DESCRIPTION

[LTTng] Add user-space trace support

	This is useful for people to derive their own domain in UST.

Change-Id: I07aebf69b3329aa37e08a2f7b23d210ab81b9e47
Signed-off-by: Ivy Mitchell xxx@xxx.xxx


COMMENTS

author: Brenden Conley
date: 2013-02-15 16:33:29.000000000

Patch Set 1:

Build Started https://hudson.eclipse.org/sandbox/job/linuxtools/1382/ 

-------------------------------------
author: Brenden Conley
date: 2013-02-15 16:51:39.000000000

Patch Set 1: Verified

Build Successful 

https://hudson.eclipse.org/sandbox/job/linuxtools/1382/ : SUCCESS

-------------------------------------
author: Braxton Mccarthy
date: 2013-02-15 19:18:34.000000000

Patch Set 1: I would prefer that you didn't submit this

(3 inline comments)



Line:9, /COMMIT_MSG -> left-indented

Line:14, lttng/org.eclipse.linuxtools.lttng2.core/META-INF/MANIFEST.MF -> It shouldn't need to depend on ctf.core directly. It only extends ctfadaptor stuff.

Line:26, lttng/org.eclipse.linuxtools.lttng2.core/src/org/eclipse/linuxtools/lttng2/core/trace/LTTngUstTrace.java -> As we discussed, this could be a LTTngTrace type, that works for any LTTng 2.0 trace. Then the LTTngKernelTrace could extend it, and we could ship an example of a trace type for a specific UST application.

Also, shouldn't the concept of CPU be moved from CtfTmfTrace to here?

-------------------------------------
author: Ivy Mitchell
date: 2013-04-12 13:18:34.000000000

Patch Set 1: (1 inline comment)



Line:26, lttng/org.eclipse.linuxtools.lttng2.core/src/org/eclipse/linuxtools/lttng2/core/trace/LTTngUstTrace.java -> The arch I would prefer is 
                        CTFTrace 
kernel  --------------A        A---- User space

People can extend the user space but really userspace is more intuitive IMO than lttng and I don't see a design advantage to do a lttng abstract tracetype, it would be +- empty. I think. 

CPU is in the events, not the trace types, that would be another patch.

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