This involves two redesigns of the map.c implementations:
1. Change of macro style and code organization
The old khash.h and map.c implementation used huge #define blocks with a
lot of backslash line continuations.
This instead uses the "implementation file" .c.h pattern. Such a file is
meant to be included multiple times, with different macros set prior to
inclusion as parameters. we already use this pattern e.g. for
eval/typval_encode.c.h to implement different typval encoders reusing a
similar structure.
We can structure this code into two parts. one that only depends on key
type and is enough to implement sets, and one which depends on both key
and value to implement maps (as a wrapper around sets, with an added
value[] array)
2. Separate the main hash buckets from the key / value arrays
Change the hack buckets to only contain an index into separate key /
value arrays
This is a common pattern in modern, state of the art hashmap
implementations. Even though this leads to one more allocated array, it
is this often is a net reduction of memory consumption. Consider
key+value consuming at least 12 bytes per pair. On average, we will have
twice as many buckets per item.
Thus old implementation:
2*12 = 24 bytes per item
New implementation
1*12 + 2*4 = 20 bytes per item
And the difference gets bigger with larger items.
One might think we have pulled a fast one here, as wouldn't the average size of
the new key/value arrays be 1.5 slots per items due to amortized grows?
But remember, these arrays are fully dense, and thus the accessed memory,
measured in _cache lines_, the unit which actually matters, will be the
fully used memory but just rounded up to the nearest cache line
boundary.
This has some other interesting properties, such as an insert-only
set/map will be fully ordered by insert only. Preserving this ordering
in face of deletions is more tricky tho. As we currently don't use
ordered maps, the "delete" operation maintains compactness of the item
arrays in the simplest way by breaking the ordering. It would be
possible to implement an order-preserving delete although at some cost,
like allowing the items array to become non-dense until the next rehash.
Finally, in face of these two major changes, all code used in khash.h
has been integrated into map.c and friends. Given the heavy edits it
makes no sense to "layer" the code into a vendored and a wrapper part.
Rather, the layered cake follows the specialization depth: code shared
for all maps, code specialized to a key type (and its equivalence
relation), and finally code specialized to value+key type.
This reduces the total number of khash_t instantiations from 22 to 8.
Make the khash internal functions take the size of values as a runtime
parameter. This is abstracted with typesafe Map containers which
are still specialized for both key, value type.
Introduce `Set(key)` type for when there is no value.
Refactor shada.c to use Map/Set instead of khash directly.
This requires `map_ref` operation to be more flexible.
Return pointers to both key and value, plus an indicator for new_item.
As a bonus, `map_key` is now redundant.
Instead of Map(cstr_t, FileMarks), use a pointer map as the FileMarks struct is
humongous.
Make `event_strings` actually work like an intern pool instead of wtf it
was doing before.
problem: can we have Serde?
solution: we have Serde at home
This by itself is just a change of notation, that could be quickly
merged to avoid messy merge conflicts, but upcoming changes are planned:
- keysets no longer need to be defined in one single file. `keysets.h` is
just the initial automatic conversion of the previous `keysets.lua`.
keysets just used in a single api/{scope}.h can be moved to that file, later on.
- Typed dicts will have more specific types than Object. this will
enable most of the existing manual typechecking boilerplate to be eliminated.
We will need some annotation for missing value, i e a boolean will
need to be represented as a TriState (none/false/true) in some cases.
- Eventually: optional parameters in form of a `Dict opts` final
parameter will get added in some form to metadata. this will require
a discussion/desicion about type forward compatibility.
Allow Include What You Use to remove unnecessary includes and only
include what is necessary. This helps with reducing compilation times
and makes it easier to visualise which dependencies are actually
required.
Work on https://github.com/neovim/neovim/issues/549, but doesn't close
it since this only works fully for .c files and not headers.
