Conflicts:
* build.zig
* src/Compilation.zig
* src/codegen/spirv/spec.zig
* src/link/SpirV.zig
* test/stage2/darwin.zig
- this one might be problematic; start.zig looks for `main` in the
root source file, not `_main`. Not sure why there is an underscore
there in master branch.
Use .bold instead of .white, the effect is the same for light-on-dark
terminals but greatly improves the readability for dark-on-light ones.
Closes#8761
Flush all the register windows to stack before starting the stack walk,
we may otherwise try to read garbage and crash and burn.
Add a few comptime annotations to debloat some functions.
This temporary patch fixes a segfault caused by miscompilation
by the LLD when generating stubs for initialization of thread local
storage. We effectively bypass TLS in the default panic handler
so that no segfault is generated and the stack trace is correctly
reported back to the user.
Note that, this is linked directly to a bigger issue with LLD
ziglang/zig#7527 and when resolved, we only need to remove the
`comptime` code path introduced with this patch to use the default
panic handler that relies on TLS.
Co-authored-by: Andrew Kelley <andrew@ziglang.org>
* move concurrency primitives that always operate on kernel threads to
the std.Thread namespace
* remove std.SpinLock. Nobody should use this in a non-freestanding
environment; the other primitives are always preferable. In
freestanding, it will be necessary to put custom spin logic in there,
so there are no use cases for a std lib version.
* move some std lib files to the top level fields convention
* add std.Thread.spinLoopHint
* add std.Thread.Condition
* add std.Thread.Semaphore
* new implementation of std.Thread.Mutex for Windows and non-pthreads Linux
* add std.Thread.RwLock
Implementations provided by @kprotty
* split std.ResetEvent into:
- ResetEvent - requires init() at runtime and it can fail. Also
requires deinit().
- StaticResetEvent - can be statically initialized and requires no
deinitialization. Initialization cannot fail.
* the POSIX sem_t implementation can in fact fail on initialization
because it is allowed to be implemented as a file descriptor.
* Completely define, clarify, and explain in detail the semantics of
these APIs. Remove the `isSet` function.
* `ResetEvent.timedWait` returns an enum instead of a possible error.
* `ResetEvent.init` takes a pointer to the ResetEvent instead of
returning a copy.
* On Darwin, `ResetEvent` is implemented using Grand Central Dispatch,
which is exposed by libSystem.
stage2 changes:
* ThreadPool: use a single, pre-initialized `ResetEvent` per worker.
* WaitGroup: now requires init() and deinit() and init() can fail.
- Add a `reset` function.
- Compilation initializes one for the work queue in creation and
re-uses it for every update.
- Rename `stop` to `finish`.
- Simplify the implementation based on the usage pattern.
stderr_mutex may still be held when we reach the signal handler, grab
our own stderr handle to print the error messages and avoid deadlocking.
Closes#7247
* Move leb128 out of debug and remove trivial *mem functions as discussed in #5588
* Turns out one of the *Mem functions was used by MachO. Replaced with trivial use of FixedBufferStream.
* Smaller startup sequence for ppc64
* Terminate the frame-pointer chain when executing _start
* Make the stack traces work on ppc64
* Make the stack traces coloured on ppc64, some ioctls numbers are
different and the whole set of constants should be audited.
* std.Mutex API is improved to not have init() deinit(). This API is
designed to support static initialization and does not require any
resource cleanup. This also happens to work around some kind of
stage1 behavior that wasn't letting the new allocator mutex code
get compiled.
* the general purpose allocator now returns a bool from deinit()
which tells if there were any leaks. This value is used by the test
runner to fail the tests if there are any.
* self-hosted compiler is updated to use the general purpose allocator
when not linking against libc.
`std.GeneralPurposeAllocator` is now available. It is a function that
takes a configuration struct (with default field values) and returns an
allocator. There is a detailed description of this allocator in the
doc comments at the top of the new file.
The main feature of this allocator is that it is *safe*. It
prevents double-free, use-after-free, and detects leaks.
Some deprecation compile errors are removed.
The Allocator interface gains `old_align` as a new parameter to
`resizeFn`. This is useful to quickly look up allocations.
`std.heap.page_allocator` is improved to use mmap address hints to avoid
obtaining the same virtual address pages when unmapping and mapping
pages. The new general purpose allocator uses the page allocator as its
backing allocator by default.
`std.testing.allocator` is replaced with usage of this new allocator,
which does leak checking, and so the LeakCheckAllocator is retired.
stage1 is improved so that the `@typeInfo` of a pointer has a lazy value
for the alignment of the child type, to avoid false dependency loops
when dealing with pointers to async function frames.
The `std.mem.Allocator` interface is refactored to be in its own file.
`std.Mutex` now exposes the dummy mutex with `std.Mutex.Dummy`.
This allocator is great for debug mode, however it needs some work to
have better performance in release modes. The next step will be setting
up a series of tests in ziglang/gotta-go-fast and then making
improvements to the implementation.
* breaking changes to the API. Some of the weird decisions from before
are changed to what would be more expected.
- `get` returns `?V`, use `getEntry` for the old API.
- `put` returns `!void`, use `fetchPut` for the old API.
* HashMap now has a comptime parameter of whether to store hashes with
entries. AutoHashMap has heuristics on whether to set this parameter.
For example, for integers, it is false, since equality checking is
cheap, but for strings, it is true, since equality checking is
probably expensive.
* The implementation has a separate array for entry_index /
distance_from_start_index. Entries no longer has holes; it is an
ArrayList, and iteration is simpler and more cache coherent.
This is inspired by Python's new dictionaries.
* HashMap is separated into an "unmanaged" and a "managed" API. The
unmanaged API is where the actual implementation is; the managed API
wraps it and provides a more convenient API, storing the allocator.
* Memory usage: When there are less than or equal to 8 entries, HashMap
now incurs only a single pointer-size integer as overhead, opposed to
using an ArrayList.
* Since the entries array is separate from the indexes array, the holes
in the indexes array take up less room than the holes in the entries
array otherwise would. However the entries array also allocates
additional capacity for appending into the array.
* HashMap now maintains insertion order. Deletion performs a "swap
remove". It's now possible to modify the HashMap while iterating.
