080
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Matthew Khouzam [Wed, 6 Apr 2016 22:14:00 +0000 (18:14 -0400)]
os.linux: Do not rely on static initialisation order for layouts
Expose DefaultLayout#getInstance() instead of DefaultLayout#INSTANCE
This solves a weird problem where the layout could be null.
To reproduce this problem, define and run the following test suite:
@RunWith(Suite.class)
@Suite.SuiteClasses({
VirtualMachineAnalysisTest.class,
LttngKernelAnalysisTest.class,
PartialStateSystemTest.class,
StateSystemInMemoryTest.class,
StateSystemFullHistoryTest.class
})
public class TestSuite { }
Change-Id: I772ee0944334283d4b36118001a858322921d98a
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/69864
Reviewed-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Tested-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-by: Hudson CI
Alexandre Montplaisir [Mon, 4 Apr 2016 23:05:13 +0000 (19:05 -0400)]
lttng: Use long values for integer fields in the Graph test trace
Since state providers expect all integer fields to be Long objects
(as ctf.tmf does), test traces should also supply long values.
"cpu" should remain an int however.
Change-Id: Icee091a56710cd4a3c16d4f00ab00ca792b13d1e
Signed-off-by: Alexandre Montplaisir <alexmonthy@efficios.com>
Reviewed-on: https://git.eclipse.org/r/69866
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Matthew Khouzam [Sun, 20 Mar 2016 21:12:01 +0000 (17:12 -0400)]
graph: add messages to IllegalStateExceptions
Change-Id: If8efc90b6e55395448abbf3ced6695678dc8b496
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/68876
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Reviewed-by: Hudson CI
Geneviève Bastien [Mon, 21 Mar 2016 13:42:42 +0000 (09:42 -0400)]
analysis.io: Unit tests for the I/O analysis
It uses a few XML kernel stub traces to cover the various cases and events
of the IO analysis, then the content of the state system is verified at certain
times and for certain attributes.
Change-Id: I33f8deb62d9df6958fe88012ada3db5f077d0c09
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Reviewed-on: https://git.eclipse.org/r/69668
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Houssem Daoud [Mon, 21 Mar 2016 13:42:01 +0000 (09:42 -0400)]
analysis.io: Introduce the input/output linux analysis
This analysis uses the block_* events to track the requests to the disks
and the number of sectors read and written. It also stores the reads and
writes by process from the syscalls.
Change-Id: I6e1b00d53eb3a217b01a906d4d1fd566e53481ef
Signed-off-by: Geneviève Bastien <gbastien+lttng@versatic.net>
Signed-off-by: Houssem Daoud <houssemmh@gmail.com>
Reviewed-on: https://git.eclipse.org/r/69667
Reviewed-by: Hudson CI
Reviewed-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Tested-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Matthew Khouzam [Wed, 6 Apr 2016 01:23:13 +0000 (21:23 -0400)]
linux.core: Simplify KernelThreadInformationProvider
This makes KernelthreadInformationProvider only check pertinent state
values. The logic should be 100% equivalent, but the code should be
much more legible.
Change-Id: I96a962c6f2fa80b2a57d753a324c60c394dae349
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/69975
Reviewed-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Tested-by: Genevieve Bastien <gbastien+lttng@versatic.net>
Reviewed-by: Hudson CI
Matthew Khouzam [Thu, 28 Jan 2016 20:40:29 +0000 (15:40 -0500)]
ctf: mini bitbuffer optimisation
Change-Id: I10806685649ae1f46d679189e6d4f022a060060e
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
Reviewed-on: https://git.eclipse.org/r/65415
Reviewed-by: Hudson CI
Reviewed-by: Francis Giraldeau <francis.giraldeau@gmail.com>
This page took 0.047335 seconds and 5 git commands to generate.