From: Matt Mackall <[EMAIL PROTECTED]> Date: Tue, 6 Sep 2005 15:32:50 -0700
> Unfortunately, netpoll fundamentally requires everything to work with > interrupts off. This can't be changed. It could be called from the > context of an oops, or worse, by the kgdb stub at a breakpoint. If > drivers require interrupts enabled for hard_start_xmit, they simply > can't work with netpoll. I really sense that nepoll needs to "fundamentally" change to defer all of it's actions to software interrupt context. Yes, you lose some reliability in cases I've described in other postings to this thread, but the locking problems all vanish and you no longer need to disable interrupts illegally when calling these core driver functions. It is illegal not just from a latency perspective, drivers really do want inerrupts enabled during ->hard_start_xmit() just to _function_ correctly. Some spin on jiffies increasing for example, and that's perfectly fine and needs to work properly. Mentioning the latency of the serial console, in support of netpoll's interrupt disabling, is quite a straw man. - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
