Re: CONFIG_PREEMPT_SOFTIRQS

From: Steven Rostedt <rostedt@xxxxxxxxxxx>
Date: Tue, 31 Jan 2006 13:10:55 -0500

On Tue, 2006-01-31 at 20:19 +0530, Vinod KK wrote:
> Hi,
>
> I am using linux 2.6.10 (with Ingo Molnars real time patches) on my mips board.
>
> When I try running high network traffic (10% traffic on a 100Mbps
> link) on my target board with the CONFIG_PREEMPT_DESKTOP=y and
> CONFIG_PREEMPT_SOFTIRQS=y options enabled I notice that the console
> freezes and recovers only after the traffic stops. I do not notice
> this behaviour with the CONFIG_PREEMPT_SOFTIRQS option disabled.
>
> I understand that the CONFIG_PREEMPT_SOFTIRQS option puts all softirq
> processing to the ksoftirqd daemon, but i do not see why this should
> stall the console. I believe there should be some task which is
> starving for CPU, but I do not know which one.
>
> The console is on an 16550 serial port.
>
> Could someone please give me some pointers on where I should start looking?

2.6.10 with Ingo's patch is pretty old. The -rt patch is very much a
moving target today and is not yet stable. So the only real answer I
can give you is to go ahead and download
http://people.redhat.com/mingo/realtime-preempt/patch-2.6.15-rt16
and try that out.

-- Steve

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

References:
- CONFIG_PREEMPT_SOFTIRQS
  - From: Vinod KK

Prev by Date: Re: 2.6.15-rt16
Next by Date: Re: [PATCH 8/12] generic hweight{32,16,8}()
Previous by thread: CONFIG_PREEMPT_SOFTIRQS
Next by thread: [PATCH] OOM kill: children accounting
Index(es):
- Date
- Thread

Relevant Pages

CONFIG_PREEMPT_SOFTIRQS
... I am using linux 2.6.10 (with Ingo Molnars real time patches) on my mips board.
... When I try running high network traffic (10% traffic on a 100Mbps ... CONFIG_PREEMPT_SOFTIRQS=y
options enabled I notice that the console ... (Linux-Kernel)
[PATCH] interrupt driven hvc_console as vio device (final)
... Here is the revised hvc_console patch which fixes the issues you were ... This
is an hvc_console patch which provides driver and ppc64 ... console output functionality
prior to early console init (pre mem init ... -Added early discovery of vterm/vty adapters
by doing a bus walk on ... (Linux-Kernel)
Re: TIOCCONS security revisited
... > In August 2004, Olaf Dabrunz posted a patch, which appears to have got ...
The problem is that TIOCCONS causes security problems. ... the ioctl TIOCCONS allows
any user to redirect console output to ... (Linux-Kernel)
Re: kern/71142; VESA 1024x768; (Poor Console resolution on Compaq 2103US)
... I chose to stick with FreeBSD over Slackware ... >for my laptop for
the simplicity in its init system ... >proper console resolution was a pain as I favour
... >still some issues that the patch created in making use ... (freebsd-current)
Re: kern/71142; VESA 1024x768; (Poor Console resolution on Compaq 2103US)
... I chose to stick with FreeBSD over Slackware ... >for my laptop for
the simplicity in its init system ... >proper console resolution was a pain as I favour
... >still some issues that the patch created in making use ... (freebsd-newbies)