[Buildroot] [autobuild.buildroot.net] Your build results for 2018-01-04

[Buildroot] [autobuild.buildroot.net] Your build results for 2018-01-04

[Thomas Petazzoni]

Hello,

On Fri, 5 Jan 2018 09:43:09 +0100, Kurt Van Dijck wrote:

> > Results for the 'master' branch
> > ===============================
> > 
> > Build failures related to your packages:
> > 
> >       xtensa |             nilfs-utils-v2.2.7 | http://autobuild.buildroot.net/results/a9fe41ed428d26ef5500cec9f762c0bae3eebd39  
> 
> I just ran this on my local branch, including the patches I sent
> yesterday (and are still pending), and that went ok, so I propose
> to wait and see until those 2 patches are accepted/denied.
> 
> Just a question.
> I don't understand why xtensa fails today and not yesterday ...
> Is there a random toolchain selection everyday for each package to save
> buildserver resources?

The autobuilders only do random testing. Essentially, what they do is:

 - Pick a random architecture/toolchain configuration among the one
   available in support/config-fragments/autobuild/

 - Run "make randpackageconfig" to generate on top of the
   architecture/toolchain configuration a random selection of packages.

 - Build

So the results every day are different, since it's completely random
what testing is being done. This allows us to test a very wide variety
of package/toolchain/architecture combinations, since doing an
exhaustive testing is simply impossible when you have thousands of
Config.in options.

Best regards,

Thomas
-- 
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com

[Buildroot] [autobuild.buildroot.net] Your build results for 2018-01-04

[Kurt Van Dijck]

> Hello,
> 
> On Fri, 5 Jan 2018 09:43:09 +0100, Kurt Van Dijck wrote:
> 
> > > Results for the 'master' branch
> > > ===============================
> > > 
> > > Build failures related to your packages:
> > > 
> > >       xtensa |             nilfs-utils-v2.2.7 | http://autobuild.buildroot.net/results/a9fe41ed428d26ef5500cec9f762c0bae3eebd39  
> > 
> > I just ran this on my local branch, including the patches I sent
> > yesterday (and are still pending), and that went ok, so I propose
> > to wait and see until those 2 patches are accepted/denied.
> > 
> > Just a question.
> > I don't understand why xtensa fails today and not yesterday ...
> > Is there a random toolchain selection everyday for each package to save
> > buildserver resources?
> 
> The autobuilders only do random testing. Essentially, what they do is:
> 
>  - Pick a random architecture/toolchain configuration among the one
>    available in support/config-fragments/autobuild/
> 
>  - Run "make randpackageconfig" to generate on top of the
>    architecture/toolchain configuration a random selection of packages.
> 
>  - Build
> 
> So the results every day are different, since it's completely random
> what testing is being done. This allows us to test a very wide variety
> of package/toolchain/architecture combinations, since doing an
> exhaustive testing is simply impossible when you have thousands of
> Config.in options.

I see.
I understand the impossibility :-)
Thanks for sharing.

Kurt