Sherman, Regarding micro benchmark, did you ever get together with monica from performance team so that she can start building suites of micro benchmarks? I think we need to start doing this so that they help(take over) performance validation in the future. Of course developers should still do it before checkin but benchmarks need to be run regularly.
Thanks. -----Original Message----- From: Xueming Shen <[email protected]> Sender: [email protected] Date: Fri, 01 Apr 2011 16:04:35 To: Dave Bristor<[email protected]>; BATEMAN, ALAN<[email protected]>; core-libs-dev<[email protected]> Subject: Review Request for 6751338: ZIP inflater/deflater performance Dave, Alan, Here is the final webrev based on Dave's patch and the jdk1.5 code that does not have the change for 6206933. JPRT job result suggests no new testing failure and my "non-scientific" benchmark test (to use GZIPOu/InputStream to compress/ decompress the rt.jar) does show relative performance gain. Will try to run more tests the weekend, but here is the webrev. http://cr.openjdk.java.net/~sherman/6751338/webrev/ Background Info: This fix is basically to back out the fix for #6206933 we made back to jdk5, which is to use malloc+GetByteArrayuRegion to replace the original GetPrimitiveArrayCritical/ ReleasePrimitiveArrayCritical() pair when access the java byte array at native code Inflater/Deflater.c, to mainly workaround the GC/Critical... issue discussed in #6186200. The change for #6206933 itself has triggered lots of performance issues since its integration, some fixed, some still outstanding. The GC rfe#6186200 has been fixed long time ago, after couple weeks of discussion/debating, we all agreed that it's the time to back out#6206933. -Sherman
