Hi,
We are running VPP v22.02 (having DPDK v21.11 and compiled with MLX5 support)
and an interface ens2 (PCI : :0b:00.0) having mlx5_core driver. We are able
to bring this interface inside VPP as HundredGigabitEthernetb/0/0 but when we
try to bring this interface's state up, we are getting
> -Original Message-
> From: Ruifeng Wang
> Sent: Wednesday, June 29, 2022 7:41 PM
> To: Slava Ovsiienko ; Ali Alnubani
> ; Matan
> Azrad
> Cc: dev@dpdk.org; Honnappa Nagarahalli ;
> sta...@dpdk.org; nd
> ; nd ; nd
> Subject: RE: [PATCH] net/mlx5: fix risk in Rx descriptor read in NEON
Adds the __extension__ attribute to resolve the following warning
with gcc 4.8.5 on CentOS 7:
lib/net/rte_macsec.h:38:2: error: type of bit-field 'short_length' is
a GCC extension [-Werror=pedantic]
Fixes: 0aaf097585a6 ("ethdev: add MACsec flow item")
Cc: gak...@marvell.com
Signed-off-by: Ali A
>> Enabled the flag pmd_iova_as_va in cnxk driver build files as they work
>> with IOVA as VA. Updated cn9k and cn10k soc build configurations to
>> enable the IOVA as VA build by default.
>>
>> Signed-off-by: Shijith Thotton
>> ---
>> config/arm/meson.build | 8 +++-
>> driver
> -邮件原件-
> 发件人: Feifei Wang
> 发送时间: Tuesday, September 27, 2022 10:48 AM
> 抄送: dev@dpdk.org; nd ; Feifei Wang
>
> 主题: [PATCH v2 0/3] Direct re-arming of buffers on receive side
>
> Currently, the transmit side frees the buffers into the lcore cache and the
> receive side allocates buffer
>> Swapped position of mbuf next pointer and second dynamic field (dynfield2)
>> if the build is configured to use IOVA as VA. This is to move the mbuf
>> next pointer to first cache line. kni library is disabled for this
>> change as it depends on the offset value of next pointer.
>>
>> Signed-off
>> > > mbuf physical address field is not used in builds which only uses VA. It
>> > > is used to expand the dynamic field area.
>> > >
>> > > Signed-off-by: Shijith Thotton
>> >
>> > We cannot condition the use of the dynamic field.
>> > I think it is enough justification to reject this patch.
>> IOVA mode in DPDK is either PA or VA. The new build option iova_as_va
>> configures the mode to VA at compile time and prevents setting it to PA
>> at runtime. For now, all drivers which are not always enabled are
>> disabled with this option. Supported driver can set the flag
>> pmd_iova_as_va
This patch is to fix the tunnel TSO not enabling issue, simplify
the logic of calculating 'Tx Buffer Size' of data descriptor with IPSec,
and fix handling that the mbuf size exceeds the TX descriptor
hardware limit(1B-16KB) which causes malicious behavior to the NIC.
Fixes: 1e728b01120c ("net/iavf
If producer thread is only acting as a bridge between NIC and DLB, then
performance can be greatly improved by bypassing the fence instruction.
DLB enqueue API calls memory fence once per enqueue burst. If prodcuer
thread is just reading from NIC and sending to DLB without updating
the read buffer
For best performance, applications running on certain cores should use
the DLB device locally available on the same tile along with other
resources. To allocate optimal resources, probing is done for each
producer port (PP) for a given CPU and the best performing ports are
allocated to producers. T
This commit implements the changes required for using suggested
port type hint feature. Each port uses different credit quanta
based on port type specified using port configuration flags.
Each port has separate quanta defined in dlb2_priv.h
Producer and consumer ports will need larger quanta value
Hi Olivier,
Thanks for the review.
>On Wed, Sep 21, 2022 at 07:26:16PM +0530, Shijith Thotton wrote:
>> This is a continuation of the discussions[1] to add mbuf physical address
>> field to
>dynamic field.
>> Previous version was to add PA field to dynamic field area based on the EAL
>IOVA mode
> > Subject: [PATCH] security: remove get_userdata function pointer
> >
> > Removes get_userdata function pointer as it is being unused
> > and make fast accessing method that uses dynamic field to get
> > userdata as default for optimizing security path.
> >
> > Signed-off-by: Srujana Challa
> Ap
We cannot configure the queues when the device is started. So, that
part is still before the device starts. I just moved
bond_ethdev_8023ad_flow_set() after the device start.
On Thu, Sep 29, 2022 at 9:05 AM Usman Tanveer wrote:
>
> when dedicated queues are enable with bonding mode 4 (mlx5), the
when dedicated queues are enable with bonding mode 4 (mlx5), the
application sets the flow, which cannot be set if the device is not
started. This fixed the issue by starting the device just before
setting the flow. Because device should be started to set the flow.
Also it does not effect other dri
If producer thread is only acting as a bridge between NIC and DLB, then
performance can be greatly improved by bypassing the fence instruction.
DLB enqueue API calls memory fence once per enqueue burst. If prodcuer
thread is just reading from NIC and sending to DLB without updating
the read buffer
This commit implements the changes required for using suggested
port type hint feature. Each port uses different credit quanta
based on port type specified using port configuration flags.
Each port has separate quanta defined in dlb2_priv.h
Producer and consumer ports will need larger quanta value
For best performance, applications running on certain cores should use
the DLB device locally available on the same tile along with other
resources. To allocate optimal resources, probing is done for each
producer port (PP) for a given CPU and the best performing ports are
allocated to producers. T
Hi, Akhil
On 2022/9/26 下午4:36, 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
Cipher algorithms:
* ``RTE_CRYPTO_CIPHER_AES_ECB``
* ``RTE_CRYPTO_CIPHER_AES_CBC``
* ``RTE_CRYPTO_CIPHER_AES_XTS``
* ``RTE_CRYPTO_CIPHER_DES_CBC``
Signed-off-by: Zhangfei Gao
---
doc/guides/cryptodevs/features/uadk.ini | 10 +
doc/guides/cryptodevs/uadk.rst | 6 +
drivers/crypto/uad
Example:
sudo dpdk-test --vdev=crypto_uadk --log-level=6
RTE>>cryptodev_uadk_autotest
RTE>>quit
Signed-off-by: Zhangfei Gao
---
app/test/test_cryptodev.c | 7 +++
app/test/test_cryptodev.h | 1 +
2 files changed, 8 insertions(+)
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptod
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 can directly
Hash algorithms:
* ``RTE_CRYPTO_AUTH_MD5``
* ``RTE_CRYPTO_AUTH_MD5_HMAC``
* ``RTE_CRYPTO_AUTH_SHA1``
* ``RTE_CRYPTO_AUTH_SHA1_HMAC``
* ``RTE_CRYPTO_AUTH_SHA224``
* ``RTE_CRYPTO_AUTH_SHA224_HMAC``
* ``RTE_CRYPTO_AUTH_SHA256``
* ``RTE_CRYPTO_AUTH_SHA256_HMAC``
* ``RTE_CRYPTO_AUTH_SHA384``
* ``RTE_CR
This commit adds the enqueue and dequeue operations.
Signed-off-by: Zhangfei Gao
---
drivers/crypto/uadk/uadk_crypto_pmd.c | 53 ++-
1 file changed, 51 insertions(+), 2 deletions(-)
diff --git a/drivers/crypto/uadk/uadk_crypto_pmd.c
b/drivers/crypto/uadk/uadk_crypto_pmd
Support the basic dev control operations: configure, close, start,
stop and get info, as well as queue pairs operations.
Signed-off-by: Zhangfei Gao
---
drivers/crypto/uadk/uadk_crypto_pmd.c | 213 --
1 file changed, 204 insertions(+), 9 deletions(-)
diff --git a/drivers
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 can directly
Hi Stephen,
Please see inline.
Thanks,
Anoob
> -Original Message-
> From: Stephen Hemminger
> Sent: Thursday, September 29, 2022 3:51 AM
> To: Anoob Joseph
> Cc: Honnappa Nagarahalli ; Jerin Jacob
> Kollanukkaran ; dev@dpdk.org;
> sean.morris...@intel.com
> Subject: [EXT] Re: [PATCH] r
Not sure these are failing now in RHEL platform, seems fail is unrelated to the
commit, I will check and resubmit new version.
-Original Message-
From: Sevincer, Abdullah
Sent: Wednesday, September 28, 2022 6:33 PM
To: dev@dpdk.org
Cc: jer...@marvell.com; Sevincer, Abdullah
Subject: [
> > > Primary process could set up queues state correctly when starting
> > > port, while secondary process not. Under multi-process scenario,
> > "stream_init"
> > > function would get wrong queues state for secondary process.
> > >
> > > This commit is to get queues state from ethdev which is loc
I agree with you. Can you fix it?
Thanks
Yunjian
> -Original Message-
> From: Slava Ovsiienko [mailto:viachesl...@nvidia.com]
> Sent: Tuesday, September 27, 2022 3:36 AM
> To: wangyunjian ; dev@dpdk.org
> Cc: Matan Azrad ; Raslan Darawsheh
> ; Dmitry Kozlyuk ;
> Huangshaozhang ; s
https://bugs.dpdk.org/show_bug.cgi?id=1087
Bug ID: 1087
Summary: [Bug] tcp data len is incorrectly calculated in
gro_tcp4_reassemble function
Product: DPDK
Version: unspecified
Hardware: All
OS: All
From: Xuan Ding
This patch applies rte_vhost_async_dma_unconfigure() API
to manually free DMA vchannels instead of waiting
until the program ends to be released.
Signed-off-by: Xuan Ding
---
examples/vhost/main.c | 45 ++-
examples/vhost/main.h | 1 +
2
From: Xuan Ding
This patch adds a new API rte_vhost_async_dma_unconfigure() to unconfigure
DMA vchannels in vhost async data path.
Lock protection are also added to protect DMA vchannels configuration and
unconfiguration from concurrent calls.
Signed-off-by: Xuan Ding
---
doc/guides/prog_guid
From: Xuan Ding
This patchset introduces a new API rte_vhost_async_dma_unconfigure()
to help user to manually free the DMA vchannel finished to use.
Note: this API should be called after async channel unregister.
v3:
* Rebase to latest DPDK.
* Refine some descriptions in the doc.
* Fix one bug
This commit implements the changes required for using suggested
port type hint feature. Each port uses different credit quanta
based on port type specified using port configuration flags.
Each port has separate quanta defined in dlb2_priv.h
Producer and consumer ports will need larger quanta value
If producer thread is only acting as a bridge between NIC and DLB, then
performance can be greatly improved by bypassing the fence instruction.
DLB enqueue API calls memory fence once per enqueue burst. If prodcuer
thread is just reading from NIC and sending to DLB without updating
the read buffer
For best performance, applications running on certain cores should use
the DLB device locally available on the same tile along with other
resources. To allocate optimal resources, probing is done for each
producer port (PP) for a given CPU and the best performing ports are
allocated to producers. T
On Mon, 29 Aug 2022 22:21:51 +0530
Anoob Joseph wrote:
> Build fails if RTE_LOG_DP_LEVEL is set to RTE_LOG_DEBUG. Fix the
> same by including the required header when RTE_LOG_DP_LEVEL is
> set to RTE_LOG_DEBUG.
>
> ../lib/rcu/rte_rcu_qsbr.h:678:40: error: expected ‘)’ before ‘PRIu64’
> 678 |
>
> > >>
> > >> Build fails if RTE_LOG_DP_LEVEL is set to RTE_LOG_DEBUG. Fix the
> > >> same by including the required header when RTE_LOG_DP_LEVEL is set
> > >> to RTE_LOG_DEBUG.
> > >>
> > >> ../lib/rcu/rte_rcu_qsbr.h:678:40: error: expected ‘)’ before ‘PRIu64’
> > >>678 |"%s: status: l
Acked-by: Kai Ji
> -Original Message-
> From: Gowrishankar Muthukrishnan
> Sent: Tuesday, September 27, 2022 8:31 AM
> To: dev@dpdk.org
> Cc: Anoob Joseph ; Zhang, Roy Fan
> ; Dooley, Brian ; Akhil
> Goyal ; jer...@marvell.com; Gowrishankar
> Muthukrishnan ; Kiran Kumar K
>
> Subject: [
This commit implements the changes required for using suggested
port type hint feature. Each port uses different credit quanta
based on port type specified using port configuration flags.
Each port has separate quanta defined in dlb2_priv.h
Producer and consumer ports will need larger quanta value
If producer thread is only acting as a bridge between NIC and DLB, then
performance can be greatly improved by bypassing the fence instruction.
DLB enqueue API calls memory fence once per enqueue burst. If prodcuer
thread is just reading from NIC and sending to DLB without updating
the read buffer
For best performance, applications running on certain cores should use
the DLB device locally available on the same tile along with other
resources. To allocate optimal resources, probing is done for each
producer port (PP) for a given CPU and the best performing ports are
allocated to producers. T
> Akhil Goyal (3):
> net: add MACsec header
> ethdev: add MACsec flow item
> security: support MACsec
Applied with some formatting changes, thanks.
On Wed, 28 Sep 2022 14:52:47 +0200
Olivier Matz wrote:
> On Wed, Sep 28, 2022 at 09:24:51AM +0200, Thomas Monjalon wrote:
> > 21/09/2022 15:56, Shijith Thotton:
> > > mbuf physical address field is not used in builds which only uses VA. It
> > > is used to expand the dynamic field area.
> > >
Bruce, David, Thomas,
PING. Please ack or review this simple patch, so it can be merged.
Details were already discussed on the list with Thomas.
NB: The test errors in Patchwork are bogus: "ERROR: Could not detect Ninja v1.5
or newer" is clearly not related to the patch.
-Morten
> From: Morte
28/09/2022 14:52, Olivier Matz:
> On Wed, Sep 28, 2022 at 09:24:51AM +0200, Thomas Monjalon wrote:
> > 21/09/2022 15:56, Shijith Thotton:
> > > mbuf physical address field is not used in builds which only uses VA. It
> > > is used to expand the dynamic field area.
> > >
> > > Signed-off-by: Shijit
This commit implements the changes required for using suggested
port type hint feature. Each port uses different credit quanta
based on port type specified using port configuration flags.
Each port has separate quanta defined in dlb2_priv.h
Producer and consumer ports will need larger quanta value
If producer thread is only acting as a bridge between NIC and DLB, then
performance can be greatly improved by bypassing the fence instruction.
DLB enqueue API calls memory fence once per enqueue burst. If prodcuer
thread is just reading from NIC and sending to DLB without updating
the read buffer
For best performance, applications running on certain cores should use
the DLB device locally available on the same tile along with other
resources. To allocate optimal resources, probing is done for each
producer port (PP) for a given CPU and the best performing ports are
allocated to producers. T
For best performance, applications running on certain cores should use
the DLB device locally available on the same tile along with other
resources. To allocate optimal resources, probing is done for each
producer port (PP) for a given CPU and the best performing ports are
allocated to producers. T
If producer thread is only acting as a bridge between NIC and DLB, then
performance can be greatly improved by bypassing the fence instruction.
DLB enqueue API calls memory fence once per enqueue burst. If prodcuer
thread is just reading from NIC and sending to DLB without updating
the read buffer
This patch addresses an issue of enqueuing more than
max_enq_depth and not able to dequeuing events equal
to max_cq_depth in a single call of rte_event_enqueue_burst
and rte_event_dequeue_burst.
Apply fix for restricting enqueue of events to max_enq_depth
so that in a single rte_event_enqueue_burs
Added support for MACsec in rte_security for offloading
MACsec Protocol operation to inline NIC device or a crypto device.
To support MACsec we cannot just make one security session and
send with the packet to process it. MACsec specifications suggest,
it has 3 different entities - SECY Entity, SC
A new flow item is defined for MACsec flows which can be
offloaded to an inline device. If the flow matches with
MACsec header, device will process as per the security
session created using rte_security APIs.
If an error comes while MACsec processing in HW, PMD will
notify with the events defined i
Added MACsec protocol header to be used for supporting
MACsec protocol offload in hardware or directly in the application.
Signed-off-by: Akhil Goyal
Acked-by: Olivier Matz
Acked-by: Ori Kam
---
doc/api/doxy-api-index.md | 3 +-
lib/net/meson.build | 1 +
lib/net/rte_macsec.h | 6
Added support for MACsec in rte_security for offloading
MACsec Protocol operation to inline NIC device or a crypto device.
To support MACsec we cannot just make one security session and
send with the packet to process it. MACsec specifications suggest,
it can have 3 different entities - SECY Entit
Acked-by: Kai Ji
> -Original Message-
> From: Arek Kusztal
> Sent: Thursday, August 18, 2022 8:42 AM
> To: dev@dpdk.org
> Cc: gak...@marvell.com; Kusztal, ArkadiuszX
> Subject: [PATCH 2/2] cryptodev: add ec point verification
>
> This addition complements point verification in ECDH ope
The code changes only update the comment description of rte_crypto_ec_point.
Can you update commit message to indicate the comments update on expected
behaviour when x.data and y.data equal to NULL.
Acked-by: Kai Ji
> -Original Message-
> From: Arek Kusztal
> Sent: Thursday, August 18
Hi Jerin,
Thanks for the feedback. I will do the suggested changes in next version of the
patch.
> -Original Message-
> From: Jerin Jacob
> Sent: Wednesday, September 28, 2022 9:38 AM
> To: Amit Prakash Shukla
> Cc: Akhil Goyal ; Fan Zhang
> ; Ray Kinsella ; dev@dpdk.org;
> Jerin Jacob
Fixed not set output length in asym pmd
when doing RSA.
Fixes: 002486db239e ("crypto/qat: refactor asymmetric session")
Signed-off-by: Arek Kusztal
---
drivers/crypto/qat/qat_asym.c | 24 ++--
1 file changed, 10 insertions(+), 14 deletions(-)
diff --git a/drivers/crypto/qat
Fixed incorrectly placed clean function in asym response.
Fixes: 002486db239e ("crypto/qat: refactor asymmetric session")
Signed-off-by: Arek Kusztal
---
drivers/crypto/qat/qat_asym.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/crypto/qat/qat_asym.c b/drivers
On Tue, Sep 27, 2022 at 11:45 PM Ori Kam wrote:
>
>
>
> > -Original Message-
> > From: Dumitrescu, Cristian
> > Sent: Tuesday, 27 September 2022 14:56
> >
> >
> >
> > > -Original Message-
> > > From: Alexander Kozyrev
> > > Sent: Monday, September 26, 2022 3:58 PM
> > > To: dev@d
On Wed, Sep 28, 2022 at 8:14 PM Gujjar, Abhinandan S
wrote:
>
>
>
> > -Original Message-
> > From: Jerin Jacob
> > Sent: Tuesday, September 27, 2022 6:56 PM
> > To: Gujjar, Abhinandan S
> > Cc: Volodymyr Fialko ; dev@dpdk.org; Jerin Jacob
> > Kollanukkaran ; Pavan Nikhilesh Bhagavatula
>
28/09/2022 17:21, Claudio Fontana:
> On 9/28/22 16:37, Maxime Coquelin wrote:
> > The title should be reworded, maybe something like below?
> > "vhost: fix possible out of bound access in buffer vectors"
>
> possible, I leave it to you and other maintainers here now to figure out.
Maxime is sugge
Hi Andrew,
Did you get a chance to review this patch? Please let me know your thoughts on
it.
Regards,
Yuan
> -Original Message-
> From: Wang, YuanX
> Sent: Monday, September 26, 2022 5:41 PM
> To: dev@dpdk.org; Thomas Monjalon ; Ferruh Yigit
> ; Andrew Rybchenko
>
> Cc: m...@ashroe.e
> -Original Message-
> From: Akhil Goyal
> Sent: Wednesday, September 28, 2022 1:56 PM
> To: Ji, Kai ; dev@dpdk.org
> Cc: tho...@monjalon.net; david.march...@redhat.com;
> hemant.agra...@nxp.com; Vamsi Krishna Attunuru
> ; ferruh.yi...@xilinx.com;
> andrew.rybche...@oktetlabs.ru; konsta
> -Original Message-
> From: Arek Kusztal
> Sent: Wednesday, September 28, 2022 6:48 PM
> To: dev@dpdk.org
> Cc: Akhil Goyal ; kai...@intel.com; Arek Kusztal
>
> Subject: [EXT] [PATCH v2 0/2] cryptodev: add SM3 and SM4 algorithms
>
> External Email
>
> --
Acked-by: Kai Ji
> -Original Message-
> From: Kusztal, ArkadiuszX
> Sent: Wednesday, September 28, 2022 2:18 PM
> To: dev@dpdk.org
> Cc: gak...@marvell.com; Ji, Kai ; Kusztal, ArkadiuszX
>
> Subject: [PATCH v2 1/2] cryptodev: add sm4 encryption algorithm
>
> ShangMi 4 (SM4) is a block
Acked-by: Kai Ji
> -Original Message-
> From: Kusztal, ArkadiuszX
> Sent: Wednesday, September 28, 2022 2:18 PM
> To: dev@dpdk.org
> Cc: gak...@marvell.com; Ji, Kai ; Kusztal, ArkadiuszX
>
> Subject: [PATCH v2 2/2] cryptodev: add sm3 hash algorithm
>
> ShangMi 3 (SM3) is a cryptographi
On 9/28/22 16:37, Maxime Coquelin wrote:
> Hi Claudio,
>
> The title should be reworded, maybe something like below?
> "vhost: fix possible out of bound access in buffer vectors"
possible, I leave it to you and other maintainers here now to figure out.
>
> On 8/2/22 02:49, Claudio Fontana wrote
On Tue, Sep 27, 2022 at 7:12 AM Abdullah Sevincer
wrote:
>
> For best performance, applications running on certain cores should use
> the DLB device locally available on the same tile along with other
> resources. To allocate optimal resources, probing is done for each
> producer port (PP) for a g
> -Original Message-
> From: Jerin Jacob
> Sent: Tuesday, September 27, 2022 6:56 PM
> To: Gujjar, Abhinandan S
> Cc: Volodymyr Fialko ; dev@dpdk.org; Jerin Jacob
> Kollanukkaran ; Pavan Nikhilesh Bhagavatula
> ; Shijith Thotton ;
> Hemant Agrawal ; Sachin Saxena
> ; Jayatheerthan, Jay
Hi Claudio,
The title should be reworded, maybe something like below?
"vhost: fix possible out of bound access in buffer vectors"
On 8/2/22 02:49, Claudio Fontana wrote:
in virtio_dev_split we cannot currently call desc_to_mbuf with
nr_vec == 0, or we end up trying to rte_memcpy from a source a
> -Original Message-
> From: Andrew Rybchenko
> Sent: Monday, 26 September 2022 16:07
>
> On 9/19/22 18:50, Michael Savisko wrote:
> > In some cases application may receive a packet that should have been
> > received by the kernel. In this case application uses KNI or other
> > means to
ShangMi 4 (SM4) is a block cipher used in the Chinese National Standard for
Wireless LAN WAPI and also used with Transport Layer Security.
- Added SM4 encryption algorithm.
Supported modes are ECB, CBC and CTR.
Signed-off-by: Arek Kusztal
---
doc/guides/cryptodevs/features/default.ini | 3 +++
ShangMi 3 (SM3) is a cryptographic hash function used in
the Chinese National Standard.
- Added SM3 algorithm
Signed-off-by: Arek Kusztal
---
doc/guides/cryptodevs/features/default.ini | 1 +
doc/guides/rel_notes/release_22_11.rst | 5 +
lib/cryptodev/rte_crypto_sym.h | 4 ++
ShangMi 4 (SM4) is a block cipher used in the Chinese National Standard for
Wireless LAN WAPI and also used with Transport Layer Security.
ShangMi 3 (SM3) is a cryptographic hash function used in the
Chinese National Standard.
This patcheset adds both to the Cryptodev.
v2:
- Fixed incorrect sm3 s
> -Original Message-
> From: Akhil Goyal
> Sent: Wednesday, 28 September 2022 15:23
> To: dev@dpdk.org
Acked-by: Ori Kam
Thanks,
Ori
On Tue, Sep 27, 2022 at 8:36 PM Abdullah Sevincer
wrote:
>
> This patch addresses an issue of enqueuing more than
> max_enq_depth and not able to dequeuing events equal
> to max_cq_depth in a single call of rte_event_enqueue_burst
> and rte_event_dequeue_burst.
>
> Apply fix for restricting enqueu
Hello:
In gro_tcp4_reassemble function, tcp data len is calculated:
tcp_dl = pkt->pkt_len - hdr_len;
https://github.com/DPDK/dpdk/blob/v22.07/lib/gro/gro_tcp4.c#L232
if packets < 60 bytes, pkt_len will contain padding bytes, tcp_dl is
incorrectly calculated. this will result the wrong data length
On Mon, Sep 26, 2022 at 8:35 PM wrote:
>
> From: Pavan Nikhilesh
>
> During cleanup `rte_event_port_quiesce` should be called
> irrespective of whether an event has been dequeued or not
> to flush any prefetched events.
>
> Fixes: 7da008df0c8d ("app/eventdev: use port quiescing")
>
> Signed-off-b
28/09/2022 14:22, Akhil Goyal:
> --- /dev/null
> +++ b/lib/net/rte_macsec.h
> +#ifndef _RTE_MACSEC_H_
> +#define _RTE_MACSEC_H_
[...]
> +#endif /* RTE_MACSEC_H_ */
Discrepancy spotted here. Anyway no need of underscores at all.
I'll rename to RTE_MACSEC_H while merging.
https://bugs.dpdk.org/show_bug.cgi?id=1086
Bug ID: 1086
Summary: Significant TX packet drops with Mellanox NIC (mlx5
PMD)
Product: DPDK
Version: 21.11
Hardware: x86
OS: Linux
Status: UNCONFIRMED
Add basic unit tests covering all nine uncore API's.
Signed-off-by: Tadhg Kearney
Reviewed-by: David Hunt
---
app/test/meson.build | 2 +
app/test/test_power_uncore.c | 301 +++
2 files changed, 303 insertions(+)
create mode 100644 app/test/test_power_
Add option for setting uncore frequency min/max/index, through uncore API.
This will be set for each package and die on the SKU. On exit, uncore min
and max frequency will be reverted back to previous frequencies.
Signed-off-by: Tadhg Kearney
Reviewed-by: David Hunt
---
.../sample_app_ug/l3_for
Add API to allow uncore frequency adjustment. This is done through
manipulating related uncore frequency control sysfs entries to
adjust the minimum and maximum uncore frequency values.
Nine API's are being added that are all public and experimental.
Signed-off-by: Tadhg Kearney
Reviewed-by: Davi
This is targeting 22.11 and aims to add an API to DPDK power library to allow
uncore frequency adjustment. This will be called through the l3fwd-power app
and gives the ability to set the minimum and maximum uncore frequency to both
min,
max or specific frequency index.
Signed-off-by: tadhgkearne
On Wed, Sep 28, 2022 at 05:53:06PM +0530, Jerin Jacob wrote:
> On Wed, Sep 28, 2022 at 5:38 PM Thomas Monjalon wrote:
> >
> > 28/09/2022 13:14, Jerin Jacob:
> > > On Tue, Sep 27, 2022 at 8:39 PM Bruce Richardson
> > > wrote:
> > > >
> > > > On Tue, Sep 27, 2022 at 04:36:18PM +0200, Thomas Monjalo
On Wed, Sep 28, 2022 at 05:52:51PM +0530, Akhil Goyal wrote:
> Added MACsec protocol header to be used for supporting
> MACsec protocol offload in hardware or directly in the application.
>
> Signed-off-by: Akhil Goyal
Acked-by: Olivier Matz
Thanks
> Subject: RE: [PATCH v4 2/6] crypto/scheduler: use unified session
>
> > Have the sym sessions changes been tested with the dpdk-test-crypto-perf
> tool ?
>
> I have not tested for scheduler pmd. Can you root cause the issue and send a
> fix?
> Fan may have tested it.
Can this fix be taken up a
On Wed, Sep 21, 2022 at 07:26:20PM +0530, Shijith Thotton wrote:
> Enabled the flag pmd_iova_as_va in cnxk driver build files as they work
> with IOVA as VA. Updated cn9k and cn10k soc build configurations to
> enable the IOVA as VA build by default.
>
> Signed-off-by: Shijith Thotton
> ---
> co
> Subject: [PATCH v2 0/3] security: support MACsec
>
> Added support for MACsec in rte_security for offloading
> MACsec Protocol operation to inline NIC device or a crypto device.
>
> To support MACsec we cannot just make one security session and
> send with the packet to process it. MACsec speci
On Wed, Sep 21, 2022 at 07:26:19PM +0530, Shijith Thotton wrote:
> Swapped position of mbuf next pointer and second dynamic field (dynfield2)
> if the build is configured to use IOVA as VA. This is to move the mbuf
> next pointer to first cache line. kni library is disabled for this
> change as it
On Wed, Sep 28, 2022 at 09:24:51AM +0200, Thomas Monjalon wrote:
> 21/09/2022 15:56, Shijith Thotton:
> > mbuf physical address field is not used in builds which only uses VA. It
> > is used to expand the dynamic field area.
> >
> > Signed-off-by: Shijith Thotton
>
> We cannot condition the use
Hi,
On Wed, Sep 21, 2022 at 07:26:18PM +0530, Shijith Thotton wrote:
> mbuf physical address field is not used in builds which only uses VA. It
> is used to expand the dynamic field area.
>
> Signed-off-by: Shijith Thotton
> ---
> lib/mbuf/rte_mbuf_core.h | 26 +-
> lib/
On Wed, Sep 21, 2022 at 07:26:17PM +0530, Shijith Thotton wrote:
> IOVA mode in DPDK is either PA or VA. The new build option iova_as_va
> configures the mode to VA at compile time and prevents setting it to PA
> at runtime. For now, all drivers which are not always enabled are
> disabled with this
Hi Shijith,
On Wed, Sep 21, 2022 at 07:26:16PM +0530, Shijith Thotton wrote:
> This is a continuation of the discussions[1] to add mbuf physical address
> field to dynamic field.
> Previous version was to add PA field to dynamic field area based on the EAL
> IOVA mode option. It was
> deemed uns
Added MACsec inline cases for encryption cases.
Signed-off-by: Akhil Goyal
---
app/test/meson.build |1 +
app/test/test_security_inline_macsec.c| 821 ++
.../test_security_inline_macsec_vectors.h | 2318 +
3 files changed, 3140 insert
1 - 100 of 149 matches
Mail list logo
