> This usually indicates the driver is not sending the Tx timestamp back > to the stack, especially if you increase it past a few milliseconds. >
Yes, that is what I have read, but the driver looks correct to me, and as previously mentioned I've put debug statements into the code path (here: https://github.com/torvalds/linux/blob/master/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c#L738) that show to me that this path is getting executed. I would like to know why ptp4l thinks the driver is not doing this, or if the driver code linked there is insufficient somehow. > > But skb_tstamp_tx is for software timestamping, if I recall correctly. > > > Try using the software timestamping option for ptp4l, and see if that > works first. (Should be -S) > Getting software timestamping working is just my backup plan if hardware doesn't work, but adding -S also gives the same ptp4l output. Thanks, Matthew _______________________________________________ Linuxptp-users mailing list Linuxptp-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linuxptp-users
