This happens in certain programs (e.g. __builtin.*Error and some Nettle
hash states) where low_inherit is called with name == NULL.

The test looked at limits for the wrong program (the refno relates
to the program at depth depth).

Revert it for now as it breaks working code.

This reverts commit c88883e0.

Fixes dumping of eg 7.4::Stdio.File.

Also fixes master()->Encoder::nameof() for module classes.

Fixes dumping of eg 7.8::Stdio.File.

This file was missed in the commit.

Fixes issue with getting "Parent lost, cannot clone program." from
SSL.handshake()->*derive*_master_secret() in dumped Pikes.

Note: No changes were needed to decode_value().

Use program.c:lower_inherit() in decode_value(). The new
parent flags handling in low_inherit() isn't safe to use
from decode_value(), since the program_state stack isn't
necessarily the same as when the encoded program was
initially compiled.

Fixes various dumping errors.

Renumber the low PIKE_T_* values so that PIKE_T_INT becomes zero.

This has the feature that zeroed memory becomes filled with Pike
svalues containing integer zeroes (and not NULL pointer arrays).
This will let call_c_initializers() avoid traversing the entire
identifier table for the class.

Note: The serialized representation of types (__parse_pike_type())
      is unchanged. As is the {out,in}put for {en,de}code_value().

Updates the code generators for ia32 and amd64.

Breaks the code generators for ppc32 and ppc64.

There was a potential NULL-deref when decoding destructed objects.

down on compiler warnings. The macro also renames parameters to catch accidental use.
(There are more places to clean up but I don't want to modify code that isn't compiling
on my machine.)

(encode|decode)_value now instantiates a local codec instance from
master()->Encoder or master()->Decoder, respectively. That to make it
possible to keep local state. Specifically, it fixes the problem that the
reverse lookup mappings in the encoder get populated way too early when the
master is used directly.

The codecs are only instantiated on-demand, though.

Also fixed some confusing error messages.

It's now again somewhat possible to use it to force programs through
decode_value even if the codec returns completely bogus things.

Rev: src/encode.c:1.294
Rev: src/object.c:1.309
Rev: src/program.c:1.778

Fixes dumping core due to running out of C-stack for recursive data in decode_value().

Rev: src/encode.c:1.293

Rev: src/encode.c:1.292

Rev: src/encode.c:1.291

Rev: src/encode.c:1.290