This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.
Summary: | RequestProcessor and Thread.MIN_PRIORITY tasks on Linux | ||
---|---|---|---|
Product: | platform | Reporter: | _ pkuzel <pkuzel> |
Component: | -- Other -- | Assignee: | issues@platform <issues> |
Status: | RESOLVED WONTFIX | ||
Severity: | blocker | CC: | issues, jtulach, ttran |
Priority: | P2 | Keywords: | PERFORMANCE, THREAD |
Version: | 3.x | ||
Hardware: | PC | ||
OS: | Linux | ||
Issue Type: | TASK | Exception Reporter: |
Description
_ pkuzel
2004-04-29 17:06:36 UTC
Do we want to implement this for promo-D? I remember we planned to do so... There is no simple solution for NetBeans. One may want to try -XX:+UseThreadPriorities option mentioned in http://java.sun.com/docs/hotspot/VMOptions.html. Also there are option -XX:-UseSpinning and -XX:PreBlockSpin=10 Personally I see some improvement in percetion of operations like warmup on my Linux with fedora core 2 based on 2.6.5 kernel running on JDK1.5.0 builds. I do not expect that we will change anything in promo-D timeframe. not clear to me what this issue is about. We pass thread priority down to JVM, what more can we do? WONTFIX RP fix (or transitive JVM fix) was mentioned by performance team at time when NetBeans native code workarounding this problem was dropped. Just continue pushing on proper implementation and planing teams. Linux end users would really appreciate it. As we've discussed recently there is no plan to make any change in RP now. We agreed that it would be usefull to measure and compare behaviour on Linux 2.4/JDK1.4 vs. Linux 2.6/JDK1.5 and vs. Windows and with the nice() workaround. We've seen no problems caused by ignored thread priorities recently. |