[PATCH/RFT] sata: rcar_sata: Reset SATA PHY on resume

[PATCH/RFT] sata: rcar_sata: Reset SATA PHY on resume

[Simon Horman]

In the case where power is cut on suspend the SATA PHY state needs to be
suspended on resume.

This is the case on the Salvator-X board with the r8a7795 or r8a7796 SoC.
In that environment it has been observed that SATA partitions cannot not be
accessed on resume.

Based on work by Khiem Nguyen.

Cc: Khiem Nguyen <khiem.nguyen.xt@rvc.renesas.com>
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
---
* Compile and boot tested only
---
 drivers/ata/sata_rcar.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/drivers/ata/sata_rcar.c b/drivers/ata/sata_rcar.c
index 5d38245a7a73..4f4f7788a6cf 100644
--- a/drivers/ata/sata_rcar.c
+++ b/drivers/ata/sata_rcar.c
@@ -969,16 +969,9 @@ static int sata_rcar_resume(struct device *dev)
 {
 	struct ata_host *host = dev_get_drvdata(dev);
 	struct sata_rcar_priv *priv = host->private_data;
-	void __iomem *base = priv->base;
 
 	clk_prepare_enable(priv->clk);
-
-	/* ack and mask */
-	iowrite32(0, base + SATAINTSTAT_REG);
-	iowrite32(0x7ff, base + SATAINTMASK_REG);
-	/* enable interrupts */
-	iowrite32(ATAPI_INT_ENABLE_SATAINT, base + ATAPI_INT_ENABLE_REG);
-
+	sata_rcar_init_controller(host);
 	ata_host_resume(host);
 
 	return 0;
-- 
2.7.0.rc3.207.g0ac5344

Re: [PATCH/RFT] sata: rcar_sata: Reset SATA PHY on resume

[Tejun Heo]

On Tue, Apr 18, 2017 at 04:02:53PM +0900, Simon Horman wrote:
> In the case where power is cut on suspend the SATA PHY state needs to be
> suspended on resume.
  ^
  re-initialized?

> 
> This is the case on the Salvator-X board with the r8a7795 or r8a7796 SoC.
> In that environment it has been observed that SATA partitions cannot not be
> accessed on resume.
> 
> Based on work by Khiem Nguyen.
> 
> Cc: Khiem Nguyen <khiem.nguyen.xt@rvc.renesas.com>
> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
> ---
> * Compile and boot tested only

Can somebody verify this fix?

Thanks.

-- 
tejun

Re: [PATCH/RFT] sata: rcar_sata: Reset SATA PHY on resume

[Simon Horman]

On Tue, Apr 18, 2017 at 04:17:38PM +0900, Tejun Heo wrote:
> On Tue, Apr 18, 2017 at 04:02:53PM +0900, Simon Horman wrote:
> > In the case where power is cut on suspend the SATA PHY state needs to be
> > suspended on resume.
>   ^
>   re-initialized?

Yes, sorry for the think-o.

> > This is the case on the Salvator-X board with the r8a7795 or r8a7796 SoC.
> > In that environment it has been observed that SATA partitions cannot not be
> > accessed on resume.
> > 
> > Based on work by Khiem Nguyen.
> > 
> > Cc: Khiem Nguyen <khiem.nguyen.xt@rvc.renesas.com>
> > Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
> > ---
> > * Compile and boot tested only
> 
> Can somebody verify this fix?

Yes, I should be able to do so but I need to set up the appropriate
hardware environment first. I would, of course, welcome verification by
others.

Re: [PATCH/RFT] sata: rcar_sata: Reset SATA PHY on resume

[Sergei Shtylyov]

Hello.

On 4/18/2017 10:02 AM, Simon Horman wrote:

> In the case where power is cut on suspend the SATA PHY state needs to be
> suspended on resume.
>
> This is the case on the Salvator-X board with the r8a7795 or r8a7796 SoC.
> In that environment it has been observed that SATA partitions cannot not be
> accessed on resume.
>
> Based on work by Khiem Nguyen.
>
> Cc: Khiem Nguyen <khiem.nguyen.xt@rvc.renesas.com>
> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
> ---
> * Compile and boot tested only
> ---
>  drivers/ata/sata_rcar.c | 9 +--------
>  1 file changed, 1 insertion(+), 8 deletions(-)

    Why the scope prefix in the subject is not just "sata_rcar: "? :-)

[...]

MBR, Sergei

Re: [PATCH/RFT] sata: rcar_sata: Reset SATA PHY on resume

[Simon Horman]

On Tue, Apr 18, 2017 at 12:20:20PM +0300, Sergei Shtylyov wrote:
> Hello.
> 
> On 4/18/2017 10:02 AM, Simon Horman wrote:
> 
> >In the case where power is cut on suspend the SATA PHY state needs to be
> >suspended on resume.
> >
> >This is the case on the Salvator-X board with the r8a7795 or r8a7796 SoC.
> >In that environment it has been observed that SATA partitions cannot not be
> >accessed on resume.
> >
> >Based on work by Khiem Nguyen.
> >
> >Cc: Khiem Nguyen <khiem.nguyen.xt@rvc.renesas.com>
> >Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
> >---
> >* Compile and boot tested only
> >---
> > drivers/ata/sata_rcar.c | 9 +--------
> > 1 file changed, 1 insertion(+), 8 deletions(-)
> 
>    Why the scope prefix in the subject is not just "sata_rcar: "? :-)

Thanks, I will drop that.