On Fri, Nov 19, 2021 at 5:54 PM Dumitrescu, Cristian
wrote:
> On a different point, we should probably tweak our autotests to differentiate
> between logical failures and those failures related to resources not being
> available, and flag the test result accordingly in the report. For example,
>-Original Message-
>From: dev On Behalf Of Asaf Penso
>Sent: Sunday, July 25, 2021 7:54 AM
>To: Ferruh Yigit ; Ajit Khaparde
>; NBU-Contact-Thomas Monjalon
>
>Cc: us...@dpdk.org; dpdk-dev ; David Marchand
>; Bruce Richardson
>; Jerin Jacob Kollanukkaran
>; Akhil Goyal ; Maxime Coquelin
>;
Hi Ferruh,
Thanks for your explaining.
It it by design, maybe a check and only rewrite the mac addresses when they are
explicitly set?
Thanks,
Wei
-Original Message-
From: Yigit, Ferruh
Sent: Tuesday, November 16, 2021 3:00 AM
To: Xu, Wei1 ; Lu, Wenzhuo ; Wu,
Jingjing ; Iremonger, B
Hi Sean,
A few minor nits, TBH not sure if they should be changed in this patch as this
patch is just removing "driver" and correcting capitalisation of PMD.
> diff --git a/drivers/net/hns3/hns3_ethdev.h
> b/drivers/net/hns3/hns3_ethdev.h
> index 6d30125dcc..488fe8dbbc 100644
> --- a/drivers/
Reviewed-by: Conor Walsh
Reviewed-by: Conor Walsh
22/11/2021 08:59, Rakesh Kudurumalla:
> From: Thomas Monjalon
> > 20/07/2021 18:50, Rakesh Kudurumalla:
> > > Current pmd_perf_autotest() in continuous mode tries to enqueue
> > > MAX_TRAFFIC_BURST completely before starting the test. Some drivers
> > > cannot accept complete MAX_TRAFFIC_BURST eve
DPDK 21.11 adds vfio support for DMA device in vhost. This patch
updates recommended IOVA mode in async datapath.
Signed-off-by: Xuan Ding
---
doc/guides/prog_guide/vhost_lib.rst | 9 +
1 file changed, 9 insertions(+)
diff --git a/doc/guides/prog_guide/vhost_lib.rst
b/doc/guides/prog_g
On 11/22/2021 4:00 AM, Xu, Wei1 wrote:
Hi Ferruh,
Thanks for your explaining.
It it by design, maybe a check and only rewrite the mac addresses when they are
explicitly set?
What are you trying to fix/achieve exactly?
Thanks,
Wei
-Original Message-
From: Yigit, Ferruh
Sent: Tuesd
rte_kvargs_parse() expects a NULL terminated keys array.
Fixes: 1afce3086cf4 ("bus/auxiliary: introduce auxiliary bus")
Signed-off-by: David Marchand
---
drivers/bus/auxiliary/auxiliary_params.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/bus/auxiliary/auxiliary_params.c
b/driv
From: Dapeng Yu
When VF is reset, PF will change DCF state from ON to other state, if
flow creation, destroy, or redirect command is sent to DCF at this
time, it will fail.
This patch tracks DCF state and returns try-again error to caller when
DCF state is not ON.
Cc: sta...@dpdk.org
Signed-of
On 10/21/2021 10:42 PM, vipul.as...@oracle.com wrote:
From: Vipul Ashri
failsafe crashed while sending early link_update request during
boot time initialization.
Based on debugging we found failsafe device was good but sub-
devices were progressing towards initialization and SUBOPS macro
where
On Mon, 2021-11-22 at 10:25 +0100, David Marchand wrote:
> rte_kvargs_parse() expects a NULL terminated keys array.
>
> Fixes: 1afce3086cf4 ("bus/auxiliary: introduce auxiliary bus")
>
> Signed-off-by: David Marchand
> ---
> drivers/bus/auxiliary/auxiliary_params.c | 1 +
> 1 file changed, 1 in
From: Elena Agostini
gpudev free and unregister functions return gracefully if input pointer is NULL
Fixes: 818a067baf90 ("gpudev: manage NULL pointer")
Signed-off-by: Elena Agostini
---
drivers/gpu/cuda/cuda.c | 6 --
lib/gpudev/gpudev.c | 6 ++
2 files changed, 6 insertions(+),
On Thu, Oct 21, 2021, at 23:42, vipul.as...@oracle.com wrote:
> From: Vipul Ashri
>
> failsafe crashed while sending early link_update request during
> boot time initialization.
> Based on debugging we found failsafe device was good but sub-
> devices were progressing towards initialization and SU
> -Original Message-
> From: Gregory Etelson
> Sent: Tuesday, November 16, 2021 17:45
> To: dev@dpdk.org; Gregory Etelson
> Cc: Raslan Darawsheh ; NBU-Contact-Thomas Monjalon
> ; Matan Azrad ; Slava
> Ovsiienko
> Subject: [PATCH v2] doc: add mlx5 PMD flex item documentation
>
> Describe
This patchset cleans up the use of the phrase PMD
throughout dpdk by capitalising pmd and also
removing the use of the word driver following PMD.
This patchset also removes some unnecessary
duplication of the word "the" in comments and
docs.
Sean Morrissey (3):
fix PMD wording typo
fix 'the t
Removing the use of driver following PMD as its
unnecessary.
Signed-off-by: Sean Morrissey
Signed-off-by: Conor Fogarty
Acked-by: John McNamara
Reviewed-by: Conor Walsh
---
app/test-pmd/cmdline.c| 4 +--
doc/guides/bbdevs/turbo_sw.rst| 2 +-
doc/guide
Remove the use of double "the" as it does not
make sense.
Signed-off-by: Sean Morrissey
Signed-off-by: Conor Fogarty
Acked-by: John McNamara
Reviewed-by: Conor Walsh
---
app/test/test_barrier.c | 2 +-
doc/guides/sample_app_ug/keep_alive.rst | 2 +-
drivers/bus/vmbus/rte_bus
The doc's contain references to pmd but the proper
use is to use PMD.
Signed-off-by: Sean Morrissey
Acked-by: John McNamara
Reviewed-by: Conor Walsh
---
doc/guides/cryptodevs/ccp.rst | 2 +-
doc/guides/cryptodevs/openssl.rst | 2 +-
doc/guides/cryptodevs/overview.rst | 2 +-
On Mon, Nov 22, 2021 at 4:21 PM Sean Morrissey wrote:
>
> Remove the use of double "the" as it does not
> make sense.
>
> Signed-off-by: Sean Morrissey
> Signed-off-by: Conor Fogarty
> Acked-by: John McNamara
> Reviewed-by: Conor Walsh
> ---
> app/test/test_barrier.c | 2 +-
>
22/11/2021 19:24, eagost...@nvidia.com:
> From: Elena Agostini
>
> gpudev free and unregister functions return gracefully if input pointer is
> NULL
We could add that the API doc was indicating NULL as a no-op accepted value.
Another explanation to add: cuda driver checks are removed because r
> From: Thomas Monjalon
> Date: Monday, 22 November 2021 at 12:23
> To: Elena Agostini
> Cc: dev@dpdk.org
> Subject: Re: [PATCH v3] gpudev: manage NULL pointer
> External email: Use caution opening links or attachments>
>
> 22/11/2021 19:24, eagost...@nvidia.com:
> > From: Elena Agostini
> >
>
22/11/2021 12:34, Elena Agostini:
> From: Thomas Monjalon
> > 22/11/2021 19:24, eagost...@nvidia.com:
> > > --- a/lib/gpudev/gpudev.c
> > > +++ b/lib/gpudev/gpudev.c
> > > @@ -569,6 +569,9 @@ rte_gpu_mem_free(int16_t dev_id, void *ptr)
> > > {
> > > struct rte_gpu *dev;
> > >
> > > + if
When the event thread polls traffic and a virtq is stopping, the FW loses
synchronization in the virtq indexes.
It causes LM failure on synchronization between the HOST indexes to
the GUEST indexes.
Unset the event thread before the queue stop in the LM process.
Fixes: 31b9c29c86af ("vdpa/mlx5:
On 11/20/2021 1:14 PM, Markus Theil wrote:
From: Markus Theil
Fix kni's ioctl signature to correctly match the kernel's
structs. This shaves off the (void*) casts and uses struct file*
instead of struct inode*. With the correct signature, control flow
integrity checkers are no longer confused a
On Mon, Nov 22, 2021 at 3:17 AM David Marchand
wrote:
> On Fri, Nov 19, 2021 at 5:54 PM Dumitrescu, Cristian
> wrote:
> > On a different point, we should probably tweak our autotests to
> differentiate between logical failures and those failures related to
> resources not being available, and fl
When the ETH spec is empty MLX4 PMD doesn't allow match other criteria,
which means the flow should be promisc one.
Currently, PMD validates this by setting flow->promisc bit when ETH spec
is empty and checking whether there is other rte_flow_item followed
when flow->promisc is on.
However, commi
This patch fixes segfault which was triggered when port, with indirect
actions created, was closed. Segfault was occurring only when
RTE_LIBRTE_MLX5_DEBUG was defined. It was caused by redundant decrement
of RX queues refcount:
- refcount was decremented when port was stopped and indirect actions
If there are sample action and the meter action in the same flow,
mlx5 PMD performs several levels of splitting. For example, sampling
feature splits the original flow into prefix subflow with sample action,
and suffix subflow with the rest of actions. Then, metering feature
splits the sampling suf
From: Elena Agostini
gpudev free and unregister functions return
gracefully if input pointer is NULL because
API doc was indicating NULL as a no-op
accepted value.
cuda driver checks are removed because
redundant with the checks added
in gpudev library.
Fixes: e818c4e2bf50 ("gpudev: add memory
From: Elena Agostini
gpudev free and unregister functions return
gracefully if input pointer is NULL because
API doc was indicating NULL as a no-op
accepted value.
cuda driver checks are removed because
redundant with the checks added
in gpudev library.
Fixes: e818c4e2bf50 ("gpudev: add memory
Apply segmentation offload when requested for non tunneled
packets e.g. IPsec transport mode.
Fixes: 1e728b01120c ("net/iavf: rework Tx path")
Signed-off-by: Radu Nicolau
---
drivers/net/iavf/iavf_rxtx.c | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/drivers/net/iavf
23/11/2021 00:55, eagost...@nvidia.com:
> From: Elena Agostini
>
> gpudev free and unregister functions return
> gracefully if input pointer is NULL because
> API doc was indicating NULL as a no-op
> accepted value.
>
> cuda driver checks are removed because
> redundant with the checks added
> i
From: Elena Agostini
gpudev free and unregister functions return
gracefully if input pointer is NULL because
API doc was indicating NULL as a no-op
accepted value.
cuda driver checks are removed because
redundant with the checks added
in gpudev library.
Fixes: e818c4e2bf50 ("gpudev: add memory
From: Elena Agostini
gpudev free and unregister functions return
gracefully if input pointer is NULL because
API doc was indicating NULL as a no-op
accepted value.
cuda driver checks are removed because
redundant with the checks added
in gpudev library.
Fixes: e818c4e2bf50 ("gpudev: add memory
Fix grammar, spelling and formatting of DPDK 21.11 release notes.
Signed-off-by: John McNamara
---
doc/guides/rel_notes/release_21_11.rst | 123 +
1 file changed, 65 insertions(+), 58 deletions(-)
diff --git a/doc/guides/rel_notes/release_21_11.rst
b/doc/guides/rel_note
Thomas,
Coul you also remove the boilerplate comments and the empty Known Issue section
from the final version.
On Mon, Nov 22, 2021 at 9:01 AM John McNamara wrote:
>
> Fix grammar, spelling and formatting of DPDK 21.11 release notes.
>
> Signed-off-by: John McNamara
Acked-by: Ajit Khaparde
> ---
> doc/guides/rel_notes/release_21_11.rst | 123 +
> 1 file changed, 65 insertions(+
This patch fixes segfault which was triggered when port, with indirect
actions created, was closed. Segfault was occurring only when
RTE_LIBRTE_MLX5_DEBUG was defined. It was caused by redundant decrement
of RX queues refcount:
- refcount was decremented when port was stopped and indirect actions
On 11/8/2021 10:13 AM, Tudor Cornea wrote:
The Kni kthreads seem to be re-scheduled at a granularity of roughly
1 millisecond right now, which seems to be insufficient for performing
tests involving a lot of control plane traffic.
Even if KNI_KTHREAD_RESCHEDULE_INTERVAL is set to 5 microseconds,
When authenticating with SNOW3G, KASUMI and ZUC,
the pointers for encryption/decryption keys is not set.
If a cipher algorithm such as AES-CBC is also used,
the application would seg fault.
Hence, these pointers should be set to some value by default.
Command line to replicate the issue:
./build/a
The cocci script is to help add prefix 'RTE_ETH' namespace to enum
& macro of ethdev library. It helps in automating these changes for
applications. The script won't make changes in the code comment part.
Usage: spatch
On Mon, Nov 22, 2021 at 1:24 PM wrote:
>
> From: Sunil Kumar Kori
>
> Patch updates eventdev and adapters feature matrices for
> cn10k platform.
>
> Signed-off-by: Sunil Kumar Kori
> ---
> v7:
> - Rebased to 21.11 rc3
> - Fix Tx adapter capabilities for SW driver
> v6:
> - Rebased to 21.11 rc
On Mon, Nov 22, 2021 at 1:33 PM wrote:
>
> From: Sunil Kumar Kori
>
> In this commit, check is added for event device, Rx,
> Tx, Crypto and Timer adapters tables for all supported
> drivers.
>
> Signed-off-by: Sunil Kumar Kori
Following errors[2] need to be taken care in
devtools/parse-event-su
examples.dox is built inside builddir and so doxygen generates some
dir_HASH.html that includes the builddir name and this may prevent DPDK
documentation to be correctly generated in some distributions, for
example CentOS Stream 9 and RHEL9, since the builddir includes the
architecture.
This commi
Fixing the mail address of Thomas Monjalon in To.
I used ./devtools/get-maintainer.sh, but it returned the old mail address.
On Mon, Nov 22, 2021 at 9:17 PM Timothy Redaelli wrote:
>
> examples.dox is built inside builddir and so doxygen generates some
> dir_HASH.html that includes the builddir n
Hi Ferruh,
I'm sorry for little delay, I was on vacation.
Currently I don't have a more accurate timeline for the NDK (sw+fw)
availability, but we aim to end of the Q1 2022.
The API of the libnfb should remains unchanged. For the DPDK nfb driver
it will be rather minor changes.
Martin
On Mon,
https://bugs.dpdk.org/show_bug.cgi?id=889
Bug ID: 889
Summary: Farah Eunice Fernandez
Product: DPDK
Version: 21.05
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
Priority: Normal
ENAv2 device requires write combining support which isn't supported by
the upstream vfio-pci. amzn-driver repository provided non-upstream
patch to enable this feature and it was linked directly by the ENA PMD
guide.
To avoid custom kernel patch linking, the user is now guided to the AWS
ENA PMD d
https://bugs.dpdk.org/show_bug.cgi?id=889
David Marchand (david.march...@redhat.com) changed:
What|Removed |Added
URL|https://mas75.com |
CC|
51 matches
Mail list logo
