|
|
Reopened every database and read back what is actually stored rather than
trusting the write reports: note lines, renamed command handlers, renamed RPC
thunks and entry points for all twelve TAs. Distinct handler counts are lower
than command-id counts wherever ids share a handler, as with RAP's seven ids
funnelling into one and OPM's 6 and 7 sharing one through a goto.
Noted that multi-line function comments are stored fine and only the listing view
renders the first line inline, while pseudocode shows all; checking the listing
alone briefly looked like truncation.
Also recorded the operational cause of a real failure: tmz spawned once and then
failed on every later attempt, with the pane present in the registry but absent
from slosh and readiness never arriving. TMZ is the stub with no real entry
function, and idatui documents that such a binary can land in the symbol palette
on startup, so the saved database restores that modal and the readiness probe
never completes. Deleting the .i64 and reopening works immediately. Worth
recording because the symptom looks like a spawn bug rather than restored state.
|
