> -Original Message-
> From: Sevincer, Abdullah
> 1) What is the difference between "--edev-stats-enable" and "--edev-device-
> stats" ?
> --edev-stats-enable enables xstats for evendev and exits from proc-info
> application (when this flag is enabled) after displaying xstats. I
> -Original Message-
> From: Konstantin Ananyev
> Sent: Thursday, March 2, 2023 6:22 PM
> To: dev@dpdk.org
> Subject: Re: [PATCH 1/5] ethdev: fix race-condition of proactive error
> handling
> mode
>
>
>
> >
> >> -Original Message-
> >> From: Chengwen Feng
> >> Sent: Tuesday
From: zhipeng Lu
add timestamp for log
Signed-off-by: zhipeng Lu
---
lib/eal/common/eal_common_log.c | 27 +--
1 file changed, 25 insertions(+), 2 deletions(-)
diff --git a/lib/eal/common/eal_common_log.c b/lib/eal/common/eal_common_log.c
index bd7b188ceb..0a21572954 1
From: zhipeng Lu
add timestamp for log
Signed-off-by: zhipeng Lu
---
lib/eal/common/eal_common_log.c | 25 -
1 file changed, 24 insertions(+), 1 deletion(-)
diff --git a/lib/eal/common/eal_common_log.c b/lib/eal/common/eal_common_log.c
index bd7b188ceb..2b481116b6 1006
On 3/3/2023 7:04 PM, Long Li wrote:
>> Subject: Re: [PATCH] net/mana: use RTE_LOG_DP for logs on datapath
>>
>> On 3/3/2023 2:16 AM, Long Li wrote:
Subject: Re: [PATCH] net/mana: use RTE_LOG_DP for logs on datapath
On Thu, 23 Feb 2023 10:09:17 -0800
Stephen Hemminger wrote:
>>>
On Fri, 3 Mar 2023 13:54:23 +
Ferruh Yigit wrote:
> On 3/3/2023 2:16 AM, Long Li wrote:
> >> Subject: Re: [PATCH] net/mana: use RTE_LOG_DP for logs on datapath
> >>
> >> On Thu, 23 Feb 2023 10:09:17 -0800
> >> Stephen Hemminger wrote:
> >>
> >>> On Thu, 23 Feb 2023 14:07:25 +
> >>> Fer
On Fri, Mar 03, 2023 at 03:12:19PM +0100, Thomas Monjalon wrote:
> 02/03/2023 18:17, Tyler Retzlaff:
> > On Thu, Mar 02, 2023 at 02:21:48PM +0100, Thomas Monjalon wrote:
> > > The DevX library path had to be provided through the variables
> > > DEVX_INC_PATH and DEVX_LIB_PATH.
> > > It was non-stan
> Subject: Re: [PATCH] net/mana: use RTE_LOG_DP for logs on datapath
>
> On 3/3/2023 2:16 AM, Long Li wrote:
> >> Subject: Re: [PATCH] net/mana: use RTE_LOG_DP for logs on datapath
> >>
> >> On Thu, 23 Feb 2023 10:09:17 -0800
> >> Stephen Hemminger wrote:
> >>
> >>> On Thu, 23 Feb 2023 14:07:25 +
Thanks for formalizing our process.
13/02/2023 10:26, jer...@marvell.com:
> --- /dev/null
> +++ b/content/process/_index.md
First question: is the website the best place for this process?
Inside the code guides, we have a contributing section,
but I'm not sure it is a good fit for the decision p
On 2/26/2023 5:22 PM, Konstantin Ananyev wrote:
>
>>> If ethdev enqueue or dequeue function is called during
>>> eth_dev_fp_ops_setup(), it may get pre-empted after setting the
>>> function pointers, but before setting the pointer to port data.
>>> In this case the
On 3/3/2023 1:49 AM, fengchengwen wrote:
> On 2023/3/2 21:01, Konstantin Ananyev wrote:
>>
>>
>>>
>>> This patch adds error recovery usage demo which will:
>>> 1. stop packet forwarding when the RTE_ETH_EVENT_ERR_RECOVERING event
>>>is received.
>>> 2. restart packet forwarding when the RTE_ETH
On 3/2/2023 12:08 PM, Konstantin Ananyev wrote:
>
>> In the proactive error handling mode, the PMD will set the data path
>> pointers to dummy functions and then try recovery, in this period the
>> application may still invoking data path API. This will introduce a
>> race-condition with data path
On 2/20/2023 6:08 AM, Ashok Kaladi wrote:
> If ethdev enqueue or dequeue function is called during
> eth_dev_fp_ops_setup(), it may get pre-empted after setting
> the function pointers, but before setting the pointer to port data.
> In this case the newly registered enqueue/dequeue function will us
On 3/2/23 21:22, Hernan Vargas wrote:
New feature to add BLER support for 4G in test-bbdev.
Signed-off-by: Hernan Vargas
---
app/test-bbdev/test_bbdev_perf.c | 195 ++-
1 file changed, 194 insertions(+), 1 deletion(-)
diff --git a/app/test-bbdev/test_bbdev_per
Acked-by: Tejasree Kondoj
> -Original Message-
> From: Ciara Power
> Sent: Friday, March 3, 2023 3:08 PM
> Cc: dev@dpdk.org; kai...@intel.com; pablo.de.lara.gua...@intel.com; Ciara
> Power
> Subject: [EXT] [PATCH v2 0/5] test/crypto: add ZUC256 chained tests
>
> External Email
>
> ---
1) What is the difference between "--edev-stats-enable" and
"--edev-device-stats" ?
--edev-stats-enable enables xstats for evendev and exits from proc-info
application (when this flag is enabled) after displaying xstats. It is like
show mempool flag so we don't want bunch of differen
Release status meeting minutes 2023-03-02
=
Agenda:
* Release Dates
* Subtrees
* Roadmaps
* LTS
* Defects
* Opens
Participants:
* AMD
* ARM
* Debian/Microsoft
* Intel
* Marvell
* Red Hat
Release Dates
-
The following are the proposed current
This patch adds a validation in checkpatch tool, to check if a
tracepoint is present in any new function added in ethdev library.
It uses the existing build_map_changes function to create a map
of functions. In the map, the newly added functions, added in the
experimental section are identified an
This patch series adds a validation in checkpatch tool to check if
tracepoint is present in any new function added in ethdev library.
The first patch in the series moves build_map_changes function from
check-symbol-change.sh to a new file build-symbol-map.sh.
The second patch in the series adds a
This patch moves the build_map_changes function from
check-symbol-change.sh to a new build-symbol-map.sh file.
This function can be used in other scripts by including
build-symbol-map.sh file.
Signed-off-by: Ankur Dwivedi
---
devtools/build-symbol-map.sh| 78 +
On Wed, Mar 1, 2023 at 3:01 AM Abdullah Sevincer
wrote:
>
> This commit fixes invalid memory free operation
> caused by extending xstats ID width from 32 to 64 bits.
>
> Telemetry application crashed when freeing memory,
> hence, the telemetry app could not display xstats.
>
> Fixes: 1bdfe4d76e98
On Thu, Mar 2, 2023 at 4:30 PM Volodymyr Fialko wrote:
>
> dev_fini() can return EAGAIN in case if LF is active(used by other
> devices). Ignoring return code leads to early free of resources that are
> still in use.
>
> Fixes: 8558dcaa05b8 ("event/cnxk: add build infra and device setup")
> Cc: st
On Mon, Feb 27, 2023 at 10:06 PM Akhil Goyal wrote:
>
> > Testeventdev crypto adapter symmetric tests are only attempting
> > NULL cipher algorithm. This limits crypto adapter usage with only
> > PMDs that can support NULL cipher algorithm. Also, since NULL cipher
> > algorithm doesn't perform any
Cc Maxime and Chenbo
On 3/3/23 14:19, Boleslav Stankevich wrote:
The length of TSO payload could not fit into 16 bits provided by the
IPv4 total length and IPv6 payload length fields. Thus, deduce it
from the length of the packet.
Fixes: 696573046e9 ("net/virtio: support TSO")
Cc: sta...@dpdk.o
On Wed, Mar 1, 2023 at 5:05 PM wrote:
>
> From: Pavan Nikhilesh
>
> Update CNXK event device guide to include WQE stashing parameters
> that are enabled by default.
>
> Fixes: ca88db3ae76d ("event/cnxk: support WQE stashing ")
Squashed with above patch and merged to next-eventdev
>
> Signed-off
On Wed, Feb 22, 2023 at 9:35 AM wrote:
>
> From: Satheesh Paul
>
> RTE_FLOW_ITEM_TYPE_IPV6_EXT and RTE_FLOW_ITEM_TYPE_IPV6_FRAG_EXT
> pattern items can be specified following a RTE_FLOW_ITEM_TYPE_IPV6.
> Modified layer C parsing logic to handle this.
>
> Fixes: a800675b06f9 ("net/cnxk: support IP
On 2/21/2023 10:35 AM, Chaoyong He wrote:
> The former logic of set flow actions don't consider the mask filed
> of control message passed to the firmware, and this caused the
> firmware skip the set action logic wrongly.
> This patch series aims to fix this problem.
>
> --
> v2:
> * fix the compi
02/03/2023 15:16, Zhang, Qi Z:
> From: Thomas Monjalon
> > 02/03/2023 14:52, Zhang, Qi Z:
> > > From: Thomas Monjalon
> > > > If IOVA as PA is disabled and the driver does not support IOVA as
> > > > VA, the build of the driver was disabled.
> > > > Unfortunately some drivers were building some s
02/03/2023 18:17, Tyler Retzlaff:
> On Thu, Mar 02, 2023 at 02:21:48PM +0100, Thomas Monjalon wrote:
> > The DevX library path had to be provided through the variables
> > DEVX_INC_PATH and DEVX_LIB_PATH.
> > It was non-standard and triggers some issues with recent Meson.
>
> is it possible for me
I'm not comfortable with this patch.
First, there is no comment in the code which helps to understand the logic.
Second, I'm afraid changing the value of the per-core variable _socket_id
may have an impact on some applications.
16/02/2023 03:50, You, KaisenX:
> From: Burakov, Anatoly
> > On 2/1
The library libpcap has their function 'bpf_validate' either so
there would be a multiple definition issue when linking with
librte_bpf.a and libpcap.a staticly.
You can reproduce this issue by 'meson build -Dprefer_static=true
-Denable_apps=test-pmd -Denable_drivers=net/af_xdp,net/af_packet'.
Not
On 3/3/2023 2:16 AM, Long Li wrote:
>> Subject: Re: [PATCH] net/mana: use RTE_LOG_DP for logs on datapath
>>
>> On Thu, 23 Feb 2023 10:09:17 -0800
>> Stephen Hemminger wrote:
>>
>>> On Thu, 23 Feb 2023 14:07:25 +
>>> Ferruh Yigit wrote:
>>>
Overall I am not sure if anyone is interested i
On 3/3/2023 11:31 AM, Ankur Dwivedi wrote:
> The first patch in this series adds fix for bug and coverity.
> The second patch makes change to pass structure pointer instead of the
> structure value, to avoid 64 bytes copy in function call stack for
> rte_eth_trace_xstats_get_names.
>
> v2:
> - Rem
On 3/3/2023 11:31 AM, Ankur Dwivedi wrote:
> The speed_fec_capa pointer can be null. So dereferencing the pointer is
> removed and only the pointer is captured in trace function.
> Fixed few more trace functions in which null pointer can be dereferenced.
> As a result of this fix the address saniti
On some CI runs, some service-cores tests spuriously fail as the
service lcore thread is not actually scheduled by the OS in the
given amount of time.
Increasing timeouts has not resolved the issue in the CI, so the
solution in this patch is to move them to a separate perf test
suite.
Signed-off-
> -Original Message-
> From: David Marchand
> Sent: Thursday, March 2, 2023 11:48 AM
> To: dev@dpdk.org
> Cc: Semion Simma ; Yajun Wu ;
> Maxime Coquelin ; Chenbo Xia
> ; Morten Brørup
> Subject: [PATCH] vhost: fix vring enable with VDPA
>
> For VDPA devices, vq are not locked once the d
On 3/2/23 21:22, Hernan Vargas wrote:
Added timeout to prevent infinite loop condition if the device
doesn't enqueue/dequeue.
Signed-off-by: Hernan Vargas
---
app/test-bbdev/test_bbdev_perf.c | 133 ++-
1 file changed, 113 insertions(+), 20 deletions(-)
diff -
On 3/2/23 21:22, Hernan Vargas wrote:
Add a call to queue start and queue stop specifically for testing the
bbdev API.
Signed-off-by: Hernan Vargas
---
app/test-bbdev/test_bbdev_perf.c | 19 +--
1 file changed, 13 insertions(+), 6 deletions(-)
diff --git a/app/test-bbdev/
The rte_eth_xstat_name structure is of size 64 bytes. Instead of passing
the structure as value it is passed as a pointer, to avoid copy of 64 bytes
in function call stack.
Signed-off-by: Ankur Dwivedi
Acked-by: Ferruh Yigit
---
lib/ethdev/ethdev_trace.h | 4 ++--
lib/ethdev/rte_ethdev.c | 2
The speed_fec_capa pointer can be null. So dereferencing the pointer is
removed and only the pointer is captured in trace function.
Fixed few more trace functions in which null pointer can be dereferenced.
As a result of this fix the address sanitizer error observed with
rte_eth_trace_find_next_of(
The first patch in this series adds fix for bug and coverity.
The second patch makes change to pass structure pointer instead of the
structure value, to avoid 64 bytes copy in function call stack for
rte_eth_trace_xstats_get_names.
v2:
- Removed null pointer reference in rte_eth_trace_read_clock a
The length of TSO payload could not fit into 16 bits provided by the
IPv4 total length and IPv6 payload length fields. Thus, deduce it
from the length of the packet.
Fixes: 696573046e9 ("net/virtio: support TSO")
Cc: sta...@dpdk.org
Signed-off-by: Boleslav Stankevich
Reviewed-by: Andrew Rybchenk
Hi, Maxime!
Yes, it is a fix and will add missing tags in v2.
Thanks,
Boleslav
On 02/03/2023 13:25, Maxime Coquelin wrote:
Hi Boleslav,
On 2/16/23 13:35, Boleslav Stankevich wrote:
The length of TSO payload could not fit into 16 bits provided by the
IPv4 total length and IPv6 payload length
> -Original Message-
> From: David Marchand
> Sent: Monday, February 27, 2023 4:09 PM
> To: Van Haaren, Harry
> Cc: dev@dpdk.org; dpdk...@iol.unh.edu; c...@dpdk.org;
> honnappa.nagaraha...@arm.com; mattias.ronnblom
> ; m...@smartsharesystems.com;
> roret...@linux.microsoft.com; acon...@re
> -Original Message-
> From: Abdullah Sevincer
> Subject: [PATCH v6] app/procinfo: display eventdev xstats for PMD data
>
I see you are supporting evendev stats print.
You can make the heading clear by removing "PMD data"
> +++ b/doc/guides/tools/proc_info.rst
> @@ -22,7 +22,9 @@
> -Original Message-
> From: David Marchand
> Sent: Friday, March 3, 2023 3:59 PM
> To: Jiang, YuX ; Thomas Monjalon
>
> Cc: dev (dev@dpdk.org) ; Kovacevic, Marko
> ; Mcnamara, John ;
> Richardson, Bruce ; Ferruh Yigit
> ; Kevin Traynor ;
> mattias.ronnblom
> Subject: Re: release candida
Document how to configure and run DTS.
Also add documentation related to new features: SUT setup and a brief
test suite implementation cookbook.
Signed-off-by: Juraj Linkeš
---
doc/guides/tools/dts.rst | 165 ++-
1 file changed, 163 insertions(+), 2 deletions(
The module stores the results and errors from all executions, build
targets, test suites and test cases.
The result consist of the result of the setup and the teardown of each
testing stage (listed above) and the results of the inner stages. The
innermost stage is the case, which also contains the
The config allows users to specify which test suites and test cases
within test suites to run.
Also add test suite running capabilities to dts runner.
Signed-off-by: Juraj Linkeš
---
dts/conf.yaml | 2 ++
dts/framework/config/__init__.py | 29 +
The test suite implements test cases defined in the corresponding test
plan.
Signed-off-by: Juraj Linkeš
---
dts/framework/remote_session/__init__.py | 2 +-
dts/framework/remote_session/os_session.py| 16 -
.../remote_session/remote/__init__.py | 2 +-
dts/framework/testb
The module implements the base class that all test suites inherit from.
It implements methods common to all test suites.
The derived test suites implement test cases and any particular setup
needed for the suite or tests.
Signed-off-by: Juraj Linkeš
---
dts/conf.yaml
Setup hugepages on nodes. This is useful not only on SUT nodes, but
also on TG nodes which use TGs that utilize hugepages.
The setup is opt-in, i.e. users need to supply hugepage configuration to
instruct DTS to configure them. It not configured, hugepage
configuration will be skipped. This is hel
Add methods for setting up and shutting down DPDK apps and for
constructing EAL parameters.
Signed-off-by: Juraj Linkeš
---
dts/conf.yaml | 4 +
dts/framework/config/__init__.py | 8 +
dts/framework/config/conf_yaml_schema.json| 25 ++
dts/fr
Add the ability to build DPDK and apps on the SUT, using a configured
target.
Signed-off-by: Juraj Linkeš
---
dts/framework/config/__init__.py | 2 +
dts/framework/exception.py| 17 ++
dts/framework/remote_session/os_session.py| 89 +-
dts/framewo
This is a basic capability needed to check whether the command execution
was successful or not. If not, raise a RemoteCommandExecutionError. When
a failure is expected, the caller is supposed to catch the exception.
Signed-off-by: Juraj Linkeš
---
dts/framework/exception.py|
The abstraction model in DTS is as follows:
Node, defining and implementing methods common to and the base of SUT
(system under test) Node and TG (traffic generator) Node.
Remote Session, defining and implementing methods common to any remote
session implementation, such as SSH Session.
OSSession,
Add code needed to run the HelloWorld testcase which just runs
the helloworld dpdk application.
The patchset currently heavily refactors this original DTS code needed
to run the testcase:
* The whole architecture has been redone into more sensible class
hierarchy
* DPDK build on the System under
Testcases are added for ZUC256 cipher-auth, auth-cipher,
and auth-cipher-verify.
4 byte, 8 byte, and 16 byte tags are tested for each chained type.
Signed-off-by: Ciara Power
Acked-by: Brian Dooley
---
v2: changed from bit to byte length representation
---
app/test/test_cryptodev.c
The cipher-auth test function for ZUC was not using the improved cipher
and auth capability check functions. This meant the required key and IV
lengths were not being checked, leading to problems with ZUC256 tests
running, and failing, on devices that only support ZUC128.
Fixes: 27b787132484 ("tes
The incorrect value was being passed to the ZUC authentication test
function, indicating the opposite of the intended GENERATE/VERIFY op.
This is fixed to use the auth op enum rather than a value.
Fixes: 83397b9f0739 ("test/crypto: add further ZUC test cases")
Signed-off-by: Ciara Power
Acked-by
The digest length used in ZUC tests for verifying the digest was
hardcoded at 4 bytes, which was suitable for ZUC128 only.
Now that ZUC256 is supported by these test functions, the digest length
can vary.
Using the test vector digest length directly in these
comparisons allows for variable digest
There were some ZUC256 tests for hash generate and hash verify mixed in
with the ZUC128 tests, with no indication in the test name that they
were for ZUC256.
These are now grouped with other ZUC256 tests, and all ZUC256 hash test
names are modified to be more uniform, and indicate the digest lengt
This patchset adds chained testcases for each ZUC256 digest size:
- cipher-auth
- auth-cipher
- auth-cipher-verify
To add these tests, some improvements were made to the existing code,
allowing for the user to easily differentiate ZUC256 and ZUC128 tests.
Some fixes are also included, that
> -Original Message-
> From: Jiale, SongX
> Sent: Friday, March 3, 2023 3:50 PM
> To: Deng, KaiwenX ; dev@dpdk.org
> Cc: sta...@dpdk.org; Yang, Qiming ; Zhou,
> YidingX ; Deng, KaiwenX
> ; Wu, Wenjun1 ;
> Zhang, Qi Z ; Lu, Wenzhuo
> Subject: RE: [PATCH] net/ixgbe: fix IPv6 mask in fdir
On Mon, Feb 27, 2023 at 5:08 PM David Marchand
wrote:
>
> Hello,
>
> On Fri, Feb 24, 2023 at 7:04 PM Harry van Haaren
> wrote:
> >
> > On some CI runs, some service-cores tests spuriously fail as the
> > service lcore thread is not actually scheduled by the OS in the
> > given amount of time.
> >
> -Original Message-
> From: Juraj Linkeš
> Sent: Thursday, February 23, 2023 11:29 PM
> To: tho...@monjalon.net; honnappa.nagaraha...@arm.com; Tu, Lijuan
> ; Richardson, Bruce ;
> pr...@iol.unh.edu
> Cc: dev@dpdk.org; Juraj Linkeš
> Subject: [PATCH v5 10/10] doc: update DTS setup and tes
Fix uninitialized hw_reassembly offload variable.
Fixes: d8d51d4f9bf5 ("examples/ipsec-secgw: support per SA HW reassembly")
Cc: rbhans...@marvell.com
Signed-off-by: Nithin Dabilpuram
---
examples/ipsec-secgw/sa.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/examples/ipsec-secgw/sa.c b/e
> -Original Message-
> From: Srikanth Yalavarthi
> Sent: Tuesday, February 7, 2023 9:20 PM
> To: Thomas Monjalon ; Srikanth Yalavarthi
>
> Cc: dev@dpdk.org; Shivah Shankar Shankar Narayan Rao
> ; Jerin Jacob Kollanukkaran
> ; Anup Prabhu ; Prince Takkar
> ; Parijat Shukla
> Subject: [P
From: Veerasenareddy Burru
Configure TL4 to respond to SDP channel backpressure.
Signed-off-by: Veerasenareddy Burru
---
drivers/common/cnxk/roc_nix_tm_utils.c | 4
1 file changed, 4 insertions(+)
diff --git a/drivers/common/cnxk/roc_nix_tm_utils.c
b/drivers/common/cnxk/roc_nix_tm_utils
To avoid compiler reordering stores to LMT line and ldeor,
add clobber attribute to ldeor, steor etc.
Signed-off-by: Nithin Dabilpuram
---
drivers/common/cnxk/roc_io.h | 12
1 file changed, 8 insertions(+), 4 deletions(-)
diff --git a/drivers/common/cnxk/roc_io.h b/drivers/common/c
From: Hiral
Clarifying otx2_process_msgs() RSS error messages
Signed-off-by: Hiral
---
.mailmap| 1 +
drivers/common/cnxk/roc_dev.c | 4 ++--
drivers/common/cnxk/roc_utils.c | 6 ++
3 files changed, 9 insertions(+), 2 deletions(-)
diff --git a/.mailmap b/.mailmap
From: Harman Kalra
Enabling one to one mapping between SQ to 64 QINTs per LF. So in case of
SQ interrupt event NIX can deliver associated QINT MSI-X interrupt to
software and increment respective QINT count CSR.
While for some cn10k chip models keeping the workaround (i.e. all SQs
assigned to sa
From: Rahul Bhansali
Supports creation of individual meta aura per NIX port for
CN106-B0/CN103xx SoC.
Individual pool buffer size can be passed using meta_buf_sz
devargs parameter per NIX for local meta aura creation.
Signed-off-by: Rahul Bhansali
---
doc/guides/nics/cnxk.rst |
From: Rahul Bhansali
Meta aura for RQs is created during queue enable process, so
aura handle for fastpath Rx queues should be updated after
this.
Signed-off-by: Rahul Bhansali
---
drivers/net/cnxk/cn10k_ethdev.c | 33 ++---
1 file changed, 26 insertions(+), 7 delet
From: Rahul Bhansali
Avoid priority flow control configuration when port is
started.
Signed-off-by: Rahul Bhansali
---
drivers/net/cnxk/cnxk_ethdev_ops.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/cnxk/cnxk_ethdev_ops.c
b/drivers/net/cnxk/cnxk_ethdev_o
Check and enable/disable flow control config per queue on
device start to handle cases like SSO enablement, TM changes etc.
Modify flow control config get to get status per RQ/SQ.
Also disallow changes to flow control config when device
is in started state.
Signed-off-by: Nithin Dabilpuram
---
Enable support similar to CN10KB as CN10KA_B0 is similar
to CN10KB.
Signed-off-by: Nithin Dabilpuram
---
drivers/common/cnxk/roc_api.h | 3 +++
drivers/common/cnxk/roc_cpt.h | 2 --
drivers/common/cnxk/roc_features.h | 31 +
drivers/common/cnxk/roc_nix_
Remove flow control default enable/disable from queue setup time
and move it to explicit per queue setup after device is configured
and as part of device start.
Also remove TM node ref count for flow control to avoid
ref count mismatch. For user tree, on user disabling flow control
or PFC on one S
From: Veerasenareddy Burru
map SQ's to SDP channels using round-robin policy.
Signed-off-by: Veerasenareddy Burru
---
drivers/common/cnxk/roc_nix_queue.c | 9 ++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers/common/cnxk/roc_nix_queue.c
b/drivers/common/cnxk/roc_
From: Kommula Shiva Shankar
This patch adds separate inline dev stats api,
thus avoiding expensive nix xstats call
Signed-off-by: Kommula Shiva Shankar
---
drivers/common/cnxk/roc_nix_inl.h | 1 +
drivers/common/cnxk/roc_nix_inl_dev.c | 33 +++
drivers/common/cnxk/
From: Rakesh Kudurumalla
bpids are not disabled while freeing resources for NIX
device as a result a new bpid is assigned which leads
to exhaustion of bpid's after soft exit of application.
This patch fixes the same
Signed-off-by: Rakesh Kudurumalla
---
drivers/net/cnxk/cnxk_ethdev.c | 28
From: Rakesh Kudurumalla
while sending external buffers from application
if tx_compl_ena is not provided as devargs, cq for
transmit completion is not initialized but the same
is accessed in driver resulting in segfault.
This patch fixes the same by calling callback handler
before the actual pack
When coming from vector routine, the first seg length is
set to same as pkt len assuming it is a single segment pkt.
Read just the same in cn10k_nix_prepare_mseg() that is called
in case of mbuf-fast-free offload disabled.
In CN9K, clear other data len fields to avoid using stale data.
Fixes: 8520
83 matches
Mail list logo
