See @the8472 comment's on Github:
https://github.com/rust-lang/rust/pull/60145#issuecomment-485424229
> I don't think is_reserved including ranges marked for future use is
> a good idea since those future uses may be realized at at some point
> and then old software with is_reserved filters may have false
> positives. This is not a hypothetical concern, such issues have been
> encountered before when IANA assigned previously reserved /8 address
> blocks.
As per @therealbstern's comment[0]:
The implementation of Ipv4::is_global is not complete, according to the
IANA IPv4 Special-Purpose Address Registry.
- It compares the address to 0.0.0.0, but anything in 0.0.0.0/8
should not be considered global.
- 0/8 is not global and is currently forbidden because
some systems used to treat it as the local network.
- The implementation of Ipv4::is_unspecified is correct.
0.0.0.0 is the unspecified address.
- It does not examine 100.64.0.0/10, which is "Shared Address
Space" and not global.
- Ditto 192.0.0.0/24 (IETF Protocol Assignments), except for
192.0.0.9/32 and 192.0.0.10/32, which are carved out as
globally reachable.
- 198.18.0.0/15 is for "Benchmarking" and should not be globally
reachable.
- 240.0.0.0/4 is reserved and not currently reachable
RFC 4291 is a little unclear about what is a unicast link local address.
According to section 2.4, the entire fe80::/10 range is reserved for
these addresses, but section 2.5.3 defines a stricter format for such
addresses.
After a discussion[0] is has been decided to add a different method for
each definition, so this commit:
- renames is_unicast_link_local() into is_unicast_link_local_strict()
- relaxed the check in is_unicast_link_local()
[0]: https://github.com/rust-lang/rust/issues/27709#issuecomment-400370706
SGX target: fix std unit tests
This fixes some tests and some code in the SGX sys implementation to make the `std` unit test suite pass.
#59009 must be merged first.
This commit deletes the `connect_timeout_unbound` test from the standard
library which, unfortunately, is by definition eventually going to be a
spuriously failing test. There's no way to reserve a port as unbound so
we can rely on ecosystem testing for this feature for now.
Closes#52590
Add vectored read and write support
This functionality has lived for a while in the tokio ecosystem, where
it can improve performance by minimizing copies.
r? @alexcrichton
Explain a panic in test case net::tcp::tests::double_bind
Those who try to build libstd on the Windows Subsystem for Linux experience a single failing test, where the point of failure is an explicit but anonymous panic, as reported in https://github.com/rust-lang/rust/issues/49367
This commit somewhat explains why and allows diagnosing a little.
There are two big categories of changes in here
- Removing lifetimes from common traits that can essentially never user a lifetime from an input (particularly `Drop` & `Debug`)
- Forwarding impls that are only possible because the lifetime doesn't matter (like `impl<R: Read + ?Sized> Read for &mut R`)
I omitted things that seemed like they could be more controversial, like the handful of iterators that have a `Item: 'static` despite the iterator having a lifetime or the `PartialEq` implementations where the flipped one cannot elide the lifetime.
Cosmetic improvements to doc comments
This has been factored out from https://github.com/rust-lang/rust/pull/58036 to only include changes to documentation comments (throughout the rustc codebase).
r? @steveklabnik
Once you're happy with this, maybe we could get it through with r=1, so it doesn't constantly get invalidated? (I'm not sure this will be an issue, but just in case...) Anyway, thanks for your advice so far!
Corentin Henry