fche changed the topic of #systemtap to: http://sourceware.org/systemtap; email systemtap@sourceware.org if answers here not timely, conversations may be logged
hpt has joined #systemtap
wcohen has quit [Ping timeout: 256 seconds]
wcohen has joined #systemtap
<toothe> hm...having problems.
sanoj has joined #systemtap
slowfranklin has joined #systemtap
slowfranklin has quit [Quit: slowfranklin]
slowfranklin has joined #systemtap
sanoj has quit [Ping timeout: 256 seconds]
slowfranklin has quit [Quit: slowfranklin]
orivej has quit [Ping timeout: 240 seconds]
sanoj has joined #systemtap
hpt has quit [Quit: Lost terminal]
slowfranklin has joined #systemtap
gila has joined #systemtap
higgins has quit [Quit: Leaving]
higgins has joined #systemtap
slowfranklin has quit [Quit: slowfranklin]
gila has quit [Quit: My Mac Pro has gone to sleep. ZZZzzz…]
gila has joined #systemtap
scox has quit [Ping timeout: 256 seconds]
gromero has joined #systemtap
sanoj has quit [Ping timeout: 248 seconds]
mjw has joined #systemtap
wcohen has quit [Ping timeout: 265 seconds]
slowfranklin has joined #systemtap
slowfranklin has quit [Quit: slowfranklin]
sanoj has joined #systemtap
sanoj has quit [Ping timeout: 240 seconds]
slowfranklin has joined #systemtap
gila has quit [Quit: My Mac Pro has gone to sleep. ZZZzzz…]
slowfranklin has quit [Quit: slowfranklin]
gila has joined #systemtap
slowfranklin has joined #systemtap
slowfranklin has quit [Client Quit]
wcohen has joined #systemtap
slowfranklin has joined #systemtap
mbenitez has joined #systemtap
mbenitez has joined #systemtap
mbenitez has quit [Changing host]
scox has joined #systemtap
slowfranklin has quit [Quit: slowfranklin]
slowfranklin has joined #systemtap
slowfranklin has quit [Quit: slowfranklin]
orivej has joined #systemtap
slowfranklin has joined #systemtap
brolley has joined #systemtap
gila has quit [Quit: My Mac Pro has gone to sleep. ZZZzzz…]
tromey has joined #systemtap
slowfranklin has quit [Quit: slowfranklin]
slowfranklin has joined #systemtap
brolley has left #systemtap [#systemtap]
slowfranklin has quit [Quit: slowfranklin]
<toothe> I keep getting tons of memory addresses, rather than the actual function names.
<toothe> And, I don't see how the code went from one function to another.
<fche> pls fpaste your script and sample output
orivej has quit [Read error: Connection reset by peer]
<toothe> I'm sorry! I was disconnected, didn't realize.
<toothe> a moment.
orivej has joined #systemtap
<toothe> for some odd reason, this takes a long time now. Might be a kernel issue.
<toothe> That's the script.
<toothe> That's the output.
<toothe> I have a lot of 0xfffff's there.
<fche> hm, have you tried stap --all-modules ?
<fche> instead of those -d individual bits ?
<fche> maybe -d kernel is not being implicitly selected
<toothe> sure.
* toothe runs.
<toothe> thescript.
<toothe> same outoupt.
brolley has joined #systemtap
<toothe> this is a "custom" kernel module.
<fche> and that's the copy installed, loaded, run etc. ?
<toothe> But custom, its a well-known kernel hacker, I am compiling his code separately.
<toothe> pardon?
<fche> out-of-tree modules are trickier to process w/ stap
<toothe> i see...
<fche> but if you installed the .ko under /lib/modules/.... etc, it should be smooth
<fche> anyway this wouldn't explain the output
<toothe> I didn't, I loaded it with insmod.
<fche> hm that could explain it; the stap -d FOO and even probe() statements will look for modules under /lib/modules/....
<toothe> i see.
<toothe> what about the --all-modules flag?
<fche> that's just matches up 'lsmod' and /lib/modules/....
<fche> it doesn't search the rest of your system for a wayward .ko file that may have replaced a kernel one
<toothe> so, its lsmod'd
<toothe> i see.
brolley has quit [Ping timeout: 265 seconds]
<toothe> okay, so let me go ahead and do a manual replace.
<fche> I'd think stap may give you some diagnostics at the beginning of the run if it notices mismatches
<fche> your output was the tail of the run; anything at the front?
mjw has quit [Quit: Leaving]
<toothe> ohh...it happens so fast. Let me check.
<toothe> are those preceeding messages stdout or stderr?
<toothe> i presume the latter.
<fche> should be the latter
orivej_ has joined #systemtap
orivej has quit [Ping timeout: 255 seconds]
<toothe> sorry, this is taking me a bit.
<toothe> I think...my computer just crashed lol.
<fche> bummer ... is there a crash message or anything?
<toothe> or locked up - i"m not certain.
<toothe> the entire machine.
<toothe> but it wasn't stap, it was at the rmmod.
<fche> whew? :-o
<fche> yeah some modules don't take well to being removed
<toothe> yeah heh. I dunno what its doing at the moment, honestly.
<fche> alt-sysrq dead too?
<toothe> its back.
<toothe> oh wait, maybe it didn't crash....im confused heh
<fche> hey might want to give that a good reboot, just to clean out the cobwebs
<toothe> I might. Its a personal machine anwyays.
<toothe> let me just finalize this loading script...
brolley has joined #systemtap
slowfranklin has joined #systemtap
<toothe> alright, got my script up - lets give this a shot now.
<fche> running your new module?
<toothe> yes.running this line now:
<toothe> sudo stap --all-modules -d kernel -g --suppress-time-limits -D DTP_NO_OVERLOAD ./onefunction.stp
<fche> (with --all-modules you don't need -d kernel, it's implied)
<toothe> it seems to take a while to load, but I suspect this is a kernel problem somewhere.
<toothe> ahh...
<toothe> alright, i'll check the stderr now.
<toothe> i'm really trying to figure out how the heck this code goes from function A to tunction B.
<toothe> even though I see zero connection between the two.
<fche> note that .function() probes will match inlined functions too
<toothe> Too many interrupts received, exiting.
<toothe> That's the only stderr I got.
<fche> you might be better served with a module(..).statement("*@*:*") type probe
<fche> that way you get a line-by-line trace
<toothe> I'm sorry, I didn't follow?
<fche> though you'd probably want that restricted by function name, file name, or line number range
<fche> .function() probes trigger at the start of a normal or inlined function
slowfranklin has quit [Client Quit]
<toothe> I'm confused. ISn't that exactly what I'm doing?
<fche> yes
<fche> it won't show you the call site or the call graph or super detailed line-by-line trace though
slowfranklin has joined #systemtap
<toothe> ah, okay.
<toothe> but that's secondary I believe?
<fche> yeah.
slowfranklin has quit [Client Quit]
<fche> so now anyway, are you getting more legible backtraces?
<toothe> no...I don't see how this code goes from A to B.
<toothe> I'm confused abou thtat.
<fche> right, so that's why you might consider statement-by-statement traces, at least of a handful of functions at a time
<toothe> hm...not sure how...
<fche> probve module("rtl_pci").statement("*_card_disable@*:*") { println(pp) }
<fche> er probe
<toothe> pp ?
<fche> pp()
<fche> sorry, thinking out loud.
<fche> probably have some more apropos examples though
<fche> https://sourceware.org/systemtap/examples/#profiling/linetimes.stp <-- but even that's too much probably
<fche> toothe, any luck with that kind of thing?
slowfranklin has joined #systemtap
slowfranklin has quit [Client Quit]
slowfranklin has joined #systemtap
slowfranklin has quit [Client Quit]
wcohen has quit [Ping timeout: 255 seconds]
wcohen has joined #systemtap
gila has joined #systemtap
wcohen has quit [Ping timeout: 256 seconds]
wcohen has joined #systemtap
gromero has quit [Ping timeout: 240 seconds]
nkambo has quit []
slowfranklin has joined #systemtap
tromey has quit [Quit: ERC (IRC client for Emacs 26.0.90)]
slowfranklin has quit [Quit: slowfranklin]
slowfranklin has joined #systemtap
irker727 has joined #systemtap
<irker727> systemtap: dsmith systemtap.git:refs/heads/master * release-3.2-72-g434d56f / httpd/backends.cxx: Copy user files into the docker container image. http://tinyurl.com/ybgn4wkt
mbenitez has quit [Quit: Leaving]
<irker727> systemtap: dsmith systemtap.git:refs/heads/master * release-3.2-73-gad4727a / httpd/backends.cxx httpd/docker/Makefile.am httpd/docker/Makefile.in httpd/docker/stap_build_docker_container.py httpd/docker/stap_build_docker_image.py: Renamed httpd/docker/stap_build_docker_container.py. http://tinyurl.com/y8393nm3
<irker727> systemtap: dsmith systemtap.git:refs/heads/master * release-3.2-74-g05befbb / httpd/utils.cxx: Add a small fix to execute_and_capture(). http://tinyurl.com/yazon84g
brolley has left #systemtap [#systemtap]
scox has quit [Ping timeout: 265 seconds]
slowfranklin has quit [Quit: slowfranklin]
wcohen has quit [Ping timeout: 265 seconds]
slowfranklin has joined #systemtap
<fche> drsmith_away, for some guy away, that's some tantalizing code
drsmith_away is now known as drsmith
<drsmith> just imagine what I could do if I was really here
<fche> yeah!
drsmith is now known as drsmith_away
slowfranklin has quit [Ping timeout: 240 seconds]
wcohen has joined #systemtap
orivej_ has quit [Ping timeout: 265 seconds]
orivej has joined #systemtap
scox has joined #systemtap