DANGEROUS means that what is known about the registers is EXPERIMENTAL
grade; however, if one tampers with this area without knowing something
about it.....You can "goat screw yourself".
Specifically, I have basic knowledge of the RAID/STRIPING powers of the
Promise FastTrak series. This was learned on imperical tests and crashes.
Unless you have several drives of the same kind to wreck, and run a mirror
backup copy. It is not suggested.
It is IMO that "DANGEROUS" is much nicer than "SCREW YOURSELF CAREFULLY".
Also that only appears in the "DEVELOPMENT KERNEL" or a patch of my that
is back-coded from there; therefore, "DANGEROUS" is acceptable in that
context. Do you see something like this in the "STABLE KERNEL", no.
Andre Hedrick
The Linux IDE guy
On Wed, 8 Sep 1999, Gerard Roudier wrote:
>
>
> On Wed, 8 Sep 1999, Jens Benecke wrote:
>
> > On Wed, Sep 08, 1999 at 02:26:26AM +0200, David Weinehall wrote:
> >
> > > > > o A "Prompt for drivers/code flagged DANGEROUS" option in the
> > > > > code-maturity section that is conditional depending on EXPERIMENTAL
> > > > What is the difference between experimental/dangerous, except the
> > > > latter scares people.
> > > That was kind of the idea... There is a reason for flagging code
> > > DANGEROUS. Only people who really doesn't care if their dogs die, their
> > > grandmas get abducted by aliens, their disks crashed and their screens
> > > burned out enables such code, while many wannabe hackers enables
> > > EXPERIMENTAL code.
> >
> > And could you please put _exactly_ that in the "Help" menu for the
> > CONFIG_DANGEROUS switch. I think there should be some humour in such boring
> > things as configuring a new kernel. ;)
>
> Any software is DANGEROUS for the reason that it contains bugs. Even a
> known bug may have catastrophic effects in some situations. So, you
> cannot expect unknown bugs to be more kind. :-)
> The only relevant information could be some status that reflects the tests
> that have been done, tells about the software being stable or still in
> heavy development, the number of systems that use the software with
> success, etc ..., but this information would be very complex to maintain
> for each kernel module.
>
> In my opinion, the EXPERIMENTAL status that warns user about code that a
> software still need testings or is still under heavy development is just
> fine and enough.
>
> Counting the number of file systems a software eats every second will not
> help at all, in my opinion. ;-)
>
> Gérard.
>
>
>
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.rutgers.edu
> Please read the FAQ at http://www.tux.org/lkml/
>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/