On Monday 18 November 2013 05:45 PM, Andreas Fenkart wrote:
2013/11/18 Michael Trimarchi <michael@xxxxxxxxxxxxxxxxxxxx>:
Hi Andreas
On Mon, Nov 18, 2013 at 8:53 AM, Andreas Fenkart <afenkart@xxxxxxxxx> wrote:
The am335x can't detect pending cirq in PM runtime suspend.
This patch reconfigures dat1 as a GPIO before going to suspend.
SDIO interrupts are detected with the GPIO, the GPIO will only wake
the module from suspend, SDIO irq detection will still happen through the
IP block.
Idea of remuxing the pins by Tony Lindgren as well as the implementation
of omap_hsmmc_pin_init.
Signed-off-by: Andreas Fenkart <afenkart@xxxxxxxxx>
diff --git a/Documentation/devicetree/bindings/mmc/ti-omap-hsmmc.txt b/Documentation/devicetree/bindings/mmc/ti-omap-hsmmc.txt
index 1136e6b..146f3ad 100644
--- a/Documentation/devicetree/bindings/mmc/ti-omap-hsmmc.txt
+++ b/Documentation/devicetree/bindings/mmc/ti-omap-hsmmc.txt
@@ -21,8 +21,11 @@ ti,non-removable: non-removable slot (like eMMC)
ti,needs-special-reset: Requires a special softreset sequence
ti,needs-special-hs-handling: HSMMC IP needs special setting for handling High Speed
ti,quirk-swakup-missing: SOC missing the swakeup line, will not detect
-SDIO irq while in suspend. Fallback to polling. Affected chips are
-am335x,
+SDIO irq while in suspend. The workaround is to reconfigure the dat1 line as a
+GPIO upon suspend. Beyond this option and the GPIO config, you also need to set
+named pinctrl states "default", "active" and "idle ", see example below. The
+MMC driver will then then toggle between default and idle during the runtime
+Affected chips are am335x,
------
| PRCM |
@@ -49,3 +52,24 @@ Example:
vmmc-supply = <&vmmc>; /* phandle to regulator node */
ti,non-removable;
};
+
+[am335x with with gpio for sdio irq]
+
+ mmc1_cirq_pin: pinmux_cirq_pin {
+ pinctrl-single,pins = <
+ 0x0f8 0x3f /* MMC0_DAT1 as GPIO2_28 */
+ >;
+ };
+
+ mmc1: mmc@48060000 {
+ ti,non-removable;
+ bus-width = <4>;
+ vmmc-supply = <&ldo2_reg>;
+ vmmc_aux-supply = <&vmmc>;
+ ti,quirk-swakeup-missing;
+ pinctrl-names = "default", "active", "idle";
+ pinctrl-0 = <&mmc1_pins>;
+ pinctrl-1 = <&mmc1_pins>;
+ pinctrl-2 = <&mmc1_cirq_pin>;
+ ti,cirq-gpio = <&gpio3 28 0>;
+ };
diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c
index 6b0ec55..b94ab08 100644
--- a/drivers/mmc/host/omap_hsmmc.c
+++ b/drivers/mmc/host/omap_hsmmc.c
@@ -36,6 +36,7 @@
#include <linux/mmc/core.h>
#include <linux/mmc/mmc.h>
#include <linux/io.h>
+#include <linux/irq.h>
#include <linux/gpio.h>
#include <linux/regulator/consumer.h>
#include <linux/pinctrl/consumer.h>
@@ -213,11 +214,32 @@ struct omap_hsmmc_host {
int req_in_progress;
int flags;
#define HSMMC_SDIO_IRQ_ENABLED (1 << 0) /* SDIO irq enabled */
+#define HSMMC_SWAKEUP_QUIRK (1 << 1)
+#define HSMMC_CIRQ_GPIO_ENABLED (1 << 2)
struct omap_hsmmc_next next_data;
+ struct pinctrl *pinctrl;
+ struct pinctrl_state *fixed, *active, *idle;
struct omap_mmc_platform_data *pdata;
};
+static irqreturn_t omap_hsmmc_cirq(int irq, void *dev_id)
+{
+ struct omap_hsmmc_host *host = dev_id;
+ unsigned long flags;
+
+ spin_lock_irqsave(&host->irq_lock, flags);
+ if (host->flags & HSMMC_CIRQ_GPIO_ENABLED) {
+ disable_irq_nosync(mmc_slot(host).sdio_irq);
+ host->flags &= ~HSMMC_CIRQ_GPIO_ENABLED;
+ }
+ spin_unlock_irqrestore(&host->irq_lock, flags);
+
+ pm_request_resume(host->dev); /* no use counter */
+
+ return IRQ_HANDLED;
+}
+
static int omap_hsmmc_card_detect(struct device *dev, int slot)
{
struct omap_hsmmc_host *host = dev_get_drvdata(dev);
@@ -452,10 +474,25 @@ static int omap_hsmmc_gpio_init(struct omap_mmc_platform_data *pdata)
} else
pdata->slots[0].gpio_wp = -EINVAL;
+ if (gpio_is_valid(pdata->slots[0].gpio_cirq)) {
+ pdata->slots[0].sdio_irq =
+ gpio_to_irq(pdata->slots[0].gpio_cirq);
What is this? re-assign the platform data?
Seems like, I didn't pay attention to this.
Simply kept in line how the write protection/read only pins are managed.
I'd rather not change this part, or not changing it as part of adding
sdio IRQ support it.
Maybe somebody else on the list can explain why the platform data
contains elements
that are modified during runtime.
- set_power / get_ro function callbacks
- ocr_mask.
Hi,
few params were passed via platform data in non-DT case and never cached
in internal data structure, with non-dt support going away soon, I am
planning to cleanup pdata usage in the driver when it gets to DT only support.
--
To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
