Hi Slava,
On Mon, Nov 16, 2020 at 9:03 PM Slava Ovsiienko wrote:
>
> Hi, Maxime
>
> It seems there are some typos (see below).
> Beside these ones - ACKed.
>
> > -Original Message-
> > From: Maxime Leroy
> > Sent: Monday, November 16, 2020 19:03
> > To: Matan Azrad ; Shahaf Shuler ;
> >
Distinguish inner/outer fields. And avoid too many nested conditionals
in each type's parser.
Signed-off-by: Zhirun Yan
---
drivers/net/ice/ice_fdir_filter.c | 504 --
1 file changed, 269 insertions(+), 235 deletions(-)
diff --git a/drivers/net/ice/ice_fdir_filter.c
When low and high registers are read separately, this opens the door to
a race condition:
- low register is read
- NIC updates the registers
- high register is read
Because of this, we may end up with an incorrect counter value.
Let's read the registers in one shot, as it is done in Linux kernel
s
Update the OS versions. Remove some old OS versions.
Signed-off-by: Ajit Khaparde
---
doc/guides/nics/bnxt.rst | 8 +++-
1 file changed, 3 insertions(+), 5 deletions(-)
diff --git a/doc/guides/nics/bnxt.rst b/doc/guides/nics/bnxt.rst
index d9a7d87930..898c71ef1e 100644
--- a/doc/guides/nics
On Tue, Nov 17, 2020 at 08:49:45AM +0100, Thomas Monjalon wrote:
> 17/11/2020 03:46, Honnappa Nagarahalli:
> >
> >
> > >
> > > 16/11/2020 17:16, Bruce Richardson:
> > > > On Mon, Nov 16, 2020 at 03:50:31PM +, Juraj Linkeš wrote:
> > > > > From: Thomas Monjalon
> > > > > > 13/11/2020 15:31,
Hi, Maxime
> -Original Message-
> From: Maxime Leroy
> Sent: Tuesday, November 17, 2020 10:52
> To: Slava Ovsiienko
> Cc: Matan Azrad ; Shahaf Shuler ;
> Olivier Matz ; dev@dpdk.org; Didier Pallard
>
> Subject: Re: [PATCH v3 2/2] net/mlx5: fix Rx descriptor status returned value
>
> Hi
17/11/2020 10:15, Bruce Richardson:
> On Tue, Nov 17, 2020 at 08:49:45AM +0100, Thomas Monjalon wrote:
> > 17/11/2020 03:46, Honnappa Nagarahalli:
> > >
> > >
> > > >
> > > > 16/11/2020 17:16, Bruce Richardson:
> > > > > On Mon, Nov 16, 2020 at 03:50:31PM +, Juraj Linkeš wrote:
> > > > > > F
update the qos_sched user guide and profile config file
with default subport profile.
Fixes: 802d214dc880 ("examples/qos_sched: update subport rate dynamically")
Signed-off-by: Savinay Dharmappa
Acked-by: Cristian Dumitrescu
---
v3: addressed the review comments of patch v2
---
doc/guides/samp
This patch set introduces vectorized RX/TX path for packed ring with NEON
intrinsics.
With this patch set, PVP case has 1.5% perf uplift for the packed vectorized
NEON path compared with the non-vector packed path, under 0.001% acceptable
loss with 2 cores on vhost side and 1 core on virtio side.
Split out AVX instruction based virtio packed ring Rx and Tx
implementation to a separate file.
Signed-off-by: Joyce Kong
Reviewed-by: Ruifeng Wang
---
drivers/net/virtio/meson.build | 4 +-
drivers/net/virtio/virtio_rxtx.c| 6 +-
drivers/net/virtio/virtio_rxtx_pack
Optimize packed ring Rx batch path with NEON instructions.
Signed-off-by: Joyce Kong
Reviewed-by: Ruifeng Wang
---
drivers/net/virtio/virtio_rxtx_packed.h | 15 ++
drivers/net/virtio/virtio_rxtx_packed_neon.h | 150 +++
2 files changed, 165 insertions(+)
create mode 10064
Optimize packed ring Tx batch path with NEON instructions.
Signed-off-by: Joyce Kong
Reviewed-by: Ruifeng Wang
---
drivers/net/virtio/virtio_rxtx_packed.h | 6 +-
drivers/net/virtio/virtio_rxtx_packed_neon.h | 143 +++
2 files changed, 148 insertions(+), 1 deletion(-)
di
Add NEON vectorized path selection logic. Default setting comes from
vectorized devarg, then checks each criteria.
Packed ring vectorized neon path need:
NEON is supported by compiler and host
VERSION_1 and IN_ORDER features are negotiated
mergeable feature is not negotiated
LRO of
update the user guide of ip pipeline sample application
to reflect the changes in command line arguments.
Fixes: 54a298e5f7f4 ("examples/ip_pipeline: update subport rate dynamically")
Signed-off-by: Savinay Dharmappa
Acked-by: Cristian Dumitrescu
---
v2: patch has only ip pipeline doc changes
-
On 11/16/2020 7:09 PM, Dekel Peled wrote:
< Please don't top post, reply moved to the bottom >
-Original Message-
From: Ferruh Yigit
Sent: Monday, November 16, 2020 6:27 PM
To: Xiaoyu Min ; Jasvinder Singh
; Cristian Dumitrescu
Cc: dev@dpdk.org; Dekel Peled
Subject: Re: [dpdk-dev] [P
The verbosity was meant to be set with options -v and -vv,
or possibly with the environment variables TEST_MESON_BUILD_VERBOSE
and TEST_MESON_BUILD_VERY_VERBOSE.
It is decided to keep only the options -v and -vv,
so the variables are renamed with lower case, marking them as privates.
The handling
Unfortunately, this change has broken the build for SPDK on Windows.
To use the DPDK libraries, an application needs to include the rte_*
header, which includes rte_os.h via rte_common.h. The definition of
ETOOMANYREFS clashes with the value used when building the SPDK on Windows.
The fundam
Hi,
> -Original Message-
> From: Gregory Etelson
> Sent: Monday, November 16, 2020 4:02 PM
> To: dev@dpdk.org
> Cc: Gregory Etelson ; Matan Azrad
> ; Raslan Darawsheh
> Subject: [PATCH v5 0/6] restore tunnel offload functionality in mlx5
>
> post merge fixes that restore tunnel offload
Hi,
> -Original Message-
> From: Alexander Kozyrev
> Sent: Sunday, November 15, 2020 4:26 PM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Raslan Darawsheh ; Dekel Peled
> ; Matan Azrad
> Subject: [PATCH] net/mlx5: fix packet padding config for RxQ via DevX
>
> Received packets can be align
On 11/16/2020 2:24 PM, Loftus, Ciara wrote:
On 11/12/2020 4:35 PM, Ciara Loftus wrote:
'uint64_t' is used to hold pointers in multiple locations in the
copy-mode code (used for kernels before 5.4). For a 32-bit build
this assumption is wrong and results in build errors. This commit
replaces suc
Fix assertion check warnings.
Fixes: 8bb81f2649b1 ("net/mlx5: use thread specific flow workspace")
Cc: sta...@dpdk.org
Signed-off-by: Xueming Li
Reviewed-by: Slava Ovsiienko
---
drivers/net/mlx5/mlx5_flow.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/mlx5/ml
When the representor device was set to PF1 in bonding mode, iterating
device iterator that looking for representors by bonding device failed
to match PF0 pci address with PF1 address. So detaching PF bonding
device only detached all representors on PF0.
This patch registers all representors of PF1
This patchset provides several bug fixes for rx queue count calculation for
mlx5 driver.
---
V4:
- fix typos in commit log
V3:
- fix spelling mistakes
- fix second commit for mprq case
V2:
* squash first patch and second patch
* fix wrong init of used for compressed cqes
Didier Pallard (1):
The commit d2d57605522d ("net/mlx5: fix Rx queue count calculation") is
incorrect because the count calculation is wrong for the next cqe:
Example:
Compressed Set of packets 1 | Compressed Set of packets 2
C | a | e0 | e1 | e2 | e3 | e4 | e5 | C | a | e0
There are 2 compressed set of packets
From: Didier Pallard
Three bugs in rx_queue_count function:
- One entry may contain several segments, so 'used' must be multiplied
by number of segments per entry to properly reflect the queue usage.
- The number of cqes is equals to (1U << rxq->elts_n) - 1 in SPRQ mode.
The range returned by
Hi,
> -Original Message-
> From: dev On Behalf Of Xiaoyu Min
> Sent: Monday, November 16, 2020 4:08 AM
> To: Matan Azrad ; Shahaf Shuler
> ; Slava Ovsiienko
> Cc: dev@dpdk.org; Jack Min ; sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH] net/mlx5: fix validate RSS queues types
>
> From: Xia
> -Original Message-
> From: Maxime Leroy
> Sent: Tuesday, November 17, 2020 13:27
> To: Matan Azrad ; Shahaf Shuler ;
> Slava Ovsiienko ; Olivier Matz
>
> Cc: dev@dpdk.org; Didier Pallard
> Subject: [PATCH v4 2/2] net/mlx5: fix Rx descriptor status returned value
>
> From: Didier Palla
> -Original Message-
> From: Maxime Leroy
> Sent: Tuesday, November 17, 2020 13:27
> To: Matan Azrad ; Shahaf Shuler ;
> Slava Ovsiienko ; Alexander Kozyrev
>
> Cc: dev@dpdk.org; NBU-Contact-N?lio Laranjeiro
>
> Subject: [PATCH v4 1/2] net/mlx5: fix Rx queue count calculation
>
> The co
On 11/14/2020 12:01 PM, Jiawei Wang wrote:
When testpmd enabled the verbosity for the received packets, if two packets
was received at the same time, for example, sampling packet and normal
packet, the dump output of these packets may be overlapping due to multiple
core handled the multiple queue
PSB.
> -Original Message-
> From: Ferruh Yigit
> Sent: Tuesday, November 17, 2020 12:30 PM
> To: Dekel Peled ; Xiaoyu Min
> ; Jasvinder Singh ;
> Cristian Dumitrescu
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 5/5] net/softnic: update headers size
> calculation
>
> On 11/16/2020
On 11/15/2020 11:23 AM, Gregory Etelson wrote:
testpmd provides commands to test tunnel offload rte_flow
capabilities. A command that creates a tunnel had a syntax bug that
allowed to spesify the 'create' command part several times.
Current patch fixed that fault. Correct syntax for tunnel creti
Hi Nick,
> This means that rte_os.h should not include POSIX/Linux definitions to
> avoid clashes such as the one seen with this change. It's clearly not
> sustainable if applications have to be modified every time we add more
> Windows support to the DPDK.
>
> Note that this is not an isolat
On 11/16/2020 5:32 AM, Simei Su wrote:
This patch defines new RSS offload types for eCPRI. For eCPRI with
Message Type 0, the hash field is physical channel ID.
Signed-off-by: Simei Su
---
lib/librte_ethdev/rte_ethdev.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/librte_ethdev/rt
On 11/16/2020 5:32 AM, Simei Su wrote:
This patch adds testpmd cmdline support for eCPRI.
Signed-off-by: Simei Su
---
app/test-pmd/cmdline.c | 9 ++---
app/test-pmd/config.c | 3 ++-
2 files changed, 8 insertions(+), 4 deletions(-)
diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cm
From: Xiaoyu Min
> The rte_flow_item_eth and rte_flow_item_vlan items are refined.
> The structs do not exactly represent the packet bits captured on the wire
> anymore.
> Should use real header instead of the whole struct.
>
> Replace the rte_flow_item_* with the existing corresponding rte_*_h
On 11/17/2020 7:39 AM, Andrew Rybchenko wrote:
On 11/16/20 10:23 PM, Ivan Malov wrote:
Don't free the outer match spec by its pointer in the parsing context
if it has already been tracked by an entry in the outer rule registry.
Fixes: dadff137931c ("net/sfc: support encap flow items in transfer
Hello Ferruh,
> On 11/15/2020 11:23 AM, Gregory Etelson wrote:
> > testpmd provides commands to test tunnel offload rte_flow
> > capabilities. A command that creates a tunnel had a syntax bug that
> > allowed to spesify the 'create' command part several times.
> >
> > Current patch fixed that faul
Hi,
> -Original Message-
> From: dev On Behalf Of Maxime Leroy
> Sent: Tuesday, November 17, 2020 1:27 PM
> Cc: dev@dpdk.org
> Subject: [dpdk-dev] [PATCH v4 0/2] net/mlx5: fixes for rx queue count
> calculation
>
> This patchset provides several bug fixes for rx queue count calculation f
On 11/17/2020 7:38 AM, Andrew Rybchenko wrote:
On 11/16/20 10:40 PM, Ivan Malov wrote:
The changeset that introduced common flow API thread safety
in fact introduced double locking to this particular PMD as
RTE flow API implementation in the PMD has been thread-safe
since the day zero. State thi
On 11/16/2020 2:02 PM, Gregory Etelson wrote:
Tunnel offload API is implemented for Direct Verbs environment only.
Current patch re-arranges tunnel related functions for compilation in
non Direct Verbs setups to prevent compilation failures. The patch
does not introduce new functions.
Fixes: 4e
AF_XDP will not work on 32-bit kernels before version 5.4.
Document this restriction in the driver guide.
Signed-off-by: Ciara Loftus
---
doc/guides/nics/af_xdp.rst | 1 +
1 file changed, 1 insertion(+)
diff --git a/doc/guides/nics/af_xdp.rst b/doc/guides/nics/af_xdp.rst
index 052e59a3ae..5ed24
Hi Zhirun,
I think it is great to refactor ice fdir to differentiate inner and
outer field vector in a more clear way. Thanks for your commit.
It seems there still needs some effort to complete this patch.
> -Original Message-
> From: Yan, Zhirun
> Sent: Tuesday, Novem
> > >>
> > >> On 11/16/2020 7:55 AM, Xiaoyu Min wrote:
> > >>> From: Dekel Peled
> > >>>
> > >>> The rte_flow_item_eth and rte_flow_item_vlan items were updated in
> > >> [1].
> > >>> The rte_flow_item_ipv6 item was updated in [2].
> > >>> The structs now contain additional metadata following t
Hi,
> -Original Message-
> From: Viacheslav Ovsiienko
> Sent: Friday, November 13, 2020 5:33 PM
> To: dev@dpdk.org
> Cc: Raslan Darawsheh ; Matan Azrad
>
> Subject: [PATCH] net/mlx5: add wire vport hint
>
> The kernel can use two approaches to distinguish the E-Switch
> source vport in
Hi,
> -Original Message-
> From: Alexander Kozyrev
> Sent: Monday, November 16, 2020 9:48 PM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Raslan Darawsheh ; Slava
> Ovsiienko ; Matan Azrad
> Subject: [PATCH] net/mlx5: fix WQE counter assert in free completion queue
>
> The following assert
From: Pavan Nikhilesh
SSO workslot shouldn't be flushed on Tx adapter enqueue as
application might use reference count to re-transmit or
forward the packet.
Fixes: cb7ee83b6365 ("event/octeontx2: improve single flow performance")
Cc: sta...@dpdk.org
Signed-off-by: Pavan Nikhilesh
---
drivers/
On Mon, Nov 16, 2020 at 8:12 PM Harman Kalra wrote:
>
> Syncing the data types of msix offset mbox response structure
> fields wrt rvu_pfvf structure as in msix offset mbox handler,
> respective block LFs response field is initialised with rvu_pfvf
> and mismatch in the data types of respective fi
> -Original Message-
> From: dev On Behalf Of Bruce Richardson
> Sent: Thursday, November 12, 2020 11:32 AM
> To: dev@dpdk.org
> Cc: Bruce Richardson
> Subject: [dpdk-dev] [RFC PATCH] build: allow passing extra config header
> to build
>
> To allow per-build override of some settings,
'_filters' is compared twice, second one will be always false, removing
it using the message more relevant to the '_filters'.
Fixes: 2deb6b5246d7 ("app/procinfo: add collectd format and host id")
Cc: sta...@dpdk.org
Signed-off-by: Ferruh Yigit
---
Cc: romanx.korynkev...@intel.com
---
app/proc-i
Fix a few of issues identified by cppcheck
Ferruh Yigit (4):
app/procinfo: fix redundant condition
app/procinfo: fix negative check on unsigned variable
app/procinfo: remove suspicious sizeof
app/procinfo: remove useless assignment
app/proc-info/main.c | 14 +-
1 file changed
'parse_xstats_ids()' return 'int'. The return value is assigned to
'nb_xstats_ids' unsigned value, later negative check on this variable is
wrong.
Adding interim 'int' variable for negative check.
Fixes: 7ac16a3660c0 ("app/proc-info: support xstats by ID and by name")
Cc: sta...@dpdk.org
Signed-
The intention with the "sizeof(0)" usage is not clear, but the 'stats'
already 'memset' by 'rte_cryptodev_stats_get()' API, removing 'memset'
in application.
Fixes: fe773600fe3e ("app/procinfo: add --show-crypto")
Cc: sta...@dpdk.org
Signed-off-by: Ferruh Yigit
---
Cc: vipin.vargh...@intel.com
-
'flag' initialized to '0' but it is overwritten later, initial
assignment can be removed.
Fixes: 0101a0ec6217 ("app/procinfo: add --show-mempool")
Cc: sta...@dpdk.org
Signed-off-by: Ferruh Yigit
---
Cc: vipin.vargh...@intel.com
---
app/proc-info/main.c | 2 +-
1 file changed, 1 insertion(+), 1
Hi Ferruh,
Thanks for the update
snipped
> show_mempool(char *name) {
> - uint64_t flags = 0;
> + uint64_t flags;
>
Checking the current code base it makes more sense to move the code inside `if`
condition check. Sample code shared below
```
- uint64_t flags = 0;
snpri
Thanks Ferruh
> -Original Message-
> From: Yigit, Ferruh
> Sent: Tuesday, November 17, 2020 10:45 PM
> To: Tahhan, Maryam ; Pattan, Reshma
> ; Mcnamara, John ;
> Varghese, Vipin
> Cc: Yigit, Ferruh ; dev@dpdk.org; sta...@dpdk.org
> Subject: [PATCH 3/4] app/procinfo: remove suspicious siz
DPDK relies on pkg-config(1) to provide correct parameters for
compiler and linker used in application build. Inaccurate build
parameters, produced by pkg-config from DPDK .pc files could fail
application build or cause unpredicted results during application
runtime.
Update system requirements do
l3fwd-power uses `--parse-ptype' parameter to query egress packets
type. Before that feature is eanbled l3fwd-power verifies PMD ability
to advertise supporrtred packet types with
rte_eth_dev_get_supported_ptypes().
The patch updates code for PMDs that register supported types after
dev_start.
F
Representor is a port in DPDK that is connected to a VF in such a way
that assuming there are no offload flows, each packet that is sent
from the VF will be received by the corresponding representor. While
each packet that is sent to a representor will be received by the VF.
This is very useful in
Hello,
> 13/09/2020 14:12, Ori Kam:
> > Hi Ferruh,
> > Can we proceed with this patch?
>
> Below, you said "first thing to do it update the rte_flow doc".
> So I am expecting a patch on the doc to start this discussion.
> This testpmd patch is on hold in my understanding.
>
Documentation patch
There is a discrepancy between RTE ETHDEV API and flow rules guide
regarding flow rules maintenance after port stop. RTE ETHDEV API in
librte_ethdev.h declares that flow rules will not be stored in PMD
after port stop:
> Quite start
Please note that some configuration is not stored between c
On 11/17/20 10:18 PM, Gregory Etelson wrote:
There is a discrepancy between RTE ETHDEV API and flow rules guide
regarding flow rules maintenance after port stop. RTE ETHDEV API in
librte_ethdev.h declares that flow rules will not be stored in PMD
after port stop:
> Quite start
Please not
Hi Lijuan,
I currently have issues with running the Nic Single-Core Perf Test.
I believe this is the scapy naming conflict issue, so I follow the solution
mentioned on the website to replace system scapy with trex scapy.
Then I got into another error that I wasn't able to get pass:
root@arm-dell:
Hi,
> -Original Message-
> From: dev On Behalf Of Xueming Li
> Sent: Tuesday, November 17, 2020 1:02 PM
> To: Matan Azrad ; Slava Ovsiienko
>
> Cc: dev@dpdk.org; Xueming(Steven) Li ; Asaf Penso
>
> Subject: [dpdk-dev] [PATCH] net/mlx5: set representor device to first PF in
> bonding mod
Hi,
> -Original Message-
> From: dev On Behalf Of Xueming Li
> Sent: Tuesday, November 17, 2020 1:01 PM
> To: Matan Azrad ; Slava Ovsiienko
>
> Cc: dev@dpdk.org; Xueming(Steven) Li ; Asaf Penso
> ; sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH] net/mlx5: fix assertion check warnings
>
>
Currently, header reformat action uses the hash list 32-bit key generated
in header reformat register function directly. The key will not be
recalculated in the hash list function.
As the 64-bit key is composed of the 32-bit attributes and 32-bit reformat
buffer csum, the hash list function only g
[PATCH v2 1/2] ethdev: add RSS offload types.
[PATCH v2 2/2] app/testpmd: add cmdline support for RSS types.
v2:
* Update cmd_config_rss.help_str and cmd_help_long_parsed() to add the ecpri.
* Update the documentation to add the ecpri.
Simei Su (2):
ethdev: add eCPRI RSS offload types
app/tes
This patch defines new RSS offload types for eCPRI. For eCPRI with
Message Type 0, the hash field is physical channel ID.
Signed-off-by: Simei Su
Reviewed-by: Ferruh Yigit
---
lib/librte_ethdev/rte_ethdev.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/librte_ethdev/rte_ethdev.h b/lib
This patch adds testpmd cmdline support for eCPRI.
Signed-off-by: Simei Su
---
app/test-pmd/cmdline.c | 13 -
app/test-pmd/config.c | 3 ++-
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 2 +-
3 files changed, 11 insertions(+), 7 deletions
Hi Weifeng,
Since the issue happens only in the flow flush when the port status change.
The root cause is the LSC callback should not be called before the port start
is done.
BR,
SuanmingMou
> -Original Message-
> From: dev On Behalf Of Weifeng Li
> Sent: Sunday, November 8, 2020 3:37
Hi David,
Snice some protocols are only supported by scapy 2.4.4, it's recemented that
scapy 2.4.4 as major version for DTS.
However perf users might not care these advanced protocols, to be compatible
with trex, dts captured these exceptions and let it go through.
Fix was contained in the l
Hi Yunjian,
As the code changed, I think this patch is not needed anymore.
Thanks,
SuanmingMou
On 3/31/2020 6:24 PM, wangyunjian wrote:
From: Yunjian Wang
This NULL check is unnecessary, container_of is never NULL.
Signed-off-by: Yunjian Wang
---
drivers/net/mlx5/mlx5.c | 3 ---
1 file
Currently, header reformat action uses the hash list 32-bit key generated
in header reformat register function directly. The key will not be
recalculated in the hash list function.
As the 64-bit key is composed of the 32-bit attributes and 32-bit reformat
buffer csum, the hash list function only g
Tested-by: Zhou, Jun
-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Xiaoyu Min
Sent: Monday, November 16, 2020 3:55 PM
To: Singh, Jasvinder ; Dumitrescu, Cristian
Cc: dev@dpdk.org; Dekel Peled
Subject: [dpdk-dev] [PATCH 5/5] net/softnic: update headers size cal
OK, please reject this patch.
Thanks,
Yunjian
> -Original Message-
> From: Suanming Mou [mailto:suanmi...@nvidia.com]
> Sent: Wednesday, November 18, 2020 10:16 AM
> To: wangyunjian ; dev@dpdk.org
> Cc: ma...@mellanox.com; shah...@mellanox.com;
> viachesl...@mellanox.com; Lilijun (Jerry)
In commit 681a67288 ("usertools: check if module is loaded before
binding"), script will exit if no driver is found in /sys/module/.
However, for build-in kernel driver, /sys/module/MODULENAME only
shows up if it has a version or at least one parameter. Take ixgbe
for example, after kernel commit
In commit 681a67288655 ("usertools: check if module is loaded before
binding"), script will exit if no driver is found in /sys/module/.
However, for built-in kernel driver, /sys/module/MODULENAME only
shows up if it has a version or at least one parameter. Take ixgbe
for example, after kernel comm
* Cao, Yahui [2020-11-17 23:31:40 +0800]:
> Hi Zhirun,
>
> I think it is great to refactor ice fdir to differentiate inner and
> outer field vector in a more clear way. Thanks for your commit.
> It seems there still needs some effort to complete this patch.
>
> > -Origina
HI， Ferruh,
Ok，I will send V3 patches. thanks.
在 2020/11/12 17:52, Ferruh Yigit 写道:
On 11/12/2020 2:28 AM, Min Hu (Connor) wrote:
Hi Ferruh, any suggestions?
在 2020/11/3 14:30, Min Hu (Connor) 写道:
Hi Ferruh,
I agree with your proposal. But if we remove record structures, we will
not
On Mon, Nov 16, 2020 at 12:28 PM Thomas Monjalon wrote:
>
> 16/11/2020 19:28, Radha Mohan:
> > On Fri, Nov 13, 2020 at 2:39 PM Thomas Monjalon wrote:
> > > 13/11/2020 20:18, Radha Mohan:
> > > > On Tue, Nov 10, 2020 at 11:57 PM Mahipal Challa
> > > > wrote:
> > > > From: Radha Mohan
> > > > Se
> -Original Message-
> From: Yan, Zhirun
> Sent: Wednesday, November 18, 2020 11:15 AM
> To: Cao, Yahui
> Cc: dev@dpdk.org; Zhang, Qi Z ; Wang, Xiao W
> ; Su, Simei
> Subject: [PATCH v1] net/ice: refactor flow pattern parser
>
> * Cao, Yahui [2020-11-17 23:31:40 +0800]:
>
> > Hi Z
On Mon, Nov 16, 2020 at 8:13 AM Ferruh Yigit wrote:
> On 11/16/2020 7:55 AM, Xiaoyu Min wrote:
> > From: Xiaoyu Min
> >
> > The rte_flow_item_eth and rte_flow_item_vlan items are refined.
> > The structs do not exactly represent the packet bits captured on the
> > wire anymore so should only cop
https://bugs.dpdk.org/show_bug.cgi?id=564
junx.w.z...@intel.com (junx.w.z...@intel.com) changed:
What|Removed |Added
Resolution|--- |FIXED
From: Beilei Xing
PMD configures the global register I40E_GLINT_CTL during
device initialization to work around the Rx write back
issue. But when a device is bound from DPDK to kernel,
the global register is not recovered to the original
state, it will cause kernel driver performance drop issue.
> -Original Message-
> From: David Marchand
> Sent: Thursday, November 12, 2020 7:58 PM
> To: jer...@marvell.com; Ruifeng Wang
> Cc: dev@dpdk.org; tho...@monjalon.net; Ferruh Yigit
> ; Luca Boccassi ; nd
>
> Subject: Re: [dpdk-dev] DPDK Release Status Meeting 12/11/2020
>
> On Thu, Nov
This patch series are mainly to enable checksum offloading for IXGBE NEON
vector PMD, including good and bad checksum flags. In the meanwhile, the
first patch enable VLAN stripping flag for Arm.
Feifei Wang (4):
net/ixgbe: add VLAN stripping support for Arm
net/ixgbe: support checksum flags fo
Implemented PKT_RX_VLAN_STRIPPED for vector PMD on Arm platform.
Test Results:
NICs: 82599(igb)
Dirver: ixgbe(vector)
Packet: vlan-id=1
$:./app/dpdk-testpmd -c 0x3 -w 0002:f9:00.0 -- -i --port-topology=chained \
to enable vlan stripping: --enable-hw-vlan-strip
test-pmd> set fwd rxonly
test-pmd> s
Updated desc_to_olflags_v() to support PKT_RX_IP_CKSUM_BAD and
PKT_RX_L4_CKSUM_BAD in the ol_flags of the mbuf for Arm neon vector
Rx function.
Test Results:
NICs: 82599(igb)
Dirver: ixgbe(vector)
Packet: IPv4_checksum = 0xee && UDP_checksum = 0xee
$:./app/dpdk-testpmd -c 0x3 -w 0002:f9:00.0 -- -
From: feiwan02
Add CKSUM_GOOD flag to distinguish a good checksum from an unknown one
in neon vertor RX function.
Test Results:
NICs: 82599(igb)
Dirver: ixgbe(vector)
Packet: IPv4_checksum = correct value && UDP_checksum = correct value
$:./app/dpdk-testpmd -c 0x3 -w 0002:f9:00.0 -- -i --port-t
From: feiwan02
IXGBE NEON vector PMD now supports checksum offloading, hence can be used
when DEV_RX_OFFLOAD_CHECKSUM is set.
Signed-off-by: Feifei Wang
Reviewed-by: Ruifeng Wang
---
drivers/net/ixgbe/ixgbe_rxtx_vec_neon.c | 6 --
1 file changed, 6 deletions(-)
diff --git a/drivers/net/i
On 11/18/20 3:34 AM, Ajit Khaparde wrote:
> On Mon, Nov 16, 2020 at 8:13 AM Ferruh Yigit wrote:
>
>> On 11/16/2020 7:55 AM, Xiaoyu Min wrote:
>>> From: Xiaoyu Min
>>>
>>> The rte_flow_item_eth and rte_flow_item_vlan items are refined.
>>> The structs do not exactly represent the packet bits capt
> -Original Message-
> From: Feifei Wang
> Sent: Wednesday, November 18, 2020 14:29
> To: Jerin Jacob ; Ruifeng Wang ;
> Guo, Jia
> ; Wang, Haiyue
> Cc: dev@dpdk.org; n...@arm.com; feiwan02
> Subject: [PATCH v1 3/4] net/ixgbe: implement good checksum flag for NEON
> vector
>
> From: f
This patch series are mainly to enable checksum offloading for IXGBE NEON
vector PMD, including good and bad checksum flags. In the meanwhile, the
first patch enable VLAN stripping flag for Arm.
Feifei Wang (4):
net/ixgbe: add VLAN stripping support for Arm
net/ixgbe: support checksum flags fo
Implemented PKT_RX_VLAN_STRIPPED for vector PMD on Arm platform.
Test Results:
NICs: 82599(igb)
Driver: ixgbe(vector)
Packet: vlan-id=1
$:./app/dpdk-testpmd -c 0x3 -w 0002:f9:00.0 -- -i --port-topology=chained \
to enable vlan stripping: --enable-hw-vlan-strip
test-pmd> set fwd rxonly
test-pmd> s
Updated desc_to_olflags_v() to support PKT_RX_IP_CKSUM_BAD and
PKT_RX_L4_CKSUM_BAD in the ol_flags of the mbuf for Arm neon vector
Rx function.
Test Results:
NICs: 82599(igb)
Driver: ixgbe(vector)
Packet: IPv4_checksum = 0xee && UDP_checksum = 0xee
$:./app/dpdk-testpmd -c 0x3 -w 0002:f9:00.0 -- -
Add CKSUM_GOOD flag to distinguish a good checksum from an unknown one
in neon vertor RX function.
Test Results:
NICs: 82599(igb)
Driver: ixgbe(vector)
Packet: IPv4_checksum = correct value && UDP_checksum = correct value
$:./app/dpdk-testpmd -c 0x3 -w 0002:f9:00.0 -- -i --port-topology=chained
t
IXGBE NEON vector PMD now supports checksum offloading, hence can be used
when DEV_RX_OFFLOAD_CHECKSUM is set.
Signed-off-by: Feifei Wang
Reviewed-by: Ruifeng Wang
---
drivers/net/ixgbe/ixgbe_rxtx_vec_neon.c | 6 --
1 file changed, 6 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_rxtx_v
The MLX5_ENCAPSULATION_DECISION_SIZE constant is used
to check the raw encap/decap actions for the raw header
size. The header is constructed of the rte_xxx_hdr
structures instead of rte items. Hence, constant
must be defined with rte_xxx_hdr structure sizes.
Fixes: 50f576d657d7 ("net/mlx5: fix VL
Hi, Haiyue
> -邮件原件-
> 发件人: Wang, Haiyue
> 发送时间: 2020年11月18日 15:22
> 收件人: Feifei Wang ; jer...@marvell.com; Ruifeng
> Wang ; Guo, Jia
> 抄送: dev@dpdk.org; nd
> 主题: RE: [PATCH v1 3/4] net/ixgbe: implement good checksum flag for NEON
> vector
>
> > -Original Message-
> > From: Feif
98 matches
Mail list logo
