Started at 349,880,038 allocations and 42s
- Switched to making `Symbol` in GOOS be a "fixed type", just a wrapper
around a `const char*` pointing to the string in the symbol table. This
is a step toward making a lot of things better, but by itself not a huge
improvement. Some things may be worse due to more temp `std::string`
allocations, but one day all these can be removed. On linux it saved
allocations (347,685,429), and saved a second or two (41 s).
- cache `#t` and `#f` in interpreter, better lookup for special
forms/builtins (hashtable of pointers instead of strings, vector for the
small special form list). Dropped time to 38s.
- special-case in quasiquote when splicing is the last thing in a list.
Allocation dropped to 340,603,082
- custom hash table for environment lookups (lexical vars). Dropped to
36s and 314,637,194
- less allocation in `read_list` 311,613,616. Time about the same.
- `let` and `let*` in Interpreter.cpp 191,988,083, time down to 28s.
Previously, `object` and `none` were both top-level types. This made
decompilation rather messy as they have no LCA and resulted in a lot of
variables coming out as type `none` which is very very wrong and
additionally there were plenty of casts to `object`. This changes it so
`none` becomes a child of `object` (it is still represented by
`NullType` which remains unusable in compilation).
This change makes `object` the sole top-level type, and the type that
can represent *any* GOAL object. I believe this matches the original
GOAL built-in type structure. A function that has a return type of
`object` can now return an integer or a `none` at the same time.
However, keep in mind that the return value of `(none)` is still
undefined, just as before. This also makes a cast to `object`
meaningless in 90% of the situations it showed up in (as every single
thing is already an `object`) and the decompiler will no longer emit
them. Casts to `none` are also reduced. Yay!
Additionally, state handlers also don't get the final `(none)` printed
out anymore. The return type of a state handler is completely
meaningless outside the event handler (which is return type `object`
anyway) so there are no limitations on what the last form needs to be. I
did this instead of making them return `object` to trick the decompiler
into not trying to output a variable to be used as a return value
(internally, in the decompiler they still have return type `none`, but
they have `object` elsewhere).
Fixes#1703Fixes#830Fixes#928
This fixes a long time issue with `lambda`. The `lambda` is a bit
overloaded in OpenGOAL: it's used in the implementation of `let`, and
also to define local anonymous functions.
```
(defmacro let (bindings &rest body)
`((lambda :inline #t ,(apply first bindings) ,@body)
,@(apply second bindings)))
```
```
(defmacro defun (name bindings &rest body)
(let ((docstring ""))
(when (and (> (length body) 1) (string? (first body)))
(set! docstring (first body))
(set! body (cdr body)))
`(define ,name ,docstring (lambda :name ,name ,bindings ,@body))))
```
In the first case of a `let`, a `return` from inside the `let` should
return from the functioning containing the `let`, not the scope of the
`lambda`. In the second case, we should return from the lambda. The way
we told the different between these cases was if the `lambda` was used
"immeidately", in the head of an expression (like it would be for the
`let` macro). But, this falsely triggers when an anonymous function is
used immediately: eg
```
((lambda () (return #f)))
```
should generate and call a real x86 function that returns immediately.
This should fix some death/mission failed stuff in jak 2.
Moves PC-specific entity and debug menu things to `entity-debug.gc` and
`default-menu-pc.gc` respectively and makes `(declare-file (debug))`
work as it should (no need to wrap the entire file in `(when
*debug-segment*` now!).
Also changes the DGO descriptor format so that it's less verbose. It
might break custom levels, but the format change is very simple so it
should not be difficult for anyone to update to the new format. Sadly,
you lose the completely useless ability to use DGO object names that
don't match the source file name. The horror!
I've also gone ahead and expanded the force envmap option to also force
the ripple effect to be active. I did not notice any performance or
visual drawbacks from this. Gets rid of some distracting LOD and some
water pools appearing super flat (and pitch back for dark eco).
Fixes#1424
Just missing `target-board-handler` because it errored with `Bad delay
slot in clean_up_cond_no_else_final` and I wasn't able to figure out
where the asm branch was.
Commented out `target-board-clone-anim` because it crashes even after
adding `clone-anim` and `clone-anim-once`.
* [goalc] macro expansion in integer constants
* working
* didn't break it yet
* support conditional compilation
* fix up some more small bugs
* fix duplicate evaluation of bitfield definitions
* paranoid
* clean up allocator interface to be simpler
* working on functions without spills
* working for all
* fix missing includes for windows
* more windows includes
* initialize regs to zero so printing value unintiailized by game code is repeatable
* temp
* working, but type pass got really slow
* clean up
* changelog and flip order
* clean up and add tests
* fix zero size array
* handle lambdas correctly
* another windows fix
* set up types
* cleaned up type analysis and got things working through atomic ops
* expression working, need types
* improved types and names
* getting close
* finish up dma-disasm
* fix
* new method of inline-array-class
* up to new array
* wip side effect stuff
* prepare for pop barrier stuff
* add pop barrier
* add local vars hack to compiler
* fix bug, make sort work
* add test for array
* bug fixes
* another bug fix
* refactoring env variable print for casts
* more tweaks
* updates
* final cleanup
* codacy fixes
* begin work on vf support
* split reg kind into reg hw kind and class, use class for ireg
* try test
* clang format
* add some more ops and some example functions
* better lvf on statics
* add documentation
* work in progress streaming rpc, simple test is working
* actually add the test
* debug windows failure
* windows fix maybe
* windows 2
* use str-load-status
* update types
* add some type stuff for gkernel
* more tweaks
* blah
* more little tweaks and more of gkernel
* add static structures with integer fields but not basics
* static structures
* update gkernel
* start cleanup
* fix typos
* fix syntax highlighting in doc
* lots of documentation updates
* clean and add tests
* more documentation and more error messages
* more document and try building kernel differently
* started adding simple functions in gcommon
* more tests and features
* more tests, debug windows
* debug prints for windows
* back up some regs for windows
* remove debugging prints
* add tests for various xmms
* use unaligned stores and loads to back up and restore xmm128s and also fix argument spilling bug
* add deftype
* add deref and fix some method _type_ things
* see if math works on windows
* add dgo
* windows debug
* windows debug 2
* one more debug try
* add extra debug print and change logic for slashes
* update
* again
* try again
* remove build game
* remove build game
* add back build-game
* remove runtime from test
* test
* reduce number of files
* go to c++ 14
* big stacks
* increase stack size again
* clean up cmake files
* add some more tests for let
* support static strings
* add function calling
* add prints for windows debgu
* one test only
* try swapping r14 and r15 in windows
* swap back
* disable defun for now
* fix massive bug
* fix formatting
water111