On Mon, 2011-01-31 at 13:36 +0800, Adam Wang wrote:
> While running Flickernoise, the FLASH_RESET_N doesn't act(goes low)
> under conditions below:
> 1, press 'Reboot' button

It is supposed to pulse low at this time. Either you missed the pulse
(it's only a few hundred nanoseconds) or there is a bug somewhere in the
FPGA design.

Pressing the 3 buttons at the same time should also reset, again with a
low pulse on FLASH_RESET_N.

> 2, press 'Power off' button

The flash does not reset there, this is expected.

> Thus the RP# (Reset) is ONLY occurred while M1 power on. If I am right
> then I can directly cut the route from FPGA then solder it connected
> to reset ic.

Please don't - we need the ability to reset the flash from the FPGA in
order to be able to recover from an interrupted write sequence.

>  Otherwise this cut will become complicated( thus is to add another
> AND gate for accepting both output from reset ic and fpga then AND
> gate's output goes to flash RP# pin).

To implement then AND, can't we just use a pull-up resistor and the open
drain version of the reset IC connected in parallel on the FPGA pin?
Before configuration, the FPGA has a weak pull-up on this pin which
would not interfere with the reset IC, and after configuration I can do
whatever I want with this pin.

S.


_______________________________________________
http://lists.milkymist.org/listinfo.cgi/devel-milkymist.org
IRC: #milkymist@Freenode
Twitter: www.twitter.com/milkymistvj
Ideas? http://milkymist.uservoice.com

Reply via email to