[Linux-aus] Extent of sco's complaint - birthday paradox

Greg 'groggy' Lehey grog at lemis.com
Sat Jun 7 09:54:02 UTC 2003


On Thursday,  5 June 2003 at 22:30:21 +0800, Leon Brooks wrote:
>
> Since they're addressing the same issues and with very similar
> constraints, a reasonably limited "instruction set" (C language) and
> working from similar educational bases and textbook examples, I'd
> expect many, many problems to be solved indistinguishably, yet
> without the Linux people ever seeing or even knowing that the SCO
> code exists.

That's certainly possible in BSD.  Just about anybody who writes a
block device driver writes the following line at some time or another:

     (*bdevsw[major(bp->b_dev)].d_strategy) (bp);

In the Third Edition (1972) I find:

/src/UNIX/PDP-11/Third-Edition/dmr/bio.c:       (*bdevsw[dev.d_major].d_strategy)(rbp);
/src/UNIX/PDP-11/Third-Edition/dmr/bio.c:       (*bdevsw[rbp->b_dev.d_major].d_strategy)(rbp);

In Linux the problem is that the kernel structures are so different
that it's difficult to imagine what use imported UNIX code would be.
One area would be in ELF initialization code: the data structures are
the same, and initializing them is a pretty mechanical task.  I've
compared code and found similarities.  They're not so close that you
could assume they had been copied.

It still seems much more plausible to me that SCO engineers
incorporated Linux code into System V than the other way round.
Firstly the Linux code is easier to get hold of, and secondly the
likelihood that they'd be caught is much smaller.

Greg
--
Finger grog at lemis.com for PGP public key
See complete headers for address and phone numbers
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url : http://lists.linux.org.au/pipermail/linux-aus/attachments/20030607/ebbe2c7a/attachment-0001.pgp 


More information about the linux-aus mailing list