Messages in this thread | | | Date | Tue, 18 Oct 2005 22:26:10 +0200 | From | Ingo Molnar <> | Subject | Re: 2.6.14-rc4-rt7 |
| |
* Daniel Walker <dwalker@mvista.com> wrote:
> > > I found this latency ~5 minutes after boot up, no load . It looks like > > > vgacon_scroll() has a memset like operation which can grow. > > > > do you have PRINTK_IGNORE_LOGLEVEL enabled? If yes then much of the > > printk code will run with interrupts disabled - hence non-preemptable. > > PRINTK_IGNORE_LOGLEVEL is a debugging feature for developers. I have > > added an extra explanation to the Kconfig, see below. > > I was just running a "make defconfig" and I enabled latency tracing . > PRINTK_IGNORE_LOGLEVEL defaults to off (doesn't it?), and I didn't > make any changes to it . Unless it get switched on by something else > .. > > It looks like the logic might be reversed in release_console_sem() .
yeah, it is reversed. I 'fixed' it a couple of releases ago - but in fact it was correct previously and i introduced this bug. I have fixed this in my tree and have released -rt9 with the fix.
Ingo - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |