fix end of process life
[lttv.git] / ltt / branches / poly / doc / developer / lttng-lttv-roadmap.html
... / ...
CommitLineData
1<html>
2<body>
3<center><big><big>LTTV & LTTng roadmap<small><small></center>
4<br>
5<br>
6Here are the roadmaps for the LTTV and LTTng development. I use a priority indice
7for the TODO items :<br>
8(1) : very high priority<br>
9(10): lowest priority<br>
10<br>
11<br>
12Dependencies are written between brackets [ ].<br>
13The # symbol marks who is currently working on the item.<br>
14The % 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
22file.<br>
23(3) Use a per architecture enumeration for traps.<br>
24(4) Add event specific fields support to filter.<br>
25(4) Add a periodic event interval view. (useful to verify event periodicity)<br>
26(4) create a graphical per cpu activity view.<br>
27(5) Add Python scripting hooks.<br>
28(5) Flight recorder : start lttd automatically upon GUI trace control stop.<br>
29(5) Automatically detect traces with too much processes and disable faulty operations.<br>
30(5) Event sequence detector (inspired from regular expressions).<br>
31(7) Create a hardware counter viewer (low cost rate counters : L1 cache miss,
32page faults, interrupts...). This will be a generalisation of the event rate
33view into a view of the evolution of a user definable event field.<br>
34<br>
35* TO FIX<br>
36(10) Event filter : event.time does not work.<br>
37(10) Add cancel button to LTTV filter GUI window.<br>
38(10) Sometimes, in the control flow view, a process with 0 creation time is
39created in addition to the real process itself. Seems to be caused by end of
40process life.<br>
41(10) Statistics do not take in account the time spent in the mode present at
42the beginning of the trace. Example : real time spent in system call on behalf
43of process 0.<br>
44<br>
45<br>
46<big>LTT Next Generation Roadmap<small><br>
47<br>
48* TODO<br>
49(4) Add Xen support.<br>
50 # <A HREF="mailto:Mathieu Desnoyers <compudj@krystal.dyndns.org>">Mathieu Desnoyers</A><br>
51 # <A HREF="mailto:Parisa Heidari <parisa.heidari@polymtl.ca>">Parisa Heidari</A><br>
52(4) efficient dynamic event filtering while recording trace.<br>
53 % Sensis Corp. <A HREF="mailto:Bish, Tim <Tim.Bish@Sensis.com>">Tim Bish</A><br>
54(4) instrument kernel bottom half irqsave, spinlocks, rwlocks, seqlocks, semaphores, mutexes, brlock.<br>
55(4) integrate NPTL instrumentation (see
56<A HREF="http://nptltracetool.sourceforge.net/">PTT</A>).<br>
57(5) Support CPUs with scalable frequency.<br>
58(5) Integrate LTTng and lttd with LKCD.<br>
59(7) Integrate periodical dump of perfctr hardware counters.<br>
60(8) Integrate periodical dump of SystemTAP computed information.<br>
61(8) Integrate SystemTAP logging with LTTng.<br>
62(9) Add support for setjmp/longjmp and jump tables instrumentation to
63ltt-instrument-functions.<br>
64(9) Make ltt-usertrace-fast support internal heartbeat timer.<br>
65(3) port LTTng to :<br>
66x86_64<BR>
67#Autodesk <A HREF="mailto:Francois L'Archeveque <fla@discreet.com>">Francois
68L'Archeveque</A> for 2.6.9 RedHat Enterprise (patch available in contrib, still
69has to be ported to 2.6.14, thanks!)<BR>
70# Wind River for 2.6.14<BR>
71PPC<BR>
72PPC64<BR>
73# Wind River for 2.6.14<BR>
74alpha<BR>
75user-mode Linux<BR>
76Xen<BR>
77# IBM<BR>
78MIPS<BR>
79MIPS64<BR>
80% Wind River for 2.6.14<BR>
81# Tim Bird <A HREF="mailto:tim.bird@as.sony.com">Tim Bird</A><br>
82S/390<BR>
83RTLinux<BR>
84% Wind River for 2.6.14<BR>
85<br>
86<br>
87<br>
88* TODO (low priority)<br>
89enhance RPM packages for lttng kernel<br>
90Integrate header generation (genevent) in kernel build system.<br>
91Find a different way to printk from instrumentation : forbidden from schedule
92and wakeup (causes a deadlock). -> Modify printk so it uses LTTng tracer!<br>
93Export channels via network sockets instead of writing them to disk.<br>
94Export buffers with time constraint for "live" visualisation. Use
95ltt_force_switch periodically from a timer to insure slow channels do not
96interfere with viewing.<br>
97Have an optional round-robin mode to write information into multiple channels
98from the same source.<br>
99<br>
100* Need to be discussed<br>
101Use 2.6.14 RelayFS control files.<br>
102Drop ltt-module-register and ltt-module-unregister, use exported variables.<br>
103drop ltt_filter_control, use functions pointers instead.<br>
104Merge facilities headers into one big header.<br>
105Change the name of XML files from XML to something else.<br>
106Remove ltt-base.c.<br>
107<br>
108<br>
109* lttng patch division (0.5.10+) :<br>
110<br>
111instrumentation.diff<br>
112instrumentation-arm.diff<br>
113instrumentation-i386.diff<br>
114instrumentation-mips.diff<br>
115instrumentation-powerpc.diff<br>
116instrumentation-ppc.diff<br>
117instrumentation-s390.diff<br>
118instrumentation-sh.diff<br>
119ltt-facilities-headers.diff<br>
120ltt-facilities-loader.diff<br>
121ltt-facilities.diff<br>
122ltt-relayfs.diff<br>
123ltt-core.diff<br>
124ltt-build.diff<br>
125<br>
126<br>
127Mathieu Desnoyers<br>
128
129
130</body>
131</html>
This page took 0.022486 seconds and 4 git commands to generate.