- Add tud_audio_n_get_ep_out_ff(), tud_audio_n_get_ep_in_ff(),
tud_audio_n_get_rx_support_ff(), and tud_audio_n_get_tx_support_ff()
- Change get_linear_read/write_info() to return linear and wrapped part
at once
- Adjusted affected code in audio_device.c and tested with
audio_4_channel.
Returns the RT driver to the function state of previous iteration, which
did not support the will_detach. Behavior should be fine without this
feature. This removes much of the added bloat to track state, and
handle requests in the APP_DETACH state which is no longer required.
Removes the optional bloat added to the RT driver, such as responding to
GETSTATE requests.
Fixes the DFU Mode to extract the attr bits from the functional
descriptor when opened.
Fixes some incorrect bitwise if checks.
Also, updates some naming of functions to be consistent with the rest of
the library.
With the runtime and mode portions in separate classes, a single
application should only be building with one or the other enabled. In
some applications both might be desired at build time.
The CFG_TUD_DFU_RUNTIME_AND_MODE option creates a DFU class, which asks
the application which mode to initialize to. This allows a runtime
change between RT and DFU mode, by just reinitializing tusb.
Fixes bug where the app callback was getting the length of the status
request transfer rather than the length of the data stage payload.
TODO: Right now this returns the expected length, when it really should
be returning the transfer length.
- Audio format and parameters are parsed from descriptors thus user no
longer needs to give them explicitely
- Tested for 4 channel software type I PCM encoding with 16 bit with 1
channel per FIFO and 2 channels per FIFO (this is I2S specific)
Reinhard Panhuber