StringIndexOfChar showed that the assumption that the char variant
should be at least as fast as the String variant doesn't always apply.
The implementation for indexOf(char) seems to have a different code
emission heuristics than the String variant.
added benchmark results for OpenJDK's StringIndexOf benchmark:
https://github.com/openjdk/jdk/pull/6509#issuecomment-979985594
-michael
On 25.11.21 15:05, Michael Bien wrote:
Hello again,
I was trying to run JDK's benchmarks over night (second attempt
actually) but had some difficulties to
Hello again,
I was trying to run JDK's benchmarks over night (second attempt
actually) but had some difficulties to get stable results.
This makes it difficult to compare the modified version with a
reference. I am not sure what the cause is, I have heard some intel CPUs
can't run avx
On 23.11.21 15:57, Roger Riggs wrote:
Hi Michael,
As you might expect performance of strings is very sensitive and has
been tuned extensively over the years many times.
Though this improves the performance for 1 character strings. It will
have an impact on *every other* length of string.
Hi Michael,
As you might expect performance of strings is very sensitive and has
been tuned extensively over the years many times.
Though this improves the performance for 1 character strings. It will
have an impact on *every other* length of string.
You'll need to show that it does not
Hello,
I kept forgetting which variants of the String methods perform better
with single-char-Strings and which with char (IDEs had the tendency to
suggest the wrong variant since it changed between JDK releases). So i
wrote JMH benchmarks and noticed that the last method with a performance