Re: [patch] forcedeth: fix locking bug with netconsole

[Ingo Molnar]

* Jeff Garzik <jgarzik@xxxxxxxxx> wrote:

> Ingo Molnar wrote:
>> Subject: forcedeth: fix locking bug with netconsole
>> From: Ingo Molnar <mingo@xxxxxxx>
>>
>> while using netconsole on forcedeth, lockdep noticed the following
>> locking bug:
> [...]
>> the fix is to disable/restore irqs instead of disable/enable.
>
> So we make _the_ TX fast path slower, to accomodate a path that is 
> rarely executed?

no strong opinion but a single PCI mmio access far outweighs any flags 
save/restore, so i'd not fight about something like that. I've never 
seen flags save/restore costs show up in NMI-driven profiles that had 
networking load.

	Ingo
--
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/