Length/Type Field Error Checks - 9.0 English

Enabled

The default operation is with the length/type error checking enabled. In this mode, the following checks are made on all frames received. If either of these checks fails, the frame is marked as bad.

• A value in the length/type field that is ≥ decimal 46 but less than decimal 1,536 (a Length interpretation) is checked against the actual data length received.
• A value in the length/type field that is < decimal 46 is checked to see that the data field is padded to exactly 46 bytes (so that the resultant frame is minimum frame size: 64 bytes total in length).

Furthermore, if padding is indicated (the length/type field is less than decimal 46) and User-Supplied FCS Passing is disabled, the length value in the length/type field is used to de-assert rx_axis_mac_tvalid after the indicated number of data bytes that removes the padding bytes from the frame.

Disabled

When the length/type error checking is disabled and the length/type field has a length interpretation, the MAC does not check the length value against the actual data length received. A frame containing only this error is marked as good. However, if the length/type field is less than decimal 46, the MAC marks a frame as bad if it is not the minimum frame size of 64 bytes. Furthermore, if padding is indicated and User-Supplied FCS Passing is disabled, a length value in the length/type field is not used to de-assert rx_axis_mac_tvalid . Instead, rx_axis_mac_tvalid is de-asserted before the start of the FCS field. In this way, no padding is removed from the frame.