You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
If ddtc_miss(sva internal signal) rises then in the same cycle ddt_walk should go high as well.
Below is the counter-example of the assertion failure, you can see that ddtc_miss rises in cycle 16 but ddt_walk didn't rise here. After doing a "why analysis" I found the reason for assertion failure. The reason for the failure is that while loading dc when we have a translation of pdtp.ppn, trans_type is UNTRANSLATED_RX and pte.x == 0 that should result in a page fault exception but IOMMU asserted the updated_dc_o signal which is wrong.
The text was updated successfully, but these errors were encountered:
I wrote an assertion that states:
Below is the counter-example of the assertion failure, you can see that
ddtc_miss
rises in cycle 16 butddt_walk
didn't rise here. After doing a "why analysis" I found the reason for assertion failure. The reason for the failure is that while loading dc when we have a translation of pdtp.ppn, trans_type is UNTRANSLATED_RX and pte.x == 0 that should result in a page fault exception but IOMMU asserted theupdated_dc_o
signal which is wrong.The text was updated successfully, but these errors were encountered: