Someone wrote an article on using syscalltrack!
This coupled with the latest interest in syscalltrack on lkml and Amir’s port to FreeBSD makes me want to get back to hacking on it…
edit: I am sorry, forgot to give credit where credit is due: Thanks ideawerkz for letting me know!
Muli Ben-Yehuda's journal 
Comment by ideawerkz — January 4, 2004 @ 1:30 AM |
why sad? if it’s because I forgot to give you credit, I’m really sorry, I meant to and got distracted.
Comment by mulix — January 4, 2004 @ 1:39 AM |
huh? no, please do not misunderstand. giving credit or not is no big deal. i am trying to see what i can work on syscalltrack.
Comment by ideawerkz — January 4, 2004 @ 1:49 AM
so why sad? there’s plenty to do!
Comment by mulix — January 4, 2004 @ 2:05 AM
Hehe, that is so true 🙂
Comment by ideawerkz — January 4, 2004 @ 2:43 AM
Haha, damn, you are making me look like small boy 🙂
Comment by ideawerkz — January 4, 2004 @ 1:50 AM
not at all! giving credit properly is very important to me, so having done wrong in that aspect, I apologize profusely. No slight on you intended, and I hope none was perceived.
Comment by mulix — January 4, 2004 @ 2:11 AM
You are kind. Thanks 🙂
Comment by ideawerkz — January 4, 2004 @ 2:43 AM
So why does linus refuse to add this main kernel branch ?
Comment by kalyan — January 4, 2004 @ 6:25 AM |
Allow me to point you to an URL:
http://groups.google.com/groups?dq=&hl=en&lr=&ie=UTF-8&oe=UTF-8&threadm=20040103154158.GB5531%40hh.idb.hist.no.lucky.linux.kernel&prev=/groups%3Fdq%3D%26num%3D25%26hl%3Den%26lr%3D%26ie%3DUTF-8%26oe%3DUTF-8%26group%3Dlucky.linux.kernel%26start%3D75
Comment by ideawerkz — January 4, 2004 @ 6:52 AM |
ignorant user
I dont understand OS all that much.. so kindly bear with me
As far as I can understand this is like libsafe, only this one works in passive mode (?). So how does the concept of hijacking the syscall come into the picture ?
Comment by kalyan — January 4, 2004 @ 7:42 AM
Uhm… you’re going to have to be more specific 🙂
Are you asking “why does syscalltrack need to hijack syscalls”, or “why is Linus opposed to hijacking syscalls”?
Comment by mulix — January 4, 2004 @ 2:51 PM
short answer (my interpretation of Linus’s position, not necessarily correct): because allowing modules to hijack system calls makes it far too easy for a binary only module to subvert the system completely. It’s a political issue, not a technical one. I am working on providing a correct technical solution to this problem, and then I *might* want to revisit the political subject. I’m not sure I have the stomach for the flamewar that is bound to ensue, though.
Comment by mulix — January 4, 2004 @ 6:54 AM |