I suppose the question is what is the float to be used for.  There are 
already (un)packdouble functions so passing it around to various tools 
should be fine.  I think when this has come up in other places it was 
read in as a string and translated to a double afterwards.  I couldn't 
find anywhere in the code where it was happening like this but I didn't 
look very hard.  I don't think adding the new function to read it in is 
a horrible idea, it was probably just never really needed.  So back to 
the first question, what is the float/double to e used for ;).  Perhaps 
it doesn't need to be a float.  But it may be this is the only option.

Danny

On 02/20/12 14:42, Michel Bourget wrote:
> Hi all,
>
> somehow, I need to introduce handling a "float" value in slurm.conf and,
> of course,
> carry/dislay/obtain it.  I noticed it's not there. before I go introduce it,
> is there any reason ( I am not aware of ) it's not there ,
> except for the fact " we never needed it before" ?
>
> A+
>

Reply via email to