Hi,
From: Bing Zhao
Sent: Thursday, June 13, 2024 3:45 PM
To: Slava Ovsiienko; dev@dpdk.org; Raslan Darawsheh
Cc: Ori Kam; Dariusz Sosnowski; Suanming Mou; Matan Azrad; rongwei liu; Bing
Zhao; sta...@dpdk.org
Subject: [PATCH] net/mlx5: fix mlx5 device start failure
From: Rongwei Liu
When devar
Hi,
From: Gregory Etelson
Sent: Monday, June 17, 2024 5:03 PM
To: dev@dpdk.org
Cc: Gregory Etelson; Maayan Kashani; Raslan Darawsheh; Dariusz Sosnowski; Slava
Ovsiienko; Ori Kam; Suanming Mou; Matan Azrad
Subject: [PATCH] net/mlx5: fix HWS VXLAN item validation
HWS can match entire VXLAN, VXLAN-
Hi,
From: Gregory Etelson
Sent: Monday, June 17, 2024 5:07 PM
To: dev@dpdk.org
Cc: Gregory Etelson; Maayan Kashani; Raslan Darawsheh
Subject: [PATCH 0/2] fix non-template IPv6 RSS hash
Gregory Etelson (2):
net/mlx5: refactor non-template RSS expansion
net/mlx5: fix non-template IPv6 RSS hash
Hi,
From: Gregory Etelson
Sent: Monday, June 24, 2024 8:11 AM
To: dev@dpdk.org
Cc: Gregory Etelson; Maayan Kashani; Raslan Darawsheh; Dariusz Sosnowski; Slava
Ovsiienko; Ori Kam; Suanming Mou; Matan Azrad
Subject: [PATCH] net/mlx5: fix HWS GRE OPTION item validation
1. GRE_OPTION flow item vali
On 2024-06-22 00:24, Abdullah Sevincer wrote:
This commit adds support flag to enable enqueue reorder
feature.
"Enqueue reorder" is how this feature is implemented (on DLB2), but it's
not a good description of what it does (or, allows for).
I've called this feature "independent enqueue" in
On 2024-06-22 00:24, Abdullah Sevincer wrote:
This commit adds support flag to enable enqueue reorder
feature.
When this flag is enabled in the port configuration PMD
restores dequeue order on enqueue if applications happen to
change it.
Signed-off-by: Abdullah Sevincer
---
lib/eventdev/rte_
On 2024/6/28 11:22 AM, Yutang Jiang wrote:
The AmpereOneAC04 is efficient Cloud Native CPU:
Up to 192 Cores
2MB Private L2 Cache per Core
8 channel DDR5
128 lanes PCIe Gen5
Signed-off-by: Yutang Jiang
---
config/arm/arm64_ampereoneac04_linux_gcc | 17 +
config/a
On 2024/6/29 12:11 AM, Jack Bond-Preston wrote:
Signed-off-by: Jack Bond-Preston
Reviewed-by: Wathsala Vithanage
---
MAINTAINERS | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 533f707d5f..dfc022c0c6 100644
--- a/MAINTAINERS
+++ b/MAI
On 2024/6/29 12:11 AM, Jack Bond-Preston wrote:
Update maintainers for Bitops, Ticketlock, and ARMv8 Crypto.
Signed-off-by: Jack Bond-Preston
Reviewed-by: Wathsala Vithanage
---
MAINTAINERS | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
https://bugs.dpdk.org/show_bug.cgi?id=1475
Bug ID: 1475
Summary: [dpdk-24.07] l3fwd performance drop 20% on Ubuntu24.04
Product: DPDK
Version: 24.07
Hardware: x86
OS: Linux
Status: UNCONFIRMED
Severity: no
Hi Zhichao,
On 11/04/2024 10:29, Zhichao Zeng wrote:
This patch enabled querying Forward Error Correction(FEC) capabilities,
set FEC mode and get current FEC mode functions.
Signed-off-by: Qiming Yang
Signed-off-by: Zhichao Zeng
---
v5: fix some judgments
v4: fix some logic
v3: optimize code
Adds the memory barrier properly before posting ctlq tail. This
makes sure memory writes have a chance to take place before HW starts
messing with the descriptors.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/idpf_controlq.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/dr
These patches integrate the latest changes in MEV TS IDPF Base driver.
---
v6:
- Addressed comments
---
v5:
- Removed warning from patch 6
---
v4:
- Removed 1st patch as we are not using NVME_CPF flag
- Addressed comments
---
v3:
- Removed additional whitespace changes
- Fixed warnings of CI
- Upd
Introducing new macro STRUCT_VAR_LEN to support length of
dynamic data structures. Currently it is set to 1.
Introducing another structure VIRTCHNL2_CHECK_STRUCT_VAR_LEN
to check the length of dynamic data structures.
Some fixes based on code changes is introduced to compile dpdk.
Signed-off-by:
Added virtchnl2_flow_types to be used for flow steering.
Added flow steer cap flags for vport create.
Add flow steer flow types and action types for vport create.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/virtchnl2.h | 60 ++--
1 file changed, 57 inser
Adding a new structure for the context descriptor that contains
the support for timesync packets, where the index for timestamping is set.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/idpf_lan_txrx.h | 20 +++-
1 file changed, 19 insertions(+), 1 deletion(-)
diff
The file is redundant in our implementation and is not required
further.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/idpf_common.c | 382 -
drivers/common/idpf/base/meson.build | 1 -
2 files changed, 383 deletions(-)
delete mode 100644 drivers/common
Adding an argument named type to define queue type
in idpf_vc_queue_switch(). This solves the issue of
improper queue type in virtchnl2 message.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/idpf_common_virtchnl.c | 8 ++--
drivers/common/idpf/idpf_common_virtchnl.h | 2 +-
driver
Updated the latest support for cpfl pmd in MEV TS
firmware version which is 1.4.
Signed-off-by: Soumyadeep Hore
---
doc/guides/nics/cpfl.rst | 2 ++
1 file changed, 2 insertions(+)
diff --git a/doc/guides/nics/cpfl.rst b/doc/guides/nics/cpfl.rst
index 9b7a99c894..528c809819 100644
--- a/doc/gui
> Introduced through customer's feedback in their attempt to address
> some bugs this introduces a memory barrier before posting ctlq tail.
> This makes sure memory writes have a chance to take place before HW
> starts messing with the descriptors.
>
> Signed-off-by: Soumyadeep Hore
> ---
>Fr
> With the internal Linux upstream feedback that is received on IDPF
> driver and also some references available online, it is discouraged to
> use 1-sized array fields in the structures, especially in the new
> Linux drivers that are going to be upstreamed. Instead, it is
> recommended to use
From: Ori Kam
Until now the number of pattern templates that was
supported per table was limited to 2.
This was the result of the limitation that the table
could only support 1 matcher.
which meant that we could only support merge of
Ipv4 + TCP and IPv4 + UDP.
With the added ability to use exten
Elf files generated by higher version compilers wrap multiple
symbols prefixed with "this_pmd_name".
The patch uses the regex "^this_pmd_name[0-9]+$" to match the
symbol name.
Bugzilla ID: 1466
Fixes: 6c4bf8f42432 ("buildtools: add Python pmdinfogen")
Cc: sta...@dpdk.org
Signed-off-by: Mingjin Y
On Mon, Jul 01, 2024 at 09:13:44AM +, Soumyadeep Hore wrote:
> These patches integrate the latest changes in MEV TS IDPF Base driver.
>
> ---
> v6:
> - Addressed comments
> ---
> v5:
> - Removed warning from patch 6
> ---
> v4:
> - Removed 1st patch as we are not using NVME_CPF flag
> - Addres
Acked-by: Kai Ji
From: Jack Bond-Preston
Sent: 07 June 2024 13:47
To: Ji, Kai ; Fan Zhang ; Akhil Goyal
Cc: dev@dpdk.org ; sta...@dpdk.org ; Wathsala
Vithanage
Subject: [PATCH v4 1/5] crypto/openssl: fix GCM and CCM thread unsafe ctxs
Commit 67ab783b5d70 ("c
Acked-by: Kai Ji
From: Jack Bond-Preston
Sent: 07 June 2024 13:47
To: Ji, Kai
Cc: dev@dpdk.org ; Wathsala Vithanage
Subject: [PATCH v4 2/5] crypto/openssl: only init 3DES-CTR key + impl once
Currently the 3DES-CTR cipher context is initialised for every buffer,
Acked-by: Kai Ji
From: Jack Bond-Preston
Sent: 07 June 2024 13:47
To: Ji, Kai
Cc: dev@dpdk.org ; Wathsala Vithanage
Subject: [PATCH v4 4/5] crypto/openssl: per-qp auth context clones
Currently EVP auth ctxs (e.g. EVP_MD_CTX, EVP_MAC_CTX) are allocated,
copied t
Acked-by: Kai Ji
From: Jack Bond-Preston
Sent: 07 June 2024 13:47
To: Ji, Kai
Cc: dev@dpdk.org ; Wathsala Vithanage
Subject: [PATCH v4 3/5] crypto/openssl: per-qp cipher context clones
Currently EVP_CIPHER_CTXs are allocated, copied to (from
openssl_session), a
Acked-by: Kai Ji
From: Jack Bond-Preston
Sent: 07 June 2024 13:47
To: Ji, Kai
Cc: dev@dpdk.org ; Wathsala Vithanage
Subject: [PATCH v4 5/5] crypto/openssl: only set cipher padding once
Setting the cipher padding has a noticeable performance footprint,
and it do
On Fri, Jun 28, 2024 at 9:37 AM Chaoyong He wrote:
>
> This patch series aims to fix some problems in secondary process PCI UIO
> resource map logic.
>
> ---
> v2:
> * Modify as the advice from reviewer.
> v3:
> * Modify logic as the comments of reviewers.
> v4:
> * Split the commits to make it ea
On Fri, Jun 28, 2024 at 9:37 AM Chaoyong He wrote:
>
> From: Zerun Fu
>
> In the previous logic the 'fd' was only saved in the primary process,
> but for some devices this value is also used in the secondary logic.
>
> For example, the call of 'rte_pci_find_ext_capability()' will fail in
> the se
The bulk-allocation array is used back to front, so we need to free
everything before the marker, not after it. Flip ionic_empty_array()
so that it frees from 0 to the provided index. Adjust the callers
as needed.
Fixes: 218afd825bca ("net/ionic: do bulk allocations of Rx mbufs")
CC: sta...@dpdk.o
Coverity pointed out that, if the final segment of the session key being
written is not a full segment, the loop could potentially read past the
end of the source buffer. Use RTE_MIN() to make sure to only copy as much
of the key as is left.
Coverity issue: 426432
Fixes: 6bc7f2cf6687 ("crypto/ioni
(uint16_t * uint16_t) promoted to uint64_t has a sign extension
problem reported by Coverity. Cast one arg to uint64_t first
to eliminate the sign extension.
Coverity issue: 426422
Coverity issue: 426427
Fixes: 2c1662bb53ca ("crypto/ionic: add adminq command")
Signed-off-by: Andrew Boyer
---
dr
https://bugs.dpdk.org/show_bug.cgi?id=1476
Bug ID: 1476
Summary: Frame Overhead Inconsistencies Within Drivers
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
If I would like to create a test suite that is driver agnostic, then I
would need to increase the boundaries at which I conduct each
individual test case. Right now, I test jumbo frame behavior -1, +1,
or at the MTU boundary. But, packets in TestPMD are dropped or
forwarded based on the MTU size pl
>
> > +def send_packet_and_verify(
> > +self,
> > +mac_address: str,
> > +add_vlan: bool = False,
> > +should_receive: bool = True,
> > +) -> None:
> > +"""Generate, send, and verify a packet based on specified
> > parameters.
> > +
> > +Te
https://bugs.dpdk.org/show_bug.cgi?id=1477
Bug ID: 1477
Summary: Port Rx/Tx offload test suite to new DTS
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
The python check tool (flake8) is picky about the indentation
of continuation lines, and dpdk-devbind was not following standard.
Error is:
E127 continuation line over-indented for visual indent
Fixes: 2ff801515e49 ("usertools/devbind: update octeontx2 DMA device")
Signed-off-by: Stephen
On Wed, 12 Oct 2022 17:38:17 +0500
Fidaullah Noonari wrote:
> binding with vfio driver, when IOMMU is disabled, causes program to crash.
> this patch adds a flag for noiommmu-mode. when this is set, if IOMMU is
> disabled, it changes vfio into unsafe noiommu mode and prints warning
> message.
>
Fix the warning
$ flake8 --max-line-length=100 dpdk-pmdinfo.py
dpdk-pmdinfo.py:217:40: E203 whitespace before ':'
Signed-off-by: Stephen Hemminger
---
usertools/dpdk-pmdinfo.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/usertools/dpdk-pmdinfo.py b/usertools/dpdk-pmdinfo.
> -Original Message-
> From: Mahmoud Maatouq
> Sent: Monday, June 24, 2024 21:10
> To: Dariusz Sosnowski
> Cc: Slava Ovsiienko ; Bing Zhao ;
> Ori Kam ; Suanming Mou ; Matan
> Azrad ; dev@dpdk.org
> Subject: Re: [PATCH] net/mlx5: fix memleak for resource object
>
> On 06/24, Dariusz Sosn
Hi,
Sorry for the delayed review. Just two small comments.
> -Original Message-
> From: Mahmoud Maatuq
> Sent: Monday, June 24, 2024 21:42
> To: Dariusz Sosnowski ; Slava Ovsiienko
> ; Bing Zhao ; Ori Kam
> ; Suanming Mou ; Matan Azrad
> ; Maayan Kashani
> Cc: dev@dpdk.org; Mahmoud Maat
Added global hairpin out of buffer counter (from ethtool)
Also added port-level hairpin out of buffer counter.
Shani Peretz (2):
net/mlx5: add global hairpin out of buffer counter
net/mlx5: add hairpin out of buffer counter
doc/guides/nics/mlx5.rst| 3 ++
doc/guides/rel_note
Exposing ethtool dev_internal_queue_oob counter which counts
the number of hairpin dropped packets in xstats.
Signed-off-by: Shani Peretz
Acked-by: Dariusz Sosnowski
---
doc/guides/rel_notes/release_24_07.rst | 1 +
drivers/net/mlx5/linux/mlx5_ethdev_os.c | 4
2 files changed, 5 insertion
Currently mlx5 PMD exposes rx_out_of_buffer counter that tracks
packets dropped when Rx queue was full.
To provide more granular statistics, this patch splits the
`rx_out_of_buffer` counter into two separate counters:
1. hairpin_out_of_buffer - This counter specifically tracks packets dropped
by t
https://bugs.dpdk.org/show_bug.cgi?id=1478
Bug ID: 1478
Summary: Rework Port Topology & Node Configuration In DTS
Configuration
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UN
Hey Dean,
I think this series looks good, thanks for addressing my comments! The
only other comment I have on this version is that it looks like the
formatting script is throwing some warnings, so those fixes might have
been left out of the diff when this series was submitted. I think once
the for
On Fri, Jun 28, 2024 at 10:00 AM Dean Marx wrote:
>
> Test suite for verifying VLAN filtering, stripping, and insertion
> functionality on Poll Mode Driver.
>
> Signed-off-by: Dean Marx
> ---
> +def send_vlan_packet_and_verify(self, should_receive: bool, strip: bool,
> vlan_id: int) -> None
DPDK allows for two ways to specify PCI device numbers:
a full version (":08:00.0") and a short version ("08:00.0").
The problem arises when the application uses one format (e.g., full)
when running testpmd, but then tries to use the other format (e.g., short)
in a subsequent command, leading t
Hey Dean,
Thanks for the new series. One thing I noted for the series as a whole
is that there are a few small errors being thrown by the formatting
script. One of the errors is on a method I wrote and gave you so I
apologize for that, it looks like I changed a parameter to a method
but didn't swi
On Fri, Jun 28, 2024 at 12:20 PM Dean Marx wrote:
>
> This suite tests the ability of the Poll Mode Driver to enable
> and disable Rx/Tx queues on a port.
>
> Signed-off-by: Dean Marx
> ---
> dts/tests/TestSuite_queue_start_stop.py | 95 +
> 1 file changed, 95 insertions(
Hi Maxime, Hemant,
Capturing below an extension for 24.11.
This includes bbdev api change and related change for
test application and acc/vrb PMD.
This provides the ability which was introduced in 3gPP
standard to have an alternate way to compute k0 no longer
always computed just from rv_index. W
In latest 3GPP 38.212, the k0 value is not necessaraly directly
derived from rv_index, in that case a value can be provided in
API. When this value is non null this would override the value
which would be computed purely from rv_index.
Signed-off-by: Nicolas Chautru
---
doc/guides/prog_guide/bbd
New k0 added in LDPC decoder structure can be parsed
by the bbdev test application.
Signed-off-by: Nicolas Chautru
---
app/test-bbdev/test_bbdev_perf.c | 7 +--
app/test-bbdev/test_bbdev_vector.c | 4
app/test-bbdev/test_bbdev_vector.h | 1 +
3 files changed, 10 insertions(+), 2 delet
The k0 may be provided to the PMD, which would be used
as the circular buffer starting position instead of value
which would be computed from legacy rv_index derived
computation.
Signed-off-by: Nicolas Chautru
---
drivers/baseband/acc/acc_common.h | 6 --
drivers/baseband/acc/rte_acc100_
> 19/06/2024 08:45, Wathsala Vithanage:
> > --- a/lib/eal/arm/include/rte_cpuflags_64.h
> > +++ b/lib/eal/arm/include/rte_cpuflags_64.h
> > @@ -36,6 +36,7 @@ enum rte_cpu_flag_t {
> > RTE_CPUFLAG_SVEF64MM,
> > RTE_CPUFLAG_SVEBF16,
> > RTE_CPUFLAG_AARCH64,
> > + RTE_CPUFLAG_WFXT,
> >
On Mon, 1 Jul 2024 23:01:17 +0300
Shani Peretz wrote:
> --- a/drivers/bus/pci/pci_common.c
> +++ b/drivers/bus/pci/pci_common.c
> @@ -501,6 +501,15 @@ rte_pci_dump(FILE *f)
> pci_dump_one_device(f, dev);
> }
> }
> +static int
> +pci_cmp_name(const struct rte_device *dev1, con
Hi Maxime.
Adding new support to support troubleshooting. This provides
to the application an API to dump into file information
to help troubleshoot issue on a queue. Some of it is tracked
at bbdev level and some extra information can be tracked
as an option at PMD level. This is for 24.11.
In p
This provides a new API to dump more debug information
related to the status on a given bbdev queue.
Some of this information is visible at bbdev level.
This also provides a new option dev op, to print more
information at the lower PMD level.
This helps user to troubleshoot issues related to
previo
Support the new dev op to dump operations information
related to a given queue and information on previous errors
detected by the driver and tracked internally in PMD.
Signed-off-by: Nicolas Chautru
---
drivers/baseband/acc/acc_common.h | 37 +
drivers/baseband/acc/rte_vrb_pmd.c | 128
> On Fri, Jun 28, 2024 at 9:37 AM Chaoyong He
> wrote:
> >
> > From: Zerun Fu
> >
> > In the previous logic the 'fd' was only saved in the primary process,
> > but for some devices this value is also used in the secondary logic.
> >
> > For example, the call of 'rte_pci_find_ext_capability()' wil
This patch series aims to fix some problems in secondary process PCI UIO
resource map logic.
---
v2:
* Modify as the advice from reviewer.
v3:
* Modify logic as the comments of reviewers.
v4:
* Split the commits to make it easier to understand.
v5:
* Fix the unmapping UIO resources probelm found b
From: Zerun Fu
To makes the logic clearer, rename the variable 'i' into 'map_idx'
and the variable 'j' into 'i'.
Signed-off-by: Zerun Fu
Reviewed-by: Chaoyong He
Reviewed-by: Long Wu
Reviewed-by: Peng Zhang
---
drivers/bus/pci/pci_common_uio.c | 30 +++---
1 file cha
From: Zerun Fu
For the primary process, the logic loops all BARs and will skip
the map of BAR with an invalid physical address (0), also will
assign 'uio_res->nb_maps' with the real mapped BARs number. But
for the secondary process, instead of loops all BARs, the logic
using the 'uio_res->nb_map'
From: Zerun Fu
In the previous logic the 'fd' was only saved in the primary process,
but for some devices this value is also used in the secondary logic.
For example, the call of 'rte_pci_find_ext_capability()' will fail in
the secondary process.
Fix this problem by getting and saving the value
On Tue, 2 Jul 2024 10:19:44 +0800
Chaoyong He wrote:
> - int fd, i, j;
> + int fd, i, map_idx;
> struct mapped_pci_resource *uio_res;
> struct mapped_pci_res_list *uio_res_list =
> RTE_TAILQ_CAST(rte_uio_tailq.head, mapped_pci_res_list);
> @@ -37,41 +37,
The deeper the idle state, the lower the power consumption, but the longer
the resume time. Some service are delay sensitive and very except the low
resume time, like interrupt packet receiving mode.
And the "/sys/devices/system/cpu/cpuX/power/pm_qos_resume_latency_us" sysfs
interface is used to s
The deeper the idle state, the lower the power consumption, but the longer
the resume time. Some service are delay sensitive and very except the low
resume time, like interrupt packet receiving mode.
And the "/sys/devices/system/cpu/cpuX/power/pm_qos_resume_latency_us" sysfs
interface is used to s
Add PM QoS configuration to declease the delay after sleep in case of
entering deeper idle state.
Signed-off-by: Huisong Li
Acked-by: Morten Brørup
---
examples/l3fwd-power/main.c | 28
1 file changed, 28 insertions(+)
diff --git a/examples/l3fwd-power/main.c b/exa
The la12xx driver do not have any checks for secondary process
and it causes the system to try to initialize the driver, causing
segmentation faults.
LA12xx driver do not support multi-processing.
Return when not called from Primary process.
Fixes: f218a1f92017 ("baseband/la12xx: introduce NXP LA1
On 02-07-2024 02:57, Nicolas Chautru wrote:
Hi Maxime, Hemant,
Capturing below an extension for 24.11.
This includes bbdev api change and related change for
test application and acc/vrb PMD.
This provides the ability which was introduced in 3gPP
standard to have an alternate way to compute k0
72 matches
Mail list logo
