Number of rx queue and number of rx port in lcore config are constants
during the period of l3 forward application running. But compiler has
no this information.
Copied values from lcore config to local variables and used the local
variables for iteration. Compiler can see that the local variables
Both L2 and L3 headers will be used in forward processing. And these
two headers are in the same cache line. It has the same effect for
prefetching with L2 header address and prefetching with L3 header
address.
Changed to use L2 header address for prefetching. The change showed
no measurable perfo
This series of patches include changes to l3fwd example application.
Some improvements are made for better usage of CPU cycles and memory.
v3:
Dropped 1/3 code reorg from v2.
v2:
Dropped 1/4 prefetch to write change from v1.
Dropped 4/4 data struct change from v1.
Added 1/3 code reorganize.
Updat
On Tue, Jun 1, 2021 at 3:11 AM Tomasz Duszynski wrote:
>
> This series adds initial support for baseband PHY available on SOCs
> belonging to Fusion family. BPHY is a hardware block comprising
> accelerators and DSPs specifically tailored for 5G/LTE usecases.
5G/LTE inline use cases.
>
> This se
On 2021/6/9 18:25, Ananyev, Konstantin wrote:
>>> On 4/23/21 12:46 PM, Chengchang Tang wrote:
To use the HW offloads capability (e.g. checksum and TSO) in the Tx
direction, the upper-layer users need to call rte_eth_dev_prepare to do
some adjustment to the packets before sending t
On Tue, Jun 1, 2021 at 3:13 AM Tomasz Duszynski wrote:
>
> Add baseband phy cgx/rpm skeleton driver. At this point
> it merely probes a matching device.
>
> Signed-off-by: Tomasz Duszynski
> Signed-off-by: Jakub Palider
> ---
> MAINTAINERS | 5 +
> doc/guides/rawdevs
Support AVF RSS for inner most header of GTPoGRE packet. It supports
RSS based on inner most IP src + dst address and TCP/UDP src + dst
port.
Signed-off-by: Lingyu Liu
---
drivers/net/iavf/iavf_hash.c | 154 +--
1 file changed, 127 insertions(+), 27 deletions(-)
Support AVF FDIR for inner header of GTPoGRE tunnel packet.
++---+
|Pattern |Input Set |
++---+
|eth/ipv4/gre/ipv4/gtpu/(
Add a virtchnl protocol header type to support AVF FDIR and RSS for GRE.
Signed-off-by: Lingyu Liu
---
drivers/common/iavf/virtchnl.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/common/iavf/virtchnl.h b/drivers/common/iavf/virtchnl.h
index 3a60faff93..197edce8a1 100644
--- a/driv
Support AVF RSS and FDIR for GTPoGRE packet.
Lingyu Liu (4):
net/iavf: support flow pattern for GTPoGRE
common/iavf: add header types for GRE
net/iavf: support AVF FDIR for GTPoGRE tunnel packet
net/iavf: support AVF RSS for GTPoGRE packet
---
V5 change:
- refine protocol header for tw
Add GTPoGRE pattern support for AVF FDIR and RSS.
Signed-off-by: Lingyu Liu
---
drivers/net/iavf/iavf_generic_flow.c | 600 +++
drivers/net/iavf/iavf_generic_flow.h | 80
2 files changed, 680 insertions(+)
diff --git a/drivers/net/iavf/iavf_generic_flow.c
b/driver
> -Original Message-
> From: Thomas Monjalon
> Sent: Tuesday, June 8, 2021 3:53 PM
> To: Xueming(Steven) Li
> Cc: dev@dpdk.org; Parav Pandit ; Ray Kinsella
> ; Wang Haiyue ;
> andrew.rybche...@oktetlabs.ru
> Subject: Re: [dpdk-dev] [RFC v2] bus/auxiliary: introduce auxiliary bus
>
> 1
> -Original Message-
> From: Wang, Haiyue
> Sent: Tuesday, June 8, 2021 4:41 PM
> To: NBU-Contact-Thomas Monjalon ; Xueming(Steven) Li
>
> Cc: dev@dpdk.org; Parav Pandit ; Ray Kinsella
> ; andrew.rybche...@oktetlabs.ru
> Subject: RE: [dpdk-dev] [RFC v2] bus/auxiliary: introduce auxil
Hi, Andrew and Ananyev
On 2021/6/9 17:37, Andrew Rybchenko wrote:
> On 6/9/21 12:11 PM, Ananyev, Konstantin wrote:
>>
>>>
>>>
>>> On 2021/6/8 17:49, Andrew Rybchenko wrote:
"for bonding" is redundant in the summary since it is already
"net/bonding"
On 4/23/21 12:46 PM, Chengcha
On Tue, Jun 1, 2021 at 3:12 AM Tomasz Duszynski wrote:
>
> Add support for setting link mode.
>
> Signed-off-by: Tomasz Duszynski
> Signed-off-by: Jakub Palider
> ---
> drivers/common/cnxk/roc_bphy_cgx.c | 28
> drivers/common/cnxk/roc_bphy_cgx.h | 11 +
> drivers/com
On Tue, Jun 1, 2021 at 3:12 AM Tomasz Duszynski wrote:
>
> Add support for enabling or disabling internal loopback.
>
> Signed-off-by: Tomasz Duszynski
> Signed-off-by: Jakub Palider
In all the git commit message we can shorten
From:
common/cnxk: add support for changing internal loopback
To
在 2021/6/8 18:29, Andrew Rybchenko 写道:
What is the difference of v2 from v1?
Please, try to add version changelog in the future.
ok😁
On 4/28/21 11:36 AM, Min Hu (Connor) wrote:
This patchset contains fixes for link speed in testpmd.
Huisong Li (2):
app/testpmd: add link speed check befor
在 2021/6/8 18:49, Andrew Rybchenko 写道:
On 4/28/21 11:36 AM, Min Hu (Connor) wrote:
From: Huisong Li
Currently, to check whether the configured link_speeds is valid, we
have to run "port start". In addition, if the configuration fails,
"port->dev_conf.link_speeds" maintained in testpmd cannot
From: Zhihong Peng
AddressSanitizer (ASan) is a google memory error detect
standard tool. It could help to detect use-after-free and
{heap,stack,global}-buffer overflow bugs in C/C++ programs,
print detailed error information when error happens, large
improve debug efficiency.
By referring to it
On Tue, Jun 1, 2021 at 3:12 AM Tomasz Duszynski wrote:
>
> Add support for retrieving link information.
>
> Signed-off-by: Tomasz Duszynski
> Signed-off-by: Jakub Palider
> ---
> +enum roc_bphy_cgx_eth_link_speed {
> + ROC_BPHY_CGX_ETH_LINK_SPEED_NONE,
> + ROC_BPHY_CGX_ETH_LINK_SPEED
On Tue, Jun 1, 2021 at 3:12 AM Tomasz Duszynski wrote:
>
> Messages can be exchanged between userspace software and firmware
> via set of two dedicated registers, namely scratch1 and scratch0.
>
> scratch1 acts as a command register i.e message is sent to firmware,
> while scratch0 holds response
Hello,
There was no activity that patch for a long time.
The patch is marked as failed, but we verified failed tests and concluded that
the failures can be ignored.
https://patchwork.dpdk.org/project/dpdk/patch/20210527152858.13312-1-getel...@nvidia.com/
How should I proceed with this case ?
Plea
> -Original Message-
> From: Kinsella, Ray
> Sent: Tuesday, May 11, 2021 5:48 PM
> To: Xueming(Steven) Li ; NBU-Contact-Thomas Monjalon
>
> Cc: dev@dpdk.org; Parav Pandit ; Neil Horman
> ; Wang Haiyue
>
> Subject: Re: [RFC v2] bus/auxiliary: introduce auxiliary bus
>
// snip //
>
L2 RSS support was deleted by mistake during code
refactoring. This patch adds it again.
Fixes: 38d632cbdc88 ("net/ice: refactor PF RSS")
Signed-off-by: Wenjun Wu
---
drivers/net/ice/base/ice_flow.c | 2 +-
drivers/net/ice/ice_hash.c | 17 +
2 files changed, 18 insertions(
IPv4 and IPv6 fragment ptypes are supposed to be separated from IP
other ptypes. New bitmaps for IP fragment ptypes were created, but the
IP fragment ptypes were not deleted from the previous non-frag bitmaps,
which will cause conflicts. This patch removes IP fragment ptypes from
the non-frag bitma
Readd other long options case in l2fwd_parse_args function
to ensure all long options will work well.
Fixes: fa19eb20d2126d8bc63acc8f336a353dfaf8c354 ("examples/l2fwd: add
forwarding port mapping option")
Cc: sta...@dpdk.org
Signed-off-by: SunChengLian
---
examples/l2fwd/main.c | 3 +++
1 file
2021-06-09 02:04 (UTC+0300), Dmitry Kozlyuk:
> 2021-06-04 16:44 (UTC-0700), Narcisa Ana Maria Vasile:
> [...]
> > diff --git a/lib/eal/include/rte_thread_types.h
> > b/lib/eal/include/rte_thread_types.h
> > index d67b24a563..7bb0d2948c 100644
> > --- a/lib/eal/include/rte_thread_types.h
> > +++ b/
Hi Dmitry,
Those failures are the result of a known issue with that machine. Those
tests were disabled on that machine shortly after you submitted that patch.
I've re-run the patch.
Owen Hilyard
On Wed, Jun 9, 2021 at 12:48 PM Dmitry Kozlyuk
wrote:
> 2021-06-09 12:01 (UTC-0400), Lincoln Lavoie
2021-06-09 12:01 (UTC-0400), Lincoln Lavoie:
> Hi Dmitry,
>
> If the failing test is the unit test func_reentrancy_autotest, that is
> being looked into , as it seems like the test case does not reliably run.
> It passes / fails between different systems running both on bare metal and
> in virtual
Hi Dmitry,
If the failing test is the unit test func_reentrancy_autotest, that is
being looked into , as it seems like the test case does not reliably run.
It passes / fails between different systems running both on bare metal and
in virtual environments, on different OSes and architectures.
Do y
2021-05-28 00:24 (UTC+0300), Dmitry Kozlyuk:
> Size of string constant symbol may be larger than its length
> measured up to NUL terminator. In this case pmdinfogen included padding
> bytes after NUL terminator in generated source, yielding incorrect code.
>
> Always trim string data to NUL termin
2021-05-20 20:17 (UTC+), Akhil Goyal:
> >
> > 2021-05-20 18:59 (UTC+), Akhil Goyal:
> > > > Windows system headers define `s_addr`, `min`, and `max` macros which
> > > > break structure definitions containing fields with one of these names.
> > > > Undefining those macros would break con
Hi,
> -Original Message-
> From: dev On Behalf Of Matan Azrad
> Sent: Wednesday, June 9, 2021 3:33 PM
> To: dev@dpdk.org
> Cc: Slava Ovsiienko ; sta...@dpdk.org; Bing Zhao
>
> Subject: [dpdk-dev] [PATCH] net/mlx5: fix aging counter deallocation
>
> When a counter is destroyed and used f
On 6/8/21 9:49 PM, Thomas Monjalon wrote:
> 08/06/2021 18:32, Andrew Rybchenko:
>> On 6/8/21 7:13 PM, Thomas Monjalon wrote:
>>> We will be at least 4 tomorrow Wednesday 3pm UTC.
>>> Everybody is welcome to join the community meeting:
>>> https://zoom.us/j/93391811485
>> Invite has July, 7. Typ
This patch creates a new function for handling PF probing of a secondary
process. A CPP handle is obtained for the CPP bridge service and the
service itself is also registered during secondary process
initialization. DPDK services aren't shared between processes so it is
not enough to only have the
When using rte_eth_dev_pci_generic_probe() during probing a
rte_eth_dev will be created with the name field corresponding to the PCI
address of the device. NFP4000/6000 devices only have a single PF (but
potentially multiple physical ports). This means that in a simple two
port example the rte_eth_
Hi,
> -Original Message-
> From: Li Zhang
> Sent: Wednesday, June 9, 2021 5:07 AM
> To: Ori Kam ; Slava Ovsiienko
> ; Matan Azrad ; NBU-
> Contact-Thomas Monjalon
> Cc: dev@dpdk.org; Raslan Darawsheh ; Asaf Penso
> ; sta...@dpdk.org
> Subject: [PATCH] net/mlx5: fix meter policy creation
Hi,
> -Original Message-
> From: Slava Ovsiienko
> Sent: Monday, June 7, 2021 12:37 PM
> To: dev@dpdk.org
> Cc: Raslan Darawsheh ; Matan Azrad
> ; sta...@dpdk.org
> Subject: [PATCH] common/mlx5: add provider query port support to glue
> library
>
> The rdma-core mlx5 provider introduced
On Wed, Jun 09, 2021 at 02:28:04PM +0200, David Marchand wrote:
> On Wed, Jun 9, 2021 at 2:02 PM Shanmugasundaram M wrote:
> > We are using 6Wind fastpath(fp-rte) which is using DPDK version 18.11.1.
> > We are seeing 65.5G(preallocated during startup) of virtual memory for
> > fp-rte on a 32GB
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Liguzinski,
> WojciechX
> Sent: Wednesday, 9 June 2021 10.37
>
> > From: Morten Brørup
> > Sent: Tuesday, May 25, 2021 11:17 AM
> >
> > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Liguzinski,
> > > WojciechX
> > > Sent: Monday, 24
When a counter is destroyed and used for aging action, the driver should
remove the counter object from the age-out list if it is there.
The counter memory of the list entry and of the counter shared
information is shared because, currently, shared counter cannot be used
for aging.
When the suppo
On Wed, Jun 9, 2021 at 2:02 PM Shanmugasundaram M wrote:
> We are using 6Wind fastpath(fp-rte) which is using DPDK version 18.11.1.
> We are seeing 65.5G(preallocated during startup) of virtual memory for
> fp-rte on a 32GB machine, 10G of that was huge pages (5000 mappings of 2M.)
>
> We found
This patch creates a new function for handling PF probing of a secondary
process. A CPP handle is obtained for the CPP bridge service and the
service itself is also registered during secondary process
initialization. DPDK services aren't shared between processes so it is
not enough to only have the
When using rte_eth_dev_pci_generic_probe() during probing a
rte_eth_dev will be created with the name field corresponding to the PCI
address of the device. NFP4000/6000 devices only have a single PF (but
potentially multiple physical ports). This means that in a simple two
port example the rte_eth_
BTW, I use it for primary and secondary communication, so that will use memif
to share data between primary and secondary. Thanks.
Best Regards,
Dave(Jianyue)
From: Wu, Jianyue (NSB - CN/Hangzhou)
Sent: 2021年6月9日 19:26
To: us...@dpdk.org; 'dev@dpdk.org'
Subject: Does memif support primary<->sec
Dear Team,
We are using 6Wind fastpath(fp-rte) which is using DPDK version 18.11.1.
We are seeing 65.5G(preallocated during startup) of virtual memory for
fp-rte on a 32GB machine, 10G of that was huge pages (5000 mappings of 2M.)
We found 4 strange anon (4K size) segments totaling 54.5G (16+1
Hi all,
Thanks Kevin's feedback, there are some patches missing between
v21.05-rc1..v21.05.
Will roll out rc2 to include them all, please hold test and verification.
Best Regards,
Xueming
> -Original Message-
> From: Xueming(Steven) Li
> Sent: Tuesday, June 1, 2021 3:55 PM
> Cc: dev@d
On Wed, Jun 9, 2021 at 3:48 PM Michael Pfeiffer
wrote:
>
> Including various headers may fail for ARM builds with 'Platform must
> be built with RTE_FORCE_INTRINSICS' if rte_config.h is not included
> before. Move the error message after the includes to ensure rte_config.h
> is always included.
>
Hello,
May I ask a question about memif?
When I'm trying to develop a communication mechanism for two processes, which
also need to share memzones between the processes, so need primary<->secondary.
https://doc.dpdk.org/guides/nics/memif.html
But from the guideline, it seems always use primary<->
On Tue, Jun 08, 2021 at 12:04:45PM +, Pavan Nikhilesh Bhagavatula wrote:
>
>
> >
> >+static __rte_always_inline void
> >+cn10k_nix_xmit_prepare_tstamp(uintptr_t lmt_addr, const uint64_t
> >*cmd,
> >+ const uint64_t ol_flags, const uint16_t
> >no_segdw,
> >+
On Tue, Jun 08, 2021 at 03:26:08PM +0300, Andrew Rybchenko wrote:
> On 6/7/21 8:58 PM, Nithin Dabilpuram wrote:
> > From: Srujana Challa
> >
> > Updates Rx inline IPSEC mailbox message format to make it
> > sync with latest CPT PF driver.
> >
> > Signed-off-by: Srujana Challa
> > ---
> > drive
Adding the PIE support for IP Pipeline
Signed-off-by: Liguzinski, WojciechX
---
examples/ip_pipeline/tmgr.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/examples/ip_pipeline/tmgr.c b/examples/ip_pipeline/tmgr.c
index e4e364cbc0..73da2da870 100644
--- a/examples/ip_pi
patch add support enable PIE or RED by
parsing config file.
Signed-off-by: Liguzinski, WojciechX
---
config/rte_config.h | 1 -
examples/qos_sched/app_thread.c | 1 -
examples/qos_sched/cfg_file.c | 82 ++---
examples/qos_sched/init.c | 7 +-
examples/qos_sched
Implement PIE based congestion management based on rfc8033
Signed-off-by: Liguzinski, WojciechX
---
drivers/net/softnic/rte_eth_softnic_tm.c | 6 +-
lib/sched/meson.build| 10 +-
lib/sched/rte_pie.c | 79 +
lib/sched/rte_pie.h
DPDK sched library is equipped with mechanism that secures it from the
bufferbloat problem
which is a situation when excess buffers in the network cause high latency and
latency
variation. Currently, it supports RED for active queue management (which is
designed
to control the queue length but
On Wed, Jun 09, 2021 at 09:38:04AM +0800, Huisong Li wrote:
>
> 在 2021/6/8 1:58, Nithin Dabilpuram 写道:
> > Add build infrastructure and common probe and remove for cnxk driver
> > which is used by both CN10K and CN9K SoC.
> >
> > Signed-off-by: Nithin Dabilpuram
> > ---
> > MAINTAINERS
> > On 4/23/21 12:46 PM, Chengchang Tang wrote:
> >> To use the HW offloads capability (e.g. checksum and TSO) in the Tx
> >> direction, the upper-layer users need to call rte_eth_dev_prepare to do
> >> some adjustment to the packets before sending them (e.g. processing
> >> pseudo headers when Tx
On 6/8/21 11:48 PM, Stephen Hemminger wrote:
> On Tue, 8 Jun 2021 18:55:17 +0300
> Andrew Rybchenko wrote:
>
>> On 6/8/21 6:42 PM, Stephen Hemminger wrote:
>>> On Tue, 8 Jun 2021 11:00:37 +0300
>>> Andrew Rybchenko wrote:
>>>
On 4/19/21 8:08 PM, Thomas Monjalon wrote:
> About the t
> -Original Message-
> From: Michael Pfeiffer
> Sent: Wednesday, June 9, 2021 6:18 PM
> To: Jan Viktorin ; Ruifeng Wang
> ; jer...@marvell.com
> Cc: dev@dpdk.org; Michael Pfeiffer ;
> sta...@dpdk.org; ko...@rehivetech.com; Phil Yang ;
> Joyce Kong
> Subject: [PATCH v5] eal: arm: fix out o
Including various headers may fail for ARM builds with 'Platform must
be built with RTE_FORCE_INTRINSICS' if rte_config.h is not included
before. Move the error message after the includes to ensure rte_config.h
is always included.
Fixes: de966ccdcd7f ("eal/arm: add byte order operations for ARM")
Hi,
> -Original Message-
> From: Alexander Kozyrev
> Sent: Tuesday, May 25, 2021 8:14 PM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Raslan Darawsheh ; Slava
> Ovsiienko ; Matan Azrad ;
> Ori Kam
> Subject: [PATCH] net/mlx5: allow copy from one tag to another
>
> The modify field implemen
On 6/9/21 12:11 PM, Ananyev, Konstantin wrote:
>
>>
>>
>> On 2021/6/8 17:49, Andrew Rybchenko wrote:
>>> "for bonding" is redundant in the summary since it is already
>>> "net/bonding"
>>>
>>> On 4/23/21 12:46 PM, Chengchang Tang wrote:
Currently, the TX offloading of the bonding device will
On 6/9/21 9:42 AM, Chengchang Tang wrote:
> Hi, Andrew and Ferruh
>
> On 2021/6/8 17:49, Andrew Rybchenko wrote:
>> "for bonding" is redundant in the summary since it is already "net/bonding".
>>
>> On 4/23/21 12:46 PM, Chengchang Tang wrote:
>>> To use the HW offloads capability (e.g. checksum an
>
>
> On 2021/6/8 17:49, Andrew Rybchenko wrote:
> > "for bonding" is redundant in the summary since it is already
> > "net/bonding"
> >
> > On 4/23/21 12:46 PM, Chengchang Tang wrote:
> >> Currently, the TX offloading of the bonding device will not take effect by
> >
> > TX -> Tx
> >
> >> using
> -Original Message-
> From: Morten Brørup
> Sent: Tuesday, May 25, 2021 11:17 AM
> To: Liguzinski, WojciechX ; dev@dpdk.org;
> Singh, Jasvinder ; Dumitrescu, Cristian
>
> Cc: Dharmappa, Savinay
> Subject: RE: [dpdk-dev] [RFC PATCH 1/3] sched: add pie based congestion
> management
> -Original Message-
> From: Michael Pfeiffer
> Sent: Tuesday, June 8, 2021 6:29 PM
> To: Jan Viktorin ; Ruifeng Wang
> ; jer...@marvell.com
> Cc: dev@dpdk.org; Michael Pfeiffer ;
> ko...@rehivetech.com; sta...@dpdk.org
> Subject: [PATCH v4] eal: arm: fix out of tree build
>
> Including v
From: Dapeng Yu
The rte_eth_devices array is not in share memory, it should not be
referenced by i40e_adapter which is shared by primary and secondary.
Any process set i40e_adapter->eth_dev will corrupt another process'
context.
The patch removed the field "eth_dev" from i40e_adapter.
Now, when
From: Dapeng Yu
The flag use_avx2 and use_avx512 are defined as local variables, they
will not be aware by the secondary process, then wrong data path is
selected. Fix the issue by moving them into struct i40e_adapter.
Fixes: 6ada10deac66 ("net/i40e: remove devarg use-latest-supported-vec")
Fixe
68 matches
Mail list logo
