Timeline for Why does my process take too long to die?
Current License: CC BY-SA 3.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 7, 2015 at 8:49 | comment | added | Anirban Ghoshal | @ott--On the original application, yes. However, as I mentioned, the delay was from exit_mmap() in Linux, which would all be accounted in strace under exit()/exit_group(). For the new program, whose source I've shared, I already added gettimeofday() calls before and after the munlockall(). | |
| Jul 6, 2015 at 21:20 | comment | added | ott-- | Have you already run your program with strace? It has some profiling options too. | |
| Jul 6, 2015 at 20:04 | comment | added | Anirban Ghoshal | Since in-house would involve sharing problems, I've written a program just to study this problem. Please see my edit to the original question. | |
| Jul 6, 2015 at 20:02 | history | edited | Anirban Ghoshal | CC BY-SA 3.0 | Added a whole lot more detail. |
| Jun 30, 2015 at 12:31 | answer | added | BillThor | timeline score: 3 | |
| Jun 30, 2015 at 12:20 | comment | added | Anirban Ghoshal | Also, my machine is a diskless system - completely based out of tmpfs, etc. | |
| Jun 30, 2015 at 12:19 | comment | added | Anirban Ghoshal | Fair enough. What I'm looking for is some advice on this. My application allocates around 2.25 G of memory and then proceeds to mlock() down upon it. What I suspected initially was maybe pending data transfer, etc, and so I closed all open fd's (by handling the SIGTERM). It didn't make an iota of difference. So, after that we added printk's in the kernel, and from that we found out that all delay comes from exit_mm. The delay is more or less uniform across the VMA's as the locked count in /proc/meminfo decreases uniformly. My application does fragment VM a lot, but 2 minutes is just too high. | |
| Jun 30, 2015 at 12:12 | comment | added | lcd047 | No: what happens at exit depends essentially on the state of the application before exit. | |
| Jun 30, 2015 at 12:10 | comment | added | Anirban Ghoshal | Nothing basically. I'm talking about all kinds of exits, even those upon receiving unhandled signals (KILL). What happens after that should be outside the scope of the in-house application? | |
| Jun 30, 2015 at 12:08 | review | First posts | |||
| Jun 30, 2015 at 12:33 | |||||
| Jun 30, 2015 at 12:08 | comment | added | lcd047 | It's an in-house application and you're asking on unix.SE what your application does when it exits? | |
| Jun 30, 2015 at 12:05 | history | asked | Anirban Ghoshal | CC BY-SA 3.0 |