From: Martin Spinler <[email protected]>

The resulting timestamp wasn't monotonically increasing value.

Now the timestamp is value in nanoseconds (Unix epoch).

Unfortunately there's currently no safe mechanism in nfb-framework
to get ticks from hardware to implement read_clock function.

Signed-off-by: Martin Spinler <[email protected]>
---
 doc/guides/nics/nfb.rst  |  6 ++----
 drivers/net/nfb/nfb_rx.h | 13 +++++++------
 2 files changed, 9 insertions(+), 10 deletions(-)

diff --git a/doc/guides/nics/nfb.rst b/doc/guides/nics/nfb.rst
index 96aaf64440..a9b4049654 100644
--- a/doc/guides/nics/nfb.rst
+++ b/doc/guides/nics/nfb.rst
@@ -64,10 +64,8 @@ products). The standard `RTE_ETH_RX_OFFLOAD_TIMESTAMP` flag 
can be used for this
 When the timestamps are enabled, a timestamp validity flag is set in the MBUFs
 containing received frames and timestamp is inserted into the `rte_mbuf` 
struct.
 
-The timestamp is an `uint64_t` field. Its lower 32 bits represent *seconds* 
portion of the timestamp
-(number of seconds elapsed since 1.1.1970 00:00:00 UTC) and its higher 32 bits 
represent
-*nanosecond* portion of the timestamp (number of nanoseconds elapsed since the 
beginning of the
-second in the *seconds* portion.
+The timestamp is an `uint64_t` field and holds the number of nanoseconds
+elapsed since 1.1.1970 00:00:00 UTC.
 
 
 Using the NFB PMD
diff --git a/drivers/net/nfb/nfb_rx.h b/drivers/net/nfb/nfb_rx.h
index 12769841ae..67b3b00e2a 100644
--- a/drivers/net/nfb/nfb_rx.h
+++ b/drivers/net/nfb/nfb_rx.h
@@ -13,6 +13,7 @@
 #include <rte_mbuf.h>
 #include <rte_mbuf_dyn.h>
 #include <rte_ethdev.h>
+#include <rte_time.h>
 
 #include "nfb.h"
 
@@ -202,15 +203,15 @@ nfb_eth_ndp_rx(void *queue,
                        if (nfb_timestamp_dynfield_offset >= 0) {
                                rte_mbuf_timestamp_t timestamp;
 
-                               /* nanoseconds */
-                               timestamp =
-                                       rte_le_to_cpu_32(*((uint32_t *)
-                                       (packets[i].header + 4)));
-                               timestamp <<= 32;
                                /* seconds */
-                               timestamp |=
+                               timestamp =
                                        rte_le_to_cpu_32(*((uint32_t *)
                                        (packets[i].header + 8)));
+                               timestamp *= NSEC_PER_SEC;
+                               /* nanoseconds */
+                               timestamp +=
+                                       rte_le_to_cpu_32(*((uint32_t *)
+                                       (packets[i].header + 4)));
                                *nfb_timestamp_dynfield(mbuf) = timestamp;
                                mbuf->ol_flags |= nfb_timestamp_rx_dynflag;
                        }
-- 
2.52.0

Reply via email to