Skip to content

Conversation

@tellison
Copy link
Member

IBM's Java VM doesn't have the concept of a permgen, so this option shouldn't be passed when the vendor property shows it is an IBM JDK.

@AmplabJenkins
Copy link

Can one of the admins verify this patch?

@srowen
Copy link
Member

srowen commented May 11, 2015

Does it actually fail to start if it sees this option?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Minor: This doesn't have to have the elses and thus extra level or two of nesting.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ACK

@srowen
Copy link
Member

srowen commented May 11, 2015

Jenkins, add to whitelist

@AmplabJenkins
Copy link

Merged build triggered.

@AmplabJenkins
Copy link

Merged build started.

@SparkQA
Copy link

SparkQA commented May 11, 2015

Test build #32398 has started for PR 6055 at commit 42a8c3f.

@tellison
Copy link
Member Author

The IBM VM will still start if passed -XX options that have no impact, so this is just being a good citizen in case that position changes in the future.

There are other IBM JVM changes I have that are a bit more significant. Let me know if you want an umbrella JIRA to cover running on IBM Java.

@srowen
Copy link
Member

srowen commented May 11, 2015

My only concern is complicating the code for args that don't actually cause a problem; I imagine any JVM should often be prepared to ignore -XX args it doesn't know since they are such a free-for-all. Still, not much harm in this particular change. But is there a lot more you have in mind?

@tellison
Copy link
Member Author

Not for -XX args. I have some changes to SizeEstimator, endianness, etc. for correctness, and thinking of extending java-opts to allow for vendor specific args. We can discuss on dev list and/or JIRA.

@SparkQA
Copy link

SparkQA commented May 11, 2015

Test build #32398 has finished for PR 6055 at commit 42a8c3f.

  • This patch passes all tests.
  • This patch merges cleanly.
  • This patch adds no public classes.

@AmplabJenkins
Copy link

Merged build finished. Test PASSed.

@AmplabJenkins
Copy link

Test PASSed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/32398/
Test PASSed.

@AmplabJenkins
Copy link

Merged build triggered.

@AmplabJenkins
Copy link

Merged build started.

@SparkQA
Copy link

SparkQA commented May 12, 2015

Test build #32497 has started for PR 6055 at commit d27174b.

@SparkQA
Copy link

SparkQA commented May 12, 2015

Test build #32497 has finished for PR 6055 at commit d27174b.

  • This patch passes all tests.
  • This patch merges cleanly.
  • This patch adds no public classes.

@AmplabJenkins
Copy link

Merged build finished. Test PASSed.

@AmplabJenkins
Copy link

Test PASSed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/32497/
Test PASSed.

@srowen
Copy link
Member

srowen commented May 12, 2015

Looks OK to me, excepting the minor simplification of those else branches.

@AmplabJenkins
Copy link

Merged build triggered.

@AmplabJenkins
Copy link

Merged build started.

@SparkQA
Copy link

SparkQA commented May 13, 2015

Test build #32611 has started for PR 6055 at commit 6ad4266.

@AmplabJenkins
Copy link

Merged build triggered.

@AmplabJenkins
Copy link

Merged build started.

@SparkQA
Copy link

SparkQA commented May 13, 2015

Test build #32612 has started for PR 6055 at commit 3a0fb66.

@SparkQA
Copy link

SparkQA commented May 13, 2015

Test build #32612 has finished for PR 6055 at commit 3a0fb66.

  • This patch fails Spark unit tests.
  • This patch merges cleanly.
  • This patch adds no public classes.

@AmplabJenkins
Copy link

Merged build finished. Test FAILed.

@AmplabJenkins
Copy link

Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/32612/
Test FAILed.

@SparkQA
Copy link

SparkQA commented May 13, 2015

Test build #32611 has finished for PR 6055 at commit 6ad4266.

  • This patch passes all tests.
  • This patch merges cleanly.
  • This patch adds no public classes.

@AmplabJenkins
Copy link

Merged build finished. Test PASSed.

@AmplabJenkins
Copy link

Test PASSed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/32611/
Test PASSed.

@asfgit asfgit closed this in e676fc0 May 13, 2015
jeanlyn pushed a commit to jeanlyn/spark that referenced this pull request May 28, 2015
IBM's Java VM doesn't have the concept of a permgen, so this option shouldn't be passed when the vendor property shows it is an IBM JDK.

Author: Tim Ellison <[email protected]>
Author: Tim Ellison <[email protected]>

Closes apache#6055 from tellison/MaxPermSize and squashes the following commits:

3a0fb66 [Tim Ellison] Convert tabs back to spaces
6ad4266 [Tim Ellison] Remove unnecessary else clauses to reduce nesting.
d27174b [Tim Ellison] Merge branch 'master' of https://github.com/apache/spark into MaxPermSize
42a8c3f [Tim Ellison] [MINOR] Avoid passing the PermGenSize option to IBM JVMs.
jeanlyn pushed a commit to jeanlyn/spark that referenced this pull request Jun 12, 2015
IBM's Java VM doesn't have the concept of a permgen, so this option shouldn't be passed when the vendor property shows it is an IBM JDK.

Author: Tim Ellison <[email protected]>
Author: Tim Ellison <[email protected]>

Closes apache#6055 from tellison/MaxPermSize and squashes the following commits:

3a0fb66 [Tim Ellison] Convert tabs back to spaces
6ad4266 [Tim Ellison] Remove unnecessary else clauses to reduce nesting.
d27174b [Tim Ellison] Merge branch 'master' of https://github.com/apache/spark into MaxPermSize
42a8c3f [Tim Ellison] [MINOR] Avoid passing the PermGenSize option to IBM JVMs.
nemccarthy pushed a commit to nemccarthy/spark that referenced this pull request Jun 19, 2015
IBM's Java VM doesn't have the concept of a permgen, so this option shouldn't be passed when the vendor property shows it is an IBM JDK.

Author: Tim Ellison <[email protected]>
Author: Tim Ellison <[email protected]>

Closes apache#6055 from tellison/MaxPermSize and squashes the following commits:

3a0fb66 [Tim Ellison] Convert tabs back to spaces
6ad4266 [Tim Ellison] Remove unnecessary else clauses to reduce nesting.
d27174b [Tim Ellison] Merge branch 'master' of https://github.com/apache/spark into MaxPermSize
42a8c3f [Tim Ellison] [MINOR] Avoid passing the PermGenSize option to IBM JVMs.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants