Instead of doing the dot product related math in scalar IR, we lift the
computation into a dedicated IR instruction.
On x64, we can use VDPPS which was more or less tailor made for this
purpose. This is better than manual scalar lowering that requires
reloading components from memory; it's not always a strict improvement
over the shuffle+add version (which we never had), but this can now be
adjusted in the IR lowering in an optimal fashion (maybe even based on
CPU vendor, although that'd create issues for offline compilation).
On A64, we can either use naive adds or paired adds, as there is no
dedicated vector-wide horizontal instruction until SVE. Both run at
about the same performance on M2, but paired adds require fewer
instructions and temporaries.
I've measured this using mesh-normal-vector benchmark, changing the
benchmark to just report the time of the second loop inside
`calculate_normals`, testing master vs #1504 vs this PR, also increasing
the grid size to 400 for more stable timings.
On Zen 4 (7950X), this PR is comfortably ~8% faster vs master, while I
see neutral to negative results in #1504.
On M2 (base), this PR is ~28% faster vs master, while #1504 is only
about ~10% faster.
If I measure the second loop in `calculate_tangent_space` instead, I
get:
On Zen 4 (7950X), this PR is ~12% faster vs master, while #1504 is ~3%
faster
On M2 (base), this PR is ~24% faster vs master, while #1504 is only
about ~13% faster.
Note that the loops in question are not quite optimal, as they store and
reload various vectors to dictionary values due to inappropriate use of
locals. The underlying gains in individual functions are thus larger
than the numbers above; for example, changing the `calculate_normals`
loop to use a local variable to store the normalized vector (but still
saving the result to dictionary value), I get a ~24% performance
increase from this PR on Zen4 vs master instead of just 8% (#1504 is
~15% slower in this setup).
### What's New?
* Fragment Autocomplete: a new API allows for type checking a small
fragment of code against an existing file, significantly speeding up
autocomplete performance in large files.
### New Solver
* E-Graphs have landed: this is an ongoing approach to make the new type
solver simplify types in a more consistent and principled manner, based
on similar work (see: https://egraphs-good.github.io/).
* Adds support for exporting / local user type functions (previously
they were always exported).
* Fixes a set of bugs in which the new solver will fail to complete
inference for simple expressions with just literals and operators.
### General Updates
* Requiring a path with a ".lua" or ".luau" extension will now have a
bespoke error suggesting to remove said extension.
* Fixes a bug in which whether two `Luau::Symbol`s are equal depends on
whether the new solver is enabled.
---
Internal Contributors:
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Hunter Goldstein <hgoldstein@roblox.com>
Co-authored-by: Varun Saini <vsaini@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
Closes#1492
Tested and working with the test case in the aforementioned issue, along
with the full defs of luau-lsp with no issues or type errors
In normal Luau files, you can use type aliases and type functions before
they are declared. The same extends to declaration files, **except** in
the new solver. The old solver perfectly allows this, and in fact
intentionally adds it:
db809395bf/Analysis/src/TypeInfer.cpp (L1711-L1717)
This causes *much* headache and pain for external projects that make use
of declaration files; namely, luau-lsp generates them from MaximumADHD's
API dump, which is not ordered by dependency. This means silent
error-types popping up everywhere because types are used before they are
declared. The workaround would be to make code to manually reorder class
definitions based on their dependencies with a bunch of code, but this
is clearly not ideal, and won't work for classes dependent on each
other/recursive.
The solution used here is the same as is used for type aliases - the
name binding for the class is given a blocked type before running the
rest of constraint generation on the block. Questions remain:
- Should the logic be split off of `checkAliases`?
- Should a bound type be used, or should the (blocked) binding type be
directly emplaced with the class type? What are the ramifications of
emplacing with the bound versus the raw type? One ramification was
initially ran into through an assertion because the class
`superTy`/`parent` was bound, and several pieces of code assume it is
not, so it had to be made followed.
- Is folllowing `superTy` to set `parent` the correct workaround for the
assertions thrown, or should the code expecting `parent` to be a
ClassType without following it be modified instead to follow `parent`?
- Should `scope->privateTypeBindings` also be checked for the duplicate
error? I would presume so, since having a class with the same name as a
private alias or type function should error as well?
The extraneous whitespace changes are clang-format ones done
automatically that should've been done in the last release - I can
remove them if necessary and let another sync or OSS cleanup commit fix
it.
Closes#1441
Brings behavior to parity with the old solver by filling in
definitionLocation and definitionModuleName for Luau-consuming
programs/libraries to use.
* New `vector` library! See https://rfcs.luau.org/vector-library.html
for details
* Replace the use of non-portable `strnlen` with `memchr`. `strnlen` is
not part of any C or C++ standard.
* Introduce `lua_newuserdatataggedwithmetatable` for faster tagged
userdata creation of userdata with metatables registered with
`lua_setuserdatametatable`
Old Solver
* It used to be the case that a module's result type would
unconditionally be inferred to be `any` if it imported any module that
participates in any import cycle. This is now fixed.
New Solver
* Improve inference of `table.freeze`: We now infer read-only properties
on tables after they have been frozen.
* We now correctly flag cases where `string.format` is called with 0
arguments.
* Fix a bug in user-defined type functions where table properties could
be lost if the table had a metatable
* Reset the random number seed for each evaluation of a type function
* We now retry subtyping arguments if it failed due to hidden variadics.
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Vighnesh <vvijay@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
Co-authored-by: Junseo Yoo <jyoo@roblox.com>
## What's new
* Added `math.map` function to the standard library, based on
https://rfcs.luau-lang.org/function-math-map.html
* `FileResolver` can provide an implementation of
`getRequireSuggestions` to provide auto-complete suggestions for
require-by-string
## New Solver
* In user-defined type functions, `readproperty` and `writeproperty`
will return `nil` instead of erroring if property is not found
* Fixed incorrect scope of variadic arguments in the data-flow graph
* Fixed multiple assertion failures
---
Internal Contributors:
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Hunter Goldstein <hgoldstein@roblox.com>
Co-authored-by: Varun Saini <vsaini@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
# General Updates
Fix an old solver crash that occurs in the presence of cyclic
`requires()`
## New Solver
- Improvements to Luau user-defined type function library
- Avoid asserting on unexpected metatable types
- Properties in user defined type functions should have a consistent
iteration order - in this case it is insertion ordering
# Runtime
- Track VM allocations for telemetry
---
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Hunter Goldstein <hgoldstein@roblox.com>
Co-authored-by: James McNellis <jmcnellis@roblox.com>
Co-authored-by: Varun Saini <vsaini@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
Co-authored-by: Junseo Yoo <jyoo@roblox.com>
Noticed while using luau-ast that function attributes aren't included in
the output. This PR corrects that.
---------
Co-authored-by: vegorov-rbx <75688451+vegorov-rbx@users.noreply.github.com>
Closes#1460.
This renames the `type` field of `AstStatTypeAlias` to `value` during
the JSON encoding process.
I've chosen to just rename the field in the JSON encoder rather than
rename the actual field since it's a lot further reaching. Another
option would have been to rename what the actual type of an AST node is
written to be something like `tokenType` instead of `type`, but that's a
bigger diff and technically breaking (as opposed to this one which
isn't!)
# General Updates
* Fix some cases where documentation symbols would not be available when
mouseovering at certain positions in the code
* Scaffolding to help embedders have more control over how `typeof(x)`
refines types
* Refinements to require-by-string semantics. See
https://github.com/luau-lang/rfcs/pull/56 for details.
* Fix for https://github.com/luau-lang/luau/issues/1405
# New Solver
* Fix many crashes (thanks you for your bug reports!)
* Type functions can now call each other
* Type functions all evaluate in a single VM. This should improve
typechecking performance and reduce memory use.
* `export type function` is now forbidden and fails with a clear error
message
* Type functions that access locals in the surrounding environment are
now properly a parse error
* You can now use `:setindexer(types.never, types.never)` to delete an
indexer from a table type.
# Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Hunter Goldstein <hgoldstein@roblox.com>
Co-authored-by: Varun Saini <vsaini@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
In this update, we continue to improve the overall stability of the new
type solver. We're also shipping some early bits of two new features,
one of the language and one of the analysis API: user-defined type
functions and an incremental typechecking API.
If you use the new solver and want to use all new fixes included in this
release, you have to reference an additional Luau flag:
```c++
LUAU_DYNAMIC_FASTINT(LuauTypeSolverRelease)
```
And set its value to `645`:
```c++
DFInt::LuauTypeSolverRelease.value = 645; // Or a higher value for future updates
```
## New Solver
* Fix a crash where scopes are incorrectly accessed cross-module after
they've been deallocated by appropriately zeroing out associated scope
pointers for free types, generic types, table types, etc.
* Fix a crash where we were incorrectly caching results for bound types
in generalization.
* Eliminated some unnecessary intermediate allocations in the constraint
solver and type function infrastructure.
* Built some initial groundwork for an incremental typecheck API for use
by language servers.
* Built an initial technical preview for [user-defined type
functions](https://rfcs.luau-lang.org/user-defined-type-functions.html),
more work still to come (including calling type functions from other
type functions), but adventurous folks wanting to experiment with it can
try it out by enabling `FFlag::LuauUserDefinedTypeFunctionsSyntax` and
`FFlag::LuauUserDefinedTypeFunction` in their local environment. Special
thanks to @joonyoo181 who built up all the initial infrastructure for
this during his internship!
## Miscellaneous changes
* Fix a compilation error on Ubuntu (fixes#1437)
---
Internal Contributors:
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Hunter Goldstein <hgoldstein@roblox.com>
Co-authored-by: Jeremy Yoo <jyoo@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh <vvijay@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
Co-authored-by: Junseo Yoo <jyoo@roblox.com>
In this update we improve overall stability of the new type solver and
address some type inference issues with it.
If you use the new solver and want to use all new fixes included in this
release, you have to reference an additional Luau flag:
```c++
LUAU_DYNAMIC_FASTINT(LuauTypeSolverRelease)
```
And set its value to `644`:
```c++
DFInt::LuauTypeSolverRelease.value = 644; // Or a higher value for future updates
```
## New Solver
* Fixed a debug assertion failure in autocomplete (Fixes#1391)
* Fixed type function distribution issue which transformed `len<>` and
`unm<>` into `not<>` (Fixes#1416)
* Placed a limit on the possible normalized table intersection size as a
temporary measure to avoid hangs and out-of-memory issues for complex
type refinements
* Internal recursion limits are now respected in the subtyping
operations and in autocomplete, to avoid stack overflow crashes
* Fixed false positive errors on assignments to tables whose indexers
are unions of strings
* Fixed memory corruption crashes in subtyping of generic types
containing other generic types in their bounds
---
Internal Contributors:
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
Fix for https://github.com/luau-lang/luau/issues/1406
While it is good to let ``index`` wait for the pending-expansion. To
re-produce the issue you need more than just this code:
https://i.imgur.com/b3OmUGF.png
It needs this, else it won't crash.
```lua
local function ProblemCauser(key: Keys, value)
PlayerData[key] = value
end
```
But regarding "pending things", I'd recommend **generalized functions**
for sanity checks like these, since there will be more cases of similar
issues I believe. But I am 100% sure that eventually this issue here can
maybe be prevented if looking at the Constraints. _(And optimization)_
Not sure if ``index`` needs the table fully completed, or if it is
preferred that the info is available based on **how much info is
available at the current position in the code**.
But if this gets done, I hope that they'll be connected to the Solver
Logger, because I actually refined mine with colors and more info _(yet
need to finish that)_ to understand the Luau Source Code more and to
debug issues.
---------
Co-authored-by: aaron <aweiss@hey.com>
Fixes https://github.com/luau-lang/luau/issues/1387
Was suggested by @alexmccord
I changed ``singletons[0]`` to ``singletons.front()``, unsure if that
makes a huge difference, and then I added the rest of the things needed
for the return type.
Maybe it's also the ideal location since doing it before looping through
``keys`` won't add the string into the type arena.
I put comments next to it based on how I thought it would make sense.
``LUAU_ASSERT`` seems to trigger when there's only one entry being put
inside a UnionType. It's as if it was put there for quality.
Allow edits by maintainers is enabled.
I tested this with a quick Unit Test something like
```lua
local test: keyof<typeof({a="test"})>
```
## New Solver
* The type functions `keyof` and `index` now also walk the inheritance
chain when they are used on class types like Roblox instances.
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
### What's new
* Light update this week, mostly fast flag cleanups.
### New Solver
* Rename flag to enable new solver from
`DebugLuauDeferredConstraintResolution` to `LuauSolverV2`
* Added support for magic functions for the new type checker (as opposed
to the type inference component)
* Improved handling of `string.format` with magic function improvements
* Cleaning up some of the reported errors by the new type checker
* Minor refactoring of `TypeChecker2.cpp` that happens to make the diff
very hard to read.
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh <vvijay@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
Co-authored-by: Junseo Yoo <jyoo@roblox.com>
### What's new
* Fixed many of the false positive errors in indexing of table unions
and table intersections
* It is now possible to run custom checks over Luau AST during
typechecking by setting `customModuleCheck` in `FrontendOptions`
* Fixed codegen issue on arm, where number->vector cast could corrupt
that number value for the next time it's read
### New Solver
* `error` type now behaves as the bottom type during subtyping checks
* Fixed the scope that is used in subtyping with generic types
* Fixed `astOriginalCallTypes` table often used by LSP to match the old
solver
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
# What's Changed?
- Variety of bugfixes in the new solver
## New Solver
- Fix an issue where we would hit a recursion limit when applying long
chains of type refinements.
- Weaken the types of `table.freeze` and `table.clone` in the new solver
so we can accept common code patterns like `local a = table.freeze({x=5,
x=0})` at the expense of accepting code like `table.freeze(true)`.
- Don't warn when the # operator is used on a value of type never
## VM
- Fix a bug in lua_resume where too many values might be removed from
stack when resume throws an error
---
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
Co-authored-by: Junseo Yoo <jyoo@roblox.com>
New Solver
* Fix some type inference issues surrounding updates to upvalues eg
```luau
local x = 0
function f()
x = x + 1
end
```
* User-defined type function progress
* Bugfixes for normalization of negated class types. eg `SomeClass &
(class & ~SomeClass)`
* Fixes to subtyping between tables and the top `table` type.
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Vighnesh <vvijay@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
Co-authored-by: Junseo Yoo <jyoo@roblox.com>
# What's Changed?
- Code refactoring with a new clang-format
- More bug fixes / test case fixes in the new solver
## New Solver
- More precise telemetry collection of `any` types
- Simplification of two completely disjoint tables combines them into a
single table that inherits all properties / indexers
- Refining a `never & <anything>` does not produce type family types nor
constraints
- Silence "inference failed to complete" error when it is the only error
reported
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Dibri Nsofor <dnsofor@roblox.com>
Co-authored-by: Jeremy Yoo <jyoo@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh <vvijay@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
# What's Changed?
- Telemetry support for usage of any type in old/new solver
- Bug fixes and flag removals with the new solver
## New Solver
- Fixed constraint ordering bug to infer types more accurately
- Improved inferring a call to `setmetatable()`
## VM
- Restored global metatable lookup for `typeof` on lightuserdata to fix
unintentional API change (Fixes#1335)
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Dibri Nsofor <dnsofor@roblox.com>
Co-authored-by: Jeremy Yoo <jyoo@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh <vvijay@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
# What's Changed?
- Bugfixes in the new solver
## New Solver
- Equality graphs(E-Graphs) data structures were added
- Refactored even more instances of "type family" with "type function"
- `table.insert` no longer spuriously warns while selecting an overload
for reasonable arguments.
- Add time tracing for the new solver
- Miscellaneous fixes to unit tests
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Jeremy Yoo <jyoo@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
Co-authored-by: Junseo Yoo <jyoo@roblox.com>
Working towards a full e-graph implementation as described by the [egg
paper](https://arxiv.org/pdf/2004.03082).
The type system has a couple of places where e-graphs would've been
useful and solved some classes of problems trivially. For example:
1. Normalization and simplification cannot handle cyclic types due to
the nature of their implementation.
2. Normalization can't tell when two tables or functions are equivalent,
but simplification theoretically can albeit not implemented.
3. Normalization requires deep normalization for inhabitance check,
whereas simplification would've returned the `never` type itself
indicating uninhabited.
4. Simplification requires constraint ordering to have perfect timing to
simplify.
5. Adding a rewrite rule requires implementing it twice, once in
simplification and once again in normalization with completely different
code design making it hard to verify that their behavior is materially
equivalent.
6. In cases where we must cache for performance, two different types
that are isomorphic have different cache entries resulting in cache
misses.
7. Type family reduction can handle cyclic type families, but only if
the cycle is not obscured by a different type family instance. (`t1
where t1 = union<number, add<t1, number>>` is irreducible)
I think we're getting the point!
---
Currently the implementation is missing a few features that makes
e-graphs actually useful. Those will be coming in a future PR.
1. Pattern matching,
6. Applying rewrites,
7. Rewrite until saturation, and
8. Extracting the best e-node according to some cost function.
# What's Changed?
- Performance improvement in the old solver
- Bugfixes in the new solver
## Old Solver
- Mark types that do not need instantiation when being exported to
prevent unnecessary work from being done
## New Solver
- Refactored instances of "type family" with "type function"
- Index-out-of-bounds bug fix in the resolution resolver
- Subtyping reasonings are merged only if all failed
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh <vvijay@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
# What's Changed?
- Mostly stability and bugfixes with the new solver.
## New Solver
- Typechecking with the new solver should respect the no-check hot
comment.
- Record type alias locations and property locations of table
assignments
- Maintain location information for exported table types
- Stability fixes for normalization
- Report internal constraint solver errors.
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
# What's Changed?
- Fix#1137 by appropriately retaining additional metadata from
definition files throughout the type system.
- Improve Frontend for LSPs by appropriately allowing the cancellation
of typechecking while running its destructor.
## New Solver
- Added support for the `rawget` type function.
- Reduced overall static memory usage of builtin type functions.
- Fixed a crash where visitors could mutate a union or intersection type
and fail to invalidate iteration over them in doing so.
- Revised autocomplete functionality to not rely on a separate run of
the type solver when using the new solver.
- Implemented a more relaxed semantic rule for casting.
- Fixed some smaller crashes in the new solver.
## Native Code Generation
- Add additional codegen specialization for `math.sign`
- Cleaned up a large number of outstanding fflags in the code.
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: James McNellis <jmcnellis@roblox.com>
Co-authored-by: Jeremy Yoo <jyoo@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh <vvijay@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
### What's new
* Added lint warning for using redundant `@native` attributes on
functions inside a `--!native` module
* Improved typechecking speed in old solver for modules with large types
### New Solver
* Fixed the length type function sealing the table prematurely
* Fixed crashes caused by general table indexing expressions
### VM
* Added support for a specialized 3-argument fast-call instruction to
improve performance of `vector` constructor, `buffer` writes and a few
`bit32` methods
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh <vvijay@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
### What's new
* A bug in exception handling in GCC(11/12/13) on MacOS prevents our
test suite from running.
* Parser now supports leading `|` or `&` when declaring `Union` and
`Intersection` types (#1286)
* We now support parsing of attributes on functions as described in the
[rfc](https://github.com/luau-lang/rfcs/pull/30)
* With this change, expressions such as `local x = @native function(x)
return x+1 end` and `f(@native function(x) return x+1 end)` are now
valid.
* Added support for `@native` attribute - we can now force native
compilation of individual functions if the `@native` attribute is
specified before the `function` keyword (works for lambdas too).
### New Solver
* Many fixes in the new solver for crashes and instability
* Refinements now use simplification and not normalization in a specific
case of two tables
* Assume that compound assignments do not change the type of the
left-side operand
* Fix error that prevented Class Methods from being overloaded
### VM
* Updated description of Garbage Collector invariant
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
### What's new
* Implemented parsing logic for attributes
* Added `lua_setuserdatametatable` and `lua_getuserdatametatable` C API
methods for a faster userdata metatable fetch compared to
`luaL_getmetatable`. Note that metatable reference has to still be
pinned in memory!
### New Solver
* Further improvement to the assignment inference logic
* Fix many bugs surrounding constraint dispatch order
### Native Codegen
* Add IR lowering hooks for custom host userdata types
* Add IR to create new tagged userdata objects
* Remove outdated NativeState
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
These were written before compiler optimizations and array invariant. It
is now impossible for t[1] to be stored in the hash part, as this would
violate the array invariant that says that elements 1..#t are stored in
the array.
For ipairs, it doesn't traverse the hash part anymore now, so we adjust
the code to make sure no elements outside of the 1..#t slice are
covered. For table.find, we can use find-with-offset to still access the
hash part.
Fixes#1283.
Implements the [Leading `|` and `&` in types](https://rfcs.luau-lang.org/syntax-leading-bar-and-ampersand.html) RFC.
The changes to the parser are exactly as described in the RFC.
---------
Co-authored-by: Alexander McCord <11488393+alexmccord@users.noreply.github.com>
The 'findBindingAtPosition' AstQuery function can be used to lookup a
local or global binding.
Inside of this function is a check to "Ignore this binding if we're
inside its definition. e.g. local abc = abc -- Will take the definition
of abc from outer scope".
However, this check is incorrect when we are looking up a global binding
at the start of a file.
Consider a complete file with the contents:
```lua
local x = stri|ng.char(1)
```
and we pass the location of the marker `|` as the position to the find
binding position. We will pick up the global binding of the definition
`string` coming from a builtin source (either defined via C++ code or a
definitions file and loaded into the global scope).
The global binding `string` will have a zero position: `0,0,0,0`.
However, the `findBindingLocalStatement` check works by looking up the
AstAncestry at the binding's defined begin position *in the current
source module*. This will then incorrectly return the local statement
for `local x`, as that is at the start of the source code. Then in turn,
we assume we are in the `local abc = abc` case, and end up skipping over
the correct binding.
We fix this by checking if the binding is at the global position. If so,
we early exit because it is impossible for a global binding to be
defined in a local statement.
### What's new?
* Remove a case of unsound `table.move` optimization
* Add Luau stack slot reservations that were missing in REPL (fixes
#1273)
### New Type Solver
* Assignments have been completely reworked to fix a case of cyclic
constraint dependency
* When indexing, if the fresh type's upper bound already contains a
compatible indexer, do not add another upper bound
* Distribute type arguments over all type families sans `eq`, `keyof`,
`rawkeyof`, and other internal type families
* Fix a case where `buffers` component weren't read in two places (fixes
#1267)
* Fix a case where things that constitutes a strong ref were slightly
incorrect
* Fix a case where constraint dependencies weren't setup wrt `for ...
in` statement
### Native Codegen
* Fix an optimization that splits TValue store only when its value and
its tag are compatible
* Implement a system to plug additional type information for custom host
userdata types
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh <vvijay@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
### What's new?
* Removed new `table.move` optimization because of correctness problems.
### New Type Solver
* Improved error messages for type families to describe what's wrong in
more detail, and ideally without using the term `type family` at all.
* Change `boolean` and `string` singletons in type checking to report
errors to the user when they've gotten an impossible type (indicating a
type error from their context).
* Split debugging flags for type family reduction
(`DebugLuauLogTypeFamilies`) from general solver logging
(`DebugLuauLogSolver`).
* Improve type simplification to support patterns like `(number |
string) | (string | number)` becoming `number | string`.
### Native Code Generation
* Use templated `luaV_doarith` to speedup vector operation fallbacks.
* Various small changes to better support arm64 on Windows.
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: James McNellis <jmcnellis@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh <vvijay@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
### New Type Solver
* Fixed crash in numeric binary operation type families
* Results of an indexing operation are now comparable to `nil` without a
false positive error
* Fixed a crash when a type that failed normalization was accessed
* Iterating on a free value now implies that it is iterable
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: James McNellis <jmcnellis@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
# What's changed?
* Optimize table.maxn. This function is now 5-14x faster
* Reserve Luau stack space for error message.
## New Solver
* Globals can be type-stated, but only if they are already in scope
* Fix a stack overflow that could occur when normalizing certain kinds
of recursive unions of intersections (of unions of intersections...)
* Fix an assertion failure that would trigger when the __iter metamethod
has a bad signature
## Native Codegen
* Type propagation and temporary register type hints
* Direct vector property access should only happen for names of right
length
* BytecodeAnalysis will only predict that some of the vector value
fields are numbers
---
## Internal Contributors
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
table.concat is idiomatic and should be the fastest way to concatenate
all table array elements together, but apparently you can beat it by
using `string.format`, `string.rep` and `table.unpack`:
```lua
string.format(string.rep("%*", #t), table.unpack(t))
```
... this just won't do, so we should fix table.concat performance.
The deficit comes from two places:
- rawgeti overhead followed by other stack accesses, all to extract a
string from what is almost always an in-bounds array lookup
- addlstring overhead in case separator is empty (extra function calls)
This change fixes this by using a fast path for in-bounds array lookup
for a string. Note that `table.concat` also supports numbers (these need
to be converted to strings which is a little cumbersome and has innate
overhead), and out-of-bounds accesses*. In these cases we fall back to
the old implementation.
To trigger out-of-bounds accesses, you need to skip the past-array-end
element (which is nil per array invariant), but this is achievable
because table.concat supports offset+length arguments. This should
almost never come up in practice but the per-element branches et al are
fairly cheap compared to the eventual string copy/alloc anyway.
This change makes table.concat ~2x faster when the separator is empty;
the table.concat benchmark shows +40% gains but it uses a variety of
string separators of different lengths so it doesn't get the full
benefit from this change.
---------
Co-authored-by: vegorov-rbx <75688451+vegorov-rbx@users.noreply.github.com>
# What's changed?
### New Type Solver
- Unification of two fresh types no longer binds them together.
- Replaced uses of raw `emplace` with `emplaceType` to catch cyclic
bound types when they are created.
- `SetIndexerConstraint` is blocked until the indexer result type is not
blocked.
- Fix a case where a blocked type got past the constraint solver.
- Searching for free types should no longer traverse into `ClassType`s.
- Fix a corner case that could result in the non-testable type `~{}`.
- Fix incorrect flagging when `any` was a parameter of some checked
function in nonstrict type checker.
- `IterableConstraint` now consider tables without `__iter` to be
iterables.
### Native Code Generation
- Improve register type info lookup by program counter.
- Generate type information for locals and upvalues
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: James McNellis <jmcnellis@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh <vvijay@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
# What's changed?
* Improved the actual message for the type errors for `cannot call
non-function` when attempting to call a union of functions/callable
tables. The error now correctly explains the issue is an inability to
determine the return type of the call in this situation.
* Resolve an issue where tables and metatables were not correctly being
cloned during instantiation (fixes#1176).
* Refactor `luaM_getnextgcopage` to `luaM_getnextpage` (generally
removing `gco` prefix where appropriate).
* Optimize `table.move` between tables for large ranges of elements.
* Reformat a bunch of code automatically using `clang-format`.
### New Type Solver
* Clean up minimally-used or unused constraints in the constraint solver
(`InstantiationConstraint`, `SetOpConstraint`,
`SingletonOrTopTypeConstraint`).
* Add a builtin `singleton` type family to replace
`SingletonOrTopTypeConstraint` when inferring refinements.
* Fixed a crash involving type path reasoning by recording when type
family reduction has taken place in the path.
* Improved constraint ordering by blocking on unreduced types families
that are not yet proven uninhabitable.
* Improved the handling of `SetIndexerConstraints` for both better
inference quality and to resolve crashes.
* Fix a crash when normalizing cyclic unions of intersections.
* Fix a crash when normalizing an intersection with the negation of
`unknown`.
* Fix a number of crashes caused by missing `follow` calls on `TypeId`s.
* Changed type family reduction to correctly use a semantic notion of
uninhabited types, rather than checking for `never` types specifically.
* Refactor the `union` and `intersect` type families to be variadic.
### Native Code Generation
* Improve translation for userdata key get/set and userdata/vector
namecall.
* Provide `[top level]` and `[anonymous]` as function names to
`FunctionStats` as appropriate when no function name is available.
* Disable unwind support on Android platforms since it is unsupported.
*
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh <vvijay@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
# What's changed?
* Support for new 'require by string' RFC with relative paths and
aliases in now enabled in Luau REPL application
### New Type Solver
* Fixed assertion failure on generic table keys (`[expr] = value`)
* Fixed an issue with type substitution traversing into the substituted
parts during type instantiation
* Fixed crash in union simplification when that union contained
uninhabited unions and other types inside
* Union types in binary type families like `add<a | b, c>` are expanded
into `add<a, c> | add<b, c>` to handle
* Added handling for type family solving creating new type families
* Fixed a bug with normalization operation caching types with unsolved
parts
* Tables with uninhabited properties are now simplified to `never`
* Fixed failures found by fuzzer
### Native Code Generation
* Added support for shared code generation between multiple Luau VM
instances
* Fixed issue in load-store propagation and new tagged LOAD_TVALUE
instructions
* Fixed issues with partial register dead store elimination causing
failures in GC assists
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: James McNellis <jmcnellis@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
# What's Changed
## New Type Solver
- Many more fixes to crashes, assertions, and hangs
- Annotated locals now countermand the inferred types of locals, meaning
that for a type `type MyType = number | string`, `local foo : MyType =
5` behaves the same as `local foo = 5 :: MyType`, where before, foo
would be assigned the type of the value on the rhs.
- Type Normalization now respects resource limits.
- Subtyping between classes and cyclic tables now supported
## Native Code Generation
- Work on the Native Code Generation(NCG) allocator continues
---
# Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: James McNellis <jmcnellis@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
# What's Changed
## New Type Solver
- Many fixes to crashes, assertions, and hangs
- Binary type family aliases now have a default parameter
- Added a debug check for unsolved types escaping the constraint solver
- Overloaded functions are no longer inferred
- Unification creates additional subtyping constraints for blocked types
- Attempt to guess the result type for type families that are too large
to resolve timely
## Native Code Generation
- Fixed `IrCmd::CHECK_TRUTHY` lowering in a specific case
- Detailed compilation errors are now supported
- More work on the new allocator
---
# Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Alexander McCord <amccord@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: James McNellis <jmcnellis@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
Co-authored-by: Vighnesh Vijay <vvijay@roblox.com>
# What's changed
### Debugger
* Values after a 'continue' statement should not be accessible by
debugger in the 'until' condition
### New Type Solver
* Many fixes to crashes and hangs
* Better bidirectional inference of table literal expressions
### Native Code Generation
* Initial steps toward a shared code allocator
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
# What's Changed
* Fix a case where the stack wasn't completely cleaned up where
`debug.info` errored when passed `"f"` option and a thread.
* Fix a case of uninitialized field in `luaF_newproto`.
### New Type Solver
* When a local is captured in a function, don't add a new entry to the
`DfgScope::bindings` if the capture occurs within a loop.
* Fix a poor performance characteristic during unification by not trying
to simplify an intersection.
* Fix a case of multiple constraints mutating the same blocked type
causing incorrect inferences.
* Fix a case of assertion failure when overload resolution encounters a
return typepack mismatch.
* When refining a property of the top `table` type, we no longer signal
an unknown property error.
* Fix a misuse of free types when trying to infer the type of a
subscript expression.
* Fix a case of assertion failure when trying to resolve an overload
from `never`.
### Native Code Generation
* Fix dead store optimization issues caused by partial stores.
---
### Internal Contributors
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
---------
Co-authored-by: Aaron Weiss <aaronweiss@roblox.com>
Co-authored-by: Andy Friesen <afriesen@roblox.com>
Co-authored-by: Vighnesh <vvijay@roblox.com>
Co-authored-by: Aviral Goel <agoel@roblox.com>
Co-authored-by: David Cope <dcope@roblox.com>
Co-authored-by: Lily Brown <lbrown@roblox.com>
Co-authored-by: Vyacheslav Egorov <vegorov@roblox.com>
When the input is a constant, we use a fairly inefficient sequence of
fmov+fcvt+dup or, when the double isn't encodable in fmov,
adr+ldr+fcvt+dup.
Instead, we can use the same lowering as X64 when the input is a
constant, and load the vector from memory. However, if the constant is
encodable via fmov, we can use a vector fmov instead (which is just one
instruction and doesn't need constant space).
Fortunately the bit encoding of fmov for 32-bit floating point numbers
matches that of 64-bit: the decoding algorithm is a little different
because it expands into a larger exponent, but the values are
compatible, so if a double can be encoded into a scalar fmov with a
given abcdefgh pattern, the same pattern should encode the same float;
due to the very limited number of mantissa and exponent bits, all values
that are encodable are also exact in both 32-bit and 64-bit floats.
This strategy is ~same as what gcc uses. For complex vectors, we
previously used 4 instructions and 8 bytes of constant storage, and now
we use 2 instructions and 16 bytes of constant storage, so the memory
footprint is the same; for simple vectors we just need 1 instruction (4
bytes).
clang lowers vector constants a little differently, opting to synthesize
a 64-bit integer using 4 instructions (mov/movk) and then move it to the
vector register - this requires 5 instructions and 20 bytes, vs ours/gcc
2 instructions and 8+16=24 bytes. I tried a simpler version of this that
would be more compact - synthesize a 32-bit integer constant with
mov+movk, and move it to vector register via dup.4s - but this was a
little slower on M2, so for now we prefer the slightly larger version as
it's not a regression vs current implementation.
On the vector approximation benchmark we get:
- Before this PR (flag=false): ~7.85 ns/op
- After this PR (flag=true): ~7.74 ns/op
- After this PR, with 0.125 instead of 0.123 in the benchmark code (to
use fmov): ~7.52 ns/op
- Not part of this PR, but the mov/dup strategy described above: ~8.00
ns/op
