Re: [dmidecode] contribute to dmidecode for spec 2.8 and 3.0

[sai]

Hi Jean,

Thanks. I've submitted the patch. I wasn't referring to the struct in 
the code but rather the strings used were different from what was 
available in the specification (in Type 4, Table 22, Meaning Column). Is 
it because they are ENUMs ?


Also, is it possible to build a next version and make it available as 
dmidecide-maj.min.tar.gz for 2.8 and 3.0 ? So that buildroot can 
automatically pick it up from 
http://download.savannah.gnu.org/releases/dmidecode/ ?

--
Sai



On 2015-08-27 01:40, Jean Delvare wrote:
> Hi Sai,
>
> On Wed, 26 Aug 2015 11:23:01 -0600, address@hidden wrote:
> > Yes, thanks for the information and the ticket about Contribution
> > information. I think I spoke too soon :). I was using dmidecode-2.12 
> > via
> > buildroot and thought that was on the HEAD of the source tree. I
> > realized after I did a git clone of the code that all that I needed 
> > was
> > already there.
>
> OK, great.
>
> > I found a typo that the 2.8 specification corrected that I didn't see 
> > in
> > the code so I'll submit that :).
>
> Please do!
>
> > I was also wondering why aren't the names such as '(7.5.2) Processor
> > Family' are not the same as those in the specification ?
>
> This particular example seems to match just fine, both specification
> versions 2.8.0 and 3.0.0 have "7.5.2 Processor Information - Processor
> Family" and the source code has:
>
> static const char *dmi_processor_family(const struct dmi_header *h, u16 
> ver)
> {
>         const u8 *data = h->data;
>         unsigned int i, low, high;
>         u16 code;
>
>         /* 7.5.2 */
>         static struct {
>                 int value;
>                 const char *name;
>         } family2[] = {
>
> So what exactly is bothering you?