<ul>
<li><a href="#intro" name="TOCintro">Introduction</a></li>
+<ul>
+<li><a href="#arch" name="TOCarch">Supported architectures</a></li>
+</ul>
<li><a href="#section1" name="TOCsection1">Installing LTTng and LTTV from
sources</a></li>
refer to :
<a
href="http://ltt.polymtl.ca/svn/trunk/lttv/doc/developer/lttng-lttv-compatibility.html">LTTng+LTTV versions compatibility</a>
-The lttng patch is necessary to have the tracing hooks in the kernel.
+
+The ongoing work had the Linux Kernel Markers integrated in the mainline Linux
+kernel since Linux 2.6.24 and the Tracepoints since 2.6.28. In its current
+state, the lttng patchset is necessary to have the trace clocksource, the
+instrumentation and the LTTng high-speed data extraction mechanism added to the
+kernel.
<br>
<br>
-Supported architectures :
+<h3><a href="#TOCarch" name="arch">Supported architectures</a></h3>
<br>
LTTng :<br>
<br>
<li> ARM (with limited timestamping precision, e.g. 1HZ. Need
architecture-specific support for better precision)
<li> MIPS
+<li> sh (partial architecture-specific instrumentation)
+<li> sparc64 (partial architecture-specific instrumentation)
+<li> s390 (partial architecture-specific instrumentation)
+<li> Other architectures supported without architecture-specific instrumentation
+and with low-resolution timestamps.<br>
<br>
<br>
LTTV :<br>
</PRE>
<p>
(note : to see if the buffers has been filled, look at the dmesg output after
-lttctl -R or after stopping tracing from the GUI, it will show an event lost
+lttctl -D or after stopping tracing from the GUI, it will show an event lost
count. If it is the case, try using larger buffers. See lttctl --help to learn
how. lttv now also shows event lost messages in the console when loading a trace
with missing events or lost subbuffers.)
success with such tracing approach to fix "rare disk delay" issues and
VM-related issues presented in this article :
<ul>
- <li> "Linux Kernel Debugging on Google-sized clusters at Ottawa Linux
- Symposium 2007"
- http://ltt.polymtl.ca/papers/bligh-Reprint.pdf
+ <li> <a href="http://ltt.polymtl.ca/papers/bligh-Reprint.pdf">Linux Kernel
+Debugging on Google-sized clusters at Ottawa Linux
+ Symposium 2007</a>
</ul>
<li> IBM Research have had problems with Commercial Scale-out applications,
which are being an increasing trend to split large server workloads.