the
application may edit the packet headers itself afterwards).
Signed-off-by: Vivien Didelot
Signed-off-by: Patrick Keroulas
---
doc/guides/rel_notes/release_20_08.rst | 1 +
drivers/net/pcap/rte_eth_pcap.c| 32 +---
lib/librte_pdump/rte_pdump.c |
Use start time and device frequency to translate raw Rx hardware
timestamps into nanoseconds.
Signed-off-by: Patrick Keroulas
---
lib/librte_pdump/rte_pdump.c | 25 +
1 file changed, 25 insertions(+)
diff --git a/lib/librte_pdump/rte_pdump.c b/lib/librte_pdump
integrates the
nanosecond converter that derives from device frequency and start time.
v2 -> v3:
- replace ib_verbs nanosecond converter with more generic method
based on device frequency and start time.
Patrick Keroulas (3):
net/mlx5: query device frequency
ethdev: add API to qu
Signed-off-by: Patrick Keroulas
---
lib/librte_ethdev/rte_ethdev.c | 12
lib/librte_ethdev/rte_ethdev.h | 17 +
lib/librte_ethdev/rte_ethdev_core.h | 5 +
lib/librte_ethdev/rte_ethdev_version.map | 2 ++
4 files changed, 36 insertions
Get clock frequency (constant) from HCA attributes and add an accessor.
Signed-off-by: Patrick Keroulas
---
drivers/common/mlx5/mlx5_devx_cmds.c| 2 ++
drivers/common/mlx5/mlx5_devx_cmds.h| 1 +
drivers/net/mlx5/linux/mlx5_ethdev_os.c | 22 ++
drivers/net/mlx5
Hello Olivier and Slava,
On Tue, Jul 7, 2020 at 10:47 AM Olivier Matz wrote:
> For the pmd pcap part, the dynamic Tx timestamp flag that is being
> introduced by Slava [1] may fit your needs: ie. when the flag is set, it
> uses the provided timestamp which should be in nanosecs.
>
> [1] https://p
On Fri, Jun 26, 2020 at 2:41 AM Olivier Matz wrote:
> Should mlx5dv_get_clock_info() be called for each convesion? In the man
> page [1], we can read this:
>
>If the clock_info becomes too old then time conversion will return
>wrong conversion results. The user must ensure that
>
On Fri, Jun 26, 2020 at 2:48 AM Olivier Matz wrote:
> I don't get why you expect that timestamp to be in nanoseconds.
> The conversion is done in librte_pdump (in the previous patch),
> but it won't work if the library is not used, right?
>
You're right Olivier. I took advantage of the definition
vey the nanoseconds to pdump, since doc says: "unit and time
reference are not normalized but are always the same for a given port"
Patrick Keroulas (3):
net/mlx5: add counter-to-ns converter from libibverbs
ethdev: add API to convert raw timestamps to nsec
net/pcap: dump hardwa
ff-by: Patrick Keroulas
---
lib/librte_ethdev/rte_ethdev.c | 12
lib/librte_ethdev/rte_ethdev.h | 17 +
lib/librte_ethdev/rte_ethdev_core.h | 5 +
lib/librte_ethdev/rte_ethdev_version.map | 2 ++
lib/librte_mbuf/rte_mbuf_core.h
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
Signed-off-by: Vivien Didelot
---
drivers/net/pcap/rte_eth_pcap.c
While some devices update their own clock info to provide current time,
mlx5dv part of libibverbs already handles this and also converts any raw
counter cycle to nanoseconds.
Signed-off-by: Patrick Keroulas
---
drivers/common/mlx5/linux/mlx5_glue.c | 16 ++
drivers/common/mlx5/linux
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
Signed-off-by: Vivien Didelot
---
drivers/net/pcap/rte_eth_pcap.c
While some devices update their own clock info to provide current time,
mlx5dv part of libibverbs already handles this and also converts any raw
counter cycle to nanoseconds.
Signed-off-by: Patrick Keroulas
---
drivers/common/mlx5/mlx5_glue.c | 16
drivers/common/mlx5
ff-by: Patrick Keroulas
---
lib/librte_ethdev/rte_ethdev.c | 12
lib/librte_ethdev/rte_ethdev.h | 17 +
lib/librte_ethdev/rte_ethdev_core.h | 5 +
lib/librte_ethdev/rte_ethdev_version.map | 2 ++
lib/librte_mbuf/rte_mbuf_core.h
integrates the
nanosecond converter that is already implemented by ibverbs. The
conversion is performed is pdump rx callback.
Patrick Keroulas (3):
net/mlx5: add timestamp-to-ns converter from libibverbs
ethdev: add API to convert raw timestamps to nsec
net/pcap: dump hardware timestamps
On Wed, Jun 3, 2020 at 3:48 AM Slava Ovsiienko wrote:
>
> > From: PATRICK KEROULAS
> > * rdma-code, libibverbs-dev: 28.0
> > * NIC Part Number: MCX516A-CDA_Ax
> > * ConnectX-5 Ex EN
> > * FW: 16.25.1020
>
> It looks like outdated firmware, ple
On Tue, May 26, 2020 at 12:00 PM Slava Ovsiienko
wrote:
> ConnectX HW timestamp is the captured value of internal 64-bit counter
> running at the frequency,
> reported in the device_frequency_khz field of struct
> mlx5_ifc_cmd_hca_cap_bits{}.
> This structure is queried in mlx5_devx_cmd_query_hc
On Tue, May 26, 2020 at 12:00 PM Slava Ovsiienko
wrote:
>> Hi, Patrick
>
> ConnectX HW timestamp is the captured value of internal 64-bit counter
running at the frequency,
> reported in the device_frequency_khz field of struct
mlx5_ifc_cmd_hca_cap_bits{}.
> This structure is queried in mlx5_devx_c
> > > > mlx5 part of libibverbs includes a ts-to-ns converter which takes the
> > > > instantaneous clock info. It's unused in dpdk so far. I've tested it in
> > > > the
> > > > device/port init routine and the result looks reliable. Since this
> > > > approach
> > > > looks very simple, compared
> > I'm trying to build an accurate capture device based on Mellanox
> > Connect-X5 with following requirements:
> > - capture every incoming packets with hardware timestamps
> > - output: pcap with timestamps in nanoseconds
> > My problem is that the packets forwarded to `dpdk-pdump` carry raw
> >
Hello,
I'm trying to build an accurate capture device based on Mellanox
Connect-X5 with following requirements:
- capture every incoming packets with hardware timestamps
- output: pcap with timestamps in nanoseconds
My problem is that the packets forwarded to `dpdk-pdump` carry raw
timestamps from
22 matches
Mail list logo
