On Thu, Jul 22, 2010 at 11:15 PM, Stefan Roese [off-list ref] wrote:

Hi Marri,

On Friday 23 July 2010 02:57:18 tmarri@amcc.com wrote:

quoted

From: Tirumala Marri <redacted>

=A0 This patch will add ADMA support for DMA engine and HW offload for
=A0 XOR/ADG (RAID-5/6) functionalities.
=A0 1. It supports memcpy, xor, GF(2) based RAID-6.
=A0 2. It supports interrupt based DMA completions.
=A0 3. Also supports memcpy in RAID-1 case.

=A0 Kernel version: 2.6.35-rc5

=A0 Testing:
=A0 =A0 Created RAID-5/6 arrays usign mdadm.
=A0 =A0 And ran raw IO and filesystem IO to the RAID array.
=A0 =A0 Chunk size 4k,64k was tested.
=A0 =A0 RAID rebuild , disk fail, resync tested.

=A0 File names:
=A0 =A0 This code is similar to ppc440spe . So I named the files as
=A0 =A0 drivers/dma/ppc4xx/adma1.c and drivers/dma/ppc4xx/adma1.h

As you describe above, a lot of the code seems to be copied from
drivers/dma/ppc4xx/adma.c/h. Wouldn't it make more sense to factor out th=

e

common code instead of duplicating it?

Yes, and you might look to drivers/dma/iop-adma.c as an example of a
way to support similar hardware with a single code base.

--
Dan