> -Original Message-
> From: Jerin Jacob
> Sent: Tuesday, October 15, 2019 11:57 AM
> To: Kiran Kumar Kokkilagadda
> Cc: Adrien Mazarguil ; Wenzhuo Lu
> ; Jingjing Wu ; Bernard
> Iremonger ; John McNamara
> ; Marko Kovacevic ;
> Thomas Monjalon ; Ferruh Yigit
> ; Andrew Rybchenko ;
> Ol
Series-
Acked-by: Hemant Agrawal
Acked-by: Hemant Agrawal
Acked-by: Hemant Agrawal
When VIRTIO_F_IN_ORDER feature is negotiated, vhost can optimize dequeue
function by only update first used descriptor.
Signed-off-by: Marvin Liu
diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 85ccc02da..88632caff 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b
Flush used flags when batched enqueue function is finished. Descriptor's
flags are pre-calculated as they will be reset by vhost.
Signed-off-by: Marvin Liu
diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index a60b88d89..bf3c30f43 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/
Optimize vhost device packed ring dequeue function by splitting batch
and single functions. No-chained and direct descriptors will be handled
by batch and other will be handled by single as before.
Signed-off-by: Marvin Liu
diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net
Buffer used ring updates as many as possible in vhost dequeue function
for coordinating with virtio driver. For supporting buffer, shadow used
ring element should contain descriptor's flags. First shadowed ring
index was recorded for calculating buffered number.
Signed-off-by: Marvin Liu
diff --
Buffer vhost enqueue shadowed ring flush action buffered number exceed
one batch. Thus virtio can receive packets at a faster frequency.
Signed-off-by: Marvin Liu
diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index 96bf763b1..a60b88d89 100644
--- a/lib/librte_vhost/vhost.h
+++
Add vhost packed ring zero copy batch and single dequeue functions like
normal dequeue path.
Signed-off-by: Marvin Liu
diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 5cdca9a7f..01d1603e3 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_ne
Add batch dequeue function like enqueue function for packed ring, batch
dequeue function will not support chained descritpors, single packet
dequeue function will handle it.
Signed-off-by: Marvin Liu
diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index 18d01cb19..96bf763b1 1006
Optimize vhost device packed ring enqueue function by splitting batch
and single functions. Packets can be filled into one desc will be
handled by batch and others will be handled by single as before.
Signed-off-by: Marvin Liu
diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_
Add vhost single packet dequeue function for packed ring and meanwhile
left space for shadow used ring update function.
Signed-off-by: Marvin Liu
Reviewed-by: Maxime Coquelin
diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index a8130dc06..e1b06c1ce 100644
--- a/lib/l
Batch enqueue function will first check whether descriptors are cache
aligned. It will also check prerequisites in the beginning. Batch
enqueue function do not support chained mbufs, single packet enqueue
function will handle it.
Signed-off-by: Marvin Liu
diff --git a/lib/librte_vhost/virtio_net
Create macro for adding unroll pragma before for each loop. Batch
functions will be contained of several small loops which can be
optimized by compilers' loop unrolling pragma.
Signed-off-by: Marvin Liu
diff --git a/lib/librte_vhost/Makefile b/lib/librte_vhost/Makefile
index 8623e91c0..30839a001
When vhost doing [de]nqueue, vq's local variable last_[used/avail]_idx
will be inceased. Adding inline functions can avoid duplicated codes.
Signed-off-by: Marvin Liu
diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index 5131a97a3..22a3ddc38 100644
--- a/lib/librte_vhost/vhost.h
Add vhost enqueue function for single packet and meanwhile left space
for flush used ring function.
Signed-off-by: Marvin Liu
Reviewed-by: Maxime Coquelin
diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 42b662080..142c14e04 100644
--- a/lib/librte_vhost/virtio_n
Packed ring has more compact ring format and thus can significantly
reduce the number of cache miss. It can lead to better performance.
This has been approved in virtio user driver, on normal E5 Xeon cpu
single core performance can raise 12%.
http://mails.dpdk.org/archives/dev/2018-April/095470.ht
>
> 11/10/2019 14:40, Akhil Goyal:
> > Hi All,
> >
> > This patchset would need ack from more vendors as it will impact user
> experience
> > on a key example application which is normally demonstrated to customers.
> >
> > IPSec library is still evolving and there are new functionality added ev
On Tue, Oct 15, 2019 at 11:51 AM wrote:
>
> From: Kiran Kumar K
>
> Add new rte_flow_item_higig2_hdr in order to match higig2 header.
> It is a layer 2.5 protocol and used in broadcom switches.
> Header format is based on the following document.
> http://read.pudn.com/downloads558/doc/comm/230146
From: Kiran Kumar K
Add new rte_flow_item_higig2_hdr in order to match higig2 header.
It is a layer 2.5 protocol and used in broadcom switches.
Header format is based on the following document.
http://read.pudn.com/downloads558/doc/comm/2301468/HiGig_protocol.pdf
Signed-off-by: Kiran Kumar K
--
> -Original Message-
> From: Maxime Coquelin [mailto:maxime.coque...@redhat.com]
> Sent: Friday, October 11, 2019 10:22 PM
> To: Liu, Yong ; Bie, Tiwei ; Wang,
> Zhihong ; step...@networkplumber.org;
> gavin...@arm.com
> Cc: dev@dpdk.org
> Subject: Re: [PATCH v4 03/14] vhost: add batch en
Hi all,
In the latest features of the MLX5 PMD (i.e., DPDK 19.08), there is LRO
hardware support.
However, I cannot make it work with our dual port 100GbE Mellanox ConnectX5
(MT27800) card.
Let me first describe the configuration of my testbed: First,I upgraded
OFED to the latest 4.7-1.0.0 with f
On Mon, Oct 14, 2019 at 06:15:42PM +0300, Andrew Rybchenko wrote:
> > diff --git a/drivers/net/virtio/virtio_rxtx.c
> > b/drivers/net/virtio/virtio_rxtx.c
> > index 27ead19fb..822cce06d 100644
> > --- a/drivers/net/virtio/virtio_rxtx.c
> > +++ b/drivers/net/virtio/virtio_rxtx.c
> > @@ -597,9 +597,
When config i40e rx queue, the temporary variable to store max packet
length is not big enough which leads to integer overflow issue. This
patch is to fix the issue by removing the variable and using the
expression directly since the variable only used once.
Fixes: 4861cde46116 ("i40e: new poll mo
From: Rajesh Ravi
Support external custom memory added to heap to be used with vfio with
--iso-cmem option.Type1 memory mapping was by passed for external memory.
But an exception is added to allow external custom memory to be used with
vfio.
Signed-off-by: Rajesh Ravi
Reviewed-by: Jonathan Ric
-Original Message-
From: Andrew Rybchenko
Sent: Monday, October 14, 2019 12:38 PM
To: Kiran Kumar Kokkilagadda ; Adrien Mazarguil
; John McNamara ; Marko
Kovacevic ; Thomas Monjalon ;
Ferruh Yigit
Cc: dev@dpdk.org; ajit.khapa...@broadcom.com
Subject: [EXT] Re: [dpdk-dev] [PATCH] eth
NAK, this is not true for ice.
Paul
> -Original Message-
> From: dev On Behalf Of Yang, Qiming
> Sent: Monday, October 14, 2019 6:49 PM
> To: Min, JiaqiX ; dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] doc: add known issue for ice
>
>
> > -Original Message-
> > From: Min, Jiaq
This patch fixed the bug that a failure appeared when config
rx_offload crc_strip using command "port config all crc-strip
on|off". The reason is that this command was removed in Commit
e5db17a1e54e. The current command is "port config
rx_offload keep_crc on|off" instead.
In this patch, some code
> -Original Message-
> From: Min, JiaqiX
> Sent: Tuesday, October 15, 2019 2:22 AM
> To: dev@dpdk.org; Yang, Qiming
> Cc: Min, JiaqiX
> Subject: [PATCH] doc: add known issue for ice
>
> Add known issue for incorrect Rx statistics when packet is oversize.
>
> Signed-off-by: Min JiaqiX
Hi Kevin,
> -Original Message-
> From: Kevin Traynor [mailto:ktray...@redhat.com]
> Sent: Tuesday, October 15, 2019 1:42 AM
> To: Zhang, Xiao ; dev@dpdk.org
> Cc: Xing, Beilei ; Zhang, Qi Z ;
> sta...@dpdk.org; Stokes, Ian
> Subject: Re: [dpdk-dev] net/i40e: fix vlan packets drop
>
> On
> -Original Message-
> From: Yigit, Ferruh
> Sent: Tuesday, October 15, 2019 00:39
> To: Wang, Haiyue ; dev@dpdk.org; Ye, Xiaolong
>
> Cc: Kinsella, Ray ; Iremonger, Bernard
> ; Sun,
> Chenmin
> Subject: Re: [PATCH v3 1/4] ethdev: add the API for getting burst mode
> information
>
> O
Hi Kevin,
Will update the patch according to your comments. Thanks.
Xiao
> -Original Message-
> From: Kevin Traynor [mailto:ktray...@redhat.com]
> Sent: Monday, October 14, 2019 6:36 PM
> To: Zhang, Xiao ; dev@dpdk.org
> Cc: Xing, Beilei ; Zhang, Qi Z ;
> sta...@dpdk.org
> Subject: Re: [
Hi, Bernard,
In the previous version, the command 'crc_strip' and 'keep_crc' used the same
macro DEV_RX_OFFLOAD_KEEP_CRC. After Commit e5db17a1e54e, "port config all
crc-strip on|off" is removed and the new command does not enable 'crc_strip'.
So I wanted to change the name to 'keep_crc' to us
Hi,
> -Original Message-
> From: Yigit, Ferruh
> Sent: Tuesday, October 15, 2019 00:39
> To: Wang, Haiyue ; dev@dpdk.org; Ye, Xiaolong
>
> Cc: Kinsella, Ray ; Iremonger, Bernard
> ; Sun,
> Chenmin
> Subject: Re: [PATCH v3 1/4] ethdev: add the API for getting burst mode
> information
>
Hi, Ferruh,
From the log in Commit e5db17a1e54e, I thought crc_strip was still available in
command "port config rx_offload ...", so I did such modification.
Since it was removed, I will rework on it.
There is no problem on 'keep_crc'.
Thanks!
-Original Message-
From: Yigit, Ferruh
Se
Hi Konstantin,
Thank you for the feedback.
>
> > >
> > > Current APIs assume ring elements to be pointers. However, in many
> > > use cases, the size can be different. Add new APIs to support
> > > configurable ring element sizes.
> > >
> > > Signed-off-by: Honnappa Nagarahalli
> > > R
On 10/14, Haiyue Wang wrote:
>v1 -> v2:
> Split into two patches, one is fixing a commit
>
>Haiyue Wang (2):
> net/ice: fix the missed memory free when dev is closed
> net/ice: avoid the parsed devargs value being overwritten
>
> drivers/net/ice/ice_ethdev.c | 44 +--
The rte_atomic64_exchange operation for ppc_64 incorrectly linked
back to a 32 bit generic operation (__atomic_exchange_4) rather than
the 64 bit generic operation (__atomic_exchange_8). As a result,
only the link speed was passed to the application, not the link
state, link duplex, on link autone
Here's another link to the thread where this was discussed last year..
Igor was actually on this thread as well...
https://mails.dpdk.org/archives/dev/2018-August/110383.html
On Mon, Oct 14, 2019 at 4:01 PM Dan Gora wrote:
>
> My original patch to add this feature was basically the same thing as
> >
> > Current APIs assume ring elements to be pointers. However, in many use
> > cases,
> > the size can be different. Add new APIs to support configurable ring element
> > sizes.
> >
> > Signed-off-by: Honnappa Nagarahalli
> > Reviewed-by: Dharmik Thakkar
> > Reviewed-by: Gavin Hu
> > Revi
My original patch to add this feature was basically the same thing as
this: setting the link status via a KNI ioctl. That method was
rejected after _much_ discussion and we eventually settled on the
currently implementation.
My original patch was here: Message-Id: <20180628225548.21885-1...@adax.c
On Tue, Aug 13, 2019 at 2:20 PM Allain Legacy
wrote:
>
> Replacing full license text with SPDX tag.
>
> Signed-off-by: Allain Legacy
> ---
> app/test/test_cfgfile.c | 32 ++--
> 1 file changed, 2 insertions(+), 30 deletions(-)
>
> diff --git a/app/test/test_cfgfile.c
From: Pavan Nikhilesh
Add event dev main loop based on enabled l2fwd options and eventdev
capabilities.
Signed-off-by: Pavan Nikhilesh
---
examples/l2fwd-event/l2fwd_common.c | 6 +
examples/l2fwd-event/l2fwd_common.h | 1 +
examples/l2fwd-event/l2fwd_event.c | 276 +++
From: Pavan Nikhilesh
Setup service cores for eventdev and Rx/Tx adapter when they don't have
internal port capability.
Signed-off-by: Pavan Nikhilesh
---
examples/l2fwd-event/l2fwd_event.c | 82 ++
examples/l2fwd-event/l2fwd_event.h | 1 +
examples/l2fwd-event/mai
From: Sunil Kumar Kori
Add documentation for l2fwd-event example.
Update release notes.
Signed-off-by: Sunil Kumar Kori
---
MAINTAINERS | 1 +
doc/guides/rel_notes/release_19_11.rst| 6 +
doc/guides/sample_app_ug/index.rst| 1 +
doc/g
From: Sunil Kumar Kori
Add event eth Rx/Tx adapter setup for both generic and internal port
event device pipelines.
Signed-off-by: Sunil Kumar Kori
Signed-off-by: Pavan Nikhilesh
---
examples/l2fwd-event/l2fwd_event.c| 3 +
examples/l2fwd-event/l2fwd_event.h| 16 +++
From: Pavan Nikhilesh
Add graceful teardown that addresses both event mode and poll mode.
Signed-off-by: Pavan Nikhilesh
---
examples/l2fwd-event/main.c | 50 +
1 file changed, 40 insertions(+), 10 deletions(-)
diff --git a/examples/l2fwd-event/main.c b/exa
From: Sunil Kumar Kori
Add event device device setup based on event eth Tx adapter
capabilities.
Signed-off-by: Sunil Kumar Kori
Signed-off-by: Pavan Nikhilesh
---
examples/l2fwd-event/l2fwd_event.c| 3 +
examples/l2fwd-event/l2fwd_event.h| 16
examples/l2fwd-eve
From: Sunil Kumar Kori
Add event device queue and port setup based on event eth Tx adapter
capabilities.
Signed-off-by: Sunil Kumar Kori
Signed-off-by: Pavan Nikhilesh
---
examples/l2fwd-event/l2fwd_event.c| 9 +-
examples/l2fwd-event/l2fwd_event.h| 10 ++
examples/
From: Pavan Nikhilesh
Add the default l2fwd poll mode routines similar to examples/l2fwd.
Signed-off-by: Sunil Kumar Kori
Signed-off-by: Pavan Nikhilesh
---
MAINTAINERS | 5 +
examples/Makefile | 1 +
examples/l2fwd-event/Makefile | 59
From: Pavan Nikhilesh
Add infra to select event device as a mode to process packets through
command line arguments. Also, allow the user to select the schedule type
to be RTE_SCHED_TYPE_ORDERED, RTE_SCHED_TYPE_ATOMIC or
RTE_SCHED_TYPE_PARALLEL.
Usage:
`--mode="eventdev"` or `--mode="poll"`
`--e
From: Sunil Kumar Kori
Add infra to split eventdev framework based on event Tx adapter
capability.
If event Tx adapter has internal port capability then we use
`rte_event_eth_tx_adapter_enqueue` to transmitting packets else
we use a SINGLE_LINK event queue to enqueue packets to a service
core whi
From: Pavan Nikhilesh
This patchset adds a new application to demonstrate the usage of event
mode. The poll mode is also available to help with the transition.
The following new command line parameters are added:
--mode: Dictates the mode of operation either poll or event.
--eventq_sched: Dict
On 14/10/2019 08:53, Xiao Zhang wrote:
> vlan packets with ip length bigger then 1496 will not be received by
> i40e due to wrong packets size checking. This patch fixes the issue by
> correcting the maximum frame size during checking.
>
> Fixes: 35b2d13fd6fd ("net: add rte prefix to ether defines
On Thu, 19 Sep 2019 15:17:27 +0200
Stephen Hemminger wrote:
> A useful feature of netvsc PMD is the ability to see how many
> packets
> were processed through the VF device. This patch set adds a similar
> (but more limited) capability to failsafe driver.
>
> Since failsafe doesn't have top leve
> -Original Message-
> From: Aaron Conole [mailto:acon...@redhat.com]
> Sent: Monday, October 14, 2019 3:54 PM
> To: Van Haaren, Harry
> Cc: David Marchand ; dev@dpdk.org
> Subject: Re: [dpdk-dev] [BUG] service_lcore_en_dis_able from
> service_autotest failing
>
> Aaron Conole writes:
>
On Mon, Oct 14, 2019 at 09:46:50PM +0530, Jerin Jacob wrote:
> On Mon, Oct 14, 2019 at 5:05 PM Bruce Richardson
> wrote:
> >
> > In other places in DPDK, e.g. EAL, the architecture-specific files for arm
> > are just stored in a folder called "arm". Rename the "arm64" folder for
> > performance th
On 10/14/2019 4:35 PM, Haiyue Wang wrote:
> Some PMDs have more than one RX/TX burst paths, add the ethdev API
> that allows an application to retrieve the mode information about
> Rx/Tx packet burst such as Scalar or Vector, and Vector technology
> like AVX2.
>
> Signed-off-by: Haiyue Wang
> Ack
> -Original Message-
> From: Dumitrescu, Cristian
> Sent: Monday, October 14, 2019 3:24 PM
> To: Singh, Jasvinder ; dev@dpdk.org
> Subject: RE: [PATCH v4 00/17] sched: subport level configuration of pipe nodes
>
>
>
> > -Original Message-
> > From: Singh, Jasvinder
> > Sent: M
To support high bandwidth NICs, all rates (port, subport level
token bucket and traffic class rates, pipe level token bucket
and traffic class rates) and stats counters defined in public
data structures (rte_sched.h) are modified to support 64-bit
values.
Signed-off-by: Jasvinder Singh
Signed-off
Modify qos sample app to allow different subports of the same port
to have different configuration in terms of number of pipes, pipe
queue sizes, etc.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
examples/qos_sched/app_thread.c | 20 ++-
examples/qos_sched/cfg_file.c
Remove redundant data structure fields from port level data
structures and update the release notes.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
doc/guides/rel_notes/release_19_11.rst | 7 -
lib/librte_sched/rte_sched.c | 42 +-
lib/
Modify internal structure and functions to support 64-bit
values for rates and stats parameters.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
lib/librte_sched/rte_approx.c | 57
lib/librte_sched/rte_approx.h | 3 +-
lib/librte_sched/rte_sched.c
Modify ip pipeline traffic management function to allow different
subports of the same port to have different configuration in terms
of number of pipes, pipe queue sizes, etc.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
examples/ip_pipeline/cli.c | 71 ++-
Modify softnic traffic management function to allow different
subports of the same port to have different configuration in
terms of number of pipes, pipe queue sizes, etc.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
drivers/net/softnic/rte_eth_softnic_tm.c | 54 ++
Modify tests function to allow different subports of the same port
to have different configuration in terms of number of pipes, pipe
queue sizes, etc.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
app/test/test_sched.c | 35 ++-
1 file change
Modify pipe queue stats read function to allow different subports
of the same port to have different configuration in terms of number
of pipes, pipe queue sizes, etc.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
lib/librte_sched/rte_sched.c | 28 +++
Modify scheduler packet dequeue operation to allow different
subports of the same port to have different configuration in terms
of number of pipes, pipe queue sizes, etc.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
lib/librte_sched/rte_sched.c | 51 +++
Modify pipe level functions to allow different subports of the same
port to have different configuration in terms of number of pipes,
pipe queue sizes, etc.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
lib/librte_sched/rte_sched.c | 103 +
Update memory footprint compute function for allowing subports of
the same port to have different configuration in terms of number of
pipes, pipe queue sizes, etc.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
lib/librte_sched/rte_sched.c | 100 ++---
Modify scheduler packet enqueue operation of the scheduler to allow
different subports of the same port to have different configuration
in terms of number of pipes, pipe queue sizes, etc.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
lib/librte_sched/rte_sched.c | 277 +
Modify packet grinder functions of the schedule to allow different
subports of the same port to have different configuration in terms
of number of pipes, pipe queue sizes, etc.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
lib/librte_sched/rte_sched.c | 223
Remove pipes configuration from the port level to allow different
subports of the same port to have different configuration in terms
of number of pipes, pipe queue sizes, etc.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
lib/librte_sched/rte_sched.c | 254 ++---
Add pipes configuration from the port level to allow different
subports of the same port to have different configuration in terms
of number of pipes, pipe queue sizes, etc.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
lib/librte_sched/rte_sched.c | 388
This patchset refactors the dpdk qos sched library to allow subport
level configuration flexibility of the pipe nodes.
Currently, all parameters for the pipe nodes (subscribers)
configuration are part of the port level structure which forces
all groups of subscribers (pipes) in different subports
Add pipe configuration parameters to subport level structure to
allow different subports of the same port to have different
configuration in terms of number of pipes, pipe queue sizes, etc.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
lib/librte_sched/Makefile| 2 +-
Update internal structures related to port and subport to allow
different subports of the same port to have different configuration
in terms of number of pipes, pipe queue sizes, etc.
Signed-off-by: Jasvinder Singh
Signed-off-by: Lukasz Krakowiak
---
lib/librte_sched/rte_sched.c | 106 +
On 10/14/2019 5:10 PM, Ferruh Yigit wrote:
> On 9/25/2019 10:36 AM, Igor Ryzhov wrote:
>> Current implementation doesn't allow us to update KNI carrier if the
>> interface is not yet UP in kernel. It means that we can't use it in the
>> same thread which is processing rte_kni_ops.config_network_if,
On Mon, Oct 14, 2019 at 5:05 PM Bruce Richardson
wrote:
>
> In other places in DPDK, e.g. EAL, the architecture-specific files for arm
> are just stored in a folder called "arm". Rename the "arm64" folder for
> performance thread example to align to this naming.
>
> Signed-off-by: Bruce Richardson
On 9/25/2019 10:36 AM, Igor Ryzhov wrote:
> Current implementation doesn't allow us to update KNI carrier if the
> interface is not yet UP in kernel. It means that we can't use it in the
> same thread which is processing rte_kni_ops.config_network_if, which is
> very convenient, because it allows u
Hello guys,
The api rte_intr_ack that has been introduced to fix a race condition
observed with (at least) qede drivers/hw.
This is an experimental api in master but it still fixes a problem, so
I'd like to see this in stable branches.
Opinions?
Thanks.
--
David Marchand
On Wed, Aug 14, 2019 at 10:30 AM Phil Yang wrote:
>
> Enable both c11 atomic and non c11 atomic lock-free stack for aarch64.
>
> Introduced a new header to reduce the ifdef clutter across generic and c11
> files. The rte_stack_lf_stubs.h contains stub implementations of
> __rte_stack_lf_count, __r
On Wed, Aug 14, 2019 at 10:29 AM Phil Yang wrote:
>
> Add 128b atomic compare and swap test for aarch64 and x86_64.
>
> Signed-off-by: Phil Yang
> Reviewed-by: Honnappa Nagarahalli
> Acked-by: Gage Eads
> Acked-by: Jerin Jacob
> Tested-by: Jerin Jacob
> ---
> app/test/test_atomic.c | 125
>
On Wed, Aug 14, 2019 at 10:29 AM Phil Yang wrote:
>
> Add 128-bit atomic compare exchange on aarch64.
A bit short, seeing the complexity of the code and the additional
RTE_ARM_FEATURE_ATOMICS config flag.
Comments inline.
>
> Suggested-by: Jerin Jacob
> Signed-off-by: Phil Yang
> Reviewed-by
Add the 'Burst mode' section into command 'show rxq|txq info
' to show the Rx/Tx burst mode description like:
"Burst mode: Vector AVX2 Scattered"
Signed-off-by: Haiyue Wang
Acked-by: Bernard Iremonger
Reviewed-by: Xiaolong Ye
---
app/test-pmd/config.c | 29 +
1 f
Retrieve burst mode options according to the selected Rx/Tx burst
function name.
Signed-off-by: Haiyue Wang
Acked-by: Bernard Iremonger
Reviewed-by: Xiaolong Ye
---
doc/guides/nics/features/ice.ini | 1 +
drivers/net/ice/ice_ethdev.c | 2 ++
drivers/net/ice/ice_rxtx.c | 58
Retrieve burst mode options according to the selected Rx/Tx burst
function name.
Signed-off-by: Haiyue Wang
Acked-by: Bernard Iremonger
Reviewed-by: Xiaolong Ye
---
doc/guides/nics/features/i40e.ini | 1 +
drivers/net/i40e/i40e_ethdev.c| 2 +
drivers/net/i40e/i40e_ethdev.h| 4 ++
dr
Some PMDs have more than one RX/TX burst paths, add the ethdev API
that allows an application to retrieve the mode information about
Rx/Tx packet burst such as Scalar or Vector, and Vector technology
like AVX2.
Signed-off-by: Haiyue Wang
Acked-by: Bernard Iremonger
Reviewed-by: Xiaolong Ye
---
v2 -> v3:
1) Add 'Burst mode info' feature matrix.
v1 -> v2:
1). Change the return type of dev_ops callback from 'void' to
'int'.
2). Update the ice/i40 PMD patches with clear git commit log.
RFCv3 -> v1:
https://patchwork.dpdk.org/patch/59103/
The transmitter packets counter was not updated
correctly in the loop inside the tx_burst routines.
Fixes: f32a3f5216a3 ("net/mlx5: fix completion queue overflow for large burst")
Cc: sta...@dpdk.org
Signed-off-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5_rxtx.c | 13 +++--
1 file
On 14/10/2019 16:15, Andrew Rybchenko wrote:
> Hi,
>
> as far as I can see the patch introduces regressions.
>
> CC Kevin and Luca to be careful with stable branches patches.
>
Thanks Andrew. Just to confirm, it's not on any stable branches. Will
note not to backport until it is resolved on mas
On Mon, Oct 14, 2019 at 04:32:54PM +0200, David Marchand wrote:
> On Fri, Oct 11, 2019 at 10:56 PM Seth Howell wrote:
> >
> > The makefile in drivers/bus/pci specified rte_ethdev as a dependency for
> > the library. However there are no actual symbols from librte_ethdev used
> > in librte_bus_pci.
Hi,
as far as I can see the patch introduces regressions.
CC Kevin and Luca to be careful with stable branches patches.
See details below.
Andrew.
On 9/23/19 5:05 PM, Marvin Liu wrote:
If reserve virtio header room by function rte_pktmbuf_prepend, both
segment data length and packet length o
If the default dev args 'proto_xtr' is not in the first position, it
will overwrite the parsed queue map value, so use an new variable to
save the default.
And enhance the error message printing to show the right information.
Signed-off-by: Haiyue Wang
---
drivers/net/ice/ice_ethdev.c | 42
The original protocol extraction memory free is removed by introducing
new design for releasing port upon close.
Fixes: bd513ece3c40 ("net/ice: release port upon close")
Signed-off-by: Haiyue Wang
---
drivers/net/ice/ice_ethdev.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net
v1 -> v2:
Split into two patches, one is fixing a commit
Haiyue Wang (2):
net/ice: fix the missed memory free when dev is closed
net/ice: avoid the parsed devargs value being overwritten
drivers/net/ice/ice_ethdev.c | 44 +---
drivers/net/ice/ice_ethde
Aaron Conole writes:
> "Van Haaren, Harry" writes:
>
>>> -Original Message-
>>> From: Aaron Conole [mailto:acon...@redhat.com]
>>> Sent: Wednesday, September 4, 2019 8:56 PM
>>> To: David Marchand
>>> Cc: Van Haaren, Harry ; dev@dpdk.org
>>> Subject: Re: [dpdk-dev] [BUG] service_lcore_e
> On Oct 14, 2019, at 7:47 AM, David Marchand wrote:
>
> On Mon, Oct 14, 2019 at 3:49 PM Harris, James R
> wrote:
>> On 10/14/19, 4:18 AM, "David Marchand" wrote:
>>
>>>On Fri, Aug 16, 2019 at 9:19 PM Jim Harris
>>> wrote:
>>>
>>> The code checks both rte_mp_request_sync() return
>>
On Mon, Oct 14, 2019 at 3:49 PM Harris, James R
wrote:
> On 10/14/19, 4:18 AM, "David Marchand" wrote:
>
> On Fri, Aug 16, 2019 at 9:19 PM Jim Harris
> wrote:
> >
> > The code checks both rte_mp_request_sync() return
> > code and that the number of messages in the reply
> >
1 - 100 of 208 matches
Mail list logo
