RE: [PATCH] [scsi] Remove __GFP_DMA

["Salyzyn, Mark" <mark_salyzyn@adaptec.com>]

The 31 bit limit for some of these cards is a problem, we currently only
do __GFP_DMA for bounce buffer sg elements allocated for user supplied
references in ioctls.

I figure we should be using pci_alloc_consistent calls for these
allocations to more accurately acquire memory within the 31 bit limit if
necessary, we could switch to these to remove the need for the __GFP_DMA
flag in the aacraid driver?

-- Mark

-----Original Message-----
From: linux-scsi-owner@vger.kernel.org
[mailto:linux-scsi-owner@vger.kernel.org] On Behalf Of James Bottomley
Sent: Wednesday, May 23, 2007 11:18 AM
To: Aubrey Li
Cc: Christoph Lameter; Bernhard Walle; linux-scsi@vger.kernel.org;
Andrew Morton; linux-kernel@vger.kernel.org
Subject: Re: [PATCH] [scsi] Remove __GFP_DMA

On Wed, 2007-05-23 at 10:41 +0800, Aubrey Li wrote:
> On 5/23/07, Christoph Lameter <clameter@sgi.com> wrote:
> > On Mon, 21 May 2007, Bernhard Walle wrote:
> >
> > > [PATCH] [scsi] Remove __GFP_DMA
> > >
> > > After 821de3a27bf33f11ec878562577c586cd5f83c64, it's not necessary
to alloate a
> > > DMA buffer any more in sd.c.
> > >
> > > Signed-off-by: Bernhard Walle <bwalle@suse.de>
> >
> > Great that avoids a DMA kmalloc slab. Any other GFP_DMAs left in the
scsi
> > layer?
> >
> > Acked-by: Christoph Lameter <clameter@sgi.com>
> 
> Yes, here is another patch

I'll defer to Mark on this one.  However, please remember that you can't
just blindly remove GFP_DMA ... there are some cards which require it.

Aacraid is one example ... it has a set of cards that can only DMA to 31
bits.  For them, the GFP_DMA is necessary:  The allocation in question
is a scatterlist, which must be within the device DMA mask.

James


-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html