Auto merge of #136367 - workingjubilee:rollup-pud5uj0, r=workingjubilee

Rollup of 16 pull requests

Successful merges:

 - #133266 (ci: fix explanation why LLVM download is disabled for windows-gnu)
 - #135768 (tests: Port `symbol-mangling-hashed` to rmake.rs)
 - #135836 (bootstrap: only build `crt{begin,end}.o` when compiling to MUSL)
 - #135840 (omit unused args warnings for intrinsics without body)
 - #135900 (Manually walk into WF obligations in `BestObligation` proof tree visitor)
 - #136146 (Explicitly choose x86 softfloat/hardfloat ABI)
 - #136154 (Use +secure-plt for powerpc-unknown-linux-gnu{,spe})
 - #136163 (Fix off-by-one error causing slice::sort to abort the program)
 - #136266 (fix broken release notes id)
 - #136283 (Update encode_utf16 to mention it is native endian)
 - #136309 (set rustc dylib on manually constructed rustc command)
 - #136314 (Use proper type when applying deref adjustment in const)
 - #136339 (CompileTest: Add Directives to Ignore `arm-unknown-*` Targets)
 - #136348 (miri: make float min/max non-deterministic)
 - #136351 (Add documentation for derive(CoercePointee))
 - #136358 (`#[optimize(none)]` implies `#[inline(never)]`)

Failed merges:

 - #135994 (Rename rustc_middle::Ty::is_unsafe_ptr to is_raw_ptr)

r? `@ghost`
`@rustbot` modify labels: rollup

Author: bors

RELEASES.md

@@ -16,8 +16,7 @@ Version 1.84.1 (2025-01-30)
 Version 1.84.0 (2025-01-09)
 ==========================
 
-<a id="
-Language"></a>
+<a id="1.84.0-Language"></a>
 
 Language
 --------

compiler/rustc_const_eval/src/interpret/intrinsics.rs

@@ -747,7 +747,13 @@ impl<'tcx, M: Machine<'tcx>> InterpCx<'tcx, M> {
     {
         let a: F = self.read_scalar(&args[0])?.to_float()?;
         let b: F = self.read_scalar(&args[1])?.to_float()?;
-        let res = self.adjust_nan(a.min(b), &[a, b]);
+        let res = if a == b {
+            // They are definitely not NaN (those are never equal), but they could be `+0` and `-0`.
+            // Let the machine decide which one to return.
+            M::equal_float_min_max(self, a, b)
+        } else {
+            self.adjust_nan(a.min(b), &[a, b])
+        };
         self.write_scalar(res, dest)?;
         interp_ok(())
     }
@@ -762,7 +768,13 @@ impl<'tcx, M: Machine<'tcx>> InterpCx<'tcx, M> {
     {
         let a: F = self.read_scalar(&args[0])?.to_float()?;
         let b: F = self.read_scalar(&args[1])?.to_float()?;
-        let res = self.adjust_nan(a.max(b), &[a, b]);
+        let res = if a == b {
+            // They are definitely not NaN (those are never equal), but they could be `+0` and `-0`.
+            // Let the machine decide which one to return.
+            M::equal_float_min_max(self, a, b)
+        } else {
+            self.adjust_nan(a.max(b), &[a, b])
+        };
         self.write_scalar(res, dest)?;
         interp_ok(())
     }