Hi Ciara,
Looks good.
From: Ciara Power
> Make is no longer supported for compiling DPDK, references are now
> removed in the documentation.
>
> Signed-off-by: Ciara Power
Acked-by: Matan Azrad
This patch adds the cryptodev raw API test support to unit test.
In addition a new test-case for QAT PMD for the test type is
enabled.
Signed-off-by: Fan Zhang
Acked-by: Adam Dybkowski
---
app/test/test_cryptodev.c | 787 --
app/test/test_cryptodev.h
This patch adds raw data-path APIs for enqueue and dequeue
operations to cryptodev. The APIs support flexible user-define
enqueue and dequeue behaviors.
Signed-off-by: Fan Zhang
Signed-off-by: Piotr Bronowski
Acked-by: Adam Dybkowski
---
doc/guides/cryptodevs/features/default.ini| 1 +
d
This patch updates QAT PMD to add raw data-path API support.
Signed-off-by: Fan Zhang
Acked-by: Adam Dybkowski
---
doc/guides/cryptodevs/features/qat.ini | 1 +
doc/guides/rel_notes/release_20_11.rst | 4 +
drivers/crypto/qat/meson.build | 1 +
drivers/crypto/qat/qat_sym.h
This patch updates ``rte_crypto_sym_vec`` structure to add
support for both cpu_crypto synchrounous operation and
asynchronous raw data-path APIs. The patch also includes
AESNI-MB and AESNI-GCM PMD changes, unit test changes and
documentation updates.
Signed-off-by: Fan Zhang
---
app/test/test_c
The Crypto Raw data-path APIs are a set of APIs designed to enable external
libraries/applications to leverage the cryptographic processing provided by
DPDK crypto PMDs through the cryptodev API but in a manner that is not
dependent on native DPDK data structures (eg. rte_mbuf, rte_crypto_op, ... e
This patch adds the cryptodev raw API test support to unit test.
In addition a new test-case for QAT PMD for the test type is
enabled.
Signed-off-by: Fan Zhang
Acked-by: Adam Dybkowski
---
app/test/test_cryptodev.c | 787 --
app/test/test_cryptodev.h
This patch updates ``rte_crypto_sym_vec`` structure to add
support for both cpu_crypto synchrounous operation and
asynchronous raw data-path APIs. The patch also includes
AESNI-MB and AESNI-GCM PMD changes, unit test changes and
documentation updates.
Signed-off-by: Fan Zhang
Acked-by: Adam Dybko
This patch adds raw data-path APIs for enqueue and dequeue
operations to cryptodev. The APIs support flexible user-define
enqueue and dequeue behaviors.
Signed-off-by: Fan Zhang
Signed-off-by: Piotr Bronowski
Acked-by: Adam Dybkowski
---
doc/guides/cryptodevs/features/default.ini| 1 +
d
This patch updates QAT PMD to add raw data-path API support.
Signed-off-by: Fan Zhang
Acked-by: Adam Dybkowski
---
doc/guides/cryptodevs/features/qat.ini | 1 +
doc/guides/rel_notes/release_20_11.rst | 4 +
drivers/crypto/qat/meson.build | 1 +
drivers/crypto/qat/qat_sym.h
The Crypto Raw data-path APIs are a set of APIs designed to enable external
libraries/applications to leverage the cryptographic processing provided by
DPDK crypto PMDs through the cryptodev API but in a manner that is not
dependent on native DPDK data structures (eg. rte_mbuf, rte_crypto_op, ... e
The API ``rte_security_session_create`` takes only single
mempool for session and session private data. So the
application need to create mempool for twice the number of
sessions needed and will also lead to wastage of memory as
session private data need more memory compared to session.
Hence the A
Hi Lukasz,
Thanks for the review.
> > diff --git a/app/test/test_security.c b/app/test/test_security.c
> > index 77fd5adc6..ed7de348f 100644
> > --- a/app/test/test_security.c
> > +++ b/app/test/test_security.c
> > @@ -237,6 +237,7 @@ static struct mock_session_create_data {
> > struct rte_se
Hi David,
> Hi Akhil
>
> > -Original Message-
> > From: akhil.go...@nxp.com
> > Sent: Thursday, September 3, 2020 9:10 PM
>
>
>
> > diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c index
> > 70bf6fe2c..6d7da1408 100644
> > --- a/app/test/test_cryptodev.c
> > +++ b/app
Hi Fan,
OK got your point. Review the changes in the documentation of APIs that I
suggested in the 2/4 patch according to the below explanation.
Regards,
Akhil
From: Zhang, Roy Fan
Sent: Sunday, October 11, 2020 2:21 AM
To: Akhil Goyal ; dev@dpdk.org
Cc: Dybkowski, AdamX
Subject: Re: [dpdk-de
Hi Akhil,
For your “always return 1” question:
The way dequeue_burst API works is we may not know how many ops to dequeue
without parsing the first user data (e.g. a structure containing n_ops data).
It is up to the user to provide a callback function to return the number of ops
- either by pa
Hi Akhil,
Will upate in v11.
regards,
Fan
发件人: Akhil Goyal
发送时间: Saturday, October 10, 2020 8:38:39 PM
收件人: Zhang, Roy Fan ; dev@dpdk.org
抄送: Bronowski, PiotrX ; Dybkowski, AdamX
; Ananyev, Konstantin
主题: RE: [dpdk-dev v11 2/4] cryptodev: add raw crypto data-
On Sat, Oct 10, 2020 at 12:42 PM Wei Hu (Xavier)
wrote:
> From: Chengchang Tang
>
> This patch adds checking whether the related Tx or Rx queue has been
> setupped in the queue-related API functions to avoid illegal address
> access. And validity check of the queue_id is also added in the API
>
Hi Fan,
> +static uint32_t
> +get_raw_dp_dequeue_count(void *user_data __rte_unused)
> +{
> + return 1;
Why is this 1 always? There could be jobs >1 which are processed.
> +}
> +
> +static void
> +post_process_raw_dp_op(void *user_data, uint32_t index __rte_unused,
> + uint8_
Hi Fan,
>
> This patch adds raw data-path APIs for enqueue and dequeue
> operations to cryptodev. The APIs support flexible user-define
> enqueue and dequeue behaviors.
>
> Signed-off-by: Fan Zhang
> Signed-off-by: Piotr Bronowski
> Acked-by: Adam Dybkowski
> ---
> doc/guides/cryptodevs/featu
> This patch updates fips validation GCM test capabilities:
>
> - In NIST GCMVS spec GMAC test vectors are the GCM ones with
> plaintext length as 0 and uses AAD as input data. Originally
> fips_validation tests treats them both as GCM test vectors.
> This patch introduce automatic test type recog
W dniu 10.10.2020 o 10:12, David Marchand pisze:
> Hello Lukasz,
>
> On Sat, Oct 10, 2020 at 1:26 AM Lukasz Wojciechowski
> wrote:
>> W dniu 09.10.2020 o 23:41, Lukasz Wojciechowski pisze:
>> More bad news - same issue just appeared on travis for v6.
>> Good news we can reproduce it.
>>
>> Is th
In all distributor tests there is a chance that tests
will send packets to distributor with rte_distributor_process()
before workers are started and requested for packets.
This patch ensures that all packets are delivered to workers
by calling rte_distributor_process() in loop until number
of succ
rte_distributor_flush() is using total_outstanding()
function to calculate if it should still wait
for processing packets. However in burst mode
only backlog packets were counted.
This patch fixes that issue by counting also in flight
packets. There are also sum fixes to properly keep
count of in
The patch clears distributors returns buffer
in clear_returns() by setting start and count to 0.
Fixes: 775003ad2f96 ("distributor: add new burst-capable library")
Cc: david.h...@intel.com
Cc: sta...@dpdk.org
Signed-off-by: Lukasz Wojciechowski
Acked-by: David Hunt
---
lib/librte_distributor/r
Fix improper indexes while comparing tags.
In the find_match_scalar() function:
* j iterates over flow tags of following packets;
* w iterates over backlog or in flight tags positions.
Fixes: 775003ad2f96 ("distributor: add new burst-capable library")
Cc: david.h...@intel.com
Cc: sta...@dpdk.org
All of the former tests analyzed only statistics
of packets processed by all workers.
The new test verifies also if packets are processed
on workers as expected.
Every packets processed by the worker is marked
and analyzed after it is returned to distributor.
This test allows finding issues in mat
Instead of making delays in test code and waiting
for worker hopefully to reach proper states,
synchronize worker shutdown test cases with spin lock
on atomic variable.
Fixes: c0de0eb82e40 ("distributor: switch over to new API")
Cc: david.h...@intel.com
Cc: sta...@dpdk.org
Signed-off-by: Lukasz W
After introducing burst API there were some artefacts in the
API documentation from legacy single API.
Also the rte_distributor_poll_pkt() function return values
mismatched the implementation.
Fixes: c0de0eb82e40 ("distributor: switch over to new API")
Cc: david.h...@intel.com
Cc: sta...@dpdk.org
During quit_workers function distributor's main core processes
some packets to wake up pending worker cores so they can quit.
As quit_workers acts also as a cleanup procedure for next test
case it should also collect these packages returned by workers'
handlers, so the cyclic buffer with returned p
Statistics of handled packets are cleared and read on main lcore,
while they are increased in workers handlers on different lcores.
Without synchronization occasionally showed invalid values.
This patch uses atomic acquire/release mechanisms to synchronize.
Fixes: c3eabff124e6 ("distributor: add
Sanity tests with mbuf alloc and shutdown tests assume that
mbufs passed to worker cores are freed in handlers.
Such packets should not be returned to the distributor's main
core. The only packets that should be returned are the packets
send after completion of the tests in quit_workers function.
In the single legacy version of the distributor synchronization
requires continues exchange of buffers between distributor
and workers. Empty buffers are sent if only handshake
synchronization is required.
However calls to the rte_distributor_return_pkt()
with 0 buffers in single mode were ignored
The sanity test with worker shutdown delegates all bufs
to be processed by a single lcore worker, then it freezes
one of the lcore workers and continues to send more bufs.
The freezed core shuts down first by calling
rte_distributor_return_pkt().
The test intention is to verify if packets assigned
The burst version of distributor implementation was missing proper
handling of worker shutdown. A worker processing packets received
from distributor can call rte_distributor_return_pkt() function
informing distributor that it want no more packets. Further calls to
rte_distributor_request_pkt() or
Synchronization of data exchange between distributor and worker cores
is based on 2 handshakes: retptr64 for returning mbufs from workers
to distributor and bufptr64 for passing mbufs to workers.
Without proper order of verifying those 2 handshakes a deadlock may
occur. This can happen when worker
rte_distributor_request_pkt and rte_distributor_get_pkt dereferenced
oldpkt parameter when in RTE_DIST_ALG_SINGLE even if number
of returned buffers from worker to distributor was 0.
This patch passes NULL to the legacy API when number of returned
buffers is 0. This allows passing NULL as oldpkt p
During review and verification of the patch created by Sarosh Arif:
"test_distributor: prevent memory leakages from the pool" I found out
that running distributor unit tests multiple times in a row causes fails.
So I investigated all the issues I found.
There are few synchronization issues that mi
rte_distributor_return_pkt function which is run on worker cores
must wait for distributor core to clear handshake on retptr64
before using those buffers. While the handshake is set distributor
core controls buffers and any operations on worker side might overwrite
buffers which are unread yet.
Sam
On Sat, 10 Oct 2020 15:12:12 +0800
"Wei Hu (Xavier)" wrote:
> + if (dev->data->rx_queues[rx_queue_id] == NULL) {
> + RTE_ETHDEV_LOG(ERR, "Rx queue %"PRIu16" of device with
> port_id=%"
> + PRIu16" has not been setupped\n",
> +
net/bonding: LACP Packet statistics support
Store the LACP packets sent and received for each slave.
This can be used for debug purposes from any DPDK application.
Signed-Off-By: Kiran K N
Change-Id: Iae82bd7d0879a4c4333a292c96d431798c56e301
---
drivers/net/bonding/eth_bond_8023ad_private.h |
> >> Add two new power management intrinsics, and provide an implementation
> >> in eal/x86 based on UMONITOR/UMWAIT instructions. The instructions
> >> are implemented as raw byte opcodes because there is not yet widespread
> >> compiler support for these instructions.
> >>
>
Hi Ciara,
> -Original Message-
> From: Power, Ciara
> Sent: Wednesday, September 30, 2020 21:04
> To: dev@dpdk.org
> Cc: Power, Ciara ; Zhao1, Wei ;
> Guo, Jia
> ; Wang, Haiyue
> Subject: [PATCH v3 11/18] net/ixgbe: add checks for max SIMD bitwidth
>
> When choosing a vector path to ta
Hi David,
> > > This patch adds support for run-time selection of the optimal
> > > architecture-specific CRC path, based on the supported instruction
> > > set(s) of the CPU.
> > >
> > > The compiler option checks have been moved from the C files to the
> > > meson script. The rte_cpu_get_flag
> -Original Message-
> From: dev On Behalf Of Harry van Haaren
> Sent: Monday, September 14, 2020 11:10 PM
> To: dev@dpdk.org
> Cc: pbhagavat...@marvell.com; Harry van Haaren
>
> Subject: [dpdk-dev] [PATCH v2] eal: add new prefetch write variants
>
> This commit adds a new rte_prefetch
https://bugs.dpdk.org/show_bug.cgi?id=553
Bug ID: 553
Summary: iavf_fdir/negative_case: run command with invalid port
'flow flush 2', it can't print error info or warning
info.
Product: DPDK
Version: 20.11
On 10/9/2020 5:47 PM, Ferruh Yigit wrote:
> On 10/9/2020 4:03 AM, jiawe...@trustnetic.com wrote:
> > Hi Ferruh,
> >
> > For the syntax/style check issue, should I fix all the errors and warnings
> > or
> just fix the errors?
> > It seems to be a lot of warnings.
> >
>
> [Please don't top post, it
> -Original Message-
> From: dev On Behalf Of Mairtin o Loingsigh
> Sent: Friday, October 9, 2020 9:51 PM
> To: jasvinder.si...@intel.com; bruce.richard...@intel.com;
> pablo.de.lara.gua...@intel.com; konstantin.anan...@intel.com
> Cc: dev@dpdk.org; brendan.r...@intel.com; mairtin.oloing
Tested-by: Jiang, YuX
Best Regards
Jiang yu
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Maxime Coquelin
> Sent: Friday, October 9, 2020 3:36 PM
> To: Zhang, Roy Fan ; dev@dpdk.org
> Cc: Xia, Chenbo ; Liu, Changpeng
> ; Yigit, Ferruh ;
> sta...@dpdk
add non temporal load and temporal store for mprq memcpy.
define RTE_LIBRTE_MLX5_NTLOAD_TSTORE_ALIGN_COPY in build
configuration to enable this optimization.
Signed-off-by: Aman Kumar
---
drivers/net/mlx5/meson.build | 1 +
drivers/net/mlx5/mlx5.c | 12
drivers/net/mlx5/mlx5.h
add non temporal store for few WQE fields to optimize
data path. Define RTE_LIBRTE_MLX5_NT_STORE in build
configurations to enable this optimization.
Signed-off-by: Aman Kumar
---
drivers/net/mlx5/meson.build | 1 +
drivers/net/mlx5/mlx5.c | 17 ++
drivers/net/mlx5/mlx5.h
add non temporal store for few WQE fields to optimize
data path. Define RTE_LIBRTE_MLX5_NT_STORE in build
configurations to enable this optimization.
Signed-off-by: Aman Kumar
---
drivers/net/mlx5/meson.build | 1 +
drivers/net/mlx5/mlx5.c | 17 ++
drivers/net/mlx5/mlx5.h
add non temporal load and temporal store for mprq memcpy.
define RTE_LIBRTE_MLX5_NTLOAD_TSTORE_ALIGN_COPY in build
configuration to enable this optimization.
Signed-off-by: Aman Kumar
---
drivers/net/mlx5/meson.build | 1 +
drivers/net/mlx5/mlx5.c | 12
drivers/net/mlx5/mlx5.h
From: Kishore Padmanabha
Added support in the ULP mapper to enable parent child flow
creation and destroy. This feature enables support for the vxlan
decap functionality.
Signed-off-by: Kishore Padmanabha
Reviewed-by: Mike Baucom
Reviewed-by: Ajit Khaparde
---
drivers/net/bnxt/tf_ulp/ulp_flo
From: Mike Baucom
Name changes due to consolidating the template table processing
and hence are not necessary.
- chip before type in name
- removal of class in key field info
Signed-off-by: Mike Baucom
Reviewed-by: Ajit Khaparde
Reviewed-by: Kishore Padmanabha
---
drivers/net/bnxt/tf_ulp/ul
From: Mike Baucom
FDB type is now driven by the caller, not the template.
So remove it.
Signed-off-by: Mike Baucom
Reviewed-by: Ajit Khaparde
Reviewed-by: Kishore Padmanabha
---
.../tf_ulp/ulp_template_db_stingray_act.c | 18 ++---
.../tf_ulp/ulp_template_db_stingray_class.c | 69 +
On Sat, Oct 10, 2020 at 10:12 AM David Marchand
wrote:
>
> Hello Lukasz,
>
> On Sat, Oct 10, 2020 at 1:26 AM Lukasz Wojciechowski
> wrote:
> > W dniu 09.10.2020 o 23:41, Lukasz Wojciechowski pisze:
> > More bad news - same issue just appeared on travis for v6.
> > Good news we can reproduce it.
>
From: Venkat Duvvuru
Currently, we are only registering to this event if the function
is a trusted VF. This patch extends it for PFs as well.
Fixes: 322bd6e70272 ("net/bnxt: add port representor infrastructure")
Cc: sta...@dpdk.org
Signed-off-by: Venkat Duvvuru
Reviewed-by: Somnath Kotur
Revi
From: Kishore Padmanabha
The default and regular flows are stored in the same flow table
instead of different flow tables. This should help code reuse
and reducing the number of allocations.
So combine default and regular flows in flow database.
Signed-off-by: Kishore Padmanabha
Reviewed-by: Mi
From: Kishore Padmanabha
Added support for parent child flow database apis. This
feature adds support to enable vxlan decap support where
flows needs to maintain parent-child flow relationship.
Signed-off-by: Kishore Padmanabha
Reviewed-by: Mike Baucom
Reviewed-by: Ajit Khaparde
---
drivers/
From: Mike Baucom
The table processing has been consolidated to be able to reuse the same
code for action and classification template processing.
Signed-off-by: Mike Baucom
Reviewed-by: Kishore Padmanabha
Reviewed-by: Ajit Khaparde
---
drivers/net/bnxt/tf_ulp/bnxt_ulp_flow.c | 1 +
d
From: Kishore Padmanabha
The session details that is shared among multiple ports
need to be outside the bnxt structure.
Fixes: 70e64b27af5b ("net/bnxt: support ULP session manager cleanup")
Cc: sta...@dpdk.org
Signed-off-by: Kishore Padmanabha
Reviewed-by: Mike Baucom
Reviewed-by: Ajit Khapar
From: Venkat Duvvuru
1. Implement HWRM_FUNC_VF_RESOURCE_CFG command and use it to
reserve resources for VFs when NEW RM is enabled.
2. Invoke PF’s FUNC_CFG before configuring VFs resources.
3. Don’t consider max_rx_em_flows in max_l2_ctx calculation
when VFs are configured.
4. Issue HWRM_FU
Hello Lukasz,
On Sat, Oct 10, 2020 at 1:26 AM Lukasz Wojciechowski
wrote:
> W dniu 09.10.2020 o 23:41, Lukasz Wojciechowski pisze:
> More bad news - same issue just appeared on travis for v6.
> Good news we can reproduce it.
>
> Is there a way to delegate a job for travis other way than sending a
From: Kishore Padmanabha
Added support in the ULP mapper to enable parent child flow
creation and destroy. This feature enables support for the vxlan
decap functionality.
Signed-off-by: Kishore Padmanabha
Reviewed-by: Mike Baucom
Reviewed-by: Ajit Khaparde
---
drivers/net/bnxt/tf_ulp/ulp_flo
Fixes and enchancements in the bnxt PMD, mostly in the
TRUFLOW layer, including templates to add support for
Stingray device.
v2:
- squashed patch patch 13 to patch 7
- updated and fixed commit logs
- updated docs and release notes where necessary
Kishore Padmanabha (4):
net/bnxt: fix the corru
From: Mike Baucom
FDB type is now driven by the caller, not the template.
So remove it.
Signed-off-by: Mike Baucom
Reviewed-by: Ajit Khaparde
Reviewed-by: Kishore Padmanabha
---
.../tf_ulp/ulp_template_db_stingray_act.c | 18 ++---
.../tf_ulp/ulp_template_db_stingray_class.c | 69 +
From: Kishore Padmanabha
Added support for parent child flow database apis. This
feature adds support to enable vxlan decap support where
flows needs to maintain parent-child flow relationship.
Signed-off-by: Kishore Padmanabha
Reviewed-by: Mike Baucom
Reviewed-by: Ajit Khaparde
---
drivers/
From: Mike Baucom
Name changes due to consolidating the template table processing
and hence are not necessary.
- chip before type in name
- removal of class in key field info
Signed-off-by: Mike Baucom
Reviewed-by: Ajit Khaparde
Reviewed-by: Kishore Padmanabha
---
drivers/net/bnxt/tf_ulp/ul
From: Venkat Duvvuru
Currently, we are only registering to this event if the function
is a trusted VF. This patch extends it for PFs as well.
Fixes: 322bd6e70272 ("net/bnxt: add port representor infrastructure")
Cc: sta...@dpdk.org
Signed-off-by: Venkat Duvvuru
Reviewed-by: Somnath Kotur
Revi
From: Mike Baucom
The table processing has been consolidated to be able to reuse the same
code for action and classification template processing.
Signed-off-by: Mike Baucom
Reviewed-by: Kishore Padmanabha
Reviewed-by: Ajit Khaparde
---
drivers/net/bnxt/tf_ulp/bnxt_ulp_flow.c | 1 +
d
09/10/2020 22:32, Ferruh Yigit:
> On 10/6/2020 9:33 AM, Olivier Matz wrote:
> > On Mon, Oct 05, 2020 at 01:23:08PM +0100, Ferruh Yigit wrote:
> >> On 9/28/2020 4:43 PM, Stephen Hemminger wrote:
> >>> On Mon, 28 Sep 2020 17:24:26 +0200
> >>> Thomas Monjalon wrote:
> 28/09/2020 15:53, Ferruh Yi
From: Kishore Padmanabha
The default and regular flows are stored in the same flow table
instead of different flow tables. This should help code reuse
and reducing the number of allocations.
So combine default and regular flows in flow database.
Signed-off-by: Kishore Padmanabha
Reviewed-by: Mi
From: Venkat Duvvuru
1. Implement HWRM_FUNC_VF_RESOURCE_CFG command and use it to
reserve resources for VFs when NEW RM is enabled.
2. Invoke PF’s FUNC_CFG before configuring VFs resources.
3. Don’t consider max_rx_em_flows in max_l2_ctx calculation
when VFs are configured.
4. Issue HWRM_FU
Fixes and enchancements in the bnxt PMD, mostly in the
TRUFLOW layer, including templates to add support for
Stingray device.
v2:
- squashed patch patch 13 to patch 7
- updated and fixed commit logs
- updated docs and release notes where necessary
Kishore Padmanabha (4):
net/bnxt: fix the corru
From: Kishore Padmanabha
The session details that is shared among multiple ports
need to be outside the bnxt structure.
Fixes: 70e64b27af5b ("net/bnxt: support ULP session manager cleanup")
Cc: sta...@dpdk.org
Signed-off-by: Kishore Padmanabha
Reviewed-by: Mike Baucom
Reviewed-by: Ajit Khapar
On 2020/10/5 17:11, Maxime Coquelin wrote:
On 10/5/20 10:57 AM, Thomas Monjalon wrote:
24/09/2020 07:41, Stephen Hemminger:
On Fri, 11 Sep 2020 17:54:48 +0200
Thomas Monjalon wrote:
As decided in the Technical Board in November 2019,
the kernel module igb_uio is moved to the dpdk-kmods re
From: Chengchang Tang
This patch adds checking whether the related Tx or Rx queue has been
setupped in the queue-related API functions to avoid illegal address
access. And validity check of the queue_id is also added in the API
functions rte_eth_dev_rx_intr_enable and rte_eth_dev_rx_intr_disable.
78 matches
Mail list logo
