Re: Memory allocation question

[george anzinger <george@mvista.com>]

Rik van Riel wrote:
> 
> On Wed, 21 Nov 2001, george anzinger wrote:
> > Alan Cox wrote:
> > >
> > > > size chuncks.  Currently I am using kmalloc() to allocate a page at a
> > > > time.  I don't want to have to worry about mapping/unmapping etc.  I
> > >
> > > Use get_free_page() to get page sized chunks
> >
> > What about __get_free_page() ?  I don't need or want the clear page
> > (performance issues).
> 
> get_free_page() doesn't clear the page afaics.
> 
In mm.h (2.4.13 kernel)....

#define __get_free_page(gfp_mask) \
		__get_free_pages((gfp_mask),0)

#define __get_dma_pages(gfp_mask, order) \
		__get_free_pages((gfp_mask) | GFP_DMA,(order))

/*
 * The old interface name will be removed in 2.5:
 */
#define get_free_page get_zeroed_page
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Where as __get_free_page() does not zero.  I know this is an old kernel,
but...


What am I missing?
-- 
George           george@mvista.com
High-res-timers: http://sourceforge.net/projects/high-res-timers/
Real time sched: http://sourceforge.net/projects/rtsched/