JDK-8069020 : [TESTBUG] New stress tests takes too long to run in JPRT
  • Type: Bug
  • Component: hotspot
  • Sub-Component: compiler
  • Affected Version: 9
  • Priority: P1
  • Status: Closed
  • Resolution: Not an Issue
  • OS: generic
  • CPU: generic
  • Submitted: 2015-01-14
  • Updated: 2015-02-19
  • Resolved: 2015-02-19
The Version table provides details related to the release that this issue/RFE will be addressed.

Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.

To download the current JDK release, click here.
JDK 9
9Resolved
Related Reports
Relates :  
JDK-8059551 - JEP-JDK-8043304: Test task: stress tests
Relates :  
JDK-8069021 - Exclude compiler/codecache/stress tests from JPRT runs
Description
After 8059551 push hotspot_compiler_2 tests run in JPRT takes very long time (instead of 10-15min):

hotspot$ grep compiler_2 JobStatus.txt
  linux_i586_2.6-fastdebug-c1-hotspot_compiler_2     success(57m 22s)
        STATS:    name=hotspot_compiler_2 run=101 pass=101 fail=0
  linux_i586_2.6-fastdebug-c2-hotspot_compiler_2     success(54m 46s)
        STATS:    name=hotspot_compiler_2 run=101 pass=101 fail=0
  linux_x64_2.6-fastdebug-c2-hotspot_compiler_2     success(54m 38s)
        STATS:    name=hotspot_compiler_2 run=101 pass=101 fail=0
  macosx_x64_10.7-fastdebug-c2-hotspot_compiler_2     success(01H 04m 19s)
        STATS:    name=hotspot_compiler_2 run=101 pass=101 fail=0
  solaris_sparcv9_5.11-fastdebug-c2-hotspot_compiler_2     running(03m 04s)
  solaris_x64_5.11-fastdebug-c2-hotspot_compiler_2     success(57m 28s)
        STATS:    name=hotspot_compiler_2 run=102 pass=102 fail=0
  windows_i586_6.1-fastdebug-c1-hotspot_compiler_2     success(01H 01m 36s)
        STATS:    name=hotspot_compiler_2 run=101 pass=101 fail=0
  windows_i586_6.1-fastdebug-c2-hotspot_compiler_2     success(01H 00m 26s)
        STATS:    name=hotspot_compiler_2 run=101 pass=101 fail=0
  windows_x64_6.1-fastdebug-c2-hotspot_compiler_2     success(58m 20s)
        STATS:    name=hotspot_compiler_2 run=101 pass=101 fail=0
Comments
This tests are timed tests. Test calculate overall execution time as following: Execution time = (JTREG_DEFAULT_TIMEOUT * TIMEOUT_FACTOR) * 0.8 This time is used for each @run in test. So for big jtreg timeout and big timeout factor (can be customized w/ options in nightly template, in JPRT runs and etc) tests can take expected a lot of time.
19-02-2015
I am removing these test from JPRT runs (8069021) but run time of tests should be fixed.
14-01-2015
-------------------------------------------------- TEST: compiler/codecache/stress/OverloadCompileQueueTest.java build: 3.403 seconds compile: 3.403 seconds build: 0.0 seconds main: 0.833 seconds build: 0.0 seconds main: 440.027 seconds build: 0.0 seconds main: 440.562 seconds TEST RESULT: Passed. Execution successful -------------------------------------------------- TEST: compiler/codecache/stress/RandomAllocationTest.java build: 3.126 seconds compile: 3.126 seconds build: 0.001 seconds main: 0.833 seconds build: 0.0 seconds main: 432.012 seconds build: 0.0 seconds main: 432.179 seconds TEST RESULT: Passed. Execution successful -------------------------------------------------- TEST: compiler/codecache/stress/UnexpectedDeoptimizationTest.java build: 2.841 seconds compile: 2.841 seconds build: 0.0 seconds main: 0.851 seconds build: 0.0 seconds main: 431.802 seconds build: 0.0 seconds main: 431.869 seconds TEST RESULT: Passed. Execution successful
14-01-2015