> On Sat, 12 Oct 2024 10:40:59 +0800
> Chaoyong He wrote:
>
> > The original logic keeps using the same name parameter when malloc
> > memory in secondary process, which may cause error when using multiple
> > PF cards.
> >
> > Fixes: 3b00109d2b65 ("net/nfp: add PF ID used to format symbols")
>
On 9/5/2024 8:33 AM, Andrew Rybchenko wrote:
> On 9/5/24 09:46, Jie Hai wrote:
>> From: Chengwen Feng
>>
>> Verify queue_id for rte_eth_tx_done_cleanup API.
>
> If I'm not mistaken the function is considered data path API (fast).
> If so, it should not validate it's parameters as in rte_eth_tx_bu
On Tue, 30 Jul 2024 13:22:35 +0100
Konstantin Ananyev wrote:
> From: Konstantin Ananyev
>
> With commit: ACL mode now can use send_packets_multi().
> What I missed with that changes: send_packets_multi() can't deal
> properly with input dst_port[i] == BAD_PORT (though it can set
> it itself), a
The comment of some control messages are not right, which conflict
with the data structure and may confuse the other developers.
Signed-off-by: Chaoyong He
Reviewed-by: Long Wu
Acked-by: Stephen Hemminger
---
drivers/net/nfp/flower/nfp_flower_cmsg.h | 14 +++---
1 file changed, 7 inser
Fix one memory leak problem in the logic of VF initialization.
Fixes: d81e2b514dc9 ("net/nfp: move device info into process private data")
Cc: sta...@dpdk.org
Signed-off-by: Chaoyong He
Reviewed-by: Long Wu
Reviewed-by: Peng Zhang
Acked-by: Stephen Hemminger
---
drivers/net/nfp/nfp_ethdev_vf
The return value of 'nfp_eth_set_fec()' is three ways, the original
logic considered it as two ways wrongly.
Fixes: 37bd1b843a20 ("net/nfp: support setting FEC mode")
Cc: zerun...@corigine.com
Cc: sta...@dpdk.org
Signed-off-by: Chaoyong He
Reviewed-by: Long Wu
Reviewed-by: Peng Zhang
Acked-by:
On Thu, Oct 10, 2024 at 7:41 PM Stephen Hemminger
wrote:
> > With the addition of SKIP_FILES in
> > commit 7e421ae345f4 ("devtools: support skipping forbid rule check"),
> > it is possible to avoid false positive on adding
> > attribute wrapper in rte_common.h.
> >
> > Signed-off-by: David Marchan
On Thu, Oct 10, 2024 at 1:27 PM Vladimir Medvedkin
wrote:
>
> Currently, for DIR24-8 algorithm, the tbl8 group is freed even though the
> readers might be using the tbl8 group entries. The freed tbl8 group can
> be reallocated quickly. As a result, lookup may be performed incorrectly.
>
> To addre
11/10/2024 10:04, Mattias Rönnblom:
> On 2024-10-10 23:24, Thomas Monjalon wrote:
> > Hello,
> >
> > This new feature looks to bring something interesting to DPDK.
> > There was a good amount of discussion and review,
> > and there is a real effort of documentation.
> >
> > However, some choices
> -Original Message-
> From: Ferruh Yigit
> Sent: Friday, October 11, 2024 02:03
> To: Konstantin Ananyev ; Dariusz Sosnowski
> ; NBU-Contact-Thomas Monjalon (EXTERNAL)
> ; Andrew Rybchenko
> Cc: dev@dpdk.org; Bruce Richardson
> Subject: Re: [RFC 0/4] ethdev: rework config restore
>
>
On Thu, Oct 10, 2024 at 1:56 PM Mattias Rönnblom wrote:
>
> On 2024-10-10 12:45, David Marchand wrote:
> > On Fri, Sep 20, 2024 at 12:57 PM Mattias Rönnblom
> > wrote:
> >> + static int \
> >> + run_parallel_test_and_modify ## size(
> >> +/**
> >> + * Get pointer to lcore variable instance of the current thread.
> >> + *
> >> + * May only be used by EAL threads and registered non-EAL threads.
> >> + */
> >> +#define RTE_LCORE_VAR_VALUE(handle) \
> >
> > RTE_LCORE_VAR_LOCAL?
> >
>
> Why is that better?
>
> Maybe Morten can re
Introduce DPDK per-lcore id variables, or lcore variables for short.
An lcore variable has one value for every current and future lcore
id-equipped thread.
The primary use case is for statically allocating
small, frequently-accessed data structures, for which one instance
should exist for each l
Add functional test suite to exercise the API.
Signed-off-by: Mattias Rönnblom
Acked-by: Morten Brørup
Acked-by: Chengwen Feng
Acked-by: Stephen Hemminger
--
PATCH v6:
* Update FOREACH invocations to match new API.
RFC v5:
* Adapt tests to reflect the removal of the GET() and SET() macro
Replace keeping PRNG state in a RTE_MAX_LCORE-sized static array of
cache-aligned and RTE_CACHE_GUARDed struct instances with keeping the
same state in a more cache-friendly lcore variable.
Signed-off-by: Mattias Rönnblom
Acked-by: Morten Brørup
Acked-by: Konstantin Ananyev
Acked-by: Chengwen F
Add basic micro benchmark for lcore variables, in an attempt to assure
that the overhead isn't significantly greater than alternative
approaches, in scenarios where the benefits aren't expected to show up
(i.e., when plenty of cache is available compared to the working set
size of the per-lcore dat
Replace static array of cache-aligned structs with an lcore variable,
to slightly benefit code simplicity and performance.
Signed-off-by: Mattias Rönnblom
Acked-by: Morten Brørup
Acked-by: Konstantin Ananyev
Acked-by: Chengwen Feng
Acked-by: Stephen Hemminger
--
PATCH v6:
* Update FOREACH
Keep per-lcore power intrinsics state in a lcore variable to reduce
cache working set size and avoid any CPU next-line-prefetching causing
false sharing.
Signed-off-by: Mattias Rönnblom
Acked-by: Morten Brørup
Acked-by: Konstantin Ananyev
Acked-by: Chengwen Feng
Acked-by: Stephen Hemminger
--
This patch set introduces a new API for static
per-lcore id data allocation.
Please refer to the API documentation for both a
rationale for this new API, and a comparison to the alternatives
available.
The adoption of this API would affect many different DPDK modules, but
the author updated onl
>
> External email: Use caution opening links or attachments
>
>
> On 10/10/2024 1:08 PM, Dariusz Sosnowski wrote:
> >> -Original Message-
> >> From: Ferruh Yigit
> >> Sent: Thursday, October 10, 2024 01:17
> >> To: Dariusz Sosnowski ; Konstantin An
Replace static array of cache-aligned structs with an lcore variable,
to slightly benefit code simplicity and performance.
Signed-off-by: Mattias Rönnblom
Acked-by: Morten Brørup
Acked-by: Konstantin Ananyev
Acked-by: Chengwen Feng
Acked-by: Stephen Hemminger
--
PATCH v7:
* Update to match
When using checkpatches.sh locally, verify that there is an In-Reply-To
header when the patch is a respin (i.e. v2, v3, etc.). This is currently
only enforced by the upstream CI but cannot be verified locally.
This cannot be verified when checking commit ids since --in-reply-to is
a git-format-pat
11/10/2024 10:09, Morten Brørup:
> > > +static void *
> > > +lcore_var_alloc(size_t size, size_t align)
> > > +{
> > > + void *handle;
> > > + unsigned int lcore_id;
> > > + void *value;
> > > +
> > > + offset = RTE_ALIGN_CEIL(offset, align);
> > > +
> > > + if (offset + size > RTE_MAX_LCORE_VAR) {
On Thu, Oct 3, 2024 at 5:50 PM Van Haaren, Harry
wrote:
> > From: David Marchand
> > Sent: Thursday, October 3, 2024 10:13 AM
> > To: Mattias Rönnblom ; Van Haaren, Harry
> >
> > Cc: dev@dpdk.org ; step...@networkplumber.org
> > ; suanmi...@nvidia.com ;
> > tho...@monjalon.net ; sta...@dpdk.o
From: Vamsi Attunuru
Some DMA controllers offer the ability to configure priority
level for the DMA channels, allowing for the prioritization
of DMA command execution based on channel importance.
This patch supports such strict priority configuration. If the
dmadev supports, it should advertise
Before this patch, ethdev layer assumed that all drivers require that
it has to forcefully restore:
- MAC addresses
- promiscuous mode setting
- all multicast mode setting
upon rte_eth_dev_start().
This patch introduces a new callback to eth_dev_ops -
get_restore_flags().
Drivers implementing th
Use get_restore_flags() internal API introduced in previous commits
in rte_eth_dev_start(), to restore only the configuration
requested by the driver.
Signed-off-by: Dariusz Sosnowski
---
lib/ethdev/rte_ethdev.c | 31 +--
1 file changed, 21 insertions(+), 10 deletions
mlx5 PMD does not require configuration restore
on rte_eth_dev_start().
Add implementation of get_restore_flags() indicating that.
Signed-off-by: Dariusz Sosnowski
---
drivers/net/mlx5/mlx5.c| 2 ++
drivers/net/mlx5/mlx5.h| 3 +++
drivers/net/mlx5/mlx5_ethdev.c | 19 +++
Extract promiscuous and all multicast configuration restore
to separate functions.
This change will allow easier integration of disabling
these procedures for supporting PMDs in follow up commits.
Signed-off-by: Dariusz Sosnowski
---
lib/ethdev/rte_ethdev.c | 34 +
This patch series reworks the config restore procedure,
so that drivers are able to enable/disable certain parts of it.
Drivers can provide get_restore_flags() callback,
which will indicate to ethdev library what configuration to restore.
If callback is not defined, then ethdev assumes that
all co
On Sun, Sep 8, 2024 at 10:43 AM Doug Foster wrote:
>
> This simplifies the implementation of rte_rcu_qsbr_thread_register()
> and rte_rcu_thread_unregister() functions. The simplified implementation
> is easier to read.
>
> Signed-off-by: Doug Foster
> Reviewed-by: Honnappa Nagarahalli
> Reviewe
On 10/11/2024 9:02 AM, Bruce Richardson wrote:
> On Fri, Oct 11, 2024 at 06:34:06AM +, Mingjin Ye wrote:
>> Add ice support for new ethdev API to adjust frequency for IEEE1588
>> PTP. Also, this patch reworks code for converting software update
>> to hardware update.
>>
>> Signed-off-by: Simei
On 10/11/2024 7:34 AM, Mingjin Ye wrote:
> [1/3] ethdev: add frequency adjustment API
> [2/3] net/ice: add frequency adjustment support for PTP
> [3/3] examples/ptpclient: add frequency adjustment
> ---
> v2: rte_eth_timesync_adjust_freq marked as experimental.
> ---
> v3: Add more description for
On 10/11/2024 7:34 AM, Mingjin Ye wrote:
> This patch adds freq adjustment API for PTP high accuracy.
>
> Signed-off-by: Simei Su
> Signed-off-by: Mingjin Ye
>
Reviewed-by: Ferruh Yigit
Applied to dpdk-next-net/main, thanks.
On 10/10/2024 8:45 AM, Chaoyong He wrote:
> Enhance the NFP service framework when using the flower firmware, we can
> using the alarm now when there is no service core.
>
> Long Wu (2):
> net/nfp: rename flower service flag
> net/nfp: enhance the flower service framework
>
Series applied to
On 10/11/2024 6:29 PM, Stephen Hemminger wrote:
> Some patches related to recent increase in possible queues.
>
> Stephen Hemminger (3):
> net/tap: handle increase in mp_max_fds
> net/tap: add static assert to make sure max queues less than fd limit
> net/tap: increase the maximum allowable
On 10/9/2024 11:32 AM, Bruce Richardson wrote:
> Add function to allow querying a node in the scheduler tree. Returns
> the parameters as were given to the add function. Adding this function
> allows apps to just query the hierarchy rather than having to maintain
> their own copies of it internall
On 10/9/2024 11:32 AM, Bruce Richardson wrote:
> Add support for an ethdev TM query function to allow apps to get details
> of the TM nodes they previously configured. This patchset includes:
>
> * ethdev changes to add the API
> * implementation of the function in "ice" pmd
> * testpmd command to
On 2024/10/11 22:13, David Marchand wrote:
> On Wed, Oct 9, 2024 at 6:50 AM Chengwen Feng wrote:
>>
>> The rte_kvargs_process() was used to parse key-value (e.g. socket_id=0),
>> it also supports to parse only-key (e.g. socket_id). But many drivers's
>> callback can only handle key-value, it will
On Fri, Oct 11, 2024 at 06:34:06AM +, Mingjin Ye wrote:
> Add ice support for new ethdev API to adjust frequency for IEEE1588
> PTP. Also, this patch reworks code for converting software update
> to hardware update.
>
> Signed-off-by: Simei Su
> Signed-off-by: Mingjin Ye
> ---
> doc/guides/
On 2024-10-10 23:24, Thomas Monjalon wrote:
Hello,
This new feature looks to bring something interesting to DPDK.
There was a good amount of discussion and review,
and there is a real effort of documentation.
However, some choices done in this implementation
were not explained or advertised eno
Mattias,
Please note that most of Thomas' questions are in the interest of the general
public, considered requests for further documentation.
> Do you have benchmarks results of the modules using such variables
> (power, random, service)?
> It would be interesting to compare time efficiency and
On 10/10/2024 3:13 PM, Serhii Iliushyk wrote:
> The list of updates provided by the patchset:
> * Update the supported version of the FPGA to 9563.55.49
> * Fix Coverity issues
> * Fix issues related to release 24.07
> * Extended and fixed the implementation of the logging
> * Added NT fl
On 10/10/2024 7:42 AM, Chaoyong He wrote:
> This patch series aims to add two commands for the testpmd application:
> - testpmd> set port eeprom magic \
> value offset
> - testpmd> set port led
>
> ---
> v5:
> * Use 'accept_risk' rather than 'confirm' to make it clear enough.
> *
Hi Thomas and Ferruh,
Kindly ping for merge.
/Huisong
在 2024/9/12 11:07, fengchengwen 写道:
Series-reviewed-by: Chengwen Feng
Thanks chengwen.
On 2024/9/12 10:38, Huisong Li wrote:
The deeper the idle state, the lower the power consumption, but the longer
the resume time. Some service are de
Now that max MP fds has increased to 253 it is possible that
the number of queues the TAP device can handle is less than that.
Therefore the code to handle MP message should only allow the
number of queues it can handle.
Coverity issue: 445386
Signed-off-by: Stephen Hemminger
---
drivers/net/tap
The ethdev layer already validates that the number of requested
queues is less than the reported max queues.
Signed-off-by: Stephen Hemminger
---
drivers/net/tap/rte_eth_tap.c | 16
1 file changed, 16 deletions(-)
diff --git a/drivers/net/tap/rte_eth_tap.c b/drivers/net/tap/rte
The default of 16 is too low for larger systems.
Upper bound is the maximum number of fd's that can be passed to
secondary process (253) and the maximum number of queue's allowed
by Linux kernel (256). Both these manifest constants, are not
exposed in visible kernel API.
Signed-off-by: Stephen He
Some patches related to recent queue limit changes.
v2 - up the limit to maximum Linux can support
dont use static_assert here
get rid of unreachable checks in configure
Stephen Hemminger (3):
net/tap: handle increase in mp_max_fds
net/tap: increase the maximum allowable queues
ne
On Sat, 12 Oct 2024 02:01:12 +0100
Ferruh Yigit wrote:
> On 10/11/2024 6:29 PM, Stephen Hemminger wrote:
> > Some patches related to recent increase in possible queues.
> >
> > Stephen Hemminger (3):
> > net/tap: handle increase in mp_max_fds
> > net/tap: add static assert to make sure max q
On 2024-10-10 10:30, David Marchand wrote:
From: Mattias Rönnblom
Introduce a set of functions and macros that operate on sets of bits,
kept in arrays of 64-bit words.
RTE bitset is designed for bitsets which are larger than what fits in
a single machine word (i.e., 64 bits).
Tiny detail: th
, Aug 16, 2024 at 17:09:
From: Pavan Nikhilesh
Add node fastpath error counters advertised during
node registration.
Signed-off-by: Pavan Nikhilesh
Reviewed-by: Robin Jarry
Hi Pavan,
, Aug 16, 2024 at 17:09:
From: Pavan Nikhilesh
Add error counters for ip4 LPM lookup failures in
ip4_lookup node.
Signed-off-by: Pavan Nikhilesh
[snip]
diff --git a/lib/node/node_private.h b/lib/node/node_private.h
index 1de7306792..36b2a733db 100644
--- a/lib/node/node_private
On Thu, Oct 10, 2024 at 1:26 PM Vladimir Medvedkin
wrote:
> diff --git a/lib/fib/meson.build b/lib/fib/meson.build
> index 6795f41a0a..8c03496cdc 100644
> --- a/lib/fib/meson.build
> +++ b/lib/fib/meson.build
> @@ -25,40 +25,28 @@ if dpdk_conf.has('RTE_ARCH_X86_64') and binutils_ok
> # linked
> Rather than re-querying the HW each time a CPU flag is requested, we can
> just save the return value in the flags array. This should speed up
> repeated querying of CPU flags, and provides a workaround for a reported
> issue where errors are seen with constant querying of the AVX-512 CPU
> fl
On Sat, Oct 5, 2024 at 6:14 PM Stephen Hemminger
wrote:
> On Fri, 13 Sep 2024 16:06:00 +0300
> Oleksandr Nahnybida wrote:
>
> > Adjust test to check if pcapng works with chained mbufs
> >
> > Signed-off-by: Oleksandr Nahnybida
> Acked-by: Stephen Hemminger
Series applied, thanks Oleksandr.
-
On Fri, Oct 4, 2024 at 12:11 AM Stephen Hemminger
wrote:
>
> If dumpcap (in legacy pcap mode) tried to handle a large segmented
> frame it would core dump because rte_pktmbuf_read() would return NULL.
> Fix by using same logic as in pcap PMD.
>
> Fixes: cbb44143be74 ("app/dumpcap: add new packet c
On Fri, Oct 11, 2024 at 12:42:01PM +, Konstantin Ananyev wrote:
>
>
> > Rather than re-querying the HW each time a CPU flag is requested, we can
> > just save the return value in the flags array. This should speed up
> > repeated querying of CPU flags, and provides a workaround for a reported
On Tue, Sep 17, 2024 at 10:10 AM Morten Brørup
wrote:
>
> Bursts of up to 64, 128 and 256 packets are not uncommon, so increase the
> maximum tested get and put burst sizes from 32 to 256.
> For convenience, also test get and put burst sizes of
> RTE_MEMPOOL_CACHE_MAX_SIZE.
>
> Some applications
> Before this patch, ethdev layer assumed that all drivers require that
> it has to forcefully restore:
>
> - MAC addresses
> - promiscuous mode setting
> - all multicast mode setting
>
> upon rte_eth_dev_start().
>
> This patch introduces a new callback to eth_dev_ops -
> get_restore_flags()
On Fri, Oct 11, 2024 at 1:29 PM David Marchand
wrote:
> > @@ -214,6 +215,7 @@ rte_fib_create(const char *name, int socket_id, struct
> > rte_fib_conf *conf)
> > rte_strlcpy(fib->name, name, sizeof(fib->name));
> > fib->rib = rib;
> > fib->type = conf->type;
> > + fib
On Wed, Jul 5, 2023 at 1:18 PM Mattias Rönnblom
wrote:
>
> Announce the removal of the single-event enqueue and dequeue
> operations from the eventdev ABI.
>
> Signed-off-by: Mattias Rönnblom
>
> ---
> PATCH v2: Fix commit subject prefix.
> ---
> doc/guides/rel_notes/deprecation.rst | 8
07/08/2024 15:43, Bruce Richardson:
> On Thu, Aug 01, 2024 at 12:01:54PM +0200, Robin Jarry wrote:
> > Fix the following error when using dpdk as a subproject:
> >
> > subprojects/dpdk/buildtools/subproject/meson.build:28:56:
> > ERROR: Unknown function "file".
> >
> > This was obviously never
On Wed, Jul 19, 2023 at 5:13 PM Cristian Dumitrescu
wrote:
>
> Postpone the deprecation of the legacy pipeline, table and port
> library API and gradual stabilization of the new API.
>
> Signed-off-by: Cristian Dumitrescu
> ---
> doc/guides/rel_notes/deprecation.rst | 21 +
>
On Wed, Aug 28, 2024 at 3:42 PM Robin Jarry wrote:
>
> Change the color of arrows leading to sink nodes to dark orange. Remove
> the node oval shape around the sink nodes and make their text dark
> orange. This results in a much more readable output for large graphs.
> See the link below for an ex
On Mon, Sep 30, 2024 at 11:28 AM Robin Jarry wrote:
>
> In some cases, the node context data is used to store two pointers
> because the data is larger than the reserved 16 bytes. Having to define
> intermediate structures just to be able to cast is tedious. And without
> intermediate structures,
On Thu, Oct 10, 2024 at 11:51 AM David Marchand
wrote:
>
> On Thu, Sep 5, 2024 at 6:22 PM Stephen Hemminger
> wrote:
> >
> > The TAP and XDP driver both are limited to only 8 queues when
> > because of the small limit imposed by EAL. Increase the limit
> > now since this release allows changing A
On Fri, 11 Oct 2024 14:37:35 +0200
Morten Brørup wrote:
> > From: Robin Jarry [mailto:rja...@redhat.com]
> > Sent: Thursday, 10 October 2024 22.08
> >
> > Morten Brørup, Oct 06, 2024 at 10:18:
> > > This has been discussed before, but I want to double check...
> > >
> > > If - sometime in the
https://bugs.dpdk.org/show_bug.cgi?id=1563
Bug ID: 1563
Summary: Many I219 variants not supported by E1000 druver
Product: DPDK
Version: 24.11
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
> -Original Message-
> From: Robin Jarry
> Sent: Friday, October 11, 2024 3:24 PM
> To: Pavan Nikhilesh Bhagavatula ; Jerin Jacob
> ; Nithin Kumar Dabilpuram
> ; Kiran Kumar Kokkilagadda
> ; zhirun@intel.com; Zhirun Yan
>
> Cc: dev@dpdk.org
> Subject: [EXTERNAL] Re: [PATCH v4 3/5]
A number of small fixes and other changes to enable Tx scheduler
enhancements to our DPDK driver have been added to the base code.
Upstream these changes for 24.11 RC2. Most have previously been
submitted as part of the scheduler changes [1]
[1] https://patches.dpdk.org/project/dpdk/list/?series=3
The Tx scheduler in the ice driver can be configured to have large
numbers of child nodes at a given layer, but the driver code implicitly
limited the number of nodes to 255 by using a u8 datatype for the number
of children. Increase this to a 16-bit value throughout the code.
Signed-off-by: Bruce
From: Jacob Keller
When removing bypass mode, the code for E822 bypass was completely
removed in error. This code should be maintained in DPDK so re-add the
necessary functions.
Fixes: ce9ad8c5bc6d ("net/ice/base: remove PHY port timer bypass mode")
Cc: sta...@dpdk.org
Signed-off-by: Jacob Kell
From: Fabio Pricoco
Refactor while loop to add a check that the values read are in the
correct range.
Fixes: 6c1f26be50a2 ("net/ice/base: add control queue information")
Cc: sta...@dpdk.org
Signed-off-by: Fabio Pricoco
Signed-off-by: Bruce Richardson
---
drivers/net/ice/base/ice_controlq.c |
From: Dave Ertman
If there is more than one VLAN defined when any reset that affects the
PF is initiated, after the reset rebuild, no traffic will pass on any
VLAN but the last one created.
This is caused by the iteration though the VLANs during replay each
clearing the vsi_map bitmap of the VSI
The ice_sched_node type has got a field for the vsi to which the node
belongs. This field was not getting set in "ice_sched_add_node", so add
a line configuring this field for each node from its parent node.
Similarly, when searching for a qgroup node, we can check for each node
that the VSI inform
In a number of places throughout the driver code, we want to confirm
that a scheduler node is indeed a child of another node. Currently, this
is confirmed by searching down the tree from the base until the desired
node is hit, a search which may hit many irrelevant tree nodes when
recursing down wr
DPDK should support more than just 9-level or 5-level topologies, so
remove the checks for those particular settings.
Signed-off-by: Bruce Richardson
---
drivers/net/ice/base/ice_ddp.c | 33 -
1 file changed, 33 deletions(-)
diff --git a/drivers/net/ice/base/ice_
If DCB support is disabled via DDP image, there will not be any traffic
class (TC) nodes in the scheduler tree immediately above the root level.
To allow the driver to work with this scenario, we allow use of the root
node as a dummy TC0 node in case where there are no TC nodes in the
tree. For use
Rather than computing from the number of HW layers the layer of the VSI,
we can instead just read that info from the VSI node itself. This allows
the layer to be changed at runtime.
Signed-off-by: Bruce Richardson
---
drivers/net/ice/base/ice_sched.c | 2 +-
1 file changed, 1 insertion(+), 1 del
> On Fri, Aug 16, 2024 at 5:10 PM wrote:
> >
> > From: Pavan Nikhilesh
> >
> > Add ability for Nodes to advertise error counters
> > during registration.
> >
> > Signed-off-by: Pavan Nikhilesh
>
> Such a series deserve a cover letter.
>
> It also deserves a RN update, and I see no removal of t
Before this patch, ethdev layer assumed that all drivers require that
it has to forcefully restore:
- MAC addresses
- promiscuous mode setting
- all multicast mode setting
upon rte_eth_dev_start().
This patch introduces a new callback to eth_dev_ops -
get_restore_flags().
Drivers implementing th
Use get_restore_flags() internal API introduced in previous commits
in rte_eth_dev_start(), to restore only the configuration
requested by the driver.
Signed-off-by: Dariusz Sosnowski
---
lib/ethdev/rte_ethdev.c | 31 +--
1 file changed, 21 insertions(+), 10 deletions
This patch series reworks the config restore procedure,
so that drivers are able to enable/disable certain parts of it.
Drivers can provide get_restore_flags() callback,
which will indicate to ethdev library what configuration to restore.
If callback is not defined, then ethdev assumes that
all co
mlx5 PMD does not require configuration restore
on rte_eth_dev_start().
Add implementation of get_restore_flags() indicating that.
Signed-off-by: Dariusz Sosnowski
---
drivers/net/mlx5/mlx5.c| 2 ++
drivers/net/mlx5/mlx5.h| 2 ++
drivers/net/mlx5/mlx5_ethdev.c | 18
Extract promiscuous and all multicast configuration restore
to separate functions.
This change will allow easier integration of disabling
these procedures for supporting PMDs in follow up commits.
Signed-off-by: Dariusz Sosnowski
---
lib/ethdev/rte_ethdev.c | 34 +
On 10/11/2024 7:34 AM, Mingjin Ye wrote:
> This patch adds PI servo controller to support frequency
> adjustment API for IEEE1588 PTP.
>
> For example, the command for starting ptpclient with PI controller is:
> dpdk-ptpclient -a :81:00.0 -c 1 -n 3 -- -T 0 -p 0x1 -c 1
>
> Signed-off-by: Simei
On 10/11/2024 2:38 AM, Stephen Hemminger wrote:
> Change the documentation to match current usage of this feature
> in the NIC table. Moved this sub heading to be after basic
> stats because the queue stats reported now are in the same structure.
>
> Although the "Stats per Queue" feature was orig
On 10/11/2024 7:51 PM, Dariusz Sosnowski wrote:
> This patch series reworks the config restore procedure,
> so that drivers are able to enable/disable certain parts of it.
> Drivers can provide get_restore_flags() callback,
> which will indicate to ethdev library what configuration to restore.
>
>
On Thu, 2024-10-10 at 19:17 +0200, David Marchand wrote:
> On Thu, Oct 10, 2024 at 7:16 PM Thomas Monjalon wrote:
> >
> > With the annotations added to the allocation functions,
> > more issues are detected at compilation time:
> >
> > nfb_rx.c:133:28: error: pointer 'rxq' used after 'rte_free'
> Add support for programming PMU counters and reading their values
> in runtime bypassing kernel completely.
>
> This is especially useful in cases where CPU cores are isolated
> i.e run dedicated tasks. In such cases one cannot use standard
> perf utility without sacrificing latency and perfo
> From: Robin Jarry [mailto:rja...@redhat.com]
> Sent: Thursday, 10 October 2024 22.01
>
> Hi Morten,
>
> Morten Brørup, Oct 06, 2024 at 11:02:
> > Personally, I would prefer following the convention of rte_ether
> functions to return boolean (as int)...
> >
> > static inline int
> rte_is_ signed
In ICE PMD, previously the ready bitmap checking before reading
PHY timestamp was not present. This caused incorrect Tx
timestamping.
The ready bitmap checking is enabled and PHY timestamp is read once
the ready bitmap gives positive value.
Fixes: 881169950d80 ("net/ice/base: implement initial PT
> >
> > > > > > > 1. rte_ring_generic_pvt.h:
> > > > > > > =
> > > > > > >
> > > > > > > pseudo-c-code //
> > > > > > > related armv8 instructions
> > > > > > >
> >
On Wed, Oct 9, 2024 at 6:50 AM Chengwen Feng wrote:
>
> The rte_kvargs_process() was used to parse key-value (e.g. socket_id=0),
> it also supports to parse only-key (e.g. socket_id). But many drivers's
> callback can only handle key-value, it will segment fault if handles
> only-key. so the patch
Acked-by: Chengwen Feng
On 2024/10/11 17:13, Vamsi Krishna wrote:
> From: Vamsi Attunuru
>
> Some DMA controllers offer the ability to configure priority
> level for the DMA channels, allowing for the prioritization
> of DMA command execution based on channel importance.
>
> This patch support
Hi Pavan,
, Aug 16, 2024 at 17:09:
From: Pavan Nikhilesh
Add ability for Nodes to advertise error counters
during registration.
Signed-off-by: Pavan Nikhilesh
---
v2 Changes:
- Fix compilation.
v3 Changes:
- Resend as 1/5 didn't make it through.
v4 Changes:
- Address review comments.
- Rebas
Hi Vladimir,
Vladimir Medvedkin, Oct 10, 2024 at 13:26:
Previously when running rte_fib_lookup IPv4 addresses must have been in
host byte order.
This patch adds a new flag RTE_FIB_FLAG_LOOKUP_BE that can be passed on
fib create, which will allow to have IPv4 in network byte order on
lookup.
Si
On Fri, 11 Oct 2024 11:56:04 +
Konstantin Ananyev wrote:
> > + *
> > + * @return
> > + * 0 in case of success, negative value otherwise.
> > + */
> > +__rte_experimental
> > +int
> > +rte_pmu_init(void);
> > +
> > +/**
> > + * @warning
> > + * @b EXPERIMENTAL: this API may change without pr
On Fri, 11 Oct 2024 10:18:54 +0200
Mattias Rönnblom wrote:
> This patch set introduces a new API for static
> per-lcore id data allocation.
>
> Please refer to the API documentation for both a
> rationale for this new API, and a comparison to the alternatives
> available.
>
> The adoption of
1 - 100 of 167 matches
Mail list logo
