Messages in this thread | | | From | est@hyperrea ... | Subject | fs (ext2) latency constructive despair :) | Date | Sat, 31 Jul 1999 09:18:24 -0700 (PDT) |
| |
Greetings, people,
I've found that unlinking a 200M file leads fairly reliably on my system to >50ms drop-outs (as reported by Benno's latencytest (www.gardena.net/benno/linux/audio) or my own rtcskips). This holds whether I use 2.2.10 or 2.3.11, with or without Andrea and Ingo's patches, using scsi or ide drives.
I can see why the unlink operation would take that long, but I'm having trouble seeing why POSIX real-time processes wouldn't be able to run at any point during it. Can anyone tell me where exactly they're getting locked out? Is my understanding that lock_kernel() is a no-op on a uniprocessor incorrect? I wish to move to a more advanced state of ignorance on this topic. :)
Best regards and thanks in advance,
Eric
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |