On Thu, Sep 28, 2023 at 9:41 PM wrote:
>
> From: Pavan Nikhilesh
DMA adapter merge created conflicts, Please rebase to next-event tree.
[for-main]dell[dpdk-next-eventdev] $ git pw series apply 29675
Failed to apply patch:
Applying: eventdev: introduce link profiles
Using index info to reconstr
On Mon, Oct 2, 2023 at 11:36 PM Harry van Haaren
wrote:
>
> This commit changes the logic in the scheduler to always
> reset reorder-buffer (and QID/FID) entries when writing
> them. This avoids stale ROB/QID/FID data re-use, which
> previously caused ordering issues.
>
> Before this commit, relea
On Thu, Sep 28, 2023 at 9:01 PM Bruce Richardson
wrote:
>
> By reducing the iterations for the final stage of the self-test, the SW
> eventdev tests can be fast enough for consideration in the fast-tests
> suite. This enables them to be run as part of the regular patch CI
> tests, and therefore in
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> Sent: Monday, 2 October 2023 20.37
] rte_ether_unformat_addr changes
>
> This patchset makes rte_ether_unformat_addr allow other formats
> for MAC address. Need to remove some inputs from existing
> cmdline_etheraddr test, and add a n
> -Original Message-
> From: Srikanth Yalavarthi
> Sent: Monday, October 2, 2023 3:29 PM
> To: Srikanth Yalavarthi
> Cc: dev@dpdk.org; Shivah Shankar Shankar Narayan Rao
> ; Anup Prabhu ;
> Prince Takkar
> Subject: [PATCH v4 3/3] mldev: drop input and output size get APIs
>
> Drop sup
> -Original Message-
> From: Srikanth Yalavarthi
> Sent: Wednesday, September 20, 2023 12:49 PM
> To: Srikanth Yalavarthi
> Cc: dev@dpdk.org; Shivah Shankar Shankar Narayan Rao
> ; Anup Prabhu ;
> Prince Takkar
> Subject: [PATCH v2 3/3] mldev: drop input and output size get APIs
>
> D
> -Original Message-
> From: Srikanth Yalavarthi
> Sent: Monday, October 2, 2023 3:32 PM
> To: Srikanth Yalavarthi
> Cc: dev@dpdk.org; Shivah Shankar Shankar Narayan Rao
> ; Anup Prabhu ;
> Prince Takkar
> Subject: [PATCH v1 1/1] app/mldev: enable support for pre-quantized I/O
>
> From:
> -Original Message-
> From: Srikanth Yalavarthi
> Sent: Wednesday, August 30, 2023 9:21 PM
> To: Srikanth Yalavarthi
> Cc: dev@dpdk.org; Shivah Shankar Shankar Narayan Rao
> ; Anup Prabhu ;
> Prince Takkar
> Subject: [PATCH v1 1/1] app/mldev: report device not found as error
>
> Repo
On Sat, Sep 30, 2023 at 1:06 AM Amit Prakash Shukla
wrote:
>
> This series adds support for event DMA adapter library. API's defined
> as part of this library can be used by the application for DMA transfer
> of data using event based mechanism.
>
> v8:
> - Re-arranged DMA adapter section in relea
Hi Olivier,
Let me know if you have any comments on this patch.
Regards,
Rakesh
> -Original Message-
> From: Rakesh Kudurumalla
> Sent: Wednesday, July 26, 2023 11:25 AM
> To: Olivier Matz
> Cc: dev@dpdk.org; Jerin Jacob Kollanukkaran ; Nithin
> Kumar Dabilpuram ; Rakesh Kudurumalla
>
This add some basic tests for rte_unformat_ether_addr
and other functions in rte_ether.
Signed-off-by: Stephen Hemminger
---
app/test/meson.build | 1 +
app/test/test_net_ether.c | 169 ++
2 files changed, 170 insertions(+)
create mode 100644 app/test/
This updates rte_ether_addr_unformat() to accept more types
formats for MAC address. It allows IEEE, IETF and Cisco
formats with leading zeros as well.
Signed-off-by: Stephen Hemminger
---
lib/net/rte_ether.c | 87 ++---
lib/net/rte_ether.h | 9 +++--
2 f
Some of the ethernet address formats which were invalid will
now become valid inputs when rte_ether_unformat_addr is modified
to allow leading zeros.
Also, make local variables static.
Signed-off-by: Stephen Hemminger
---
app/test/test_cmdline_etheraddr.c | 11 ---
1 file changed, 4 ins
This patchset makes rte_ether_unformat_addr allow other formats
for MAC address. Need to remove some inputs from existing
cmdline_etheraddr test, and add a new test in test suite
to cover this. There is some overlap between the two tests
but that is fine.
Stephen Hemminger (3):
test: remove som
Currently when adding default route as first route
it is not added and no error is reported.
When we enter the loop over an address range in
the general case we will eventually reach the check
for when the range has ended, and exit the loop.
However when adding default route as first route,
since a
When FIB contains any route covering last ip address
(255.255.255.255), upon adding a new default route
the entire fib table will be overwritten with corresponding
default route next hop.
Previous fix added check for ledge against 0 for case
if default route is added as a first route, however
this
Hi Maxime,
More generally the highest priority is the VRB2 serie:
1. https://patches.dpdk.org/project/dpdk/list/?series=29702
2. https://patches.dpdk.org/project/dpdk/list/?series=29558
3. https://patches.dpdk.org/project/dpdk/list/?series=29537
4. https://patches.dpdk.org/project/dpdk
Hi Maxime,
Kind reminder to review these series for 23.11 in order of priority:
1. https://patches.dpdk.org/project/dpdk/list/?series=29558
2. https://patches.dpdk.org/project/dpdk/list/?series=29537
3. https://patches.dpdk.org/project/dpdk/list/?series=29705
Thanks,
Hernan
On Mon, Oct 02, 2023 at 02:15:51PM +0100, Ferruh Yigit wrote:
> On 10/2/2023 2:01 PM, Bruce Richardson wrote:
> > On Mon, Oct 02, 2023 at 12:48:52PM +, Shibin Koikkara Reeny wrote:
> >> Shared UMEM feature is missing in the af_xdp driver build
> >> after the commit 33d66940e9ba ("build: use C11
On 10/2/2023 2:01 PM, Bruce Richardson wrote:
> On Mon, Oct 02, 2023 at 12:48:52PM +, Shibin Koikkara Reeny wrote:
>> Shared UMEM feature is missing in the af_xdp driver build
>> after the commit 33d66940e9ba ("build: use C11 standard").
>>
>> Runtime Error log while using Shared UMEM feature:
On Mon, Oct 02, 2023 at 12:48:52PM +, Shibin Koikkara Reeny wrote:
> Shared UMEM feature is missing in the af_xdp driver build
> after the commit 33d66940e9ba ("build: use C11 standard").
>
> Runtime Error log while using Shared UMEM feature:
> rte_pmd_af_xdp_probe(): Initializing pmd_af_xdp f
Shared UMEM feature is missing in the af_xdp driver build
after the commit 33d66940e9ba ("build: use C11 standard").
Runtime Error log while using Shared UMEM feature:
rte_pmd_af_xdp_probe(): Initializing pmd_af_xdp for net_af_xdp0
init_internals(): Shared UMEM feature not available. Check kernel
With the meson test suites being reworked to be automatically generated,
rather than having to be hard-coded in the meson.build file, we need to
update our contributing guidelines to take account of this change.
Signed-off-by: Bruce Richardson
---
NOTE: this patch contains only changes needed to
On 2023-09-07 17:24, Stephen Hemminger wrote:
The per-lcore PRNG was not initializing the rand_state of all
the lcores. Any usage of rte_random by a non-EAL lcore would
"/../ by an unregistered non-EAL thread /../"
use rand_states[RTE_MAX_LCORE] which was never initialized.
Fix by using RTE_
On 2023-10-02 10:25, David Marchand wrote:
Hello guys,
On Mon, Aug 28, 2023 at 2:07 PM Anatoly Burakov
wrote:
Currently, the rte_rand() state is initialized with seed, and each
rand state is initialized up until RTE_MAX_LCORE'th rand state. However,
rand state also has one extra rand state re
This commit adds a unit test for an issue identified
where ordered history-list entries are not correctly
cleared when the returned event is of op RELEASE type.
The result of the history-list bug is that a future event
which re-uses that history-list slot, but has an op type
of FORWARD will incorr
This commit changes the logic in the scheduler to always
reset reorder-buffer (and QID/FID) entries when writing
them. This avoids stale ROB/QID/FID data re-use, which
previously caused ordering issues.
Before this commit, release events left the history-list
in an inconsistent state, and future e
On Wed, Sep 06, 2023 at 10:31:00AM +0800, Jieqiang Wang wrote:
> __mm_cmpeq_epi16 returns 0x if the corresponding 16-bit elements are
> equal. In original SSE2 implementation for function compare_signatures,
> it utilizes _mm_movemask_epi8 to create mask from the MSB of each 8-bit
> element, wh
From: Anup Prabhu
Enabled support for pre-quantized input and output in ML
test application.
Signed-off-by: Anup Prabhu
---
Depends-on: series-29710 ("Spec changes to support multi I/O models")
app/test-mldev/ml_options.c| 8
app/test-mldev/ml_options.h| 28 +
> -Original Message-
> From: Jerin Jacob
> Sent: 29 September 2023 09:10
> To: Srikanth Yalavarthi
> Cc: dev@dpdk.org; Shivah Shankar Shankar Narayan Rao
> ; Anup Prabhu ;
> Prince Takkar ; Srikanth Yalavarthi
>
> Subject: [EXT] Re: [PATCH v3 4/4] mldev: update release notes for 23.11
>
Introduce IO layout in ML device specification. IO layout
defines the expected arrangement of model input and output
buffers in the memory. Packed and Split layout support is
added in the specification.
Updated rte_ml_op to support array of rte_ml_buff_seg
pointers to support packed and split I/O
Drop support and use of ML input and output size get functions,
rte_ml_io_input_size_get and rte_ml_io_output_size_get.
These functions are not required, as the model buffer size can
be computed from the fields of updated rte_ml_io_info structure.
Signed-off-by: Srikanth Yalavarthi
---
doc/guid
Updated rte_ml_io_info to support shape of arbitrary
number of dimensions. Dropped use of rte_ml_io_shape
and rte_ml_io_format. Introduced new fields nb_elements
and size in rte_ml_io_info.
Updated drivers and app/mldev to support the changes.
Signed-off-by: Srikanth Yalavarthi
---
app/test-mld
This series implements changes to mldev spec to extend support
for ML models with multiple inputs and outputs. Changes include
introduction of I/O layout to support packed and split buffers
for model input and output. Extended the rte_ml_model_info
structure to support multiple inputs and outputs.
On 9/30/2023 7:31 AM, Andrew Rybchenko wrote:
> On 9/27/23 13:36, Ivan Malov wrote:
>> Packet replay enables users to leverage multiple counters in
>> one flow and allows to request delivery to multiple ports.
>>
>> A given flow rule may use either one inline count action
>> and multiple indirect c
On Fri, Sep 15, 2023 at 1:52 PM Bruce Richardson
wrote:
>
> To help ensure that we don't have "orphaned" tests not in any test
> suites we can add the following checks:
>
> * In developer-mode builds, emit a warning for each test defined using
> REGISTER_TEST_COMMAND
> * In checkpatches, add a c
On Wed, Sep 27, 2023 at 10:27 AM Bruce Richardson
wrote:
>
> On Wed, Sep 27, 2023 at 08:30:05AM +0200, David Marchand wrote:
> > On Tue, Sep 26, 2023 at 5:01 PM Aaron Conole wrote:
> > > David Marchand writes:
> > > > On Tue, Sep 19, 2023 at 10:36 AM Bruce Richardson
> > > > wrote:
> > > >> > >
On Mon, Sep 25, 2023 at 6:27 PM Aaron Conole wrote:
> Bruce Richardson writes:
>
> > When called from automated tools, like meson test, it is often useful to
> > skip tests in a test suite, without having to alter the test build. To
> > do so, we add support for DPDK_TEST_SKIP environment variabl
Hi Konstantin,
I thought your comments make sense, and addressed them with a patchset
https://patches.dpdk.org/project/dpdk/list/?series=29709,
please check again.
Thanks,
Trevor Tao
At 2023-09-26 21:49:05, "Konstantin Ananyev"
wrote:
>
>Hi Trevor,
>>>
>>>
At 2023-09-18 02:04:19,
> From: David Marchand [mailto:david.march...@redhat.com]
> Sent: Monday, 2 October 2023 10.26
>
> Hello guys,
>
> On Mon, Aug 28, 2023 at 2:07 PM Anatoly Burakov
> wrote:
> >
> > Currently, the rte_rand() state is initialized with seed, and each
> > rand state is initialized up until RTE_MAX_LC
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> Sent: Thursday, 7 September 2023 17.25
>
> The per-lcore PRNG was not initializing the rand_state of all
> the lcores. Any usage of rte_random by a non-EAL lcore would
> use rand_states[RTE_MAX_LCORE] which was never initialized.
>
>
Now the port Rx mq_mode had been set to RTE_ETH_MQ_RX_RSS
by default, but some hw and/or virtual interface does not
support the RSS and offload mode presupposed, e.g., some
virtio interfaces in the cloud don't support
RSS and the error msg may like:
virtio_dev_configure(): RSS support requested bu
Add an option to enable the RX mode requirement relax
in release notes and l3fwd sample guide.
Signed-off-by: Trevor Tao
---
doc/guides/rel_notes/release_23_11.rst | 2 ++
doc/guides/sample_app_ug/l3_forward.rst | 4 +++-
2 files changed, 5 insertions(+), 1 deletion(-)
diff --git a/doc/guides/
Now the port Rx offload mode is set to RTE_ETH_RX_OFFLOAD_CHECKSUM
by default, but some hw and/or virtual interface does not support
the offload mode presupposed, e.g., some virtio interfaces in
the cloud may only partly support RTE_ETH_RX_OFFLOAD_UDP_CKSUM/
RTE_ETH_RX_OFFLOAD_TCP_CKSUM,
but not RT
https://bugs.dpdk.org/show_bug.cgi?id=1276
David Marchand (david.march...@redhat.com) changed:
What|Removed |Added
Status|CONFIRMED |RESOLVED
> From: David Marchand [mailto:david.march...@redhat.com]
> Sent: Monday, 2 October 2023 09.34
>
> On Fri, Aug 4, 2023 at 6:16 PM Stephen Hemminger
> wrote:
> >
> > The ring used to store mbufs needs to be multiple producer,
> > multiple consumer because multiple queues might on multiple
> > core
On Mon, Jul 24, 2023 at 3:37 PM Tomasz Duszynski wrote:
>
> Use proper naming when dealing with a raw device.
>
> Fixes: c88b3f2558ed ("rawdev: introduce raw device library")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Tomasz Duszynski
Acked-by: Hemant Agrawal
Applied, thanks.
--
David Marchand
On Thu, Aug 24, 2023 at 12:52 PM Tomasz Duszynski
wrote:
>
> Remove device from the list before doing actual cleanup to avoid use
> after free.
>
> Bugzilla ID: 1276
> Fixes: 17c839f74da3 ("bus: add platform bus")
>
> Signed-off-by: Tomasz Duszynski
Applied, thanks.
--
David Marchand
Hello guys,
On Mon, Aug 28, 2023 at 2:07 PM Anatoly Burakov
wrote:
>
> Currently, the rte_rand() state is initialized with seed, and each
> rand state is initialized up until RTE_MAX_LCORE'th rand state. However,
> rand state also has one extra rand state reserved for non-EAL threads,
> which is
Hello Stephen,
On Thu, Sep 21, 2023 at 6:24 AM Stephen Hemminger
wrote:
>
> The computation of timestamp is more easily done in pdump
> than pcapng. The initialization is easier and makes the pcapng
> library have no global state.
>
> It also makes it easier to add HW timestamp support later.
>
>
On Fri, Aug 4, 2023 at 6:16 PM Stephen Hemminger
wrote:
>
> The ring used to store mbufs needs to be multiple producer,
> multiple consumer because multiple queues might on multiple
> cores might be allocating and same time (consume) and in
> case of ring full, the mbufs will be returned (multiple
51 matches
Mail list logo
