Jakub Konka
e4039cecc7
x64: fix (un)wrapping error unions + refactor
2022-03-06 19:02:02 +01:00
Luuk de Gram
43cb19ea4d
wasm: Implement @wasmMemoryGrow builtin
...
Similarly to the other wasm builtin, this implements the grow variation where the memory
index is a comptime known value. The operand as well as the result are runtime values.
This also verifies during semantic analysis the target we're building for is wasm, or else
emits a compilation error. This means that other backends do not have to handle this AIR instruction,
other than the wasm and LLVM backends.
2022-03-03 16:33:46 -07:00
Jakub Konka
3ec74a1cd8
codegen: handle elem_ptr when lowering to memory
...
* x64: handle storing from-to non-stack memory
2022-03-02 20:04:34 -05:00
Jakub Konka
77072d1a17
x64: fix bug in lowering optionals directly to immediate
2022-03-02 17:40:37 +01:00
Jakub Konka
c9d1db7e8e
x64: fix incorrect calc of rdi spill stack loc for backpatching
2022-03-02 14:53:02 +01:00
Jakub Konka
350bf9db13
x64: fix intCast to properly clear out dest register
2022-03-02 14:34:43 +01:00
Jakub Konka
d35cae551e
x64: rectify and add missing optionals bits
...
Includes changes/additions to:
* `wrap_optional`
* `optional_payload`
* `isNull` helper
2022-03-02 14:05:29 +01:00
Jakub Konka
e8eb9778cc
codegen: lower field_ptr to memory across linking backends
...
This requires generating an addend for the target relocation as
the field pointer might point at a field inner to the container.
2022-03-01 22:03:18 +01:00
Jakub Konka
52205a3c16
x64: fix array to slice impl
2022-03-01 17:59:39 +01:00
Jakub Konka
caa4e30ef4
x64: impl airMemset using inline memset
2022-03-01 15:21:10 +01:00
Jakub Konka
5a6f4395e6
x64: impl airMemcpy using inline memcpy
2022-03-01 14:29:48 +01:00
Jakub Konka
cfbc3537ef
x64: pass more behavior tests
2022-02-28 23:20:05 +01:00
Jakub Konka
66d28d0f20
x64: implement get_union_tag for register
2022-02-28 23:20:05 +01:00
Jakub Konka
decc90e0e7
x64: clean up loadMemPtrIntoRegister abstraction
2022-02-28 23:20:05 +01:00
Jakub Konka
12cdb36c5b
codegen: fix padding calculation for error unions when lowering
...
* do not track `rdi` register before `call` inst, but instead freeze
it from further use, until `call` has been realised
* pass more error union tests
2022-02-28 23:20:05 +01:00
Jakub Konka
05431d7c4a
x64: impl unwrap_errunion_payload and unwrap_errunion_err for register
2022-02-28 23:20:05 +01:00
Andrew Kelley
d5100dc815
stage2: fix frame_address AIR instruction
...
Various places were assuming different union tags. Now it is
consistently a no-op instruction, just like the similar
instruction ret_addr.
2022-02-28 13:38:33 -07:00
Veikka Tuominen
90bce11f62
stage2: implement @frameAddress
2022-02-28 13:09:14 -07:00
Jakub Konka
16f9774d2d
x64: fix switch condition mir; pass more union tests
2022-02-28 15:25:40 +01:00
Jakub Konka
a61ac9ecbf
x64: fix store with ABI size > 8 on stack; pass union tests
2022-02-28 12:14:41 +01:00
Jakub Konka
57a823582c
x64: impl airGetUnionTag
2022-02-28 11:40:25 +01:00
Jakub Konka
4d0c48738b
x64: make lowerUnnamedConst a fallthrough condition
2022-02-28 11:40:25 +01:00
Jakub Konka
1dc05e9e77
x64: impl airSetUnionTag
2022-02-28 11:40:21 +01:00
Andrew Kelley
2687b8f7f4
stage2: implement @unionInit
...
The ZIR instruction `union_init_ptr` is renamed to `union_init`.
I made it always use by-value semantics for now, not taking the time to
invest in result location semantics, in case we decide to change the
rules for unions. This way is much simpler.
There is a new AIR instruction: union_init. This is for a comptime known
tag, runtime-known field value.
vector_init is renamed to aggregate_init, which solves a TODO comment.
2022-02-26 20:59:36 -07:00
Jakub Konka
e0f5627d4a
x64+aarch64: check for pointer to zero-bit type when lowering decl
...
Unless the pointer is a pointer to a function, if the pointee type
has zero-bits, we need to return `MCValue.none` as the `Decl` has
not been lowered to memory, and therefore, any GOT reference will be
wrong.
2022-02-25 21:59:19 +01:00
Jakub Konka
4b14384989
aarch64: check if type has runtime bits before allocating mem ptr
2022-02-25 21:59:19 +01:00
Veikka Tuominen
b034c45b2b
stage2: implement fieldParentPtr
2022-02-24 19:48:34 +02:00
Jakub Konka
2f0299c3cf
x64: account for multiple returns from functions
...
This is necessary to correctly adjust for spilling of the %rdi
register in the callee.
2022-02-23 19:29:47 +01:00
Jakub Konka
d657ede324
x64: disable printing results on macos until I fix the linker
...
Hopefully, this will make the CI green, and in the meantime I can
fix the bugs in the MachO linker.
2022-02-22 21:57:44 +01:00
Jakub Konka
3be6c79ca2
x64: spill compare flags between blocks, extern calls, and cmp insts
...
This is just the first step towards the final solution as to get
here I had omit a safety assert check in `getResolvedInst` helper.
2022-02-22 21:56:34 +01:00
Jakub Konka
b9b4e4671f
x64: add naive impl of else in switch
2022-02-22 21:56:34 +01:00
Jakub Konka
2617117221
x64: fix lowering of error unions (we didn't pad to alignment)
...
* fix returning large values on stack from procedure calls - we
need to explicitly specify source and dest base registers for
`genSetStack` as well
2022-02-22 21:56:34 +01:00
Jakub Konka
87a0d754be
x64: add basic, naive impl of switch
2022-02-22 21:56:34 +01:00
Jakub Konka
5ba66911fa
x64: add basic impl of shl for integers
2022-02-22 21:56:34 +01:00
Jakub Konka
71dda25f14
x64: add basic impl of minimum builtin for ints
2022-02-22 21:56:34 +01:00
Jakub Konka
57c9eec232
x64: implement unwrap_err_payload and unwrap_err_err for stack offset
2022-02-22 21:56:34 +01:00
Jakub Konka
f4f23e307c
codegen: lower error_set and error_union
2022-02-22 21:56:34 +01:00
Jakub Konka
358b544157
x64: recover address in rdi if was spilled
2022-02-22 21:56:34 +01:00
Jakub Konka
630da643c6
x64: pass pointer to stack for return value in .rdi register
2022-02-22 21:56:34 +01:00
Jakub Konka
478f770053
x64: first, really horrible attempt at returning values on stack
2022-02-22 21:56:34 +01:00
Jakub Konka
e582c0a043
x64: reorder resolving call params: return values first, then params
2022-02-22 21:56:34 +01:00
Jakub Konka
8bc95b22dc
x64: sub is non-commutative
2022-02-22 21:56:34 +01:00
Andrew Kelley
74303a3d95
Merge pull request #10925 from Vexu/stage2
...
stage2: support anon init through error unions and optionals
2022-02-21 14:18:17 -05:00
Veikka Tuominen
27c63bf433
stage2: implement errunion_payload_ptr_set
2022-02-19 20:48:00 +02:00
Jakub Konka
da86839af0
x64: clean up implementation of divs, mod, rem for integers
2022-02-19 14:24:03 +01:00
Jakub Konka
bd396d7e07
x64: add unsigned div and move logic into a helper fn
2022-02-19 14:24:03 +01:00
Jakub Konka
2c13a4b87e
x64: implement div_exact for ints (signed+unsigned)
2022-02-19 14:24:03 +01:00
Andrew Kelley
2e1c16d649
Merge pull request #10924 from ziglang/air-independence-day
...
AIR independence day
2022-02-19 02:57:48 -05:00
Andrew Kelley
4e1e5ab622
stage2: make AIR not reference ZIR for inline assembly
...
Instead it stores all the information it needs to into AIR.
closes #10784
2022-02-18 19:41:32 -07:00
Cody Tapscott
ef417f19e1
stage2: Implement @bitReverse and @byteSwap
...
This change implements the above built-ins for Sema and the LLVM
backend. Other backends have had placeholders added for lowering.
2022-02-18 14:28:32 -07:00
