Re: [PATCH] [media] coda: Improve runtime PM support

From: Philipp Zabel
Date: Thu Sep 11 2014 - 13:43:28 EST


Hi Ulf,

thanks for the patch!

Am Mittwoch, den 10.09.2014, 15:43 +0200 schrieb Ulf Hansson:
> For several reasons it's good practice to leave devices in runtime PM
> active state while those have been probed.

It would be nice to mention those reasons.

> In this cases we also want to prevent the device from going inactive,
> until the firmware has been completely installed, especially when using
> a PM domain.
>
> Signed-off-by: Ulf Hansson <ulf.hansson@xxxxxxxxxx>
> ---
> drivers/media/platform/coda/coda-common.c | 42 ++++++++-----------------------
> 1 file changed, 11 insertions(+), 31 deletions(-)
>
> diff --git a/drivers/media/platform/coda/coda-common.c b/drivers/media/platform/coda/coda-common.c
> index 0997b5c..361f28d 100644
> --- a/drivers/media/platform/coda/coda-common.c
> +++ b/drivers/media/platform/coda/coda-common.c
> @@ -1703,39 +1703,16 @@ static void coda_fw_callback(const struct firmware *fw, void *context)
> memcpy(dev->codebuf.vaddr, fw->data, fw->size);
> release_firmware(fw);
>
> - if (pm_runtime_enabled(&pdev->dev) && pdev->dev.pm_domain) {
> - /*
> - * Enabling power temporarily will cause coda_hw_init to be
> - * called via coda_runtime_resume by the pm domain.
> - */
> - ret = pm_runtime_get_sync(&dev->plat_dev->dev);
> - if (ret < 0) {
> - v4l2_err(&dev->v4l2_dev, "failed to power on: %d\n",
> - ret);
> - return;
> - }
> -
> - ret = coda_check_firmware(dev);
> - if (ret < 0)
> - return;
> -
> - pm_runtime_put_sync(&dev->plat_dev->dev);
> - } else {
> - /*
> - * If runtime pm is disabled or pm_domain is not set,
> - * initialize once manually.
> - */
> - ret = coda_hw_init(dev);
> - if (ret < 0) {
> - v4l2_err(&dev->v4l2_dev, "HW initialization failed\n");
> - return;
> - }
> -
> - ret = coda_check_firmware(dev);
> - if (ret < 0)
> - return;
> + ret = coda_hw_init(dev);
> + if (ret < 0) {
> + v4l2_err(&dev->v4l2_dev, "HW initialization failed\n");
> + return;
> }
>
> + ret = coda_check_firmware(dev);
> + if (ret < 0)
> + return;
> +

Won't this keep the PM domain active indefinitely if hw_init or
check_firmware fails? Other than that, this is a great change.

> dev->alloc_ctx = vb2_dma_contig_init_ctx(&pdev->dev);
> if (IS_ERR(dev->alloc_ctx)) {
> v4l2_err(&dev->v4l2_dev, "Failed to alloc vb2 context\n");
> @@ -1771,6 +1748,7 @@ static void coda_fw_callback(const struct firmware *fw, void *context)
> v4l2_info(&dev->v4l2_dev, "codec registered as /dev/video[%d-%d]\n",
> dev->vfd[0].num, dev->vfd[1].num);
>
> + pm_runtime_put_sync(&pdev->dev);
> return;
>
> rel_m2m:
> @@ -1998,6 +1976,8 @@ static int coda_probe(struct platform_device *pdev)
>
> platform_set_drvdata(pdev, dev);
>
> + pm_runtime_get_noresume(&pdev->dev);
> + pm_runtime_set_active(&pdev->dev);

Should we have a comment here why this is necessary?

regards
Philipp

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/