• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 98 subscribers
  • Views 247 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TMS320C6746: About UHPI Write Timing

O.H
Mastermind 6430 points
Part Number: TMS320C6746


Hi,

This question is related to "Figure 6-55. UHPI Write Timing (HAS Not Used, Tied High)" in the data sheet (SPRS591F).

After the falling edge of UHPI_HR / W, something like noise occurs, changes from low to high, then changes from high to low again.

In this case, if the time from the falling edge of the noise to the falling edge of HSTROBE satisfies the setup time (1:tsu), is it okay?

*See the figure below

Best regards,

O.H

  • 0
    TI__Guru**** 161600 points

    Hi,

    In this case, if the time from the falling edge of the noise to the falling edge of HSTROBE satisfies the setup time (1:tsu), is it okay?

    I don't think so, because if you have a spike in the UHPI_HR/W  can you guarantee that setup & hold times are met in the next sequence as shown in the figure above. At some point you will start getting corrupted data on the line in my opinion.

    Also did they verify that noise is generated after every falling edge of UHPI_HR / W?

    Best Regards,
    Yordan

  • 0 in reply to Yordan Kovachev
    Mastermind 6430 points

    Hi,Yordan

    Thanks for your reply.

    For the signal of UHPI_HR / W in "Figure 6-55", There is no specification of High / Low in the grayed out part,so we believe that the signal before the falling edge ofUHPI_HR / W is unspecified.

    Therefore, I thought that there was no problem with the signal as shown in the figure below, but does it mean that there is a possibility of acquiring corrupted data due to noise or something, and that the operation cannot be guaranteed?

    >Also did they verify that noise is generated after every falling edge of UHPI_HR / W?

    I will check it.

    Best regards,

    O.H

  • 0 in reply to O.H
    TI__Guru**** 161600 points

    Hi,

    Having in mind the description of the UHPI_HR/W signal from Section 20.2.3 Signal Descriptions in the device TRM:
      "HPI read/write. On the falling edge of internal HSTRB, UHPI_HR/W indicates whether the current access is to be a read or write operation.
    Driving UHPI_HR/W high indicates the transfer is a read from the HPI, while driving UHPI_HR/W low indicates a write to the HPI."

    I'd say that there is a possibility to corrupt the data transfer through the HPI at some point if this noise is present.

    >Also did they verify that noise is generated after every falling edge of UHPI_HR / W?

    I will check it.

    Let me know when you get customer's feedback.

    Best Regards,
    Yordan