Master sends EchoReply for unknown reasons #344
Labels
No Milestone
No Assignees
3 Participants
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: M-Labs/artiq-zynq#344
Loading…
Reference in New Issue
Block a user
No description provided.
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
#322 seemed to touch on the topic before, but it was a different issue causing the timeouts with 2 satellites.
As described in https://forum.m-labs.hk/d/775-destn-communication-failed-timed-out-with-artiq8-zynq
Reproduced with Kasli-SoC master and both 2.0 and SoC satellite by @srenblad
Happens exactly with
self.core.wait_until_mu(now_mu())
running.Since that is a loop that is constantly reading the RTIO counter from CSR, it seems like there may be a conflict and data on the AXI/CSR bus may be incorrect, when read/written by two cores at the same time.
Not relevant for configs with Kasli v2 as master.
If it's AXI data corruption it can produce a bunch of problems on every variant, just more subtle ones.
See also https://forum.m-labs.hk/d/854-drtio-link-issue-after-gateware-update
We should figure out what is actually going on.