[KJ] some *seriously* deceased code in the tree

[KJ] some *seriously* deceased code in the tree

[Robert P. J. Day]

  over here:

http://www.fsdev.dreamhosters.com/wiki/index.php?title=If_0_or_if_1_directives

i mentioned the depressing number of "#if 0" and "#if 1" directives in
the tree, many of which i'm *sure* have long since lost their value.
as one amusing example, i just ran across this in
arch/parisc/mm/fault.c:

======================#if 0
/* This is the treewalk to find a vma which is the highest that has
 * a start < addr.  We're using find_vma_prev instead right now, but
 * we might want to use this at some point in the future.  Probably
 * not, but I want it committed to CVS so I don't lose it :-)
 */
                        while (tree != vm_avl_empty) {
                                if (tree->vm_start > addr) {
                                        tree = tree->vm_avl_left;
                                } else {
                                        prev = tree;
                                        if (prev->vm_next = NULL)
                                                break;
                                        if (prev->vm_next->vm_start > addr)
                                                break;
                                        tree = tree->vm_avl_right;
                                }
                        }
#endif
======================
  in the first place, how old does kernel code have to be to have been
originally checked in via *CVS*?  :-)

  more importantly, though, this strikes me as a misuse of the
official kernel tree -- as an archival dumping ground for snippets of
code that have no immediate value but might someday be useful so, what
the heck, let's throw it in there and, who knows, we might get back to
it but maybe not.

  IMHO, that's the sort of thing that deserves to be ripped out
without a moment's hesitation.

  thoughts?

-- 
====================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA

http://www.fsdev.dreamhosters.com/wiki/index.php?title=Main_Page
====================================
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/kernel-janitors

Re: [KJ] some *seriously* deceased code in the tree

[Tobias Klauser]

On 2007-02-03 at 10:43:36 +0100, Robert P. J. Day <rpjday@mindspring.com> wrote:
[...]
>   in the first place, how old does kernel code have to be to have been
> originally checked in via *CVS*?  :-)

Some subprojects use or used CVS as their SCM system.

Cheers, Tobias
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/kernel-janitors

Re: [KJ] some *seriously* deceased code in the tree

[Matthew Wilcox]

On Sat, Feb 03, 2007 at 04:43:36AM -0500, Robert P. J. Day wrote:
> as one amusing example, i just ran across this in
> arch/parisc/mm/fault.c:
> 
> ======================> #if 0
> /* This is the treewalk to find a vma which is the highest that has
>  * a start < addr.  We're using find_vma_prev instead right now, but
>  * we might want to use this at some point in the future.  Probably
>  * not, but I want it committed to CVS so I don't lose it :-)
>  */
>                         while (tree != vm_avl_empty) {
>                                 if (tree->vm_start > addr) {
>                                         tree = tree->vm_avl_left;
>                                 } else {
>                                         prev = tree;
>                                         if (prev->vm_next = NULL)
>                                                 break;
>                                         if (prev->vm_next->vm_start > addr)
>                                                 break;
>                                         tree = tree->vm_avl_right;
>                                 }
>                         }
> #endif
> ======================> 
>   in the first place, how old does kernel code have to be to have been
> originally checked in via *CVS*?  :-)

The parisc-linux project switched from CVS to git in August 2006.
Since you asked so nicely, I originally checked in this code in December
2000, so it's just over 6 years old.

>   more importantly, though, this strikes me as a misuse of the
> official kernel tree -- as an archival dumping ground for snippets of
> code that have no immediate value but might someday be useful so, what
> the heck, let's throw it in there and, who knows, we might get back to
> it but maybe not.
> 
>   IMHO, that's the sort of thing that deserves to be ripped out
> without a moment's hesitation.

What harm does it do to leave it alone?  There's about a billion more
important things to spend time on than this.
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/kernel-janitors

Re: [KJ] some *seriously* deceased code in the tree

[Robert P. J. Day]

On Sat, 3 Feb 2007, Matthew Wilcox wrote:

> On Sat, Feb 03, 2007 at 04:43:36AM -0500, Robert P. J. Day wrote:

> >   more importantly, though, this strikes me as a misuse of the
> > official kernel tree -- as an archival dumping ground for snippets
> > of code that have no immediate value but might someday be useful
> > so, what the heck, let's throw it in there and, who knows, we
> > might get back to it but maybe not.
> >
> >   IMHO, that's the sort of thing that deserves to be ripped out
> > without a moment's hesitation.
>
> What harm does it do to leave it alone?  There's about a billion
> more important things to spend time on than this.

*technically*, of course, it doesn't do any harm at all.  but given
that people on the KJ list might be looking for small projects, what's
wrong with identifying and removing code that has absolutely no value
anymore?

i mean, any software project that has over 2500 "#if 0" directives
scattered through it wouldn't suffer from a little cleanup.

rday

-- 
====================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA

http://www.fsdev.dreamhosters.com/wiki/index.php?title=Main_Page
====================================
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/kernel-janitors