Based on my (still limited) understanding of the driver source, ECP
writes use the DMA and FIFO because parport_pc_probe_port() at
parport_pc.c:1621 makes parport_pc_ecp_write_block_pio() the write
handler, and this function takes care of turning on and off the FIFO
and DMA.
A note from Wed, 27 Oct 1999 ("ECP Read") explains that the
corresponding code for reading is not used, because of certain
uncertainties about how to deal with the data in the FIFO.
Right now, this is causing a problem for me, and I'd like to pitch in
to fix it, but I need to understand a few things better.
1) What specifically are the problems regarding the FIFO? I have a
set of specs and datasheets that I can consult to see if these can
be resolved.
2) What is the status of the parport_pc_ecp_read_block_pio() code? Is
it up-to-date with the rest of the driver and simply unused, or do
I need to do something to bring it up-to-date?
3) Can it be installed as the ecp_read_data handler in the same way
that is done for write?
4) Is anyone else working on this so I could coordinate with them to
avoid stepping on toes?
Thanks,
Greg Nelson
Princeton Gamma-Tech, Inc.
ghn@qtmsys.com
-- To unsubscribe, send mail to: linux-parport-request@torque.net --
-- with the single word "unsubscribe" in the body of the message. --
This archive was generated by hypermail 2b29 : Thu Mar 09 2000 - 16:06:04 EST