On Mon, 18 May 2026 20:28:43 GMT, Joe Darcy <[email protected]> wrote:
>> Prototype a fully functional `unsigned int128` type as a Valhalla value >> class. >> >> --------- >> - [x] I confirm that I make this contribution in accordance with the >> [OpenJDK Interim AI Policy](https://openjdk.org/legal/ai). > > src/java.base/share/classes/java/math/UInt128.java line 178: > >> 176: * Returns the instance representing {@code v}. >> 177: * >> 178: * @param v must meet 0 ≤ {@code v} < 2<sup>{@link #SIZE}</sup>. > > As a small issue, I don't know if having "<=" as a single character, as it > seems to be here, goes against general encoding rules for JDK sources. JDK source code has [switched to UTF-8](https://bugs.openjdk.org/browse/JDK-8301971) some time ago. > src/java.base/share/classes/java/math/UInt64.java line 186: > >> 184: * @return the result described above. >> 185: */ >> 186: static boolean lessThan(long x, long y) { > > Is it intentional that this method is _not_ public? > > Having a `@see` link to `Long.compareUnsigned`, if the obvious relationship > between the methods holds, would be helpful. The idea is that the `public` API point is `compareTo()`. But it's no deal to make the `static` comparison methods `public` as well. ------------- PR Review Comment: https://git.openjdk.org/valhalla/pull/2420#discussion_r3262410613 PR Review Comment: https://git.openjdk.org/valhalla/pull/2420#discussion_r3262421042
