understanding the transceiver "RX word aligner pattern"

[IMura]

I am using the Arria10 transceiver using the standard PCS, 8bits FGPA/Standard PCS int width and 10 bit PCS/PMA int width. Other parameters as follow :

10b/8b Encoder and decoder enabled

Rx word aligner mode : Synchronous state machine

Rx word aligner pattern length : 10

RX word aligner pattern :0x17C

I really need help understanding the "RX word aligner pattern :0x17C". If i do a simple loopback and send 0x7C on my 8bit tx parallel data input, '1 ' on my tx_data_k input, there is no 'pattern' detected on the receiving side. However if I write 0xBC instead, pattern is detected and RX word is aligned. Where is this 0xBC specified? what happened to the custom pattern 0x17C?

[CheePin_C_Intel]

Hi, As I understand it, you have some inquiries related to the RX word alignment pattern. For your information, when the 8b10b blocks are enabled, the pattern used in the RX word alignment is the 10b encoded pattern of the 8b input data to TX parallel input. For example, 0x17C is the 10b version of 0xBC. Therefore, at the TX parallel input, you will feed 0xBC. TX will encode this data into 0x17C and send it to the RX through serial link. Please let me know if there is any concern. Thank you.

[IMura]

Thank you! I was expecting that but prior to your response, I forgot to rearrange the bits to be LSB->MSB before checking it on the 10b8b conversion tables. Now everything is clear, thank you.

[CheePin_C_Intel]

Hi, Thanks for your update. Glad to hear that you have managed to resolve the issue. I would set this case to close. Thank you very much.