Messages in this thread | | | Date | Sun, 18 Oct 1998 09:03:43 +0200 (CEST) | From | MOLNAR Ingo <> | Subject | jiffies wraparound |
| |
On Sun, 18 Oct 1998, Alan Cox wrote:
> > It's a beauty wart, it's been there forever, and we know that machines > > actually tend to survive it fine. So far I have _one_ person who has > > 1.2 survives mostly, 2.0 survives mostly, 2.1.x doesnt survive
well i have tried to simulate a jiffies wraparound with pre2-126, and it has survived more or less. There are a few strange things that happen near and after the wraparound:
- the IDE driver got confused somewhat:
8:49am up 496 days, 23:47, 5 users, load average: 0.57, 0.43, 0.22
hda: timeout waiting for DMA hda: irq timeout: status=0xd0 { Busy } hda: DMA disabled changing jiffies from fffefd63 to ffff0d63. ide0: reset timed-out, status=0xd0 hda: status timeout: status=0xd0 { Busy } hda: drive not ready for command ide0: reset: success
- when the wraparound happens, all network connections and all terminal lines are killed at once.
- gpm goes crazy, mouse is very jumpy, but this can be fixed by restarting gpm.
otherwise things look mostly fine, no network connection is stuck, no process is hanging, no kernel oops. The time.c change in 2.1.126 has indeed made time handling more robust, while 2.1.125 oopses with division by zero, the 2.1.126 time code survived this quick 'time travel' without problems. Apart from processes getting killed during wraparound, there is no big problem, is there?
-- mingo
- 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/
| |