Nor Flash

[WiMos]

I had an issue with the 4READ command.

Data is output one clock later than expected.

It seems that the change_addr signal is to blame: it fires depending on the ADS bit at 24 or32 bit boundary.

For commands configurable between 3/4 byte addresses this is correct, but i.m.o. 4READ always is 4 byte address,

so change_addr should always fire at32 bit boundary.

Please check whether assumption above is correct.

Regards,  Wim

My code mod to remove issue:

IF (current_state /= STANDBY) AND
    (Instruct /= WRR) AND (Instruct /= WRAR) THEN
 IF ((Instruct = READ) AND to_nat(WRR_reg_in) = 0) OR Instruct /= READ THEN
-- IF ADS = '1' AND address_cnt = 32 THEN
    IF address_cnt = 32 THEN
      change_addr <= '1', '0' AFTER 1 ns;
    ELSIF ADS = '0' AND address_cnt = 24 THEN
      change_addr <= '1', '0' AFTER 1 ns;
   END IF;
END IF;

[AS_36]

Hi Wim,

1. Could you please tell me the version of the model you are using?
2. Could you also tell me whether you are using SDF file or not?

Thank you and Regards.

[WiMos]

Hi,

I do not use the SDF file, only vhdl sources, compiled in Questasim.

Model version:

# Release Date (DD/MM/YYYY) => 08/06/2018
# Release Version => 1.3

[AS_36]

Hi,

I would request you to repeat the simulation using the SDF file. If the issue persists, please do let me know.

Regards.