C99 introduced designated initializers for structs. Omitted fields are
implicitly initialized to zero. Some C APIs are designed with this in
mind. Defaulting to zero values for translated struct fields permits Zig
code to comfortably use such an API.
Closes#8165
This introduces the concept of a "weak global name" into translate-c.
translate-c consists of two passes. The first is important, because it
discovers all global names, which are used to prevent naming conflicts:
whenever we see an identifier in the second pass, we can mangle it if it
conflicts with any global or any other in-scope identifier.
Unfortunately, this is a bit tricky for structs, unions, and enums. In
C, these types are not represented by normal identifers, but by separate
tags - `struct foo` does not prevent an unrelated identifier `foo`
existing. In general, we want to translate type names to user-friendly
ones such as `struct_foo` and `foo` where possible, but we can't
guarantee such names will not conflict with real variable names.
This is where weak global names come in. In the initial pass, when a
global type declaration is seen, `struct_foo` and `foo` are both added
as weak global names. This essentially means that we will use these
names for the type *if possible*, but if there is another global with
the same name, we will mangle the type name instead. Then, when actually
translating the declaration, we check whether there's a "true" global
with a conflicting name, in which case we mangle our name. If the
user-friendly alias `foo` conflicts, we do not attempt to mangle it: we
just don't emit it, because a mangled alias isn't particularly helpful.
Consider this C macro:
```c
#define FOO(x) struct x
```
Previously, translate-c did not detect that the `x` in the body referred
to the argument, so wrongly translated this code as using the
nonexistent `struct_x`. Since undefined identifiers are noticed in
AstGen, this prevents the translated file from being usable at all.
translate-c now instead detects this case and emits an appropriate
compile error in the macro's place.
Most of this migration was performed automatically with `zig fmt`. There
were a few exceptions which I had to manually fix:
* `@alignCast` and `@addrSpaceCast` cannot be automatically rewritten
* `@truncate`'s fixup is incorrect for vectors
* Test cases are not formatted, and their error locations change
If the C code had variables that were named the same as the prefixes used
for name mangling, such as "tmp" or "ref", then the codegen would generate
incorrect code in some cases. This was because these aliases were immediately
visible to expressions that actually needed to use the original name.
I introduced the concept of reserving aliases without enabling them. An alias
that isn't enabled isn't visible to expression translation, but is still
reserved so that sub-expressions generate aliases that don't overlap.
Add test cases to cover the cases that would break before this change.
Co-authored-by: Veikka Tuominen <git@vexu.eu>
Superceeds PR #12735 (now supporting all packed structs in GNU C)
Fixes issue #12733
This stops translating C packed struct as a Zig packed struct.
Instead use a regular `extern struct` with `align(1)`.
This is because (as @Vexu explained) Zig packed structs are really just integers (not structs).
Alignment issue is more complicated. I think @ifreund was the
first to notice it in his comment on PR #12735
Justification of my interpretion of the C(lang) behavior
comes from a careful reading of the GCC docs for type & variable attributes:
(clang emulates gnu's packed attribute here)
The final line of the documentation for __attribute__ ((aligned)) [on types] says:
> When used on a struct, or struct member, *the aligned attribute can only increase the alignment*; in order to decrease it, the packed attribute must be specified as well.
This implies that GCC uses the `packed` attribute for alignment purposes
in addition to eliminating padding.
The documentation for __attribute__((packed)) [on types], states:
> This attribute, attached to a struct, union, or C++ class type definition, specifies that each of its members (other than zero-width bit-fields) is placed to minimize the memory required. **This is equivalent to specifying the packed attribute on each of the members**.
The key is resolving this indirection, and looking at the documentation
for __attribute__((packed)) [on fields (wierdly under "variables" section)]:
> The packed attribute specifies that a **structure member should have the smallest possible alignment** — one bit for a bit-field and one byte otherwise, unless a larger value is specified with the aligned attribute. The attribute does not apply to non-member objects.
Furthermore, alignment is the only effect of the packed attribute mentioned in the GCC docs (for "common" architecture).
Based on this, it seems safe to completely substitute C 'packed' with Zig 'align(1)'.
Target-specific or undocumented behavior potentially changes this.
Unfortunately, the current implementation of `translate-c` translates as
`packed struct` without alignment info.
Because Zig packed structs are really integers (as mentioned above),
they are the wrong interpretation and we should be using 'extern struct'.
Running `translate-c` on the following code:
```c
struct foo {
char a;
int b;
} __attribute__((packed));
struct bar {
char a;
int b;
short c;
__attribute__((aligned(8))) long d;
} __attribute__((packed));
```
Previously used a 'packed struct' (which was not FFI-safe on stage1).
After applying this change, the translated structures have align(1)
explicitly applied to all of their fields AS EXPECTED (unless explicitly overriden).
This makes Zig behavior for `tranlsate-c` consistent with clang/GCC.
Here is the newly produced (correct) output for the above example:
```zig
pub const struct_foo = extern struct {
a: u8 align(1),
b: c_int align(1),
};
pub const struct_bar = extern struct {
a: u8 align(1),
b: c_int align(1),
c: c_short align(1),
d: c_long align(8),
};
```
Also note for reference: Since the last stable release (0.9.1),
there was a change in the language spec
related to the alignment of packed structures.
The docs for Zig 0.9.1 read:
> Packed structs have 1-byte alignment.
So the old behavior of translate-c (not specifying any alignment) was possibly correct back then.
However the current docs read:
> Packed structs have the same alignment as their backing integer
Suggsestive both to the change to an integer-backed representation
which is incompatible with C's notation.
This makes translate-c lower discards as `_ = @TypeOf(foo);` to avoid
tripping the "pointless discard" error.
Ideally, translate-c would avoid emitting pointless discards, in which
case this commit can be reverted, however, that is a separate
enhancement.
* Fix incorrect result when the first digit after the decimal point is not 0-9 - eg 0x0.ap0
* Fix compiler panic when the number starts with `0X` with a capital `X` - eg 0X0p0
* Fix compiler panic when the number has a decimal point immediately after `0x` - eg 0x.0p0
