fche changed the topic of #systemtap to: http://sourceware.org/systemtap; email systemtap@sourceware.org if answers here not timely, conversations may be logged
<fche> jhg_, hard to say under what conditions it will or won't work
<fche> stap is interpreting dwarf bytecode that a magic compiler produces for us, plus magic stack/register dumps a magic kernel gives us
<fche> so there's a lot of magic involved
<fche> and yeah sometimes we can get data for inlined functions, sometimes not
Meths has quit [Ping timeout: 268 seconds]
Meths has joined #systemtap
hpt has joined #systemtap
hpt has quit [Ping timeout: 268 seconds]
hpt has joined #systemtap
hpt has quit [Ping timeout: 265 seconds]
hpt has joined #systemtap
<jhg_> I turned off optimization for the kernel module that calls the kernel function. this boundary seems like it would be fairly explicit, as it is a dynamically loaded module.
<jhg_> script and output: https://hastebin.com/okexafidoh.go
<jhg_> it tracks the child by thread id
hpt has quit [Ping timeout: 260 seconds]
hpt has joined #systemtap
hpt has quit [Ping timeout: 264 seconds]
hpt has joined #systemtap
_whitelogger has joined #systemtap
slowfranklin has joined #systemtap
slowfranklin has quit [Quit: slowfranklin]
slowfranklin has joined #systemtap
aryehw has joined #systemtap
pwithnall has joined #systemtap
orivej has quit [Ping timeout: 260 seconds]
aryehw has quit [Ping timeout: 240 seconds]
scox has quit [Ping timeout: 240 seconds]
pwithnall has quit [Quit: pwithnall]
aryehw has joined #systemtap
hpt has quit [Ping timeout: 244 seconds]
mjw has joined #systemtap
wcohen_ has quit [Remote host closed the connection]
aryehw has quit [Remote host closed the connection]
aryehw has joined #systemtap
drsmith has left #systemtap [#systemtap]
drsmith has joined #systemtap
wcohen has joined #systemtap
orivej has joined #systemtap
tromey has joined #systemtap
brolley has joined #systemtap
aryehw has quit [Ping timeout: 240 seconds]
irker882 has joined #systemtap
<irker882> systemtap: wcohen systemtap.git:refs/heads/master * release-3.3-68-ge180d97 / stap-exporter/scripts/example1.stp stap-exporter/scripts/example2.stp: Adjust stap-exporter probe points to work with newer Linux 4.17 kernels http://tinyurl.com/yawb5y3l
<irker882> systemtap: wcohen systemtap.git:refs/heads/master * release-3.3-69-gaa0349e / stap-exporter/scripts/cpu_throttle.stp: Correct typos in the cpu_throttle.stp comments. http://tinyurl.com/y8ojsn58
<fche> wcohen, why don't those tests use syscall.FOO aliases?
<wcohen> fche, they probably could use the non-dwarf versions of syscall.FOO
<wcohen> the probe originally using kernel.function("sys_read") changed to kernel.kprobe("sys_read") to avoid requiring kernel-debuginfo, then tweak to work with new kernels.
* wcohen changing fixing it to use syscall.read and syscall.open
<irker882> systemtap: wcohen systemtap.git:refs/heads/master * release-3.3-70-g954ab72 / stap-exporter/scripts/example1.stp stap-exporter/scripts/example2.stp: Use the predefined syscall.read and syscall.open in the stap-exporter examples http://tinyurl.com/y84qjjgn
<fche> thanks
<fche> yeah the modern syscall.* aliases adapt nicely
pwithnall has joined #systemtap
pwithnall has quit [Ping timeout: 256 seconds]
pwithnall has joined #systemtap
scox has joined #systemtap
slowfranklin has left #systemtap [#systemtap]
gromero has joined #systemtap
dmalcolm_ has quit [Read error: Connection reset by peer]
dmalcolm has joined #systemtap
slowfranklin has joined #systemtap
slowfranklin has quit [Client Quit]
pwithnall has quit [Quit: pwithnall]
orivej has quit [Ping timeout: 260 seconds]
orivej has joined #systemtap
tromey has quit [Quit: ERC (IRC client for Emacs 26.1.50)]
<irker882> systemtap: wcohen systemtap.git:refs/heads/master * release-3.3-71-g3f56638 / stap-exporter/exporter.conf stap-exporter/scripts/also_ran.stp: Add the also_ran.stp script to the stap-export examples http://tinyurl.com/y8la79k8
<fche> wcohen, by the way, is there a reason these exporter samples need to be isolated from the testsuite.examples ?
<fche> ISTM they could easily be in the latter location
<fche> and stap-exporter would be configured with stap --example FOO.stp (using jafeer's recent searching mode, built with this in mind)
<wcohen> fche, certainly it would make sense to have them in testsuite.examples.
<fche> at one point the idea was not to make exporter-only-purposed scripts, but to make generally useful scripts exporter-friendly
<wcohen> fche, just put these examples in testsuite/systemtap.examples? make the scripts so they provide output both through /proc/systemtap/* and terminal? or just make people aware that the info is pulled out out of /proc/systemtap/*/...
<fche> output through both would be ideal
<fche> at probe-end if nothing else; wouldn't want to impose too frequent/high cost on an exporter-bound script in generating periodic stdout reports that are ignored
<wcohen> also how would these be grouped in testsuite/systemtap.examples? another directory for exporter-friendly scripts? or use another keyword type for "output:" of the meta data and place the scripts in the existing current directories.
<fche> suggest grouping them topically as usual
<fche> /general/ if nothing else
<wcohen> okay and have a keyword to make it easier to find the stap-exporter friendly scripts.
<fche> exactly
<fche> maybe 'prometheus'
<wcohen> okay.
wcohen has quit [Ping timeout: 264 seconds]
pwithnall has joined #systemtap
gromero has quit [Ping timeout: 264 seconds]
brolley has left #systemtap [#systemtap]
mjw has quit [Quit: Leaving]
wcohen has joined #systemtap
wcohen has quit [Ping timeout: 264 seconds]
wcohen has joined #systemtap
pwithnall has quit [Quit: pwithnall]