summaryrefslogtreecommitdiff
path: root/src/codegen/mod.rs
AgeCommit message (Collapse)Author
2018-01-22codegen: Try to reasonably handle enum : bool.Emilio Cobos Álvarez
Just use the repr name we generate, since we generate constants for that. It's not worth trying to guess the actual type to use IMO. Bindings lose a bit of portability I guess, but that's really a lost bet already, so instead of special-casing bool and map constants, let's use a consistent representation everywhere. Fixes #1145
2018-01-19codegen: Don't generate repr(C) for enums.Emilio Cobos Álvarez
That's only undefined for enums with fields. Fixes #1224 See also: https://botbot.me/mozilla/rustc/2018-01-19/?msg=95934948&page=2
2018-01-06codegen: Be consistent about variadic signatures.Emilio Cobos Álvarez
Fixes #1216.
2017-12-29Don't generate symbols for pure virtual functions.Emilio Cobos Álvarez
Fixes #1197.
2017-12-12Always add `repr(C)` to rustified enumsNick Fitzgerald
If we don't, then eddyb's recent layout optimizations will do unexpected things to them. We also need to handle empty `enum`s without variants. When `repr(C)` is on a Rust `enum`, it cannot be empty, so we need to add a dummy.
2017-11-27Properly handle namespaces for enum configuration optionsTamir Duberstein
...by using canonical_path rather than canonical_name. Fixes #1125.
2017-11-23Auto merge of #1158 - glyn:large-bitfield-units, r=emiliobors-servo
Support bitfield allocation units larger than 64 bits Individual bitfields are still limited to at most 64 bits, but this restriction can be weakened when Rust supports `u128`. This implements issue #816. Usage notes: * Since common code is added to each generated binding, a program which uses more than one binding may need to work around the duplication by including each binding in its own module. * The values created by bitfield allocation unit constructors can be assigned directly to the corresponding struct fields with no need for transmutation. Implementation notes: `__BindgenBitfieldUnit` represents a bitfield allocation unit using a `Storage` type accessible as a slice of `u8`. The alignment of the unit is inherited from an `Align` type by virtue of the field: ``` align: [Align; 0], ``` The position of this field in the struct is irrelevant. It is assumed that the alignment of the `Storage` type is no larger than the alignment of the `Align` type, which will be true if the `Storage` type is, for example, an array of `u8`. This assumption is checked in a debug assertion. Although the double underscore (__) prefix is reserved for implementations of C++, there are precedents for this convention elsewhere in bindgen and so the convention is adopted here too. Acknowledgement: Thanks to @fitzgen for an initial implementation of `__BindgenBitfieldUnit` and code to integrate it into bindgen. r? @emilio
2017-11-21Support bitfield allocation units larger than 64 bitsNick Fitzgerald
Individual bitfields are still limited to at most 64 bits, but this restriction can be weakened when Rust supports u128. This implements issue #816. Usage notes: * Since common code is added to each generated binding, a program which uses more than one binding may need to work around the duplication by including each binding in its own module. * The values created by bitfield allocation unit constructors can be assigned directly to the corresponding struct fields with no need for transmutation. Implementation notes: __BindgenBitfieldUnit represents a bitfield allocation unit using a Storage type accessible as a slice of u8. The alignment of the unit is inherited from an Align type by virtue of the field: align: [Align; 0], The position of this field in the struct is irrelevant. The alignment of the Storage type is intended to be no larger than the alignment of the Align type, which will be true if the Storage type is, for example, an array of u8. Although the double underscore (__) prefix is reserved for implementations of C++, there are precedents for this convention elsewhere in bindgen and so the convention is adopted here too. Acknowledgement: Thanks to @fitzgen for an initial implementation of __BindgenBitfieldUnit and code to integrate it into bindgen.
2017-11-18Generate ptr::null rather than zero literalTamir Duberstein
2017-11-13Avoid bitfield getters and setters accessing memory beyond "self"Glyn Normington
This fixes https://github.com/rust-lang-nursery/rust-bindgen/issues/954.
2017-11-02Detect `#pragma pack(...)` and make `pack(n)` where `n > 1` opaqueNick Fitzgerald
This is a bandaid for #537. It does *not* fix the underlying issue, which requires `#[repr(packed = "N")]` support in Rust. However, it does make sure that we don't generate type definitions with the wrong layout, or fail our generated layout tests.
2017-11-01"Alignment of field" -> "Offset of field"Nick Fitzgerald
It isn't checking alignment at all; it's checking offsets.
2017-10-30Refactor derive_partialeq_or_partialord.Sergey Pepyakin
2017-10-28ir: References have no implicit template parameters.Emilio Cobos Álvarez
Fixes #1113.
2017-10-25Compute sizedness with a fixed-point analysisNick Fitzgerald
This fixes a couple bugs where we weren't properly adding an `_address` byte. It also helps pave the way for computing implicit fields (such as padding, `_address`, vtable pointers, etc) in its own pass, before codegen. Fixes #768
2017-10-24Auto merge of #1085 - pepyakin:builtin-clone-impls, r=fitzgenbors-servo
Derive `Clone` along with `Copy` on Rust 1.21 Fixes #934 r? @fitzgen or @emilio
2017-10-24Derive `Clone` along with `Copy` on latest stable.Sergey Pepyakin
2017-10-24Auto merge of #1091 - jrmuizel:linkage, r=emiliobors-servo
Store function linkage in the ir This lets us capture 'static inline' functions in the ir and filter them later down the pipeline. This is the first step on the way to handling these functions better.
2017-10-24Store function linkage in the irJeff Muizelaar
This lets us capture 'static inline' functions in the ir and filter them later down the pipeline. This is the first step on the way to handling these functions better.
2017-10-23Remove `CompInfo::needs_explicit_vtable` and use `HasVtable::has_vtable_ptr` ↵Nick Fitzgerald
instead
2017-10-13Handle unsigned integer constants greater than u32::MAX in codegenNick Fitzgerald
We were not checking signed-ness and emitting the appropriate types. Fixes #1040
2017-10-12Auto merge of #1079 - fitzgen:move-self-into-gen, r=pepyakinbors-servo
Move `self` into `ir::BindgenContext::gen` Small clean up. See each commit for details. r? @pepyakin
2017-10-12Make `ir::BindgenContext::gen` take ownership of `self`Nick Fitzgerald
`bindgen` follows a pipeline architecture, and we only ever generate bindings once. By taking ownership of `self`, we can enforce this. We can also remove checks inside `gen` for whether we have resolved type refs or not, since we now know that we haven't because it is guaranteed to only be called the one time.
2017-10-12Rename `bitfield` to `bitfield_width`Nick Fitzgerald
Its more clear what the methods/fields are returning/storing when we add "width" to the name.
2017-10-11Handle explicit align=1.Sergey Pepyakin
2017-10-11Refactor requires_explicit_alignSergey Pepyakin
2017-10-10Auto merge of #1071 - pepyakin:emit-hex-bitfield-masks, r=fitzgenbors-servo
Emit hex bitfield masks r? @fitzgen
2017-10-10Emit hex bitfield masksSergey Pepyakin
2017-10-10Auto merge of #1065 - liranringel:thiscall, r=fitzgenbors-servo
Add support for the thiscall ABI Fixes https://github.com/rust-lang-nursery/rust-bindgen/issues/541 The thiscall ABI is experimental, so in order to use it nightly is required and also the following statement: `#![feature(abi_thiscall)]` That's a problem because the `tests_expectations` crate (in the tests folder) tries to compile it (and stable is required).
2017-10-10Add support for the thiscall ABILiran Ringel
2017-10-09Fix bitfields in bindgen unionsSergey Pepyakin
2017-10-05Eagerly generate bitfield accessor namesSergey Pepyakin
Also clean a bit.
2017-10-04Remove early anonymous bitfield filtering and consolidate name methodOliver Geller
2017-10-03Derive partialeq "manually" when possibleSergey Pepyakin
Remove derive-partialeq-template-inst test. Add comments. Don't implement PartialEq for incomplete arrays Handle opaque bases and template instantiations Extract constrain_type. Extract `is whitelisted?` check Add failing partialeq-anonfield join for comps Fix: return insert if not whitelisted Delegate TypeRefs and alias to constrain_join. Delegate Template instantiations to constrain_join Add derive-partialeq-pointer.hpp test Update comment. Fix layout alignment larger that array limit Add missing test for derive-partialeq-anonfield.rs Clean Clean Fix typo in opaque-template-inst-member test Remove redudant stmt Add comment on can_supersede. Format impl_partialeq and leave a comment Extract requires_storage into it's own function. Clean
2017-10-02Tighten up `is_unsized` and `has_vtable` checks to operated on TypeIdNick Fitzgerald
2017-10-02Make methods/constructors/destructors use FunctionIdNick Fitzgerald
And also allow ID comparison across ID types, as this makes implementing the above much easier.
2017-10-02Introduce ModuleId to strongly type IDs pointing at ModulesNick Fitzgerald
2017-10-02A bunch of parsing things should return TypeIdNick Fitzgerald
2017-10-02Make `Enum::repr` into a `TypeId`Nick Fitzgerald
2017-10-02Make base members use TypeId rather than ItemIdNick Fitzgerald
2017-10-02Make a bunch more methods take generic idsNick Fitzgerald
2017-10-02Make comp fields contain `TypeId`sNick Fitzgerald
2017-10-02Make functions which take an ItemId generic to take any kind of idNick Fitzgerald
2017-10-02Make `TypeKind::{Alias,TemplateAlias,Array,Pointer,Reference}` use `TypeId`Nick Fitzgerald
This commit makes certain `TypeKind`s that can only reference other types use `TypeId` instead of `ItemId`.
2017-10-02Turn `build_templated_path` into `build_path`Nick Fitzgerald
None of the callers were passing template parameters.
2017-10-02Remove unused parameter to `codegen::utils::type_from_named`Nick Fitzgerald
2017-10-02Introduce the `TypeId` newtype over `ItemId`Nick Fitzgerald
This commit also makes `BindgenContext::resolve_type` take a `TypeId`, and adds unchecked conversions everywhere that we call it. Next, I'm going to go through the code base, replacing these unchecked conversions with checked ones, and tightening up types as I go.
2017-09-25codegen: Avoid generating wrong type for enums with unknown representations.Emilio Cobos Álvarez
2017-09-22Deanonymize fieldsSergey Pepyakin
Clean. Add timer. Clean codegen
2017-09-22Auto merge of #1016 - pepyakin:impl-debug-fixes, r=fitzgenbors-servo
Impl_debug fixes r? @fitzgen
generated by cgit v1.2.3 (git 2.46.0) at 2025-07-06 00:54:47 +0000