17/10/2022 10:32, Andrew Rybchenko:
> On 10/6/22 21:35, Hanumanth Pothula wrote:
> > Presently, Rx metadata is sent to PMD by default, leading
> > to a performance drop as processing for the same in rx path
> > takes extra cycles.
> >
> > Hence, introducing command line argument, 'nic-to-pmd-rx-me
This patchset introduced the idpf (Infrastructure Data Path Function)
PMD in DPDK for Intel® IPU E2000 (Device ID: 0x1452).
The Intel® IPU E2000 targets to deliver high performance under real
workloads with security and isolation.
Please refer to
https://www.intel.com/content/www/us/en/products/net
Support device init and add the following dev ops:
- dev_configure
- dev_close
- dev_infos_get
Signed-off-by: Beilei Xing
Signed-off-by: Xiaoyun Li
Signed-off-by: Xiao Wang
Signed-off-by: Wenjun Wu
Signed-off-by: Junfeng Guo
---
MAINTAINERS| 9 +
doc/guides/n
Add support for tx_queue_setup ops.
In the single queue model, the same descriptor queue is used by SW to
post buffer descriptors to HW and by HW to post completed descriptors
to SW.
In the split queue model, "RX buffer queues" are used to pass
descriptor buffers from SW to HW while Rx queues are
Add support for rx_queue_setup ops.
Signed-off-by: Beilei Xing
Signed-off-by: Xiaoyun Li
Signed-off-by: Junfeng Guo
---
drivers/net/idpf/idpf_ethdev.c | 11 +
drivers/net/idpf/idpf_rxtx.c | 400 +
drivers/net/idpf/idpf_rxtx.h | 46
3 files changed, 45
Add dev ops dev_start, dev_stop and link_update.
Signed-off-by: Beilei Xing
Signed-off-by: Xiaoyun Li
Signed-off-by: Junfeng Guo
---
drivers/net/idpf/idpf_ethdev.c | 55 ++
drivers/net/idpf/idpf_rxtx.c | 20 +
2 files changed, 75 insertions(+)
dif
Add support for these device ops:
- rx_queue_start
- tx_queue_start
Signed-off-by: Beilei Xing
Signed-off-by: Xiaoyun Li
Signed-off-by: Junfeng Guo
---
drivers/net/idpf/idpf_ethdev.c | 40 ++-
drivers/net/idpf/idpf_ethdev.h | 9 +
drivers/net/idpf/idpf_rxtx.c | 237 +++--
dr
Add support for these device ops:
- rx_queue_stop
- tx_queue_stop
Signed-off-by: Beilei Xing
Signed-off-by: Xiaoyun Li
Signed-off-by: Junfeng Guo
---
doc/guides/nics/features/idpf.ini | 1 +
drivers/net/idpf/idpf_ethdev.c| 14 ++-
drivers/net/idpf/idpf_rxtx.c | 148 ++
Add support for queue operations:
- rx_queue_release
- tx_queue_release
Signed-off-by: Beilei Xing
Signed-off-by: Xiaoyun Li
Signed-off-by: Junfeng Guo
---
drivers/net/idpf/idpf_ethdev.c | 2 +
drivers/net/idpf/idpf_rxtx.c | 81 ++
drivers/net/idpf/idpf_rxt
Add dev ops mtu_set.
Signed-off-by: Beilei Xing
Signed-off-by: Junfeng Guo
---
doc/guides/nics/features/idpf.ini | 1 +
drivers/net/idpf/idpf_ethdev.c| 14 ++
2 files changed, 15 insertions(+)
diff --git a/doc/guides/nics/features/idpf.ini
b/doc/guides/nics/features/idpf.ini
Add basic Rx support in split queue mode and single queue mode.
Signed-off-by: Beilei Xing
Signed-off-by: Xiaoyun Li
Signed-off-by: Junfeng Guo
---
drivers/net/idpf/idpf_ethdev.c | 2 +
drivers/net/idpf/idpf_rxtx.c | 273 +
drivers/net/idpf/idpf_rxtx.h |
Add basic Tx support in split queue mode and single queue mode.
Signed-off-by: Beilei Xing
Signed-off-by: Xiaoyun Li
Signed-off-by: Junfeng Guo
---
drivers/net/idpf/idpf_ethdev.c | 3 +
drivers/net/idpf/idpf_ethdev.h | 1 +
drivers/net/idpf/idpf_rxtx.c | 357 +
Parse packet type during receiving packets.
Signed-off-by: Wenjun Wu
Signed-off-by: Junfeng Guo
---
drivers/net/idpf/idpf_ethdev.c | 6 +
drivers/net/idpf/idpf_ethdev.h | 6 +
drivers/net/idpf/idpf_rxtx.c | 11 ++
drivers/net/idpf/idpf_rxtx.h | 5 +
drivers/net/idpf/idpf_vchnl.c | 2
Enable write back on ITR expire, then packets can be received one by
one.
Signed-off-by: Beilei Xing
Signed-off-by: Junfeng Guo
---
drivers/net/idpf/idpf_ethdev.c | 116 +
drivers/net/idpf/idpf_ethdev.h | 13
drivers/net/idpf/idpf_vchnl.c | 111 +++
Add RSS support.
Signed-off-by: Beilei Xing
Signed-off-by: Junfeng Guo
---
drivers/net/idpf/idpf_ethdev.c | 116 +
drivers/net/idpf/idpf_ethdev.h | 26
drivers/net/idpf/idpf_vchnl.c | 97 +++
3 files changed, 239 insertions(+)
Add Rx offloading support:
- support CHKSUM and RSS offload for split queue model
- support CHKSUM offload for single queue model
Signed-off-by: Beilei Xing
Signed-off-by: Xiaoyun Li
Signed-off-by: Junfeng Guo
---
doc/guides/nics/features/idpf.ini | 5 ++
drivers/net/idpf/idpf_ethdev.c
Add Tx offloading support:
- support TSO
Signed-off-by: Beilei Xing
Signed-off-by: Xiaoyun Li
Signed-off-by: Junfeng Guo
---
doc/guides/nics/features/idpf.ini | 1 +
drivers/net/idpf/idpf_ethdev.c| 4 +-
drivers/net/idpf/idpf_rxtx.c | 128 +-
drivers/n
Add support of AVX512 vector data path for single queue model.
Signed-off-by: Wenjun Wu
Signed-off-by: Junfeng Guo
---
doc/guides/nics/idpf.rst| 19 +
drivers/net/idpf/idpf_ethdev.c | 3 +-
drivers/net/idpf/idpf_ethdev.h | 5 +
drivers/net/idpf/idpf_rxtx.c
Add support for timestamp offload.
Signed-off-by: Wenjing Qiao
Signed-off-by: Junfeng Guo
---
doc/guides/nics/features/idpf.ini | 1 +
drivers/net/idpf/idpf_ethdev.c| 5 +-
drivers/net/idpf/idpf_ethdev.h| 3 ++
drivers/net/idpf/idpf_rxtx.c | 65 ++
drivers/ne
On Thu, Oct 20, 2022 at 1:52 PM Ferruh Yigit wrote:
> >> Honestly I think the status quo is OK:
> >> We have some aliases in some PMD for some historical reason
> >> and everybody looks OK with that. Isn't it?
> >>
> >
> > Well, the inconsistency bugs me a little, but if others feel the status quo
On Tue, Jul 26, 2022 at 11:37 PM Stephen Hemminger
wrote:
>
> There is an option for recording RSS hash with packets in the
> pcapng standard. This implements this for all received packets.
>
> There is a corner case that can not be addressed with current
> DPDK API's. If using rte_flow() and some
Hi Morten,
On Wed, Oct 26, 2022 at 04:44:36PM +0200, Morten Brørup wrote:
> Add __rte_cache_aligned to the objs array.
>
> It makes no difference in the general case, but if get/put operations are
> always 32 objects, it will reduce the number of memory (or last level
> cache) accesses from five
On 10/27/2022 8:58 AM, David Marchand wrote:
On Thu, Oct 20, 2022 at 1:52 PM Ferruh Yigit wrote:
Honestly I think the status quo is OK:
We have some aliases in some PMD for some historical reason
and everybody looks OK with that. Isn't it?
Well, the inconsistency bugs me a little, but if oth
> -Original Message-
> From: Mingjin Ye
> Sent: Thursday, October 27, 2022 1:10 AM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Zhou, YidingX ; Ye, MingjinX
> ; Richardson, Bruce ;
> Konstantin Ananyev ; Yang, Qiming
> ; Zhang, Qi Z ; Lu, Wenzhuo
> ; Junyu Jiang ; Rong, Leyi
> ; Wisam Jaddo ;
> -Original Message-
> From: Mingjin Ye
> Sent: Thursday, October 27, 2022 1:10 AM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Zhou, YidingX ; Ye, MingjinX
> ; Singh, Aman Deep ;
> Zhang, Yuying
> Subject: [PATCH v4 1/2] app/testpmd: fix vlan offload of rxq
>
> After setting vlan offload i
> -Original Message-
> From: Richardson, Bruce
> Sent: Tuesday 25 October 2022 11:39
> To: David Marchand
> Cc: Mattias Rönnblom ; dev@dpdk.org; Morten
> Brørup ; Power, Ciara
>
> Subject: Re: [PATCH v3 1/4] telemetry: support boolean type
>
> On Tue, Oct 25, 2022 at 11:43:27AM +0200
On 10/27/2022 7:13 AM, Thomas Monjalon wrote:
27/10/2022 00:57, Ferruh Yigit:
On 10/26/2022 10:53 PM, Thomas Monjalon wrote:
13/10/2022 18:30, Long Li:
Subject: [PATCH] net/mana: disable driver by default
Driver is disabled by default because its dependencies are not upstreamed
yet, code is a
On 10/24/2022 8:44 AM, Spike Du wrote:
When testpmd quit with mlx5 avail_thresh enabled, a rte timer handler
delays to reconfigure rx queue to re-arm this event. However at the same
time, testpmd is destroying rx queues.
It's never a valid use case for mlx5 avail_thresh. Before testpmd quit,
us
> From: Olivier Matz [mailto:olivier.m...@6wind.com]
> Sent: Thursday, 27 October 2022 10.35
>
> Hi Morten,
>
> On Wed, Oct 26, 2022 at 04:44:36PM +0200, Morten Brørup wrote:
> > Add __rte_cache_aligned to the objs array.
> >
> > It makes no difference in the general case, but if get/put operatio
> -邮件原件-
> 发件人: Thomas Monjalon
> 发送时间: Friday, October 21, 2022 4:42 AM
> 收件人: David Marchand
> 抄送: Hunt, David ; Ruifeng Wang
> ; dev@dpdk.org; nd ; Feifei
> Wang
> 主题: Re: 回复: [PATCH v1 3/3] examples/l3fwd-power: enable PMD power
> mgmt on Arm
>
> 11/10/2022 09:56, Feifei Wang:
> >
> -邮件原件-
> 发件人: Stephen Hemminger
> 发送时间: Friday, October 21, 2022 6:10 AM
> 收件人: Feifei Wang
> 抄送: David Hunt ; dev@dpdk.org; nd
> ; Ruifeng Wang
> 主题: Re: [PATCH v1 3/3] examples/l3fwd-power: enable PMD power mgmt
> on Arm
>
> On Thu, 25 Aug 2022 14:42:51 +0800
> Feifei Wang wrote:
https://bugs.dpdk.org/show_bug.cgi?id=1116
Bug ID: 1116
Summary: Failsafe PMD: Crashes with multiple secondary
processes
Product: DPDK
Version: 21.11
Hardware: All
OS: Linux
Status: UNCONFIRMED
This library has no maintainer and, for now, nobody expressed interest
in taking over.
Mark this experimental library as deprecated and announce plan for
removal in v23.11.
Signed-off-by: David Marchand
---
MAINTAINERS | 3 +--
doc/guides/prog_guide/flow_classify_
> This patchset extends Intel QuickAssist Technology asymmetric crypto PMD.
> Following features were added:
> - ECDH algorithm handling
> - EC point verification
>
> Depends-on: series-25284 ("crypto/qat: fix uncleared cookies in asym")
>
Series applied to dpdk-next-crypto
Thanks.
> > Subject: [dpdk-dev v1] crypto/qat: fix of qat build request session in mp
> >
> > This patch fix the session pointer passed in set_session() when ctx has NULL
> > build request pointer in multi-processes scenario.
> >
> > Fixes: fb3b9f492205 ("crypto/qat: rework burst data path")
> > Cc: sta...
> As the queue pair used in secondary process needs to be set up by
> the primary process, this patch adds an IPC register function to help
> secondary process to send out queue-pair setup request to primary
> process via IPC request messages. A new "qp_in_used_pid" param stores
> the PID to provid
> Subject: RE: [EXT] [PATCH 1/2] app/test-crypto-perf: fix number of scheduler
> sessions
>
> > The scheduler PMD needs 1 session header,
> > along with a session per worker.
> >
> > After the session rework, this change was made to other apps,
> > for example l2fwd-crypto, but was missed in test-
> Subject: [PATCH v2] app/test: add support to run multi-seg inline IPsec test
>
> Add support to run multi-seg inline ipsec test using
> new test command `inline_ipsec_sg_autotest`
>
> Signed-off-by: Nithin Dabilpuram
> ---
Acked-by: Akhil Goyal
Applied to dpdk-next-crypto
Thanks.
> Acked-by: Gowrishankar Muthukrishnan
>
> > When parsing request files check for file type. This fix will remove
> > dependence on command line parameter for using libjansson
> >
> > Fixes: 0f42f3d6034c ("examples/fips_validation: share callback with multiple
> > keys")
> > Cc: gmuthukri...@marv
19/10/2022 20:42, Stephen Hemminger:
> Tech Board and Governing Board approved license a exception
> for new Google driver. More general approval for MIT
> usage will be handled as needed.
>
> Signed-off-by: Stephen Hemminger
> ---
> +4. MIT 10/19/202210/18/2022
On 10/27/2022 10:46 AM, David Marchand wrote:
This library has no maintainer and, for now, nobody expressed interest
in taking over.
Mark this experimental library as deprecated and announce plan for
removal in v23.11.
Signed-off-by: David Marchand
Acked-by: Ferruh Yigit
Add check for null hash algorithm digest size. Digest size should be 4B
or request will be rejected.
Fixes: 1703e94ac5ce ("qat: add driver for QuickAssist devices")
Cc: declan.dohe...@intel.com
Cc: sta...@dpdk.org
Signed-off-by: Brian Dooley
---
drivers/crypto/qat/qat_sym_session.c | 19
> Introduce a new crypto PMD for hardware accelerators based on UADK [1].
>
> UADK is a framework for user applications to access hardware accelerators.
> UADK relies on IOMMU SVA (Shared Virtual Address) feature, which share
> the same page table between IOMMU and MMU.
> Thereby user application
Hi lihuisong,
This means that queue offloads need to update by recalling dev_configure
and setup target queues.
Can you tell me, where is the limitation?
Thanks,
Mingjin
> -Original Message-
> From: lihuisong (C)
> Sent: 2022年10月26日 17:53
> To: Ye, MingjinX ; dev@dpdk.org
> Cc: sta...@d
This add the support of hardware reassembly per SA basis.
In SA rule, new parameter reassembly_en is added to enable
HW reassembly per SA.
For example:
sa in aead_algo aead_key mode ipv4-tunnel src
dst type inline-protocol-offload port_id reassembly_en
Stats counter frag_dropped will represe
Hi All,
Could you please review and provide suggestions if any.
Thanks,
Mingjin
> -Original Message-
> From: Ye, MingjinX
> Sent: 2022年10月25日 1:27
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Zhou, YidingX ; Ye, MingjinX
> ; Singh, Aman Deep
> ; Zhang, Yuying
> Subject: [PATCH v4 1/2] app
On Tue, Oct 18, 2022 at 3:58 AM Stephen Hemminger
wrote:
>
> Remove some of the older experimental tags for 22.11
>
> Stephen Hemminger (3):
> eal: remove experimental from rte_epoll_wait_interruptible
> eal: make rte_log_list_types not experimental
> rwlock: make trylock operations no long
On Thu, Oct 20, 2022 at 5:34 PM Aaron Conole wrote:
> > Save some cpu time and disk by testing linking against static and shared
> > library in single environments.
> >
> > The .ci/linux-build.sh is modified so it reconfigures an existing build
> > directory: an empty DEF_LIB= means that static an
On Thu, Oct 27, 2022 at 11:22:07AM +0200, Morten Brørup wrote:
> > From: Olivier Matz [mailto:olivier.m...@6wind.com]
> > Sent: Thursday, 27 October 2022 10.35
> >
> > Hi Morten,
> >
> > On Wed, Oct 26, 2022 at 04:44:36PM +0200, Morten Brørup wrote:
> > > Add __rte_cache_aligned to the objs array
On Wed, Oct 12, 2022 at 2:05 PM Markus Theil wrote:
>
> From: Markus Theil
>
> Add missing include in order to make C++ compilers
> happy.
>
We have build checks for headers (-Dcheck_includes meson option), like for C++:
$ cat $BUILDDIR/buildtools/chkincs/chkincs-cpp.p/rte_telemetry.cpp
#includ
> From: Olivier Matz [mailto:olivier.m...@6wind.com]
> Sent: Thursday, 27 October 2022 13.43
>
> On Thu, Oct 27, 2022 at 11:22:07AM +0200, Morten Brørup wrote:
> > > From: Olivier Matz [mailto:olivier.m...@6wind.com]
> > > Sent: Thursday, 27 October 2022 10.35
> > >
> > > Hi Morten,
> > >
> > > On
>
> This library has no maintainer and, for now, nobody expressed interest
> in taking over.
> Mark this experimental library as deprecated and announce plan for
> removal in v23.11.
>
> Signed-off-by: David Marchand
> ---
Acked-by: Konstantin Ananyev
> --
> 2.37.3
27/10/2022 09:34, Thomas Monjalon:
> 17/10/2022 10:32, Andrew Rybchenko:
> > On 10/6/22 21:35, Hanumanth Pothula wrote:
> > > Presently, Rx metadata is sent to PMD by default, leading
> > > to a performance drop as processing for the same in rx path
> > > takes extra cycles.
> > >
> > > Hence, int
> -Original Message-
> From: Brian Dooley
> Sent: Thursday 27 October 2022 11:50
> To: Ji, Kai
> Cc: dev@dpdk.org; sta...@dpdk.org; gak...@marvell.com; Kusztal,
> ArkadiuszX ; Dooley, Brian
> ; Doherty, Declan
> Subject: [PATCH v1] crypto/qat: fix null hash algorithm digest size
>
>
On Fri, Oct 14, 2022 at 9:51 AM Markus Theil wrote:
>
> Makes apps configurable from meson, like already
> possible for drivers.
>
> Signed-off-by: Markus Theil
With these new options, it's hard to tell which and why some
application is built (or not).
Can we have an output similar to libraries
On 10/27/2022 2:06 PM, Huang, ZhiminX wrote:
-Original Message-
From: Mingjin Ye
Sent: Thursday, October 27, 2022 1:10 AM
To: dev@dpdk.org
Cc: sta...@dpdk.org; Zhou, YidingX ; Ye, MingjinX
; Singh, Aman Deep ;
Zhang, Yuying
Subject: [PATCH v4 1/2] app/testpmd: fix vlan offload of rxq
We were always waiting the full timeout because we always waited for
the MAGIC_PROMPT, which is only usable for cleaning the session of
previous data. That also meant that the actual prompt was appended to
the output, resulting in output being '[PEXPECT]#' when the executed
command produced empty o
> > Add check for null hash algorithm digest size. Digest size should be 4B or
> > request will be rejected.
> >
> > Fixes: 1703e94ac5ce ("qat: add driver for QuickAssist devices")
> > Cc: declan.dohe...@intel.com
> > Cc: sta...@dpdk.org
> > Signed-off-by: Brian Dooley
> > ---
>
> Acked-by: Ciara
> Subject: [PATCH v2] examples/ipsec-secgw: support per SA HW reassembly
>
> This add the support of hardware reassembly per SA basis.
> In SA rule, new parameter reassembly_en is added to enable
> HW reassembly per SA.
> For example:
> sa in aead_algo aead_key mode ipv4-tunnel src
> dst type
27/10/2022 11:46, David Marchand:
> This library has no maintainer and, for now, nobody expressed interest
> in taking over.
> Mark this experimental library as deprecated and announce plan for
> removal in v23.11.
>
> Signed-off-by: David Marchand
It adds enough warnings to reach potential user
On Thu, Oct 27, 2022 at 3:13 PM David Marchand
wrote:
>
> On Fri, Oct 14, 2022 at 9:51 AM Markus Theil
> wrote:
> >
> > Makes apps configurable from meson, like already
> > possible for drivers.
> >
> > Signed-off-by: Markus Theil
>
> With these new options, it's hard to tell which and why some
> Adding code snippet using literalinclude so that to keep
> automatically these structures in doc in sync with the
> bbdev source code.
>
> Signed-off-by: Nicolas Chautru
Applied to dpdk-next-crypto
Thanks.
On Thu, Oct 27, 2022 at 02:11:29PM +0200, Morten Brørup wrote:
> > From: Olivier Matz [mailto:olivier.m...@6wind.com]
> > Sent: Thursday, 27 October 2022 13.43
> >
> > On Thu, Oct 27, 2022 at 11:22:07AM +0200, Morten Brørup wrote:
> > > > From: Olivier Matz [mailto:olivier.m...@6wind.com]
> > > >
Hi Jie,
Thanks for the patch.
On 10/25/2022 7:35 AM, Jie Wang wrote:
For NIC I40E_10G-10G_BASE_T_X722, when testpmd is configured with
link speed, it cannot receive jumbo frame packets.
Why only the jumbo frame are effected and not other pkts.
Because it has changed the link status of the
The following changes since commit 9082336048e8c5779c1b4f8930041a42e9197348:
net/nfp: support new solution for tunnel decap action (2022-10-25 10:53:33
+0200)
are available in the Git repository at:
http://dpdk.org/git/next/dpdk-next-crypto
for you to fetch changes up to ee18061c78af29975
>
> As the queue pair used in secondary process needs to be set up by
> the primary process, this patch adds an IPC register function to help
> secondary process to send out queue-pair setup request to primary
> process via IPC request messages. A new "qp_in_used_pid" param stores
> the PID to
When built with debug enabled (RTE_LIBRTE_MEMPOOL_DEBUG defined), the
performance of mempools with caches is improved as follows.
Accessing objects in the mempool is likely to increment either the
put_bulk and put_objs or the get_success_bulk and get_success_objs
debug statistics counters.
By add
When built with debug enabled (RTE_LIBRTE_MEMPOOL_DEBUG defined), the
performance of mempools with caches is improved as follows.
Accessing objects in the mempool is likely to increment either the
put_bulk and put_objs or the get_success_bulk and get_success_objs
debug statistics counters.
By add
Release status meeting minutes 2022-10-27
=
Agenda:
* Release Dates
* Subtrees
* Roadmaps
* LTS
* Defects
* Opens
Participants:
* ARM
* Canonical [No]
* Debian/Microsoft
* Intel
* Marvell
* Nvidia
* Red Hat
* Xilinx/AMD
Release Dates
-
The fo
https://bugs.dpdk.org/show_bug.cgi?id=1117
Bug ID: 1117
Summary: af-packet driver fix breaks rte_pktmbuf_prepend()
Product: DPDK
Version: 21.11
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
Hello,
IBM - Power Systems
DPDK 22.11.0-rc1
* Basic PF on Mellanox: No new issues or regressions were seen.
* Performance: not tested.
* OS: RHEL 8.5 kernel: 4.18.0-348.el8.ppc64le
with gcc version 8.5.0 20210514 (Red Hat 8.5.0-10)
RHEL 9.0 kernel: 5.14.0-70.13.1.el9_0.ppc64le
On 2022/10/27 下午6:53, Akhil Goyal wrote:
Introduce a new crypto PMD for hardware accelerators based on UADK [1].
UADK is a framework for user applications to access hardware accelerators.
UADK relies on IOMMU SVA (Shared Virtual Address) feature, which share
the same page table between IOMMU
> -Original Message-
> From: Andrew Rybchenko
> Sent: Tuesday, October 25, 2022 6:04 PM
> To: Guo, Junfeng ; Zhang, Qi Z
> ; Wu, Jingjing ; Xing, Beilei
>
> Cc: dev@dpdk.org; Li, Xiaoyun
> Subject: Re: [PATCH v11 15/18] net/idpf: add support for Rx offloading
>
> On 10/24/22 16:12, Ju
Hi All,
Could you please review and provide suggestions if any.
Thanks,
Yuan
> -Original Message-
> From: Wang, YuanX
> Sent: Tuesday, October 18, 2022 10:50 PM
> To: andrew.rybche...@oktetlabs.ru; Singh, Aman Deep
> ; Zhang, Yuying
> Cc: dev@dpdk.org; Ding, Xuan ; Tang, Yaqi
> ; Wang,
在 2022/10/27 19:02, Ye, MingjinX 写道:
Hi lihuisong,
This means that queue offloads need to update by recalling dev_configure
and setup target queues.
Why not update queue offloads in PMD?
Can you tell me, where is the limitation?
According to other Rx/Tx offload configurations, this may not b
Hi Singh,
> -Original Message-
> From: Singh, Aman Deep
> Sent: Thursday, October 27, 2022 11:38 PM
> To: Wang, Jie1X ; dev@dpdk.org
> Cc: Yang, SteveX ; Zhang, Qi Z ;
> Yang, Qiming ; Zhang, Yuying
> ; sta...@dpdk.org
> Subject: Re: [PATCH] app/testpmd: fix testpmd receive jumbo frame pa
Normally, the Rx/Tx offload capability of bonding interface is
the intersection of the capability of all slave devices. And
Rx/Tx offloads configuration of slave device comes from bonding
interface. But now there is a risk that slave device retains its
previous offload configurations which is not w
The status_0 field of the rx descriptor is incorrectly parsed as the error
field, resulting in a parsing error of offload features.
This patch fixes parsing of wrong fields.
Fixes: daa02b5cddbb ("mbuf: add namespace to offload flags")
Fixes: dbf3c0e77a22 ("net/ice: handle Rx flex descriptor")
Cc:
The status_0 field of the rx descriptor is incorrectly parsed as the error
field, which causes the parsing error of offload features.
This patch is to fixe the parsing of wrong fields.
Fixes: daa02b5cddbb ("mbuf: add namespace to offload flags")
Fixes: dbf3c0e77a22 ("net/ice: handle Rx flex descr
Some capabilities (like, rx_offload_capa and tx_offload_capa) of
bonding device in dev_info is zero when no slave is added. And its
capability will be updated when add a new slave device.
The capability to update dynamically can introduce some probelms if
not handled properly. For example, the rec
https://bugs.dpdk.org/show_bug.cgi?id=1101
jiang,yu (yux.ji...@intel.com) changed:
What|Removed |Added
CC||yux.ji...@intel.com
Some capabilities (like, rx_offload_capa and tx_offload_capa) of bonding
device in dev_info is zero when no slave is added. And its capability will
be updated when add a new slave device.
The capability to update dynamically may introduce some problems if not
handled properly. For example, the rec
Add __rte_cache_aligned to the objs array.
It makes no difference in the general case, but if get/put operations are
always 32 objects, it will reduce the number of memory (or last level
cache) accesses from five to four 64 B cache lines for every get/put
operation.
For readability reasons, an ex
When built with debug enabled (RTE_LIBRTE_MEMPOOL_DEBUG defined), the
performance of mempools with caches is improved as follows.
Accessing objects in the mempool is likely to increment either the
put_bulk and put_objs or the get_success_bulk and get_success_objs
debug statistics counters.
By add
Add __rte_cache_aligned to the objs array.
It makes no difference in the general case, but if get/put operations are
always 32 objects, it will reduce the number of memory (or last level
cache) accesses from five to four 64 B cache lines for every get/put
operation.
For readability reasons, an ex
When built with debug enabled (RTE_LIBRTE_MEMPOOL_DEBUG defined), the
performance of mempools with caches is improved as follows.
Accessing objects in the mempool is likely to increment either the
put_bulk and put_objs or the get_success_bulk and get_success_objs
debug statistics counters.
By add
On 10/27/22 16:22, David Marchand wrote:
On Thu, Oct 27, 2022 at 3:13 PM David Marchand
wrote:
On Fri, Oct 14, 2022 at 9:51 AM Markus Theil wrote:
Makes apps configurable from meson, like already
possible for drivers.
Signed-off-by: Markus Theil
With these new options, it's hard to tell wh
On 10/27/22 14:04, David Marchand wrote:
On Wed, Oct 12, 2022 at 2:05 PM Markus Theil wrote:
From: Markus Theil
Add missing include in order to make C++ compilers
happy.
We have build checks for headers (-Dcheck_includes meson option), like for C++:
$ cat $BUILDDIR/buildtools/chkincs/chkin
88 matches
Mail list logo
