> -Original Message-
> From: Liu, Yong
> Sent: Wednesday, April 7, 2021 11:25 AM
> To: maxime.coque...@redhat.com; Xia, Chenbo
> Cc: dev@dpdk.org; Liu, Yong ; sta...@dpdk.org
> Subject: [PATCH 1/2] vhost: fix accessing uninitialized variables
>
> This patch fixs coverity issue by adding
Hi Thomas,
Thanks for the review.
On Tue, Apr 6, 2021 at 5:24 PM Thomas Monjalon wrote:
> 26/03/2021 11:47, Michal Krawczyk:
> > From: Stanislaw Kardach
> >
> > rte_timer_subsystem_init() may return -EALREADY if the timer subsystem
> > was already initialized. This can happen i.e. in PMD code
On Thu, Apr 1, 2021 at 6:39 PM Burakov, Anatoly
wrote:
>
> On 25-Mar-21 1:32 PM, David Marchand wrote:
> > Hello,
> >
> > On Mon, Apr 20, 2020 at 4:13 PM Li Feng wrote:
> >>
> >> Cool, thank you, Anatoly and Kozlyuk.
> >>
> >> I haven't found how Windows implements the rte_mem_virt2phy.
> >>
> >>
From: Smadar Fuks
This patch updates release notes for added support of
OCTEON TX2 flow action to port id
Signed-off-by: Smadar Fuks
---
doc/guides/rel_notes/release_21_05.rst | 4
1 file changed, 4 insertions(+)
diff --git a/doc/guides/rel_notes/release_21_05.rst
b/doc/guides/rel_notes
> -Original Message-
> From: dev On Behalf Of Matan Azrad
> Sent: Thursday, March 18, 2021 5:13 PM
> To: dev@dpdk.org
> Cc: Maxime Coquelin
> Subject: [dpdk-dev] [PATCH] vdpa/mlx5: improve interrupt management
>
> The driver should notify the guest for each traffic burst detected by CQ
>
If no offload flags are set, then 0 == 0 & is true.
This means that if DEV_TX_OFFLOAD_MBUF_FAST_FREE is not
set (and no other flags are set) then this check will still
enable `tx_simple_allowed`.
This patch adds the 'offloads' flag not equal zero to avoid this
scenario.
Bugzilla ID: 665
Fixes: 3
Hi Cheng,
Some comments are inline.
> -Original Message-
> From: Jiang, Cheng1
> Sent: Wednesday, March 31, 2021 10:06 PM
> To: maxime.coque...@redhat.com; Xia, Chenbo
> Cc: dev@dpdk.org; Hu, Jiayu ; Yang, YvonneX
> ; Wang, Yinan ; Jiang,
> Cheng1
> Subject: [PATCH v3] vhost: add suppo
> -Original Message-
> From: Liu, Yong
> Sent: Wednesday, April 7, 2021 11:25 AM
> To: maxime.coque...@redhat.com; Xia, Chenbo
> Cc: dev@dpdk.org; Liu, Yong ; sta...@dpdk.org
> Subject: [PATCH 2/2] vhost: fix async enqueue accessing uninitialized
> variables
>
> This patch fixs coverity
Hi,
> -Original Message-
> From: Ajit Khaparde
> Sent: Wednesday, April 7, 2021 11:16 AM
> To: NBU-Contact-Thomas Monjalon
> Cc: Bing Zhao ; Andrew Rybchenko
> ; Ori Kam ;
> ferruh.yi...@intel.com; dev@dpdk.org
> Subject: Re: [dpdk-dev] [RFC] ethdev: introduce action context APIs
>
> On
Add "Flow dump" in features/default.ini and features/mlx5.ini.
Add testpmd CLI and API changes in release_notes.
Signed-off-by: Haifei Luo
---
doc/guides/nics/features/default.ini | 1 +
doc/guides/nics/features/mlx5.ini | 1 +
doc/guides/rel_notes/release_21_05.rst | 5 -
3 files cha
add mlx5 APIs for single flow dump feature
Signed-off-by: Haifei Luo
---
drivers/common/mlx5/linux/meson.build | 2 ++
drivers/common/mlx5/linux/mlx5_glue.c | 13 +
drivers/common/mlx5/linux/mlx5_glue.h | 1 +
drivers/common/mlx5/mlx5_devx_cmds.c
Modify API mlx5_flow_dev_dump to support the feature.
Modify mlx5_socket since one extra arg flow_ptr is added.
Signed-off-by: Haifei Luo
---
drivers/net/mlx5/linux/mlx5_socket.c | 30 --
drivers/net/mlx5/mlx5_flow.c | 29 +++--
2 files
Add support for single flow dump.
The CLIs to dump one rule: flow dump PORT rule ID
to dump all: flow dump PORT all
Examples:
testpmd> flow dump 0 all
testpmd> flow dump 0 rule 0
Signed-off-by: Haifei Luo
---
app/test-pmd/cmdline_flow.c | 55 +
app/tes
Previous implementations support dump all the flows. Add new arg
rte_flow in rte_flow_dev_dump to dump one flow.
Signed-off-by: Haifei Luo
---
app/test-pmd/config.c| 2 +-
doc/guides/nics/mlx5.rst | 9 ++---
doc/guides/prog_guide/rte_flow.rst | 24
Dump internal representation information of all flows is supported.
It is useful to dump one flow. To implement this requirement,
add this CLI to dump one rule: flow dump PORT rule ID
and the CLI to dump all: flow dump PORT all
Examples:
testpmd> flow dump 0 all
testpmd> flow dump 0 r
Enable GTPU pattern for CVL switch filter. Support teid and
qfi field of GTPU pattern. Patterns without inner l3/l4 field
support outer dst/src ip. Patterns with inner l3/l4 field only
support inner dst/src ip and inner dst/src port.
+---+---
This patch fixs coverity issue in async enqueue function by adding
initialization step before using temporary virtio header.
Coverity issue: 366123
Fixes: cd6760da1076 ("vhost: introduce async enqueue for split ring")
Cc: sta...@dpdk.org
Signed-off-by: Marvin Liu
diff --git a/lib/librte_vhost/v
This patch fixs coverity issue by adding initialization step before
using temporary virtio header.
Coverity issue: 366181
Fixes: fb3815cc614d ("vhost: handle virtually non-contiguous buffers in Rx-mrg")
Cc: sta...@dpdk.org
Signed-off-by: Marvin Liu
Reviewed-by: Maxime Coquelin
diff --git a/lib
> -Original Message-
> From: Xia, Chenbo
> Sent: Tuesday, April 6, 2021 2:18 PM
> To: Liu, Yong ; maxime.coque...@redhat.com
> Cc: dev@dpdk.org; sta...@dpdk.org
> Subject: RE: [PATCH] vhost: fix accessing uninitialized variables
>
> Hi Marvin,
>
> > -Original Message-
> > From:
Hello,
> -Original Message-
> From: Ajit Khaparde
> Sent: Wednesday, March 24, 2021 7:27 AM
> To: Bing Zhao
> Cc: Ori Kam ; NBU-Contact-Thomas Monjalon
> ; Ferruh Yigit ; Andrew
> Rybchenko ; dpdk-dev
> Subject: Re: [dpdk-dev] [RFC] ethdev: introduce conntrack flow
> action and item
>
For platforms that don't support turbo boost,rte_power_turbo_status()
returns "-ENOTSUP" (like power_kvm_vm_turbo_status()). So don't allow
check_power_turbo() to continue if
rte_power_turbo_status(TEST_POWER_LCORE_ID)!=1
Fixes: aeaeaf5f2d62 ("test/power: add cases for turbo feature")
Cc: lukaszx
Hi
> -Original Message-
> From: Yu, DapengX
> Sent: Thursday, April 1, 2021 16:29
> To: Li, Xiaoyun ; Zhang, Qi Z
> Cc: dev@dpdk.org; Yu, DapengX ; sta...@dpdk.org
> Subject: [PATCH] app/testpmd: fix queue Rx and Tx offload reconfig cmd
>
> From: Dapeng Yu
>
> Configure per queue rx o
> >>
> >> 在 2021/3/25 22:40, Thomas Monjalon 写道:
> >>> 25/03/2021 14:26, oulijun:
> 在 2021/3/25 20:19, Thomas Monjalon 写道:
> > 25/03/2021 12:17, oulijun:
> >> Ruifeng has reviewed-by. Does anyone else have any comments? Is
> >> it necessary to send the V2?
> >
> > v2 for
Looks good to me. Thank you for fixing this bug for us.
Reviewed-by: Alexander Kozyrev
> -Original Message-
> From: dev On Behalf Of Feifei Wang
> Sent: Thursday, March 18, 2021 3:19
> Cc: dev@dpdk.org; n...@arm.com; Feifei Wang
> Subject: [dpdk-dev] [PATCH v1 0/4] refactor SMP barrier
Mellanox hardware can only modify any packet field in 32-bit chunks,
which means 4 such chunks are needed to modify an IPv6 address.
The modification order of these chunks starts from the most significant
bits for the IPv6 address. That leads to confusing results when trying
to modify either source
Extend the range of immediate value used in the MODIFY_FIELD action
from 32 to 64 bits to conform to the rte_flow_action_modify_data spec.
Apply appropriate big endian conversion to the immediate value
according to a destination field bit width.
Fixes: 641dbe4fb053 ("net/mlx5: support modify field
The MODIFY_FIELD RTE action requires the extended metadata support
in order to manipulate on METADATA register as well as on MARK register.
Check if it is supported and reject the MODIFY_FIELD action if it is not
just like it was done before for the MARK register modifications.
Fixes: 1fc5eded4c (
> -Original Message-
> From: Thomas Monjalon
> Sent: Wednesday, April 7, 2021 12:23 AM
> To: John Hurley ; Suanming Mou
>
> Cc: Ori Kam ; dev@dpdk.org; Slava Ovsiienko
> ; Matan Azrad ; Raslan
> Darawsheh
> Subject: Re: [dpdk-dev] [PATCH v3 4/4] regex/mlx5: prevent wrong calculation
>
Hi Thomas,
I agree with the "ifcvf" name, as there might be more than one driver in a
single driver dir.
Thanks,
Xiao
> -Original Message-
> From: Xu, Rosen
> Sent: Tuesday, April 6, 2021 5:48 PM
> To: Thomas Monjalon ; Wang, Xiao W
>
> Cc: dev@dpdk.org; Richardson, Bruce ;
> Hemant A
>
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Tom Barbette
> > Sent: Wednesday, March 31, 2021 10:53 AM
> >
> > Le 31-03-21 à 02:44, Honnappa Nagarahalli a écrit :
> > > - Ability to tune the values of #defines
> > > * Few prominent points discussed
> > > - This will result i
> Subject: Re: [PATCH v5] pflock: implementation of phase-fair reader writer
> locks
>
> On Tue, 6 Apr 2021 21:56:05 +
> Honnappa Nagarahalli wrote:
>
> > How does the autotest script does this?
>
> This was copy-paste from original. Think the original was wrong here.
Ok, would it be poss
2021-04-06 16:34 (UTC+0200), Morten Brørup:
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Dmitry Kozlyuk
> > Sent: Sunday, April 4, 2021 1:41 AM
> >
> > POSIX sleep(3) is missing from Windows.
> > Add generic rte_thread_sleep() to suspend current OS thread.
> >
> > Signed-off-by: Dmitr
06/04/2021 22:09, David Marchand:
> On Tue, Apr 6, 2021 at 8:53 PM Thomas Monjalon wrote:
> >
> > About the title, it is not a single file but one file per lib/driver.
> > I think it would better convey the idea by saying "remove .def files".
>
> Ok.
>
> >
> > 06/04/2021 19:59, David Marchand:
>
On Tue, 6 Apr 2021 21:56:05 +
Honnappa Nagarahalli wrote:
> How does the autotest script does this?
This was copy-paste from original. Think the original was wrong here.
>
> This is a new type of reader-writer lock that provides better fairness
> guarantees which better suited for typical DPDK applications.
> A pflock has two ticket pools, one for readers and one for writers.
>
> Phase fair reader writer locks ensure that neither reader nor writer will be
> sta
>
> The creation of control threads used a pthread barrier for synchronization.
> This patch fixes a race condition where the pthread barrier could get
> destroyed while one of the threads has not yet returned from the
> pthread_barrier_wait function, which could result in undefined behaviour.
>
On Tue, Apr 6, 2021 at 8:53 PM Thomas Monjalon wrote:
>
> About the title, it is not a single file but one file per lib/driver.
> I think it would better convey the idea by saying "remove .def files".
Ok.
>
> 06/04/2021 19:59, David Marchand:
> > Rather than have two files that keeps getting out
This patch adds tests for predictable RSS feature
Signed-off-by: Vladimir Medvedkin
---
app/test/test_thash.c | 468 +-
1 file changed, 462 insertions(+), 6 deletions(-)
diff --git a/app/test/test_thash.c b/app/test/test_thash.c
index a6aadd1..28a
This patch implements predictable RSS functionality.
Signed-off-by: Vladimir Medvedkin
---
lib/librte_hash/rte_thash.c | 577 ++--
lib/librte_hash/rte_thash.h | 42
lib/librte_hash/version.map | 1 +
3 files changed, 602 insertions(+), 18 deletions
This patch adds predictable RSS API.
It is based on the idea of searching partial Toeplitz hash collisions.
Signed-off-by: Vladimir Medvedkin
---
lib/librte_hash/meson.build | 3 +-
lib/librte_hash/rte_thash.c | 96 ++
lib/librte_hash/rte_thash.h | 138
This patch series introduces predictable RSS feature.
It is based on the idea of searching for partial hash collisions
within Toeplitz hash.
The Toeplitz hash function is a homomorphism between (G, ^) and (H, ^),
where (G, ^) - is a group of tuples and (H, ^) is a group of hashes
with respect to X
About the title, it is not a single file but one file per lib/driver.
I think it would better convey the idea by saying "remove .def files".
06/04/2021 19:59, David Marchand:
> Rather than have two files that keeps getting out of sync, let's
> annotate the version.map to generate the Windows expor
Rather than have two files that keeps getting out of sync, let's
annotate the version.map to generate the Windows export file.
Note: Existing version.map annotation achieved with:
$ for dir in lib/librte_eal drivers/common/mlx5; do
./buildtools/map-list-symbol.sh $dir/*.map |
while read fi
On 4/6/2021 5:40 PM, Ferruh Yigit wrote:
On 4/2/2021 2:26 AM, Pallavi Kadam wrote:
This patch-set enables building the Intel ice PMD on Windows.
v3 changes:
- Rebase on latest head
- Replaced 'unsigned long long' with 'size_t'
- Moved cflag '-fno-asynchronous-unwind-tables' above th
On Tue, Apr 6, 2021 at 6:51 PM Bruce Richardson
wrote:
> > -# special case, allow override if an def file already exists alongside map
> > file
> > -override_file = join(dirname(args[1]), basename(args[2]))
> > -if exists(override_file):
> > -with open(override_file) as f_in:
> >
On Tue, Apr 06, 2021 at 06:32:30PM +0200, David Marchand wrote:
> Rather than have two files that keeps getting out of sync, let's
> annotate the version.map to generate the Windows export file.
>
> Note: EAL version.map annotation achieved with:
> $ ./buildtools/map-list-symbol.sh lib/librte_eal/
Hello Bill,
>-Original Message-
>From: dev On Behalf Of Dong Zhou
>Sent: Thursday, April 1, 2021 4:23 PM
>To: Jiawei(Jonny) Wang ; Raslan Darawsheh
>
>Cc: dev@dpdk.org
>Subject: [dpdk-dev] [PATCH] doc: update push/pop VLAN support in mlx5
>guide
>
>Updates the documentation for push/pop V
On Tue, Apr 6, 2021 at 6:33 PM David Marchand wrote:
> diff --git a/buildtools/map_to_win.py b/buildtools/map_to_win.py
> index 2a6cb88605..34434d96fa 100644
> --- a/buildtools/map_to_win.py
> +++ b/buildtools/map_to_win.py
> @@ -7,7 +7,7 @@
>
>
> def is_function_line(ln):
> -return ln.starts
On 4/2/2021 2:26 AM, Pallavi Kadam wrote:
This patch-set enables building the Intel ice PMD on Windows.
v3 changes:
- Rebase on latest head
- Replaced 'unsigned long long' with 'size_t'
- Moved cflag '-fno-asynchronous-unwind-tables' above the assignment
"c_args
On Tue, Apr 6, 2021 at 6:04 PM Thomas Monjalon wrote:
>
> 06/04/2021 15:11, Thomas Monjalon:
> > --- a/lib/librte_eal/version.map
> > +++ b/lib/librte_eal/version.map
> > # added in 21.05
> > + rte_log_list_types;
>
> The symbol should be added in the .def as well.
>
Yes, and I'm fed up
Rather than have two files that keeps getting out of sync, let's
annotate the version.map to generate the Windows export file.
Note: EAL version.map annotation achieved with:
$ ./buildtools/map-list-symbol.sh lib/librte_eal/version.map |
while read file version sym; do
! git grep -qw $sym li
On 4/6/2021 1:57 AM, Min Hu (Connor) wrote:
According to the suggestion of our legal department,
to standardize the copyright license of our code to
avoid potential copyright risks, we make a unified
modification to the "Hisilicon", which was nonstandard,
in the main modules we maintain.
We chan
30/03/2021 03:39, Suanming Mou:
> From: John Hurley
>
> A recent change adds support for scattered mbuf and UMR support for regex.
> Part of this commit makes the pi and ci counters of the regex_sq a quarter
> of the length in non umr mode, effectively moving them from 16 bits to
> 14. The new ge
Instead of calculating the address of a packed descriptor based on the
vq->desc_packed and vq->last_used_idx every time, store that base
address in desc_base. On arm, this saves 176 bytes in code size of
function in which vhost_flush_enqueue_batch_packed gets inlined.
---
lib/librte_vhost/virtio_n
The creation of control threads used a pthread barrier for
synchronization. This patch fixes a race condition where the pthread
barrier could get destroyed while one of the threads has not yet
returned from the pthread_barrier_wait function, which could result in
undefined behaviour. The barrier ha
On 4/4/2021 8:45 AM, Raslan Darawsheh wrote:
qfi field is 8 bits which represent single bit for
PPP (paging Policy Presence) single bit for RQI
(Reflective QoS Indicator) and 6 bits for qfi
(QoS Flow Identifier) based on RFC 38415-g30
This update the doxygen format and the mask for qfi
to proper
06/04/2021 15:11, Thomas Monjalon:
> --- a/lib/librte_eal/version.map
> +++ b/lib/librte_eal/version.map
> # added in 21.05
> + rte_log_list_types;
The symbol should be added in the .def as well.
Hi Olivier,
Hi Honnappa,
Thanks for your input Honnappa. I've made the changes to
completely remove the barrier. However, I have not moved the
call to pthread_setaffinity_np to the control thread; I
think we still want to report the result of this function
to the caller of rte_ctrl_thread_create
There is no need to pass a pointer to an mbuf pointer.
Signed-off-by: Balazs Nemeth
---
lib/librte_vhost/virtio_net.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index e1696c0c0..179c57b46 100644
---
Now that all allocation and freeing has been moved together, use the
faster bulk versions instead of handling packets one by one.
Signed-off-by: Balazs Nemeth
---
lib/librte_vhost/virtio_net.c | 10 ++
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/lib/librte_vhost/virtio_
Move allocation out further and perform all allocation in one loop. The
same goes for freeing packets. This is to prepare for use of bulk
versions of these functions.
Signed-off-by: Balazs Nemeth
---
lib/librte_vhost/virtio_net.c | 18 ++
1 file changed, 6 insertions(+), 12 delet
In preparation for subsequent patches, move mbuf allocation out and
rename virtio_dev_pktmbuf_alloc to virtio_dev_pktmbuf_prep. This
function now receives an already allocated mbuf pointer.
Signed-off-by: Balazs Nemeth
---
lib/librte_vhost/virtio_net.c | 54 ++-
1
Use the faster bulk versions of alloc and free in virtio packed path.
Balazs Nemeth (4):
vhost: move allocation of mbuf outside of packet enqueue
vhost: perform all mbuf allocations in one loop
vhost: allocate and free packets in bulk
vhost: remove unnecessary level of indirection
lib/li
06/04/2021 17:31, Stanisław Kardach:
> Hi Thomas,
>
> Thanks for the review.
>
> On Tue, Apr 6, 2021 at 5:24 PM Thomas Monjalon wrote:
>
> > 26/03/2021 11:47, Michal Krawczyk:
> > > From: Stanislaw Kardach
> > >
> > > rte_timer_subsystem_init() may return -EALREADY if the timer subsystem
> > >
26/03/2021 11:47, Michal Krawczyk:
> From: Stanislaw Kardach
>
> rte_timer_subsystem_init() may return -EALREADY if the timer subsystem
> was already initialized. This can happen i.e. in PMD code (see
> eth_ena_dev_init). This is not an error, rather a notification as the
> initialization functio
Add the implementation for Marvell CN10k mempool batch dequeue op.
Signed-off-by: Ashwin Sekhar T K
---
drivers/mempool/cnxk/cn10k_mempool_ops.c | 72 +++-
1 file changed, 71 insertions(+), 1 deletion(-)
diff --git a/drivers/mempool/cnxk/cn10k_mempool_ops.c
b/drivers/mempoo
Add the implementation for Marvell CN10k get count op.
Signed-off-by: Ashwin Sekhar T K
---
drivers/mempool/cnxk/cn10k_mempool_ops.c | 28 +++-
1 file changed, 27 insertions(+), 1 deletion(-)
diff --git a/drivers/mempool/cnxk/cn10k_mempool_ops.c
b/drivers/mempool/cnxk/cn10k
Add the implementation for Marvell CN10k mempool batch enqueue op.
Signed-off-by: Ashwin Sekhar T K
---
drivers/mempool/cnxk/cn10k_mempool_ops.c | 28 +++-
1 file changed, 27 insertions(+), 1 deletion(-)
diff --git a/drivers/mempool/cnxk/cn10k_mempool_ops.c
b/drivers/mempoo
Marvell CN10k mempool supports batch enqueue/dequeue which can
dequeue up to 512 pointers and enqueue up to 15 pointers using
a single instruction.
These batch operations require a DMA memory to enqueue/dequeue
pointers. This patch adds the initialization of this DMA memory.
Signed-off-by: Ashwin
Add Marvell CN10k mempool ops and implement CN10k mempool alloc.
CN10k has 64 bytes L1D cache line size. Hence the CN10k mempool
alloc does not make the element size an odd multiple L1D cache
line size as NPA requires the element sizes to be multiples of
128 bytes.
Signed-off-by: Ashwin Sekhar T
Add Marvell CN9k mempool enqueue/dequeue. Marvell CN9k
supports burst dequeue which allows to dequeue up to 32
pointers using pipelined casp instructions.
Signed-off-by: Pavan Nikhilesh
Signed-off-by: Ashwin Sekhar T K
---
doc/guides/mempool/cnxk.rst | 4 +++
drivers/mempool/cnxk/c
Add Marvell CN9k mempool ops and implement CN9k mempool
alloc which makes sure that the element size always occupy
odd number of cachelines to ensure even distribution among
of elements among L1D cache sets.
Signed-off-by: Pavan Nikhilesh
Signed-off-by: Ashwin Sekhar T K
---
drivers/mempool/cnx
Register the CNXk mempool plt init callback which will set the
appropriate mempool ops to be used for the platform.
Signed-off-by: Ashwin Sekhar T K
---
drivers/mempool/cnxk/cnxk_mempool_ops.c | 15 +++
1 file changed, 15 insertions(+)
diff --git a/drivers/mempool/cnxk/cnxk_mempool_
Add generic CNXk mempool ops which will enqueue/dequeue
from pool one element at a time.
Signed-off-by: Pavan Nikhilesh
Signed-off-by: Ashwin Sekhar T K
---
drivers/mempool/cnxk/cnxk_mempool.h | 26
drivers/mempool/cnxk/cnxk_mempool_ops.c | 171
drivers/mempoo
Add the implementation for CNXk mempool device
probe and remove.
Signed-off-by: Pavan Nikhilesh
Signed-off-by: Ashwin Sekhar T K
---
doc/guides/mempool/cnxk.rst | 23 +
drivers/mempool/cnxk/cnxk_mempool.c | 132 +++-
2 files changed, 151 insertions(+), 4 del
Add the meson based build infrastructure for Marvell
CNXK mempool driver along with stub implementations
for mempool device probe.
Also add Marvell CNXK mempool base documentation.
Signed-off-by: Pavan Nikhilesh
Signed-off-by: Jerin Jacob
Signed-off-by: Nithin Dabilpuram
Signed-off-by: Ashwin
This patchset adds the mempool/cnxk driver which provides the support for the
integrated mempool device found in Marvell CN10K SoC.
The code includes mempool driver functionality for Marvell CN9K SoC as well,
but right now it is not enabled. The future plan is to deprecate existing
mempool/octeont
18/02/2021 19:45, Ferruh Yigit:
> On 2/18/2021 5:58 PM, Ajit Khaparde wrote:
> > On Wed, Feb 17, 2021 at 2:49 AM Thomas Monjalon wrote:
> >> 17/02/2021 11:37, Ferruh Yigit:
> >>> On 2/17/2021 5:57 AM, Asaf Penso wrote:
> From: Ferruh Yigit
> > On 2/7/2021 10:52 AM, Asaf Penso wrote:
> >>
Hi Abhinandan,
Please see inline.
Thanks,
Anoob
> >
> > Advertise crypto adapter forward mode capability and set crypto
> > adapter enqueue function in driver.
> >
> > Signed-off-by: Shijith Thotton
[snip]
> > +
> > + if (!ev->sched_type)
> > + otx2_ssogws_head_wait(tag_op);
> > +
06/04/2021 16:28, David Marchand:
> On Tue, Apr 6, 2021 at 3:12 PM Thomas Monjalon wrote:
> > +void
> > +rte_log_list_types(FILE *out, const char *prefix)
> > +{
> > + struct rte_log_dynamic_type *sorted_types;
> > + const size_t type_size = sizeof(rte_logs.dynamic_types[0]);
> > +
On 4/4/2021 10:49 AM, Salem Sol wrote:
From: Jiawei Wang
With the current code the VXLAN/NVGRE parsing routine
stored the configuration of the header on stack, this
might lead to overwriting the data on the stack.
This patch stores the external data of vxlan and nvgre encap
into global data as
On Tue, Apr 06, 2021 at 10:20:37AM -0400, Aaron Conole wrote:
> Bruce Richardson writes:
>
> > On Tue, Apr 06, 2021 at 08:33:07AM -0400, Aaron Conole wrote:
> >> Thomas Monjalon writes:
> >>
> >> > 17/03/2021 15:44, Aaron Conole:
> >> >> The hugepage test really needs to check multiple things o
Avoid race with unregister interrupt handler if interrupt
source has some active callbacks at the moment, use wrapper
around rte_intr_callback_unregister() to check for -EAGAIN
return value and to loop until rte_intr_callback_unregister()
succeeds.
Signed-off-by: Renata Saiakhova
Acked-by: Anatol
For pci_vfio, while removing the device by rte_dev_remove(),
pci_vfio_disable_notifier() will call rte_intr_callback_unregister(),
which may return -EAGAIN, if an interrupt source (kernel) has some active
callbacks right now. As a result, the callback (req notifier) can be never
unregistered,
and
On 4/4/2021 10:49 AM, Salem Sol wrote:
Updates the documentation with the added support for sample actions VXLAN
and NVGRE encap in E-Switch steering flow.
Signed-off-by: Salem Sol
---
doc/guides/nics/mlx5.rst | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/doc/guides
From: Satheesh Paul
Added support for allocating RSS group and setting
it as action of an NPC rule for rte_flow.
Signed-off-by: Satheesh Paul
---
drivers/common/cnxk/roc_npc.c | 159 +
drivers/common/cnxk/roc_npc.h | 20 +
drivers/common/cnxk/
From: Pavan Nikhilesh
Add TIM LF IRQ register and un-register functions.
Signed-off-by: Pavan Nikhilesh
---
drivers/common/cnxk/meson.build| 1 +
drivers/common/cnxk/roc_tim.c | 51 ++
drivers/common/cnxk/roc_tim_irq.c | 104 +
d
From: Pavan Nikhilesh
Add TIM device init, fini which are used to attach TIM LF
resources to the RVU PF/VF and TIM LF alloc and free.
Signed-off-by: Pavan Nikhilesh
---
drivers/common/cnxk/meson.build| 1 +
drivers/common/cnxk/roc_api.h | 3 +
drivers/common/cnxk/roc_platform.c |
From: Pavan Nikhilesh
Add sso debug dump support. This dumps all SSO LF register values
to a given file handle.
Signed-off-by: Pavan Nikhilesh
---
drivers/common/cnxk/meson.build | 1 +
drivers/common/cnxk/roc_sso.h | 4 +++
drivers/common/cnxk/roc_sso_debug.c | 68
From: Pavan Nikhilesh
Add support to registering and un-registering SSO HWS and
HWGRP IRQs.
Signed-off-by: Pavan Nikhilesh
---
drivers/common/cnxk/meson.build| 1 +
drivers/common/cnxk/roc_sso.c | 41 ++
drivers/common/cnxk/roc_sso_irq.c | 164 +
From: Pavan Nikhilesh
Add SSO HWGRP interface for configuring XAQ pool, setting priority
and internal HW buffer limits for each HWGRP.
Signed-off-by: Pavan Nikhilesh
---
drivers/common/cnxk/roc_sso.c | 110
drivers/common/cnxk/roc_sso.h | 21 ++
On 4/1/2021 5:13 AM, Jiawei(Jonny) Wang wrote:
Hello Ferruh,
-Original Message-
From: Ferruh Yigit
Sent: Wednesday, March 31, 2021 8:08 PM
To: Salem Sol ; dev@dpdk.org
Cc: Jiawei(Jonny) Wang ; Ori Kam ;
Xiaoyun Li
Subject: Re: [dpdk-dev] [PATCH v3 1/8] app/testpmd: store VXLAN/NVGRE
e
From: Pavan Nikhilesh
Add SSO HWS interface for setting/unsetting links, retrieving
base address and nanoseconds to getwork timeout.
Signed-off-by: Pavan Nikhilesh
---
drivers/common/cnxk/roc_sso.c | 128 -
drivers/common/cnxk/roc_sso.h | 6 ++
d
From: Pavan Nikhilesh
Add SSO device init and fini which attach SSO LF resources to the
RVU PF/VF and SSO HWS and HWGRP LFs alloc, free.
Signed-off-by: Pavan Nikhilesh
---
drivers/common/cnxk/meson.build | 1 +
drivers/common/cnxk/roc_api.h | 3 +
drivers/common/cnxk/roc_idev.c
From: Kiran Kumar K
Adding support initialize and fini the npc. Further, adding APIs to
create and destroy the npc rules.
Signed-off-by: Kiran Kumar K
---
drivers/common/cnxk/meson.build | 1 +
drivers/common/cnxk/roc_npc.c | 713
drivers/common/cnx
From: Kiran Kumar K
Adding npc parsing API support to parse different patterns and actions.
Based on the pattern and actions ltype values will be chosen and
mcam data will be configured at perticular offsets.
Signed-off-by: Kiran Kumar K
---
drivers/common/cnxk/meson.build | 1 +
drivers
From: Kiran Kumar K
Adding mcam utility functions like reading KEX and reserving and writing
mcam rules.
Signed-off-by: Kiran Kumar K
---
drivers/common/cnxk/meson.build| 1 +
drivers/common/cnxk/roc_npc_mcam.c | 708 +
drivers/common/cnxk/roc_npc_priv
From: Kiran Kumar K
Adding NPC helper APIs to manage MCAM like pre allocating the mcam,
configuring the rules, shifting mcam rules and preparing the data for
mcam based on KEX.
Signed-off-by: Kiran Kumar K
---
drivers/common/cnxk/meson.build | 1 +
drivers/common/cnxk/roc_npc_priv.h |
From: Kiran Kumar K
Adding initial support for programming NPC. NPC is Network Parser
and CAM unit that provides Rx and Tx packet parsing and packet
manipulation functionality on Marvell CN9K and CN10K SoC's. It is
mapped to RTE Flow in DPDK.
Signed-off-by: Kiran Kumar K
---
drivers/common/cnx
Add support to dump TM HW registers and hierarchy on error.
This patch also adds support for misc utils such as API to
query TM HW resource availability, resource pre-allocation
and static priority support on root node.
Signed-off-by: Nithin Dabilpuram
---
drivers/common/cnxk/roc_nix.h
1 - 100 of 276 matches
Mail list logo
