Peter Memishian writes:
> 
>  > If you have any questions or concerns about this overall direction
>  > (mirroring some user-level commands into mdb dcmd analogs, with
>  > different but "familiar" and "useful" output), please do speak up.
> 
> I think it's reasonable, with the understanding that those differences are
> likely to grow.  In particular, as we move to an increasingly sophisticated
> userland for networking (e.g., consider how dladm and NWAM profiles will
> interact in the future, and how dladm and dlmgmtd already interact),
> ::dladm will likely need to diverge more from dladm(1M).  Getting userland
> right is hard enough without mirrored dcmds in the picture, so I think
> we'll need to just accept those limitations when they arise.

Right.  I expect it to be just roughly analogous and useful as long as
you don't get too fancy in your demands.

In other words, the dcmd output will include things that you wouldn't
display in any user-space command (such as data structure addresses),
it won't contain things that are only accessible in the user space
code (such as SMF repository bits), and many of the options (such as
"-o" and "-R") are meaningless, so only a few basic features will be
present.

The idea is just to reduce (perhaps just slightly) the number of odd
things that kernel engineers need to remember when looking at dumps.
Replicating the entire command (with all bells and whistles) inside
mdb is intentionally not the goal.

-- 
James Carlson, Solaris Networking              <[email protected]>
Sun Microsystems / 35 Network Drive        71.232W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 781 442 1677
_______________________________________________
networking-discuss mailing list
[email protected]

Reply via email to