[ipxe-devel] SAN Boot Windows XP with USB-NIC.

[Muralidhar.Appalla at Emulex.Com]

Michael,

I am sorry I did not closely looked at your statement. Yes, this is going to break our UNDI driver. We give IRQ as non zero in UNDI_GET_INFO structure but flags will be saying IRQ_NOT_SUPPORTED. So you probably cannot use the logic that you put.

Thanks
Murali


-----Original Message-----
From: Michael Brown [mailto:mbrown at fensystems.co.uk] 
Sent: Thursday, March 10, 2011 3:22 PM
To: Appalla, Muralidhar
Cc: jaspers.cheng at msa.hinet.net; ipxe-devel at ipxe.org
Subject: Re: [ipxe-devel] SAN Boot Windows XP with USB-NIC.

On Thursday 10 Mar 2011 23:04:59 Muralidhar.Appalla at emulex.com wrote:
> Yes, our device supports IRQ. This information that is populated from UNDI
>  is directly read from PCI Configuration space and populated. But we do not
>  enable interrupts in our UNDI stack but rather do polling. Our OS drivers
>  definitely use this IRQ.
> 
> The statement to check whether IRQ is zero seems to me also very safe to
>  do.

Sorry, I think I'm misunderstanding something.  You say that the information 
provided in UNDI is directly read from PCI configuration space, in which case 
it *will* include a non-zero IRQ number.  That means that if I apply the 
following logic:

> > i.e. assume that the NIC supports interrupts if *either* the ServiceFlags
> > indicate SUPPORTED_IRQ *or* the IRQ number is non-zero.

then that would cause iPXE to assume that your UNDI driver *does* generate 
interrupts, and so it will fail.

It sounds as though this idea (to assume that the NIC supports interrupts if 
either the ServiceFlags indicate SUPPORTED_IRQ or the IRQ number is non-zero) 
would *break* iPXE running over the Emulex UNDI stack.

Am I missing something?

Michael