[alsa-devel] [PATCH] ASoC: fsl-sai: Add device tree nodes and its availiable check
SAIs that are not connected on the board should have a status = "disabled" property in their device tree nodes.
Signed-off-by: Xiubo Li Li.Xiubo@freescale.com --- sound/soc/fsl/fsl_sai.c | 3 +++ 1 file changed, 3 insertions(+)
diff --git a/sound/soc/fsl/fsl_sai.c b/sound/soc/fsl/fsl_sai.c index cdd3fa8..6b2ab91 100644 --- a/sound/soc/fsl/fsl_sai.c +++ b/sound/soc/fsl/fsl_sai.c @@ -404,6 +404,9 @@ static int fsl_sai_probe(struct platform_device *pdev) struct resource *res; int ret;
+ if (!np || !of_device_is_available(np)) + return -ENODEV; + sai = devm_kzalloc(&pdev->dev, sizeof(*sai), GFP_KERNEL); if (!sai) return -ENOMEM;
On Fri, Jan 10, 2014 at 10:05:50AM +0800, Xiubo Li wrote:
- if (!np || !of_device_is_available(np))
return -ENODEV;
I would expect the of_device_is_available() check to be done by the driver core rather than by individual drivers - every single driver should have that check. Is this not happening? The check for np is OK though.
- if (!np || !of_device_is_available(np))
return -ENODEV;I would expect the of_device_is_available() check to be done by the driver core rather than by individual drivers - every single driver should have that check. Is this not happening? The check for np is OK though.
Yes, I hasn't found the core driver does that for now...
And IMO this is needed since the SAI driver is only base dts. And maybe the device is not presently operational, but it might become operational in the future (for example, something is not plugged in, or switched off). Or A serious error was detected in the device, and it is unlikely to become operational without repair...
And for now for the individual drivers, before calling the of_device_is_available the '!np' check is need, because: While in __of_device_is_available:
status = of_get_property(device, "status", &statlen); if (status == NULL) return 1;
The status value returned from 'of_get_property()' will be NULL in two cases: Firstly: the 'device' value (device node) is NULL. Secondly: the 'status' property is actaully not exist.
If the device node is NULL, the __of_device_is_available will return true too, that will mean the absent state of the 'status' property.
Thanks,
-- Best Regards, Xiubo
On Mon, Jan 13, 2014 at 06:15:24AM +0000, Li.Xiubo@freescale.com wrote:
I would expect the of_device_is_available() check to be done by the driver core rather than by individual drivers - every single driver should have that check. Is this not happening? The check for np is OK though.
Yes, I hasn't found the core driver does that for now...
Are you sure, we actively use the ability to mark devices as unavailable on some systems?
And IMO this is needed since the SAI driver is only base dts. And maybe the device is not presently operational, but it might become operational in the future (for example, something is not plugged in, or switched off). Or A serious error was detected in the device, and it is unlikely to become operational without repair...
I'm sorry but I don't understand what you're saying here at all. These are all things that might happen but I don't see the connection between them and the benefits of implementing things in the core - surely the more functionality there is the more useful it's going to be to factor all the code out into the driver core?
participants (3)
-
Li.Xiubo＠freescale.com -
Mark Brown -
Xiubo Li