34950 Commits

Author SHA1 Message Date
Andrew Kelley
8e27821b60 simplify diagnostics 2025-09-24 20:01:18 -07:00
Andrew Kelley
d1820d2a3e delete enough aro to make it compile 2025-09-24 20:01:18 -07:00
Andrew Kelley
328280b566 move translate-c helpers 2025-09-24 20:01:18 -07:00
Andrew Kelley
1bdcdbd996 delete all the translate-c tests
the ziglang/translate-c package has its own test suite, so these are
redundant
2025-09-24 20:01:17 -07:00
Andrew Kelley
f49a54745b compiler: update aro and translate-c to latest; delete clang translate-c 2025-09-24 19:57:28 -07:00
Alex Rønne Petersen
91b0adc4c1
Merge pull request #25231 from taylordotfish/bugfix/ppc-restore_rt
Fix `restore_rt` on PowerPC and remove unnecessary clobbers
2025-09-25 01:31:24 +02:00
Jacob Young
4a344de651 MappedFile: fix the insert range path not updating the root node 2025-09-24 16:09:46 -07:00
Jacob Young
94d319a10f x86_64: support more in/out forms
Closes #25303
2025-09-24 16:09:14 -07:00
Jacob Young
561e556aaf target: check for backend support for the new linker
Closes #25343
2025-09-24 16:08:26 -07:00
rpkak
9b3b7aa911 Integrate libc-test cases into the build system
zig build test-libc -Dlibc-test-path=/path/to/libc-test
2025-09-24 16:05:18 -07:00
rpkak
4fb08986cb optimize std.mem.swap 2025-09-24 12:29:43 -07:00
taylor.fish
c73df65ded Don't specify clobbers in restore_rt
Per @alexrp, this is unnecessary in naked functions.
2025-09-23 21:03:28 -07:00
taylor.fish
7f49dae284 Fix PowerPC restore_rt
Clang fails to compile the CBE translation of this code ("non-ASM
statement in naked function"). Similar to the implementations of
`restore_rt` on x86 and ARM, when the CBE is in use, this commit employs
alternative inline assembly that avoids using non-immediate input
operands.
2025-09-23 20:47:34 -07:00
Carter Snook
5bf52a6f50
std: always allow spawning processes when an env map is explicitly provided (#25092)
In a library, the two `builtin.link_libc` and `builtin.output_mode ==
.Exe` checks could both be false. Thus, you would get a compile error
even if you specified an `env_map` at runtime. This change turns the
compile error into a runtime panic and updates the documentation to
reflect the runtime requirement.
2025-09-24 03:12:28 +02:00
rpkak
bc512648db use copy_file_range syscall on linux 2025-09-24 03:08:12 +02:00
Kyle Schwarz
6260d2772f glibc: guard inet-fortified.h 2025-09-24 02:55:50 +02:00
taylor.fish
bdca0f93c4 Fix PowerPC syscalls causing invalid code from CBE
Fixes #25209.

On PowerPC, some registers are both inputs to syscalls and clobbered by
them. An example is r0, which initially contains the syscall number, but
may be overwritten during execution of the syscall.

musl and glibc use a `+` (read-write) constraint to indicate this, which
isn't supported in Zig. The current implementation of PowerPC syscalls
in the Zig standard library instead lists these registers as both inputs
and clobbers, but this results in the C backend generating code that is
invalid for at least some C compilers, like GCC, which doesn't support
the specifying the same register as both an input and a clobber.

This PR changes the PowerPC syscall functions to list such registers as
inputs and outputs rather than inputs and clobbers. Thanks to jacobly0
who pointed out that it's possible to have multiple outputs; I had
gotten the wrong idea from the documentation.
2025-09-24 02:54:55 +02:00
Andrew Kelley
ba137783ed
forbid trivial local address returned from functions (#25333)
progress towards #25312
2025-09-23 23:37:53 +00:00
Alex Rønne Petersen
7b92d5f405
std.pie: fix register constraint in getDynamicSymbol() for s390x (#25327)
If the compiler happens to pick `ret = r0`, then this will assemble to
`ag r0, 0` which is obviously not what we want. Using `a` instead of `r` will
ensure that we get an appropriate address register, i.e. `r1` through `r15`.

Re-enable pie_linux for s390x-linux which was disabled in
ed7ff0b693037078f451a7c6c1124611060f4892.
2025-09-22 18:36:47 +02:00
Alex Rønne Petersen
26825e9506
Merge pull request #25324 from alexrp/freebsd
Some changes to prepare for FreeBSD CI
2025-09-22 05:04:31 +02:00
Alex Rønne Petersen
16c18b835e
ci: stop building FreeBSD module tests on x86_64-linux
They're now built on the x86_64-freebsd machine.
2025-09-22 01:39:29 +02:00
alexrp
3675074c7c
ci: add x86_64-freebsd scripts 2025-09-22 01:37:32 +02:00
alexrp
b3432c2796
test: disable some stack trace tests on FreeBSD 2025-09-22 01:37:32 +02:00
alexrp
b2d2b441b2
test: disable test-link on FreeBSD
https://github.com/ziglang/zig/issues/25323
2025-09-22 01:37:32 +02:00
alexrp
f7d62009ff
std.posix: remove bogus assert that SIGRTMAX < NSIG 2025-09-22 01:37:32 +02:00
alexrp
0700ec35bd
compiler: don't use self-hosted backend on any BSD yet
There are some blocking bugs in the self-hosted ELF linker.
2025-09-22 01:37:32 +02:00
Jacob Young
f58200e3f2 Elf2: create a new linker from scratch
This iteration already has significantly better incremental support.

Closes #24110
2025-09-21 14:09:14 -07:00
Andrew Kelley
2a97e0af6d CI: stop testing x86_64-macos
There are two reasons for this:

1. Apple is about to drop support for this target. Zig will keep support
   but move it to a lower tier - one that does not require continuous CI
   testing. Support for this target will be maintained by the enthusiasm
   of contributors but will not block other bug fixes and enhancements.
2. This is our only non-self-hosted action runner. We are migrating away
   from GitHub soon at which point this runner will no longer be
   available.
2025-09-21 12:36:50 -07:00
Alex Rønne Petersen
4d8dcccbd4
ci: temporarily disable riscv64-linux
GitHub sucks:

    Sep 20 20:49:21 ganymede runsvc.sh[82817]: An error occured: Runner version v2.326.0 is deprecated and cannot receive messages.
    Sep 20 20:49:21 ganymede runsvc.sh[82817]: Runner listener exited with error code 1
    Sep 20 20:49:21 ganymede runsvc.sh[82817]: Runner listener exit with terminated error, stop the service, no retry needed.
2025-09-21 17:29:57 +02:00
Frank Denis
bdc31c9561 aarch64/zonCast: don't return a pointer to a stack element
Elements are computed at comptime, so don't declare them as "var".
2025-09-21 05:01:41 -07:00
Andrew Kelley
594cb38fcb
Merge pull request #25302 from ziglang/growCapacity
std: remove loop from growCapacity
2025-09-21 04:55:39 -07:00
Andrew Kelley
010d9a63f2
Merge pull request #25154 from ziglang/no-decl-val-3
rework byval ZIR instructions; forbid runtime vector indexes
2025-09-21 01:49:28 -07:00
Andrew Kelley
633162eb0c add behavior test: comptime C pointer to optional pointer 2025-09-20 18:58:32 -07:00
Andrew Kelley
1c9ac9dbb7 add behavior test: avoid unused field function body compile error 2025-09-20 18:51:43 -07:00
Andrew Kelley
500afbf076 add behavior test: resist alias of explicit copy...
...of array passed as arg

closes #22906
2025-09-20 18:51:26 -07:00
Andrew Kelley
ed7ff0b693 allow some test cases to regress
tracked by #24061 - these should be re-enabled once that is solved.
2025-09-20 18:33:01 -07:00
mlugg
0e16d933be fix rebase error 2025-09-20 18:33:01 -07:00
mlugg
20925f2957 Revert "frontend: another packedStructFieldPtrInfo fix"
This reverts commit dedccecda944f88a5278c12c24ffbea46126de63.
2025-09-20 18:33:01 -07:00
Jacob Young
f81a721e41 standalone: fix misaligned stack crash 2025-09-20 18:33:01 -07:00
Jacob Young
5144f10ec9 aarch64: fix behavior failures 2025-09-20 18:33:01 -07:00
Jacob Young
60cdacaff2 x86_64: rewrite vector element pointer access 2025-09-20 18:33:01 -07:00
Andrew Kelley
2ba03e98c8 disable failing stage2_aarch64 behavior tests 2025-09-20 18:33:01 -07:00
Andrew Kelley
5d8c1fb6ab frontend: another packedStructFieldPtrInfo fix
it was calculating host integer size in a wrong way. just use integer
abi size
2025-09-20 18:33:01 -07:00
mlugg
baaf715d21 cases: update for new error
The latest bugfix reverted this case to its old behavior (which is a
reasonable behavior to have).
2025-09-20 18:33:01 -07:00
mlugg
fcc7e378f8 Revert "delete failing test case"
This reverts commit ba4d4602ab9bb0dc17fc0d57141d9324bdbb356d.
2025-09-20 18:33:01 -07:00
mlugg
2f9d8d5a9b Zcu: fix analysis of type of decl with inferred type
If the `nav_ty` is resolved by the `nav_val`, then we need to also mark
the `nav_ty` as in progress when we begin resolving the `nav_val`.
2025-09-20 18:33:01 -07:00
Andrew Kelley
d7d50496d9 langref: don't assume too much about pointer to packed struct field 2025-09-20 18:33:01 -07:00
Andrew Kelley
725dec6aa7 Sema: add missed logic to checkPtrAttributes
It wasn't checking bit pointer data.
2025-09-20 18:33:01 -07:00
Andrew Kelley
7ffe068a79 delete failing test case
Matthew can revert this commit when he's ready to tackle the assertion
failure
2025-09-20 18:33:01 -07:00
Jacob Young
cbbb67c5df x86_64: fix safety crashes in storeRegs 2025-09-20 18:33:01 -07:00