| Age | Commit message (Collapse) | Author |
|---|
|
Now that we're close to supporting all the types, get rid of the `else`
prong and explicitly list out those types that are not yet implemented.
Thanks @g-w1
|
|
They are lowered directly as the integer tag type, with no typedef.
|
|
|
|
|
|
|
|
|
|
See #6542 for more details. Upon investigation, this change is not
needed.
|
|
Introduce `ResultLoc.none_or_ref` which is used by field access
expressions to avoid unnecessary loads when the field access itself
will do the load. This turns:
```zig
p.y - p.x - p.x
```
from
```zir
%14 = load(%4) node_offset:8:12
%15 = field_val(%14, "y") node_offset:8:13
%16 = load(%4) node_offset:8:18
%17 = field_val(%16, "x") node_offset:8:19
%18 = sub(%15, %17) node_offset:8:16
%19 = load(%4) node_offset:8:24
%20 = field_val(%19, "x") node_offset:8:25
```
to
```zir
%14 = field_val(%4, "y") node_offset:8:13
%15 = field_val(%4, "x") node_offset:8:19
%16 = sub(%14, %15) node_offset:8:16
%17 = field_val(%4, "x") node_offset:8:25
```
Much more compact. This requires `Sema.zirFieldVal` to support both
pointers and non-pointers.
C backend: Implement typedefs for struct types, as well as the following
TZIR instructions:
* mul
* mulwrap
* addwrap
* subwrap
* ref
* struct_field_ptr
Note that add, addwrap, sub, subwrap, mul, mulwrap instructions are all
incorrect currently and need to be updated to properly handle wrapping
and non wrapping for signed and unsigned.
C backend: change indentation delta to 1, to make the output smaller and
to process fewer bytes.
I promise I will add a test case as soon as I fix those warnings that
are being printed for my test case.
|
|
|
|
rework ZIR memory layout; overhaul source locations
|
|
|
|
Wanted to make sure those new test cases still pass.
Also grab that CI fix so we can get those green check marks.
|
|
|
|
|
|
This way we don't have to pass src to every function and we can simply
use the first node as the lazy source location for all the todo
errors.
|
|
|
|
|
|
The LLVM backend is now structured into 3 different structs, namely
Object, DeclGen and FuncGen. Object represents an object that is
generated by the LLVM backend. DeclGen is responsible for generating
a decl and FuncGen is responsible for generating llvm instructions
from tzir in a function.
|
|
this allows to compile with ninja
|
|
(with a lot of things commented out)
|
|
There are some `@panic("TODO")` in there but I'm trying to get the
branch to the point where collaborators can jump in.
Next is to repair the seam between LazySrcLoc and codegen's expected
absolute file offsets.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
We can now codegen optionals! This includes the following instructions:
- is_null
- is_null_ptr
- is_non_null
- is_non_null_ptr
- optional_payload
- optional_payload_ptr
- br_void
Also includes a test for optionals.
|
|
|
|
|
|
stage2: Add support for extern functions for the wasm backend
|
|
This type is not widely applicable enough to be a public part of the
public interface of the std.
The current implementation in only fully utilized by the zig fmt
implementation, which could benefit by even tighter integration as
will be demonstrated in the next commit. Therefore, move the current
io.AutoIndentingStream to lib/std/zig/render.zig.
The C backend of the self hosted compiler also use this type currently,
but it does not require anywhere near its full complexity. Therefore,
implement a greatly simplified version of this interface in
src/codegen/c.zig.
|
|
Previously, the registers included r0, r1, r2, r3 which are not
included in the callee saved registers according to the Procedure Call
Standard for the ARM Architecture.
|
|
|
|
|
|
Stage 2: SPIR-V setup
|
|
stage2: wasm control flow
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
breaks correctly
|
|
stage1: don't filter test blocks with empty label
|
