Re: Error report on device malfuncion

Itay Gazit wrote:
> Continue to the irc talk I had with Marty.
> 
> I am doing verification to a driver I wrote under gPXE.
> 
> I noticed that if I disable my device while tftp transfer, the boot process
> stuck! and doesn't report on an error!
> 
> We should discuss, whether gPXE should catch such errors and exit to the
> gPXE shell or ignore them.
> 
> Itay

Hi Itay,

Thanks for writing.  As we discussed on IRC, the current gPXE driver API 
does have a way to indicate to core that there has been a device failure 
that would fatal.

It's something to consider.  I would think that if such a thing were to 
be added, it would likely be added to the _poll() routine, since that is 
what is called to see if there are packets for us.

Let's see if anyone else has any thoughts on this.

Since this would require an API change, I have CC'ed 
Etherboot-Developers so folks there can weigh in.

Marty


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/