[PATCH] detect path to dmidecode in configure (fd.o bug #9318)

William Jon McCann mccann at jhu.edu
Sat Feb 17 07:44:38 PST 2007


Hi,

On 2/16/07, Danny Kukawka <danny.kukawka at web.de> wrote:
> On Thursday 15 February 2007 20:21, David Zeuthen wrote:
> > On Fri, 2007-02-09 at 19:43 +0100, Danny Kukawka wrote:
> > > this patch adds detection of the location of dmidecode to configure. If
> > > dmidecode could not get found we fall back to the same path as already
> > > hardcoded in probe-smbios.c. This would fix fd.o bug #9318
> >
> > I guess that's fine although I'm not too fond of requiring the tool to
> > be available during build (e.g. it forces a BuildRequires in the RPM
> > spec file). But I suppose it's useful for some people. Please commit.
> > Thanks.
>
> The alternative is the attached patch which check the path to dmidecode on
> runtime. This should be no really problem with performance since the addon
> get started only one time - on hald startup.
>
> I can live also with this patch.

Comments on the patch.

Isn't it better to use the following?

const char *path;
static const char *possible_locations[];

Jon


More information about the hal mailing list