[dpdk-dev] [PATCH 1/3] mbuf: embedding timestamp into the packet

[Oleg Kuporosov]

The hard requirement of financial services industry is accurate
timestamping aligned with the packet itself. This patch is to satisfy
this requirement:

- include uint64_t timestamp field into rte_mbuf with minimal impact to
  throughput/latency. Keep it just simple uint64_t in ns (more than 580
  years) would be enough for immediate needs while using full
  struct timespec with twice bigger size would have much stronger
  performance impact as missed cacheline0.

- it is possible as there is 6-bytes gap in 1st cacheline (fast path)
  and moving uint16_t vlan_tci_outer field to 2nd cacheline.

- such move will only impact for pretty rare usable VLAN RX stripping
  mode for outer TCI (it used only for one NIC i40e from the whole set and
  allows to keep minimal performance impact for RX/TX timestamps.

Signed-off-by: Oleg Kuporosov <olegk at mellanox.com>
---
 lib/librte_mbuf/rte_mbuf.h | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h
index 23b7bf8..1e1f2ed 100644
--- a/lib/librte_mbuf/rte_mbuf.h
+++ b/lib/librte_mbuf/rte_mbuf.h
@@ -851,8 +851,7 @@ struct rte_mbuf {
 
 	uint32_t seqn; /**< Sequence number. See also rte_reorder_insert() */
 
-	/** Outer VLAN TCI (CPU order), valid if PKT_RX_QINQ_STRIPPED is set. */
-	uint16_t vlan_tci_outer;
+	uint64_t timestamp;       /**< Packet's timestamp, usually in ns */
 
 	/* second cache line - fields only used in slow path or on TX */
 	MARKER cacheline1 __rte_cache_min_aligned;
@@ -885,6 +884,9 @@ struct rte_mbuf {
 		};
 	};
 
+	/** Outer VLAN TCI (CPU order), valid if PKT_RX_QINQ_STRIPPED is set. */
+	uint16_t vlan_tci_outer;
+
 	/** Size of the application private data. In case of an indirect
 	 * mbuf, it stores the direct mbuf private data size. */
 	uint16_t priv_size;
-- 
1.8.3.1