Tue, 21 Mar 2017 16:03:37 +0100*** empty log message *** expecco_head_5844
sr [Tue, 21 Mar 2017 16:03:37 +0100] rev 330
*** empty log message ***

Sat, 18 Mar 2017 04:55:07 +0000update tags
convert-repo [Sat, 18 Mar 2017 04:55:07 +0000] rev 329
update tags

Fri, 17 Mar 2017 15:12:41 +0100*** empty log message ***
sr [Fri, 17 Mar 2017 15:12:41 +0100] rev 328
*** empty log message ***

Sat, 11 Mar 2017 21:41:58 +0000Workaround for bug (?) in ProcessorScheduler jv
Jan Vrany <jan.vrany@fit.cvut.cz> [Sat, 11 Mar 2017 21:41:58 +0000] rev 327
Workaround for bug (?) in ProcessorScheduler

"Due to some bug (?) in ProcessorScheduler, we need to set
exitWhenNoMoreUserProcesses to prevent process dispatcher
to exit prematurely.

Sometimes, ProcessorScheduler>>checkForEndOfSispatch thought
there's no runnable user proless even though test runner did not
finish (was merely waiting for external command to finish).
Happened sometimes. not always.

Thu, 23 Feb 2017 13:36:24 +0000Test report: print test execution time in result summary jv
Jan Vrany <jan.vrany@fit.cvut.cz> [Thu, 23 Feb 2017 13:36:24 +0000] rev 326
Test report: print test execution time in result summary

Mon, 13 Feb 2017 12:00:51 +0100Workaround for (presumably) race in TestReport jv
Jan Vrany <jan.vrany@fit.cvut.cz> [Mon, 13 Feb 2017 12:00:51 +0100] rev 325
Workaround for (presumably) race in TestReport

On SWING CI, sometimes when test times out, the whole
test run crashes on DNU as `outcome` instvar is nil.
This is maybe because of some race.

As a quick and dirty workaround, create a new one if
outcome is nil. Proper solution is to revise the code
and figure out what's wrong, indeed. This is left as
future work.

Mon, 05 Dec 2016 06:48:09 +0100Merge jv
Merge Script [Mon, 05 Dec 2016 06:48:09 +0100] rev 323
Merge

Mon, 14 Nov 2016 23:43:14 +0000Autoscale testcase-provided timeout to compensate for slooow machines jv
Jan Vrany <jan.vrany@fit.cvut.cz> [Mon, 14 Nov 2016 23:43:14 +0000] rev 322
Autoscale testcase-provided timeout to compensate for slooow machines

Each test case has a timeout to guard against runaway tests. However
on really slow machines the timeout us not big enough. To compensate for this,
asses the "speed" of machine running tests and scale default timeout
if machine is slower than some (arbitrary) norm.

The speed assesment is done by measuring time to run (arbitrary) benchmark
code. This has the advantage to reflect actual machine load, not only
hardvare spec.

However, we may need to play with these magic numbers to make it working.
Generally a workaround.

Wed, 09 Nov 2016 04:12:36 +0100*** empty log message *** expecco_2_10_0 expecco_2_10_0_41 expecco_2_11_0 expecco_ALM_1_10_0 expecco_ALM_1_10_0_8 expecco_ALM_1_11_0 expecco_ALM_1_11_0_2 expecco_ALM_1_11_2
Claus Gittinger <cg@exept.de> [Wed, 09 Nov 2016 04:12:36 +0100] rev 321
*** empty log message ***

Wed, 17 Aug 2016 03:50:58 +0000update tags
convert-repo [Wed, 17 Aug 2016 03:50:58 +0000] rev 320
update tags