[PATCH 05/07] sata_mv: mv_fill_sg fixes v2

[Mark Lord <liml@rtr.ca>]

Grant Grundler wrote:
> On Fri, Jan 30, 2009 at 3:50 PM, Mark Lord <liml@rtr.ca> wrote:
>> Fix mv_fill_sg() to zero out the reserved word (required for ATAPI),
>> and to include a memory barrier.  This may also help with problems
>> reported by Jens on the PPC platform.
>>
>> Signed-off-by: Mark Lord <mlord@pobox.com>
>>
>> --- old/drivers/ata/sata_mv.c   2009-01-30 18:31:03.000000000 -0500
>> +++ linux/drivers/ata/sata_mv.c 2009-01-30 18:32:08.000000000 -0500
>> @@ -1364,12 +1364,13 @@
>>                        u32 offset = addr & 0xffff;
>>                        u32 len = sg_len;
>>
>> -                       if ((offset + sg_len > 0x10000))
>> +                       if ((offset + len > 0x10000))
> 
> Why the double parens?
..

No reason, no harm -- they were there already.

But if we're being pedantic, here it is again without them:

(REPOST, without the redundant parenthesis)

Fix mv_fill_sg() to zero out the reserved word (required for ATAPI),
and to include a memory barrier.  This may also help with problems
reported by Jens on the PPC platform.

Signed-off-by: Mark Lord <mlord@pobox.com>

--- old/drivers/ata/sata_mv.c	2009-01-30 18:31:03.000000000 -0500
+++ linux/drivers/ata/sata_mv.c	2009-01-30 18:32:08.000000000 -0500
@@ -1364,12 +1364,13 @@
 			u32 offset = addr & 0xffff;
 			u32 len = sg_len;
 
-			if ((offset + sg_len > 0x10000))
+			if (offset + len > 0x10000)
 				len = 0x10000 - offset;
 
 			mv_sg->addr = cpu_to_le32(addr & 0xffffffff);
 			mv_sg->addr_hi = cpu_to_le32((addr >> 16) >> 16);
 			mv_sg->flags_size = cpu_to_le32(len & 0xffff);
+			mv_sg->reserved = 0;
 
 			sg_len -= len;
 			addr += len;
@@ -1381,6 +1382,7 @@
 
 	if (likely(last_sg))
 		last_sg->flags_size |= cpu_to_le32(EPRD_FLAG_END_OF_TBL);
+	mb(); /* ensure data structure is visible to the chipset */
 }
 
 static void mv_crqb_pack_cmd(__le16 *cmdw, u8 data, u8 addr, unsigned last)