On Wed, 8 May 2024 08:19:54 GMT, Daniel Jeliński <djelin...@openjdk.org> wrote:
> Replace the custom unsigned divide / remainder functions with the > compiler-optimized Long.divideUnsigned / remainderUnsigned. > > No new tests. Existing tier1-3 tests continue to pass. > > I added a new set of divide benchmarks. Results in thread. > > I also removed the BigDecimal.toString benchmarks. They no longer serve their > purpose - toString caches the returned value, so we were only benchmarking > the cache access time. src/java.base/share/classes/java/math/BigDecimal.java line 5683: > 5681: tmp = (dividendHi << shift) | (dividendLo >>> 64 - shift); > 5682: long u2 = tmp & LONG_MASK; > 5683: long q1, r_tmp; Nit: There doesn't seem to be a reason for splitting the q1 and r_tmp declarations from their assignments, why not the following instead? (Also applies to the cases below) long q1 = Long.divideUnsigned(tmp, v1); long r_tmp = Long.remainderUnsigned(tmp, v1); ------------- PR Review Comment: https://git.openjdk.org/jdk/pull/19134#discussion_r1596137061
