On 04/29/16 09:23, Maxim Uvarov wrote:
On 04/29/16 06:35, Bill Fischofer wrote:
+/* Note that this implementation does not including any code or
conditionally
+ * modified code that is endian specific, yet it works equally well
on BIG or
+ * LITTLE endian machines. The reason that this works is primarily
because
+ * a 16-bit one's complement sum happens to be "endian-agnostic".
Specifically
+ * if one does a sum of 16-bit pkt values on a big endian machine
and then on
+ * a little endian machine, they will not agree. But after turning
it into
+ * a one's complement sum by adding the carry bits in and truncating to
+ * 16-bits (which may need to be done more than once), the final
16-bit results
+ * will be byte-swapped versions of the other. Then after storing
the result
+ * back into the pkt (as a 16-bit value), the final byte pattern
will be
+ * identical for both machines. */
+
+int odph_udp_tcp_chksum(odp_packet_t odp_pkt,
+ odph_chksum_op_t op,
+ uint16_t *chksum_ptr)
That is not doxygen function description. Needed to check if you broke
doxygen
validation for helpers.
Maxim.
Oh, because that is not and separated by empty line. So everything is
good with it.
Maxim.
_______________________________________________
lng-odp mailing list
[email protected]
https://lists.linaro.org/mailman/listinfo/lng-odp
