Hi,

On 05/15/2018 01:59 AM, Adrian Hunter wrote:

On 15/05/18 00:56, Thomas Hebb wrote:

quoted

Currently, runtime_suspend() unconditionally disables the clock gates
for the controller, which means that it's unable to receive interrupts
generated by connected SDIO cards.

We currently get / put runtime pm with enable / disable of the SDIO IRQ
(refer sdhci_enable_sdio_irq()) so are you sure this is needed?

You're correct; this patch is unnecessary. I wrote it before
923713b35745 ("mmc: sdhci: Disable runtime pm when the sdio_irq is
enabled"), and it was needed then. Sorry for the noise.

FYI, sdhci-esdhc-imx still checks the IRQ in its suspend/resume
functions. That's one of the things that misled me to think this patch
was still relevant.

quoted

Signed-off-by: Thomas Hebb <redacted>
---
 drivers/mmc/host/sdhci-pxav3.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/mmc/host/sdhci-pxav3.c b/drivers/mmc/host/sdhci-pxav3.c
index a34434166ca7..59760f3cc1d7 100644
--- a/drivers/mmc/host/sdhci-pxav3.c
+++ b/drivers/mmc/host/sdhci-pxav3.c

@@ -562,9 +562,11 @@ static int sdhci_pxav3_runtime_suspend(struct device *dev)
 	if (host->tuning_mode != SDHCI_TUNING_MODE_3)
 		mmc_retune_needed(host->mmc);
 
-	clk_disable_unprepare(pxa->clk_io);
-	if (!IS_ERR(pxa->clk_core))
-		clk_disable_unprepare(pxa->clk_core);
+	if (!sdhci_sdio_irq_enabled(host)) {
+		clk_disable_unprepare(pxa->clk_io);
+		if (!IS_ERR(pxa->clk_core))
+			clk_disable_unprepare(pxa->clk_core);
+	}
 
 	return 0;
 }

@@ -575,9 +577,11 @@ static int sdhci_pxav3_runtime_resume(struct device *dev)
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
 	struct sdhci_pxa *pxa = sdhci_pltfm_priv(pltfm_host);
 
-	clk_prepare_enable(pxa->clk_io);
-	if (!IS_ERR(pxa->clk_core))
-		clk_prepare_enable(pxa->clk_core);
+	if (!sdhci_sdio_irq_enabled(host)) {
+		clk_prepare_enable(pxa->clk_io);
+		if (!IS_ERR(pxa->clk_core))
+			clk_prepare_enable(pxa->clk_core);
+	}
 
 	return sdhci_runtime_resume_host(host);
 }