[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20120625070614.GA13876@core.coreip.homeip.net>
Date: Mon, 25 Jun 2012 00:06:14 -0700
From: Dmitry Torokhov <dmitry.torokhov@...il.com>
To: Roland Stigge <stigge@...com.de>
Cc: axel.lin@...il.com, riyer@...dia.com, michael.hennerich@...log.com,
grant.likely@...retlab.ca, linux-input@...r.kernel.org,
linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
kevin.wells@....com, srinivas.bakki@....com,
devicetree-discuss@...ts.ozlabs.org, rob.herring@...xeda.com,
aletes.xgr@...il.com
Subject: Re: [PATCH v6] input: keyboard: Add keys driver for the LPC32xx SoC
HI Roland,
On Sun, Jun 24, 2012 at 04:44:38PM +0200, Roland Stigge wrote:
> This patch adds a driver for the key scan interface of the LPC32xx SoC
>
> Signed-off-by: Roland Stigge <stigge@...com.de>
>
This looks _very_ good, I just have a couple more comments:
> +
> +static int lpc32xx_kscan_open(struct input_dev *dev)
> +{
> + struct lpc32xx_kscan_drv *kscandat = input_get_drvdata(dev);
> +
> + return clk_prepare_enable(kscandat->clk);
Don't you need to clear IRQ here, the same way you do it in resume?
> +}
> +
> +static void lpc32xx_kscan_close(struct input_dev *dev)
> +{
> + struct lpc32xx_kscan_drv *kscandat = input_get_drvdata(dev);
> +
> + clk_disable_unprepare(kscandat->clk);
And same here.
> +}
> +
> +int lpc32xx_parse_dt(struct platform_device *pdev)
Static? Also why don't you pass kscandat instead of platform device?
> +{
> + struct lpc32xx_kscan_drv *kscandat = platform_get_drvdata(pdev);
> + struct device_node *np = pdev->dev.of_node;
> + u32 rows, columns;
> +
> + of_property_read_u32(np, "keypad,num-rows", &rows);
> + of_property_read_u32(np, "keypad,num-columns", &columns);
> + if (!rows || rows != columns) {
> + dev_err(&pdev->dev,
> + "rows and columns must be specified and be equal!\n");
> + return -EINVAL;
> + }
> +
> + kscandat->matrix_sz = rows;
> + kscandat->row_shift = get_count_order(columns);
> +
> + of_property_read_u32(np, "nxp,debounce-delay-ms", &kscandat->deb_clks);
> + of_property_read_u32(np, "nxp,scan-delay-ms", &kscandat->scan_delay);
> + if (!kscandat->deb_clks || !kscandat->scan_delay) {
> + dev_err(&pdev->dev, "debounce or scan delay not specified\n");
> + return -ENXIO;
> + }
> +
> + kscandat->keymap = kzalloc(sizeof(kscandat->keymap[0]) *
> + (rows << kscandat->row_shift), GFP_KERNEL);
> + if (!kscandat->keymap) {
> + dev_err(&pdev->dev, "could not allocate memory for keymap\n");
> + return -ENOMEM;
> + }
I think this allocation belongs to the probe().
> +
> + return 0;
> +}
> +
> +static int __devinit lpc32xx_kscan_probe(struct platform_device *pdev)
> +{
> + struct lpc32xx_kscan_drv *kscandat;
> + struct resource *res;
> + int error;
> + int irq;
> +
> + kscandat = kzalloc(sizeof(struct lpc32xx_kscan_drv), GFP_KERNEL);
> + if (!kscandat) {
> + dev_err(&pdev->dev, "failed to allocate memory\n");
> + return -ENOMEM;
> + }
> +
> + res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> + if (!res) {
> + dev_err(&pdev->dev, "failed to get platform I/O memory\n");
> + error = -EBUSY;
> + goto out1;
> + }
> +
> + res = request_mem_region(res->start, resource_size(res), pdev->name);
> + if (!res) {
> + dev_err(&pdev->dev, "failed to request I/O memory\n");
> + error = -EBUSY;
> + goto out1;
> + }
> + kscandat->io_p_start = res->start;
> + kscandat->io_p_size = resource_size(res);
> +
> + kscandat->kscan_base = ioremap(res->start, resource_size(res));
> + if (!kscandat->kscan_base) {
> + dev_err(&pdev->dev, "failed to remap I/O memory\n");
> + error = -EBUSY;
> + goto out2;
> + }
> +
> + /* Get the key scanner clock */
> + kscandat->clk = clk_get(&pdev->dev, NULL);
> + if (IS_ERR(kscandat->clk)) {
> + dev_err(&pdev->dev, "failed to get clock\n");
> + error = PTR_ERR(kscandat->clk);
> + goto out3;
> + }
> + clk_enable(kscandat->clk);
Why not clk_prepare_enable()? Also I think you need to move it down, to
where you actually write to the device's registers.
> +
> + irq = platform_get_irq(pdev, 0);
> + if ((irq < 0) || (irq >= NR_IRQS)) {
> + dev_err(&pdev->dev, "failed to get platform irq\n");
> + error = -EINVAL;
> + goto out4;
> + }
> + error = request_irq(irq, lpc32xx_kscan_irq, 0, pdev->name, kscandat);
> + if (error) {
> + dev_err(&pdev->dev, "failed to request irq\n");
> + goto out4;
> + }
> +
> + kscandat->input = input_allocate_device();
> + if (kscandat->input == NULL) {
> + dev_err(&pdev->dev, "failed to allocate device\n");
> + error = -ENOMEM;
> + goto out5;
> + }
This might be too late - IRQ is alreday registered and clock is enabled
- what is to stop interrupts to be delivered? I'd allocate the input
device together with kscandat structure.
> +
> + platform_set_drvdata(pdev, kscandat);
> +
> + error = lpc32xx_parse_dt(pdev);
> + if (error) {
> + dev_err(&pdev->dev, "failed to parse device tree\n");
> + goto out6;
> + }
> +
> + /* Setup key input */
> + kscandat->input->evbit[0] = BIT_MASK(EV_KEY);
> + kscandat->input->name = pdev->name;
> + kscandat->input->phys = "matrix-keys/input0";
> + kscandat->input->id.vendor = 0x0001;
> + kscandat->input->id.product = 0x0001;
> + kscandat->input->id.version = 0x0100;
> + kscandat->input->open = lpc32xx_kscan_open;
> + kscandat->input->close = lpc32xx_kscan_close;
> + kscandat->input->dev.parent = &pdev->dev;
> +
> + error = matrix_keypad_build_keymap(NULL, NULL, kscandat->matrix_sz,
> + kscandat->matrix_sz,
> + kscandat->keymap, kscandat->input);
> + if (error) {
> + dev_err(&pdev->dev, "failed to build keymap\n");
> + goto out6;
> + }
> +
> + input_set_drvdata(kscandat->input, kscandat);
> + input_set_capability(kscandat->input, EV_MSC, MSC_SCAN);
> +
> + error = input_register_device(kscandat->input);
> + if (error) {
> + dev_err(&pdev->dev, "failed to register input device\n");
> + goto out6;
> + }
> +
I think this needs to be done before you register input device as as
soon as it is registered it may be used.
> + /* Configure the key scanner */
> + writel(kscandat->deb_clks, LPC32XX_KS_DEB(kscandat->kscan_base));
> + writel(kscandat->scan_delay, LPC32XX_KS_SCAN_CTL(kscandat->kscan_base));
> + writel(LPC32XX_KSCAN_FTST_USE32K_CLK,
> + LPC32XX_KS_FAST_TST(kscandat->kscan_base));
> + writel(kscandat->matrix_sz,
> + LPC32XX_KS_MATRIX_DIM(kscandat->kscan_base));
> + writel(1, LPC32XX_KS_IRQ(kscandat->kscan_base));
> + clk_disable(kscandat->clk);
clk_disable_unprepare()?
> + return 0;
> +
> +out6:
> + input_free_device(kscandat->input);
> +out5:
> + free_irq(irq, pdev);
> +out4:
> + clk_disable(kscandat->clk);
> + clk_put(kscandat->clk);
> +out3:
> + iounmap(kscandat->kscan_base);
> +out2:
> + release_mem_region(kscandat->io_p_start, kscandat->io_p_size);
> +out1:
> + kfree(kscandat);
> +
> + return error;
> +}
> +
> +static int __devexit lpc32xx_kscan_remove(struct platform_device *pdev)
> +{
> + struct lpc32xx_kscan_drv *kscandat = platform_get_drvdata(pdev);
> +
> + input_unregister_device(kscandat->input);
Are we certain it is safe to free input device before freeing IRQ? I see
we have close(), still I'd feel safer if you swapper unregister and
free_irq().
> + free_irq(platform_get_irq(pdev, 0), pdev);
> + clk_put(kscandat->clk);
> + iounmap(kscandat->kscan_base);
> + release_mem_region(kscandat->io_p_start, kscandat->io_p_size);
> + kfree(kscandat);
> +
> + return 0;
> +}
> +
> +#ifdef CONFIG_PM_SLEEP
> +static int lpc32xx_kscan_suspend(struct device *dev)
> +{
> + struct platform_device *pdev = to_platform_device(dev);
> + struct lpc32xx_kscan_drv *kscandat = platform_get_drvdata(pdev);
> +
> + /* Clear IRQ and disable clock */
> + writel(1, LPC32XX_KS_IRQ(kscandat->kscan_base));
> + clk_disable_unprepare(kscandat->clk);
> +
This may race with open()/close(). You need to take input_dev->mutex and
avoid touching the device if it sis not opened (users == 0).
> + return 0;
> +}
> +
> +static int lpc32xx_kscan_resume(struct device *dev)
> +{
> + struct platform_device *pdev = to_platform_device(dev);
> + struct lpc32xx_kscan_drv *kscandat = platform_get_drvdata(pdev);
> +
> + /* Enable clock and clear IRQ */
> + clk_prepare_enable(kscandat->clk);
> + writel(1, LPC32XX_KS_IRQ(kscandat->kscan_base));
> +
Same here.
Thanks!
--
Dmitry
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists