Hello,
In current AF_XDP PMD implementation, we cannot use it with software NICs like
VETH. Because it allocates umem with RTE_MEMZONE_IOVA_CONTIG.
Since AF_XDP itself is usable with any network devices, I think it is better to
have option to take umem with RTE_MEMZONE_SIZE_HINT_ONLY. It is als
Runtime requests to install an rte_flow with MARK action should fail
if a vector RX function was already chosen for the device during
configuration time.
Currently, i40e rte_flow driver would successfully install the flow
with MARK action, even when vector RX functions are in use. However,
those v
Hi Qi,
> -Original Message-
> From: Zhang, Qi Z
> Sent: Thursday, May 23, 2019 7:49 PM
> To: Ergin, Mesut A ; Xing, Beilei
>
> Cc: dev@dpdk.org
> Subject: RE: [PATCH] net/i40e: Fail rte_flow MARK requests if RX func was
> vectorized
>
> Hi Mesut:
>
> > -Original Message-
> > Fro
On Fri, 24 May 2019 19:32:14 +0200
Thomas Monjalon wrote:
> 24/05/2019 19:11, Stephen Hemminger:
> > On Fri, 24 May 2019 19:05:20 +0200
> > Thomas Monjalon wrote:
> >
> > > > > > device drivers on Hyper-V/Azure. If the build system supported
> > > > > > dependencies
> > > > > > (like Linux
On Fri, May 24, 2019 at 05:55:20PM +0100, Ferruh Yigit wrote:
> Current design requires kernel drivers and they need to be probed by
> Linux up to some level so that they can be usable by DPDK for ethtool
> support, this requires maintaining the Linux drivers in DPDK.
>
> Also ethtool support is l
Hi,
> > > > i40e_rx_vec_dev_conf_condition_check_default(struct rte_eth_dev *dev)
> > > > if (rxmode->offloads & DEV_RX_OFFLOAD_VLAN_EXTEND)
> > > > return -1;
> > > >
> > > > + /* Should not override if vector was already disallowed */
> > >
> > > It is possible a de
24/05/2019 19:11, Stephen Hemminger:
> On Fri, 24 May 2019 19:05:20 +0200
> Thomas Monjalon wrote:
>
> > > > > device drivers on Hyper-V/Azure. If the build system supported
> > > > > dependencies
> > > > > (like Linux kbuild) this would not be necessary. Meson only does
> > > > > dynamic depe
On 5/24/2019 5:03 PM, Stephen Hemminger wrote:
> From: Stephen Hemminger
>
> This driver should use dynamic log level not RTE_LOGTYPE_PMD.
> Other drivers were converted back in 18.02.
>
> This is really a bug, all other drivers use dynamic log levels
> by now.
>
> Signed-off-by: Stephen Hemmin
On Fri, 24 May 2019 19:05:20 +0200
Thomas Monjalon wrote:
> > > > device drivers on Hyper-V/Azure. If the build system supported
> > > > dependencies
> > > > (like Linux kbuild) this would not be necessary. Meson only does
> > > > dynamic dependencies
> > > > so that doesn't help.
> > > >
> >
24/05/2019 18:48, Stephen Hemminger:
> On Fri, 24 May 2019 18:38:53 +0200
> Thomas Monjalon wrote:
> > 24/05/2019 18:07, Stephen Hemminger:
> > > On Fri, 24 May 2019 14:26:40 +0100
> > > Ferruh Yigit wrote:
> > >
> > > > On 5/23/2019 11:01 PM, Stephen Hemminger wrote:
> > > > > Several users
On 24/05/2019 17:46, dwilder wrote:
> Hi Kevin-
>
> I tested 8.11.2 RC1 on IBM Power9 hardware. No issues or performance
> degradation were observed.
>
Thanks David. Good news on a Friday is always welcome :-)
> Tests run:
> * Single port stability test using l3fwd (16 cpus) and TRex, teste
On Fri, 24 May 2019 18:38:53 +0200
Thomas Monjalon wrote:
> 24/05/2019 18:07, Stephen Hemminger:
> > On Fri, 24 May 2019 14:26:40 +0100
> > Ferruh Yigit wrote:
> >
> > > On 5/23/2019 11:01 PM, Stephen Hemminger wrote:
> > > > Several users have run into problems where the MLX drivers were n
On 5/24/2019 5:29 PM, Ferruh Yigit wrote:
> From: Thomas Monjalon
I don't know what I did wrong to manage this, I am sending a new version.
>
> Current design requires kernel drivers and they need to be probed by
> Linux up to some level so that they can be usable by DPDK for ethtool
> support,
On Fri, 24 May 2019 17:29:22 +0100
Ferruh Yigit wrote:
> From: Thomas Monjalon
>
> Current design requires kernel drivers and they need to be probed by
> Linux up to some level so that they can be usable by DPDK for ethtool
> support, this requires maintaining the Linux drivers in DPDK.
>
> Al
Hi Kevin-
I tested 8.11.2 RC1 on IBM Power9 hardware. No issues or performance
degradation were observed.
Tests run:
* Single port stability test using l3fwd (16 cpus) and TRex, tested 64
and 1500 byte packets at a 0.0% drop rate for 4 hours each.
System tested:
- IBM Power9 Model 8335-1
24/05/2019 18:07, Stephen Hemminger:
> On Fri, 24 May 2019 14:26:40 +0100
> Ferruh Yigit wrote:
>
> > On 5/23/2019 11:01 PM, Stephen Hemminger wrote:
> > > Several users have run into problems where the MLX drivers were not
> > > enabled in their build. And then trying to run their DPDK
> > > app
On Fri, 24 May 2019 14:26:40 +0100
Ferruh Yigit wrote:
> On 5/23/2019 11:01 PM, Stephen Hemminger wrote:
> > Several users have run into problems where the MLX drivers were not
> > enabled in their build. And then trying to run their DPDK
> > application on Azure. What happens is that all packets
Setting RTE_MAX_LCORE to reflect the largest available configuration.
Adding defines for missing RTE_MACHINE_CPUFLAGs.
Signed-off-by: David Wilder
---
config/ppc_64/meson.build | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/config/ppc_64/meson.build b/config/ppc_64/meson.
From: Stephen Hemminger
This driver should use dynamic log level not RTE_LOGTYPE_PMD.
Other drivers were converted back in 18.02.
This is really a bug, all other drivers use dynamic log levels
by now.
Signed-off-by: Stephen Hemminger
---
v2 - use rte_log and fix debug logs
drivers/net/mlx4/m
The API to prepare checksum offloads employs outer
IP checksum flag to tell regular IPv4 packets from
tunnel packets with outer IPv4 encapsulation. This
flag cannot serve as a marker for the said purpose
because a packet can have outer IPv4 encapsulation
and may not have outer IP checksum offload r
From: Ajit Khaparde
Update HWRM API to version 1.10.0.74
Signed-off-by: Ajit Khaparde
Reviewed-by: Lance Richardson
---
drivers/net/bnxt/hsi_struct_def_dpdk.h | 1399
1 file changed, 1168 insertions(+), 231 deletions(-)
diff --git a/drivers/net/bnxt/hsi_struct_def_dp
From: Ajit Khaparde
Update release doc briefly describing updates to bnxt PMD.
Signed-off-by: Ajit Khaparde
---
doc/guides/rel_notes/release_19_08.rst | 6 ++
1 file changed, 6 insertions(+)
diff --git a/doc/guides/rel_notes/release_19_08.rst
b/doc/guides/rel_notes/release_19_08.rst
inde
From: Ajit Khaparde
Update HWRM version to 1.10.0.48
Signed-off-by: Ajit Khaparde
Reviewed-by: Lance Richardson
---
drivers/net/bnxt/hsi_struct_def_dpdk.h | 3156 +---
1 file changed, 2874 insertions(+), 282 deletions(-)
diff --git a/drivers/net/bnxt/hsi_struct_def_dpdk.h
From: Ajit Khaparde
Update HWRM API to version 1.10.0.19
Signed-off-by: Ajit Khaparde
Reviewed-by: Lance Richardson
---
drivers/net/bnxt/hsi_struct_def_dpdk.h | 3112 +++-
1 file changed, 2514 insertions(+), 598 deletions(-)
diff --git a/drivers/net/bnxt/hsi_struct_def_dp
Fix reta query op to only return table entries as identfied by the
provided mask. Translate firmware group IDs to queue numbers.
Removed extraneous code from bnxt_reta_query_op().
Fixes: d819382543f3 ("net/bnxt: add RSS redirection table operations")
Signed-off-by: Lance Richardson
Reviewed-by:
Fix the reta update function to only update table entries that are
selected by the update mask. Translate queue number to firmware
group ID when updating an entry.
Fixes: d819382543f3 ("net/bnxt: add RSS redirection table operations")
Signed-off-by: Lance Richardson
Reviewed-by: Ajit Khaparde
--
From: Ajit Khaparde
We are trying to update the indirection table for all the VNICs.
We should update the table only for the default vnic0.
This patch fixes the above issue.
Fixes: d819382543f3 ("net/bnxt: add RSS redirection table operations")
Signed-off-by: Ajit Khaparde
Reviewed-by: Rahul Gu
Introduce vector mode support for the bnxt pmd.
Signed-off-by: Lance Richardson
Signed-off-by: Ajit Khaparde
---
v2:
* Squashed with v1 patch 4 ("fix double counting VLAN tags").
* Dropped two unnecessary coding style changes from bnxt_txr.h.
config/common_base | 1 +
drive
This patchset bsae on top of the previous submission adds the following:
1) Support for vector mode TX and RX.
2) HWRM API update (split into multiple patches).
3) Fixes for RSS reta update and query.
It also updates the release notes.
v2:
* Squashed patches 3 and 4 from v1 patchset.
* Adde
To allow sharing of tx_bds_in_hw() and bnxt_tx_avail() between
vector-mode and non-vector transmit functions, move these functions
into bnxt_txr.h.
Signed-off-by: Lance Richardson
Reviewed-by: Ajit Khaparde
Reviewed-by: Maxime Coquelin
---
drivers/net/bnxt/bnxt_txr.c | 15 ---
driv
In preparation for a bnxt vector-mode driver, compute and store
scattered_rx status for the device when started.
Signed-off-by: Lance Richardson
Reviewed-by: Ajit Khaparde
Reviewed-by: Maxime Coquelin
---
drivers/net/bnxt/bnxt_ethdev.c | 23 +++
1 file changed, 23 insertion
On 5/21/2019 10:27 PM, Stephen Hemminger wrote:
> This driver should use dynamic log level not RTE_LOGTYPE_PMD.
> Other drivers were converted back in 18.02.
>
> This is really a bug, all other drivers use dynamic log levels
> by now.
+1, we should switch it to dynamic log much earlier
Matan, Sh
On 5/23/2019 8:45 AM, Olivier Matz wrote:
> Add the ability to pass custom options to checkpatch script. An example
> of use is to change the output format so it can run in emacs compilation
> mode:
>
> DPDK_CHECKPATCH_PATH=/path/to/linux/scripts/checkpatch.pl \
> DPDK_CHECKPATCH_OPTIONS='--
On 5/23/2019 2:37 AM, Beilei Xing wrote:
> This patch allows VF to get/update/reset pctype
> mapping info.
>
> Signed-off-by: Beilei Xing
> ---
> v2 changes:
> - Allow VF update/reset pctype mapping.
>
> drivers/net/i40e/rte_pmd_i40e.c | 8
> 1 file changed, 8 deletions(-)
>
> diff -
On 5/20/19 7:51 AM, Tiwei Bie wrote:
On Fri, May 17, 2019 at 05:06:11PM +0200, Maxime Coquelin wrote:
[...]
+static void
+copy_vnet_hdr_from_desc(struct virtio_net_hdr *hdr,
+ struct buf_vector *buf_vec)
+{
+ uint64_t len;
+ uint64_t remain = sizeof(struct virtio_
On 5/23/2019 8:45 AM, Olivier Matz wrote:
> Add the ability to pass custom options to checkpatch script. An example
> of use is to change the output format so it can run in emacs compilation
> mode:
>
> DPDK_CHECKPATCH_PATH=/path/to/linux/scripts/checkpatch.pl \
> DPDK_CHECKPATCH_OPTIONS='--
writes:
> From: Pavan Nikhilesh
>
> This patch adds an optimized arm64 instruction based routine to leverage
> CPU pipeline characteristics of octeontx2. The theme is to fill the
> pipeline with CASP operations as much HW can do so that HW can do alloc()
> HW ops in full throttle.
>
> Cc: Olivie
On 5/23/2019 11:01 PM, Stephen Hemminger wrote:
> Several users have run into problems where the MLX drivers were not
> enabled in their build. And then trying to run their DPDK
> application on Azure. What happens is that all packets
> go over the slow path, and failsafe repeatedly probes for neve
On 5/24/2019 2:05 AM, Xu, Rosen wrote:
> Hi,
>
>> -Original Message-
>> From: Pei, Andy
>> Sent: Thursday, May 23, 2019 17:15
>> To: dev@dpdk.org
>> Cc: Pei, Andy ; Zhang, Roy Fan
>> ; Zhang, Qi Z ; Wu,
>> Jingjing ; Xing, Beilei ;
>> Yigit,
>> Ferruh ; Xu, Rosen
>> Subject: [PATCH] net/
On 5/24/2019 2:10 AM, Ruifeng Wang (Arm Technology China) wrote:
> Hi,
>
>> -Original Message-
>> From: Liron Himi
>> Sent: Thursday, May 23, 2019 18:51
>> To: Ruifeng Wang (Arm Technology China) ;
>> tho...@monjalon.net
>> Cc: dev@dpdk.org; nd ; Liron Himi
>> Subject: RE: [dpdk-dev] [PA
On 5/24/2019 2:55 AM, Zhao1, Wei wrote:
>
> Hi, Ferruh
>
>> -Original Message-
>> From: Yigit, Ferruh
>> Sent: Tuesday, May 21, 2019 11:43 PM
>> To: Zhao1, Wei ; dev@dpdk.org
>> Cc: sta...@dpdk.org; Peng, Yuan ; Lu, Wenzhuo
>>
>> Subject: Re: [dpdk-stable] [PATCH] app/testpmd: fix offloa
On 5/21/2019 5:13 PM, Olivier Matz wrote:
> The rte_net headers conflict with the libc headers, because
> some definitions are duplicated, sometimes with few differences.
>
> This patchset adds the rte_ (or RTE_) prefix to all structures, functions
> and defines in rte_net library. This is a big c
On 5/21/2019 5:13 PM, Olivier Matz wrote:
> Also rename arp_hrd, arp_pro, arp_hln, arp_pln and arp_op fields
> to avoid conflict with the #defines in gnu libc.
>
> Signed-off-by: Olivier Matz
<...>
> @@ -614,7 +614,7 @@ mode6_debug(const char __attribute__((unused)) *info,
> struct ether_hdr *
On 5/21/2019 5:13 PM, Olivier Matz wrote:
> Add 'RTE_' prefix to defines:
> - rename ETHER_ADDR_LEN as RTE_ETHER_ADDR_LEN.
> - rename ETHER_TYPE_LEN as RTE_ETHER_TYPE_LEN.
> - rename ETHER_CRC_LEN as RTE_ETHER_CRC_LEN.
> - rename ETHER_HDR_LEN as RTE_ETHER_HDR_LEN.
> - rename ETHER_MIN_LEN as RTE_E
On 24-May-19 10:34 AM, Reshma Pattan wrote:
Add new telemetry mode support for l3fwd-power.
This is a standalone mode, in this mode l3fwd-power
does simple l3fwding along with calculating
empty polls, full polls, and busy percentage for
each forwarding core. The aggregation of these
values of all
Add new telemetry mode support for l3fwd-power.
This is a standalone mode, in this mode l3fwd-power
does simple l3fwding along with calculating
empty polls, full polls, and busy percentage for
each forwarding core. The aggregation of these
values of all cores is reported as application
level teleme
After setting up the link on x552/X557-AT 10GBASE-T NICs, sometimes the
link does not get set up properly and as a result all the subsequent
calls to ixgbe_check_link() from ixgbe_dev_link_update_share() fail.
Introduce a delay time of 1s in ixgbe_setup_mac_link_t_X550em() before
beginning to set
Hi,
> -Original Message-
> From: dev On Behalf Of lir...@marvell.com
> Sent: Saturday, May 18, 2019 05:34
> To: tho...@monjalon.net
> Cc: dev@dpdk.org; Liron Himi
> Subject: [dpdk-dev] [PATCH 1/2] mk: add Marvell ARMADA architecture
> based on armv8-a
>
> From: Liron Himi
>
> This pat
48 matches
Mail list logo
