fche changed the topic of #systemtap to: http://sourceware.org/systemtap; email systemtap@sourceware.org if answers here not timely, conversations may be logged
scox has joined #systemtap
hpt has joined #systemtap
hpt has quit [Ping timeout: 264 seconds]
hpt has joined #systemtap
ananth has joined #systemtap
hpt has quit [Ping timeout: 272 seconds]
ravi_ has joined #systemtap
ravi_ has quit [Ping timeout: 252 seconds]
hpt has joined #systemtap
ravi_ has joined #systemtap
hchiramm has joined #systemtap
ravi_ has quit [Read error: Connection reset by peer]
ego_ has joined #systemtap
srikar_away is now known as srikar
ravi_ has joined #systemtap
hchiramm has quit [Ping timeout: 250 seconds]
hkshaw has quit [Quit: Leaving.]
ravi_ has quit [Ping timeout: 250 seconds]
hchiramm has joined #systemtap
ravi_ has joined #systemtap
hchiramm has quit [Ping timeout: 250 seconds]
ravi_ has quit [Ping timeout: 252 seconds]
ravi_ has joined #systemtap
hchiramm has joined #systemtap
ravi_ has quit [Remote host closed the connection]
ravi_ has joined #systemtap
mjw has joined #systemtap
jhg_ has quit [Ping timeout: 260 seconds]
hpt has quit [Ping timeout: 258 seconds]
jhg_ has joined #systemtap
pmuldoon has quit [Ping timeout: 264 seconds]
pmuldoon has joined #systemtap
scox has quit [Ping timeout: 252 seconds]
ph7 has joined #systemtap
ph7 has quit [Ping timeout: 264 seconds]
ph7 has joined #systemtap
hkshaw has joined #systemtap
mjw has quit [Ping timeout: 240 seconds]
ph7 has quit [Quit: Leaving.]
ph7 has joined #systemtap
hpt has joined #systemtap
mjw has joined #systemtap
ph7 has quit [Client Quit]
ph7 has joined #systemtap
wcohen has quit [Remote host closed the connection]
ph7 has quit [Quit: Leaving.]
mjw has quit [Ping timeout: 276 seconds]
wcohen has joined #systemtap
scox has joined #systemtap
mjw has joined #systemtap
ravi_ has quit [Ping timeout: 264 seconds]
mjw has quit [Client Quit]
mjw has joined #systemtap
hpt has quit [Quit: Lost terminal]
ph7 has joined #systemtap
wcohen has quit [Ping timeout: 258 seconds]
ego_ has quit [Ping timeout: 258 seconds]
drsmith has joined #systemtap
hkshaw has quit [Ping timeout: 240 seconds]
srikar is now known as srikar_away
ego_ has joined #systemtap
wcohen has joined #systemtap
ph7 has quit [Quit: Leaving.]
hkshaw has joined #systemtap
ph7 has joined #systemtap
srikar_away is now known as srikar
tromey has joined #systemtap
ananth has quit [Quit: Leaving]
ph7 has quit [Read error: Connection timed out]
ph7 has joined #systemtap
ego_ has quit [Ping timeout: 240 seconds]
dmalcolm has quit [Read error: Connection reset by peer]
ph7 has quit [Quit: Leaving.]
<jhg_> more interesting behavior. I manually created a kernel modules entry: /lib/modules/`uname -r`/kernel/fs/openafs/openafs.ko, depmod -a, then stap -v -l 'module("openafs").function("*")' shows the functions just fine
<jhg_> if I run the module and its daemon, probe module("openafs").function("*").call {} does not trigger anything
<jhg_> stap -vv stapfile.stp shows that it focuses on the right kmod
dmalcolm has joined #systemtap
<jhg_> hmm
<jhg_> even the full path to the module in the /lib/modules/... does not fire
<fche> is there perhaps more than one openafs.ko module in play?
<fche> stap compares buildids at startup and rejects probing mismatched binaries; anything in dmesg?
<jhg_> could be multiple .ko's in play. I'll delete the extras.
<fche> stap -vv should identify the one(s) it finds/probes, probably
<jhg_> nothing obvious in dmesg. just what I'd expect for loading and unloading kmods all the time.
<jhg_> the path is correct in stap -vv "focused on..."
<fche> ok; wonder if that's the same one that your modprobe/ afs userspace caused to load
<fche> try running stap with -DDEBUG_SYMBOLS=1
<jhg_> output for the startup command: Loading kernel module: /lib/modules/4.4.15/kernel/fs/openafs/openafs.ko
<jhg_> focused on module 'openafs' = [0x10000-0x23cc10, bias 0 file /lib/modules/4.4.15/kernel/fs/openafs/openafs.ko ELF machine |x86_64 (code 62)
<fche> try also stap -DDEBUG_KPROBES
srikar is now known as srikar_away
srikar_away is now known as srikar
<jhg_> how do I get stap to rebuild from scratch (no cache)?
<fche> stap --poison-cache
<fche> but it should detect the need for it automagically (from buildids etc)
<jhg_> still doesn't see it
<fche> with -DDEBUG_KPROBES, are you seeing "+kprobe ..." type messages?
<fche> hey what the heck try also -DDEBUG_SYMBOLS too
<jhg_> negative on the kprobes
ananth has joined #systemtap
<jhg_> well... I suppose I'll go back to absolute paths in the build tree, because at least they worked... once =)
<jhg_> not sure if this is users error or if I should file a bug
<fche> not sure.
<fche> another bit of information gathering to try: stap -k ... then share the generated stap-symbols.h file
<jhg_> well, at least it notices when the .ko is unloaded and loaded again: stapkp_refresh:774: refresh 74 probes with module libafs
<fche> hm libafs != openafs
<fche> what's up there?
<jhg_> perhaps this is a naming issue with building the module?
<jhg_> it was built as libafs.ko, I renamed it openafs.ko and put it into /lib/modules/... to give it a different namespace
<jistone> but it's module struct will probably still have the old name
<jistone> *its
<fche> jhg_, yeah, that will probably mess lots of things up, not just stap
* jhg_ looks
<jhg_> can't find that quickly. easy test is to rename it back to libafs.ko, same directory, depmod -a and see where we stand
<fche> (and rmmod the previous version)
* jhg_ nods
<jhg_> oh yes... there it is. =)
<jhg_> thanks fche jistone. I can see again.
<fche> we cured the blind! :-)
<fche> HIGH FIVE
* jhg_ high fives!
<jistone> o/
<fche> hahaha. I wonder how to better automate this ...
<fche> could stap look into the .ko file and note / warn as to the actual kernel-level MODULE_NAME therein?
<jistone> \o <- my imaginary friend who doesn't leave me hanging
<jhg_> \o
<fche> \o/ guy with two hands
<jistone> thanks :)
* jhg_ has never done the ascii-five before
<jhg_> cool
<jhg_> fche: probably not worth adding the complexity to detect it. I am just good at breaking things that already work.
<jistone> I bet unicode has emojis for this
<fche> you're probably not the only one ;-)
<fche> U+1F4A9 ?
<jistone> I'm not putting my hand near that
<jhg_>
hchiramm has quit [Ping timeout: 252 seconds]
zodbot has quit [Quit: reboot to free memory]
zodbot has joined #systemtap
srikar is now known as srikar_away
lberk has quit [Ping timeout: 250 seconds]
hchiramm has joined #systemtap
mjw has quit [Ping timeout: 250 seconds]
mjw has joined #systemtap
lberk has joined #systemtap
hchiramm has quit [Ping timeout: 250 seconds]
mjw has quit [Quit: Leaving]
hchiramm has joined #systemtap
hchiramm has quit [Read error: Connection timed out]
detter has joined #systemtap
hchiramm has joined #systemtap
wcohen has quit [Ping timeout: 244 seconds]
hchiramm has quit [Ping timeout: 250 seconds]
hchiramm has joined #systemtap
scox has quit [Ping timeout: 276 seconds]
tromey has quit [Quit: ERC (IRC client for Emacs 25.0.95.2)]
drsmith is now known as drsmith_away
wcohen has joined #systemtap
hchiramm has quit [Ping timeout: 250 seconds]
hchiramm has joined #systemtap
detter has quit [Quit: Leaving]
hchiramm has quit [Ping timeout: 250 seconds]
hchiramm has joined #systemtap
hchiramm_ has joined #systemtap