Re: [PATCH v2 05/10] DMA, CMA: support arbitrary bitmap granularity

From: Minchan Kim
Date: Thu Jun 12 2014 - 02:42:51 EST


On Thu, Jun 12, 2014 at 03:43:55PM +0900, Joonsoo Kim wrote:
> On Thu, Jun 12, 2014 at 03:06:10PM +0900, Minchan Kim wrote:
> > On Thu, Jun 12, 2014 at 12:21:42PM +0900, Joonsoo Kim wrote:
> > > ppc kvm's cma region management requires arbitrary bitmap granularity,
> > > since they want to reserve very large memory and manage this region
> > > with bitmap that one bit for several pages to reduce management overheads.
> > > So support arbitrary bitmap granularity for following generalization.
> > >
> > > Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@xxxxxxx>
> > >
> > > diff --git a/drivers/base/dma-contiguous.c b/drivers/base/dma-contiguous.c
> > > index bc4c171..9bc9340 100644
> > > --- a/drivers/base/dma-contiguous.c
> > > +++ b/drivers/base/dma-contiguous.c
> > > @@ -38,6 +38,7 @@ struct cma {
> > > unsigned long base_pfn;
> > > unsigned long count;
> > > unsigned long *bitmap;
> > > + int order_per_bit; /* Order of pages represented by one bit */
> >
> > Hmm, I'm not sure it's good as *general* interface even though it covers
> > existing usecases.
> >
> > It forces a cma area should be handled by same size unit. Right?
> > It's really important point for this patchset's motivation so I will stop
> > review and wait other opinions.
>
> If you pass 0 to order_per_bit, you can manage cma area in every
> size(page unit) you want. If you pass certain number to order_per_bit,
> you can allocate and release cma area in multiple of such page order.
>
> I think that this is more general implementation than previous versions.

Fair enough.

--
Kind regards,
Minchan Kim
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/