f250802ce7
compiler: introduce @Restrict builtin
...
conservative, incomplete change
2025-08-31 20:40:15 -07:00
fc23fe90ce
std.os.windows: avoid dragging in baggage on non-windows
...
particularly noticeable when usize is not 4 or 8 bytes
2025-08-31 20:39:47 -07:00
ec36e0609f
delete behavior test that depends on std.fmt
...
behavior tests should have minimal dependency on std
2025-08-31 12:49:18 -07:00
150169f1e0
std.fmt: delete deprecated APIs
...
std.fmt.Formatter -> std.fmt.Alt
std.fmt.format -> std.Io.Writer.print
2025-08-31 12:49:18 -07:00
7fdd60df1c
Populate MSG struct for OpenBSD ( #25076 )
...
* update the MSG struct with the correct values for openbsd
* add comment with link to sys/sys/socket.h
---------
Co-authored-by: Brandon Mercer <bmercer@eutonian.com >
2025-08-30 22:04:32 +00:00
b7104231af
Merge pull request #25077 from ziglang/GenericReader
...
std.Io: delete GenericReader, AnyReader, FixedBufferStream; and related API breakage
2025-08-30 12:43:52 -07:00
151314346d
Merge pull request #25055 from ziglang/llvm21
...
LLVM 21
2025-08-30 21:00:53 +02:00
e2fdaea0b3
Revert "std.Io.Reader: work around llvm backend bug"
...
This reverts commit 530cc2c111
2025-08-30 06:04:25 -07:00
66b43234bb
std: disable sha3-512 single test on RISC-V with V support
...
https://github.com/ziglang/zig/issues/25083
2025-08-30 13:27:25 +02:00
07da9567e6
compiler: fix macos build
2025-08-30 00:48:50 -07:00
fadd268a60
upgrade more old API uses
2025-08-30 00:48:50 -07:00
9a0970a12b
rework std.Io.Writer.Allocating to support runtime-known alignment
...
Also, breaking API changes to:
* std.fs.Dir.readFileAlloc
* std.fs.Dir.readFileAllocOptions
2025-08-30 00:48:50 -07:00
f6879c35af
std.posix.test: fix mmap() test to use actual page size instead of 4096
2025-08-30 06:36:41 +02:00
59ee4e8e54
std.hash.xxhash: disable xxhash3 test on all mips64 targets
...
See also dd1de18f96https://github.com/ziglang/zig/issues/23807
2025-08-30 06:36:41 +02:00
af78c55676
std: skip some failing tests on hexagon
2025-08-30 06:36:41 +02:00
367be3777c
std.Thread: make unreachable errors in sleep() clearer
2025-08-30 06:36:41 +02:00
5b80900a5d
std.zig.system: update aarch64 and powerpc cpu detection for LLVM 21
2025-08-30 06:36:41 +02:00
0b9f5e2549
std.Target: fix alignment for int/long types on m68k
2025-08-30 06:36:41 +02:00
fdd32a2257
std.Target: add managarm OS tag
2025-08-30 06:36:41 +02:00
464eef390b
std.Target: make ultrasparc3 the baseline for sparc64-solaris
...
https://github.com/llvm/llvm-project/pull/149990
2025-08-30 06:36:41 +02:00
978555eea4
std.Target.x86: purge avx10.n-256, rename avx10.n_512 to avx10.n, require evex512 for avx512f
...
Intel has abandoned AVX10.N/128,256; AVX10.N is now always 512-bit.
2025-08-30 06:36:41 +02:00
2c0cc81e74
std.Target: bump feature bit count from 288 to 317
...
/lib/std/debug.zig:559:14: error: reached unreachable code
if (!ok) unreachable; // assertion failure
^~~~~~~~~~~
/lib/std/Target/riscv.zig:335:21: note: called at comptime here
std.debug.assert(len <= CpuFeature.Set.needed_bit_count);
~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2025-08-30 06:36:41 +02:00
587eddda28
std.Target: update CPU features to LLVM 21
2025-08-30 06:36:41 +02:00
79f267f6b9
std.Io: delete GenericReader
...
and delete deprecated alias std.io
2025-08-29 17:14:26 -07:00
558bea2a76
std.Io: delete CountingReader
2025-08-29 11:11:59 -07:00
43fbc37a49
std.debug.Pdb: migrate more towards new Reader API
...
There was some bug in this branch, and rather than diagnosing it, I
fully finished porting over to new Reader API. Did it fix the bug?
2025-08-28 22:41:06 -07:00
7da9e4b35e
std.tz: fix redundant endian handling
...
I didn't notice the check+swap before.
2025-08-28 18:30:57 -07:00
530cc2c111
std.Io.Reader: work around llvm backend bug
...
tracked by #25067 and I already have a fix cooking in another branch
2025-08-28 18:30:57 -07:00
8023f3dceb
fix not discarding delimiter
...
perhaps these APIs have the defaults backwards, eh?
2025-08-28 18:30:57 -07:00
f7884961c2
update more to avoid GenericWriter
2025-08-28 18:30:57 -07:00
ea3471288a
update GenericWriter usage found by test-cases
2025-08-28 18:30:57 -07:00
888f00e856
std.crypto.ml_kem: update to not use GenericWriter
2025-08-28 18:30:57 -07:00
e9a271cba3
std.tz: update to new Reader API
2025-08-28 18:30:57 -07:00
9860dd475a
std: delete most remaining uses of GenericWriter
2025-08-28 18:30:57 -07:00
cc931660eb
link.MachO: update to not use GenericWriter
2025-08-28 18:30:57 -07:00
2dc6ddd7e8
std.Io.Writer: add toArrayList/fromArrayList
2025-08-28 18:30:57 -07:00
57dbc9e74a
std.Io: delete GenericWriter
2025-08-28 18:30:57 -07:00
224fca7e0e
process.totalSystemMemory: Avoid overflow on Linux when totalram is a 32-bit usize
...
Fixes #25038
2025-08-28 17:05:39 +02:00
2b73c28cec
Reader.appendRemaining: Take ownership of the full allocated slice
...
Before this commit, calling appendRemaining with an ArrayList where list.items.len != list.capacity could result in illegal behavior if the Writer.Allocating resized the list during the appendRemaining call.
Fixes #25057
2025-08-28 07:56:50 -07:00
73a0b5441b
AstGen: forward result type through unary float builtins
...
Uses a new `float_op_result_ty` ZIR instruction tag.
2025-08-28 15:46:12 +01:00
a31950aa57
std.debug: remove @frameAddress() "UAF"
...
We can't call `@frameAddress()` and then immediately `return`! That
invalidates the frame. This *usually* isn't a problem, because the stack
walk `next` call will *probably* have a stack frame and it will
*probably* be at the exact same address, but neither of those is a
guarantee. On powerpc, presumably some unfortunate inlining was going
on, so this frame was indeed invalidated when we started walking frames.
We need to explicitly pass `@frameAddress` into any function which will
return before we actually walk the stack. Pretty simple patch.
Resolves : #24970
2025-08-28 10:56:11 +01:00
151c7dc74b
std.os.uefi: fix type error at MemoryType.format()
2025-08-28 07:50:38 +01:00
417c68d249
std.os.linux: powerpc syscalls clobber ctr and xer
...
https://git.musl-libc.org/cgit/musl/commit/?id=f6944eb3c4ce1c97dc39dc36d32390dc9f70b67b
2025-08-28 07:41:22 +02:00
50edad37ba
Merge pull request #25034 from ziglang/lzma
...
std.compress: update lzma, lzma2, and xz to new I/O API
2025-08-27 06:49:45 -07:00
12a58087a4
Fix TLS 1.2 client key exchange to use negotiated named group ( #25007 )
...
The TLS 1.2 implementation was incorrectly hardcoded to always send the
secp256r1 public key in the client key exchange message, regardless of
which elliptic curve the server actually negotiated.
This caused TLS handshake failures with servers that preferred other curves
like X25519.
This fix:
- Tracks the negotiated named group from the server key exchange message
- Dynamically selects the correct public key (X25519, secp256r1, or
secp384r1) based on what the server negotiated
- Properly constructs the client key exchange message with the
appropriate key size for each curve type
Fixes TLS 1.2 connections to servers like ziglang.freetls.fastly.net
that prefer X25519 over secp256r1.
2025-08-27 11:18:40 +02:00
ae2622bf82
std.c: add correct SOMAXCONN for BSDs
...
Note the previous "28" here for openbsd was some kind of copy
error long ago. That's the value of KERN.SOMAXCONN, which is an
entirely different thing.
2025-08-27 09:46:37 +02:00
68f590d430
std.compress.xz: fix 32-bit targets
2025-08-26 21:07:09 -07:00
668299f0db
std: update xz unit tests to new I/O API
2025-08-26 21:00:58 -07:00
980445f08b
std.compress.lzma: fix unpacked size checking logic
2025-08-26 21:00:58 -07:00
722e066173
std.compress.xz.Decompress: some tests passing
2025-08-26 21:00:58 -07:00
Andrew Kelley