DSW limits the rate of migrations on a per-port basis. Hence, as the
number of cores grows, so does the total migration capacity.
In high core-count systems, this allows for a situation where flows
are migrated to a lightly loaded port which recently already received
a number of new flows (from ot
Performance and statistics improvements for the distributed software
(DSW) event device.
Mattias Rönnblom (8):
event/dsw: reduce latency in low-load situations
event/dsw: reduce max flows to speed up load balancing
event/dsw: extend statistics
event/dsw: improve migration mechanism
event
Reduce the maximum number of DSW flows from 32k to 8k, to be able
rebalance load faster.
Signed-off-by: Mattias Rönnblom
---
drivers/event/dsw/dsw_evdev.h | 16 ++--
1 file changed, 14 insertions(+), 2 deletions(-)
diff --git a/drivers/event/dsw/dsw_evdev.h b/drivers/event/dsw/dsw_e
Allowing moving multiple flows in one migration transaction, to
rebalance load more quickly.
Introduce a threshold to avoid migrating flows between ports with very
similar load.
Simplify logic for selecting which flow to migrate. The aim is now to
move flows in such a way that the receiving port
On dequeue, polling the control ring once is enough.
Fixes: f6257b22e767 ("event/dsw: add load balancing")
Suggested-by: Ola Liljedahl
Signed-off-by: Mattias Rönnblom
---
drivers/event/dsw/dsw_event.c | 5 -
1 file changed, 5 deletions(-)
diff --git a/drivers/event/dsw/dsw_event.c b/drive
DSW keeps an internal port load estimate, used by the load balancing
mechanism. As a side effect, it keeps track of the total number of
busy cycles since startup. This metric is indirectly exposed in the
form of DSW xstats' "port__event_proc_latency", which is the total
number of busy cycles divide
On Mon, Mar 9, 2020 at 11:47 AM Sunil Kumar Kori wrote:
>
> If user passes log-level eal parameter to enable log level based on regex
> then in case of error message is being printed for pattern match instead of
> regex. Following is the warning message thrown:
>
> Compiling C object
> 'lib/76b5a
Extend DSW xstats.
To allow visualization of migrations, track the number flow
immigrations in "port__immigrations". The "port__migrations"
retains legacy semantics, but is renamed "port__emigrations".
Expose the number of events currently undergoing processing
(i.e. pending releases) at a partic
In DSW, in case a port can't produce any events for the application to
consume, the port is considered idle.
To slightly reduce wall-time latency, flush the port's output buffer
in case of such an empty dequeue.
Signed-off-by: Mattias Rönnblom
---
drivers/event/dsw/dsw_event.c | 10 +-
Remove unnecessary read barrier (and misleading comment) on control
message dequeue.
Fixes: f6257b22e767 ("event/dsw: add load balancing")
Suggested-by: Ola Liljedahl
Signed-off-by: Mattias Rönnblom
---
drivers/event/dsw/dsw_event.c | 5 -
1 file changed, 5 deletions(-)
diff --git a/drive
> -Original Message-
> From: Mattias Rönnblom
> Sent: Monday, March 9, 2020 12:21 PM
> To: Jerin Jacob Kollanukkaran
> Cc: dev@dpdk.org; stefan.sundkv...@ericsson.com; ola.liljed...@arm.com;
> Mattias Rönnblom
> Subject: [PATCH 5/8] event/dsw: avoid migration waves in large systems
>
>
On Sat, Mar 7, 2020 at 8:34 PM Gavin Hu wrote:
>
> Hi Jerin,
>
> > -Original Message-
> > From: Jerin Jacob
> > Sent: Friday, March 6, 2020 3:45 PM
> > To: Gavin Hu
> > Cc: dpdk-dev ; nd ; David Marchand
> > ; tho...@monjalon.net;
> > jer...@marvell.com; Ye, Xiaolong ; Honnappa
> > Nagar
Any DPDK public header file which includes stdbool.h may conflict with
local definition of bool, if any, which further results in compilation
error. To avoid, used standard stdbool.h instead of defining bool
internally.
I observed this issue during a development where I included rte_uuid.h
into rt
> -Original Message-
> From: Mattias Rönnblom
> Sent: Monday, March 9, 2020 1:28 PM
> To: Jerin Jacob Kollanukkaran
> Cc: dev@dpdk.org; Stefan Sundkvist ;
> ola.liljed...@arm.com
> Subject: [EXT] Re: [PATCH 5/8] event/dsw: avoid migration waves in large
> systems
>
> On 2020-03-09 08:17,
On 2020-03-09 08:17, Jerin Jacob Kollanukkaran wrote:
>> -Original Message-
>> From: Mattias Rönnblom
>> Sent: Monday, March 9, 2020 12:21 PM
>> To: Jerin Jacob Kollanukkaran
>> Cc: dev@dpdk.org; stefan.sundkv...@ericsson.com; ola.liljed...@arm.com;
>> Mattias Rönnblom
>> Subject: [PATC
On Mon, Mar 9, 2020 at 1:23 PM Sunil Kumar Kori wrote:
>
> Any DPDK public header file which includes stdbool.h may conflict with
> local definition of bool, if any, which further results in compilation
> error. To avoid, used standard stdbool.h instead of defining bool
> internally.
>
> I observe
Hi,
On Mon, Mar 09, 2020 at 11:01:25AM +0800, Tonghao Zhang wrote:
> On Sat, Mar 7, 2020 at 8:54 PM Andrew Rybchenko
> wrote:
> >
> > On 3/7/20 3:51 PM, Andrew Rybchenko wrote:
> > > On 3/6/20 4:37 PM, Jerin Jacob wrote:
> > >> On Fri, Mar 6, 2020 at 7:06 PM wrote:
> > >>> From: Tonghao Zhang
>
From: Alvin Zhang
Below ops are added too:
mac_addr_add
mac_addr_remove
mac_addr_set
set_mc_addr_list
mtu_set
promiscuous_enable
promiscuous_disable
allmulticast_enable
allmulticast_disable
rx_queue_setup
rx_queue_release
rx_queue_count
rx_descriptor_done
rx_descriptor_status
tx_descriptor_status
From: Alvin Zhang
Add functions and definitions that are OS specified.
Add readme too.
Signed-off-by: Alvin Zhang
---
drivers/net/igc/Makefile | 46 +++
drivers/net/igc/base/README| 23
drivers/net/igc/base/e1000_osdep.c | 64 +
drivers/net/igc/base/e1000_
From: Alvin Zhang
Implement device detection and loading.
Signed-off-by: Alvin Zhang
---
MAINTAINERS | 7 +
config/common_base | 7 +
doc/guides/nics/features/igc.ini| 8 +
doc/guides/nics/igc.rst | 39 +
doc/g
From: Alvin Zhang
Bellow ops are implemented:
dev_configure
dev_start
dev_stop
dev_close
dev_reset
dev_set_link_up
dev_set_link_down
link_update
fw_version_get
dev_led_on
dev_led_off
Signed-off-by: Alvin Zhang
---
doc/guides/nics/features/igc.ini | 4 +
drivers/net/igc/igc_ethdev.c | 644
From: Alvin Zhang
Implement base status, extend status and per queue status API.
Below ops are added:
stats_get
xstats_get
xstats_get_by_id
xstats_get_names_by_id
xstats_get_names
stats_reset
xstats_reset
queue_stats_mapping_set
Signed-off-by: Alvin Zhang
---
doc/guides/nics/features/igc.ini
From: Alvin Zhang
Below ops ware added:
vlan_filter_set
vlan_offload_set
vlan_tpid_set
vlan_strip_queue_set
Signed-off-by: Alvin Zhang
---
doc/guides/nics/features/igc.ini | 2 +
drivers/net/igc/igc_ethdev.c | 169 +++
drivers/net/igc/igc_ethdev.h
From: Alvin Zhang
Setup NIC to generate MSI-X interrupts.
Set the IVAR register to map interrupt causes to vectors.
Implement interrupt enable/disable functions.
Signed-off-by: Alvin Zhang
---
doc/guides/nics/features/igc.ini | 1 +
drivers/net/igc/igc_ethdev.c | 170
From: Alvin Zhang
Below ops are added:
reta_update
reta_query
rss_hash_update
rss_hash_conf_get
Signed-off-by: Alvin Zhang
---
doc/guides/nics/features/igc.ini | 2 +
drivers/net/igc/igc_ethdev.c | 171 +++
drivers/net/igc/igc_ethdev.h | 9 +++
d
From: Alvin Zhang
Update feature list too.
Signed-off-by: Alvin Zhang
---
doc/guides/nics/features/igc.ini | 1 +
drivers/net/igc/igc_ethdev.c | 121 +++
2 files changed, 122 insertions(+)
diff --git a/doc/guides/nics/features/igc.ini b/doc/guides/nic
From: Alvin Zhang
Update feature list too.
Signed-off-by: Alvin Zhang
---
doc/guides/nics/features/igc.ini | 1 +
drivers/net/igc/Makefile | 1 +
drivers/net/igc/igc_ethdev.c | 5 +
drivers/net/igc/igc_ethdev.h | 15 +++
drivers/net/igc/igc_filter.c | 237 +++
From: Alvin Zhang
Support configure of hash filter.
Signed-off-by: Alvin Zhang
---
drivers/net/igc/igc_filter.c | 155 +++
drivers/net/igc/igc_txrx.c | 77 -
drivers/net/igc/igc_txrx.h | 4 ++
3 files changed, 235 insertions(+)
From: Alvin Zhang
Add L3 protocol type and L4 destination port filter.
Signed-off-by: Alvin Zhang
---
drivers/net/igc/igc_ethdev.h | 38 +
drivers/net/igc/igc_filter.c | 341 +++
drivers/net/igc/igc_filter.h | 3 +
3 files changed, 382 insertions(
From: Alvin Zhang
Support putting all TCP SYN packets into a specified queue.
Signed-off-by: Alvin Zhang
---
drivers/net/igc/igc_ethdev.h | 18 ++
drivers/net/igc/igc_filter.c | 129 +++
drivers/net/igc/igc_filter.h | 3 +
3 files changed, 150 ins
From: Alvin Zhang
Below type of flows are supported:
ether-type filter,
2-tuple filter,
SYN filter,
RSS
Signed-off-by: Alvin Zhang
---
drivers/net/igc/Makefile | 1 +
drivers/net/igc/igc_ethdev.c | 3 +
drivers/net/igc/igc_ethdev.h | 27 ++
drivers/net/igc/igc_filter.c | 7 +
driver
Hi, Alvin
Thanks for the patch, before going through the whole series, one comment is
that for such big patch set, better to send it with cover letter which can give
some background, simple intro about the patch structure and changlog as well
for later version.
Thanks,
Xiaolong
On 03/09, alvinx.
On 3/9/2020 8:16 AM, Jerin Jacob wrote:
> On Mon, Mar 9, 2020 at 1:23 PM Sunil Kumar Kori wrote:
>>
>> Any DPDK public header file which includes stdbool.h may conflict with
>> local definition of bool, if any, which further results in compilation
>> error. To avoid, used standard stdbool.h instea
With this patch it is possible to connect 2 DPDK memifs into loopback,
i.e. when they have the same id and different roles, as for example:
"--vdev=net_memif0,role=master,id=0"
"--vdev=net_memif1,role=slave,id=0"
Signed-off-by: Július Milan
---
drivers/net/memif/memif_socket.c | 30
On 3/7/2020 3:56 PM, Gavin Hu wrote:
> Declaring zero-length arrays in other contexts, including as interior
> members of structure objects or as non-member objects, is discouraged.
> Accessing elements of zero-length arrays declared in such contexts is
> undefined and may be diagnosed.[1]
>
> Fix
On Mon, Mar 9, 2020 at 4:27 PM Olivier Matz wrote:
>
> Hi,
>
> On Mon, Mar 09, 2020 at 11:01:25AM +0800, Tonghao Zhang wrote:
> > On Sat, Mar 7, 2020 at 8:54 PM Andrew Rybchenko
> > wrote:
> > >
> > > On 3/7/20 3:51 PM, Andrew Rybchenko wrote:
> > > > On 3/6/20 4:37 PM, Jerin Jacob wrote:
> > > >
On 2020-03-09 09:12, Jerin Jacob Kollanukkaran wrote:
>> -Original Message-
>> From: Mattias Rönnblom
>> Sent: Monday, March 9, 2020 1:28 PM
>> To: Jerin Jacob Kollanukkaran
>> Cc: dev@dpdk.org; Stefan Sundkvist ;
>> ola.liljed...@arm.com
>> Subject: [EXT] Re: [PATCH 5/8] event/dsw: avoid
> From: Alex Williamson
> Sent: Friday, March 6, 2020 2:23 AM
>
> On Tue, 25 Feb 2020 03:08:00 +
> "Tian, Kevin" wrote:
>
> > > From: Alex Williamson
> > > Sent: Thursday, February 20, 2020 2:54 AM
> > >
> > > With the VF Token interface we can now expect that a vfio userspace
> > > driver
> From: Alex Williamson
> Sent: Friday, March 6, 2020 2:18 AM
>
> On Tue, 25 Feb 2020 02:59:37 +
> "Tian, Kevin" wrote:
>
> > > From: Alex Williamson
> > > Sent: Thursday, February 20, 2020 2:54 AM
> > >
> > > If we enable SR-IOV on a vfio-pci owned PF, the resulting VFs are not
> > > fully
> From: Alex Williamson
> Sent: Friday, March 6, 2020 1:34 AM
>
> Hi Kevin,
>
> Sorry for the delay, I've been out on PTO...
>
> On Tue, 25 Feb 2020 02:33:27 +
> "Tian, Kevin" wrote:
>
> > > From: Alex Williamson
> > > Sent: Thursday, February 20, 2020 2:54 AM
> > >
> > > Changes since v1
> From: Alex Williamson
> Sent: Monday, March 9, 2020 8:46 AM
>
> On Sat, 7 Mar 2020 01:04:41 +
> "Tian, Kevin" wrote:
>
> > > From: Alex Williamson
> > > Sent: Friday, March 6, 2020 11:39 PM
> > >
> > > On Fri, 6 Mar 2020 08:32:40 +
> > > "Tian, Kevin" wrote:
> > >
> > > > > From: Al
> From: Alex Williamson
> Sent: Saturday, March 7, 2020 6:18 AM
>
> On Fri, 6 Mar 2020 07:57:19 +
> "Tian, Kevin" wrote:
>
> > > From: Alex Williamson
> > > Sent: Friday, March 6, 2020 2:23 AM
> > >
> > > On Tue, 25 Feb 2020 03:08:00 +
> > > "Tian, Kevin" wrote:
> > >
> > > > > From: A
> From: Alex Williamson
> Sent: Friday, March 6, 2020 11:39 PM
>
> On Fri, 6 Mar 2020 08:32:40 +
> "Tian, Kevin" wrote:
>
> > > From: Alex Williamson
> > > Sent: Friday, March 6, 2020 2:18 AM
> > >
> > > On Tue, 25 Feb 2020 02:59:37 +
> > > "Tian, Kevin" wrote:
> > >
> > > > > From: A
> From: Tian, Kevin
> Sent: Friday, March 6, 2020 3:57 PM
>
> > From: Alex Williamson
> > Sent: Friday, March 6, 2020 2:23 AM
> >
> > On Tue, 25 Feb 2020 03:08:00 +
> > "Tian, Kevin" wrote:
> >
> > > > From: Alex Williamson
> > > > Sent: Thursday, February 20, 2020 2:54 AM
> > > >
> > > > Wi
> From: Alex Williamson
> Sent: Monday, March 9, 2020 8:46 AM
>
> On Sat, 7 Mar 2020 01:35:23 +
> "Tian, Kevin" wrote:
>
> > > From: Alex Williamson
> > > Sent: Saturday, March 7, 2020 6:18 AM
> > >
> > > On Fri, 6 Mar 2020 07:57:19 +
> > > "Tian, Kevin" wrote:
> > >
> > > > > From: Ale
> From: Tian, Kevin
> Sent: Monday, March 9, 2020 9:22 AM
>
> > From: Alex Williamson
> > Sent: Monday, March 9, 2020 8:46 AM
> >
> > On Sat, 7 Mar 2020 01:04:41 +
> > "Tian, Kevin" wrote:
> >
> > > > From: Alex Williamson
> > > > Sent: Friday, March 6, 2020 11:39 PM
> > > >
> > > > On Fri,
On 3/5/2020 3:48 PM, Andrew Rybchenko wrote:
> From: Igor Romanov
>
> Call correct cleanup operations on failure in Rx queue start function.
>
> Fixes: df1bfde4ff0d ("net/sfc: factor out libefx-based Rx datapath")
> Fixes: 28944ac098aa ("net/sfc: implement Rx queue start and stop operations")
>
On Mon, Mar 09, 2020 at 04:55:28PM +0800, Tonghao Zhang wrote:
> On Mon, Mar 9, 2020 at 4:27 PM Olivier Matz wrote:
> >
> > Hi,
> >
> > On Mon, Mar 09, 2020 at 11:01:25AM +0800, Tonghao Zhang wrote:
> > > On Sat, Mar 7, 2020 at 8:54 PM Andrew Rybchenko
> > > wrote:
> > > >
> > > > On 3/7/20 3:51
Any DPDK public header file which includes stdbool.h may conflict with
local definition of bool, if any, which further results in compilation
error. To avoid, used standard stdbool.h instead of defining bool
internally.
I observed this issue during a development where I included rte_uuid.h
into rt
Updated. Sent next version(v3).
Regards
Sunil Kumar Kori
>-Original Message-
>From: Ferruh Yigit
>Sent: Monday, March 9, 2020 2:14 PM
>To: Jerin Jacob ; Sunil Kumar Kori
>
>Cc: Alfredo Cardigliano ; dpdk-dev ;
>dpdk stable
>Subject: [EXT] Re: [dpdk-dev] [PATCH v2 1/1] net/ionic: use sta
Hi,
> -Original Message-
> From: dev On Behalf Of
> xiangxia.m@gmail.com
> Sent: Friday, March 6, 2020 4:27 PM
> To: Matan Azrad ; Ali Alnubani
>
> Cc: dev@dpdk.org; Tonghao Zhang ;
> sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH dpdk-dev v3] common/mlx5: fix dynamic loading
> referen
We nead isb rather than dsb to sync system counter to cntvct_el0.
Signed-off-by: Haifeng Lin
---
lib/librte_eal/common/include/arch/arm/rte_atomic_64.h | 3 +++
lib/librte_eal/common/include/arch/arm/rte_cycles_64.h | 2 ++
2 files changed, 5 insertions(+)
diff --git a/lib/librte_eal/common/inc
On Mon, Mar 9, 2020 at 10:14 AM Linhaifeng wrote:
>
> We nead isb rather than dsb to sync system counter to cntvct_el0.
I'll leave the arm maintainers look at this, but I have a comment on the form.
>
> Signed-off-by: Haifeng Lin
> ---
> lib/librte_eal/common/include/arch/arm/rte_atomic_64.h
We should use isb rather than dsb to sync system counter to cntvct_el0.
Signed-off-by: Haifeng Lin
---
lib/librte_eal/common/include/arch/arm/rte_atomic_64.h | 3 +++
lib/librte_eal/common/include/arch/arm/rte_cycles_64.h | 2 ++
2 files changed, 5 insertions(+)
diff --git a/lib/librte_eal/common/
Hi Jerin,
> -Original Message-
> From: Jerin Jacob
> Sent: Monday, March 9, 2020 3:36 PM
> To: Gavin Hu
> Cc: dpdk-dev ; nd ; David Marchand
> ; tho...@monjalon.net; jer...@marvell.com;
> Ye, Xiaolong ; Honnappa Nagarahalli
> ; Ruifeng Wang
> ; Phil Yang ; Joyce Kong
> ; Steve Capper
>
This series are updates and fixes for hns3 PMD driver.
Chengchang Tang (3):
net/hns3: remove unnecessary restriction on setting VF's MTU
net/hns3: support promiscuous and allmulticast mode for VF
net/hns3: fix promiscuous mode for PF
Hongbo Zheng (1):
net/hns3: support TCP segment offload
From: Hongbo Zheng
This patch adds TCP segment offload support for hns3 PMD driver.
Signed-off-by: Hongbo Zheng
Signed-off-by: Wei Hu (Xavier)
---
doc/guides/nics/features/hns3.ini| 1 +
doc/guides/nics/features/hns3_vf.ini | 1 +
doc/guides/nics/hns3.rst | 1 +
drivers/
From: "Wei Hu (Xavier)"
Currently, when receiving and transmitting packets based on hns3 network
engine there are probably unexpected and redundant Tx interrupts if Rx
interrupt is enabled.
The root cause as below:
Tx and Rx queues with the same number share the interrupt vector in hns3
network
From: Chengchang Tang
The hns3 PF/VF devices on the same port share the hardware MTU
configuration. Currently, we send mailbox to inform hns3 PF kernel
ethdev driver to finish hardware MTU configuration in hns3 VF PMD
driver, there is no need to stop the port for hns3 VF device, and
the MTU value
From: Chengchang Tang
Currently, we only support VF device is bound to vfio_pci or igb_uio and
then driven by DPDK driver when PF is driven by kernel mode hns3 ethdev
driver, VF is not supported when PF is driven by hns3 DPDK driver.
This patch adds promiscuous and allmulticast mode support for
From: Chengchang Tang
Currently, when promiscuous mode is enabled, it is just allowed to accept
all the unicast and broadcast packets in hns3 PF PMD driver. It should
also be able to receive multicast packets.
Fixes: 19a3ca4c99cf ("net/hns3: add start/stop and configure operations")
Cc: sta...@d
No need to export dsb and dmb.
EAL memory barriers are the public API.
%s/\<\(d.b\)\>(\([^()]*\))$/asm volatile("\1 \2" : : : "memory")/
Signed-off-by: David Marchand
---
.../common/include/arch/arm/rte_atomic_64.h | 19 ---
1 file changed, 8 insertions(+), 11 deletions(-)
di
Hi Ferruh,
> -Original Message-
> From: Ferruh Yigit
> Sent: Monday, March 9, 2020 4:55 PM
> To: Gavin Hu ; dev@dpdk.org
> Cc: nd ; david.march...@redhat.com; tho...@monjalon.net;
> ktray...@redhat.com; jer...@marvell.com; Honnappa Nagarahalli
> ; Ruifeng Wang
> ; Phil Yang ; Joyce Kong
>
Bitwise operation APIs are defined and used in a lot of PMDs,
which caused a huge code duplication. To reduce duplication,
this patch consolidates them into a common API family and use
it for all the PMDs(we started with a few selectively).
v7:
1. Change the API's head file to 'rte_pmd_bitops.h'
Bitwise operation APIs are defined and used in a lot of PMDs,
which caused a huge code duplication. To reduce duplication,
this patch consolidates them into a common API family.
Signed-off-by: Joyce Kong
Reviewed-by: Gavin Hu
Reviewed-by: Phil Yang
Acked-by: Morten Brørup
---
MAINTAINERS
Remove its own bit operation APIs and use the common one,
this can reduce the code duplication largely.
Signed-off-by: Joyce Kong
Reviewed-by: Gavin Hu
---
drivers/net/bnx2x/bnx2x.c| 253 +--
drivers/net/bnx2x/bnx2x.h| 10 +-
drivers/net/bnx2x/ecore_sp.h
Add test cases of bit operations which used by PMDs.
Signed-off-by: Joyce Kong
Reviewed-by: Gavin Hu
Reviewed-by: Phil Yang
---
MAINTAINERS| 1 +
app/test/Makefile | 1 +
app/test/autotest_data.py | 6 ++
app/test/meson.build | 2 +
app/test/test_pmd_bit
Remove its own bit operation APIs and use the common one,
this can reduce the code duplication largely.
Signed-off-by: Joyce Kong
Reviewed-by: Gavin Hu
---
drivers/net/axgbe/axgbe_common.h | 29 +
drivers/net/axgbe/axgbe_ethdev.c | 14 +++---
drivers/net/axgb
Remove its own bit operation APIs and use the common one,
this can reduce the code duplication largely.
Signed-off-by: Joyce Kong
Reviewed-by: Gavin Hu
---
drivers/net/hinic/Makefile| 1 +
drivers/net/hinic/base/hinic_compat.h | 33 +--
drivers/net/hinic/hin
Remove its own bit operation APIs and use the common one,
this can reduce the code duplication largely.
Signed-off-by: Joyce Kong
Reviewed-by: Gavin Hu
---
drivers/net/qede/base/bcm_osal.c | 22 +
drivers/net/qede/base/bcm_osal.h | 14 -
drivers/net/qede/ba
On Thu, Feb 27, 2020 at 04:46:09PM +, Ferruh Yigit wrote:
>On 2/19/2020 8:18 AM, Július Milan wrote:
>> With this patch it is possible to connect 2 DPDK memifs into loopback,
>> i.e. when they have the same id and different roles, as for example:
>> "--vdev=net_memif0,role=master,id=0"
>> "
Shows the example of how to connect 2 memif ports into loopback and test
the communication.
Signed-off-by: Július Milan
---
doc/guides/nics/memif.rst | 18 ++
1 file changed, 18 insertions(+)
diff --git a/doc/guides/nics/memif.rst b/doc/guides/nics/memif.rst
index 4d7f0086a..08a
Remove function ice_clear_queues, since all equivalent code
has already be executed durring ice_rx|tx_queue_stop.
Also function ice_rx|tx_queue_release_mbufs simpley wrapped a
function pointer call which is not necessary, remove them.
Signed-off-by: Qi Zhang
---
drivers/net/ice/ice_ethdev.c |
We should use isb rather than dsb to sync system counter to cntvct_el0.
Signed-off-by: Haifeng Lin
---
lib/librte_eal/common/include/arch/arm/rte_atomic_64.h | 3 +++
lib/librte_eal/common/include/arch/arm/rte_cycles_64.h | 2 ++
2 files changed, 5 insertions(+)
diff --git a/lib/librte_eal/comm
We should use isb rather than dsb to sync system counter to cntvct_el0.
Signed-off-by: Linhaifeng
---
lib/librte_eal/common/include/arch/arm/rte_atomic_64.h | 3 +++
lib/librte_eal/common/include/arch/arm/rte_cycles_64.h | 2 ++
2 files changed, 5 insertions(+)
diff --git a/lib/librte_eal/commo
On 3/9/2020 12:09 PM, Július Milan wrote:
> On Thu, Feb 27, 2020 at 04:46:09PM +, Ferruh Yigit wrote:
>> On 2/19/2020 8:18 AM, Július Milan wrote:
>>> With this patch it is possible to connect 2 DPDK memifs into loopback,
>>> i.e. when they have the same id and different roles, as for example:
On 3/9/2020 10:46 AM, Július Milan wrote:
> With this patch it is possible to connect 2 DPDK memifs into loopback,
> i.e. when they have the same id and different roles, as for example:
> "--vdev=net_memif0,role=master,id=0"
> "--vdev=net_memif1,role=slave,id=0"
>
> Signed-off-by: Július Milan
On 3/9/2020 9:45 AM, Gavin Hu wrote:
> Hi Ferruh,
>
>> -Original Message-
>> From: Ferruh Yigit
>> Sent: Monday, March 9, 2020 4:55 PM
>> To: Gavin Hu ; dev@dpdk.org
>> Cc: nd ; david.march...@redhat.com; tho...@monjalon.net;
>> ktray...@redhat.com; jer...@marvell.com; Honnappa Nagarahall
Via code inspection, I found that some partially initialized
stack variables were being passed along to called functions,
which could eventually result in those uninitialized members
being used. To fix this, make sure the local variables are
zeroed out before partially initializing them. This sho
Send correct recp_list to ice_remove_mac_rule. ICE_SW_LKUP_ETHERTYPE
rule list was sent instead of ICE_SW_LKUP_MAC. That caused problem
with adding new mac rule on VF, because rule wasn't removed correctly.
Fixes: c7dd15931183 ("net/ice/base: add virtual switch code")
Cc: sta...@dpdk.org
Signed-o
Main changes:
1) Support GTPU uplink and downlink
2) add new ACL module
3) couple fixes and code clean
Qi Zhang (28):
net/ice/base: fix uninitialized stack variables
net/ice/base: add and update E822 device IDs
net/ice/base: fix removing MAC rule
net/ice/base: read PSM clock frequency fro
Add the device IDs for the Intel(R) Ethernet Connection E822-L and
E822-X SKUs. Update the codenames and branding strings for the previous
C822N device IDs which should be using E822-C.
Signed-off-by: Jacob Keller
Signed-off-by: Paul M Stillwell Jr
Signed-off-by: Qi Zhang
---
drivers/net/ice/b
Add new api function which allow user to choose port on which
vlan and ethertype rule going to be added.
Signed-off-by: Michal Swiatkowski
Signed-off-by: Paul M Stillwell Jr
Signed-off-by: Qi Zhang
---
drivers/net/ice/base/ice_switch.c | 162 ++
1 file chang
Read the GLGEN_CLKSTAT_SRC register to determine which PSM clock
frequency is selected. This ensures that the rate limiter profile
calculations will be correct.
Signed-off-by: Ben Shelton
Signed-off-by: Paul M Stillwell Jr
Signed-off-by: Qi Zhang
---
drivers/net/ice/base/ice_common.c | 1 +
Use enum ice_flow_field directly so no need to be converted
from u16 for ice_flow_xtract_fld
Signed-off-by: Paul M Stillwell Jr
Signed-off-by: Qi Zhang
---
drivers/net/ice/base/ice_flow.c | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ice/base/ice_flow.c
This is a collection of minor fixes that were found during code review.
Changes are:
- Call ice_hweight8() instead of calculating it ourselves in
ice_bits_max_set().
- Call ice_test_and_clear_bit() over calling ice_is_bit_set() then
ice_clear_bit() in ice_rem_vsi_rss_list().
- Remove 'chrs' var
For structures using the common C "struct hack" technique to create a
flexible length structure member at the end of the structure, use the
ice_struct_size macro to determine the length of the structure instead of
open coding the calculation.
Signed-off-by: Bruce Allan
Signed-off-by: Paul M Still
As title, these macros are added for future use.
Signed-off-by: Paul M Stillwell Jr
Signed-off-by: Qi Zhang
---
drivers/net/ice/base/ice_adminq_cmd.h | 26 ++
1 file changed, 26 insertions(+)
diff --git a/drivers/net/ice/base/ice_adminq_cmd.h
b/drivers/net/ice/base/ice
The variable name 'type' is not very descriptive. Replace instances of
those with a variable name that is more descriptive or replace it if not
needed.
Signed-off-by: Bruce Allan
Signed-off-by: Paul M Stillwell Jr
Signed-off-by: Qi Zhang
---
drivers/net/ice/base/ice_flex_pipe.c | 8
Refactor function ice_prof_bld_xlt2, a switch statement is better suited
for this situation and eliminates the need for the "found" variable.
Signed-off-by: Tony Nguyen
Signed-off-by: Paul M Stillwell Jr
Signed-off-by: Qi Zhang
---
drivers/net/ice/base/ice_flex_pipe.c | 21 +---
As title.
Signed-off-by: Paul M Stillwell Jr
Signed-off-by: Qi Zhang
---
drivers/net/ice/base/ice_type.h | 8
1 file changed, 8 insertions(+)
diff --git a/drivers/net/ice/base/ice_type.h b/drivers/net/ice/base/ice_type.h
index 3e24bb1dc..4979580ec 100644
--- a/drivers/net/ice/base/ice
Add a new MAC type ICE_MAC_E810 to distinguish E810 devices from other
devices. MAC types for all other devices will be ICE_MAC_GENERIC till
there's a need to distinguish further between devices.
Signed-off-by: Anirudh Venkataramanan
Signed-off-by: Paul M Stillwell Jr
Signed-off-by: Qi Zhang
--
Adds the Driver Shared Parameters (0x0C90) AQ command.
Signed-off-by: Lev Faerman
Signed-off-by: Paul M Stillwell Jr
Signed-off-by: Qi Zhang
---
drivers/net/ice/base/ice_adminq_cmd.h | 17 +
1 file changed, 17 insertions(+)
diff --git a/drivers/net/ice/base/ice_adminq_cmd.h
b
Enable GTPU uplink and downlink flag usage.
TCAM with different GTPU extend header flag can be saperated.
Signed-off-by: Dan Nowlin
Signed-off-by: Paul M Stillwell Jr
Signed-off-by: Qi Zhang
---
drivers/net/ice/base/ice_flex_pipe.c | 68 +++---
drivers/net/ice/base/
Adds functions to check for link override firmware support and get
the override settings for a port. Link override allows a user to force
link settings that are not normally supported.
Firmware support is version dependent so a function to check support has
been added.
The link FC settings will u
Adds masks indicating AN clauses to the Get PHY Capabilities
command. Changes the name of the low_power_ctrl field to be
properly descriptive of it being an AN field.
Signed-off-by: Lev Faerman
Signed-off-by: Paul M Stillwell Jr
Signed-off-by: Qi Zhang
---
drivers/net/ice/base/ice_adminq_cmd.h
Fix abbreviations as found by abbrevcheck
Signed-off-by: Tony Nguyen
Signed-off-by: Paul M Stillwell Jr
Signed-off-by: Qi Zhang
---
drivers/net/ice/base/ice_dcb.c | 8 +++---
drivers/net/ice/base/ice_fdir.c | 6 ++---
drivers/net/ice/base/ice_flex_pipe.c | 52 ++---
The ice_read_sr_buf_aq function implements logic to correctly break
apart NVM reads into 4Kb chunks. Additionally, it ensures that each read
never crosses a Shadow RAM sector boundary. This logic is useful when
reading the flat NVM as a byte-addressable stream.
Extract that logic in terms of bytes
Currently the NVM and Option ROM version information is stored in
a minimal format. The ice_get_nvm_version function exists to extract
this information for display.
This needlessly complicates using these fields as the extraction
function must be called to parse the NVM and Option ROM data. Furthe
In this patch, we will modify the ice_copy_phy_caps_to_cfg(...) function
to conditionally fill up the
ice_aqc_set_phy_cfg_data.module_compliance_enforcement with correct
value, based on the PHY persistent feature.
Apply the ice_copy_phy_caps_to_cfg() function inside ice_set_fc()
Signed-off-by: Ch
1 - 100 of 195 matches
Mail list logo
