On Fri, 13 Oct 2023 20:56:01 +0300 Vladimir Oltean wrote: > > > I'm not really sure what you're referring to, with SOF_..._DMA. > > > The DMA, if presented as a PHC as I am proposing, would play the role of > > > the hardware timestamp provider (think SOF_TIMESTAMPING_TX_HARDWARE | > > > SOF_TIMESTAMPING_RX_HARDWARE), so there will be no driver-visible > > > special socket option flags for DMA timestamping. > > > > Each packet may want different timestamp tho, especially on Tx it > > should be fairly easy for socket to request to get "real" MAC stamps, > > while most get cheaper DMA stamps. Currently some drivers run flow > > matching to find PTP packets and automatically give them better quality > > timestamps :( > > > > Even if at the config level we use PHCs we need to translate that into > > some SKBTX_* bit, don't we? > > I think Richard had something to say about that being wishful thinking: > https://lore.kernel.org/netdev/ZGw46hrpiqCVNeXS@xxxxxxxxxxxxxxxxxx/ 🤷️ > On RX I'm not sure how you'd know in advance if the packet is going to > be routed to a socket that wants DMA or MAC timestamps. And having a > socket with hardware timestamps from one provider in one direction, and > another provider in the other direction, is.... not sane as a kernel API? For DC NICs all the timestamps are based on the same PHC. The use case is to get MAC/precise stamps for PTP and DMA/rough stamps for TCP CC. Agreed that stamps from different PHCs in different directions would be weird. Thinking about it again we want the ability to configure the rx filter per stamping point. MAC stamps PTP frames and DMA stamps all the rest. For Tx we need to know whether app wanted DMA or MAC stamp (more or less whether it's PTP again, without running a classifier on the packet).
