Re: [PATCH 1/6] perf tools: Factor features display code

[Jiri Olsa]

On Wed, Feb 19, 2014 at 06:59:36PM -0300, Arnaldo Carvalho de Melo wrote:
> Em Wed, Feb 19, 2014 at 04:52:54PM +0100, Jiri Olsa escreveu:
> > Currently the we display all detected features/libraries
> > by following rules:
> >  - if one of the features is missing
> >  - if it's build from clean tree
> > 
> > This patch changes changes this behavior in several ways.
> > 
> > - We no longer display all detected features, only detected
> >   libraries are displayed by default:
> > 
> >   $ make
> >     BUILD:   Doing 'make -j4' parallel build
> > 
> >   Auto-detecting system features:
> >   ...                         dwarf: [ on  ]
> >   ...                         glibc: [ on  ]
> >   ...                          gtk2: [ on  ]
> >   ...                      libaudit: [ on  ]
> >   ...                        libbfd: [ on  ]
> >   ...                        libelf: [ on  ]
> >   ...                       libnuma: [ on  ]
> >   ...                       libperl: [ on  ]
> >   ...                     libpython: [ on  ]
> >   ...                      libslang: [ on  ]
> >   ...                     libunwind: [ on  ]
> 
> I like it, testing now, one suggestion: Since this is just about
> libraries, ditch the "lib' prefix, and make the header more
> informative, making it look like this:
> 
>    Auto-detecting system libraries that enables features:
>    ...                         dwarf: [ on  ]
>    ...                             c: [ on  ]
>    ...                          gtk2: [ on  ]
>    ...                         audit: [ on  ]
>    ...                           bfd: [ on  ]
>    ...                           elf: [ on  ]
>    ...                          numa: [ on  ]
>    ...                          perl: [ on  ]
>    ...                        python: [ on  ]
>    ...                         slang: [ on  ]
>    ...                        unwind: [ on  ]

yep, we could prettify it somehow.. but those names are tightly
bound to tests in config/feature-checks/Makefile .. so it'd
mean bigger change ;-)

>  
> Also, do we look first for dwarf, then for glibc?

the order is artifical.. we check all together.. almost ;-)

we can sort the output to make more sense, like from the most
crucial lib..

jirka
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/