ltt/branches/poly/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 <big>LTTV Roadmap<small><br>
  17 <br>
  18 * TODO<br>
  19 (3) Add cluster support.<br>
  20         # <A HREF="mailto:Eric Clement <eric.clement@polymtl.ca>">Eric Clement</A><br>
  21 (3) Make LTTV aware of type formats (visual separators) defined in the XML
  22 file.<br>
  23 (3) Use a per architecture enumeration for traps.<br>
  24 (3) Change the byte pair "facility, event" id for a short combining the
  25 informatinon.<br>
  26 (4) Statistics per time window.<br>
  27 (4) Add Xen per physical CPU view.<br>
  28 (4) Add Xen per vcpu view.<br>
  29 (4) Disable plugins when threshold reached (i.e. too much process in control
  30 flow view). Draw, and, when the threshold is reached, stop drawing. The global
  31 statistics view can inhibit showing the per process stats.<br>
  32 (4) Add a visual artifact : PID 0 could be named swapper instead of UNNAMED for
  33 cpus > 0.<br>
  34 (4) Add event specific fields support to filter.<br>
  35 (4) Add a periodic event interval view. (useful to verify event periodicity)<br>
  36 (4) create a graphical per cpu activity view.<br>
  37 (4) Filter by target process.<br>
  38 (4) Compensate for time spent in probes in LTTV analysis.<br>
  39 (4) Add CPU, network, disk, memory usage histogram. [Per interval statistics]<br>
  40 (5) Add Python scripting hooks.<br>
  41 (5) Add GUI interface to take an hybrid trace.<br>
  42 (5) Flight recorder : start lttd automatically upon GUI trace control stop.<br>
  43 (5) Automatically detect traces with too much processes and disable faulty operations.<br>
  44 (5) Event sequence detector (inspired from regular expressions).<br>
  45 (7) Create a hardware counter viewer (low cost rate counters : L1 cache miss,
  46 page faults, interrupts...). This will be a generalisation of the event rate
  47 view into a view of the evolution of a user definable event field.<br>
  48 <br>
  49 * TO FIX<br>
  50 (10) Add cancel button to LTTV filter GUI window.<br>
  51 (10) Sometimes, in the control flow view, a process with 0 creation time is
  52 created in addition to the real process itself. Seems to be caused by end of
  53 process life.<br>
  54 (10) Statistics do not take in account the time spent in the mode present at
  55 the beginning of the trace. Example : real time spent in system call on behalf
  56 of process 0.<br>
  57 <br>
  58 <br>
  59 <big>LTT Next Generation Roadmap<small><br>
  60 <br>
  61 * TODO<br>
  62 (1) CPU Hotplug support. (Only ltt-heartbeat needs to be fixed).<br>
  63 (1) Add Xen support.<br>
  64         # <A HREF="mailto:Mathieu Desnoyers <compudj@krystal.dyndns.org>">Mathieu Desnoyers</A><br>
  65 (1) Integrate SystemTAP logging with LTTng.<br>
  66 (2) Test and post to LKML Linux Kernel Markers.<br>
  67 (3) Change the byte pair "facility, event" id for a short combining the
  68 informatinon.<br>
  69 (4) efficient dynamic event filtering while recording trace.<br>
  70         % Sensis Corp. <A HREF="mailto:Bish, Tim <Tim.Bish@Sensis.com>">Tim Bish</A><br>
  71 (4) instrument kernel bottom half irqsave, spinlocks, rwlocks, seqlocks, semaphores, mutexes, brlock.<br>
  72 (4) integrate NPTL instrumentation (see
  73 <A HREF="http://nptltracetool.sourceforge.net/">PTT</A>).<br>
  74 (4) Probe calibration kernel module.<br>
  75 (4) Make page faults detect nested fault without nesting 4 times in the page
  76 fault handler.<br>
  77 (5) Support CPUs with scalable frequency.<br>
  78 (5) Add boot time tracing support.<br>
  79 (5) Integrate LTTng and lttd with LKCD.<br>
  80 (7) Integrate periodical dump of perfctr hardware counters.<br>
  81 (8) Integrate periodical dump of SystemTAP computed information.<br>
  82 (9) Rethink facility logging code (genevent output). Maybe unneeded because of
  83 markers/probes mechanism.<br>
  84 (9) Add support for setjmp/longjmp and jump tables instrumentation to
  85 ltt-instrument-functions.<br>
  86 (9) Make ltt-usertrace-fast support internal heartbeat timer.<br>
  87 (3) port LTTng to :<br>
  88 alpha<BR>
  89 user-mode Linux<BR>
  90 Xen<BR>
  91 # IBM<BR>
  92 S/390<BR>
  93 RTLinux<BR>
  94 % Wind River for 2.6.14<BR>
  95 sh4<br>
  96 <br>
  97 <br>
  98 <br>
  99 * TODO (low priority)<br>
 100 enhance RPM packages for lttng kernel<br>
 101 Integrate header generation (genevent) in kernel build system.<br>
 102 Export channels via network sockets instead of writing them to disk.<br>
 103 Export buffers with time constraint for "live" visualisation. Use
 104 ltt_force_switch periodically from a timer to insure slow channels do not
 105 interfere with viewing.<br>
 106 Have an optional round-robin mode to write information into multiple channels
 107 from the same source.<br>
 108 <br>
 109 * Need to be discussed<br>
 110 Use 2.6.14 RelayFS control files.<br>
 111 Drop ltt-module-register and ltt-module-unregister, use exported variables.<br>
 112 drop ltt_filter_control, use functions pointers instead.<br>
 113 Merge facilities headers into one big header.<br>
 114 Change the name of XML files from XML to something else.<br>
 115 Remove ltt-base.c.<br>
 116 <br>
 117 Mathieu Desnoyers<br>
 118
 119
 120 </body>
 121 </html>