Ok, here you go; the explanation is below. This is from net-2.6.24 of
early this AM. I saw a patch you posted that is derived from Krishna;
although it hasnt showed up in the tree - i have considered those
changes and this patch adds a little more optimization in case of
errors.

I will send you a patch to kill LLTX the sooner this shows up somewhere.

cheers,
jamal

On Wed, 2007-03-10 at 09:18 -0400, jamal wrote:

> The cleanup is to break up the code so it is functionally more readable
> from a perspective of the 4 distinct parts in ->hard_start_xmit():
> 
> a) packet formatting (example: vlan, mss, descriptor counting, etc.)
> b) chip-specific formatting
> c) enqueueing the packet on a DMA ring
> d) IO operations to complete packet transmit, tell DMA engine to chew
> on, tx completion interrupts, set last tx time, etc.
> 
> Each of those steps sitting in different functions accumulates state
> that is used in the next steps. cb stores this state because it a
> scratchpad the driver owns. You could create some other structure and
> pass it around the iteration, but why waste more bytes.