https://bz.apache.org/bugzilla/show_bug.cgi?id=63916
--- Comment #36 from Mark Thomas <ma...@apache.org> --- That is the right code. Testing locally on Linux: java -showversion org.apache.tomcat.Bug63916NioPoller openjdk version "1.8.0_222" OpenJDK Runtime Environment (AdoptOpenJDK)(build 1.8.0_222-b10) OpenJDK 64-Bit Server VM (AdoptOpenJDK)(build 25.222-b10, mixed mode) Selector provider: sun.nio.ch.EPollSelectorProvider@4cd646ad Default send buffer size is [1313280] Writing 10MB took [259] milliseconds And with specific buffer sizes 16k 379ms 8k 402ms 4k 56,091ms On MacOS s2laptop:classes mthomas$ java -version openjdk version "1.8.0_212" OpenJDK Runtime Environment (AdoptOpenJDK)(build 1.8.0_212-b03) OpenJDK 64-Bit Server VM (AdoptOpenJDK)(build 25.212-b03, mixed mode) s2laptop:classes mthomas$ java org.apache.tomcat.Bug63916NioPoller Selector provider: sun.nio.ch.KQueueSelectorProvider@227974ab Default send buffer size is [146988] Writing 10MB took [364] milliseconds And with specific buffer sizes 16k 332ms 8k 309ms 4k 319ms 2k 356ms 1k 20,041ms Both Oracle Java 7 and OpenJDK 13 give more reasonable results of around 440 ms. There looks to be an issue with the JVM. Given how bad these figures are I agree with Michael's suggestion to add something to the docs. I'll also open a JRE bug. -- You are receiving this mail because: You are the assignee for the bug. --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
