Are the rx_missed_errors/rx_out_of_buffer counters always showing 0 no
matter how fast you push your generator?
I had a similar issue with missing counters on DPDK 20.11 that was fixed on
21.05 by applying this patch:
mend to keep the package versions provided for your distro and
> downgrade dpdk if necessary.
>
>
> On Mon, Aug 2, 2021 at 4:18 PM Gerry Wan wrote:
> >
> > I am running Ubuntu 20.04, and the mlx5dv.h is indeed in
> `/usr/include/infiniband/`.
> >
> > The steps
/github.com/linux-rdma/rdma-core/blob/stable-v28/providers/mlx5/mlx5dv.h
> I expect this header to be installed in `/usr/include/infiniband/`.
> Please verify.
>
> Maybe something wrong with ninja, try meson or make.
>
> On Mon, Aug 2, 2021 at 2:35 PM Gerry Wan wrote:
> >
>
properly remove everything
> related
> to OFED with the uninstaller. Then, install rdma-core and libibverbs-dev
> which
> includes infiniband/mlx5dv.h IIRC.
> Regards,
> Patrick
>
> On Sun, Aug 1, 2021 at 8:23 PM Gerry Wan wrote:
>
>> I recently tried to reinstall MLX OFED 5
I recently tried to reinstall MLX OFED 5.2-2.2.0.0 on my bare metal machine
Ubuntu 20.04 (previously using the command ./mlnxofedinstall --dpdk
--upstream-libs --with-mft --with-kernel-mft). However, it now fails with
the message
Checking SW Requirements...
Removing old packages...
Uninstalling
My understanding of an increasing imissed counter is that it indicates your
processing logic in the lcore is not fast enough to handle the rate of
incoming packets, and is independent of the number of free mbufs. I would
guess that using the pipeline model (passing mbufs between lcores via
rings)
common/mlx5: add DevX command to query WQ
> common/mlx5: add DevX commands for queue counters
> net/mlx5: fix imissed statistics
>
>
> Regards,
> Asaf Penso
>
> >-Original Message-
> >From: users On Behalf Of Tom Barbette
> >Sent: Tuesday, Apr
ays remain 0.
Any workarounds for this? This is an important statistic for my use case.
On Fri, Apr 2, 2021 at 5:03 PM Gerry Wan wrote:
> I have a simple forwarding experiment using a mlx5 NIC directly connected
> to a generator. I am noticing that at high enough throughput,
> rx_good_packets
I have a simple forwarding experiment using a mlx5 NIC directly connected
to a generator. I am noticing that at high enough throughput,
rx_good_packets + rx_phy_discard_packets may not equal rx_phy_packets.
Where are these packets being dropped?
Below is an example xstats where I receive at
?
If this is likely the case, how can I fix this to work with shared
libraries?
On Sun, Dec 13, 2020 at 1:04 PM Gerry Wan wrote:
> OK, I was able to build it this way.
> However, it seems that it still does not solve my problem where the
> application detects no ports available. I am
u can do when doing ninja is to disable mxl4
> compilation:
>
>
>
> meson build -D disable_driver=net/mlx4; ninja -C build
>
>
>
> Kindest regards,
>
> Raslan Darawsheh
>
>
>
> *From:* Gerry Wan
> *Sent:* Sunday, December 13, 2020 1:25 PM
> *To:* R
l be
> built automatically.
>
> As for your question if you still don't need to bind mlx5 driver that's
> true, its still valid.
>
> Kindest regards,
> Raslan Darawsheh
>
> > -Original Message-----
> > From: users On Behalf Of Gerry Wan
> > Sent: Su
Hello,
I am trying to update my DPDK version from 20.08 to 20.11. I am using the
mlx5 driver, and previously to enable it I had to set
CONFIG_RTE_LIBRTE_MLX5_PMD=y in config/common_base. In 20.11, I see no
option to do so. I was able to build my application (and the sample
applications), but
id <https://aka.ms/ghei36>
>
> ------
> *From:* Gerry Wan
> *Sent:* Sunday, October 25, 2020 9:14:32 PM
> *To:* Matan Azrad
> *Cc:* Alireza Sanaee ; Asaf Penso ;
> users@dpdk.org
> *Subject:* Re: [dpdk-users] RSS with VLAN tagged packets
>
&g
that contain the API changes,
is there a way I can download and apply them all?
On Sun, Oct 25, 2020 at 10:58 AM Matan Azrad wrote:
> This one will solve it:
>
> https://patches.dpdk.org/patch/82100/
>
>
>
> *From:* Alireza Sanaee
> *Sent:* Sunday, October 25, 2020 5:36 PM
et you know as soon as we have it.
>
> Thanks,
> Asaf Penso
>
> Regards,
> Asaf Penso
>
>
> ------
> *From:* users on behalf of Gerry Wan <
> ger...@stanford.edu>
> *Sent:* Sunday, October 25, 2020, 00:05
> *To:* users@dpdk.org
Hello,
I am configuring RSS through through rte_eth_dev_configure using
rte_eth_conf with mq_ mode = ETH_MQ_RSS and specifying rss_hf = ETH_RSS_IP
| ETH_RSS_TCP | ETH_RSS_UDP. I've noticed that even single VLAN tagged
packets do not get hashed.
Is there a flag I am missing to do this? Is RSS
Hi,
I'm testing out the maximum RX throughput my application can handle and am
using rte_eth_stats_get() to measure the point at which RX packets start
getting dropped. I have a traffic generator directly connected to my RX
port.
I've noticed that imissed is the number of packets that are
Hello,
I'm trying to query per-flow statistics using RTE_FLOW_ACTION_TYPE_COUNT on
a Mellanox ConnectX-5 port. I tried extending the flow_filtering sample
application with:
struct rte_flow_query_count count = {
.reset = 1,
.hits_set = 1,
.bytes_set = 1,
.hits = 0,
.bytes = 0,
Regards,
> Asaf Penso
>
> -Original Message-
> From: users On Behalf Of Erez Ferber
> Sent: Monday, July 27, 2020 9:25 PM
> To: Gerry Wan
> Cc: users@dpdk.org
> Subject: Re: [dpdk-users] COUNT action not supported on mlx5
>
> Hi,
>
> One possible co
Hello,
I'm trying to query per-flow statistics using RTE_FLOW_ACTION_TYPE_COUNT on
a Mellanox ConnectX-5 port. I tried extending the flow_filtering sample
application with:
struct rte_flow_query_count count = {
.reset = 1,
.hits_set = 1,
.bytes_set = 1,
.hits = 0,
.bytes = 0,
21 matches
Mail list logo