* AstGen: Move `refToIndex` and `indexToRef` to Zir
* ZIR: the switch_block_*_* instruction tags are collapsed into one
switch_block tag which uses 4 bits for flags, and reduces the
scalar_cases_len field from 32 to 28 bits.
This freed up more ZIR tags, 2 of which are now used for
`switch_cond` and `switch_cond_ref` for producing the switch
condition value. For example, for union values it returns the
corresponding enum value.
* switching with multiple cases and ranges is not yet supported because
I want to change the ZIR encoding to store index pointers into the
extra array rather than storing prong indexes. This will avoid O(N^2)
iteration over prongs.
* AstGen now adds a `switch_cond` on the operand and then passes the
result of that to the `switch_block` instruction.
* Sema: partially implement `switch_capture_*` instructions.
* Sema: `unionToTag` notices if the enum type has only one possible value.
The remaining uses of this result location were causing a bunch of errors
problems where the pointers returned from rvalue and lvalue expressions
would be confused, allowing for extra pointers on rvalue expressions.
For example:
```zig
const X = struct {a: i32};
var x: X = .{.a = 1};
var ptr = &x;
_ = x.a;
```
In the last line, the lookup of x with result location .none_or_ref would
return a double pointer (**X). This would be dereferenced one, after which
a relative pointer to `a` would be fetched and derefenced to get the final
result.
However, this also allows us to manually construct a double pointer, and
fetch the field of the inner type of that:
```zig
_ = &(&(x)).a;
```
This problem also manifests itself with element access. There are two obvious
ways to fix the problem, both of which include replacing the usage of
.none_or_ref for field- and element accesses with something which
deterministically produce either a pointer or value: either result location
.ref or .none. In the former case, this would be paired with .elem_ptr, and
in the latter case with .elem_val.
Note that the stage 1 compiler does not have this problem, because there is
no equivalent of .elem_val and .field_val. In this way it is equivalent to
using the result location .ref for field- and element accesses.
In this case i have used .none, as this matches language behaviour more
closely.
Dereferencing single pointers is now handled outside of the main switch,
which allows deduplication of some cases. This also implements the
relevant operations for pointers to types and pointers to slices.
* do not add linkage scope to aliased exported symbols - this is
not respected on macOS
* special-case `MachO.openPath` in `link.File.openPath` as on macOS
we always link with zld
* redirect to `MachO.flushObject` when linking relocatable objects
in MachO linker whereas move the entire linking logic into
`MachO.flushModule`
When returning an error set or an error union from a function which has
an inferred error set, it populates the error names in addition to the
set of functions. This can have false negatives, meaning that after
checking the map of an unresolved error set, one must do full error set
resolution before emitting a compile error.
* Sema: fix returned operands not coercing to the function return type
in some cases.
- When returning an error or an error union from a function with an
inferred error set, it will now populate the inferred error set.
- Implement error set coercion for the common case of inferred error
set to inferred error set, without forcing a full resolution.
* LLVM backend: update instruction lowering that handles error unions
to respect `isByRef`.
- Also implement `wrap_err_union_err`.
* Relax compile error for "unable to export type foo" to allow
integers, structs, arrays, and floats. This will need to be further
improved to do the same checks as we do for C ABI struct field types.
* LLVM backend: fix extern variables
* LLVM backend: implement AIR instruction `wrap_err_union_payload`
* Sema: implement peer type resolution for optionals and null.
* Rename `Module.optionalType` to `Type.optional`.
* LLVM backend: re-use anonymous values. This is especially useful when
isByRef()=true because it means re-using the same generated LLVM globals.
* LLVM backend: rework the implementation of is_null and is_non_null
AIR instructions. Generate slightly better LLVM code, and also fix
the behavior for optionals whose payload type is 0-bit.
* LLVM backend: improve `cmp` AIR instruction lowering to support
pointer-like optionals.
* `Value`: implement support for equality-checking optionals.
* LLVM backend: The `alloc` AIR instruction as well as pointer
constants which point to a 0-bit element type now call a common
codepath to produce a `*const llvm.Value` which is a non-zero pointer
with a bogus-but-properly-aligned address.
* LLVM backend: improve the lowering of optional types.
* Type: `hasCodeGenBits()` now returns `true` for pointers even when
it returns `false` for their element types.
Effectively, #6706 is now implemented in stage2 but not stage1.
* saturating shl - check for negative rhs at comptime
- adds expected compile_errors case for negative rhs
* add expected compile error for sat shl assign
* std.os: take advantage of `@minimum`. It's probably time to
deprecate `std.min` and `std.max`.
* New AIR instructions: min and max
* Introduce SIMD vector support to stage2
* Add `@Type` support for vectors
* Sema: add `checkSimdBinOp` which can be re-used for other arithmatic
operators that want to support vectors.
* Implement coercion from vectors to arrays.
- In backends this is handled with bitcast for vector to array,
however maybe we want to reduce the amount of branching by
introducing an explicit AIR instruction for it in the future.
* LLVM backend: implement lowering vector types
* Sema: Implement `slice.ptr` at comptime
* Value: improve `numberMin` and `numberMax` to support floats in
addition to integers, and make them behave properly in the presence
of NaN.
* `Module.Union.getFullyQualifiedName` returns a sentinel-terminated
slice so that backends that need null-termination do not need an
additional copy.
* Module.Union: implement a `getLayout` function which returns
information about ABI size and alignment so that the LLVM backend can
properly lower union types into llvm types.
* Sema: `resolveType` now returns `error.GenericPoison` rather than a
Type with tag `generic_poison`. Callsites that want to allow that
need to bypass this higher-level function.
* Sema: implement coercion of enums and enum literals to unions.
* Sema: fix comptime mutation of pointers to unions
* LLVM backend: fully implement proper lowering of union types and
values according to the union layout, and update the handling of AIR
instructions that deal with unions to support union layouts.
* LLVM backend: handle `decl_ref_mut`
- Maybe this should be unreachable since comptime vars should be
changed to be non-mutable when they go out of scope, but it's
harmless for the LLVM backend to support lowering the value.
* Type: fix `requiresComptime` for optionals, pointers, and some other
types. This function is still wrong for structs, unions, and enums.
* Make `alloc` AIR instructions call `resolveTypeLayout`.
* `Sema.zirResolveInferredAlloc` now calls `requireRuntimeBlock` in the
case that it operates on a non-comptime instruction.
* `Type.abiSize` and `Type.abiAlignment` now return 0 for `void`
* Sema: implement `resolveTypeFields` for unions.
* LLVM Backend: support `ptr_elem_ptr` when the element type is 0-bit.
* Type: improve `abiAlignment` implementation for structs to properly
handle fields with non-default alignment.
* Value: implement hashing array, vector, and structs.
The logic for `buildAlloca` had a null deref when the latest alloca was
the last instruction in the entry block. Now the logic is simplified to
always insert alloca instructions first before all other instructions.
There is no longer a need to track `entry_block` or `latest_alloca_inst`;
these fields are deleted frem `FuncGen`.
* apply late symbol resolution for globals - instead of resolving
the exact location of a symbol in locals, globals or undefs,
we postpone the exact resolution until we have a full picture
for relocation resolution.
* fixup stubs to defined symbols - this is currently a hack rather
than a final solution. I'll need to work out the details to make
it more approachable. Currently, we preemptively create a stub
for a lazy bound global and fix up stub offsets in stub helper
routine if the global turns out to be undefined only. This is quite
wasteful in terms of space as we create stub, stub helper and lazy ptr
atoms but don't use them for defined globals.
* change log scope to .link for macho.
* remove redundant code paths from Object and Atom.
* drastically simplify the contents of Relocation struct (i.e., it is
now a simple superset of macho.relocation_info), clean up relocation
parsing and resolution logic.
