On Mon, Aug 29, 2011 at 05:54:23PM +0800, Eric Miao wrote:

On Mon, Aug 29, 2011 at 5:37 PM, Wolfram Sang [off-list ref] wrote:

quoted

On Mon, Aug 29, 2011 at 04:51:46PM +0800, Eric Miao wrote:

quoted

On Mon, Aug 29, 2011 at 4:36 PM, Wolfram Sang [off-list ref] wrote:

quoted

On Mon, Aug 29, 2011 at 03:18:55PM +0800, Richard Zhu wrote:

quoted

On imx53 AHCI, soft reset fails with IPMS set when PMP
is enabled but SATA HDD/ODD is connected to SATA port,
do soft reset again to port 0.
So the 'ahci_pmp_retry_srst_ops' is required when imx53
ahci is present.

Signed-off-by: Richard Zhu <redacted>
---
?drivers/ata/ahci_platform.c | ? 44 +++++++++++++++++++++++++++++++++++++-----
?1 files changed, 38 insertions(+), 6 deletions(-)

diff --git a/drivers/ata/ahci_platform.c b/drivers/ata/ahci_platform.c
index 6fef1fa..f32c91e 100644
--- a/drivers/ata/ahci_platform.c
+++ b/drivers/ata/ahci_platform.c

@@ -23,6 +23,41 @@

?#include <linux/ahci_platform.h>
?#include "ahci.h"

+enum ahci_type {
+ ? ? AHCI, ? ? ? ? ? /* standard platform ahci */
+ ? ? IMX53_AHCI, ? ? /* ahci on i.mx53 */

How about making all the IMX*-naming more generic because other SoC might need
this somewhen, too?

From a practical point of view, we can start with what we know already.
As the SATA controller on "other SoC" so far, we don't know yet if they
show the same issue (which we have to use ahci_pmp_retry_srst_ops).

So Wolfram, how about we consider a generic one once more controllers
are being added?

If we do so later, and we then also change the platform_device_id to something
generic, we then also have to change all users, too. Is it so bad to change
IMX53_AHCI (and imx53-ahci and alike) to something like "ahci-pmp-retry-srst"
(or similar) now?

There are other fields in ata_port_info which might be different between
boards. Using a SoC specific name as an index to an array of ata_port_info
will be more flexible, as is adopted in drivers/ata/ahci.c - the PCI driver.

Well, that file also has generic fixups:

       [board_ahci_ign_iferr] =
...

which is what I had in mind thinking that the flaw was not too imx53 specific.
If you think I am wrong, then be it. I mainly wanted to avoid the fixups which
might be needed later.

Regards,

   Wolfram

-- 
Pengutronix e.K.                           | Wolfram Sang                |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20110829/a13976b0/attachment.sig>