| Message ID | 20150116080722.GA7813@dtor-ws |
|---|---|
| State | Accepted |
| Headers | show |
On Fri, Jan 16, 2015 at 12:07:22AM -0800, Dmitry Torokhov wrote: > Trying to register an I2C device asynchronously (via async_schedule() call) > results in an ugly warning from request_module() warning about potential > deadlock (because request_module tries to wait for async works to > complete). While we could try to switch to request_module_nowait(), other > buses, as well as I2C itself when not using device tree, do not try to load > modules, but rather rely on the standard infrastructure (udev) to execute > module loading, and we should be doing the same. > > Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com> Sounds reasonable. Applied to for-next, thanks!
diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c index 39d25a8..706caea 100644 --- a/drivers/i2c/i2c-core.c +++ b/drivers/i2c/i2c-core.c @@ -1419,8 +1419,6 @@ static struct i2c_client *of_i2c_register_device(struct i2c_adapter *adap, if (of_get_property(node, "wakeup-source", NULL)) info.flags |= I2C_CLIENT_WAKE; - request_module("%s%s", I2C_MODULE_PREFIX, info.type); - result = i2c_new_device(adap, &info); if (result == NULL) { dev_err(&adap->dev, "of_i2c: Failure registering %s\n",
Trying to register an I2C device asynchronously (via async_schedule() call) results in an ugly warning from request_module() warning about potential deadlock (because request_module tries to wait for async works to complete). While we could try to switch to request_module_nowait(), other buses, as well as I2C itself when not using device tree, do not try to load modules, but rather rely on the standard infrastructure (udev) to execute module loading, and we should be doing the same. Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com> --- drivers/i2c/i2c-core.c | 2 -- 1 file changed, 2 deletions(-)