[2476] in SIPB_Linux_Development

home help back first fref pref prev next nref lref last post

Re: Kernel interface changes (was Re: cdrecord problems on

daemon@ATHENA.MIT.EDU (Kev)
Thu Feb 4 09:06:57 1999

To: Stephen Harris <sweh@mpn.com>
Cc: alan@lxorguk.ukuu.org.uk (Alan Cox), efoo@MIT.EDU, warlord@MIT.EDU,
        xiphmont@MIT.EDU, linux-kernel@vger.rutgers.edu, linux-dev@MIT.EDU,
        jered@MIT.EDU, nemo@MIT.EDU, linux-scsi@vger.rutgers.edu,
        cox@idecnet.com
In-Reply-To: Your message of "Thu, 04 Feb 1999 07:30:09 GMT."
             <199902040730.HAA31679@nebula.mpn.com> 
Date: Thu, 04 Feb 1999 09:06:30 EST
From: Kev <klmitch@MIT.EDU>

> > If you want to make this work even better run a lot of 2.0 binaries on 2.2,
> > check they all still work. If not find out why. Similarly people should be
> 
> I don't understand why Linux is taking so much flack here.   Going from 2.0
> to 2.2 is a MAJOR CHANGE.  Hell, even Solaris 2.4 programs don't run properly
> under Solaris 2.6 (and not even under 2.5.1 - try out the Lotus Notes 4.0
> client, for example).

This thread isn't about going from 2.0 to 2.2; it's about going from
2.0.33 to 2.0.34.  In the past, changes such as this caused major
compatibility headaches; this must be prevented from happening in the
future if Linux expects to get anywhere.

> Even with the ..x patches, the user has a choice; backwards compatibility with
> BUGS, or a bugless mode with the possibility of an unfortunate binary side
> effect.  Heh, Alan, for each of your patches have a kernel compile configuration
> option "enable XYZ bug mode for compatibility" :-)

Again, this isn't what has happened in the past; people have *reordered*
*structure* *members*.  This isn't fixing a bug!  This is randomly and
gratuitously breaking binary compatibility *in* *a* *stable* *release*
for no benefit!  THIS is what must stop happening if Linux expects to
get anywhere.

> And anyway, this isn't linux-scsi related...

Actually, this whole thread started because one developer I know kept
getting the SCSI interface changed out from under him, again, *in* *a*
*stable* *release*; therefore, it is certainly linux-scsi related.
-- 
Kevin L. Mitchell <klmitch@mit.edu>
-------------------------  -. .---- --.. ..- -..-  --------------------------
http://web.mit.edu/klmitch/www/               (PGP keys availiable from here)
    RSA AE87D37D/1024:  DE EA 1E 99 3F 2B F9 23  A0 D8 05 E0 6F BA B9 D2
    DSS ED0DB34E/1024: D9BF 0E74 FDCB 43F5 C597  878F 9455 EC24 ED0D B34E
    DH  2A2C31D4/2048: 1A77 4BA5 9E32 14AE 87DA  9FEC 7106 FC62 2A2C 31D4


home help back first fref pref prev next nref lref last post