On 6/12/24 17:00, Anatoly Burakov wrote:
From: Ian Stokes
Refactor __ice_aq_get_set_rss_lut():
get:
- make use params->lut_size only as a size of params->lut;
- return LUT size via params->lut_size;
set:
- remove option to set RSS LUT smaller than available
(eg forbid PF LUT sized 512
> To optimize AES-GCM IPsec operation within crypto/mlx5,
> the DPDK API typically supplies AES_GCM AAD/Payload/Digest
> in separate locations, potentially disrupting their
> contiguous layout. In cases where the memory layout fails
> to meet hardware (HW) requirements, an UMR WQE is initiated
> ah
> On 4/26/24 09:48, Chaoyong He wrote:
> > From: Xinying Yu
> >
> > After migrate to new host, the vring information was
>
> migrating*
>
> is recovered*
>
Will fix in the v2 patch series.
Thanks for your review.
> > recovered by the value in offset 'NFP_NET_CFG_TX_USED_INDEX'
> > and 'NFP_NE
> Subject: [PATCH v4] net/mana: fix uninitialized scalar variable
>
> [You don't often get email from mahmoudmatook...@gmail.com. Learn why
> this is important at https://aka.ms/LearnAboutSenderIdentification ]
>
> in case we didn't enter or finished the loop without getting ifname we ret
> shoul
> On 4/26/24 09:48, Chaoyong He wrote:
> > From: Xinying Yu
> >
> > Create the relay vring on host and then set the address of Rx used
> > ring to the VF config bar. So the device can DMA the used ring
> > information to host rather than directly to VM.
> >
> > Use 'NFP_NET_CFG_CTRL_LM_RELAY' noti
> On 4/26/24 09:48, Chaoyong He wrote:
> > From: Xinying Yu
> >
> > Add the basic logic of software live migration.
> >
> > Unset the ring notify area to stop the direct IO datapath if the
> > device support, then we can setup the vring relay to help the live
> > migration.
> >
> > Signed-off-by:
> On 4/26/24 09:48, Chaoyong He wrote:
> > From: Xinying Yu
> >
> > The ctrl words of vDPA is locate on the extend word, so should use the
> > 'nfp_ext_reconfig()' rather than 'nfp_reconfig()'.
>
> located*
> extended*
>
> so it should*
>
Will fix in the v2 patch series.
Thanks for your review
On Fri, 14 Jun 2024 at 01:59, Akhil Goyal wrote:
>
> > Introduce a new compress PMD for hardware accelerators based on UADK [1].
> >
> > UADK is a framework for user applications to access hardware accelerators.
> > UADK relies on IOMMU SVA (Shared Virtual Address) feature, which share
> > the sam
The IPsec operation shrinks AAD directly before payload in
enqueue burst and restores the memory in dequeue burst.
This commit adds the support of OOP mode follows the
similar strategy.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
doc/guides/cryptodevs/mlx5.rst| 3 ++
drivers
To optimize AES-GCM IPsec operation within crypto/mlx5,
the DPDK API typically supplies AES_GCM AAD/Payload/Digest
in separate locations, potentially disrupting their
contiguous layout. In cases where the memory layout fails
to meet hardware (HW) requirements, an UMR WQE is initiated
ahead of the G
To optimize AES-GCM IPsec operation within crypto/mlx5,
the DPDK API typically supplies AES_GCM AAD/Payload/Digest
in separate locations, potentially disrupting their
contiguous layout. In cases where the memory layout fails
to meet hardware (HW) requirements, an UMR WQE is initiated
ahead of the G
> -Original Message-
> From: Akhil Goyal
> Sent: Friday, June 14, 2024 2:02 AM
> To: Suanming Mou
> Cc: dev@dpdk.org
> Subject: RE: [PATCH 0/2] crypto/mlx5: optimize AES-GCM IPsec operation
>
> > Hi Akhil,
> >
> > Sorry, I just noticed seems you were not added to the list by maintainer
Prior to this change, queue page lists (QPLs) were kept as device
resources, being stored in the gve_priv struct. This does not make
sense because each QPL inherently belongs to a single queue.
This change moves all QPL resources into the queues themselves, and
couples QPL allocation/registration
Prior to this change, queue page lists (QPLs) were kept as device
resources, being stored in the gve_priv struct. This does not make
sense because each QPL inherently belongs to a single queue.
This change moves all QPL resources into the queues themselves, and
couples QPL allocation/registration
On Wed, 5 Jul 2023 17:35:14 +0800
Kaisen You wrote:
>
> +/* get cmdline form PID. Read process info form /proc/$PID. */
> +static char *get_cmdline_from_pid(pid_t pid, char *buf, int len)
> +{
> + char filename[PATH_MAX];
> + char *name = NULL;
> + int fd;
> + int ret;
> +
> +
Prior to this change, queue page lists (QPLs) were kept as device
resources, being stored in the gve_priv struct. This does not make
sense because each QPL inherently belongs to a single queue.
This change moves all QPL resources into the queues themselves, and
couples QPL allocation/registration
https://bugs.dpdk.org/show_bug.cgi?id=1459
Bug ID: 1459
Summary: Port over dynamic_queue test suite to new DTS
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
From: Jeremy Spewock
The InteractiveShell class currently relies on being cleaned up and
shutdown at the time of garbage collection, but this cleanup of the class
does no verification that the session is still running prior to cleanup.
So, if a user were to call this method themselves prior to ga
The argument is '--lcores' but it is being logged as '--lcore'
in several places.
e.g.
$ dpdk-testpmd -l 8,10,12 --lcores 8,10,12
EAL: Option --lcore is ignored, because (-l) is set!
Fix the spelling in logs and tests.
Fixes: 117eaa70584b ("eal: add error check for core options")
Signed-off-by:
Hello,
This is a reminder that the DTS meeting for next Wednesday has been moved
to next Thursday, 6/20, at 13:00 UTC, to adjust for the American Juneteenth
Holiday. Thanks non-Americans for adjusting.
I've modified the Linux Foundation event, so if that is writing to your
calendar, your event sh
From: Jeremy Spewock
There are methods within DTS currently that support updating the MTU of
ports on a node, but the methods for doing this in a linux session rely
on the ip command and the port being bound to the kernel driver. Since
test suites are run while bound to the driver for DPDK, there
Rework 'lcores' and 'memory_channels' into a new 'dpdk_config'
subsection in an effort to make these attributes SUT specific; the
traffic generator, more often than not, does not need this information.
Ideally, if such information is needed, then it will be listed in the
'traffic_generator' compone
The 'arch' attribute in the conf.yaml is unnecessary, as this can be
readily discovered within the constructor of any given node. Since OS is
determined within user configuration, finding system arch can be done
both reliably and easily within the framework.
For Linux/Posix systems, the 'uname' co
Removed use_first_core from the conf.yaml in favor of determining this
within the framework. use_first_core continue to serve a purpose in that
it is only enabled when core 0 is explicitly provided in the
configuration. Any other configuration, including "" or "any," will
omit core 0.
Documentatio
Remove the list of devices from the schema, as these are unuesed.
Likewise, removed build-target information since these is not currently
used, and it is unlikely to be used in the future. Adjustments to the
dts.rst are made to reflect these changes.
Bugzilla ID: 1360
Signed-off-by: Nicholas Pratt
A good amount of the attributes listed in the conf.yaml are either
currently unused or unneeded. The goal of this patch is to eliminate
minutiea from the config that may make the overall execution process
more difficult and tedious.
It should be noted that more improvements are possible here, and
> >> diff --git a/drivers/crypto/ionic/meson.build
> b/drivers/crypto/ionic/meson.build
> >> new file mode 100644
> >> index 00..4114e13e53
> >> --- /dev/null
> >> +++ b/drivers/crypto/ionic/meson.build
> >> @@ -0,0 +1,13 @@
> >> +# SPDX-License-Identifier: BSD-3-Clause
> >> +# Copyright 20
> Introduce a new compress PMD for hardware accelerators based on UADK [1].
>
> UADK is a framework for user applications to access hardware accelerators.
> UADK relies on IOMMU SVA (Shared Virtual Address) feature, which share
> the same page table between IOMMU and MMU.
> Thereby user applicatio
From: Jeremy Spewock
Interactive shells are managed in a way currently where they are closed
and cleaned up at the time of garbage collection. Due to there being no
guarantee of when this garbage collection happens in Python, there is no
way to consistently know when an application will be closed
On Thu, 13 Jun 2024 17:51:14 +0100
Ferruh Yigit wrote:
> > Hi Sivaprasad,
> >
> > Is this '(lcoreid_t)' cast required? Because of integer promotion I
> > think result will be correct without casting.
> >
> > (And without integer promotion considered, casting needs to be done on
> > one of the v
On Thu, Jun 13, 2024 at 6:14 PM Kevin Traynor wrote:
>
> The argument is '--lcores' but it is being logged as '--lcore'
> in several places.
>
> e.g.
> $ dpdk-testpmd -l 8,10,12 --lcores 8,10,12
> EAL: Option --lcore is ignored, because (-l) is set!
>
> Fix the spelling in logs and tests.
>
> Fixe
> Hi Akhil,
>
> Sorry, I just noticed seems you were not added to the list by maintainer
> script.
> Can you please take a look at the series when available.
>
Please resend. CI is complaining about apply failure.
in case we didn't enter or finished the loop without getting ifname we
ret shouldn't have an arbitrary value.
Coverity issue: 424690
Fixes: 84497839d9ca ("net/mana: support MTU update")
Cc: lon...@microsoft.com
Signed-off-by: Mahmoud Maatuq
---
v4:
* used full name in the sign off
v3:
* add Fix
From: Jeremy Spewock
Some NICs tested in DPDK allow for the scattering of packets without an
offload and others enforce that you enable the scattered_rx offload in
testpmd. The current version of the suite for testing support of
scattering packets only tests the case where the NIC supports testin
From: Jeremy Spewock
v4:
* comment and formatting adjustments based on feedback on previous
version
* switch class inheritance order for interactive shells and singe
active interactive shells
* method decorator for starting and stopping ports in testpmd shell
that gets used for the up
> > > From: Kundapura, Ganapati [mailto:ganapati.kundap...@intel.com]
> > > Sent: Thursday, 30 May 2024 16.22
> > >
> > > Hi,
> > >
> > > > From: Akhil Goyal
> > > > Sent: Thursday, May 30, 2024 5:17 PM
> > > >
> > > > > > > #if may not be needed in application.
> > > > > > > Test should be skippe
On 6/12/2024 12:37 AM, Ferruh Yigit wrote:
> On 6/6/2024 12:27 PM, Sivaprasad Tummala wrote:
>> With modern CPUs, it is possible to have higher
>> CPU count thus we can have higher RTE_MAX_LCORES.
>> In testpmd application, the current config forwarding
>> cores option "--nb-cores" is hard limited
On Thu, 13 Jul 2023 13:17:16 +0800
Kaisen You wrote:
> Add NIC exception parameter handling to dpdk_test process,
> program exits when carrying unbound vfio NIC parameters.
>
> Fixes: 50247fe03fe0 ("test/timer: exercise new APIs in secondary process")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Ka
On Mon, 12 Jun 2023 19:47:16 +
Wathsala Vithanage wrote:
> For improved performance over the current C11 based ring
> implementation, two ACQUIRE fences between the load of the old_head
> and the load of the const_tail in __rte_ring_move_prod_head and the
> load of old_head and load of the pr
On 4/19/2024 12:30 PM, Ferruh Yigit wrote:
> On 4/16/2024 10:55 AM, Sivaprasad Tummala wrote:
>> diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
>> index ba1007ace6..6b28c22c96 100644
>> --- a/app/test-pmd/config.c
>> +++ b/app/test-pmd/config.c
>> @@ -4785,9 +4785,9 @@ fwd_stream_on_oth
On Thu, 6 Jun 2024 10:44:22 +
Shaiq Wani wrote:
> Check whether CPFL PMD runs on Host or ACC
>
> ---
> v2 Changes:
> -Changed implementation based on review comment.
> v3 Changes:
> -Fixed indentation.
> v4 Changes:
> -Fix ipu_imc and ipu_acc to ipu-imc and ipu-acc.
> v5 Changes:
> -Updated
Hello,
On Thu, Jun 13, 2024 at 4:02 PM David Marchand
wrote:
>
> Hello Raju,
>
> On Thu, Jun 13, 2024 at 3:58 PM David Marchand
> wrote:
> >
> > It looks like Rahul left Chelsio (at least, his mail address is
> > rejected).
> > Remove him from the maintainers, which makes this driver an orphan.
> On Jun 13, 2024, at 6:07 AM, Akhil Goyal wrote:
>
> Caution: This message originated from an External Source. Use proper caution
> when opening attachments, clicking links, or responding.
>
>
>> diff --git a/drivers/crypto/ionic/meson.build
>> b/drivers/crypto/ionic/meson.build
>> new fil
Hello Raju,
On Thu, Jun 13, 2024 at 3:58 PM David Marchand
wrote:
>
> It looks like Rahul left Chelsio (at least, his mail address is
> rejected).
> Remove him from the maintainers, which makes this driver an orphan.
>
> Signed-off-by: David Marchand
As we have no other contact than Rahul, I am
It looks like Rahul left Chelsio (at least, his mail address is
rejected).
Remove him from the maintainers, which makes this driver an orphan.
Signed-off-by: David Marchand
---
MAINTAINERS | 1 -
1 file changed, 1 deletion(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index c9adff9846..fe44179ccd 1
On Wed, Mar 6, 2024 at 11:14 PM Tyler Retzlaff
wrote:
>
> Clean up direct uses of GCC __attribute__((__pure__)) and
> __attribute__((__may_alias__)). A checkpatch check already warns for
> new additions.
>
> Tyler Retzlaff (5):
> eal: add rte pure attribute macro
> lpm: use rte macro instead o
On Thu, Jun 13, 2024 at 3:44 PM David Marchand
wrote:
>
> On Wed, Jun 12, 2024 at 10:16 AM David Marchand
> wrote:
> >
> > On Wed, Mar 6, 2024 at 11:14 PM Tyler Retzlaff
> > wrote:
> > >
> > > Use existing __rte_may_alias macro from rte_common.h instead of
> > > directly using __attribute__((__m
On 6/12/2024 7:23 PM, Mahmoud Maatuq wrote:
> From: mmaatuq
>
> in case we didn't enter or finished the loop without getting ifname we
> ret shouldn't have an arbitrary value.
>
> Coverity issue: 424690
> Fixes: 84497839d9ca ("net/mana: support MTU update")
> Cc: lon...@microsoft.com
>
> Signed
Hello Rahul,
On Wed, Jun 12, 2024 at 10:16 AM David Marchand
wrote:
>
> On Wed, Mar 6, 2024 at 11:14 PM Tyler Retzlaff
> wrote:
> >
> > Use existing __rte_may_alias macro from rte_common.h instead of
> > directly using __attribute__((__may_alias__)).
> >
> > Signed-off-by: Tyler Retzlaff
> > --
On Wed, Jun 12, 2024 at 9:43 AM Morten Brørup
wrote:
> >
> > Flex arrays are part of the C99 standard.
> > No need to mark them with __extension__.
> >
> > Signed-off-by: David Marchand
> Acked-by: Morten Brørup
CI is happy, and so am I.
Applied, thanks.
--
David Marchand
The method will not create any new flow rule implicitly during split
stage, but only generate needed items, actions and attributes based
on the detection.
All the rules will still be created in the flow_hw_list_create().
In the meanwhile, once the mark action is specified in the FDB rule,
a new r
From: Rongwei Liu
When devargs "allow_duplicate_pattern=0" is specified, PMD won't
allow duplicated flows to be inserted and return EEXIST as rte_errno.
The queue/RSS split table is shared globally by all representors and
PMD didn't prepend port information into it, so all the following ports
tr
>
> On 5/23/2024 5:26 PM, Konstantin Ananyev wrote:
> > From: Konstantin Ananyev
> >
> > ../drivers/net/ice/ice_rxtx.c:1871:29: warning: variable length array used
> > [-Wvla]
> >
> > Here VLA is used as a temp array for mbufs that will be used as a split
> > RX data buffers.
> > As at any giv
> diff --git a/drivers/crypto/ionic/meson.build
> b/drivers/crypto/ionic/meson.build
> new file mode 100644
> index 00..4114e13e53
> --- /dev/null
> +++ b/drivers/crypto/ionic/meson.build
> @@ -0,0 +1,13 @@
> +# SPDX-License-Identifier: BSD-3-Clause
> +# Copyright 2021-2024 Advanced Micro
On 4/26/24 09:48, Chaoyong He wrote:
From: Xinying Yu
Add the software assisted vDPA live migration feature
into NFP document.
Signed-off-by: Xinying Yu
Reviewed-by: Chaoyong He
Reviewed-by: Long Wu
Reviewed-by: Peng Zhang
---
doc/guides/vdpadevs/nfp.rst | 9 +
1 file changed
On 4/26/24 09:48, Chaoyong He wrote:
From: Xinying Yu
Setup the vring relay thread to monitor the interruption from
device. And do the dirty page logging or notify device according
to event data.
Signed-off-by: Xinying Yu
Reviewed-by: Chaoyong He
Reviewed-by: Long Wu
Reviewed-by: Peng Zh
On 4/26/24 09:48, Chaoyong He wrote:
From: Xinying Yu
After migrate to new host, the vring information was
migrating*
is recovered*
recovered by the value in offset 'NFP_NET_CFG_TX_USED_INDEX'
and 'NFP_NET_CFG_RX_USED_INDEX'.
Signed-off-by: Xinying Yu
Reviewed-by: Chaoyong He
Reviewe
On 4/26/24 09:48, Chaoyong He wrote:
From: Xinying Yu
Create the relay vring on host and then set the address of Rx
used ring to the VF config bar. So the device can DMA the
used ring information to host rather than directly to VM.
Use 'NFP_NET_CFG_CTRL_LM_RELAY' notify the device side. And
On 4/26/24 09:48, Chaoyong He wrote:
From: Xinying Yu
Add the interrupt setup logic of vring relay.
The epoll fd is provided here so host can get the interrupt from device
on Rx direction, all other operations on vring relay are based on this.
Signed-off-by: Xinying Yu
Reviewed-by: Chaoyo
On 4/26/24 09:48, Chaoyong He wrote:
From: Xinying Yu
Add the basic logic of software live migration.
Unset the ring notify area to stop the direct IO datapath if the
device support, then we can setup the vring relay to help the
live migration.
Signed-off-by: Xinying Yu
Reviewed-by: Chaoy
On 4/26/24 09:48, Chaoyong He wrote:
From: Xinying Yu
In order to add the new configuration logic of software live
migration, split the datapath update logic into two parts,
queue configuration and VF configuration.
Signed-off-by: Xinying Yu
Reviewed-by: Chaoyong He
Reviewed-by: Long Wu
On 4/26/24 09:48, Chaoyong He wrote:
From: Xinying Yu
The ctrl words of vDPA is locate on the extend word, so should
use the 'nfp_ext_reconfig()' rather than 'nfp_reconfig()'.
located*
extended*
so it should*
Also replace the misuse of 'NFP_NET_CFG_CTRL_SCATTER' macro
with 'NFP_NET_CFG
On 4/26/24 09:48, Chaoyong He wrote:
From: Xinying Yu
Reconfigure the NIC will fail because lack of the
initialization logic of queue configuration pointer.
Fix this by adding the correct initialization logic.
Fixes: d89f4990c14e ("vdpa/nfp: add hardware init")
Cc: chaoyong...@corigine.com
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
---
examples/l3fwd-power/main.c | 29 +
1 file changed, 29 insertions(+)
diff --git a/examples/l3fwd-power/main.c b/examples/l3fwd-power/main.c
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
Hi Ferruh,
> On 5/23/2024 5:26 PM, Konstantin Ananyev wrote:
> > From: Konstantin Ananyev
> >
> > To continue further work on VLA replacement for these series:
> > https://patches.dpdk.org/project/dpdk/list/?series=31887
> > extra few patches to remove VLA for gro lib and ixgbe and ice PMDs.
> >
>
> On 5/23/2024 5:26 PM, Konstantin Ananyev wrote:
> > From: Konstantin Ananyev
> >
> > ../lib/gro/rte_gro.c:182:34: warning: variable length array used [-Wvla]
> > ../lib/gro/rte_gro.c:363:34: warning: variable length array used [-Wvla]
> >
> > In both cases the pattern is the same: we use un
68 matches
Mail list logo
