Hi,

On 13.04.2015, at 13:29, Frank Shearar <frank.shearar at gmail.com> wrote:

> On 10 April 2015 at 16:21, Tobias Pape <Das.Linux at gmx.de> wrote:
>> Hi Eliot,
>> 
>> On 10.04.2015, at 15:46, Eliot Miranda <eliot.miranda at gmail.com> wrote:
>> 
>>> 
>>> 
>>> 
>>> On Apr 9, 2015, at 3:21 PM, Tobias Pape <Das.Linux at gmx.de> wrote:
>>> 
>>>> 
>>>> On 10.04.2015, at 00:17, Tobias Pape <Das.Linux at gmx.de> wrote:
>>>> 
>>>>> Updated on the CI :)
>>>> 
>>>> aaand reverted. We got a crash:
>>>>  http://build.squeak.org/job/SqueakTrunk/1245/console
>>> 
>> 
>> 
>>> Yes.  I get a crash in updating with the v3 VM.  So far no problems with the Spur VM.  But Tibias, unless the bleeding edge gets tested how will I spit these regressions?  I need to know.  So can we not have two builds, one with the last known good and one with the bleeding edge?
>> 
>> Sure, this is possible.
>> probably something like a matrix build.
>> Frank is more savvy with the infrastructure, tho.
> 
> I'd say that SqueakTrunk should use the latest stable Cog - its main
> purpose is to test _trunk's_ bleeding edge. We do also have
> SqueakTrunkOnSpur, which should track Spur's/Cog's bleeding edge.
> 

Ack :)

> Which means we should split the versions in the build scripts. Maybe
> we should just have a COG_VERSION and a SPUR_COG_VERSION as a
> not-too-stinky quick win?


Probably… I don't know yet. Have to make up my mind :)

> 
> frank
> 
>> best
>>        -Tobias
>> 
>> 
>>> Otherwise I'm flying blind.
>>> 
>>>> 
>>>> Best
>>>>  -Tobias
>>>> 
>>>>> Best
>>>>>  -Tobias
>>>>> On 09.04.2015, at 23:28, Eliot Miranda <eliot.miranda at gmail.com> wrote: