I've been having a problem with vanilla IPC Notify. It's been plaguing us for months, but has become more critical and after a week I haven't been able to identify it.
This is a development from a previous thread here: e2e.ti.com/.../lp-am243-freertos-task-crashes-without-errors-stack-issue-or-something-else
For a long time with out project it was working flawlessly across all R4F cores and the M4F core. Then, using one launchpad sending a message to one specific core caused it to hard fault - sometimes the interrupt handler would at least start to execute, other times it wouldn't start at all. Early on I could even step through the execution, which strangely would avoid any hard fault from occurring. Eventually this stopped happening and we haven't been able to replicate it since.
Then we noticed that different launchpads would work differently with the same code. Currently I have access to two boards: one where an IPC message to any core will cause that core to hard fault, and another where only two of the R4F cores will hard fault but the others work without problems.
My inclination is to believe there's a hardware problem and our boards are just old(?) given the same code on two cores is operating differently (we've ordered some new boards from TI which are on their way) but I'm not convinced as I can still run the IPC Notify demo without problems.
What more can I do to diagnose this further in the meantime? Our development has stalled entirely since our cores can't coordinate with one another anymore. Thanks in advance.
