RVM
RVM-182

invalid frame address

Details

Type: Bug
Status: Closed
Priority: Major
Resolution: Cannot Reproduce
Affects Version/s: 2.9.0.1
Fix Version/s: None
Component/s: None
Labels:
None
Environment:
rvmx86lnx32.anu.edu.au

Number of attachments :
0

Description

This error started appearing in dacapo after the change to new thread model. See below for examples. It appeared intermittently prior to that.

To find the other tests where this also happened you can execute the following SQL on the cattrack database.
SELECT
('http://jikesrvm.anu.edu.au/cattrack/results/' || hosts.name || '/' || test_runs.variant || '.' || test_runs.id || '/' || build_configurations.name || '/' || test_configurations.name || '/' || groups.name || '/' || test_cases.name || '/' || test_case_executions.name ||'/Output.txt') AS uri,
test_runs.start_time
FROM test_runs
RIGHT JOIN hosts ON hosts.id = test_runs.host_id
RIGHT JOIN build_configurations ON build_configurations.test_run_id = test_runs.id
RIGHT JOIN test_configurations ON test_configurations.build_configuration_id = build_configurations.id
RIGHT JOIN groups ON groups.test_configuration_id = test_configurations.id
RIGHT JOIN test_cases ON test_cases.group_id = groups.id
RIGHT JOIN test_case_executions ON test_case_executions.test_case_id = test_cases.id
WHERE
test_case_executions.id IN (SELECT owner_id FROM test_case_execution_outputs WHERE output LIKE '%invalid frame address%')
ORDER BY test_runs.start_time DESC

Sample output of running SQL

Activity

Ascending order - Click to sort in descending order

Hide

Permalink

Ian Rogers added a comment - 11/Sep/07 8:43 AM

Is this still an issue?

Show

Ian Rogers added a comment - 11/Sep/07 8:43 AM Is this still an issue?

Hide

Permalink

Peter Donald added a comment - 11/Sep/07 7:34 PM

It still occurred as of two days ago so it looks like it is still an issue

Show

Peter Donald added a comment - 11/Sep/07 7:34 PM It still occurred as of two days ago so it looks like it is still an issue

Hide

Permalink

Ian Rogers added a comment - 12/Sep/07 3:06 AM

We seem to have had a run of bad MarkCompact regression tests, and RC is known to have problems. Of the other regressions we seem to get the bad frame address after some other catastrophe has occurred. Hopefully this will work itself out. I'm not sure if we can do a particular item to get it fixed for the 2.9.2 release.

Show

Ian Rogers added a comment - 12/Sep/07 3:06 AM We seem to have had a run of bad MarkCompact regression tests, and RC is known to have problems. Of the other regressions we seem to get the bad frame address after some other catastrophe has occurred. Hopefully this will work itself out. I'm not sure if we can do a particular item to get it fixed for the 2.9.2 release.

Hide

Permalink

Ian Rogers added a comment - 18/Apr/08 7:00 PM

Without cattrack searches I can't see if this is still an issue. Moving to 2.9.4 to not impede the 2.9.3 release.

Show

Ian Rogers added a comment - 18/Apr/08 7:00 PM Without cattrack searches I can't see if this is still an issue. Moving to 2.9.4 to not impede the 2.9.3 release.

Hide

Permalink

Ian Rogers added a comment - 27/Jun/08 1:29 PM

I suspect this issue is dead but still cannot search cattrack.

Show

Ian Rogers added a comment - 27/Jun/08 1:29 PM I suspect this issue is dead but still cannot search cattrack.

Hide

Permalink

David Grove added a comment - 10/Jul/08 7:53 AM

I believe this is obsolete.

Show

David Grove added a comment - 10/Jul/08 7:53 AM I believe this is obsolete.

People

Assignee:

Unassigned

Reporter:

Peter Donald

Vote (0)

Watch (0)

Dates

Created:

20/Aug/07 8:42 PM

Updated:

10/Jul/08 7:53 AM

Resolved:

10/Jul/08 7:53 AM