>-Original Message-
>From: Thomas Monjalon [mailto:tho...@monjalon.net]
>Sent: 8. januar 2018 16:16
>To: Finn Christensen
>Cc: dev@dpdk.org
>Subject: Re: [dpdk-dev] Napatech pmd
>
>Hi,
>
>08/01/2018 14:08, Finn Christensen:
>> Hi Thomas,
>>
>> Thanks for bringing this discussion up again.
On 12/13/2017 12:05 PM, Hemant Agrawal wrote:
> Signed-off-by: Hemant Agrawal
<...>
> @@ -0,0 +1,37 @@
> +/*-
> + * Copyright 2017 NXP.
> + *
> + * SPDX-License-Identifier:BSD-3-Clause
I guess latest agreement was without break.
> + */
> +
> +#ifndef _PMD_DPAA_H_
> +#define _PMD_DP
On 12/13/2017 12:05 PM, Hemant Agrawal wrote:
> DPAA hardware support two kinds of queues:
> 1. Pull mode queue - where one needs to regularly pull the packets.
> 2. Push mode queue - where the hw pushes the packet to queue. These are
>high performance queues, but limitd in number.
limited
>
On 12/13/2017 12:05 PM, Hemant Agrawal wrote:
> From: Ashish Jain
>
> Setting correct frame size in dpaa_dev_mtu_set
> api call. Also setting correct max frame size in
> hardware in dev_configure for jumbo frames
>
> Signed-off-by: Ashish Jain
> Acked-by: Hemant Agrawal
<...>
> @@ -111,19 +1
On 12/13/2017 12:05 PM, Hemant Agrawal wrote:
> Fixes: 05ba55bc2b1a ("net/dpaa: add packet dump for debugging")
> Fixes: 37f9b54bd3cf ("net/dpaa: support Tx and Rx queue setup")
> Cc: sta...@dpdk.org>
> Signed-off-by: Hemant Agrawal
Hi Hemant,
fix coverity issues is not very helpful as commit ti
On 12/13/2017 12:05 PM, Hemant Agrawal wrote:
> This patch series add various improvement and performance related
> optimizations for DPAA PMD
>
> Ashish Jain (2):
> net/dpaa: fix the mbuf packet type if zero
> net/dpaa: set the correct frame size in device MTU
>
> Hemant Agrawal (11):
> ne
09/01/2018 11:31, Mordechay Haimovsky:
> From: Guo, Jia [mailto:jia@intel.com]
> > On 1/9/2018 8:39 AM, Thomas Monjalon wrote:
> > > At last there is the kernel binding effort - this one will probably
> > > be ignored for 18.02, because it is another huge topic.
> > > Without bothering with ker
Hi,
> -Original Message-
> From: Maxime Coquelin [mailto:maxime.coque...@redhat.com]
> Sent: Tuesday, January 9, 2018 4:50 PM
> To: Wang, Xiao W ; y...@fridaylinux.org
> Cc: Bie, Tiwei ; dev@dpdk.org;
> step...@networkplumber.org
> Subject: Re: [dpdk-dev] [PATCH v7 3/3] net/virtio: support
> From: Maxime Coquelin [mailto:maxime.coque...@redhat.com]
> On 01/09/2018 03:26 PM, Xiao Wang wrote:
> > When live migration is done, for the backup VM, either the virtio
> > frontend or the vhost backend needs to send out gratuitous RARP packet
> > to announce its new network location.
> >
> > T
On 12/11/2017 1:27 PM, Daly, Lee wrote:
>> Subject: [dpdk-dev] [PATCH 1/2] lib: refactor basic stats code
>>
>> Moved the code to get the basic stats names and values into static functions.
>>
>> Signed-off-by: Elza Mathew
>
> Reviewed-by: Lee Daly
Series applied to dpdk-next-net/master, thanks
On 12/18/2017 9:03 AM, santosh wrote:
>
> On Thursday 14 December 2017 02:01 PM, Pavan Nikhilesh wrote:
>> Advertise mempool/octeontx as the only supported mempool ops when the
>> application checks using `rte_eth_dev_pool_ops_supported`.
>>
>> Signed-off-by: Pavan Nikhilesh
>> ---
>
> Acked-by:
On 12/15/2017 5:51 PM, Ferruh Yigit wrote:
> On 12/15/2017 3:34 AM, Radu Nicolau wrote:
>> Functions like nl_recev and nl_send name clash functions in the
>> libnl library (https://www.infradead.org/~tgr/libnl/).
>> All functions declared in tap_netlink.h were decorated with tap_
>> for consistency
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Pablo de Lara
> Sent: Wednesday, December 13, 2017 1:14 PM
> To: Doherty, Declan
> Cc: dev@dpdk.org; De Lara Guarch, Pablo
> Subject: [dpdk-dev] [PATCH] app/crypto-perf: support IMIX
>
> Add support for IMIX pe
So far, about hot plug in dpdk, we already have hot plug add/remove
api and fail-safe driver to offload the fail-safe work from the app
user. But there are still lack of a general event api, since the interrupt
event, which hot plug related with, is diversity between each device and
driver, such as
use testpmd for example, to show app how to request and use
uevent monitoring to handle the hot removal event and the
hot insertion event.
Signed-off-by: Jeff Guo
---
v8->v7:
fix some defind issue
---
app/test-pmd/testpmd.c | 178 +
app/test-pmd/te
when the hot plug out of the device, the uio resource have been
invalid, so to avoid the system hung, no need to do the invalid
operation in igb_uio pci release function.
Signed-off-by: Jeff Guo
---
lib/librte_eal/linuxapp/igb_uio/igb_uio.c | 6 ++
1 file changed, 6 insertions(+)
diff --git
This patch aim to add a general uevent mechanism in eal device layer,
to enable all linux kernel object hot plug monitoring, so user could use
these APIs to monitor and read out the device status info that sent from
the kernel side, then corresponding to handle it, such as detach or attach
the devi
Hi Shahaf,
> -Original Message-
> From: Shahaf Shuler [mailto:shah...@mellanox.com]
> Sent: Tuesday, January 9, 2018 6:02 PM
> To: Lu, Wenzhuo ; Wu, Jingjing
> ; Yigit, Ferruh
> Cc: dev@dpdk.org
> Subject: RE: [dpdk-dev] [PATCH v3 04/10] app/testpmd: convert to new
> Ethdev Tx offloads AP
This patch fixes the reset of the service core,
that when rte_service_lcore_del() is called, the
lcore_role is restored to RTE.
This issue was reported as when running the unit tests, an
error was thrown that "failed to allocate lcore". Investigating
revealed that the state of the service-cores af
This patch fixes a potential bug, which was not consistently
showing up in the unit tests. The issue was that the service-
lcore being started was not in a "WAIT" state, and hence EAL
would return -EBUSY instead of launching the lcore.
In order to ensure a core is in a launch-ready state, the appl
09/01/2018 09:25, Guo, Jia:
> On 1/9/2018 8:39 AM, Thomas Monjalon wrote:
> >> +int
> >> +_rte_dev_callback_process(struct rte_device *device,
> >> + enum rte_eal_dev_event_type event,
> >> + void *cb_arg, void *ret_param)
> >
> > cb_arg must be an opaque paramete
change the xmit_hash functions to handle bursts of packet instead of
single packets at a time, and update the affected tx_burst functions.
Signed-off-by: Declan Doherty
Signed-off-by: Keith Wiles
---
V2: rebased to HEAD of next-net
drivers/net/bonding/rte_eth_bond_api.c | 3 -
drivers/ne
So, how can separate the patch into more small patch, use stake or null
implement in function. I think we should consider if it is a economic way now,
if I could explain more detail in code for you all not very familiar the
background? I have sent v8, please check, thanks all.
Best regards,
Je
09/01/2018 12:03, Wang, Xiao W:
> > From: Maxime Coquelin [mailto:maxime.coque...@redhat.com]
> > On 01/09/2018 03:26 PM, Xiao Wang wrote:
> > > +#define RARP_PKT_SIZE64
> > > +static int
> > > +make_rarp_packet(struct rte_mbuf *rarp_mbuf, const struct ether_addr
> > *mac)
> > > +{
> > > + stru
Hi Shahaf,
> -Original Message-
> From: Shahaf Shuler [mailto:shah...@mellanox.com]
> Sent: Tuesday, January 9, 2018 6:04 PM
> To: Lu, Wenzhuo ; Wu, Jingjing
> ; Yigit, Ferruh
> Cc: dev@dpdk.org
> Subject: RE: [dpdk-dev] [PATCH v3 10/10] app/testpmd: fix on the flight
> VLAN configuration
> From: Pavan Nikhilesh [mailto:pbhagavat...@caviumnetworks.com]
> Sent: Tuesday, January 9, 2018 9:41 AM
> To: Van Haaren, Harry
> Cc: dev@dpdk.org; Pavan Nikhilesh ;
> sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH] service: fix number mapped cores count
>
> When adding service the number of mapp
09/01/2018 12:39, Guo, Jia:
> So, how can separate the patch into more small patch, use stake or null
> implement in function. I think we should consider if it is a economic way
> now, if I could explain more detail in code for you all not very familiar the
> background? I have sent v8, please c
Got it, seems that would be more complex work to do later, still need to
debugging the feature more to fulfill for more driver.
Best regards,
Jeff Guo
From: Mordechay Haimovsky [mailto:mo...@mellanox.com]
Sent: Tuesday, January 9, 2018 6:32 PM
To: Guo, Jia ; Thomas Monjalon
Cc: dev@dpdk.org; st
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Tomasz Duszynski
> Sent: Tuesday, January 9, 2018 9:09 AM
> To: dev@dpdk.org
> Cc: m...@semihalf.com; d...@marvell.com; nsams...@marvell.com;
> jianbo@arm.com; Tomasz Duszynski
> Subject: [dpdk-dev] [PATCH v3
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Tomasz Duszynski
> Sent: Tuesday, January 9, 2018 9:09 AM
> To: dev@dpdk.org
> Cc: m...@semihalf.com; d...@marvell.com; nsams...@marvell.com;
> jianbo@arm.com; Tomasz Duszynski
> Subject: [dpdk-dev] [PATCH v3
> -Original Message-
> From: De Lara Guarch, Pablo
> Sent: Wednesday, December 13, 2017 3:55 PM
> To: Doherty, Declan
> Cc: dev@dpdk.org; De Lara Guarch, Pablo
> ; sta...@dpdk.org
> Subject: [PATCH] cryptodev: fix function prototype
>
> qp_detach_session function was using the attach_se
> -Original Message-
> From: De Lara Guarch, Pablo
> Sent: Wednesday, December 13, 2017 3:55 PM
> To: Doherty, Declan
> Cc: dev@dpdk.org; De Lara Guarch, Pablo
> ; sta...@dpdk.org
> Subject: [PATCH] cryptodev: add missing CPU flag string
>
> CPU flag AVX512 was added in a previous relea
Hi Shahaf,
> -Original Message-
> From: Shahaf Shuler [mailto:shah...@mellanox.com]
> Sent: Tuesday, January 9, 2018 6:08 PM
> To: Lu, Wenzhuo ; Wu, Jingjing
> ; Yigit, Ferruh
> Cc: dev@dpdk.org
> Subject: RE: [dpdk-dev] [PATCH v3 08/10] app/testpmd: remove txqflags
>
> Tuesday, January
On 1/9/2018 7:38 PM, Thomas Monjalon wrote:
09/01/2018 09:25, Guo, Jia:
On 1/9/2018 8:39 AM, Thomas Monjalon wrote:
+int
+_rte_dev_callback_process(struct rte_device *device,
+ enum rte_eal_dev_event_type event,
+ void *cb_arg, void *ret_param)
cb_
Your comments about split it totally make sense ,no doubt that, but my question
is that if split api with the funcational , so the function part should be set
null implement or stake. Any other good idea or tip for that.
Best regards,
Jeff Guo
-Original Message-
From: Thomas Monjalon
From: Stephen Hemminger
All PMD should be using dynamic log levels.
Signed-off-by: Stephen Hemminger
---
config/common_base | 1 -
drivers/net/avp/avp_ethdev.c | 10 ++
drivers/net/avp/avp_logs.h | 9 -
3 files changed, 14 insertions(+), 6 deletions(-)
diff --git
From: Stephen Hemminger
Signed-off-by: Stephen Hemminger
---
config/common_base | 2 --
drivers/net/e1000/e1000_logs.h | 16 ++--
drivers/net/e1000/em_ethdev.c | 15 +++
3 files changed, 21 insertions(+), 12 deletions(-)
diff --git a/config/common_base b/c
From: Stephen Hemminger
Replace compile time option for init and driver log with
dynamic value.
Signed-off-by: Stephen Hemminger
Signed-off-by: Ferruh Yigit
---
v2: remove remaining RTE_LIBRTE_BNX2X_DEBUG usage
---
config/common_base | 2 --
doc/guides/nics/bnx2x.rst|
From: Stephen Hemminger
Convert all drivers possible to dynamic logging.
Signed-off-by: Stephen Hemminger
---
config/common_base | 2 --
drivers/net/ixgbe/ixgbe_ethdev.c | 15 +++
drivers/net/ixgbe/ixgbe_logs.h | 16 ++--
3 files changed, 21 insertions(
From: Stephen Hemminger
Replace compile time configuration with runtime.
Signed-off-by: Stephen Hemminger
---
config/common_base | 2 --
drivers/net/vmxnet3/vmxnet3_ethdev.c | 15 +++
drivers/net/vmxnet3/vmxnet3_logs.h | 17 ++---
3 files changed, 2
build error:
.../drivers/net/nfp/nfp_net.c:1497:8: error:
‘dev_data’ undeclared (first use in this function); did you mean
‘dev_t’?
dev_data->port_id);
^
Fixes: 7ccf1115f0ab ("net/nfp: fix jumbo settings")
Cc: sta...@dpdk.org
Signed-off-by: Ferruh Yigit
---
drivers/net/n
From: Stephen Hemminger
Signed-off-by: Stephen Hemminger
Tested-by: Maxime Coquelin
Reviewed-by: Maxime Coquelin
---
config/common_base | 2 --
drivers/net/virtio/virtio_ethdev.c | 15 +++
drivers/net/virtio/virtio_logs.h | 19 +++
3 files change
From: Stephen Hemminger
Signed-off-by: Stephen Hemminger
---
config/common_base | 2 --
drivers/net/fm10k/fm10k_ethdev.c | 15 +++
drivers/net/fm10k/fm10k_logs.h | 16 ++--
3 files changed, 21 insertions(+), 12 deletions(-)
diff --git a/config/common_ba
From: Stephen Hemminger
Use dynamic rather static configuration for logging.
Also include ena_assert_msg.
Signed-off-by: Stephen Hemminger
---
config/common_base | 1 -
drivers/net/ena/ena_ethdev.c | 15 +++
drivers/net/ena/ena_logs.h | 14 ++
drivers
From: Stephen Hemminger
Signed-off-by: Stephen Hemminger
Signed-off-by: Ferruh Yigit
---
v2: remove unsed macro RTE_LIBRTE_NFP_NET_DEBUG_INIT
---
config/common_base | 3 ++-
drivers/net/nfp/nfp_net.c | 14 ++
drivers/net/nfp/nfp_net_logs.h | 24 ++
From: Stephen Hemminger
This driver is mostly like others with slightly different logging
macros. The semantics were retained, with some minor reformatting.
Signed-off-by: Stephen Hemminger
---
config/common_base | 2 --
drivers/net/qede/qede_ethdev.c | 15
drivers/ne
From: Stephen Hemminger
This driver mostly uses the common pattern. Convert this to the
dynamic logging.
Signed-off-by: Stephen Hemminger
---
config/common_base| 2 --
drivers/net/liquidio/lio_ethdev.c | 15 +++
drivers/net/liquidio/lio_logs.h | 20 --
On Tue, Jan 09, 2018 at 11:38:04AM +, Harry van Haaren wrote:
> This patch fixes the reset of the service core,
> that when rte_service_lcore_del() is called, the
> lcore_role is restored to RTE.
>
Title seems awkward, how about "fix lcore role after delete"?
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Lu, Wenzhuo
> Sent: Tuesday, January 9, 2018 11:37 AM
> To: Shahaf Shuler ; Wu, Jingjing
> ; Yigit, Ferruh
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 04/10] app/testpmd: convert to new Ethdev
> Tx o
OPDL ring is the core infrastructure of OPDL PMD. OPDL ring library
provide the core data structure and core helper function set. The Ring
implements a single ring multi-port/stage pipelined packet distribution
mechanism. This mechanism has the following characteristics:
• No multiple queue cost,
The OPDL (Ordered Packet Distribution Library) eventdev is a specific
implementation of the eventdev API. It is particularly suited to packet
processing workloads that have high throughput and low latency
requirements. All packets follow the same path through the device.
The order which packets f
This commit adds a OPDL implementation of the eventdev API. The
implementation here is intended to enable the community to use
the OPDL infrastructure under eventdev API.
The main components of the implementation is three files:
- opdl_evdev.c Creation, configuration, etc
- opdl_e
Signed-off-by: Liang Ma
Signed-off-by: Peter Mccarthy
---
drivers/event/opdl/opdl_evdev.c | 108
1 file changed, 108 insertions(+)
diff --git a/drivers/event/opdl/opdl_evdev.c b/drivers/event/opdl/opdl_evdev.c
index cad000a..11ac8fc 100644
--- a/drivers/
Signed-off-by: Liang Ma
Signed-off-by: Peter Mccarthy
---
drivers/event/opdl/opdl_evdev.c | 152
1 file changed, 152 insertions(+)
diff --git a/drivers/event/opdl/opdl_evdev.c b/drivers/event/opdl/opdl_evdev.c
index 11ac8fc..56e6910 100644
--- a/drivers/
Signed-off-by: Liang Ma
Signed-off-by: Peter Mccarthy
---
drivers/event/opdl/opdl_evdev.c | 66 +
1 file changed, 66 insertions(+)
diff --git a/drivers/event/opdl/opdl_evdev.c b/drivers/event/opdl/opdl_evdev.c
index 56e6910..c42e813 100644
--- a/drivers/e
This commit adds unit test inside the OPDL PMD. There is a PMd parameter
"self_test" can be used to triger the test when vdev bus probe opdl device
e.g.
sudo ./app/test --vdev="event_opdl0,self_test=1"
Signed-off-by: Liang Ma
Signed-off-by: Peter Mccarthy
---
drivers/event/opdl/Makefile
Signed-off-by: Liang Ma
Signed-off-by: Peter Mccarthy
---
drivers/event/dpaa2/dpaa2_eventdev.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/event/dpaa2/dpaa2_eventdev.c
b/drivers/event/dpaa2/dpaa2_eventdev.c
index 13e7122..e437edc 100644
--- a/drivers/event/d
Signed-off-by: Liang Ma
Signed-off-by: Peter Mccarthy
---
drivers/event/octeontx/ssovf_evdev.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/event/octeontx/ssovf_evdev.c
b/drivers/event/octeontx/ssovf_evdev.c
index b80a6c0..d85b4fb 100644
--- a/drivers/event/o
this commitd add three new eventdev capability flags
RTE_EVENT_DEV_CAP_NONSEQ_MODE
Event device is capable of operating in none sequential mode. The path
of the event is not necessary to be sequential. Application can change
the path of event at runtime.if the flag is not set, then event each eve
Signed-off-by: Liang Ma
Signed-off-by: Peter Mccarthy
---
MAINTAINERS| 6 ++
doc/guides/rel_notes/release_18_02.rst | 11 +++
2 files changed, 17 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index f0baeb4..1b8d617 100644
--- a/MAINTAINERS
+++ b/M
Signed-off-by: Liang Ma
Signed-off-by: Peter Mccarthy
---
drivers/event/sw/sw_evdev.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/event/sw/sw_evdev.c b/drivers/event/sw/sw_evdev.c
index 1ef6340..aed521b 100644
--- a/drivers/event/sw/sw_evdev.c
+++ b/drivers/ev
On 1/9/2018 12:10 PM, Ferruh Yigit wrote:
> From: Stephen Hemminger
>
> All PMD should be using dynamic log levels.
>
> Signed-off-by: Stephen Hemminger
Series applied to dpdk-next-net/master, thanks.
Add the description about opdl pmd
Signed-off-by: Liang Ma
Signed-off-by: Peter Mccarthy
---
doc/guides/eventdevs/index.rst | 1 +
doc/guides/eventdevs/opdl.rst | 162 +
2 files changed, 163 insertions(+)
create mode 100644 doc/guides/eventdevs/opdl.r
On 12/19/2017 12:25 PM, Pavan Nikhilesh wrote:
> Implement dynamic logging for cavium specific thunder nicvf and octeontx
> platforms.
>
> Pavan Nikhilesh (5):
> net/thunderx: implement dynamic logging
> net/octeontx: implement dynamic logging
> mempool/octeontx: implement dynamic logging
>
-Original Message-
> Date: Tue, 9 Jan 2018 12:20:18 +
> From: Liang Ma
> To: jerin.ja...@caviumnetworks.com
> CC: dev@dpdk.org, harry.van.haa...@intel.com, bruce.richard...@intel.com,
> deepak.k.j...@intel.com, john.ge...@intel.com, peter.mccar...@intel.com,
> sea...@gmail.com
> Subj
On Tue, Jan 09, 2018 at 10:20:05AM +0100, Thomas Monjalon wrote:
> 09/01/2018 02:32, Neil Horman:
> > On Fri, Jan 05, 2018 at 11:00:52AM -0500, Neil Horman wrote:
> > > On Fri, Jan 05, 2018 at 03:08:52PM +0100, Thomas Monjalon wrote:
> > > > 04/01/2018 13:56, Neil Horman:
> > > > > On Sat, Dec 30,
Hi Yanglong,
Sorry, I was just confused by the description of the patch.
The code itself seems ok... probably need to rephrase the description
(remove mentioning l3fwd?).
Konstantin
> Hi, Konstantin
>
> Thanks for your comments!
> Do you means that tx_q is must less than rx_q when the SRIOV i
Hi Jeff,
On Tue, Jan 09, 2018 at 12:08:52PM +, Guo, Jia wrote:
> Your comments about split it totally make sense ,no doubt that, but my
> question is that if split api with the funcational , so the function part
> should be set null implement or stake. Any other good idea or tip for that.
>
On 12/19/2017 6:01 PM, Pavan Nikhilesh wrote:
> The channel to port id map is used by event octeontx to map the received
> wqe to the respective ethdev port.
>
> Signed-off-by: Pavan Nikhilesh
> Acked-by: Santosh Shukla
Series applied to dpdk-next-net/master, thanks.
> +DPDK_18.04 {
Fixed whi
On 1/9/2018 12:47 PM, Ferruh Yigit wrote:
> On 12/19/2017 6:01 PM, Pavan Nikhilesh wrote:
>> The channel to port id map is used by event octeontx to map the received
>> wqe to the respective ethdev port.
>>
>> Signed-off-by: Pavan Nikhilesh
>> Acked-by: Santosh Shukla
>
> Series applied to dpdk-
Hi Anoob,
On Fri, Jan 05, 2018 at 11:48:50AM +0530, Anoob Joseph wrote:
> Hi Adrien,
>
>
> On 12/21/2017 07:52 PM, Adrien Mazarguil wrote:
> > On Thu, Dec 21, 2017 at 12:12:29PM +0200, Boris Pismenny wrote:
> >
> > > On 12/21/2017 10:06 AM, Anoob Joseph wrote:
> >
> > > > I can see the benefit
> -Original Message-
> From: Zhang, Roy Fan
> Sent: Tuesday, January 9, 2018 11:26 AM
> To: De Lara Guarch, Pablo ; Doherty,
> Declan
> Cc: dev@dpdk.org; De Lara Guarch, Pablo
>
> Subject: RE: [dpdk-dev] [PATCH] app/crypto-perf: support IMIX
>
>
>
> > -Original Message-
> > F
On Tue, Jan 09, 2018 at 07:34:17AM +, Yao, Lei A wrote:
>
>
> > -Original Message-
> > From: Chen, Junjie J
> > Sent: Tuesday, January 9, 2018 7:04 PM
> > To: y...@fridaylinux.org; maxime.coque...@redhat.com; Wang, Xiao W
> > ; Bie, Tiwei ; Yao, Lei A
> >
> > Cc: dev@dpdk.org; Chen,
On Fri, Jan 05, 2018 at 06:10:39AM -0800, Tonghao Zhang wrote:
> This patch adds the name for vhost-user reconnect thread.
> It can help us to know whether the thread is running.
>
> Signed-off-by: Tonghao Zhang
> ---
> lib/librte_vhost/socket.c | 8
> 1 file changed, 8 insertions(+)
>
Signed-off-by: Ferruh Yigit
---
v2: Changed syntax for SPDX header
---
doc/guides/nics/pcap_ring.rst | 31 ++-
drivers/net/pcap/Makefile | 35 ---
drivers/net/pcap/rte_eth_pcap.c | 36
3 files
On Fri, Jan 05, 2018 at 02:28:06AM -0800, Zhike Wang wrote:
> From: zhike wang
>
> The max_mtu is kept as zero in case no CRTL channel, which leads
> to failure when calling virtio_mtu_set().
>
> Signed-off-by: Zhike Wang
Applied to dpdk-next-virtio.
Thanks.
--yliu
> ---
> drivers/n
Intel VT-d supports different address widths for the IOVAs, from
39 bits to 56 bits.
While recent processors support at least 48 bits, VT-d emulation
currently only supports 39 bits. It makes DMA mapping to fail in this
case when using VA as IOVA mode, as user-space virtual addresses uses
up to 47
On Thu, Jan 04, 2018 at 02:33:32PM +0800, Zhiyong Yang wrote:
> When binding X710 NIC (i40e driver) to DPDK, vhost sample startups
> failure.
> The sample requires that MAX_QUEUES should be defined no less than 320.
> So, the patch redefines MAX_QUEUES 320 to fix the issue.
It just makes the issue
This patch series add various improvement and performance related
optimizations for DPAA PMD
v2:
- fix the spelling of PORTALS
- Add Akhil's patch wich is required for crypto
- minor improvement in push mode patch
Akhil Goyal (1):
bus/dpaa: support for enqueue frames of multiple queues
Ash
From: Ashish Jain
Populate the mbuf field packet_type which is required
for calculating checksum while transmitting frames
Fixes: 8cffdcbe85aa ("net/dpaa: support scattered Rx")
Cc: sta...@dpdk.org
Signed-off-by: Ashish Jain
Signed-off-by: Hemant Agrawal
---
drivers/net/dpaa/dpaa_rxtx.c | 19
This patch update the logic and expose the soc value
register, so that it can be used by other modules as well.
Signed-off-by: Hemant Agrawal
---
drivers/bus/dpaa/dpaa_bus.c | 12
drivers/bus/dpaa/rte_bus_dpaa_version.map | 8
drivers/bus/dpaa/rte_dpaa_bus.h
From: Nipun Gupta
The settings are tuned for performance.
Signed-off-by: Nipun Gupta
Acked-by: Hemant Agrawal
---
drivers/bus/dpaa/base/qbman/qman.c | 10 --
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/drivers/bus/dpaa/base/qbman/qman.c
b/drivers/bus/dpaa/base/qbman
Replace the byte based tail queue congestion support
with frame count based congestion groups.
It can easily map to number of RX descriptors for a queue.
Signed-off-by: Hemant Agrawal
---
drivers/bus/dpaa/rte_bus_dpaa_version.map | 5 ++
drivers/net/dpaa/dpaa_ethdev.c| 98 ++
HW portal is a processing context in DPAA. This patch allow
creation of a queue specific HW portal context.
Signed-off-by: Hemant Agrawal
---
drivers/bus/dpaa/base/qbman/qman.c| 69 --
drivers/bus/dpaa/base/qbman/qman_driver.c | 153 +-
drivers/bu
From: Ashish Jain
Setting correct frame size in dpaa_dev_mtu_set
api call. Also setting correct max frame size in
hardware in dev_configure for jumbo frames
Signed-off-by: Ashish Jain
Acked-by: Hemant Agrawal
---
drivers/net/dpaa/dpaa_ethdev.c | 20 +---
drivers/net/dpaa/dpaa_
From: Nipun Gupta
Signed-off-by: Nipun Gupta
Acked-by: Hemant Agrawal
---
drivers/bus/dpaa/base/qbman/qman.c | 7 ---
drivers/bus/dpaa/include/fsl_qman.h | 2 ++
2 files changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers/bus/dpaa/base/qbman/qman.c
b/drivers/bus/dpaa/base/qbma
From: Nipun Gupta
change the TX budget to 7 to sync best with the hw.
Signed-off-by: Nipun Gupta
Acked-by: Hemant Agrawal
---
drivers/net/dpaa/dpaa_ethdev.h | 2 +-
drivers/net/dpaa/dpaa_rxtx.c | 5 +++--
2 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/net/dpaa/dpaa_e
From: Nipun Gupta
Optimize it for best case. Create a function
for TX offloads to be used in multiple legs.
Signed-off-by: Nipun Gupta
Acked-by: Hemant Agrawal
---
drivers/net/dpaa/dpaa_rxtx.c | 73
1 file changed, 46 insertions(+), 27 deletions(-)
From: Nipun Gupta
Signed-off-by: Nipun Gupta
Signed-off-by: Hemant Agrawal
---
drivers/net/dpaa/dpaa_rxtx.c | 48
drivers/net/dpaa/dpaa_rxtx.h | 2 +-
2 files changed, 23 insertions(+), 27 deletions(-)
diff --git a/drivers/net/dpaa/dpaa_rxtx.c b/d
Signed-off-by: Hemant Agrawal
---
drivers/bus/dpaa/base/qbman/qman.c| 22 ++
drivers/bus/dpaa/include/fsl_qman.h | 7 +++
drivers/bus/dpaa/rte_bus_dpaa_version.map | 1 +
3 files changed, 30 insertions(+)
diff --git a/drivers/bus/dpaa/base/qbman/qman.c
b/
Signed-off-by: Hemant Agrawal
---
app/test-pmd/Makefile | 4
app/test-pmd/cmdline.c | 7 +++
2 files changed, 11 insertions(+)
diff --git a/app/test-pmd/Makefile b/app/test-pmd/Makefile
index 82b3481..34125e5 100644
--- a/app/test-pmd/Makefile
+++ b/app/test-pmd/Makefile
@@ -43,6 +43,1
Signed-off-by: Hemant Agrawal
---
drivers/net/dpaa/dpaa_ethdev.c | 17 +
1 file changed, 17 insertions(+)
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index 5d94af5..de016ab 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_et
Signed-off-by: Hemant Agrawal
---
drivers/net/dpaa/Makefile | 3 +++
drivers/net/dpaa/dpaa_ethdev.c| 42 +++
drivers/net/dpaa/rte_pmd_dpaa.h | 35 ++
drivers/net/dpaa/rte_pmd_dpaa_version.map | 8 ++
4
fix the soc id path and missing fclose
Fixes: cf0fab1d2ca5 ("net/dpaa: support firmware version get API")
Cc: sta...@dpdk.org
Signed-off-by: Hemant Agrawal
---
drivers/net/dpaa/dpaa_ethdev.c | 14 +-
drivers/net/dpaa/dpaa_ethdev.h | 2 +-
2 files changed, 6 insertions(+), 10 deleti
DPAA hardware support two kinds of queues:
1. Pull mode queue - where one needs to regularly pull the packets.
2. Push mode queue - where the hw pushes the packet to queue. These are
high performance queues, but limitd in number.
This patch add the driver support for push m de queues.
Signed-o
v8:
- Add a helper in lib/librte_net to make rarp packet, it's used by
both vhost and virtio.
v7:
- Improve comment for state_lock.
- Rename spinlock variable 'sl' to 'lock'.
v6:
- Use rte_pktmbuf_alloc() instead of rte_mbuf_raw_alloc().
- Remove the 'len' parameter in calling virtio_send_comma
From: Akhil Goyal
Signed-off-by: Akhil Goyal
Signed-off-by: Nipun Gupta
Acked-by: Hemant Agrawal
---
drivers/bus/dpaa/base/qbman/qman.c| 66 +++
drivers/bus/dpaa/include/fsl_qman.h | 14 +++
drivers/bus/dpaa/rte_bus_dpaa_version.map | 1 +
3 file
Signed-off-by: Sunil Kumar Kori
Signed-off-by: Hemant Agrawal
Signed-off-by: Nipun Gupta
---
doc/guides/nics/dpaa.rst | 11
drivers/net/dpaa/dpaa_ethdev.c | 64 +-
drivers/net/dpaa/dpaa_ethdev.h | 2 +-
drivers/net/dpaa/dpaa_rxtx.c | 34
The virtio_send_command function may be called from app's configuration
routine, but also from an interrupt handler called when live migration is
done on the backup side. So this patch makes control queue thread-safe
first.
Signed-off-by: Xiao Wang
---
drivers/net/virtio/virtio_ethdev.c | 7
Suggested-by: Maxime Coquelin
Signed-off-by: Xiao Wang
---
lib/librte_net/Makefile| 1 +
lib/librte_net/rte_arp.c | 42 ++
lib/librte_net/rte_arp.h | 14 +
lib/librte_net/rte_net_version.map | 6 ++
4 files cha
1 - 100 of 404 matches
Mail list logo
