Skip to content
This repository was archived by the owner on Aug 5, 2022. It is now read-only.

Conversation

@pafonso
Copy link
Contributor

@pafonso pafonso commented Feb 18, 2015

No description provided.

clero and others added 16 commits January 29, 2015 11:31
BZ: 213233

The setCriterionState command of the Test Platform was
setting undesired value when asking for multiple states
to an InclusiveCriterion.
The method used in setCriterionStateByLexicalSpace was
to parse arguments before to get numerical value for
each of them with the getNumericalValue method from
ISelectionCriterionTypeInterface. Then, different
masks were aggregated to set the criterion's state.
Regarding the getNumericalValue method behaviour which
parses automatically value arguments, this method leads
to undesired bytes in the mask which leads to undesired
criterion's state.

The solution is to aggregate the setCriterionState
arguments when more than one arg are used.
As getNumericalValue use | to split his first parameter,
the code has been modified to use this behaviour.
getNumericalValue is now called only one
time and the returned mask is directly used to set
criterion's state.

Change-Id: Ie3a1d79a6f66928a2418f11233ff0d1ef5c3e900
Signed-off-by: Jules Clero <jules.clero@intel.com>
BZ: 213233

TestPlatform.cpp has been edited regarding uncrustify
suggestions.

Change-Id: Iac75c5923d5bf9836730ca6f21789967a98895e9
Signed-off-by: Jules Clero <jules.clero@intel.com>
BZ: 207091

The parameter platform manager used to initialize the element tree.
However, this is too early for plugins, since they need information from
the core which have not been initialized yet.

This patch delays the initialization of the element tree by adding it to
the load() function of the ParameterMgr.

Change-Id: I98e0e3a20ac0af2736eddc4c8349911eb1d1a691
Signed-off-by: Guillaume Denneulin <guillaume.denneulin@intel.com>
Signed-off-by: Mattijs Korpershoek <mattijsx.korpershoek@intel.com>
Reviewed-on: https://android.intel.com/215426
Reviewed-by: De Chivre, Renaud <renaud.de.chivre@intel.com>
Reviewed-by: Benavoli, Patrick <patrick.benavoli@intel.com>
Reviewed-by: Rocard, KevinX <kevinx.rocard@intel.com>
Tested-by: Rocard, KevinX <kevinx.rocard@intel.com>
BZ: 208504

CParameterAccessContext::outRawFormatIsHex() was missing the const modifier.

Change-Id: I3120a231d2dff8f4274b95cd2d3c8ff0817bf8ac
Signed-off-by: Patrick Benavoli <patrick.benavoli@intel.com>
BZ: 208504

Some source files were mistakenly assigned the executable flag.

Change-Id: Iad85123c331d2599126ed46f070fe10614f7c360
Signed-off-by: Patrick Benavoli <patrick.benavoli@intel.com>
BZ: 208504

Concerns the handle to remote processor library.
In PFW coding style, the variables are named like in english language:
The actual nature of the object needs to lie at the end of the name.

Change-Id: If41218c3456d07c81a953182488c6ae71e31b6d8
Signed-off-by: Patrick Benavoli <patrick.benavoli@intel.com>
BZ: 208504

Before this patch there was a crash in CElement destructor in Ubuntu upon
destuction of parameter framework instance.
The crash was caused by the attempt to call unmapped code: The subsystem
plugin libraries were getting unloaded before the execution of the subsystem
destructors.
Note:
In Android the issue is not seen since library unload is not implemented.

This patch ensures SystemClass destructor destroys the subsystems explicitely
unloading the plugins.

Change-Id: I19dad262b384bdbd63c7c319a41a5d547d0e75e9
Signed-off-by: Patrick Benavoli <patrick.benavoli@intel.com>
BZ: 209937

The "listCriteria" command returns the list of criteria in a human readable
way. However, external tools that require to know this list of criteria have
trouble parsing this output. An XML output would solve the issue.

Changed "listCriteria" command to accept "XML" as first and only argument.
In the end, listCriteria command accepts the 3 following forms:
- listCriteria => will list the criteria states and type content in a human
readable format
- listCriteria csv|CSV => will list the criteria states and type content in a
CSV format
- listCriteria xml|XML => will list the criteria states and type content in an
XML format

Removed python binding accordingly.

Change-Id: Ib060ec0a5d1ff87ba6c25caf4e0d5839a7927715
Signed-off-by: Patrick Benavoli <patrick.benavoli@intel.com>
Signed-off-by: Sebastien Gonzalve <sebastien.gonzalve@intel.com>
BZ: 209937

In case an error is encountered during the XML encoding operation,
no error message is created to log the issue.

This patch updates XmlStringDocSink.cpp::doProcess() so that the
error string is filled in case of errors.

Change-Id: Ie93e50f857506b8637b1bb0dd8d411a01f16ec49
Signed-off-by: Patrick Benavoli <patrick.benavoli@intel.com>
BZ: 209937

This patch solves 1 issue regarding XML export feature:
- ParameterMgr.cpp::exportElementToXMLString(): This method was receiving a
useless bValidateWithSchema argument. Indeed, elements in memory are
necessarily valid (unlike documents coming from external sources).

Change-Id: I55bf84c8ff53c37c9fc55e60f7f64e8776a43dc7
Signed-off-by: Patrick Benavoli <patrick.benavoli@intel.com>
From time to time, on Buildbot, the PFW generation build step fails
due to remote processor socket connection timeout.

In order to avoid deconnection when the peer is slow,
this patch removes the remote processor socket connection timeout.

Change-Id: I56b5b4b846a7489accd56ba3a3224dc1ffc1085e
Tracked-On: https://jira01.devtools.intel.com/browse/IMINAN-5511
Signed-off-by: Kevin Rocard <kevinx.rocard@intel.com>
Reviewed-on: https://android.intel.com:443/289012
During host domain generation, a port must be chosen before spawning a servers,
producing an error if someone else uses it before the server.

This race was workaround by retiring 10 times before failing.
This divided the failure probability by 10, but is no longer sufficient.

A correct implementation without this race is currently been developed
(without socket) but until then change the retry number from 10 to 1000.

Signed-off-by: Kevin Rocard <kevin.rocard@intel.com>
Tracked-On: https://jira01.devtools.intel.com/browse/IMINAN-17358
Change-Id: Ie80e4508952ef150eb999d9a6fa7e60b61825f35
Reviewed-on: https://android.intel.com:443/319856
Some parameters need to be converted to
logarithm.
This can be the case for volumes that come
from HAL as mere decimal values.

This patch adds the LogarithmAdaptationParameter
type.

Change-Id: Ic64232816a82c56a6ddffa5fdccda356eb4ec9c7
Tracked-On: https://jira01.devtools.intel.com/browse/IMINAN-16954
Signed-off-by: Piotr Diop <piotrx.diop@intel.com>
Reviewed-on: https://android.intel.com:443/320371
The directory where to generate output may not exists leading
generation to failure.

This patch adds the directory creation and thus makes sure generation
won't fail for this reason.

Change-Id: I1cee75d6d7921315dc3a928245499bde447c3639
Tracked-On: https://jira01.devtools.intel.com/browse/IMINAN-19676
Signed-off-by: Sebastien Gonzalve <sebastien.gonzalve@intel.com>
libremote-processor headers are needed by the parameter core.

This patch adds explicitily the dependency to this module.

Change-Id: I7a223a0abd35f9524a047ca92e1029c5cef38e38
Tracked-On: https://jira01.devtools.intel.com/browse/IMINAN-19676
Signed-off-by: Philippe Afonso <philippex.afonso@intel.com>
When refactoring Element::toXml, we forgot to make the necessary adaptation in
SystemClass::toXml. The latter does not need to set the Name attribute anymore
because this is taken care of by the base toXml method.

SystemClass::toXml now only calls the base and can therefore be completely
removed.

Change-Id: Ibd12feaac2c101dcc22e8716e332cb02f03ba99a
Tracked-On: https://jira01.devtools.intel.com/browse/IMINAN-19676
Signed-off-by: David Wagner <david.wagner@intel.com>
dawagner added a commit that referenced this pull request Feb 18, 2015
@dawagner dawagner merged commit 8a46614 into intel:master Feb 18, 2015
@krocard
Copy link
Contributor

krocard commented Mar 2, 2015

What is iminleg ? For people outside of Intel, this pull request is exceptionally obscure, no description, no commentary (internal reviewed), not even a nice branch name (like upstream_from_legacy_branch).
:-(
Edit: Just to be clear. I am not criticizing the patches but the way they were merged.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

7 participants