[ https://issues.apache.org/jira/browse/HARMONY-4705?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12524146 ] 

Aleksey Shipilev commented on HARMONY-4705:
-------------------------------------------

On the Dacapo benchmark attached patch gives boosts on jython (~10%), antlr (~7%), pmd(~5%) @ 2xWoodcrest, Windows x86.

> [drlvm][jni] JNI transition uses ineffective TLS get
> ----------------------------------------------------
>
>                 Key: HARMONY-4705
>                 URL: https://issues.apache.org/jira/browse/HARMONY-4705
>             Project: Harmony
>          Issue Type: Improvement
>          Components: DRLVM
>            Reporter: Aleksey Shipilev
>         Attachments: JNI-tls-1.patch
>
>
> For now, JNI transition uses ineffective scheme of TLS getting - it is done via several recursive calls, finally reaching hythread_self().
> The idea is to inline helper for TLS get instead of calling get_thread_ptr_stub(). Disassembly shows that 10 push/pops and 3 calls are completely eliminated there, keeping in mind that TLS get occurs at least 3 times on each JNI transition this give us a huge boost, eliminating up to 30 push/pops and 9 calls.
> JNI microtest (calling several millions of empty JNI methods) testifies this:
> $ ../../Builds/Harmony-clean/bin/java -cp . -Xmx128m -Xms128m nalog.nalog
> iteration: 0  millis:8763
> iteration: 1  millis:8722
> iteration: 2  millis:8833
> $ ../../Builds/Harmony-fastTLS-1/bin/java -cp . -Xmx128m -Xms128m nalog.nalog
> iteration: 0  millis:6780
> iteration: 1  millis:6689
> iteration: 2  millis:6870

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.