LTTV & LTTng roadmap

Here are the roadmaps for the LTTV and LTTng development. I use a priority indice for the TODO items :
(1) : very high priority
(10): lowest priority


Dependencies are written between brackets [ ].
The # symbol marks who is currently working on the item.
The % symbol marks who is interested in the realisation of the item.

LTTV Roadmap

* TODO
(2) Finish pre-computing of states to a file.
(3) Add cluster support.
# Eric Clement
(3) Make LTTV aware of type formats (visual separators) defined in the XML file.
(3) Use a per architecture enumeration for traps.
(3) Change the byte pair "facility, event" id for a short combining the informatinon.
(4) Statistics per time window.
(4) Disable plugins when threshold reached (i.e. too much process in control flow view). Draw, and, when the threshold is reached, stop drawing. The global statistics view can inhibit showing the per process stats.
(4) Add a visual artifact : PID 0 could be named swapper instead of UNNAMED for cpus > 0.
(4) Add event specific fields support to filter.
(4) Add a periodic event interval view. (useful to verify event periodicity)
(4) create a graphical per cpu activity view.
(4) Filter by target process.
(4) Add CPU, network, disk, memory usage histogram. [Per interval statistics]
(5) Add Python scripting hooks.
(5) Flight recorder : start lttd automatically upon GUI trace control stop.
(5) Automatically detect traces with too much processes and disable faulty operations.
(5) Event sequence detector (inspired from regular expressions).
(7) Create a hardware counter viewer (low cost rate counters : L1 cache miss, page faults, interrupts...). This will be a generalisation of the event rate view into a view of the evolution of a user definable event field.

* TO FIX
(4) Fix scroll backward in detailed event list (introduced with "stop" functionnality.
(10) Add cancel button to LTTV filter GUI window.
(10) Sometimes, in the control flow view, a process with 0 creation time is created in addition to the real process itself. Seems to be caused by end of process life.
(10) Statistics do not take in account the time spent in the mode present at the beginning of the trace. Example : real time spent in system call on behalf of process 0.


LTT Next Generation Roadmap

* TODO
(3) Find a way to make logging know when it causes a minor page fault recursively (without ignoring silently all nested events).
(3) Have various subbuffer sizes for high and low rate channels.
(3) Change the byte pair "facility, event" id for a short combining the informatinon.
(3) Add sysenter instrumentation on x86. Verify x86_64.
(3) #IBM Fix ppc64 syscall time accounting. Maybe caused by a missing syscall return instrumentation ?
(3) Process and kernel stack dump for x86_64, for native and compat.
(4) Add Xen support.
# Mathieu Desnoyers
# Parisa Heidari
(4) efficient dynamic event filtering while recording trace.
% Sensis Corp. Tim Bish
(4) instrument kernel bottom half irqsave, spinlocks, rwlocks, seqlocks, semaphores, mutexes, brlock.
(4) In flight recorder mode, have the low traffic channels saved as normal tracefiles and high traffic channels being flight recorder.
(4) Try to use my own non LOCK prefixed version of atomic operations : we are using per-CPU variables, so it should make it possible.
(4) integrate NPTL instrumentation (see PTT).
(5) Support CPUs with scalable frequency.
(5) Integrate LTTng and lttd with LKCD.
(7) Integrate periodical dump of perfctr hardware counters.
(8) Integrate periodical dump of SystemTAP computed information.
(8) Integrate SystemTAP logging with LTTng.
(9) Add support for setjmp/longjmp and jump tables instrumentation to ltt-instrument-functions.
(9) Make ltt-usertrace-fast support internal heartbeat timer.
(3) port LTTng to :
x86_64
#Autodesk Francois L'Archeveque for 2.6.9 RedHat Enterprise (patch available in contrib, still has to be ported to 2.6.14, thanks!)
# Wind River for 2.6.14
PPC
PPC64
# Wind River for 2.6.14
alpha
user-mode Linux
Xen
# IBM
MIPS
MIPS64
% Wind River for 2.6.14
# Tim Bird Tim Bird
S/390
RTLinux
% Wind River for 2.6.14



* TODO (low priority)
enhance RPM packages for lttng kernel
Integrate header generation (genevent) in kernel build system.
Export channels via network sockets instead of writing them to disk.
Export buffers with time constraint for "live" visualisation. Use ltt_force_switch periodically from a timer to insure slow channels do not interfere with viewing.
Have an optional round-robin mode to write information into multiple channels from the same source.

* Need to be discussed
Use 2.6.14 RelayFS control files.
Drop ltt-module-register and ltt-module-unregister, use exported variables.
drop ltt_filter_control, use functions pointers instead.
Merge facilities headers into one big header.
Change the name of XML files from XML to something else.
Remove ltt-base.c.


* lttng patch division (0.5.10+) :

instrumentation.diff
instrumentation-arm.diff
instrumentation-i386.diff
instrumentation-mips.diff
instrumentation-powerpc.diff
instrumentation-ppc.diff
instrumentation-s390.diff
instrumentation-sh.diff
ltt-facilities-headers.diff
ltt-facilities-loader.diff
ltt-facilities.diff
ltt-relayfs.diff
ltt-core.diff
ltt-build.diff


Mathieu Desnoyers