#18064: Workaround for multiple matmul hangs:

[ncvetkovicTT]

Ticket

#18064 #18065 #17099 #16673

Problem description

Disabling instruction cache doesn't happen in time for subsequent TTI instruction not to end up in the cache. In order to properly disable I$, we need to disable branch prediction first. Since reprogramming the REPLAY buffers needs to happen when the cache is disabled, SFPI compiler cannot rely on REPLAY buffers. These things introduce multiple matmul hangs.

What's changed

• Guard CSRRS by disabling branch prediction
• Add a compiler flag which makes the SFPI compiler not to use replay buffers

Checklist

• All post commit CI passes
• Blackhole Post commit CI passes (if applicable) - run
• Model regression CI passes (if applicable)
• Device performance regression CI passes (if applicable)
• (For models and ops writers) Full new models tests CI passes (if applicable)
• New/Existing tests provide coverage for changes

[ncvetkovicTT]

@nathan-TT Is the place where I need to add the compiler flag correct?

[nathan-TT]

@nathan-TT Is the place where I need to add the compiler flag correct?

looks right.