On Tue, 21 Apr 2026 06:14:20 GMT, Stuart Marks <[email protected]> wrote:

>> jengebr has updated the pull request incrementally with one additional 
>> commit since the last revision:
>> 
>>   PR feedback - types in putHashMapEntries
>
> test/micro/org/openjdk/bench/java/util/HashMapConstructorBenchmark.java line 
> 120:
> 
>> 118:             HashMap<BigInteger, Integer> temp = new HashMap<>(source);
>> 119:             if (temp.size() != mapSize)
>> 120:                 throw new RuntimeException();
> 
> Interesting. It looks like test code here... but why? After pondering this a 
> bit I imagine that simply storing the new HashMap into a local and then 
> overwriting it on the next loop might result in the JIT compiler eliminating 
> a bunch of dead code, so adding the if-statement and the size check might 
> defeat this. But I'm kind of guessing here. A comment would be useful; or, 
> perhaps there's  away to prevent the compiler from doing this? Use a 
> blackhole? Kind of strange to do this in setup code, but maybe it's 
> appropriate.

Yes, it's effectively a blackhole.  I'll address via either comment or JMH 
Blackhole.

-------------

PR Review Comment: https://git.openjdk.org/jdk/pull/28243#discussion_r3117575976

Reply via email to