Add support to provide link type as part of link status
if available from ethdev driver. Link type such as Fibre,
Twisted-Pair etc is generally available information in PMD or
lower layer software and is a useful information for DPDK
applications. Similar info is also available via ethtool
for Linu
> Offset should be aligned first before checking if there is free space for
> another
> write.
>
> Bugzilla ID: 1665
> Fixes: ebaee6409702 ("trace: simplify trace point headers")
>
> Signed-off-by: Oleksandr Nahnybida
> ---
> lib/eal/include/rte_trace_point.h | 4 +---
> 1 file changed, 1 inse
> -Original Message-
> From: Tanzeel Ahmed
> Sent: Saturday, 29 March 2025 22:54
> To: Ori Kam
> Cc: dev@dpdk.org; Tanzeel Ahmed
> Subject: [PATCH] examples/flow_filtering: fix make clean
>
> External email: Use caution opening links or attachments
>
>
> make clean is unable to dele
Provide link type along with link status get.
Signed-off-by: Nithin Dabilpuram
---
drivers/net/cnxk/cnxk_ethdev.c | 21 -
drivers/net/cnxk/cnxk_ethdev.h | 1 +
drivers/net/cnxk/cnxk_link.c | 10 +++---
3 files changed, 28 insertions(+), 4 deletions(-)
diff --git a/dri
Speed capability currently is updating as
all speeds capable. Get it from CGX fw data
supported link modes instead.
Signed-off-by: Nithin Dabilpuram
---
drivers/common/cnxk/hw/nix.h | 54 ++
drivers/common/cnxk/roc_mbox.h | 24 +++---
drivers/common/cnxk/roc_nix
On Wed, Apr 2, 2025 at 11:35 AM Oleksandr Nahnybida
wrote:
>
> Offset should be aligned first before checking if there is free space for
> another write.
>
> Bugzilla ID: 1665
> Fixes: ebaee6409702 ("trace: simplify trace point headers")
This commit ebaee6409702 ("trace: simplify trace point head
for DPAA/FSLMC changes
Acked-by: Hemant Agrawal
On 03-04-2025 22:28, David Marchand wrote:
Declaring inline helpers in version.map is unnecessary.
There is no exported symbol and this will probably be treated as an
error by MSVC linker.
rte_dpaa2_dev_type is an enum token.
roc_se_ctx_swap has no implementation (leftover from a previous rewo
https://bugs.dpdk.org/show_bug.cgi?id=1690
Bug ID: 1690
Summary: Calling rte_eth_dev_get_module_info() and
rte_eth_dev_get_module_eeprom() for MLX ConnectX-6
spawn errors when transceiver not inserted
Product: DPDK
So far, each DPDK library (or driver) exposing symbols in an ABI had to
maintain a version.map and use some macros for symbol versioning,
specially crafted with the GNU linker in mind.
This series proposes to rework the whole principle, and instead rely on
marking the symbol exports in the source
Declaring inline helpers in version.map is unnecessary.
There is no exported symbol and this will probably be treated as an
error by MSVC linker.
eal_log_journal has no implementation and can be removed.
Signed-off-by: David Marchand
---
lib/eventdev/version.map | 12
lib/graph/ver
Rather than maintain a file in parallel of the code, symbols to be
exported can be marked with a token RTE_EXPORT_*SYMBOL.
>From those marks, the build framework generates map files only for
symbols actually compiled (which means that the WINDOWS_NO_EXPORT hack
becomes unnecessary).
The build fra
Hi Stephen,
I was testing Grout on one of our baremetal servers today while DTS was
running at the same time (this was an accident - I thought I had it marked
offline in our CI system). Unfortunately as a result DTS conflicted with
Grout which was using the same mellanox card, and a false fail was
Switch to dynamically generated version maps.
As the map files get generated, tooling around checking, converting,
updating etc.. static version maps can be removed.
Signed-off-by: David Marchand
Acked-by: Aaron Conole
---
.github/workflows/build.yml | 1 -
MAINTAINERS
Display the version when a symbol was introduced.
This is needed for scripting the conversion from static to dynamically
generated version maps.
It is also useful when listing experimental symbols.
Signed-off-by: David Marchand
---
buildtools/map-list-symbol.sh | 2 +-
1 file changed, 1 insertio
For versioning symbols:
- MSVC uses pragmas on the symbol,
- GNU linker uses special asm directives,
To accommodate both GNU linker and MSVC linker, introduce new macros for
exporting and versioning symbols that will surround the whole function.
This has the advantage of hiding all the ugly detai
On Wed, Apr 2, 2025 at 10:06 AM Thomas Monjalon wrote:
>
> 28/03/2025 11:52, David Marchand:
> > --- /dev/null
> > +++ b/lib/eal/common/eal_symbol_exports.h
> > @@ -0,0 +1,16 @@
> > +/* SPDX-License-Identifier: BSD-3-Clause
> > + * Copyright (c) 2025 Red Hat, Inc.
> > + */
> > +
> > +#ifndef EAL_S
With gcc, the macro __rte_used translates to __attribute__((used)).
MSVC has something to the same effect, but harder to use and with some
limitations (one being that it cannot be used with "static"). Therefore,
it makes sense to avoid __rte_used in some cases.
The functions modified in this patch
On 2025/4/2 22:45, Morten Brørup wrote:
>> From: Dengdui Huang [mailto:huangdeng...@huawei.com]
>> Sent: Wednesday, 2 April 2025 14.25
>>
>> The communication pipe may be unavailable, for example,
>> when the lcore role is ROLE_OFF or ROLE_NON_EAL.
>> So check whether the pipe is available before
This patch defines rte_node specific dynamic field structure
(rte_node_mbuf_dynfield_t)
rte_node_mbuf_dynfield_t structure holds two types of fields
- Persistent data fields which are preserved across graph walk.
Currently size of persistent data fields is zero.
- Overloadable data fields which
- Used global node mbuf in ip[4|6]_lookup/rewrite nodes
- Redefine node_mbuf_priv1() to rte_node_mbuf_overload_fields_get()
Signed-off-by: Nitin Saxena
---
lib/node/ip4_lookup.c | 14 ---
lib/node/ip4_rewrite.c| 15 +---
lib/node/ip6_lookup.c |
在 2025/4/2 16:47, Jie Hai 写道:
I am moving on to other things and dengdui is going to
take over the role of hns3 maintainer. Update the
MAINTAINERS accordingly.
Signed-off-by: Jie Hai
Thanks for your contributions to hns3.
Acked-by: Huisong Li
---
MAINTAINERS | 2 +-
1 file changed, 1 i
Reviewed-by: Luca Vizzarro
I'm happy with the change. The original intention of the comment was to
make sure that non-involved ports were still being allowed, and that
there were no side effects. The comment change somewhat becomes
redundant as it's repeating the subsequent message. So if it
> -Original Message-
> From: luca.bocca...@gmail.com
> Sent: Wednesday, March 26, 2025 7:51 PM
> To: sta...@dpdk.org
> Cc: dev@dpdk.org; Ali Alnubani ; David Christensen
> ; Hemant Agrawal ;
> Stokes, Ian ; Jerin Jacob ;
> Mcnamara, John ; Kevin Traynor
> ; Luca Boccassi ; Pei Zhang
> ; Ra
Reviewed-by: Luca Vizzarro
Hi Patrick,
this looks good, easy and quick fix. I am afraid that you can't append
to an `Iterable` type though. I've also confirmed this with mypy:
Unsupported left operand type for + ("Iterable[TestSuiteConfig]")
I guess the easiest fix is to just change the argument type to `list`
from
Thanks Luca for the info - there is also 1 other fix needed. I'll send a v2.
On Thu, Apr 3, 2025 at 9:45 AM Luca Vizzarro wrote:
> Hi Patrick,
>
> this looks good, easy and quick fix. I am afraid that you can't append
> to an `Iterable` type though. I've also confirmed this with mypy:
>
>Uns
Hello,
while benchmarking the Nvidia ConnectX-7 400Gbe NIC, I encountered some
interesting behavior: when processing short packets (72 B), the maximum
packet rate I can achieve is 230 Mpps. However, when I increase the
packet size to 512 B briefly and then bring the packet size down to 72 B
a
Align the Tx queue struct field names in idpf and
cpfl driver with the common Tx queue struct.
Signed-off-by: Shaiq Wani
---
drivers/net/intel/cpfl/cpfl_rxtx.c| 40 -
drivers/net/intel/cpfl/cpfl_rxtx_vec_common.h | 4 +-
drivers/net/intel/idpf/idpf_common_rxtx.c | 72 +++
Used the common Tx entry structure and common Tx mbuf ring replenish fn
in place of idpf-specific structure and function.
The vector driver code paths (AVX2, AVX512) use the smaller SW
ring structure.
Signed-off-by: Shaiq Wani
---
drivers/net/intel/cpfl/cpfl_rxtx.c| 2 +-
drivers/ne
Switch the idpf driver to use the common Tx free function for
AVX2 and AVX512
Signed-off-by: Shaiq Wani
---
.../net/intel/idpf/idpf_common_rxtx_avx2.c| 61 +
.../net/intel/idpf/idpf_common_rxtx_avx512.c | 230 +-
drivers/net/intel/idpf/idpf_rxtx_vec_common.h | 8 +
3
This patch fixes cipher data length, in the event of algorithm
chaining. When enqueuing crypto op to vhost backend
cipher.data.length is set correctly which is in
virtqueue_crypto_sym_pkt_header_arrange(). This field is computed
and assigned wrongly instead of using passed value. This is
rectified
In symmetric algorithm chaining, we need to consider both
cipher and auth data length for cipher source.
Fixes: 82adb12a1fce ("crypto/virtio: support burst enqueue/dequeue")
Signed-off-by: Rajesh Mudimadugula
---
drivers/crypto/virtio/virtio_rxtx.c | 20
1 file changed, 12
On Tue, Mar 18, 2025 at 1:41 PM Luca Vizzarro wrote:
>
> +def _sniff(self, recv_port: Port):
> +"""Sniff packets and use events and queue to communicate with the
> main thread.
> +
> +Raises:
> +InteractiveSSHSessionDeadError: If the SSH connection has
> been unexp
There are various counters/stats coming from ice_stats_get that appear
to only use 32 or 40 bits and once it hits that limit it "overflows"
and resets. This was causing large spikes (sudden 2^42 increase from
one second to the next) in values of ibytes and obytes, while causing
sudden drops in opac
35 matches
Mail list logo
