This commit deletes the `alloc_system` crate from the standard
distribution. This unstable crate is no longer needed in the modern
stable global allocator world, but rather its functionality is folded
directly into the standard library. The standard library was already the
only stable location to access this crate, and as a result this should
not affect any stable code.
This commit removes all jemalloc related submodules, configuration, etc,
from the bootstrap, from the standard library, and from the compiler.
This will be followed up with a change to use jemalloc specifically as
part of rustc on blessed platforms.
In investigating [an issue][1] with `panic_implementation` defined in an
executable that's optimized I once again got to rethinking a bit about the
`rustc_std_internal_symbol` attribute as well as weak lang items. We've sort of
been non-stop tweaking these items ever since their inception, and this
continues to the trend.
The crux of the bug was that in the reachability we have a [different branch][2]
for non-library builds which meant that weak lang items (and std internal
symbols) weren't considered reachable, causing them to get eliminiated by
ThinLTO passes. The fix was to basically tweak that branch to consider these
symbols to ensure that they're propagated all the way to the linker.
Along the way I've attempted to erode the distinction between std internal
symbols and weak lang items by having weak lang items automatically configure
fields of `CodegenFnAttrs`. That way most code no longer even considers weak
lang items and they're simply considered normal functions with attributes about
the ABI.
In the end this fixes the final comment of #51342
[1]: https://github.com/rust-lang/rust/issues/51342#issuecomment-414368019
[2]: 35bf1ae257/src/librustc/middle/reachable.rs (L225-L238)
This to-be-stable attribute is equivalent to `#[lang = "oom"]`.
It is required when using the alloc crate without the std crate.
It is called by `handle_alloc_error`, which is in turned called
by "infallible" allocations APIs such as `Vec::push`.
This turned out to be important on Windows.
Calling `handle_alloc_error(Layout:🆕:<[u8; 42]>())` caused:
```
Exception thrown at 0x00007FF7C70DC399 in a.exe: 0xC0000005:
Access violation reading location 0x000000000000002A.
```
0x2A equals 42, so it looks like the `Layout::size` field of type `usize`
was interpreted as a pointer to read from.
The acronym is not descriptive unless one has seen it before.
* Rename the `oom` function to `handle_alloc_error`. It was **stabilized in 1.28**, so if we do this at all we need to land it this cycle.
* Rename `set_oom_hook` to `set_alloc_error_hook`
* Rename `take_oom_hook` to `take_alloc_error_hook`
Bikeshed: `alloc` v.s. `allocator`, `error` v.s. `failure`
Per discussion in https://github.com/rust-lang/rust/issues/51245#issuecomment-393651083
This allows more flexibility in what can be done with the API. This also
splits `rtabort!` into `dumb_print` happening in the default hook and
`abort_internal`, happening in the actual oom handler after calling the
hook. Registering an empty function thus makes the oom handler not print
anything but still abort.
Cc: @alexcrichton
As discussed in
https://github.com/rust-lang/rust/issues/49668#issuecomment-384893456
and subsequent, there are use-cases where the OOM handler needs to know
the size of the allocation that failed. The alignment might also be a
cause for allocation failure, so providing it as well can be useful.
