[Boost-testing] Re: Toolset naming and regressions

From: Aleksey Gurtovoy (agurtovoy_at_[hidden])
Date: 2005-06-06 16:21:25

Rene Rivera writes:
> Stephen W. Carson wrote:
>> Rene Rivera wrote:
>> 
>>> Jeff Garland wrote:
>>>
>>>> So...we really need to standardize toolset naming for regression
>>>> testers. 
>>>
>>> It is standardized as: [toolset]-[version]-[platform]-[cpu]-[build]
>>    First time I've seen that. I think my toolsets are named fine
>> because I did what I was told to do.
>
> Don't remember when the discussion about toolset names happened. But
> it was likely when the first XSLT results came online.
>
>>    Which toolsets are marked as required for release? How do I know?
>
> Currently the list is:
>
> borland-5_6_4
> cw-9_4
> cw-8_3
> cw-9_3-darwin
> cw-9_4-darwin
> cw-9_5-darwin
> msvc
> msvc-stlport
> vc7
> vc-7_1
> vc-8_0
> mingw-3_4_2
> mingw-3_3_1
> gcc-2.95.3-linux
> gcc-2.95.3-stlport-4.5.3-linux
> gcc-3.3.5-linux
> gcc-3.4.3-linux
> gcc-4.0.0-linux
> gcc-3_3-darwin
> iw-7_1-vc6
> iw-7_1-vc6-stlp-4_5_3
> intel-win32-8_0
> intel-8.1-linux
>
> The list is at the top of boost-root/status/explicit-failures-markup.xml
>
> Hmm... I guess we should remove cw-9_[34]-darwin :-)
>
> And remove iw-*.

Please feel free to.

>
> Any chance the metacomm tests could be switched from "vc7" to "vc-7_0"?

Done, should be reflected in the reports early tomorrow.

>
>>    You wrote (in a different message) that all a toolset requires is
>> a commitment from one person to support the platform. Does that mean
>> running the regressions faithfully as I've been doing for gcc and cw
>> on the Mac? Or more than that?
>
> More than that. Obviously a release toolset needs to be consistently
> tested. But it also requires someone who is willing to debug the
> tests. Testing is awesome, but if there's no one to fix the problems
> the toolset will always show up with errors. Ultimately though it's up
> to the release manager.

Yep.

-- 
Aleksey Gurtovoy
MetaCommunications Engineering