Requested revision
Standard: | IEEE Std 802.1AS-2020 | Clause: | 11.2.19.3.4 |
Clause title: | computePropTime() |
Rationale for revision
The submitter would like to acknowledge Jingfei Lv and Silvana Rodrigues for bringing this item to the submitter’s attention.
The computation of the quantities t2 and t3 (the “2” and “3” are subscripts)in 11.2.19.3.4 is incorrect for the case of Common Mean Link Delay Service (CMLDS). This results in an incorrect value for the meanLinkDelay, D, when CMLDS is used.
Proposed text
Replace the text that describes the computation of t2 by:
t2 is the sum of (1) the ns field of the requestReceiptTimestamp, (2) the seconds field of the requestReceiptTimestamp multiplied by 109, and (3) the correctionField multiplied by s (see 11.2.19.2.13) and then divided by 216 (i.e., the correctionField is expressed in nanoseconds plus fractional nanoseconds),of the Pdelay_Resp message received in response to the Pdelay_Req message sent by the MD entity, plus (4) delayAsymmetry*neighborRateRatio if this state machine is invoked by CMLDS.
Replace the text that describes the computation of t3 by:
t3 is the sum of (1) the ns field of the responseOriginTimestamp, (2) the seconds field of the responseOriginTimestamp multiplied by 109, and (3) the correctionField divided by 216 (i.e., the correctionField is expressed in nanoseconds plus fractional nanoseconds), of the Pdelay_Resp_Follow_Up message received in response to the Pdelay_Req message sent by the MD entity, minus (4) delayAsymetry*(neighborRateRatio-1) if this state machine is invoked by CMLDS.
Impact on existing networks
There is no impact on existing networks, other than that the value of meanLinkDelay in the case where CMLDS is used will be correct.
Originator
Name: | Geoffrey M. Garner | Email: | gmgarner@alum.mit.edu |
Affiliation: | Huawei | ||
Submitted: | 2021-07-07 |