2015-05-04 14:06 GMT+02:00 Sascha Hauer [off-list ref]:

On Mon, May 04, 2015 at 01:05:27PM +0200, Matthias Brugger wrote:

quoted

2015-04-27 8:49 GMT+02:00 Sascha Hauer [off-list ref]:

quoted

The pm_runtime callbacks already enable and disable the device.
Use them in probe() and remove() instead of duplicating the
code. This allows us to concentrate more code for enabling/disabling
the UART in a single place.

Signed-off-by: Sascha Hauer <redacted>
---
 drivers/tty/serial/8250/8250_mtk.c | 69 ++++++++++++++++++++------------------
 1 file changed, 36 insertions(+), 33 deletions(-)

diff --git a/drivers/tty/serial/8250/8250_mtk.c b/drivers/tty/serial/8250/8250_mtk.c
index bcfaa8dc..2f28bd0 100644
--- a/drivers/tty/serial/8250/8250_mtk.c
+++ b/drivers/tty/serial/8250/8250_mtk.c

@@ -115,6 +115,29 @@ mtk8250_set_termios(struct uart_port *port, struct ktermios *termios,
                tty_termios_encode_baud_rate(termios, baud, baud);
 }

+static int mtk8250_runtime_suspend(struct device *dev)
+{
+       struct mtk8250_data *data = dev_get_drvdata(dev);
+
+       clk_disable_unprepare(data->uart_clk);
+
+       return 0;
+}
+
+static int mtk8250_runtime_resume(struct device *dev)
+{
+       struct mtk8250_data *data = dev_get_drvdata(dev);
+       int err;
+
+       err = clk_prepare_enable(data->uart_clk);
+       if (err) {
+               dev_warn(dev, "Can't enable clock\n");
+               return err;
+       }
+
+       return 0;
+}
+
 static void
 mtk8250_do_pm(struct uart_port *port, unsigned int state, unsigned int old)
 {

@@ -130,19 +153,12 @@ mtk8250_do_pm(struct uart_port *port, unsigned int state, unsigned int old)
 static int mtk8250_probe_of(struct platform_device *pdev, struct uart_port *p,
                           struct mtk8250_data *data)
 {
-       int err;
-
        data->uart_clk = devm_clk_get(&pdev->dev, NULL);
        if (IS_ERR(data->uart_clk)) {
                dev_warn(&pdev->dev, "Can't get uart clock\n");
                return PTR_ERR(data->uart_clk);
        }

-       err = clk_prepare_enable(data->uart_clk);
-       if (err) {
-               dev_warn(&pdev->dev, "Can't prepare clock\n");
-               return err;
-       }
        p->uartclk = clk_get_rate(data->uart_clk);

        return 0;

@@ -193,14 +209,18 @@ static int mtk8250_probe(struct platform_device *pdev)
        writel(0x0, uart.port.membase +
                        (MTK_UART_RATE_FIX << uart.port.regshift));

-       data->line = serial8250_register_8250_port(&uart);
-       if (data->line < 0)
-               return data->line;
-
        platform_set_drvdata(pdev, data);

-       pm_runtime_set_active(&pdev->dev);
        pm_runtime_enable(&pdev->dev);
+       if (!pm_runtime_enabled(&pdev->dev)) {
+               err = mtk8250_runtime_resume(&pdev->dev);
+               if (err)
+                       return err;
+       }
+
+       data->line = serial8250_register_8250_port(&uart);
+       if (data->line < 0)
+               return data->line;

Why do you delete pm_runtime_set_active here?

pm_runtime_set_active() communicates the initial state of the device to
the pm core. Previously the driver activated the device manually by
directly calling clk_prepare_enable(). Since this manual enabling of
the clock is removed in this patch, the device is no longer active,
hence pm_runtime_set_active() has to be removed.

Sounds resonable.
Acked-by: Matthias Brugger <redacted>

-- 
motzblog.wordpress.com