Hi,
> -Original Message-
> From: Dekel Peled
> Sent: Thursday, January 16, 2020 8:36 PM
> To: Matan Azrad ; Slava Ovsiienko
>
> Cc: Raslan Darawsheh ; Ori Kam
> ; dev@dpdk.org
> Subject: [PATCH] net/mlx5: add GTP item support
>
> This patch adds to MLX5 PMD support of matching on GTP it
The return check of function tap_lsc_intr_handle_set() is wrong, it should
be 0 or a positive number if success. So the intr_handle->intr_vec was not
been freed when tap_lsc_intr_handle_set() returned a positive number.
Fixes: 4870a8cdd968 ("net/tap: support Rx interrupt")
Cc: sta...@dpdk.org
Sig
The return check of function tap_lsc_intr_handle_set() is wrong, it should
be 0 or a positive number if success. So the intr_handle->intr_vec was not
been freed when tap_lsc_intr_handle_set() returned a positive number.
Fixes: 4870a8cdd968 ("net/tap: support Rx interrupt")
Cc: sta...@dpdk.org
Sig
13/01/2020 07:00, Mahipal Challa:
> This patchset adds support for OCTEON TX2 end point mode of operation.
> The driver implementation uses DPDK rawdevice sub-system.
>
> v2:
> * Updated memory barrier API's as per Gavin Hu suggestion.
>
> v3:
> * Fixed memory leak possibility issues.
>
> v4:
>
16/01/2020 18:24, Ciara Power:
> The client script for use with the telemetry library did not support
> python3, as the data being sent over the socket was in string format.
> Python3 requires the data be explicitly converted to bytes before being
> sent. Similarily, the received bytes need to be
26/12/2019 02:46, Wu, Jingjing:
> From: Li, Xiaoyun
> > All buffers and ring info should be written before tail register update.
> > This patch relocates the write memory barrier before updating tail register
> > to avoid potential issues.
> >
> > Fixes: 11b5c7daf019 ("raw/ntb: add enqueue and de
23/12/2019 03:39, Wu, Jingjing:
> From: Li, Xiaoyun
> > In prerequisites of ntb guide, the correct flag when loading igb_uio
> > module should be `wc_activate=1`, not `wc_active=1`.
> >
> > Fixes: 11b5c7daf019 ("raw/ntb: add enqueue and dequeue functions")
> > Cc: sta...@dpdk.org
> >
> > Signed-
25/12/2019 10:09, Xiaoyun Li:
> Mempool ops may register in different order when compiling. The default
> ops index is always zero but the wanted ops is ring_mp_mc. This patch
> sets best mempool ops before generating mbuf pool to fix this issue.
>
> Fixes: 5194299d6ef5 ("examples/ntb: support mor
In the root table, there is some limitation of total number of header
modify actions, 16 or 8 for each. But in other tables, there is no
such strict limitation. In an IPv6 case, the IP fields modifying
will occupy more actions than that in IPv4, so the total support
number should be increased in or
10/12/2019 21:47, Ori Kam:
>
> > -Original Message-
> > Subject: Re: [dpdk-dev] [PATCH] ethdev: allow multiple security sessions to
> > use one rte flow
> >
> > On Sun, Dec 8, 2019 at 4:19 PM Anoob Joseph wrote:
> > >
> > > The rte_security API which enables inline protocol/crypto featur
> -Original Message-
> From: Thomas Monjalon
> Sent: Monday, January 20, 2020 3:20 AM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Rao, Nikhil ; Chas Williams
> ; Richardson, Bruce ;
> Stephen Hemminger ; Pattan, Reshma
>
> Subject: Re: [dpdk-dev] [PATCH v5] app/test: fix build when ring
Hi Adam,
>
> This patch set is a first step to refactor the overly complex symmetric
> crypto unit tests. It merges many separate arrays of the tests
> for these PMDs: null, aesni_mb, aesni_gcm, openssl, qat, sw_snow3g,
> sw_kasumi, sw_zuc into one big array that's then used when running
> unit te
17/01/2020 15:34, Olivier Matz:
> The documentation says that a negative errno is returned on error, but
> in most places that's not the case.
>
> Fix the documentation and the exceptions in code. The second one
> (return from populate_virt) also fixes a memory leak.
>
> Note that testpmd was usi
17/01/2020 11:09, Andrew Rybchenko:
> On 1/17/20 1:01 PM, Olivier Matz wrote:
> > On Fri, Jan 17, 2020 at 10:51:49AM +0100, Olivier Matz wrote:
> >> When allocating a mempool which is larger than the largest
> >> available area, it can take a lot of time:
> >>
> >> a- the mempool calculate the requ
On 1/20/2020 12:41 PM, Selwin Sebastian wrote:
> V1000/R1000 processors are using the same PCI ids for the network
> device as SNOWYOWL processor but has altered register definitions
> for determining the window settings for the indirect PCS access.
> Add support to check for this hardware and if f
On 1/20/2020 11:05 AM, Thomas Monjalon wrote:
> The version of the PMD should not be overwritten with LIBABIVER.
> All internal libs must have the same version.
>
> Moreover, specifying LIBABIVER in the driver causes a symbolic link loop:
> librte_rawdev_octeontx2_ep.so.1 -> librte_rawdev_octeontx
The version of the PMD should not be overwritten with LIBABIVER.
All internal libs must have the same version.
Moreover, specifying LIBABIVER in the driver causes a symbolic link loop:
librte_rawdev_octeontx2_ep.so.1 -> librte_rawdev_octeontx2_ep.so.1
Fixes: 56d46d13f736 ("raw/octeontx2_ep: add b
V1000/R1000 processors are using the same PCI ids for the network
device as SNOWYOWL processor but has altered register definitions
for determining the window settings for the indirect PCS access.
Add support to check for this hardware and if found use the new
register values.
Signed-off-by: Selwi
20/01/2020 12:06, Ferruh Yigit:
> On 1/20/2020 11:05 AM, Thomas Monjalon wrote:
> > The version of the PMD should not be overwritten with LIBABIVER.
> > All internal libs must have the same version.
> >
> > Moreover, specifying LIBABIVER in the driver causes a symbolic link loop:
> > librte_rawdev
The three Intel SW Crypto PMDs supporting SNOW3G, ZUC and KASUMI link against
libSSO libraries.
The Intel IPSec Multi-buffer library recently integrated the code from these
libraries, in version v0.53.
This library can be downloaded from https://github.com/intel/intel-ipsec-mb.
KASUMI, SNOW3G an
Link against Intel IPSec Multi-buffer library, which
added support for ZUC-EEA3 and ZUC-EIA3 from version v0.53,
moving from libSSO ZUC library.
Signed-off-by: Pablo de Lara
---
devtools/test-build.sh | 6 ++--
doc/guides/cryptodevs/zuc.rst | 52 ++--
Link against Intel IPSec Multi-buffer library, which
added support for SNOW3G-UEA2 and SNOW3G-UIA2 from version v0.53,
moving from libSSO SNOW3G library.
Signed-off-by: Pablo de Lara
---
devtools/test-build.sh | 4 +-
doc/guides/cryptodevs/snow3g.rst | 58 +
Link against Intel IPSec Multi-buffer library, which
added support for KASUMI-F8 and KASUMI-F9 from version v0.53,
moving from libSSO KASUMI library.
Signed-off-by: Pablo de Lara
---
devtools/test-build.sh | 6 +--
doc/guides/cryptodevs/kasumi.rst | 62
On 1/20/2020 4:42 PM, Selwin Sebastian wrote:
> V1000/R1000 processors are using the same PCI ids for the network
> device as SNOWYOWL processor but has altered register definitions
> for determining the window settings for the indirect PCS access.
> Add support to check for this hardware and if fo
17/01/2020 15:57, Olivier Matz:
> rte_mempool_populate_virt() sometimes fail, when it calls
> rte_mempool_populate_iova() with an area which is too small to store one
> object. This should not be an error.
>
> I prepared a v2 which implements an ABI compatibility through symbol
> versioning, as su
+1 - thanks Dave
On 20/01/2020 04:48, Jerin Jacob Kollanukkaran wrote:
>> -Original Message-
>> From: d...@barachs.net
>> Sent: Saturday, January 18, 2020 8:45 PM
>> To: 'Ray Kinsella' ; Jerin Jacob Kollanukkaran
>> ; 'dpdk-dev'
>> Subject: [EXT] RE: [RFC] [dpdk-dev] DPDK Trace support
>
On 1/19/2020 3:53 PM, Alfredo Cardigliano wrote:
> Add basic, per queue and extended statistics for
> RX and TX, both from the adapter and the driver.
>
> Signed-off-by: Alfredo Cardigliano
> Reviewed-by: Shannon Nelson
<...>
> @@ -84,6 +84,153 @@ ionic_lif_reset(struct ionic_lif *lif)
>
On Thu, Jan 16, 2020 at 01:04:25PM +, Viacheslav Ovsiienko wrote:
> The routine rte_pktmbuf_priv_flags is introduced to fetch
> the flags from the mbuf memory pool private structure
> in unified fashion.
>
> Signed-off-by: Viacheslav Ovsiienko
Acked-by: Olivier Matz
On Thu, Jan 16, 2020 at 06:40:23PM +0530, Jerin Jacob wrote:
> On Wed, Jan 15, 2020 at 2:35 AM wrote:
> >
> > From: Jerin Jacob
> >
> > The existing optimize_object_size() function address the memory object
> > alignment constraint on x86 for better performance.
> >
> > Different (micro) architec
Meson versions 0.52 and 0.53 are being overly smart and detecting the path
"/sys/devices/system/cpu/present" in the call to cat in
app/test/meson.build and then adding it as a dependency to the build
configuration. This causes issues on systems where the timestamp of that
file always returns the cu
20/01/2020 13:24, Olivier Matz:
> On Thu, Jan 16, 2020 at 06:40:23PM +0530, Jerin Jacob wrote:
> > On Wed, Jan 15, 2020 at 2:35 AM wrote:
> > >
> > > From: Jerin Jacob
> > >
> > > The existing optimize_object_size() function address the memory object
> > > alignment constraint on x86 for better p
28/12/2019 10:40, Jörg Thalheim:
>
> Looks good to me!
>
> Olivier Matz writes:
>
> > From: Jörg Thalheim
> >
> > Enhance API documentation of rte_pktmbuf_attach_extbuf() to
> > explain that the attached mbuf is initialized with length = 0.
> >
> > Link: https://bugs.dpdk.org/show_bug.cgi?id=3
On 1/20/2020 8:30 AM, Yunjian Wang wrote:
> The return check of function tap_lsc_intr_handle_set() is wrong, it should
> be 0 or a positive number if success. So the intr_handle->intr_vec was not
> been freed when tap_lsc_intr_handle_set() returned a positive number.
>
> Fixes: 4870a8cdd968 ("net/
Hi Vladimir, Akhil,
Marvell is also observing 10-15% drop with the SAD change. I agree with Akhil's
opinion and we are not in favor of making this change.
Thanks,
Anoob
From: Akhil Goyal
Sent: Monday, January 20, 2020 12:14 PM
To: Medvedkin, Vladimir ; dev@dpdk.org
Cc: konstantin.anan...@inte
> -Original Message-
> From: Bing Zhao
> Sent: Monday, January 20, 2020 11:43 AM
> To: Ori Kam ; Slava Ovsiienko
> ; Raslan Darawsheh ;
> Matan Azrad
> Cc: dev@dpdk.org; sta...@dpdk.org
> Subject: [PATCH v3] net/mlx5: fix modify actions support limitation
>
> In the root table, there
On 1/19/2020 3:53 PM, Alfredo Cardigliano wrote:
> The patch series provides an initial version of a
> poll mode driver for Pensando network adapters.
> The driver name is ionic.
>
> v5 Changes:
> --
> - Fixed commits headlines
> - Fixed doc syntax
> - Fixed unused parameters errors
> - Fi
> -Original Message-
> From: Ori Kam
> Sent: Monday, January 20, 2020 15:03
> To: Bing Zhao ; Slava Ovsiienko
> ; Raslan Darawsheh ;
> Matan Azrad
> Cc: dev@dpdk.org; sta...@dpdk.org
> Subject: RE: [PATCH v3] net/mlx5: fix modify actions support limitation
>
>
>
> > -Original Messa
> -Original Message-
> From: Suanming Mou
> Sent: Wednesday, January 8, 2020 12:41
> To: Matan Azrad ; Shahaf Shuler
> ; Slava Ovsiienko
> Cc: dev@dpdk.org; xiangxia.m@gmail.com; Raslan Darawsheh
> ; sta...@dpdk.org
> Subject: [PATCH v2] net/mlx5: fix incorrect pointer operation in me
This patch set is a first step to refactor the overly complex symmetric
crypto unit tests. It merges many separate arrays of the tests
for these PMDs: null, aesni_mb, aesni_gcm, openssl, qat, sw_snow3g,
sw_kasumi, sw_zuc into one big array that's then used when running
unit tests on these PMDs.
In
This patch gets rid of individual functions that all call
test_blockcipher_all_tests separately for every PMD and instead
provides just one set universal for all PMDs that's basing on the
driver id from the global variable gbl_driver_id.
Signed-off-by: Adam Dybkowski
---
app/test/test_cryptodev.
This patch adds capability checks to many tests meant to be run
in the future on various PMDs. This way the code is prepared for
more thorough refactoring in order to create one big central
unit tests array.
Signed-off-by: Adam Dybkowski
---
app/test/test_cryptodev.c | 637 ++
This patch refactors most of unit tests to be contained in one
combined array, and run depending on the PMD capabilities instead of
providing multiple array with tests for individual PMDs.
Only a subset of unit tests was merged into one array - it combines
all tests originally meant to be run on th
>
> This patch set is a first step to refactor the overly complex symmetric
> crypto unit tests. It merges many separate arrays of the tests
> for these PMDs: null, aesni_mb, aesni_gcm, openssl, qat, sw_snow3g,
> sw_kasumi, sw_zuc into one big array that's then used when running
> unit tests on
20/01/2020 14:05, Ferruh Yigit:
> On 1/19/2020 3:53 PM, Alfredo Cardigliano wrote:
> > The patch series provides an initial version of a
> > poll mode driver for Pensando network adapters.
> > The driver name is ionic.
> >
> > v5 Changes:
> > --
> > - Fixed commits headlines
> > - Fixed do
This series introduces event-mode additions to ipsec-secgw. This effort
is parallel to the similar changes in l2fwd (l2fwd-event app) & l3fwd.
With this series, ipsec-secgw would be able to run in eventmode. The
worker thread (executing loop) would be receiving events and would be
submitting it ba
From: Ankur Dwivedi
The default flow created would enable security processing on all ESP
packets. If the default flow is created, SA based rte_flow creation
would be skipped.
Signed-off-by: Ankur Dwivedi
Signed-off-by: Anoob Joseph
---
examples/ipsec-secgw/ipsec-secgw.c | 61 +
Add framework for eventmode helper. Event mode involves
initialization of multiple devices like eventdev, ethdev
and etc. Add routines to initialize and uninitialize event
device. Generate a default config for event device if it
is not specified in the configuration. Currently event
helper supports
Add event device port-lcore link and specify which event queues should
be connected to the event port. Generate a default config for event
port-lcore links if it is not specified in the configuration. This
routine will check the number of available ports and then create links
according to the numbe
Add Rx adapter support. The event helper init routine will initialize
the Rx adapter according to the configuration. If Rx adapter config
is not present it will generate a default config. If there are enough
event queues available it will map eth ports and event queues 1:1
(one eth port will be con
Add Tx adapter support. The event helper init routine will initialize
the Tx adapter according to the configuration. If Tx adapter config
is not present it will generate a default config.
Signed-off-by: Anoob Joseph
Signed-off-by: Lukasz Bartosik
---
examples/ipsec-secgw/event_helper.c | 313 ++
Add routines to display the eventmode configuration and provide
an overview of the devices used.
Signed-off-by: Anoob Joseph
Signed-off-by: Lukasz Bartosik
---
examples/ipsec-secgw/event_helper.c | 207
examples/ipsec-secgw/event_helper.h | 14 +++
2 files
From: Lukasz Bartosik
In eventmode workers can be drafted differently according to the
capabilities of the underlying event device. The added functions will
receive an array of such workers and probe the eventmode properties to
choose the worker.
Signed-off-by: Anoob Joseph
Signed-off-by: Lukas
From: Lukasz Bartosik
Add driver inbound and outbound worker thread for ipsec-secgw. In driver
mode application does as little as possible. It simply forwards packets
back to port from which traffic was received instructing HW to apply
inline security processing using first outbound SA configured
From: Lukasz Bartosik
Add application inbound/outbound worker thread and
IPsec application processing code for event mode.
Exampple ipsec-secgw command in app mode:
ipsec-secgw -w 0002:02:00.0,ipsec_in_max_spi=128
-w 0002:03:00.0,ipsec_in_max_spi=128 -w 0002:0e:00.0 -w 0002:10:00.1
--log-level=8
From: Lukasz Bartosik
Add eventmode support to ipsec-secgw. With the aid of event helper
configure and use the eventmode capabilities.
Signed-off-by: Anoob Joseph
Signed-off-by: Lukasz Bartosik
---
examples/ipsec-secgw/event_helper.c | 4 +-
examples/ipsec-secgw/event_helper.h | 14 ++
exa
From: Lukasz Bartosik
Add support for Rx and Tx internal ports. When internal ports are
available then a packet can be received from eth port and forwarded
to event queue by HW without any software intervention. The same
applies to Tx side where a packet sent to an event queue can by
forwarded by
From: Lukasz Bartosik
Add command line option -s which can be used to configure number
of buffers in a pool. Default number of buffers is 8192.
Signed-off-by: Anoob Joseph
Signed-off-by: Lukasz Bartosik
---
examples/ipsec-secgw/ipsec-secgw.c | 23 +++
1 file changed, 19 in
> -Original Message-
> From: stable On Behalf Of Thomas Monjalon
> Sent: Friday, January 17, 2020 9:52 PM
> To: Trahe, Fiona ; Trybula, ArturX
>
> Cc: dev@dpdk.org; sta...@dpdk.org; shal...@marvell.com; Dybkowski, AdamX
> ; Danilewicz, MarcinX
> ; Akhil Goyal
> Subject: Re: [dpdk-stab
Hi Slava,
Some comments inline.
On Thu, Jan 16, 2020 at 01:04:26PM +, Viacheslav Ovsiienko wrote:
> Update detach routine to check the mbuf pool type.
> Introduce the special internal version of detach routine to handle
> the special case of pinned external bufferon mbuf freeing.
>
> Signed-
On Thu, Jan 16, 2020 at 01:04:27PM +, Viacheslav Ovsiienko wrote:
> The dedicated routine rte_pktmbuf_pool_create_extbuf() is
> provided to create mbuf pool with data buffers located in
> the pinned external memory. The application provides the
> external memory description and routine initiali
Hi,
> -Original Message-
> From: Suanming Mou
> Sent: Wednesday, January 8, 2020 12:41 PM
> To: Matan Azrad ; Shahaf Shuler
> ; Slava Ovsiienko
> Cc: dev@dpdk.org; xiangxia.m@gmail.com; Raslan Darawsheh
> ; sta...@dpdk.org
> Subject: [PATCH v2] net/mlx5: fix incorrect pointer operati
Hi,
> -Original Message-
> From: Bing Zhao
> Sent: Monday, January 20, 2020 11:43 AM
> To: Ori Kam ; Slava Ovsiienko
> ; Raslan Darawsheh ;
> Matan Azrad
> Cc: dev@dpdk.org; sta...@dpdk.org
> Subject: [PATCH v3] net/mlx5: fix modify actions support limitation
>
> In the root table, there
On Thu, Jan 16, 2020 at 01:04:28PM +, Viacheslav Ovsiienko wrote:
> The new mbuf pool type is added to testpmd. To engage the
> mbuf pool with externally attached data buffers the parameter
> "--mp-alloc=xbuf" should be specified in testpmd command line.
>
> The objective of this patch is just
This is rework of "[PATCH v2] net/virtio: add link speed tuning" which I
sent month ago. Compile time link_speed is reworked to devarg.
Best regards,
Ivan
refactor vdpa specific devargs parsing to more generic way
Signed-off-by: Ivan Dyukov
---
drivers/net/virtio/virtio_ethdev.c | 32 +-
1 file changed, 23 insertions(+), 9 deletions(-)
diff --git a/drivers/net/virtio/virtio_ethdev.c
b/drivers/net/virtio/virtio_ethdev.
Some applications like pktgen use link_speed to calculate
transmit rate. It limits outcome traffic to hardcoded 10G.
This patch adds link_speed devarg which allows to configure
link_speed of virtio device.
Signed-off-by: Ivan Dyukov
---
drivers/net/virtio/virtio_ethdev.c | 99 ++
20/01/2020 14:54, Mcnamara, John:
> From: stable On Behalf Of Thomas Monjalon
> > 15/01/2020 16:54, Akhil Goyal:
> > > > >
> > > > > This patch replaces existing test vector with a new one containing
> > > > > C code to fix license issue.
> > > > >
> > > > > Signed-off-by: Artur Trybula
> > > > A
Hi Akhil,
> Hi Vladimir,
> The SA lookup logic and management is purely requirement based for the
> application.
>The application may only cater to <128 SAs which can
> be handled based on the current logic.
Not always, current implementation can handle < 128 SA,
whose SPI%128 never match (le
Hi Thomas
> -Original Message-
> From: Thomas Monjalon
> Sent: Monday, January 20, 2020 2:33 PM
> To: Trahe, Fiona ; Trybula, ArturX
> ; Mcnamara,
> John
> Cc: dev@dpdk.org; sta...@dpdk.org; shal...@marvell.com; Dybkowski, AdamX
> ; Danilewicz, MarcinX
> ; Akhil Goyal
>
> Subject: Re:
The second patch will apply failed without the first patch applied, so make
them a series. The first patch is just a typo fix. The second patch fixes
the issue that the interrupts are turned off only on the Rx queues that
has traffic arrived, rather than all the Rx queues a core has taken.
Xiao Wa
Fixes: aee3bc79cc34 ("examples/l3fwd-power: enable one-shot Rx interrupt and
polling switch")
Cc: sta...@dpdk.org
Signed-off-by: Xiao Wang
---
examples/l3fwd-power/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/examples/l3fwd-power/main.c b/examples/l3fwd-power/main.
Since all related queues' interrupts are turned on before epoll, we need
to turn off all the interrupts after wakeup. This patch fixes the issue
of only turning off the interrupted queues.
Fixes: b736d64787fc ("examples/l3fwd-power: disable Rx interrupt when waking
up")
Cc: sta...@dpdk.org
Signe
17/12/2019 14:52, Dumitrescu, Cristian:
> From: Eelco Chaudron
> > Moved RFC4115 APIs to none experimental as they have been there
> > since 19.02. Also, these APIs are the same as the none RFC4115 APIs.
> >
> > Signed-off-by: Eelco Chaudron
>
> Acked-by: Cristian Dumitrescu
Applied, thanks
Replace DSW's use of regular DPDK rings (and code for
packing/unpacking control messages into void pointers) with custom
size rings.
In addition to cleaner code, this change allows DSW to support up to
the eventdev API's maximum of 255 ports by tweaking DSW_MAX_PORTS.
Signed-off-by: Mattias Rönnb
20/01/2020 13:22, Bruce Richardson:
> Meson versions 0.52 and 0.53 are being overly smart and detecting the path
> "/sys/devices/system/cpu/present" in the call to cat in
> app/test/meson.build and then adding it as a dependency to the build
> configuration. This causes issues on systems where the
Hi, Olivier
Thanks a lot for the thorough review.
There are some answers to comments, please, see below.
> >
> > /**
> > + * @internal version of rte_pktmbuf_detach() to be used on mbuf freeing.
>
> -version
> +Version
>
> > + * For indirect and regular (not pinned) external mbufs the standard
On 1/20/2020 4:01 AM, Xiaoyu Min wrote:
> Explanation of flow dump command is added into testpmd functions.
>
> Signed-off-by: Xiaoyu Min
Reviewed-by: Ferruh Yigit
Applied to dpdk-next-net/master, thanks.
Hi,
On Mon, Jan 20, 2020 at 03:41:10PM +, Slava Ovsiienko wrote:
> Hi, Olivier
>
> Thanks a lot for the thorough review.
> There are some answers to comments, please, see below.
>
> > >
> > > /**
> > > + * @internal version of rte_pktmbuf_detach() to be used on mbuf freeing.
> >
> > -versi
03/01/2020 12:52, Bruce Richardson:
> The FreeBSD CI system is reporting errors with DPDK on the current head
> of the FreeBSD development branch for FreeBSD 13. Ahead of BSD 13's
> release, we need to apply fixes to ensure DPDK continues to compile
> on the various versions.
>
> Bruce Richardson
On Sat, 18 Jan 2020 10:14:31 -0500
wrote:
> It would be well worth considering one of the vpp techniques to minimize
> trace impact:
>
> static inline ring_handler_inline (..., int is_traced)
> {
> for (i = 0; i < vector_size; i++)
> {
> if (is_traced)
> {
> do_trace_w
On 1/19/2020 3:53 PM, Alfredo Cardigliano wrote:
> Add support for managing RX filters based on MAC and VLAN.
> Hardware cannot provide the list of filters, thus we keep
> a local list.
> Add support for promisc and allmulticast modes.
>
> Signed-off-by: Alfredo Cardigliano
> Reviewed-by: Shannon
Support get_queue_num operation to get the maximum number of queues
supported by the device.
This number comes from the DevX capabilities.
Signed-off-by: Matan Azrad
---
drivers/vdpa/mlx5/mlx5_vdpa.c | 54 ++-
1 file changed, 53 insertions(+), 1 deletion(
A new Mellanox vdpa PMD will be added to support vdpa operations by
Mellanox adapters.
Both, the mlx5 PMD and the vdpa mlx5 PMD should initialize the glue.
The glue initialization should be only one per process, so all the mlx5
PMDs using the glue should share the same glue object.
Move the glue
Move the vendor information, vendor ID and device IDs from mlx5 PMD to
the common mlx5 file.
Signed-off-by: Matan Azrad
---
drivers/common/mlx5/mlx5_common.h | 21 +
drivers/net/mlx5/mlx5.h | 21 -
drivers/net/mlx5/mlx5_txq.c | 1 +
3 file
Add a new driver to support vDPA operations by Mellanox devices.
The first Mellanox devices which support vDPA operations are
ConnectX6DX and Bluefield1 HCA for their PF ports and VF ports.
This driver is depending on rdma-core like the mlx5 PMD, also it is
going to use mlx5 DevX to create HW obj
The DevX commands interfaces is included in the mlx5.h file with a lot
of other PMD interfaces.
As an arrangement to make the DevX commands shared with different PMDs,
this patch moves the DevX interface to a new file called mlx5_devx_cmds.h.
Also remove shared device structure dependency from De
Add the DevX capabilities for vDPA configuration and information of
Mellanox devices.
Signed-off-by: Matan Azrad
---
drivers/common/mlx5/mlx5_devx_cmds.c | 90
drivers/common/mlx5/mlx5_devx_cmds.h | 24 ++
drivers/common/mlx5/mlx5_prm.h | 45 +++
Move PCI detection by IB device from mlx5 PMD to the common code.
Signed-off-by: Matan Azrad
---
drivers/common/mlx5/Makefile| 2 +-
drivers/common/mlx5/mlx5_common.c | 55 +
drivers/common/mlx5/mlx5_common.h | 4 ++
drive
Add support for get_features and get_protocol_features operations.
Part of the features are reported by the DevX capabilities.
Signed-off-by: Matan Azrad
---
doc/guides/vdpadevs/features/mlx5.ini | 7
drivers/vdpa/mlx5/mlx5_vdpa.c | 66 +--
2 files
Steps:
- Prepare net/mlx5 for code sharing.
- Introduce new common lib for mlx5 devices.
- Share code from net/mlx5 to common/mlx5.
- Introduce vDPA driver for Mellanox devices.
Matan Azrad (38):
net/mlx5: separate DevX commands interface
mlx5: prepare common library
mlx5: share the mlx5 gl
Add support for rdma-core API to allocate NULL MR.
Signed-off-by: Matan Azrad
---
drivers/common/mlx5/mlx5_glue.c | 13 +
drivers/common/mlx5/mlx5_glue.h | 1 +
2 files changed, 14 insertions(+)
diff --git a/drivers/common/mlx5/mlx5_glue.c b/drivers/common/mlx5/mlx5_glue.c
index d5
Add option to create an indirect mkey by the current
mlx5_devx_cmd_mkey_create command.
Align the net/mlx5 driver usage in the above command.
Signed-off-by: Matan Azrad
---
drivers/common/mlx5/mlx5_devx_cmds.c | 43 ++--
drivers/common/mlx5/mlx5_devx_cmds.h | 12
The event queue is managed only by the kernel.
Add the rdma-core command in glue to query the kernel event queue
details.
Signed-off-by: Matan Azrad
---
drivers/common/mlx5/mlx5_glue.c | 15 +++
drivers/common/mlx5/mlx5_glue.h | 2 ++
2 files changed, 17 insertions(+)
diff --git a
The CQE has owner bit to indicate if it is in SW control or HW.
Share a CQE check for all the mlx5 drivers.
Signed-off-by: Matan Azrad
---
drivers/common/mlx5/mlx5_common.h | 41 +++
drivers/net/mlx5/mlx5_rxtx.h | 39 +
Add the next commands to glue in order to support interrupt event
channel operations associated to events in the EQ:
devx_create_event_channel,
devx_destroy_event_channel,
devx_subscribe_devx_event,
devx_subscribe_devx_event_fd,
devx_get_event.
Signed-off-by
The isolated, protected and independent direct access to the HW by
multiple processes is implemented via User Access Region (UAR)
mechanism.
The UAR is part of PCI address space that is mapped for direct access to
the HW from the CPU.
UAR is comprised of multiple pages, each page containing registe
Virtio access region(VAR) is the UAR that allocated for virtio emulation
access.
Add rdma-core operations to allocate and free VAR.
Signed-off-by: Matan Azrad
---
drivers/common/mlx5/Makefile| 5 +
drivers/common/mlx5/meson.build | 1 +
drivers/common/mlx5/mlx5_glue.c | 26 +++
Virtio emulation offload allows SW to offload the I/O operations of a
virtio virtqueue, using the device, allowing an improved performance
for its users.
While supplying all the relevant Virtqueue information (type, size,
memory location, doorbell information, etc.). The device can then
offload the
In order to free resources in the CQ and to allow to the HW to send
packets from the guest, the CQ should be polled.
In order to poll the CQ we need to trigger an interrupt for each new
CQE posted by the HW.
Register interrupt handler to poll and arm a CQ when completion event
was raised by the H
As an arrangement to the vitrio queues creation, a CQ should be created
for the virtio queue.
The design is to trigger an event for the guest and for the vdpa driver
when a new CQE is posted by the HW after the packet transition.
This patch add the basic operations to create and destroy CQs and t
1 - 100 of 174 matches
Mail list logo
