@@ -86,15 +86,21 @@ static int amba_read_periphid(struct amba_device *adev)
iounmap(tmp);
+ if (r < 0)
+ return r;
+
return adev->periphid ? 0 : -ENODEV;
}
static const struct amba_id *
amba_lookup(const struct amba_id *table, struct amba_device *dev)
{
- int ret = 0;
+ int ret;
- if (amba_read_periphid(dev))
+ ret = amba_read_periphid(dev);
+ if (ret < 0)
+ return ERR_PTR(ret);
+ if (ret)
return NULL;
while (table->mask) {
@@ -111,12 +117,17 @@ static int amba_match(struct device *dev, struct device_driver *drv)
{
struct amba_device *pcdev = to_amba_device(dev);
struct amba_driver *pcdrv = to_amba_driver(drv);
+ const struct amba_id *id;
/* When driver_override is set, only bind to the matching driver */
if (pcdev->driver_override)
return !strcmp(pcdev->driver_override, drv->name);
- return amba_lookup(pcdrv->id_table, pcdev) != NULL;
+ id = amba_lookup(pcdrv->id_table, pcdev);
+ if (IS_ERR(id))
+ return PTR_ERR(id);
+
+ return id != NULL;
}
static int amba_uevent(struct device *dev, struct kobj_uevent_env *env)
Currently, when we scan the bus for matching devices, we assume that the clock controller is probed and available, but that might be not always true. If the bus clock is unavailable, we just silently fail to detect devices. Handle this by propagating the errors up to the driver core, so that it can retry the scan later. Signed-off-by: Georgi Djakov <georgi.djakov@linaro.org> --- drivers/amba/bus.c | 17 ++++++++++++++--- 1 file changed, 14 insertions(+), 3 deletions(-)