Rollup merge of #129944 - Mark-Simulacrum:relnotes-tweak, r=pietroalbini

Add compat note for trait solver change

r? ``@pietroalbini`` ``@BoxyUwU``

cc ``@lcnr``

1 files changed, 3 insertions, 0 deletions

diff --git a/RELEASES.md b/RELEASES.md
index 29c872eb448..1b9bc62ba3a 100644
--- a/RELEASES.md
+++ b/RELEASES.md
@@ -100,6 +100,9 @@ Compatibility Notes
   The reason is that these types have different roles: `std::panic::PanicHookInfo` is the argument to the [panic hook](https://doc.rust-lang.org/stable/std/panic/fn.set_hook.html) in std context (where panics can have an arbitrary payload), while `core::panic::PanicInfo` is the argument to the [`#[panic_handler]`](https://doc.rust-lang.org/nomicon/panic-handler.html) in no_std context (where panics always carry a formatted *message*). Separating these types allows us to add more useful methods to these types, such as `std::panic::PanicHookInfo::payload_as_str()` and `core::panic::PanicInfo::message()`.
 
 * The new sort implementations may panic if a type's implementation of [`Ord`](https://doc.rust-lang.org/std/cmp/trait.Ord.html) (or the given comparison function) does not implement a [total order](https://en.wikipedia.org/wiki/Total_order) as the trait requires. `Ord`'s supertraits (`PartialOrd`, `Eq`, and `PartialEq`) must also be consistent. The previous implementations would not "notice" any problem, but the new implementations have a good chance of detecting inconsistencies, throwing a panic rather than returning knowingly unsorted data.
+* [In very rare cases, a change in the internal evaluation order of the trait
+  solver may result in new fatal overflow errors.](https://github.com/rust-lang/rust/pull/126128)
+
 
 <a id="1.81.0-Internal-Changes"></a>