Age | Commit message (Collapse) | Author | Files | Lines |
|
in finding compatibility there are 3 steps how to find
out whether the provided device driver supports currently
inserted device.
1. compare fields in the driver info structure
If there is a match (all CHECK_PARTIAL_COMPATIBILITY() macros return true),
we proceed to step 2.
If all device drivers are searched, but none is compatible ->
that means no device driver is available for currently inserted device.
2. try to call driver's init function.
If it return non-null pointer to data we may proceed to the step 3.
Otherwise, we continue in the loop handling step 1.
3. call analyze descriptor for all descriptors. When it returns true,
it means success and that the driver supports current device. From now on
poll function is allowed to be called.
If all descriptors were provided to the analyze_descriptor method and
it still returns false, it means device driver is not initialized and
should not be used with this device.
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
out: const void *
in: void *
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Leave the behaviour as before: Default: FullSpeed
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
+ merge fixes
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
+ added logging output of data that is going to be sent.
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
it was always 0 - control transfer...
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
packet_size_max0 was set only if it was 8. It was not working for 16 or 32.
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
* fixed typo/bug: "hub->state += 2" must have been
"hub->state = 2" to issue empty read
Now, after each SET_ or CLEAR_ command, empty read is issued.
+ removed forward declaration of event() - cleaning
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
TIM6 has clock running at 84MHz instead of 168MHz,
so set the prescaler accordingly.
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
forward declarations are not needed
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
This way, we don't need forward declarations
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
this was causing infinite loop, when size of the
data was not multiple of 4.
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
fixing after enum STATES refactor.
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Overflows in 6.5536 seconds
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
added comment to usbh_hubbed.h
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
Instead of 2 argument call, use function with one argument.
It is more clear, and safer for inexperienced users with libopencm3.
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|
|
First public version, date: 1.4.2015
Signed-off-by: Amir Hammad <amir.hammad@hotmail.com>
|