9160 Commits

Author SHA1 Message Date
Cody Tapscott
cbc85f4516 stage1: Fix seg-fault when slicing string literal with sentinel 2022-07-08 02:14:38 +03:00
VÖRÖSKŐI András
75c33ba85e Sema: add a note about @setEvalBranchQuota() when branch quota is exceeded
closes #11996
2022-07-08 02:13:33 +03:00
Andrew Kelley
0c78ece1c9
Merge pull request #12016 from Vexu/stage2-compile-errors
Stage2 compile error improvements
2022-07-07 14:52:28 -04:00
emma
81bbefe9b8
AstGen: fix catch payoad not checking for shadowing 2022-07-07 21:38:32 +03:00
Veikka Tuominen
5007f727e5 stage2: move C pointer allowzero error to AstGen 2022-07-07 10:50:06 +03:00
Veikka Tuominen
b5ac2b4330 Sema: improve array source location 2022-07-07 10:50:06 +03:00
Veikka Tuominen
27ee414159 Sema: improve slice source locations 2022-07-07 10:50:06 +03:00
Veikka Tuominen
2ca752ea1a Module: add .node_offset_un_op 2022-07-07 10:50:06 +03:00
Veikka Tuominen
89cef9f5f7 stage2: make line field of @src runtime known 2022-07-07 10:50:06 +03:00
Veikka Tuominen
226994cd7e Sema: misc error message fixes 2022-07-07 10:50:06 +03:00
Veikka Tuominen
c07c2d68c7 Sema: more runtime indexing comptime value checks 2022-07-07 10:50:06 +03:00
Veikka Tuominen
1569b9c165 Sema: validate pointer types 2022-07-07 10:50:05 +03:00
Veikka Tuominen
299836dbd9 Sema: panic at comptime + misc error message improvements 2022-07-07 10:50:05 +03:00
Veikka Tuominen
252388eb28 AstGen: move error_to_int, int_to_error and select to extended 2022-07-07 10:45:45 +03:00
Riccardo Binetti
0101a5f75e Sema: accept thumb and thumbeb with ARM calling conventions
Commit 3014a0d5f1dcbcfdcec3852ffd54f3c589fe3e83 added calling convention
validation, but left out thumb and thumbeb from the archs that can use the
ARM-specific calling conventions. This plus
a8a7f15106abef7c647d4de572a04b77ce048dd6 made compilation fail for thumb
targets.
2022-07-06 21:02:30 -04:00
Andrew Kelley
f3069f51aa Sema: fix type resolution during generic function instantiation
Found when trying to compile the std lib tests for x86_64-macos.
2022-07-06 17:00:17 -07:00
Cody Tapscott
86570b3e2c stage2: Fix corrupted Type when de-referencing field pointer 2022-07-05 15:02:13 -04:00
Andrew Kelley
6fc9f6c5f6 link.Coff: notice special windows symbols
in order to have a better default for subsystem. This brings stage2
behavior on par with stage1.
2022-07-04 17:20:01 -07:00
Andrew Kelley
9ecc47cd7c Sema: fix intFitsInType implementation
The function did not handle comptime_int in the case of lazy_align or
lazy_size.
2022-07-04 16:52:34 -07:00
Andrew Kelley
eba8892b84 Sema: resolve lazy value before hashing it for generic fn 2022-07-04 16:44:41 -07:00
Andrew Kelley
ecb4293afe stage2: disable unit tests when building with stage1
They trip LLVM assertions and spirv is not needed to bootstrap. Not
important for these tests to pass with stage1.
2022-07-04 16:20:33 -07:00
Andrew Kelley
d65e248ed1 stage2: ELF: improve error reporting when libc is missing
Future improvement: make plain error notes actually render as notes
rather than errors, but keep them as errors for the case of
sub-compilation errors, e.g. when compiler-rt has compilation errors.
2022-07-04 14:06:20 -07:00
Veikka Tuominen
050fef3c23 translate-c: do not try to get rid of do while loop
It might contain breaks and continues.

Closes #11994
2022-07-04 21:56:54 +03:00
Jakub Konka
5735ce39ae clang: parse --verbose flag as -v 2022-07-03 09:43:01 +02:00
Andrew Kelley
2360f8c490
Merge pull request #11974 from ziglang/fixfixfix
stage2 fixes
2022-07-01 17:29:31 -04:00
Andrew Kelley
7e2eb1326b
Merge pull request #11973 from Vexu/stage2-compile-errors
more stage2 compile error fixes
2022-07-01 17:29:18 -04:00
Jakub Konka
b79884eaf0 macho: implement pruning of unused segments and sections
This is a prelude to a more elaborate work which will implement
`-dead_strip` flag - garbage collection of unreachable atoms. Here,
when sorting sections, we also check that the section is actually
populated with some atoms, and if not, we exclude it from the final
linked image. This can happen when we do not import any symbols
from dynamic libraries in which case we will not be populating
the stubs sections or the GOT table, implying we can skip allocating
those sections. Furthermore, we also make a check that a segment
is actually occupied too, with the exception of `__TEXT` segment
which is non-optional given that it wraps the header and load commands
and thus is required by the `dyld` to perform dynamic linking, and
`__PAGEZERO` which is generally non-optional when the linked image
is an executable. For any other segment, if its section count is
zero, we mark it as dead and skip allocating it and generating
a load command for it.

This commit also includes some minor improvements to the linker such
as refactoring of the segment allocating codepaths, skipping
`__PAGEZERO` generation for dylibs, and skipping generation of zero-sized
atoms for special symbols such as `__mh_execute_header` and `___dso_handle`.
These special symbols are only allocated local and global symbol pair
and their VM addresses is set to the start of the `__TEXT` segment,
but no `Atom` is created, as it's not necessary given that they never
carry any machine code.

Finally, we now always force-link against `libSystem` which turns out
to be required for `dyld` to properly handle `LC_MAIN` load command
on older macOS versions such as 10.15.7.
2022-07-01 22:05:22 +02:00
Veikka Tuominen
2029601cb2 AstGen: use elem_{ptr,val}_node for array access syntax 2022-07-01 10:22:26 +03:00
Veikka Tuominen
a6bf8c2593 Sema: add more validation to zirFieldParentPtr 2022-07-01 10:22:25 +03:00
Veikka Tuominen
e6ebf56dd6 Sema: validate @intToEnum int operand type 2022-07-01 10:22:25 +03:00
Veikka Tuominen
ae7b32eb62 Sema: validate deref operator type and value 2022-07-01 10:22:25 +03:00
Veikka Tuominen
3c73f71177 Sema: prefer func.onwer_decl for compilelog src 2022-07-01 10:22:25 +03:00
Veikka Tuominen
3014a0d5f1 Sema: validate callconv 2022-07-01 10:22:25 +03:00
Veikka Tuominen
6d24c40b6e Sema: improve bitcast to enum error 2022-07-01 09:29:13 +03:00
Andrew Kelley
095e24e537 stage2: implement alignment calculation of vectors
closes #11856
2022-06-30 19:39:41 -07:00
Andrew Kelley
d3542be875 LLVM: be sure to never pass align(0) attribute
This can happen with pointers to zero-bit types.

This commit fixes an LLVM assertion being tripped.
2022-06-30 18:33:02 -07:00
Andrew Kelley
c030ec1884 LLVM: use unnamed struct llvm type for unions when necessary
The constant value lowering for unions was missing a check for whether
the payload was itself an unnamed struct. Lowerings of other types
already handle this case.

closes #11971
2022-06-30 18:33:02 -07:00
Andrew Kelley
152462e2e1 stage2: object format affects whether LLVM can be used 2022-06-30 18:33:02 -07:00
Andrew Kelley
1951051e3d
Merge pull request #11942 from Vexu/stage2-compile-errors
Move passing stage1 compile error tests to stage2
2022-06-30 18:39:46 -04:00
Andrew Kelley
77a334451f
Merge pull request #11967 from ziglang/runtime-float-negation
stage2: lower float negation explicitly + modify hash/eql logic for floats
2022-06-30 17:34:05 -04:00
Andrew Kelley
b3b96b5e28 LLVM: lower float negation with xor a constant
Rather than a compiler-rt call in the case that LLVM does not support
lowering the fneg instruction.
2022-06-30 11:48:37 -07:00
Veikka Tuominen
4f200eda9f stage2 llvm: ensure @tagName functions are unique 2022-06-30 21:47:25 +03:00
Veikka Tuominen
90ae37cc00 tests: move compile errors not planned for stage2 2022-06-30 09:57:38 +02:00
Veikka Tuominen
cc3336c784 Sema: add source location to coerce result ptr, fix negation error 2022-06-30 09:57:38 +02:00
Veikka Tuominen
979910dc38 Sema: validate shift amounts and switch ranges 2022-06-30 09:57:38 +02:00
Veikka Tuominen
6cadac18b8 Sema: improve auto generated union enum name 2022-06-30 09:57:38 +02:00
Veikka Tuominen
2e7dc5e151 Sema: improve vector overflow errors 2022-06-30 09:57:38 +02:00
Veikka Tuominen
03b356e34a Sema: improve @call errors 2022-06-30 09:57:38 +02:00
Veikka Tuominen
3204d00a5e move passing stage1 compile error tests to stage2 2022-06-30 09:57:38 +02:00
Andrew Kelley
c8531faaf5 stage2: hash/eql of fixed-size floats use bit pattern
Zig guarantees the memory layout of f16, f32, f64, f80, and f128 which
means for generic function purposes, values of these types need to be
compared on the basis of their bits in memory. This means nan-packing
can be used with generic functions, for example.

For comptime_float, the sign is observable, whether it is nan is
observable, but not any more kinds of bit patterns are observable.

This fixes the std.fmt tests that check printing "-nan".
2022-06-30 00:54:03 -07:00