Date: Sun, 09 Feb 2003 23:11:25 -0800 From: Terry Lambert <tlambert2@mindspring.com> To: Nate Lawson <nate@root.org> Cc: current@freebsd.org Subject: Re: bus_setup_intr() vs. ether_ifattach() race Message-ID: <3E47509D.EA90194A@mindspring.com> References: <Pine.BSF.4.21.0302092214480.29408-100000@root.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Nate Lawson wrote: > Which is the correct order to do these two functions? If the irq is > enabled before the device is attached, it seems a response cannot be sent > if a packet arrives before the attach. The right way seems to be to > attach the device before setting up an irq but does this have side > effects? The main side effect is that probes that require IRQs in order to determine if the device is there (e.g. AMD Lance, etc.) will never be able to probe true, unless the interrupt is routed to an IRQ handler. On the other hand, it should probably not be the standard IRQ handler that gets invoked, as a result of the probe, but one that is specific to probing. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3E47509D.EA90194A>