> -Original Message-
> From: dev On Behalf Of Chenbo Xia
> Sent: Tuesday, July 27, 2021 1:15 PM
> To: dev@dpdk.org
> Cc: maxime.coque...@redhat.com; sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH] net/virtio: fix uninitialized duplex mode
>
> When virtio front-end initializes, the duplex mo
> -Original Message-
> From: Gaoxiang Liu
> Sent: Monday, July 26, 2021 10:42 PM
> To: Xia, Chenbo ; dev@dpdk.org; liugaoxi...@huawei.com
> Cc: Gaoxiang Liu
> Subject: [PATCH] net/virtio: fix memory leak of interrupt handle
>
> Free memory of interrupt handle in virtio_user_dev_uninit()
> -Original Message-
> From: Maxime Coquelin
> Sent: Monday, July 26, 2021 3:58 PM
> To: dev@dpdk.org; Xia, Chenbo ; Wang, Yinan
> ; Ling, WeiX
> Cc: Maxime Coquelin ; sta...@dpdk.org
> Subject: [PATCH] vhost: fix crash on reconnect
>
> When the vhost-user frontend like Virtio-user tries
> -Original Message-
> From: Andrew Rybchenko
> Sent: Wednesday, July 21, 2021 5:22 PM
> To: Maxime Coquelin ; Xia, Chenbo
> ; Stephen Hemminger ;
> Stokes,
> Ian ; Andrew Rybchenko ;
> Yigit, Ferruh
> Cc: dev@dpdk.org; Ivan Ilchenko ; sta...@dpdk.org
> Subject: [PATCH] net/virtio: repor
> -Original Message-
> From: Naga Harish K, S V
> Sent: Saturday, July 24, 2021 7:41 PM
> To: Jayatheerthan, Jay ; jerinjac...@gmail.com
> Cc: dev@dpdk.org
> Subject: [PATCH v2] eventdev: fix event port setup in tx adapter
>
> The event port config set by application in
> rte_event_eth_tx
Comments inlined
-Original Message-
From: Jerin Jacob
Sent: 28 July 2021 11:38
To: Kundapura, Ganapati
Cc: dpdk-dev ; Jayatheerthan, Jay
Subject: Re: RFC: Enahancements to Rx adapter for DPDK 21.11
On Mon, Jul 26, 2021 at 6:37 PM Kundapura, Ganapati
wrote:
>
> A gentle ping for comm
On Mon, Jul 26, 2021 at 6:37 PM Kundapura, Ganapati
wrote:
>
> A gentle ping for comments.
>
> -Original Message-
> From: dev On Behalf Of Kundapura, Ganapati
> Sent: 23 July 2021 12:33
> To: dpdk-dev ; Jerin Jacob ;
> Jayatheerthan, Jay
> Subject: [dpdk-dev] RFC: Enahancements to Rx ad
From: Dapeng Yu
In function softnic_table_action_profile_free(), the memory referenced
by pointer "ap" in the instance of "struct softnic_table_action_profile"
is not freed.
This patch fixes it.
Fixes: a737dd4e5863 ("net/softnic: add table action profile")
Cc: sta...@dpdk.org
Signed-off-by: Da
> -Original Message-
> From: Gregory Etelson
> Sent: Tuesday, July 27, 2021 21:08
> To: dev@dpdk.org
> Cc: getel...@nvidia.com; Ajit Khaparde ;
> Olivier Matz ; Andrew Rybchenko
> ; Yigit, Ferruh ;
> Thomas Monjalon ; sta...@dpdk.org; Li, Xiaoyun
>
> Subject: [PATCH v2] app/testpmd: fix T
For ACL, there are three entry priorities: LOW, NORMAL, HIGH.
Low priority starts from the highest index, 25% of total entries;
Normal priority starts from the highest index, 50% of total entries;
High priority starts from the lowest index, 25% of total entries.
Each TCAM block has 512 entries of
Hi
> -Original Message-
> From: Gregory Etelson
> Sent: Tuesday, July 27, 2021 21:08
> To: dev@dpdk.org
> Cc: getel...@nvidia.com; Ajit Khaparde ;
> Olivier Matz ; Andrew Rybchenko
> ; Yigit, Ferruh ;
> Thomas Monjalon ; sta...@dpdk.org; Li, Xiaoyun
>
> Subject: [PATCH v2] app/testpmd: f
> Announce changes to make rte_security_set_pkt_metadata() and
> rte_security_get_userdata() inline instead of C functions and
> also addition of another field in structure rte_security_ctx for
> holding flags.
>
> Signed-off-by: Nithin Dabilpuram
> Acked-by: Akhil Goyal
> ---
>
> - Series a
Hi Konstantin,
> > > > There are two options that we considered,
> > > > 1. Extend the enum, rte_crypto_op_status, to cover warnings [1]
> > > > 2. There are reserved fields in rte_cryto_op structure. So we can use
> bits in
> > > them to indicate various cases. [2]
> > > >
> > > > Both the submit
> Allow user to provision a per security session maximum segment size
> (MSS) for use when Transmit Segmentation Offload (TSO) is supported.
> The MSS value will be used when PKT_TX_TCP_SEG or PKT_TX_UDP_SEG
> ol_flags are specified in mbuf.
>
> Signed-off-by: Declan Doherty
> Signed-off-by: Radu
>
> > Currently, there is vendor which can support bigger crypto data size.
> > Increase the default mbuf data payload size to 4K as needed.
> >
> > Signed-off-by: Suanming Mou
> > ---
> > app/test/test_cryptodev.h | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
>
> Adding more pe
> > This patch fixes the raw data path dequeue burst fail problem.
> > Previously the in case the queue is full and not all packets
> > asked to be dequeued are processed, the dequeue burst will
> > never happen.
> >
> > Fixes: c21574edc52a ("cryptodev: add dequeue count parameter in raw
> API")
>
> This patch fixes the raw data path dequeue burst fail problem.
> Previously the in case the queue is full and not all packets
> asked to be dequeued are processed, the dequeue burst will
> never happen.
>
> Fixes: c21574edc52a ("cryptodev: add dequeue count parameter in raw API")
> Cc: roy.fan.z
> The PMD destroy function was calling the release function, which frees
> cryptodev->data, and then tries to free cryptodev->data->dev_private,
> which causes the heap use after free issue.
>
> A temporary pointer is set before the free of cryptodev->data,
> which can then be used afterwards to f
Announce changes to make rte_security_set_pkt_metadata() and
rte_security_get_userdata() inline instead of C functions and
also addition of another field in structure rte_security_ctx for
holding flags.
Signed-off-by: Nithin Dabilpuram
Acked-by: Akhil Goyal
---
- Series at
https://inbox.dp
> Coverity is reporting FORWARD_NULL issue when qt data is NULL.
> Adding NULL check for this.
>
> Coverity issue: 371893
> Fixes: 5a3513caeb455 ("crypto/cnxk: add asymmetric session ops")
>
> Signed-off-by: Kiran Kumar K
Acked-by: Akhil Goyal
Applied to dpdk-next-crypto
Announce changes to make rte_security_set_pkt_metadata() and
rte_security_get_userdata() inline instead of C functions and
also addition of another field in structure rte_security_ctx for
holding flags.
Signed-off-by: Nithin Dabilpuram
Acked-by: Akhil Goyal
---
- Series at
https://inbox.dp
Added the files to illustrate the variable size header usage.
Signed-off-by: Cristian Dumitrescu
---
examples/pipeline/examples/varbit.cli | 35 ++
examples/pipeline/examples/varbit.spec | 95 ++
2 files changed, 130 insertions(+)
create mode 100644 examples/pip
Added look-ahead instruction to read a header from the input packet
without advancing the extraction pointer. This is typically used in
correlation with the special extract instruction to extract variable
size headers from the input packet: the first few header fields are
read without advancing the
Added a mechanism to extract variable size headers through a special
flavor of the extract instruction. The length of the last struct field
which has variable size is passed as argument to the instruction.
Signed-off-by: Cristian Dumitrescu
---
lib/pipeline/rte_swx_pipeline.c | 75 ++
The emit instruction that is responsible for pushing headers into the
output packet is now reading the header length from internal run-time
structures as opposed to constant value from the instruction opcode.
Signed-off-by: Cristian Dumitrescu
---
lib/pipeline/rte_swx_pipeline.c | 14 ++-
Added support for variable size headers. The last field of a struct
type can now have a variable size between 0 and N bytes. Useful to
accommodate IPv4 packets with options, etc.
Signed-off-by: Cristian Dumitrescu
---
lib/pipeline/rte_swx_pipeline.c | 107 ++---
lib/pipe
> From: Amaranath Somalapuram
>
> drop all the code duplicating the PCI bus driver
> developed for enable IOMMU in vdev.
>
> Signed-off-by: Amaranath Somalapuram
>
> ---
> V6:
> - Address the following comments from Akhil and David:
> - Spellcheck, Title, Changelog
>
Applied to dpdk-next-crypt
> > Fix possible access of an array by negative index in function
> > qat_sym_qp_setup.
> >
> > Fixes: 8f393c4ffdc1 ("common/qat: support GEN4 devices") Coverity issue:
> > 372131 Coverity issue: 372134
> >
> > Signed-off-by: Arek Kusztal
>
> Acked-by: Adam Dybkowski
Applied to dpdk-next-crypto
>
> New ucode expects const values A and B for asymmetric ECDSA
> messages. Adding roc support for this.
>
> Signed-off-by: Kiran Kumar K
> ---
Series Acked-by: Akhil Goyal
Applied to dpdk-next-crypto
TX checksum of a tunnelled packet can be calculated for outer headers
only or for both outer and inner parts. The calculation method is
determined by application.
If TX checksum calculation can be offloaded, hardware ignores
existing checksum value and replaces it with an updated result.
If TX chec
Added the files to illustrate the variable size header usage.
Signed-off-by: Cristian Dumitrescu
---
examples/pipeline/examples/varbit.cli | 35 ++
examples/pipeline/examples/varbit.spec | 95 ++
2 files changed, 130 insertions(+)
create mode 100644 examples/pip
Added look-ahead instruction to read a header from the input packet
without advancing the extraction pointer. This is typically used in
correlation with the special extract instruction to extract variable
size headers from the input packet: the first few header fields are
read without advancing the
Added a mechanism to extract variable size headers through a special
flavor of the extract instruction. The length of the last struct field
which has variable size is passed as argument to the instruction.
Signed-off-by: Cristian Dumitrescu
---
lib/pipeline/rte_swx_pipeline.c | 75 ++
Added support for variable size headers. The last field of a struct
type can now have a variable size between 0 and N bytes. Useful to
accommodate IPv4 packets with options, etc.
Signed-off-by: Cristian Dumitrescu
---
lib/pipeline/rte_swx_pipeline.c | 107 +---
lib/pipel
The emit instruction that is responsible for pushing headers into the
output packet is now reading the header length from internal run-time
structures as opposed to constant value from the instruction opcode.
Signed-off-by: Cristian Dumitrescu
---
lib/pipeline/rte_swx_pipeline.c | 14 ++-
This patch fixes the raw data path dequeue burst fail problem.
Previously the in case the queue is full and not all packets
asked to be dequeued are processed, the dequeue burst will
never happen.
Fixes: c21574edc52a ("cryptodev: add dequeue count parameter in raw API")
Cc: roy.fan.zh...@intel.com
PCI, vmbus, and auxiliary drivers printed a warning
when NUMA node had beed reported as (-1) or not reported by OS:
EAL: Invalid NUMA socket, default to 0
This message and its level might confuse users, because configuration
is valid and nothing happens that requires attention or interventi
Long,
I suspect that this has not tripped me up because I'm only using queue
per TX/RX path. But yes, for multiple queues it looks to be a problem.
Jonathan
On 7/26/21 6:16 PM, Long Li wrote:
Subject: [PATCH v2] bus/vmbus: Fix crash when handling packets in
secondary process
Have secondary
On 7/27/21 7:14 AM, Chenbo Xia wrote:
> When virtio front-end initializes, the duplex mode should be set
> unknown before reading any duplex mode information from configuration
> space. This patch fixes the issue that duplex mode is by default set
> to zero, which equals ETH_LINK_HALF_DUPLEX. Th
In order to avoid code duplication common code shared across drivers
is stored under drivers/common.
Currently if one needs to introduce changes to common and particular
driver i.e common/driver and net/driver separate patches needs to be
prepared otherwise warning about wrong headline prefix gets
> -Original Message-
> From: Thomas Monjalon
> Sent: Saturday, July 24, 2021 10:52 AM
> To: Juraj Linkeš ; Bruce Richardson
> ; Pavan Nikhilesh Bhagavatula
> ; Honnappa Nagarahalli
>
> Cc: dev@dpdk.org; jer...@marvell.com; Jan Viktorin
> ; Ruifeng Wang ; nd
>
> Subject: Re: [dpdk-dev]
On Thu, Jul 22, 2021 at 09:56:25PM +0200, Thomas Monjalon wrote:
> +Cc more people for reviews.
>
> 07/07/2021 11:40, Mohsin Kazmi:
> > Preparation the headers for the hardware offload
"of" is missing after Preparation?
> > misses the outer ipv4 checksum offload.
> > It results in bad checksum c
> -Original Message-
> From: Andrew Rybchenko
> Sent: 23 июля 2021 г. 23:07
> To: Dmitry Kozlyuk ; dev@dpdk.org
> Cc: sta...@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2] bus: clarify log for non-NUMA-aware devices
>
> External email: Use caution opening links or attachments
>
>
> On 6/
> -Original Message-
> From: Zhang, AlvinX
> Sent: Monday, July 26, 2021 5:17 PM
> To: Xing, Beilei ; Zhang, Qi Z
> Cc: dev@dpdk.org; Zhang, AlvinX ;
> sta...@dpdk.org
> Subject: [PATCH v2] net/iavf: fix virtual channel RSS command error handling
>
> Kernel PF may not respond to virtu
Hi Akhil,
> Hi Konstantin,
> > > There are two options that we considered,
> > > 1. Extend the enum, rte_crypto_op_status, to cover warnings [1]
> > > 2. There are reserved fields in rte_cryto_op structure. So we can use
> > > bits in
> > them to indicate various cases. [2]
> > >
> > > Both t
Hi,
> -Original Message-
> From: dev On Behalf Of Gregory Etelson
> Sent: Tuesday, July 20, 2021 6:38 PM
> To: dev@dpdk.org
> Cc: Gregory Etelson ; sta...@dpdk.org; Slava
> Ovsiienko ; Matan Azrad ;
> Shahaf Shuler
> Subject: [dpdk-dev] [PATCH] net/mlx5: fix representor interrupts handle
Drop queue creation and destruction were not implemented for DevX
flow engine and Verbs engine methods were used as a workaround.
Implement these methods for DevX so that there is a valid queue ID
that can be used regardless of queue configuration via API.
Cc: sta...@dpdk.org
Signed-off-by: Dmitr
It was unspecified what happens to indirect actions when a port
is stopped, possibly reconfigured, and started again. MLX5 PMD,
the first one to use indirect actions, intended to keep them across
such a sequence, but the implementation was buggy. Patches 1-3 fix
the PMD behavior, patch 4 adds commo
rte_flow_action_handle_create() did not specify what happens
with an indirect action when device is stopped, possibly reconfigured,
and started again.
It is proposed that indirect actions persisted across such a sequence.
This allows for easier API usage and better HW resources utilization
by savi
Maximum available flow priority was discovered using Verbs API
regardless of the selected flow engine. This required some Verbs
objects to be initialized in order to use DevX engine. Make priority
discovery an engine method and implement it for DevX using its API.
Cc: sta...@dpdk.org
Signed-off-b
MLX5 PMD uses reference counting to manage RX queue resources.
After port stop shared RSS actions kept references to RX queues,
preventing resource release. As a result, internal PMD mempool
for such queues had been exhausted after a number of port restarts.
Diagnostic message from rte_eth_dev_star
Ping.
On Sun, Jul 18, 2021 at 3:15 PM wrote:
>
> From: Jerin Jacob
>
> Added a diagram to document meter library components
> and added text for steps must performed by the application to
> configure the traffic meter and policing library.
>
> Signed-off-by: Jerin Jacob
> ---
> doc/guides/prog
>
> All ABIs in PCI bus driver, which are defined in rte_buc_pci.h,
> will be removed and the header will be made internal.
>
> Signed-off-by: Chenbo Xia
> ---
> doc/guides/rel_notes/deprecation.rst | 5 +
> 1 file changed, 5 insertions(+)
>
> diff --git a/doc/guides/rel_notes/deprecation.
Acked-by: John McNamara
The memory size of bandwidth config parameters is not set correctly in
memory copy process, which leads to the wrong values. This patch fixed
the size to the correct value.
Fixes: 3a6bfc37eaf4 ("net/ice: support QoS config VF bandwidth in DCF")
Signed-off-by: Ting Xu
---
drivers/net/ice/ice_dcf
Set proper selection flags for the RX hash and queue num in case of
mismatch between the L3 proto of the RSS hash types and the flow rule
pattern.
v2:
* Same code changes as v1, but separated into two commits.
Lior Margalit (2):
net/mlx5: fix RSS L4 proto selection flags settings
net/mlx5:
The RSS hash types defined in the API do not support setting the L4 proto
type (TCP or UDP) without setting the L3 proto. For example, ETH_RSS_TCP
is defined as
(ETH_RSS_NONFRAG_IPV4_TCP | \
ETH_RSS_NONFRAG_IPV6_TCP | \
ETH_RSS_IPV6_TCP_EX).
The L3 proto of the RSS hash type may be different
The selection flags for the RX hash define how the received packets will
be distributed between multiple queues.
When creating a new TIR, the queue_num is set to 1 if non of the selection
flags is set.
Applied the same to the RSS desc before checking if it matches a cached
TIR object to save cr
> -Original Message-
> From: dev On Behalf Of Arek Kusztal
> Sent: Monday, 26 July, 2021 18:48
> To: dev@dpdk.org
> Cc: gak...@marvell.com; Trahe, Fiona ; Zhang, Roy
> Fan ; Kusztal, ArkadiuszX
>
> Subject: [dpdk-dev] [PATCH] crypto/qat: fix illegal access of an array
>
> Fix possible ac
From: Rohit Raj
l3fwd uses mbufs with 2KB data size. If we enable jumbo packets, it is
not able to store packets with size greater than 2KB, hence these
packets are dropped.
This patch fixes this issue by enabling scatter for jumbo packet, if
it is supported by NIC.
If scatter is not supported
> -Original Message-
> From: Yu, DapengX
> Sent: Tuesday, July 27, 2021 9:15 AM
> To: Singh, Jasvinder ; Dumitrescu, Cristian
>
> Cc: dev@dpdk.org; Yu, DapengX ; sta...@dpdk.org
> Subject: [PATCH] net/softnic: fix null pointer dereference
>
> From: Dapeng Yu
>
> When there is no "fi
On Mon, Jul 26, 2021 at 05:56:17AM +, Xia, Chenbo wrote:
> Hi, Ferruh
>
> > -Original Message-
> > From: Yigit, Ferruh
> > Sent: Friday, July 23, 2021 8:47 PM
> > To: Xia, Chenbo ; dev@dpdk.org; tho...@monjalon.net
> > Cc: m...@ashroe.eu; nhor...@tuxdriver.com; david.march...@redhat.c
From: Kai Ji
This patch enable out-of-place auth-cipher operations where
digest should be encrypted among with the rest of raw data.
It also adds support for partially encrypted digest when using
auth-cipher operations.
Fixes: 7c87e2d7b359 ("crypto/snow3g: use IPsec library")
Cc: pablo.de.lara.g
From: Dapeng Yu
When there is no "firmware" in arguments, the "firmware" pointer is
null, and will be dereferenced by rte_strscpy().
This patch moves the code block which copies character string from
"firmware" to "p->firmware" into the "if" statements where "firmware"
argument exists and it is
Adding Marvell prefix for qlogic drivers.
Removing myself as I'm no longer responsible for qede driver
Signed-off-by: Igor Russkikh
Signed-off-by: Rasesh Mody
---
MAINTAINERS | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 4de75ac94b..bc3
Fixing ex-Aquantia email - it is now part of Marvell.
Removing Pavel Belous email - he is not in company now.
Adding Marvell tree to run this through.
Signed-off-by: Igor Russkikh
---
MAINTAINERS | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
To align with other eth device queue configuration callbacks, change RX
and TX queue release callback API parameter from queue object to device
and queue index.
Signed-off-by: Xueming Li
=
In formal patch, there should be a lot of changes to update all PMDs.
---
lib/ethd
In eth PMD driver model, each RX queue was pre-loaded with mbufs for
saving incoming packets. When number of SF or VF scale out in a switch
domain, the memory consumption became significant. Most important,
polling all ports leads to high cache miss, high latency and low
throughput.
To save memory
68 matches
Mail list logo
