nfc.c: reformulate comments of prev commit in chip-agnostic way. BTW there is more to do...

This commit is contained in:
Philippe Teuwen 2011-05-12 20:51:44 +00:00
parent e5601e562f
commit ff7696fffb

View file

@ -561,7 +561,9 @@ nfc_initiator_transceive_bits (nfc_device_t * pnd, const byte_t * pbtTx, const s
* - Overall communication with the host is heavier and slower.
*
* Timer control:
* By default timer can count up to 65535 cycles, so about 4.8ms, with a precision of about 73ns.
* By default timer configuration tries to maximize the precision, which also limits the maximum
* cycles count before saturation/timeout.
* E.g. with PN53x it can count up to 65535 cycles, so about 4.8ms, with a precision of about 73ns.
* - If you're ok with the defaults, set *cycles = 0 before calling this function.
* - If you need to count more cycles, set *cycles to the maximum you expect but don't forget
* you'll loose in precision and it'll take more time before timeout, so don't abuse!
@ -586,7 +588,9 @@ nfc_initiator_transceive_bytes_timed (nfc_device_t * pnd, const byte_t * pbtTx,
* - Overall communication with the host is heavier and slower.
*
* Timer control:
* By default timer can count up to 65535 cycles, so about 4.8ms, with a precision of about 73ns.
* By default timer configuration tries to maximize the precision, which also limits the maximum
* cycles count before saturation/timeout.
* E.g. with PN53x it can count up to 65535 cycles, so about 4.8ms, with a precision of about 73ns.
* - If you're ok with the defaults, set *cycles = 0 before calling this function.
* - If you need to count more cycles, set *cycles to the maximum you expect but don't forget
* you'll loose in precision and it'll take more time before timeout, so don't abuse!