[PATCH] Priority heap infrastructure enhancements

[Andrew Morton]

On Sat, 21 Jun 2008 13:35:38 +0530 Balbir Singh <balbir at linux.vnet.ibm.com> wrote:

> Paul Menage wrote:
> > On Wed, Jun 18, 2008 at 6:48 AM, Balbir Singh <balbir at linux.vnet.ibm.com> wrote:
> >> Some of the common code has been factored into heap_adjust() a.k.a heapify
> >> in data structures terminology.
> >>
> >> I am sending out this  patch indepedent of the memory controller changes as
> >> they deserve to be reviewed independently.
> >>
> >> One limitation of the current heap_insert() routine is that it does not
> >> insert an element which is greater than the root, when the heap slots
> >> are fully used. I'll work on and review that interface and find a suitable
> >> way to address that issue
> > 
> > How else would you want it to behave? If you have a fixed size heap
> > and it's full, then you have to drop the largest value. (Well, you
> > could in theory drop the smallest value, but there's no quick way to
> > find that.)
> > 
> 
> I would like to be able to drop the smallest value. Since we cannot drop the
> smallest value, dropping a leaf (heap->size) should be sufficiently good enough.
> I want a max heap and losing the root of the heap does not work for me.
> 
> 
> >> Comments, Flames? Please do review closely!
> >>
> >> Signed-off-by: Balbir Singh <balbir at linux.vnet.ibm.com>
> > 
> > Looks fine.
> > 
> > Reviewed-by: Paul Menage <menage at google.com>
> 
> Thanks for the review!
> 

yup, thanks.

I'll duck the patch until we have some code which uses it.  Please
retain Paul's Reviewed-by: for that occasion.