FPGA Intellectual Property
PCI Express*, Networking and Connectivity, Memory Interfaces, DSP IP, and Video IP
Announcements
Intel Support hours are Monday-Fridays, 8am-5pm PST, except Holidays. Thanks to our community members who provide support during our down time or before we get to your questions. We appreciate you!

Need Forum Guidance? Click here
Search our FPGA Knowledge Articles here.
5950 Discussions

How to send Electrical IDLE in transceiver lines?

HBhat2
New Contributor II
303 Views

Hi,

 

Using Stratix 10 Native PHY IP, I am interfacing to custom PCIe Logic and custom PCS to transmit PCIe like traffic using transceiver lanes. Here, I have some concerns on Electrical IDLE.

1) As per my understanding, Electrical IDLE can be achieved if the transmitter FIFO becomes empty. Whether my understanding is correct? or any other mechanism should be followed to achieve electrical IDLE?

2) Once the transceiver is out of Electrical IDLE do I need to re-calibrate the transceivers? Also, the word align logic should be re-run to align the word boundary (currently using bit slip logic to align the data word for the first time) after exiting the Electrical IDLE state?

 

With Regards,

HPB

0 Kudos
2 Replies
SengKok_L_Intel
Moderator
102 Views
Hi HPB, Below are my answer: 1. As per the PCIe Spec, to enter the ELECTRICAL IDLE stage, the link partner is required to send the EISO( Electrical Idle Order Set), and it can't achieve if somehow the transmitter FIFO is empty. 2. Exist from the Electrical Idle, the transceiver does not require to perform re-calibration but the word alignment has to re-run. Regards -SK
HBhat2
New Contributor II
102 Views

Hi @SengKokL_Intel​ ,

1) As I understand the spec, in electrical IDLE state, the common mode voltage shall be maintained in the serial lines. So, with respect to S-10 transceiver, in what way I can maintain common mode voltage in the transceiver serial line, but no toggling.

2) "the transceiver does not require to perform re-calibration but the word alignment has to re-run. " Now I understand and agree with this statement.

 

With Regards,

HPB

Reply