fche changed the topic of #systemtap to: http://sourceware.org/systemtap; email systemtap@sourceware.org if answers here not timely, conversations may be logged
<irker079>
systemtap: dsmith systemtap.git:refs/heads/dsmith/python * release-3.0-346-g21745df / testsuite/lib/systemtap.exp testsuite/systemtap.base/listing_mode.exp testsuite/systemtap.base/listing_mode.py testsuite/systemtap.unprivileged/test.py testsuite/systemtap.unprivileged/unprivileged_probes.exp: Add python 2 and python3 probe support to existing tests. http://tinyurl.com/ztkn5kk
drsmith has joined #systemtap
naveen has joined #systemtap
gila has joined #systemtap
gila has quit [Ping timeout: 256 seconds]
hchiramm has joined #systemtap
hchiramm_ has joined #systemtap
gila has joined #systemtap
hchiramm has quit [Ping timeout: 245 seconds]
<irker079>
systemtap: dsmith systemtap.git:refs/heads/master * release-3.0-307-gda79853 / testsuite/systemtap.pass1-4/semko.exp: Avoid creating a file called '&1' in testsuite/systemtap.pass1-4/semko.exp. http://tinyurl.com/ze3xcrt
wcohen has quit [Ping timeout: 245 seconds]
gila has quit [Ping timeout: 268 seconds]
hchiramm_ has quit [Ping timeout: 244 seconds]
wcohen has joined #systemtap
hchiramm_ has joined #systemtap
wcohen is now known as wcohen|pto
gila has joined #systemtap
gila has quit [Ping timeout: 260 seconds]
gila has joined #systemtap
jhg_ has joined #systemtap
pwithnall has quit [Ping timeout: 246 seconds]
mjw has quit [Quit: Leaving]
naveen has quit [Quit: WeeChat 1.6]
tromey has quit [Quit: ERC (IRC client for Emacs 25.1.90.1)]
drsmith has left #systemtap [#systemtap]
frdeso has joined #systemtap
<frdeso>
Hi all, I am new to Systemtap. When tracing USDT probes, when a probe is triggered is a kernel module involved to gather the payload of the probe ? Also, is there any documentation on the implementation of the USDT support?
<fche>
frdeso, hi
<fche>
the implementation of the instrumentation per se is a separate thing from how the instrumentation is used at run time
<fche>
the latter - depends on which stap --runtime is chosen; in the default kernel-module-based one, yes, the kernel will trap on the NOPs that the instrumentation leaves in object code
<fche>
in --runtime=dyninst, no kernel is involved, but instead some run-time binary rewriting is done