[RFT PATCH] mmc: tmio: make sure SDIO gets reinitialized after resume

[RFT PATCH] mmc: tmio: make sure SDIO gets reinitialized after resume

[Wolfram Sang]

To achieve that, we set the registers in the generic HW reset routine
which gets called at both, init and resume. We also make sure to move
SDIO initialization before reset gets called in probe().

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
---

This patch addresses a report I got from Shimoda-san. I can't really test it
because the SDIO cards I have all require MMC_PM_KEEP_POWER to suspend and
that we don't support. So, I send this out to ask for testing. At least it
didn't cause regressions for the setup I used. To make testing easier, this
is *not* put on top of Simon's SDHI DMA rework patches, so it should apply
on current upstream and development branches.

 drivers/mmc/host/tmio_mmc_pio.c | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)

diff --git a/drivers/mmc/host/tmio_mmc_pio.c b/drivers/mmc/host/tmio_mmc_pio.c
index a2d92f10501bdd..5e87fcaa99d430 100644
--- a/drivers/mmc/host/tmio_mmc_pio.c
+++ b/drivers/mmc/host/tmio_mmc_pio.c
@@ -229,6 +229,12 @@ static void tmio_mmc_reset(struct tmio_mmc_host *host)
 	if (host->pdata->flags & TMIO_MMC_HAVE_HIGH_REG)
 		sd_ctrl_write16(host, CTL_RESET_SDIO, 0x0001);
 	msleep(10);
+
+	if (host->pdata->flags & TMIO_MMC_SDIO_IRQ) {
+		sd_ctrl_write16(host, CTL_SDIO_IRQ_MASK, host->sdio_irq_mask);
+		sd_ctrl_write16(host, CTL_TRANSACTION_CTL, 0x0001);
+	}
+
 }
 
 static void tmio_mmc_reset_work(struct work_struct *work)
@@ -1221,6 +1227,10 @@ int tmio_mmc_host_probe(struct tmio_mmc_host *_host,
 	if (_host->native_hotplug)
 		pm_runtime_get_noresume(&pdev->dev);
 
+	_host->sdio_irq_enabled = false;
+	if (pdata->flags & TMIO_MMC_SDIO_IRQ)
+		_host->sdio_irq_mask = TMIO_SDIO_MASK_ALL;
+
 	tmio_mmc_clk_stop(_host);
 	tmio_mmc_reset(_host);
 
@@ -1237,13 +1247,6 @@ int tmio_mmc_host_probe(struct tmio_mmc_host *_host,
 
 	_host->sdcard_irq_mask &= ~irq_mask;
 
-	_host->sdio_irq_enabled = false;
-	if (pdata->flags & TMIO_MMC_SDIO_IRQ) {
-		_host->sdio_irq_mask = TMIO_SDIO_MASK_ALL;
-		sd_ctrl_write16(_host, CTL_SDIO_IRQ_MASK, _host->sdio_irq_mask);
-		sd_ctrl_write16(_host, CTL_TRANSACTION_CTL, 0x0001);
-	}
-
 	spin_lock_init(&_host->lock);
 	mutex_init(&_host->ios_lock);
 
-- 
2.11.0

RE: [RFT PATCH] mmc: tmio: make sure SDIO gets reinitialized after resume

[Yoshihiro Shimoda]

Hi Wolfram-san,

> From: Wolfram Sang
> Sent: Friday, May 19, 2017 9:31 PM
> 
> To achieve that, we set the registers in the generic HW reset routine
> which gets called at both, init and resume. We also make sure to move
> SDIO initialization before reset gets called in probe().
> 
> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
> ---
> 
> This patch addresses a report I got from Shimoda-san. I can't really test it
> because the SDIO cards I have all require MMC_PM_KEEP_POWER to suspend and
> that we don't support. So, I send this out to ask for testing. At least it
> didn't cause regressions for the setup I used. To make testing easier, this
> is *not* put on top of Simon's SDHI DMA rework patches, so it should apply
> on current upstream and development branches.

Thank you for the patch!
Hayakawa-san tested this patch on both H3/Salvator-X and E2/ALT.

So,
Tested-by: Masaharu Hayakawa <masaharu.hayakawa.ry@renesas.com>

Best regards,
Yoshihiro Shimoda

>  drivers/mmc/host/tmio_mmc_pio.c | 17 ++++++++++-------
>  1 file changed, 10 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/mmc/host/tmio_mmc_pio.c b/drivers/mmc/host/tmio_mmc_pio.c
> index a2d92f10501bdd..5e87fcaa99d430 100644
> --- a/drivers/mmc/host/tmio_mmc_pio.c
> +++ b/drivers/mmc/host/tmio_mmc_pio.c
> @@ -229,6 +229,12 @@ static void tmio_mmc_reset(struct tmio_mmc_host *host)
>  	if (host->pdata->flags & TMIO_MMC_HAVE_HIGH_REG)
>  		sd_ctrl_write16(host, CTL_RESET_SDIO, 0x0001);
>  	msleep(10);
> +
> +	if (host->pdata->flags & TMIO_MMC_SDIO_IRQ) {
> +		sd_ctrl_write16(host, CTL_SDIO_IRQ_MASK, host->sdio_irq_mask);
> +		sd_ctrl_write16(host, CTL_TRANSACTION_CTL, 0x0001);
> +	}
> +
>  }
> 
>  static void tmio_mmc_reset_work(struct work_struct *work)
> @@ -1221,6 +1227,10 @@ int tmio_mmc_host_probe(struct tmio_mmc_host *_host,
>  	if (_host->native_hotplug)
>  		pm_runtime_get_noresume(&pdev->dev);
> 
> +	_host->sdio_irq_enabled = false;
> +	if (pdata->flags & TMIO_MMC_SDIO_IRQ)
> +		_host->sdio_irq_mask = TMIO_SDIO_MASK_ALL;
> +
>  	tmio_mmc_clk_stop(_host);
>  	tmio_mmc_reset(_host);
> 
> @@ -1237,13 +1247,6 @@ int tmio_mmc_host_probe(struct tmio_mmc_host *_host,
> 
>  	_host->sdcard_irq_mask &= ~irq_mask;
> 
> -	_host->sdio_irq_enabled = false;
> -	if (pdata->flags & TMIO_MMC_SDIO_IRQ) {
> -		_host->sdio_irq_mask = TMIO_SDIO_MASK_ALL;
> -		sd_ctrl_write16(_host, CTL_SDIO_IRQ_MASK, _host->sdio_irq_mask);
> -		sd_ctrl_write16(_host, CTL_TRANSACTION_CTL, 0x0001);
> -	}
> -
>  	spin_lock_init(&_host->lock);
>  	mutex_init(&_host->ios_lock);
> 
> --
> 2.11.0

Re: [RFT PATCH] mmc: tmio: make sure SDIO gets reinitialized after resume

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 777 bytes --]

Shimoda-san,

> > This patch addresses a report I got from Shimoda-san. I can't really test it
> > because the SDIO cards I have all require MMC_PM_KEEP_POWER to suspend and
> > that we don't support. So, I send this out to ask for testing. At least it
> > didn't cause regressions for the setup I used. To make testing easier, this
> > is *not* put on top of Simon's SDHI DMA rework patches, so it should apply
> > on current upstream and development branches.
> 
> Thank you for the patch!
> Hayakawa-san tested this patch on both H3/Salvator-X and E2/ALT.
> 
> So,
> Tested-by: Masaharu Hayakawa <masaharu.hayakawa.ry@renesas.com>

Very good, thanks, also to Hayakawa-san! I'll rebase this patch and
resubmit with the tag added!

Regards,

   Wolfram


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]