My project wish-list for the next 12 months

[Wilko Bulte]

On Tue, Dec 14, 2004 at 01:47:15PM -0600, Scott M. Ferris wrote..
> On Tue, 14 Dec 2004 09:29:19 +0200, Danny Braniss <danny at cs.huji.ac.il> wrote:
> > Great!, we seem to be on the same wavelength, im now writing (at about one
> > char a minute) the login user program, and somehow - to be discovered -, the
> > socket will be passed to the kernel.
> > my main efford, at the moment, is a) to &^%$$## understand the RFC (i think
> > they used a scrambler) and b) define the data structures.
> 
> How do you plan on handling cases where the user program blocks and
> can't login again (because of a page fault for code or data,
> allocating a new socket in the kernel, allocating a new socket buffer
> in the kernel, etc)?
> 
> One of the major problems any software-only iSCSI initiator has is
> dealing with memory deadlocks.  The OS may try to write out one or
> more pages in order to free up memory.  If the iSCSI initiator needs
> to allocate memory (directly or indirectly in the TCP stack) in order
> to complete that write, you've got a circular dependency where in
> order to get free memory you need to have free memory.  
> Hardware-based iSCSI HBAs solve this by having their own memory and
> TCP stack separate from the OS.  Software-only iSCSI initiators such

Downside: TOE cards are not cheap though..

-- 
Wilko Bulte				wilko at FreeBSD.org