S27ks0641-$skew error for CSNreg

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
clab_4481021
Level 1
Level 1

Hi,

I try to integrate the verilog model for S27ks0641 with the synopsys SSI IP.

I got the following:

Warning!  Timing violation
           $skew( negedge CSNeg:414195 PS, posedge CSNeg:498597 PS,  1 : 1 PS );
            File: s27ks0641.v, line = 369
           Scope: ssi_tbe.mem_model
            Time: 498597 PS

I modified the value in the verilog file of tskew_CSNeg_CSNeg from 1 (tCSM) to 100000 to avoid it.

we run in 166MHz and the time between fall(CSNeg) and rise(CSNeg) can not be 1 ps

not sure why this value was set to such a short time?

the skew command requires a time interval of 1 ps max for CS to be active - that is not enough to deliver a Hyperbus frame.

can you please clarify?

Best wishes,

Cluny

0 Likes
1 Solution

Hi,

Could you please let me know, whether you are using the latest Verilog model for S27KS0641 available at: https://www.cypress.com/documentation/models/verilog/s27kl0641-s27ks0641-verilog ? If not, please use this model for your simulations and let us know the results.


Thanks and Regards,

Sudheesh

View solution in original post

0 Likes
3 Replies
SudheeshK
Moderator
Moderator
Moderator
250 sign-ins First question asked 750 replies posted

Hi,

Please find our comment about your query below.

"The tcsm (CS# Low maximum) time is checked with this. It is covered in the spec and recommended value is 4 us. In the Verilog sdf:

(SKEW (negedge CSNeg) (posedge CSNeg) (4000))

What is the timescale you have used? SDF timescale is set to 1 ns.

About tdevice time, as usual we do not recommend changing this because we haven't tested it, but in this case I guess it is ok. There is no other way of doing it.

Other option is to use VHDL model where tdevice_VCS parameter can be changed through instantiation."

Thanks and Regards,

Sudheesh

0 Likes

Hi Sudheesh and thank you for your reply.

regarding the tcsm:

i am using timescale of 1ps

as we might send long write/read trans the time of the assertion of CS (low) can be 3000ns end even more - so i do not undesrtand why the tcsm was set to 1ps originally (even 1 ns is too short).

as i said - i have changed it and it solved - but i want to make sure i didn't violate something and that we wont be failing when we use the real HyperRam with its defined timing.

So how a value of tcsm=1 can be enough to deliver long streams?

best wishes,

Cluny

0 Likes

Hi,

Could you please let me know, whether you are using the latest Verilog model for S27KS0641 available at: https://www.cypress.com/documentation/models/verilog/s27kl0641-s27ks0641-verilog ? If not, please use this model for your simulations and let us know the results.


Thanks and Regards,

Sudheesh

0 Likes