doc/developer/lttng-lttv-roadmap.html

   1 <html>
   2 <body>
   3 <center><big><big>LTTV & LTTng roadmap<small><small></center>
   4 <br>
   5 <br>
   6 Here are the roadmaps for the LTTV and LTTng development. I use a priority indice
   7 for the TODO items :<br>
   8 (1) : very high priority<br>
   9 (10): lowest priority<br>
  10 <br>
  11 <br>
  12 Dependencies are written between brackets [ ].<br>
  13 The # symbol marks who is currently working on the item.<br>
  14 The % symbol marks who is interested in the realisation of the item.<br>
  15 <br>
  16 <br>
  17 <big>LTT Next Generation Roadmap<small><br>
  18 <br>
  19 * TODO (high priority)<br>
  20 <BR>
  21 (1) LTTng event description: move from tracepoint/markers to Ftrace TRACE_EVENT
  22  declarations. Extend TRACE_EVENT as needed. (<A HREF="mailto: Douglas Santos
  23 <douglas.santos@polymtl.ca">Douglas Santos</A>)<BR>
  24 (1) LTTng ring buffer adaptation for FTrace.<A HREF="mailto:
  25 mathieu.desnoyers@efficios.com">Mathieu Desnoyers</A>)<BR>
  26 (1) FTrace/LTTng trace format standardization.<BR>
  27 (1) Extend NO_HZ support vs trace streaming support to other architectures (need
  28 to add cpu idle notifiers and test).<br>
  29 (1) Make sure ltt-ascii kernel text dump fits well with streaming hooked into
  30 cpu idle.<br>
  31 (1) Support CPUs with scalable frequency with a time-consistent increment and
  32 with an approach scalable to SMP. (done for ARM OMAP3 UP only, but the OMAP3
  33 approach should be tested and probably derived into an SMP implementation)<br>
  34 <br>
  35 <br>
  36 * Nice to have<br>
  37 (3) Bring stack dump in sync with new lttng.<br>
  38 (4) Dump mounts. (to fix)<br>
  39 (4) Add Xen support. (Trace buffer desallocation needs to be fixed)<br>
  40 (4) integrate NPTL instrumentation (see
  41 <A HREF="http://nptltracetool.sourceforge.net/">PTT</A>).<br>
  42 (4) Probe calibration kernel module.<br>
  43 (4) Make page faults detect nested fault without nesting 4 times in the page
  44 fault handler. (or use vmalloc_sync_all at each tracing module load)<br>
  45 (5) Add boot time tracing support.<br>
  46 (5) Integrate LTTng and lttd with LKCD.<br>
  47         # <A HREF="mailto:Lai Jiangshan <laijs@cn.fujitsu.com>>">Lai Jiangshan</A><br>
  48 (7) Integrate periodical dump of perfctr hardware counters.<br>
  49 (8) Integrate SystemTAP logging with LTTng.<br>
  50 (8) Integrate periodical dump of SystemTAP computed information.<br>
  51 (9) Add support for setjmp/longjmp and jump tables instrumentation to
  52 ltt-instrument-functions.<br>
  53 <br>
  54 <br>
  55 <br>
  56 <big>LTTV Roadmap<small><br>
  57 <br>
  58 Note: new feature development is currently done in the Linux Tools Project:
  59 LTTng Integration. Mainwhile, LTTV is maintained as a known-stable viewer.<br>
  60 <br>
  61 <br>
  62 * Nice to have<br>
  63 (4) Statistics per time window.<br>
  64 (4) Add Xen per physical CPU view.<br>
  65 (4) Add Xen per vcpu view.<br>
  66 (4) Disable plugins when threshold reached (i.e. too much process in control
  67 flow view). Draw, and, when the threshold is reached, stop drawing. The global
  68 statistics view can inhibit showing the per process stats.<br>
  69 (4) Add a visual artifact : PID 0 could be named swapper instead of UNNAMED for
  70 cpus > 0.<br>
  71 (4) Add event specific fields support to filter.<br>
  72 (4) Add a periodic event interval view. (useful to verify event periodicity)<br>
  73 (4) create a graphical per cpu activity view.<br>
  74 (4) Filter by target process.<br>
  75 (4) Compensate for time spent in probes in LTTV analysis.<br>
  76 (4) Add CPU, network, disk, memory usage histogram. [Per interval statistics]<br>
  77 (4) Add sort by process priority in the control flow view (must also instrument
  78 priority information of the processes).<br>
  79 % Airbus<br>
  80 (5) Add Python scripting hooks.<br>
  81 (5) Add GUI interface to take an hybrid trace.<br>
  82 (5) Automatically detect traces with too much processes and disable faulty operations.<br>
  83 (5) Event sequence detector (inspired from regular expressions).<br>
  84 (7) Create a hardware counter viewer (low cost rate counters : L1 cache miss,
  85 page faults, interrupts...). This will be a generalisation of the event rate
  86 view into a view of the evolution of a user definable event field.<br>
  87 <br>
  88 * TO FIX<br>
  89 (10) Add cancel button to LTTV filter GUI window.<br>
  90 (10) Sometimes, in the control flow view, a process with 0 creation time is
  91 created in addition to the real process itself. Seems to be caused by end of
  92 process life.<br>
  93 (10) Statistics do not take in account the time spent in the mode present at
  94 the beginning of the trace. Example : real time spent in system call on behalf
  95 of process 0.<br>
  96 <br>
  97 <br>
  98 Mathieu Desnoyers<br>
  99
 100
 101 </body>
 102 </html>