Currently, CDT does not ask GDB for the value of the variables it
believes to be of complex types, such as structures. However, as
described in the bug, it assumes that a typedefed pointer is a complex
structure. Because of that, it displays a value of "{...}" for it
instead of the pointer value.
By asking GDB for the value of the variable even if it's of a complex
type, CDT will always display the right thing for the value. This will
cause a few more -var-evaluate-expression calls, but their number is
still limited to what is visible in the variables view. So the impact
should be negligible.
Screenshot of before/after:
http://i.imgur.com/Sx5ZPfO.png
It is still impossible to edit the value, but this would be addressed in
another patch.
Change-Id: I92e6ead6351677e098a56d4af5bdb85fc61df080
Signed-off-by: Simon Marchi <simon.marchi@polymtl.ca>
The try/catch doesn't add anything. If getStackTrace throws, the test
will fail and will show the useful stack trace.
Change-Id: Icf453ff2ca6c10a32f264fece0f0db9af8049612
Signed-off-by: Simon Marchi <simon.marchi@polymtl.ca>
Presently, the field from MI are parsed in MIVar, then copied to
MIVariableManager.MIVariableObject, and then to ExprMetaGetVarInfo.
As result, although MI allows additional fields, a custom IDE cannot
access them, even though all of the above classes are in dsf.gdb
package.
This package firstly eliminates some code duplication, and then make
raw MITuple be propagated along the chain.
Change-Id: I57436fef65aaaf4805422df34466d10f2a205920
getAdaptable() in 4.5 M6 has changed to use generics, making casting its
result no longer necessary.
Change-Id: Idd943216a925576bb72c784b2c5a2a4b8e8b00d2
Signed-off-by: Marc Khouzam <marc.khouzam@ericsson.com>
Visualizer
This commit migrates the MulticoreVisualizer class to use the standard
DSF programming paradigm, which is to use RequestMonitors. It also
renames some methods to better fit in that programming model.
Change-Id: Ie00a5dcb810fb13a4ac526ff221cc10bd52a3fca
Signed-off-by: Marc Khouzam <marc.khouzam@ericsson.com>
I used this command to print a summary of the updates:
mvn versions:display-plugin-updates -Djava.net.useSystemProxies | grep "\\->" | sort | uniq
The command reports which maven plugin versions are outdated per cdt project (pom.xml). The extra grep and sort are just to make it display as a short summary.
Change-Id: I6abec0c1d510aac4599fb0d573d443807ee84a7b
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
isEmpty is garantied to be as fast or faster than size and
length. It is also easier to read.
Change-Id: Ia768910c21313666aeab774963c0b6e82d825b44
Signed-off-by: Matthew Khouzam <matthew.khouzam@ericsson.com>
I got a feedback that the error message when passing an unsupported
version should be more helpful.
Change-Id: I3c925c500d6de787ca9bfce421eda6109e9f6415
Signed-off-by: Simon Marchi <simon.marchi@polymtl.ca>
MIMemoryTest.writeCharVarying{Address,Offset}
When we switched SyncUtil.readMemory to use queries, it exposed a race
condition with the asserts that check for the number of event received.
Basically, nothing guarantees that when the assert is evaluated the
event had the time to arrive.
We now use a ServiceEventWaitor to wait until the event arrives (with a
timeout, of course).
At the same time, rename word_size to wordSize.
Change-Id: I6f4b51e22f46625e20bbbdbac91cf70cfd864e5e
Signed-off-by: Simon Marchi <simon.marchi@polymtl.ca>
