Hi
I came across the L1D$ acronym in the C6472 and C6455 errata. How is this different than the L1D cache (without the "$"). The L1D$ is found in about 20 places in the errata doc.
Under a specific set of circumstances, a snoop-write updates unintended data being
allocated into L1D$ from external, cacheable memory. This can lead directly to program
misbehavior. If that line is then modified by CPU accesses, a subsequent victim
writeback from L1D could commit this corrupted line to lower levels of memory. The key
requirements for this issue are:
(I could not find L1D$ in the cache users guide)
Cheers
