| Age | Commit message (Collapse) | Author |
|---|
|
Part of an effort to ship more of the compiler in source form.
|
|
Part of an effort to ship more of the compiler in source form.
|
|
This commit eliminates the `dbg_block_{begin,end}` instructions from
both ZIR and AIR. Instead, lexical scoping of `dbg_var_{ptr,val}`
instructions is decided based on the AIR block they exist within. This
is a much more robust system, and also results in a huge drop in ZIR
bytes - around 7% for Sema.zig.
This required some enhancements to Sema to prevent elision of blocks
when they are required for debug variable scoping. This can be observed
by looking at the AIR for the following simple test program with and
without `-fstrip`:
```zig
export fn f() void {
{
var a: u32 = 0;
_ = &a;
}
{
var a: u32 = 0;
_ = &a;
}
}
```
When `-fstrip` is passed, no AIR blocks are generated. When `-fno-strip`
is passed, the ZIR blocks are lowered to true AIR blocks to give correct
lexical scoping to the debug vars.
The changes here incidentally reolve #19060. A corresponding behavior
test has been added.
Resolves: #19060
|
|
The `decl_node` was offset from the wrong source node.
|
|
If an adapted string key with embedded nulls was put in a hash map with
`std.hash_map.StringIndexAdapter`, then an incorrect hash would be
entered for that entry such that it is possible that when looking for
the exact key that matches the prefix of the original key up to the
first null would sometimes match this entry due to hash collisions and
sometimes not if performed later after a grow + rehash, causing the same
key to exist with two different indices breaking every string equality
comparison ever, for example claiming that a container type doesn't
contain a field because the field name string in the struct and the
string representing the identifier to lookup might be equal strings but
have different string indices. This could maybe be fixed by changing
`std.hash_map.StringIndexAdapter.hash` to only hash up to the first
null, therefore ensuring that the entry's hash is correct and that all
future lookups will be consistent, but I don't trust anything so instead
I assert that there are no embedded nulls.
|
|
Previously, the `src_node` field of `struct_decl`, `union_decl`,
`enum_decl`, and `opaque_decl` was optional, included in trailing data
only if a flag in `Small` was set. However, this was unnecessary logic:
AstGen always provided the source node. We can simplify a few bits of
logic by making this field non-optional, moving it into non-trailing
data.
There was one place where the field was actually omitted before: the
root struct of a file was at source node 0, so the node was
coincidentally elided. Therefore, this commit has a fixed cost of 4
bytes of ZIR per file.
|
|
In most cases where AstGen is coercing to a fixed type (such as `u29`,
`type`, `std.builtin.CallingConvention) we do not necessarily require an
explicit coercion instruction. Instead, Sema knows the type that is
required, and can perform the coercion after the fact. This means we can
use the `coerced_ty` result location kind, saving unnecessary coercion
instructions and therefore ZIR bytes.
This required a few enhancements to Sema to introduce missing coercions.
|
|
`sema.src` is a failed experiment. It introduces complexity, and makes
often unwarranted assumptions about the existence of instructions
providing source locations, requiring an unreasonable amount of caution
in AstGen for correctness. Eliminating it simplifies the whole frontend.
This required adding source locations to a few instructions, but the
cost in ZIR bytes should be counteracted by the other work on this
branch.
|
|
AstGen has logic to elide leading `dbg_stmt` instructions when multiple
are emitted consecutively; however, it only applied in some cases. A
simple reshuffle here makes this logic apply universally, saving some
bytes in ZIR.
|
|
This is a small optimization to generated ZIR. In any function where the
return type is not a trivial Ref, we know it is almost certainly not
`void` (unless the user aliased it or did something else weird to fool
AstGen), and thus the return type is very likely to be required for
return value RLS at some point. Thus, we can just emit one `ret_type` at
the start of the function and use it throughout.
This sees a very small improvement in overall ZIR bytes.
|
|
Also add corresponding invaidation logic to Zcu. Therefore, the only
invalidation logic which is not yet in place is `decl_val` dependencies.
|
|
This commit changes how declarations (`const`, `fn`, `usingnamespace`,
etc) are represented in ZIR. Previously, these were represented in the
container type's extra data (e.g. as trailing data on a `struct_decl`).
However, this introduced the complexity of the ZIR mapping logic having
to also correlate some ZIR extra data indices. That isn't really a
problem today, but it's tricky for the introduction of `TrackedInst` in
the commit following this one. Instead, these type declarations now
simply contain a trailing list of ZIR indices to `declaration`
instructions, which directly encode all data related to the declaration
(including containing the declaration's body). Additionally, the ZIR for
`align` etc have been split out into their own bodies. This is not
strictly necessary, but it's much simpler to understand for an
insignificant cost in bytes, and will simplify the resolution of #131
(where we may need to evaluate the pointer type, including align etc,
without immediately evaluating the value body).
|
|
This logic was previously in Sema, which was unnecessary complexity, and meant the issue was not detected unless the declaration was semantically analyzed. This commit finishes the work which 941090d started.
Resolves: #17916
|
|
|
|
fixes #18579
|
|
|
|
|
|
|
|
|
|
|
|
|
|
This change only emits the unwrap_errunion_err instruction if the error
capture is actually used in a branch.
|
|
|
|
|
|
|
|
|
|
Closes #17817
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
When a local variable is never used as an lvalue, we can determine that
`const` would be sufficient for this variable, so emit an error in this
case. More sophisticated checking is unfortunately not possible with
Zig's current analysis model, since whether an lvalue is actually
mutated depends on semantic analysis, in which some code paths may not
be analyzed, so attempting to determine this would result in false
positive compile errors.
It's worth noting that an unfortunate consequence of this is that any
field call `a.b()` will allow `a` to be `var`, even if `b` does not take
a pointer as its first parameter - this is again a necessary compromise
because the parameter type is not known until semantic analysis.
Also update `translate-c` to not trigger these errors. This is done by
replacing the `_ = @TypeOf(x)` emitted with `_ = &x` - the reference
there means that the local is permitted to be `var`. A similar strategy
will be used to prevent compile errors in the behavior tests, where we
sometimes want to force a value to be runtime-known.
Resolves: #224
|
|
Partially addresses #17916.
|
|
There are two optimizations here, which work together to avoid a
pathological case.
The first optimization is that AstGen now records the result type of an
array multiplication expression where possible. This type is not used
according to the language specification, but instead as an optimization.
In the expression '.{x} ** 1000', if we know that the result must be an
array, then it is much more efficient to coerce the LHS to an array with
length 1 before doing the multiplication. Otherwise, we end up with a
1000-element tuple which we must coerce to an array by individually
extracting each field.
Secondly, the previous logic would repeatedly extract element/field
values from the LHS when initializing the result. This is unnecessary:
each element must only be extracted once, and the result reused.
These changes together give huge improvements to compiler performance on
a pathological case: AIR instructions go from 65551 to 15, and total AIR
bytes go from 1.86MiB to 264.57KiB. Codegen time spent on this function
(in a debug compiler build) goes from minutes to essentially zero.
Resolves: #17586
|
|
This change allows struct field inits to use layout information
of their own struct without causing a circular dependency.
`semaStructFields` caches the ranges of the init bodies in the `StructType`
trailing data. The init bodies are then resolved by `resolveStructFieldInits`,
which is called before the inits are actually required.
Within the init bodies, the struct decl's instruction is repurposed to refer
to the field type itself. This is to allow us to easily rebuild the inst_map
mapping required for the init body instructions to refer to the field type.
Thanks to @mlugg for the guidance on this one!
|
|
This commit starts by making Zir.Inst.Index a nonexhaustive enum rather
than a u32 alias for type safety purposes, and the rest of the changes
are needed to get everything compiling again.
|
|
Make distinct error limit configurable (attempt #2)
|
|
|
|
After the previous commit, these make_ptr_const ZIR instructions are
redundant.
|
|
|
|
Closes #17343
|
|
Replaces the @fabs builtin with a new @abs builtins which accepts
floats, signed integers and vectors of said types.
|
|
This commit introduces the new `ref_coerced_ty` result type into AstGen.
This represents a expression which we want to treat as an lvalue, and
the pointer will be coerced to a given type.
This change gives known result types to many expressions, in particular
struct and array initializations. This allows certain casts to work
which previously required explicitly specifying types via `@as`. It also
eliminates our dependence on anonymous struct types for expressions of
the form `&.{ ... }` - this paves the way for #16865, and also results
in less Sema magic happening for such initializations, also leading to
potentially better runtime code.
As part of these changes, this commit also implements #17194 by
disallowing RLS on explicitly-typed struct and array initializations.
Apologies for linking these changes - it seemed rather pointless to try
and separate them, since they both make big changes to struct and array
initializations in AstGen. The rationale for this change can be found in
the proposal - in essence, performing RLS whilst maintaining the
semantics of the intermediary type is a very difficult problem to solve.
This allowed the problematic `coerce_result_ptr` ZIR instruction to be
completely eliminated, which in turn also simplified the logic for
inferred allocations in Sema - thanks to this, we almost break even on
line count!
In doing this, the ZIR instructions surrounding these initializations
have been restructured - some have been added and removed, and others
renamed for clarity (and their semantics changed slightly). In order to
optimize ZIR tag count, the `struct_init_anon_ref` and
`array_init_anon_ref` instructions have been removed in favour of using
`ref` on a standard anonymous value initialization, since these
instructions are now virtually never used.
Lastly, it's worth noting that this commit introduces a slightly strange
source of generic poison types: in the expression `@as(*anyopaque, &x)`,
the sub-expression `x` has a generic poison result type, despite no
generic code being involved. This turns out to be a logical choice,
because we don't know the result type for `x`, and the generic poison
type represents precisely this case, providing the semantics we need.
Resolves: #16512
Resolves: #17194
|
|
This required a third `if (found_already == null)` in another place in
AstGen.zig for this special case of `@export`.
Fixes #17188
|
|
This also modifies AstGen so that struct types use 1 bit each from the
flags to communicate if there are nonzero inits, alignments, or comptime
fields. This allows adding a struct type to the InternPool without
looking ahead in memory to find out the answers to these questions,
which is easier for CPUs as well as for me, coding this logic right now.
|
|
AstGen emits an error when a closure over a known-runtime value crosses
a namespace boundary. This usually makes sense: however, this usage is
actually valid if the capture is within a `@TypeOf` operand. Sema
already has a special case to allow such closure within `@TypeOf` when
AstGen could not determine a value to be runtime-known. This commit
simply introduces analagous logic to AstGen to allow `var`s to cross
namespace boundaries within `@TypeOf`.
|
