RVM

Increase time out on DaCapo eclipse for rvmppclnx64

Details

  • Number of attachments :
    0

Description

We're consistently failing DaCapo eclipse on rvmppclnx64 due to timeouts. The traces [1][2] look ok so we should give the machine more time to finish the test.

[1] http://jikesrvm.anu.edu.au/cattrack/results/rvmppclnx64.anu.edu.au/commit/3813/prototype/default/dacapo/eclipse
[2] http://jikesrvm.anu.edu.au/cattrack/results/rvmppclnx64.anu.edu.au/commit/3802/prototype/default/dacapo/eclipse

Activity

Ascending order - Click to sort in descending order
Hide
Permalink
David Grove added a comment -

I'm first going to try jacking the timelimit up a little bit. If that isn't sufficient, I'm considering trying the run with the smallest possible input on the slow machines for bloat and eclipse. Some kind of test that completes is better than running something that is always going to timeout...

Show
David Grove added a comment - I'm first going to try jacking the timelimit up a little bit. If that isn't sufficient, I'm considering trying the run with the smallest possible input on the slow machines for bloat and eclipse. Some kind of test that completes is better than running something that is always going to timeout...
David Grove made changes -
Field Original Value New Value
Assignee David Grove [ dgrove ]
Hide
Permalink
David Grove added a comment -

On a hunch, I profiled bloat running on a prototype-opt image (with aos disabled) on ppc.

We spend 26% of total time in VM_DynamicTypeCheck.instanceOfResolved and instanceOfInterface. I'm contemplating implementing inline versions of the simplest cases of instanceof/checkcast for the baseline compiler. An additional 10% is going to typechecks for interface invocation (mustImplementInterface).

bloat completes in about 1200 seconds when run manually.

Show
David Grove added a comment - On a hunch, I profiled bloat running on a prototype-opt image (with aos disabled) on ppc. We spend 26% of total time in VM_DynamicTypeCheck.instanceOfResolved and instanceOfInterface. I'm contemplating implementing inline versions of the simplest cases of instanceof/checkcast for the baseline compiler. An additional 10% is going to typechecks for interface invocation (mustImplementInterface). bloat completes in about 1200 seconds when run manually.
David Grove made changes -
Status Open [ 1 ] In Progress [ 3 ]
David Grove made changes -
Fix Version/s 3.0 [ 13530 ]
Fix Version/s 2.9.4 [ 14162 ]
Hide
Permalink
David Grove added a comment -

increased timeout to 1200 seconds in r14633.

Show
David Grove added a comment - increased timeout to 1200 seconds in r14633.
Hide
Permalink
David Grove added a comment -

I believe this is fixed (eclipse is now running to completion on excalibur), but can't close until we fix the build problems with the prototype image on ppc32-linux and get a confirming test result.

Show
David Grove added a comment - I believe this is fixed (eclipse is now running to completion on excalibur), but can't close until we fix the build problems with the prototype image on ppc32-linux and get a confirming test result.
David Grove made changes -
Status In Progress [ 3 ] Resolved [ 5 ]
Resolution Fixed [ 1 ]
Ian Rogers made changes -
Status Resolved [ 5 ] Closed [ 6 ]

People

  • Assignee:
    David Grove
    Reporter:
    Ian Rogers
Vote (0)
Watch (0)

Dates

  • Created:
    Updated:
    Resolved: