>-Original Message-
>From: Andrew Rybchenko
>Sent: Tuesday, January 19, 2021 3:49 PM
>To: Xueming(Steven) Li ; NBU-Contact-Thomas
>Monjalon ; Ferruh Yigit ;
>Olivier Matz
>Cc: dev@dpdk.org; Slava Ovsiienko ; Asaf Penso
>
>Subject: Re: [PATCH v4 1/9] ethdev: introduce representor type
>
On 1/19/21 10:14 AM, Xueming Li wrote:
> SubFunction is a portion of the PCI device, created on demand, a SF
> netdev has its own dedicated queues(txq, rxq). A SF netdev supports
> eswitch representation offload similar to existing PF and VF
> representors.
>
> To support SF representor, this patc
On 1/19/21 10:14 AM, Xueming Li wrote:
> Current VF representor syntax:
> representor=2 - single representor
> representor=[0-3] - single range
>
> To prepare for more representor types, this patch adds compatible VF
> representor devargs syntax:
>
> vf#:
> representor=vf2
On 1/19/21 10:37 AM, Xueming(Steven) Li wrote:
> Hi Andrew,
>
>> -Original Message-
>> From: Andrew Rybchenko
>> Sent: Tuesday, January 19, 2021 3:25 PM
>> To: Xueming(Steven) Li ; NBU-Contact-Thomas
>> Monjalon ; Ferruh Yigit ;
>> Olivier Matz
>> Cc: dev@dpdk.org; Slava Ovsiienko ; Asaf
On 1/19/21 10:14 AM, Xueming Li wrote:
> To support extended representor syntax, this patch extends the
> representor list parsing to support for representor port range in
> devargs, examples:
>representor=[1,2,3] - single list
>representor=[1,3-5,7,9-11] - list with singles and ra
On 1/18/21 2:16 PM, Xueming Li wrote:
> To support extended representor syntax, this patch extends the
> representor list parsing to support for representor port range in
> devargs, examples:
>representor=[1,2,3] - single list
>representor=[1,3-5,7,9-11] - list with singles and ran
Hi, Wenjun
> -Original Message-
> From: Wu, Wenjun1
> Sent: Tuesday, January 19, 2021 2:59 PM
> To: dev@dpdk.org; Guo, Jia ; Zhang, Qi Z
> ; Zhang, Yuying
> Cc: Wu, Wenjun1 ; sta...@dpdk.org
> Subject: [PATCH v1] net/e1000: fix the invalid flow control mode setting
>
> E1000_CTRL regist
Hi Andrew,
>-Original Message-
>From: Andrew Rybchenko
>Sent: Tuesday, January 19, 2021 3:25 PM
>To: Xueming(Steven) Li ; NBU-Contact-Thomas
>Monjalon ; Ferruh Yigit ;
>Olivier Matz
>Cc: dev@dpdk.org; Slava Ovsiienko ; Asaf Penso
>
>Subject: Re: [PATCH v4 1/9] ethdev: introduce represent
> -Original Message-
> From: Xueming Li
> Sent: Tuesday, January 19, 2021 15:15
> Cc: dev@dpdk.org; Viacheslav Ovsiienko ;
> xuemi...@nvidia.com; Asaf Penso
> ; Ajit Khaparde ; Somnath Kotur
> ; Daley, John ; Hyong Youb
> Kim ;
> Xing, Beilei ; Guo, Jia ; Wang,
> Haiyue
> ; Matan Azrad
In case of kernel bonding device, counter was read from first bonding PF
member.
This patch reads all member PFs and sums to get bond xstats.
Signed-off-by: Xueming Li
Acked-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/linux/mlx5_ethdev_os.c | 127 +++-
1 file changed, 102
To probe representors from different kernel bonding PFs, had to specify
2 separate devargs like this:
-a 03:00.0,representor=pf0vf[0-3] -a 03:00.0,representor=pf1vf[0-3]
This patch supports range or list of PF section in devargs, so the
alternative short devargs of above is:
-a 03:00.0,rep
This patch updates representor name parsing for SF.
In sysfs, representor name stored under "phys_port_name" sysfs key,
similar to VF representor, switch port name of SF representor is
"pfsf".
For netlink message, net SF type is supported.
Examples:
pf0sf1
pf0sf[0-3]
Signed-off-by: Xueming Li
To probe representor on 2nd PF of kernel bonding device, had to specify
PF1 BDF in devarg:
,representor=0
When closing bonding device, all representors had to be closed together
and this implies all representors have to use primary PF of bonding
device. So after probing representor port on 2nd PF
SubFunction [1] is a portion of the PCI device, a SF netdev has its own
dedicated queues(txq, rxq). A SF netdev supports E-Switch representation
offload similar to existing PF and VF representors. A SF shares PCI
level resources with other SFs and/or with its parent PCI function.
This patch set in
This patch adds support for SF representor. Similar to VF representor,
switch port name of SF representor in phys_port_name sysfs key is
"pfsf".
Device representor argumnt is "representors=sf[list]", list member could
be mix of instance and range. Example:
representors=sf[0,2,4,8-12,-1]
To prob
Since kernel bonding netdev doesn't provide statistics counter that
reflects all member ports, PMD has to manually summarize counters from
each member ports.
As a preparation, this patch collects bonding member port information
and saves to shared context data.
Signed-off-by: Xueming Li
Acked-by
With kernel bonding, there was an error when setting VF MAC address
through representor. The Netlink api requires ifindex of owner PF, not
bonding device ifindex.
Uses owner PF ifindex to modify VF default MAC in case of bonding
device.
Fixes: c21e5facf7d2 ("net/mlx5: use bond index for netdev op
With kernel bonding, representors on second PF are being probed by
devargs:
,representor=pf1vf
No need to save primary PF port ID and lookup when probing sibling
ports, revert patch [1]
[1]:
commit e6818853c022 ("net/mlx5: set representor to first PF in bonding
mode")
Signed-off-by: Xuemi
On 1/18/21 2:16 PM, Xueming Li wrote:
> To support more representor type, this patch introduces representor type
> enum. The enum is subject to extend for new types upcoming.
>
> Signed-off-by: Xueming Li
> Acked-by: Viacheslav Ovsiienko
One nit below and a question below.
In any case:
Acked-
> -Original Message-
> From: Xing, Beilei
> Sent: Tuesday, January 19, 2021 15:02
> To: Wang, Haiyue ; dev@dpdk.org
> Cc: Xing, Beilei
> Subject: [PATCH] doc: add ixgbe known issue with RSS
>
> From: Beilei Xing
>
> Currently, IXGBE doesn't support RSS when double VLAN
> is enabled due
Add the ethertype offset for QinQ switch rule dummy packet to
allow matching the corresponding field.
Signed-off-by: Yuying Zhang
---
drivers/net/ice/base/ice_switch.c | 13 ++---
1 file changed, 10 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ice/base/ice_switch.c
b/drivers/
Old DPDK version or some drivers didn't support SubFunction representor.
For application to adapt different DPDK version automatically, or to be
used for different NICs, this patch introduces new eth device capability
of supporting SubFunction representor device.
Signed-off-by: Xueming Li
Acked-b
The NIC can have multiple PCIe links and can be attached to the multiple
hosts, for example the same single NIC can be shared for multiple server
units in the rack. On each PCIe link NIC can provide multiple PFs and
VFs/SFs based on these ones. To provide the unambiguous identification
of the PCIe
This patch updates kvargs parser to support lists on top of range, allow
multiple lists or range:
k1=a[1,2]b[3-5]
Signed-off-by: Xueming Li
Acked-by: Viacheslav Ovsiienko
Acked-by: Thomas Monjalon
---
app/test/test_kvargs.c | 46 +--
lib/librte_kvargs/rte_kvargs.c | 101
The NIC can have multiple PCIe links and can be attached to multiple
hosts, for example the same single NIC can be shared for multiple server
units in the rack. On each PCIe link NIC can provide multiple PFs and
VFs/SFs based on these ones. The full representor identifier consists of
three indices
With Kernel bonding, multiple underlying PFs are bonded, VFs come
from different PF, need to identify representor of VFs unambiguously by
adding PF index.
This patch introduces optional 'pf' section to representor devargs
syntax, examples:
representor=pf0vf0 - single VF representor
r
SubFunction is a portion of the PCI device, created on demand, a SF
netdev has its own dedicated queues(txq, rxq). A SF netdev supports
eswitch representation offload similar to existing PF and VF
representors.
To support SF representor, this patch introduces new devargs syntax,
examples:
represe
Current VF representor syntax:
representor=2 - single representor
representor=[0-3] - single range
To prepare for more representor types, this patch adds compatible VF
representor devargs syntax:
vf#:
representor=vf2 - single representor
representor=vf[1,3,5]- singl
To support extended representor syntax, this patch extends the
representor list parsing to support for representor port range in
devargs, examples:
representor=[1,2,3] - single list
representor=[1,3-5,7,9-11] - list with singles and ranges
Signed-off-by: Xueming Li
Acked-by: Viache
To support more representor type, this patch introduces representor type
enum. The enum is subject to extend for new types upcoming.
Signed-off-by: Xueming Li
Acked-by: Viacheslav Ovsiienko
Acked-by: Thomas Monjalon
Acked-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_ethdev.c| 7 ++
dedicated queues(txq, rxq). A SF netdev supports E-Switch representation
offload similar to existing PF and VF representors. A SF shares PCI
level resources with other SFs and/or with its parent PCI function.
>From SmartNIC perspective, when PCI device is shared for multi-host,
representors for ho
Acked-by: abhinandan.guj...@intel.com
Thanks & Regards
Abhinandan
> -Original Message-
> From: Ankur Dwivedi
> Sent: Monday, January 18, 2021 9:50 PM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Gujjar, Abhinandan S ;
> akhil.go...@nxp.com; ano...@marvell.com; jer...@marvell.com; Ankur
> Dw
From: Beilei Xing
Currently, IXGBE doesn't support RSS when double VLAN
is enabled due to FW limitation.
Signed-off-by: Beilei Xing
---
doc/guides/nics/ixgbe.rst | 5 +
1 file changed, 5 insertions(+)
diff --git a/doc/guides/nics/ixgbe.rst b/doc/guides/nics/ixgbe.rst
index c801dbae81..696
> -Original Message-
> From: Zhang,Alvin
> Sent: Tuesday, January 19, 2021 1:26 PM
> To: Zhang, Qi Z
> Cc: dev@dpdk.org; Zhang, AlvinX ; sta...@dpdk.org
> Subject: [PATCH v7] net/ixgbe: fix configuration of max frame size
>
> From: Alvin Zhang
>
> For some types of NIC, jumbo frame
> -Original Message-
> From: Wang, Haiyue
> Sent: Monday, January 18, 2021 7:38 PM
> To: dev@dpdk.org
> Cc: Yang, Qiming ; Zhang, Qi Z ;
> Fu, Qi ; Wang, Haiyue
> Subject: [PATCH v2] net/ice: refactor DCF VLAN handling
>
> Since DCF always configure the outer VLAN offloads for the tar
> -Original Message-
> From: Bruce Richardson
> Sent: Monday, January 18, 2021 10:30 PM
> To: dev@dpdk.org
> Cc: ferruh.yi...@intel.com; Bruce Richardson ;
> sta...@dpdk.org; Matan Azrad ; Shahaf Shuler
> ; Viacheslav Ovsiienko ;
> Liron Himi ; Fiona Trahe ;
> John Griffin ; Deepak Kumar
From: Alvin Zhang
For some types of NIC, jumbo frame is not supported in IOV mode,
so if a VF requests to configure the frame size to not bigger
than IXGBE_ETH_MAX_LEN, the kernel driver returns 0, but the DPDK
ixgbe PMD returens -1, this will cause the VF to fail to start
when the PF driven by D
> -Original Message-
> From: Ding, Xuan
> Sent: Tuesday, January 19, 2021 12:35 PM
> To: Zhang, Qi Z ; Wu, Jingjing ;
> Xing, Beilei
> Cc: dev@dpdk.org; sta...@dpdk.org
> Subject: RE: [PATCH v2] net/iavf: fix VIRTHCNL_OP_SET_RSS_HENA handling
>
> Hi Qi,
>
> > -Original Message---
Hi Qi,
> -Original Message-
> From: Zhang, Qi Z
> Sent: Tuesday, January 19, 2021 11:45 AM
> To: Ding, Xuan ; Wu, Jingjing ;
> Xing, Beilei
> Cc: dev@dpdk.org; sta...@dpdk.org
> Subject: RE: [PATCH v2] net/iavf: fix VIRTHCNL_OP_SET_RSS_HENA handling
>
>
>
> > -Original Message
Add eCPRI tunnel port add and rm ops to configure eCPRI UDP tunnel port
in dcf.
Signed-off-by: Jeff Guo
---
v5:
rebase patch
v4:
add doc
---
doc/guides/rel_notes/release_21_02.rst | 12 +
drivers/net/ice/ice_dcf_ethdev.c | 67 ++
2 files changed, 79 insertions(
Add eCPRI tunnel port add and rm ops to configure eCPRI UDP tunnel port
in dcf.
Signed-off-by: Jeff Guo
---
doc/guides/rel_notes/release_21_02.rst | 12 +
drivers/net/ice/ice_dcf_ethdev.c | 67 ++
2 files changed, 79 insertions(+)
diff --git a/doc/guides/rel_no
Add eCPRI tunnel port add and rm ops to configure eCPRI UDP tunnel port
in dcf.
Signed-off-by: Jeff Guo
---
doc/guides/rel_notes/release_21_02.rst | 12 +
drivers/net/ice/ice_dcf_ethdev.c | 67 ++
2 files changed, 79 insertions(+)
diff --git a/doc/guides/rel_no
Hi All,
I was finally able to reproduce the issue, but on x86_64 machine with 32
cores.
You can find a patch fixing it here:
https://inbox.dpdk.org/dev/20210119035910.8324-1-l.wojciec...@partner.samsung.com/T/#t
Best regards
Lukasz
W dniu 11.01.2021 o 15:19, Lincoln Lavoie pisze:
> Hi Jerin
The distributor library implementation uses a cyclic queue to store
packets returned from workers. These packets can be later collected
with rte_distributor_returned_pkts() call.
However the queue has limited capacity. It is able to contain only
127 packets (RTE_DISTRIB_RETURNS_MASK).
Big burst te
According to the discussion in this thread:
https://protect2.fireeye.com/v1/url?k=24ecce33-7b77f77c-24ed457c-0cc47a30
d446-e317a6beb8cfa273&q=1&e=f8bb12df-3698-4bce-a7b7-d72e22b91431&u=https%
3A%2F%2Finbox.dpdk.org%2Fdev%2FCAOE1vsOehF4ZMOWffpEv%3DQF6YOc5wXtg23PV83B
9CLiTMn8wQA%40mail.gmail.com%2F%2
I see, thank you Stephen for the information.
Le lun. 18 janv. 2021 à 16:42, Stephen Hemminger
a écrit :
> On Mon, 18 Jan 2021 14:19:51 -0500
> Adel Belkhiri wrote:
>
> > Hi all,
> >
> > I was wondering if Eventdev (or at least Pipeline) library is used by any
> > real application (not the ones
> -Original Message-
> From: dev On Behalf Of Xing, Beilei
> Sent: Monday, January 18, 2021 4:27 PM
> To: Zhang, AlvinX ; Guo, Jia ; Xie,
> WeiX
> Cc: dev@dpdk.org; Zhang, AlvinX
> Subject: Re: [dpdk-dev] [PATCH v2] net/i40e: fix out-of-scope variable
>
>
>
> > -Original Messag
> -Original Message-
> From: Ding, Xuan
> Sent: Monday, January 18, 2021 4:28 PM
> To: Zhang, Qi Z ; Wu, Jingjing ;
> Xing, Beilei
> Cc: dev@dpdk.org; Ding, Xuan ; sta...@dpdk.org
> Subject: [PATCH v2] net/iavf: fix VIRTHCNL_OP_SET_RSS_HENA handling
>
> Allow error to be returned for
The app will calculate and output used cpu time for rte flows insertion
rate. It's also need for some new insertion items, such as meter. It's
better to split this calculation and output part to a single function,
so that all new insertion items can use it.
Signed-off-by: Dong Zhou
Reviewed-by: W
Currently, test-flow-perf app cannot generate flows with meter
action. This patch introduces new parameter "--meter" to generate
flows with meter action.
Signed-off-by: Dong Zhou
Reviewed-by: Wisam Jaddo
Reviewed-by: Alexander Kozyrev
---
app/test-flow-perf/actions_gen.c | 19
app/test-f
This patch introduces new parameter "--meter" to generate flows with meter
action in test-flow-perf application.
V2:
Divide this patch to 2 patches, one includes the split calculation and output
used cpu time for all insertion items, another includes the support for meter
action.
Dong Zhou (2):
> -Original Message-
> From: Guo, Junfeng
> Sent: Monday, January 18, 2021 10:08 PM
> To: Zhang, Qi Z ; Wu, Jingjing ;
> Xing, Beilei
> Cc: dev@dpdk.org; Wang, Haiyue ; Zhang, Yuying
> ; Guo, Junfeng ; Zhao1,
> Wei
> Subject: [PATCH v3] net/ice: enable QinQ filter for switch
>
> Enab
Add eCPRI protocol flow features to ice pmd and iavf pmd.
Signed-off-by: Jeff Guo
---
doc/guides/rel_notes/release_21_02.rst | 12
1 file changed, 12 insertions(+)
diff --git a/doc/guides/rel_notes/release_21_02.rst
b/doc/guides/rel_notes/release_21_02.rst
index 3bb0b9a9ed..c84f0b
Hi David,
> -Original Message-
> From: dev On Behalf Of David Christensen
> Sent: Friday, January 15, 2021 3:03 AM
> To: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 8/9] test/vfio_user: introduce functional
> test
>
>
>
> On 1/13/21 10:14 PM, Chenbo Xia wrote:
> > This patch intro
Hi David,
> -Original Message-
> From: dev On Behalf Of David Christensen
> Sent: Friday, January 15, 2021 2:49 AM
> To: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 3/9] vfio_user: implement device and region
> related APIs
>
>
>
> On 1/13/21 10:14 PM, Chenbo Xia wrote:
> > This p
Refine the doxygen for rte_eth_dev_udp_tunnel_port_add.
Add more detail description of the impacted offload functions.
Signed-off-by: Qi Zhang
---
v2:
- reword doxygen that focus on API impact base on previous discussion.
lib/librte_ethdev/rte_ethdev.h | 11 +++
1 file changed, 11 ins
Hi David,
> -Original Message-
> From: David Marchand
> Sent: Friday, January 15, 2021 3:58 PM
> To: Xia, Chenbo
> Cc: dev ; Thomas Monjalon ; Stephen
> Hemminger ; Liang, Cunming
> ; Lu, Xiuchun ; Li, Miao
> ; Wu, Jingjing ; Xing, Beilei
>
> Subject: Re: [PATCH v2 0/9] Introduce vfio-u
Thank Stephen,
but in which the scenarios, it should wait link to up, in which
scenarios, it should not ?
By the way, how to define the "wait" time value ?
在 2021/1/19 10:26, Stephen Hemminger 写道:
On Tue, 19 Jan 2021 09:06:48 +0800
"Min Hu (Connor)" wrote:
Hi, Bruce and all,
Do you
On Tue, 19 Jan 2021 09:06:48 +0800
"Min Hu (Connor)" wrote:
> Hi, Bruce and all,
> Do you know the difference between "rte_eth_link_get" and
> "rte_eth_link_get_nowait"? I know they call funciton "link_update"
> with differenct parameter "wait_to_complete"(set 1 means wait, set 0
> means
Fix mask to include all four bits of hardware packet type
field.
Fixes: 97b1db288dd0 ("net/bnxt: use table based packet type translation")
Cc: sta...@dpdk.org
Signed-off-by: Lance Richardson
---
drivers/net/bnxt/bnxt_rxr.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drive
On 1/15/2021 5:19 PM, Rahul Lakkireddy wrote:
Patch 1 fixes issue with link status always getting set to down by
driver for every link config change request. This is not required
for some link config changes, like Tx/Rx pause on/off. Instead,
driver should rely on actual link status returned by f
On 1/17/2021 10:21 PM, Ivan Malov wrote:
An application may submit all-zeros masks for a given field
to be used in two given specifications and, in the meantime,
use different unmasked values. Later on, when compared, the
two specifications will prove unequal, and, if the rules in
question are ou
On 1/18/2021 9:40 PM, Alexander Kozyrev wrote:
Implement a generic modify rte flow API as described in RFC:
http://patches.dpdk.org/patch/85384/
This API allows modifying a destination header field with data from
a source header field. Number of bits to use from the source is specified.
Tag, Mar
Hi, Bruce and all,
Do you know the difference between "rte_eth_link_get" and
"rte_eth_link_get_nowait"? I know they call funciton "link_update"
with differenct parameter "wait_to_complete"(set 1 means wait, set 0
means not wait). But how to define the "wait" time, and why it shoud wait?
On 1/14/2021 1:33 PM, Lijun Ou wrote:
This series add TM feature support and fix some
bugs for hns3 pmd driver. Because the TM need to
use rte_atomicNN_xxx for the resetting of the
hns3_reset_data structure. Therefore it needs to
add a new updates patch for using C11 atomics
builtins for resettin
On 1/14/2021 1:33 PM, Lijun Ou wrote:
From: Chengwen Feng
This patch support RTE TM ops function for PF, which could
used to:
1. config port's peak rate.
2. config TC's peak rate.
Signed-off-by: Chengwen Feng
Signed-off-by: Lijun Ou
---
drivers/net/hns3/hns3_dcb.c| 216 ---
drive
> -Original Message-
> From: Thomas Monjalon
> Sent: Monday, January 18, 2021 5:50 PM
> To: Zhang, Qi Z
> Cc: dev@dpdk.org; Yigit, Ferruh ; or...@nvidia.com;
> getel...@nvidia.com; andrew.rybche...@oktetlabs.ru;
> ajit.khapa...@broadcom.com; jer...@marvell.com
> Subject: Re: [dpdk-dev]
On Mon, Jan 18, 2021 at 8:18 AM Alexander Kozyrev wrote:
>
> Implement a generic modify rte flow API as described in RFC:
> http://patches.dpdk.org/patch/85384/
>
> This API allows modifying a destination header field with data from
> a source header field. Number of bits to use from the source is
>-Original Message-
>From: Ajit Khaparde
>Sent: Tuesday, January 19, 2021 2:18 AM
>To: NBU-Contact-Thomas Monjalon
>Cc: Xueming(Steven) Li ; Ferruh Yigit
>; Andrew Rybchenko
>; Olivier Matz ;
>dpdk-dev ; Slava Ovsiienko ; Asaf
>Penso
>Subject: Re: [dpdk-dev] [PATCH v4 1/9] ethdev: introd
Hi Thomas,
>-Original Message-
>From: Thomas Monjalon
>Sent: Tuesday, January 19, 2021 12:18 AM
>To: Xueming(Steven) Li
>Cc: Ferruh Yigit ; Andrew Rybchenko
>; Olivier Matz ;
>dev@dpdk.org; Slava Ovsiienko ; Asaf Penso
>
>Subject: Re: [dpdk-dev] [PATCH v4 2/9] ethdev: support representor
Hi Slava,
>-Original Message-
>From: Slava Ovsiienko
>Sent: Tuesday, January 19, 2021 12:17 AM
>To: Xueming(Steven) Li
>Cc: dev@dpdk.org; Matan Azrad ; Shahaf Shuler
>; NBU-Contact-Thomas Monjalon
>; Xueming(Steven) Li ; Asaf
>Penso
>Subject: RE: [PATCH v3 7/9] net/mlx5: save bonding me
On Mon, Jan 18, 2021 at 3:18 AM Xueming Li wrote:
>
> The NIC can have multiple PCIe links and can be attached to multiple
> hosts, for example the same single NIC can be shared for multiple server
> units in the rack. On each PCIe link NIC can provide multiple PFs and
> VFs/SFs based on these one
14/01/2021 15:46, Anatoly Burakov:
> Anatoly Burakov (5):
> eal: uninline power intrinsics
> eal: avoid invalid API usage in power intrinsics
> eal: change API of power intrinsics
> eal: remove sync version of power monitor
> eal: add monitor wakeup function
>
> Liang Ma (6):
> ethdev:
14/01/2021 15:46, Anatoly Burakov:
> From: Liang Ma
>
> + Currently, this power management API is limited to mandatory mapping of 1
> + queue to 1 core (multiple queues are supported, but they must be polled
> from
> + different cores).
This is quite limited.
Not sure librte_power is the
14/01/2021 15:46, Anatoly Burakov:
> +struct rte_power_monitor_cond {
> + volatile void *addr; /**< Address to monitor for changes */
> + uint64_t val; /**< Before attempting the monitoring, the address
> +* may be read and compared against this value.
On Mon, Jan 18, 2021 at 10:15 AM Thomas Monjalon wrote:
>
> 18/01/2021 19:00, Ajit Khaparde:
> > On Mon, Jan 18, 2021 at 9:57 AM Thomas Monjalon wrote:
> > > 18/01/2021 18:42, Ajit Khaparde:
> > > > On Mon, Jan 18, 2021 at 3:17 AM Xueming Li wrote:
> > > > > +enum rte_eth_representor_type {
> >
On Mon, Jan 18, 2021 at 9:57 AM Thomas Monjalon wrote:
>
> 18/01/2021 18:42, Ajit Khaparde:
> > On Mon, Jan 18, 2021 at 3:17 AM Xueming Li wrote:
> > > +enum rte_eth_representor_type {
> > > + RTE_ETH_REPRESENTOR_NONE, /**< not a representor. */
> > > + RTE_ETH_REPRESENTOR_VF, /**<
On Mon, Jan 18, 2021 at 3:17 AM Xueming Li wrote:
>
> To support more representor type, this patch introduces representor type
> enum. The enum is subject to extend for new types upcoming.
>
> Signed-off-by: Xueming Li
> Acked-by: Viacheslav Ovsiienko
Acked-by: Ajit Khaparde
> ---
> drivers/n
On Mon, 18 Jan 2021 14:19:51 -0500
Adel Belkhiri wrote:
> Hi all,
>
> I was wondering if Eventdev (or at least Pipeline) library is used by any
> real application (not the ones in the "app" and "example" directories of
> dpdk codebase). Actually, I developed few performance analyses related to
>
Add support for the RTE_FLOW_ACTION_MODIFY_FIELD to the testpmd.
Implement CLI to create the modify_field action and supply all the
needed parameters to modify an arbitrary packet field (as well as
mark, tag or metadata) with data from another field or immediate
value.
Example of the flow is the f
Implement the generic modify flow API to allow manipulations on
an arbitrary header field (as well as mark, metadata or tag) using
data from another field or a user-specified value.
This generic modify mechanism removes the necessity to implement
a separate RTE Flow action every time we need to mod
Implement a generic modify rte flow API as described in RFC:
http://patches.dpdk.org/patch/85384/
This API allows modifying a destination header field with data from
a source header field. Number of bits to use from the source is specified.
Tag, Mark or Metadata can also be used as a source/destin
15/01/2021 14:25, Juraj Linkeš:
> build: rename Arm build variables
> build: remove unused or superfluous variables
> build: reformat and move Arm config and comments
> build: simplify how Arm flags are processed
> build: organize Arm config into dict
> build: isolate configuration for
On 1/11/21 9:02 AM, Xia, Chenbo wrote:
> Hi Maxime,
>
>> -Original Message-
>> From: Maxime Coquelin
>> Sent: Monday, December 21, 2020 5:14 AM
>> To: dev@dpdk.org; Xia, Chenbo ; olivier.m...@6wind.com;
>> amore...@redhat.com; david.march...@redhat.com
>> Cc: Maxime Coquelin
>> Subjec
Don't assume standard headroom.
Use helper variables to improve readability.
Fixes: a27d901331da ("net/ionic: add Rx and Tx handling")
Cc: sta...@dpdk.org
Signed-off-by: Andrew Boyer
---
drivers/net/ionic/ionic_rxtx.c | 20
1 file changed, 12 insertions(+), 8 deletions(-)
d
One function marked cold is in the hot path.
Make sure to always inline hot path functions.
Fixes: a27d901331da ("net/ionic: add Rx and Tx handling")
Cc: sta...@dpdk.org
Signed-off-by: Andrew Boyer
---
drivers/net/ionic/ionic_rxtx.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-
DTS, at least, expects to be able to specify L4 checksum offload
without L3 csum offload. Split up the flag checks.
Fixes: a27d901331da ("net/ionic: add Rx and Tx handling")
Cc: sta...@dpdk.org
Signed-off-by: Andrew Boyer
---
drivers/net/ionic/ionic_rxtx.c | 20
1 file chang
The IONIC FW is little-endian.
Signed-off-by: Andrew Boyer
---
drivers/net/ionic/ionic_dev.h | 2 +-
drivers/net/ionic/ionic_osdep.h | 4 ++--
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ionic/ionic_dev.h b/drivers/net/ionic/ionic_dev.h
index 8847d6cad4..bacbe3f0
This will conserve resources by reducing struct ionic_qcq.
Saving a cacheline or two in the rxq and txq structs helps when
running in embedded configurations where CPU cache space is at a
premium.
Signed-off-by: Andrew Boyer
---
drivers/net/ionic/ionic_lif.h | 4 +++-
drivers/net/ionic/ionic_
This is a hot-path function.
Remove ionic_q_has_space() while here.
Signed-off-by: Andrew Boyer
---
drivers/net/ionic/ionic_dev.c | 21 +
drivers/net/ionic/ionic_dev.h | 15 +--
2 files changed, 14 insertions(+), 22 deletions(-)
diff --git a/drivers/net/ionic/ion
This is hot-path function.
Signed-off-by: Andrew Boyer
Signed-off-by: Neel Patel
---
drivers/net/ionic/ionic_dev.c | 6 --
drivers/net/ionic/ionic_dev.h | 9 -
drivers/net/ionic/ionic_osdep.h | 1 -
3 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/drivers/net/ionic
The ionic PMD only supports TX queue version 1 or greater.
Version 1 introduced a new SGL format with support for more
fragments per descriptor.
Signed-off-by: Andrew Boyer
---
drivers/net/ionic/ionic_dev.h| 2 +-
drivers/net/ionic/ionic_ethdev.c | 8
drivers/net/ionic/ionic_lif.c
This allows the PMD to better support FW changes.
Signed-off-by: Andrew Boyer
Signed-off-by: Shannon Nelson
---
drivers/net/ionic/ionic_dev.c | 15 ++
drivers/net/ionic/ionic_dev.h | 3 ++
drivers/net/ionic/ionic_lif.c | 95 +++
drivers/net/ionic/ionic_lif.h
Using the IONIC_ARRAY_SIZE() macro makes the code clearer.
Signed-off-by: Andrew Boyer
---
drivers/net/ionic/ionic_dev.c| 10 --
drivers/net/ionic/ionic_ethdev.c | 3 +--
drivers/net/ionic/ionic_lif.c| 9 +++--
drivers/net/ionic/ionic_main.c | 32 +++--
The IONIC firmware is little-endian.
Add a new field to struct ionic_rx_filter to store the CPU-endian
match type.
Use a local variable for the VLAN when searching the hash table.
Signed-off-by: Andrew Boyer
---
drivers/net/ionic/ionic_rx_filter.c | 22 +++---
drivers/net/ionic/i
The IONIC firmware is little-endian.
Signed-off-by: Andrew Boyer
---
drivers/net/ionic/ionic_dev.c| 27 +++---
drivers/net/ionic/ionic_ethdev.c | 27 +++---
drivers/net/ionic/ionic_lif.c| 138 ---
drivers/net/ionic/ionic_main.c | 6 +-
4 files changed, 1
Only the NotifyQ uses an interrupt, so simplify the other queues.
Simplify ionic_dev_cmd_adminq_init() and ionic_cq_init().
Move ionic_intr_alloc() into ionic_notify_qcq_alloc().
Create ionic_lif_notifyq_deinit().
Simplify ionic_lif_qcq_deinit().
Remove unneeded flags and defines.
Signed-off-by:
This patch series fixes some transmit issues, adds (better) support for
big-endian systems, and improves performance by stripping down some
structures and inlining a few functions.
The endianness code has been reviewed internally but not really tested -
I do not have access to a big-endian system
Hi,
> -Original Message-
> From: Jiawei Wang
> Sent: Thursday, January 14, 2021 9:25 AM
> To: ferruh.yi...@intel.com; Slava Ovsiienko ;
> Matan Azrad ; Ori Kam
> Cc: dev@dpdk.org; Raslan Darawsheh
> Subject: [PATCH v3 0/4] Add RSS action support in the sample sub-actions list
>
> Curren
-Original Message-
From: Jerin Jacob
Sent: Monday, 11 January 2021 16:57
To: Liron Himi
Cc: Jerin Jacob Kollanukkaran ; dpdk-dev ;
Yuri Chipchev
Subject: [EXT] Re: [dpdk-dev] [PATCH v1 11/38] net/mvpp2: align checking order
External Email
--
1 - 100 of 224 matches
Mail list logo
