Timeline for Help with identifying race condition in wargame
Current License: CC BY-SA 3.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 20, 2014 at 9:54 | comment | added | Peter Andersson | @PieThon It is possible to exploit this on a single core processor. I've done it. It's just a matter of spawning a lot of these "race" processes in parallel and hoping that one process preempts the other on the exact right instruction. It's all an odds game. I did it with 500 parallel race processes running in the background. Then I brought the one surviving process with sh back to the foreground. | |
| Apr 19, 2014 at 18:11 | comment | added | PieThon |
Thanks a lot for the info - I was also trying to get the program to enter the win block with ./a python -c 'print "000badc0ffffffff ffffffffdedecade "*20000' but wasn't having much luck with it since I ran it a bunch of times with no "Win". I'm glad to hear I'm on the right track. The machine I'm running on is single core, is there anything besides spawning multiple other processes to eat CPU that I can do to cause the program to switch threads more frequently?
|
|
| Apr 19, 2014 at 18:09 | vote | accept | PieThon | ||
| Apr 19, 2014 at 16:07 | history | edited | Peter Andersson | CC BY-SA 3.0 |
added 1742 characters in body
|
| Apr 19, 2014 at 12:05 | history | edited | Peter Andersson | CC BY-SA 3.0 |
added 554 characters in body
|
| Apr 19, 2014 at 11:58 | history | edited | Peter Andersson | CC BY-SA 3.0 |
added 554 characters in body
|
| Apr 19, 2014 at 9:34 | history | edited | Peter Andersson | CC BY-SA 3.0 |
added 154 characters in body
|
| Apr 19, 2014 at 9:16 | history | answered | Peter Andersson | CC BY-SA 3.0 |