DESCRIPTION

tmf: Move the trace statistics to the analysis framework

By itself, this patch causes a bug where the Statistics view will not be
populated unless it is re-opened *after* the trace has been loaded.

The patch following this one fixes this (and unfortunately, that one
depends on changes from this one, so it wasn't trivial to re-order them).

Change-Id: I8b6c59d70e84fdc5971665e1e78d096e3abe30db
Signed-off-by: Braxton Mccarthy xxx@xxx.xxx


COMMENTS

author: Braxton Mccarthy
date: 2014-01-31 21:09:42.000000000

Uploaded patch set 3.

-------------------------------------
author: Brenden Conley
date: 2014-01-31 21:32:33.000000000

Patch Set 3:

Build Started https://hudson.eclipse.org/linuxtools/job/linuxtools-gerrit/5279/

-------------------------------------
author: Ivy Mitchell
date: 2014-01-31 21:34:56.000000000

Patch Set 3: Code-Review+1

(2 comments)

couple minor nits

Line:150, lttng/org.eclipse.linuxtools.tmf.core/src/org/eclipse/linuxtools/tmf/core/statistics/TmfStatisticsEventTypesModule.java -> commented code, should this go away?
Maybe it would be good to explain it if you want to stay?

Line:115, lttng/org.eclipse.linuxtools.tmf.ui/src/org/eclipse/linuxtools/tmf/ui/views/statistics/TmfStatisticsModule.java -> Should this be cancel? I know it's an API thing...

-------------------------------------
author: Ivy Mitchell
date: 2014-01-31 21:39:32.000000000

Patch Set 3: IP-Clean+1

We're in 2014... I think... up to your discretion.

-------------------------------------
author: Brenden Conley
date: 2014-01-31 22:12:47.000000000

Patch Set 3: Verified-1

Build Failed 

https://hudson.eclipse.org/linuxtools/job/linuxtools-gerrit/5279/ : FAILURE

-------------------------------------
author: Brenden Conley
date: 2014-01-31 23:50:00.000000000

Patch Set 3: -Verified

Build Started https://hudson.eclipse.org/linuxtools/job/linuxtools-gerrit/5285/

-------------------------------------
author: Brenden Conley
date: 2014-02-01 00:37:46.000000000

Patch Set 3: Verified+1

Build Successful 

https://hudson.eclipse.org/linuxtools/job/linuxtools-gerrit/5285/ : SUCCESS

-------------------------------------
author: Braxton Mccarthy
date: 2014-02-03 20:16:50.000000000

Patch Set 3: Code-Review+2

(2 comments)

Line:150, lttng/org.eclipse.linuxtools.tmf.core/src/org/eclipse/linuxtools/tmf/core/statistics/TmfStatisticsEventTypesModule.java -> Yeah these comments have been around for a long time. They come from LttngKernelStateProvider (or whatever its ancestor was called). If we ever want to add per process or per CPU statistics we could just uncomment these, but at the cost of a bigger state history.

I think the plan was to move the "per event type" statistics to the partial history, eventually, so once we do that we could enable those additional statistics.

Line:115, lttng/org.eclipse.linuxtools.tmf.ui/src/org/eclipse/linuxtools/tmf/ui/views/statistics/TmfStatisticsModule.java -> I agree cancel() would make a lot more sense. But that's part of the IAnalysisModuleAPI, and not really related to this patch. But that reminds me to send another patch for that!

-------------------------------------
