[RFC,v1,3/3] net/pcap: dump hardware timestamps
Checks
Commit Message
When hardware timestamping is activated, system time should no longer be
used to timestamp dumped the packets. Instead, use value held by
forwarded and assume they were converted to nanoseconds.
Signed-off-by: Patrick Keroulas <patrick.keroulas@radio-canada.ca>
Signed-off-by: Vivien Didelot <vivien.didelot@gmail.com>
---
drivers/net/pcap/rte_eth_pcap.c | 32 ++++++++++++++++++++++----------
1 file changed, 22 insertions(+), 10 deletions(-)
Comments
On 6/9/2020 3:17 PM, Patrick Keroulas wrote:
> When hardware timestamping is activated, system time should no longer be
> used to timestamp dumped the packets. Instead, use value held by
> forwarded and assume they were converted to nanoseconds.
>
> Signed-off-by: Patrick Keroulas <patrick.keroulas@radio-canada.ca>
> Signed-off-by: Vivien Didelot <vivien.didelot@gmail.com>
I think better to get the Vivien's patch that enables 'nanosecond timestamp',
with requested change first and later this patch can be on top of it to add
PKT_RX_TIMESTAMP support.
Or you can split the patch into two, as mentioned above, both are OK but I don't
want pcap nanosecond timestamp patch blocked from this hw timestamp patchset
discussions.
On 6/9/2020 3:41 PM, Ferruh Yigit wrote:
> On 6/9/2020 3:17 PM, Patrick Keroulas wrote:
>> When hardware timestamping is activated, system time should no longer be
>> used to timestamp dumped the packets. Instead, use value held by
>> forwarded and assume they were converted to nanoseconds.
>>
>> Signed-off-by: Patrick Keroulas <patrick.keroulas@radio-canada.ca>
>> Signed-off-by: Vivien Didelot <vivien.didelot@gmail.com>
>
> I think better to get the Vivien's patch that enables 'nanosecond timestamp',
> with requested change first and later this patch can be on top of it to add
> PKT_RX_TIMESTAMP support.
>
> Or you can split the patch into two, as mentioned above, both are OK but I don't
> want pcap nanosecond timestamp patch blocked from this hw timestamp patchset
> discussions.
>
Vivien's patch merged [1], can you please rebase this patchset on top of next-net?
[1] https://patches.dpdk.org/patch/71130/
@@ -45,6 +45,8 @@
#define RTE_PMD_PCAP_MAX_QUEUES 16
+#define NSEC_PER_SEC 1000000000L
+
static char errbuf[PCAP_ERRBUF_SIZE];
static struct timeval start_time;
static uint64_t start_cycles;
@@ -288,14 +290,24 @@ eth_null_rx(void *queue __rte_unused,
}
static inline void
-calculate_timestamp(struct timeval *ts) {
- uint64_t cycles;
- struct timeval cur_time;
-
- cycles = rte_get_timer_cycles() - start_cycles;
- cur_time.tv_sec = cycles / hz;
- cur_time.tv_usec = (cycles % hz) * 1e6 / hz;
- timeradd(&start_time, &cur_time, ts);
+calculate_timestamp(const struct rte_mbuf *mbuf, struct timeval *ts) {
+ if (mbuf->ol_flags & PKT_RX_TIMESTAMP) {
+ /* timestamp unit is nanoseconds but must fit in timeval */
+ ts->tv_sec = mbuf->timestamp / NSEC_PER_SEC;
+ ts->tv_usec = mbuf->timestamp % NSEC_PER_SEC;
+ }
+ else {
+ uint64_t cycles = rte_get_timer_cycles() - start_cycles;
+ struct timeval cur_time;
+ cur_time.tv_sec = cycles / hz;
+ cur_time.tv_usec = (cycles % hz) * NSEC_PER_SEC / hz;
+ ts->tv_sec = start_time.tv_sec + cur_time.tv_sec;
+ ts->tv_usec = start_time.tv_usec + cur_time.tv_usec;
+ if (ts->tv_usec > NSEC_PER_SEC) {
+ ts->tv_usec -= NSEC_PER_SEC;
+ ts->tv_sec += 1;
+ }
+ }
}
/*
@@ -331,7 +343,7 @@ eth_pcap_tx_dumper(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
caplen = sizeof(temp_data);
}
- calculate_timestamp(&header.ts);
+ calculate_timestamp(mbuf, &header.ts);
header.len = len;
header.caplen = caplen;
/* rte_pktmbuf_read() returns a pointer to the data directly
@@ -475,7 +487,7 @@ open_single_tx_pcap(const char *pcap_filename, pcap_dumper_t **dumper)
* with pcap_dump_open(). We create big enough an Ethernet
* pcap holder.
*/
- tx_pcap = pcap_open_dead(DLT_EN10MB, RTE_ETH_PCAP_SNAPSHOT_LEN);
+ tx_pcap = pcap_open_dead_with_tstamp_precision(DLT_EN10MB, RTE_ETH_PCAP_SNAPSHOT_LEN, PCAP_TSTAMP_PRECISION_NANO);
if (tx_pcap == NULL) {
PMD_LOG(ERR, "Couldn't create dead pcap");
return -1;