2026-08-29 - A setting consulted nowhere, for the fourth time
payload_args now holds the same handoff block the
declared-entry path is handed, so the two modes agree about what the
guest is looking at. It did not move the wall; it is recorded because it
is right rather than because it worked.
The second finding is the one that matters. Forcing
vsnprintf to answer a value did nothing at all -
no report, no change, no sign the knob had been read.
ThunkTable::len means the guest's imports, deliberately, so
a report cannot flatter itself by counting everything this emulator can
answer (D366) - and every diagnostic was sized by that number, so
by-name resolutions fell outside all of them.
Fourth time a setting has been consulted nowhere: D082, D166, D187, and now this. Each time the knob existed, was set, reached nothing, and the run reported no change - which reads exactly like a measurement.
The reports keep len; the diagnostics take
total. It paid immediately: the argument dump fired for the
first time and named the format klogsrv renders -
%s:%d:%s: %s - which is klog_printf's file,
line, function and message, four conversions, all inside the register
half.
The -1 read is still unexplained after six eliminations.
Every pointer into that path is guarded and the fault is unchanged, at a
fixed offset in our own binary. Next step is to look at where in
our code it faults - something this project can do to itself
and has never needed to.