• State Accepted Answer
  • +1 person also asked this people also asked this
  • Locked Locked
  • Replies 3 replies
  • Subscribers 91 subscribers
  • Views 10647 views
  • Users 0 members are here
Options
Related
How was your experience today?
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

AHB DeadLock: HREADY=0 & HTRANS=BUSY

Chanpreet Singh

If the slave asserts wait state (HREADY=0), and master asserts HTRANS =BUSY, then should the slave wait for change of HTRANS from BUSY to NONSEQ/SEQ or should the slave change the HREADY=1.  This condition happens at 'Incrementing burst of undefined length'. This is a deadlock condition; something which is not clear from AHB spec.

Parents
  • +1

    This isn't a deadlock. The slave's data phase HREADY=0 response is for the previous address phase transfer (which must have been a NONSEQ or SEQ transfer), and is not in any way related to the current address phase transfer with HTRANS=BUSY.

    So the slave should drive HREADY however it wants, adding as many wait states as it wants to the data phase of this NONSEQ/SEQ transfer.

    If the master is still signaling HTRANS=BUSY when the current data phase transfer does finally compete (HREADY=1), only then does the HTRANS=BUSY transfer get sampled, and then the requirement that the slave provides a single cycle HREADY=1, HRESP=OKAY response is relevant.

    So definitely no deadlock or contradictory requirements.

Reply
  • +1

    This isn't a deadlock. The slave's data phase HREADY=0 response is for the previous address phase transfer (which must have been a NONSEQ or SEQ transfer), and is not in any way related to the current address phase transfer with HTRANS=BUSY.

    So the slave should drive HREADY however it wants, adding as many wait states as it wants to the data phase of this NONSEQ/SEQ transfer.

    If the master is still signaling HTRANS=BUSY when the current data phase transfer does finally compete (HREADY=1), only then does the HTRANS=BUSY transfer get sampled, and then the requirement that the slave provides a single cycle HREADY=1, HRESP=OKAY response is relevant.

    So definitely no deadlock or contradictory requirements.

Children