fche changed the topic of #systemtap to: http://sourceware.org/systemtap; email systemtap@sourceware.org if answers here not timely, conversations may be logged
pwithnall has joined #systemtap
scox has joined #systemtap
pwithnall has quit [Ping timeout: 248 seconds]
gromero has quit [Ping timeout: 272 seconds]
hpt has joined #systemtap
wcohen has quit [Quit: Leaving]
lindi-_ has joined #systemtap
wcohen has joined #systemtap
higgins` has joined #systemtap
higgins has quit [*.net *.split]
lindi- has quit [*.net *.split]
iskander has quit [Ping timeout: 240 seconds]
iskander has joined #systemtap
orivej has quit [Ping timeout: 256 seconds]
irker708 has quit [Quit: transmission timeout]
slowfranklin has joined #systemtap
iskander has quit [Ping timeout: 260 seconds]
sanoj has joined #systemtap
iskander has joined #systemtap
slowfranklin has quit [Quit: slowfranklin]
fche has quit [Read error: error:1408F10B:SSL routines:SSL3_GET_RECORD:wrong version number]
fche has joined #systemtap
lberk_ has joined #systemtap
lberk has quit [*.net *.split]
lberk_ is now known as lberk
fche has quit [Ping timeout: 252 seconds]
fche has joined #systemtap
fche has quit [Changing host]
fche has joined #systemtap
sanoj has quit [Ping timeout: 248 seconds]
sanoj has joined #systemtap
gila has joined #systemtap
sanoj has quit [Ping timeout: 240 seconds]
iskander has quit [Ping timeout: 256 seconds]
gila has quit [Quit: My Mac Pro has gone to sleep. ZZZzzz…]
gila has joined #systemtap
hpt has quit [Quit: Lost terminal]
gila has quit [Quit: My Mac Pro has gone to sleep. ZZZzzz…]
gila has joined #systemtap
gila has quit [Quit: My Mac Pro has gone to sleep. ZZZzzz…]
sanoj has joined #systemtap
mjw has joined #systemtap
gromero has joined #systemtap
gila has joined #systemtap
gila has quit [Quit: My Mac Pro has gone to sleep. ZZZzzz…]
gila has joined #systemtap
sanoj has quit [Ping timeout: 256 seconds]
fche has quit [Read error: error:1408F10B:SSL routines:SSL3_GET_RECORD:wrong version number]
fche has joined #systemtap
sanoj has joined #systemtap
slowfranklin has joined #systemtap
<gromero> Hi fche. A long time ago you said to me about using systemtap:
<gromero> Mai 30 21:47:19 <fche> we'd love to hear about your successes (& failures) with it, drop us a note sometime if able
<gromero> fche: this was the result: http://seclists.org/oss-sec/2017/q4/51
<gromero> and more details also in https://www.securityfocus.com/bid/101264/references
orivej has joined #systemtap
fche has quit [Ping timeout: 248 seconds]
fche has joined #systemtap
fche has joined #systemtap
<fche> gromero, is there a little writeup of the systemtap side of the investigation?
sanoj has quit [Ping timeout: 252 seconds]
<gromero> fche: systemtap was fundamental to that finding. there is currently not a text on that (I hope to write a account of it tho). here is all my systemtap stuff I used (including crafted tapsets): https://github.com/leitao/htm_torture/tree/isolation/systemtap
<fche> htm_torture -- now there's a word I like
<gromero> fche: lol
<gromero> yup, torture was the first step. systemtap enabled me to inspect what was really happening in kernel space
<gromero> transactional memory is beast to debug on PowerPC (you can't use gdb becasue kernel aborts TM on syscalls)
<gromero> I also used systemtap to discover a register corruption on userspace when TM rollbacked. So I had to use this dirty hack:
<mjw> grin. That is how we emulate it in valgrind too "oops, your transaction aborted immediate, good luck, thanks" :)
<gromero> mjw: heh I see. yup. on scripts in like https://github.com/leitao/htm_torture/blob/isolation/systemtap/tapset/full.d I'm probing the main TM reclaim/recheckpoint functions in kernelspace, it's the only way to track the whole thing.
<gromero> HTM in PowerPC is a little bit more complicated than on x64...
<gromero> fche & community: anyway. I just would like to say thanks for the great work on systemtap and wish Happy Holidays :-)
<fche> likewise, and see you in the new year!
<gromero> fche: thanks! see you in the new year! I could not yet contribute the 128-bit register stuff :grin:
<fche> we should have some better method to let one expand this logic without physical patches
<gromero> fche: what you have in mind exactly? a sort of file format to describe the arch specific details?
<fche> something like a way of plugging in tapset embedded-c functions that extract those things
<fche> so one wouldn't have to -patch- an existing one, just make available another one to call
<fche> dunno, thinking out loud
<gromero> fche: I see
orivej has quit [Ping timeout: 260 seconds]
CME has quit [Ping timeout: 265 seconds]
CME has joined #systemtap
brolley has joined #systemtap
tromey has joined #systemtap
pwithnall has joined #systemtap
pwithnall has quit [Client Quit]
gila has quit [Quit: My Mac Pro has gone to sleep. ZZZzzz…]
orivej has joined #systemtap
drsmith_away has left #systemtap [#systemtap]
CME has quit [Ping timeout: 265 seconds]
CME has joined #systemtap
CME has quit [Read error: Connection reset by peer]
CME has joined #systemtap
slowfranklin has quit [Quit: slowfranklin]
orivej has quit [Ping timeout: 240 seconds]
slowfranklin has joined #systemtap
hpt has joined #systemtap
CME has quit [Ping timeout: 265 seconds]
CME has joined #systemtap
gila has joined #systemtap
hpt has quit [Quit: Lost terminal]
gila has quit [Quit: My Mac Pro has gone to sleep. ZZZzzz…]
orivej has joined #systemtap
tromey has quit [Quit: ERC (IRC client for Emacs 26.0.90)]
slowfranklin has quit [Quit: slowfranklin]
brolley has left #systemtap [#systemtap]
mjw has quit [Quit: Leaving]
_whitelogger has joined #systemtap