Hi I have a host and XL710 nic cards and created few VFs on these NIC. I
created two VMs on that host and assign one VF to each VM.
VM#1: It has RHEL 7.2 installed and I use DPDK version 2.2.0
SRIOV ports on this VM works OK. testmpd works.
The SRIOV interfaces comes up with below logs when I in
Declan, you are the bonding maintainer, please confirm this fix.
13/07/2017 11:48, Jianbo Liu:
> On 13 July 2017 at 16:00, Herbert Guan wrote:
> > When test case "test_balance_l23_tx_burst_ipv4_toggle_ip_addr" is
> > calling balance_l23_tx_burst(), the ip_addr instead of mac_addr
> > should be to
On Tuesday 11 July 2017 03:31 PM, Jerin Jacob wrote:
> When we run DPDK on guest or VFIO mode on host,
> the dpdk library or device will not be directly accessing
> the physical address. Instead, the device does go through
> an IO address translation memory management unit. On x86,
> we call it as
21/07/2017 05:29, Wu, Jingjing:
>
> > when creating a bond device in testpmd, a name for a device must meet the
> > correct convention described in the documentation:
> > The device name must start with the net_bonding prefix followed by numbers
> > or letters.
> > Change for ALB mempool allocatio
> -Original Message-
> From: Guo, Jia
> Sent: Friday, July 21, 2017 11:07 AM
> To: Xing, Beilei ; Wu, Jingjing
> Cc: dev@dpdk.org; Guo, Jia ; sta...@dpdk.org
> Subject: [PATCH v2] net/i40e: fix link down and negotiation issue
>
> Enable the functions set link down and set link up in i40
18/07/2017 17:47, Shreyansh Jain:
> On Thursday 22 June 2017 06:18 PM, Hemant Agrawal wrote:
> > In case the alloc api is not able to allocate the required
> > number of buffer, it can return '0', which will not indicate
> > the failure to the calling function.
> > This patch fix the return value t
On Thursday 20 July 2017 03:59 PM, Santosh Shukla wrote:
> An API/ABI change is planned for 17.11 to change following
>
> * Remove unused flag param from rte_mempool_generic_get and _put.
> * Change data type for mempool 'flag' from int to unsigned int.
> Refer [1].
> * Add struct rte_mempool *
18/07/2017 17:45, Shreyansh Jain:
> On Thursday 13 July 2017 07:48 PM, Hemant Agrawal wrote:
> > In order to support multiprocess applications, pool data is to be
> > allocated on dynamic memory instead of existing usages of global
> > variable.
> >
> > Signed-off-by: Nipun Gupta
>
> Acked-by: S
13/07/2017 13:28, Hemant Agrawal:
> Disabling NUMA warnings on non-NUMA systems.
>
> "EAL: eal_parse_sysfs_value(): cannot open sysfs value
> /sys/bus/pci/devices/:00:00.0/numa_node
> EAL: numa_node is invalid or not present. Set it 0 as default
> EAL: cannot open /proc/self/numa_maps, c
This patch sends the HWRM command to free a filter in the hardware,
before using it again.
Fixes: f92735db1e4c ("net/bnxt: add L2 filter alloc/init/free")
Signed-off-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_hwrm.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/net/bnxt/bnxt_h
Add code to check for invalid filter_id in bnxt_hwrm_clear_filter
Fixes: f92735db1e4c ("net/bnxt: add L2 filter alloc/init/free")
Signed-off-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_hwrm.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/
We are wrongly using a Rx side HWRM command set_rx_mask to configure
VLAN anti-spoof. This being a Tx side feature, this patch
tries to fix it.
Since the HWRM command to do it ringt is available only in
the newer firmware versions, the patch verifies the firmware
version before attempting to send
Check if the vnic_id and rss_rule is not invalid before passing it
to the firmware to cleanup the VNIC. Log a message if the vnic_id
is invalid.
Fixes: db678d5c2b54 ("net/bnxt: add HWRM VNIC configure")
Signed-off-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_hwrm.c | 19 ++-
1 fi
This patch resets the VF stats during initialization
Fixes: b7778e8a1c00 ("net/bnxt: refactor to properly allocate resources for
PF/VF")
Signed-off-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_hwrm.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/bnxt/bnxt_h
remove the unnecessary rte_delay in bnxt_set_hwrm_link_config
Fixes: 7bc8e9a227cc ("net/bnxt: support async link notification")
Signed-off-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_hwrm.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_h
Fix use of local variable to avoid segfault.
cnt was incorrectly tested and decremented in the loop that removes
a VLAN from the table.
Fixes: 36735a932ca7 ("support set VF QOS and MAC anti spoof")
Signed-off-by: Stephen Hurd
Signed-off-by: Ajit Khaparde
--
v1->v2: incorporate review feedback.
1) handle_async_event is a DEBUG level log message.
2) Log "Unable to get default VNIC for VF %d" at INFO level.
Fixes: 36735a932ca7 ("support set VF QOS and MAC anti spoof")
Signed-off-by: Stephen Hurd
Signed-off-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_cpr.c | 2 +-
drivers/net/bnxt/r
This patch set fixes some of the issues found during testing.
Please apply.
v1->v2: incorporate review feedback
net/bnxt: fix log levels for non error conditions.
net/bnxt: fix to avoid a segfault
net/bnxt: fix vnic cleanup
net/bnxt: fix set link config
net/bnxt: reset VF stats during i
Enable the functions set link down and set link up in i40e by check
phy_type, and fix the issue of auto negotiation failed in XXV710 when
bind kernel driver after unbind from dpdk driver by modify the speed
setting distinguish from set link up and down. With this fix, if unbind
dpdk to bind kernel
Yes, I have already use cpu_to_le32 to translator, but I agree first defend it
as uint32_t might be better to read . I will send v2 , thanks for review.
Best regards,
Jeff Guo
-Original Message-
From: Wu, Jingjing
Sent: Friday, July 21, 2017 10:26 AM
To: Guo, Jia ; Xing, Beilei
Cc: de
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Michal Jastrzebski
> Sent: Thursday, July 13, 2017 5:49 PM
> To: dev@dpdk.org
> Cc: Jain, Deepak K ; Doherty, Declan
> ; Jastrzebski, MichalX K
>
> Subject: [dpdk-dev] [PATCH] app/testpmd: fix bonding initializat
> -Original Message-
> From: Guo, Jia
> Sent: Thursday, July 20, 2017 4:45 PM
> To: Xing, Beilei ; Wu, Jingjing
> Cc: dev@dpdk.org; Guo, Jia
> Subject: [PATCH] net/i40e: fix link down and negotiation issue
>
> Enable the functions set link down and set link up in i40e by check phy_type
Mirror rule id should not exceed the boundary defined by
IXGBE_MAX_MIRROR_RULES.
Fixes: fe3a45fd4104 ("ixgbe: add VMDq support")
Cc: sta...@dpdk.org
Signed-off-by: Qi Zhang
---
drivers/net/ixgbe/ixgbe_ethdev.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/net/ixgbe/ixgbe_ethdev
20/07/2017 15:17, Daniel Mrzyglod:
> There were mising :: so file was parsed in wrong way
>
> Fixes: c735b831b0c1 ("app/testpmd: add cmd for dedicated LACP Rx/Tx queues")
>
> Signed-off-by: Daniel Mrzyglod
Applied, thanks
> > >> Set UIO info device file operations open and release. Call pci reset
> > >> function inside open and release to clear device state at start and end.
> > >> Copied this behaviour from vfio_pci kernel module code. With this patch,
> > >> it is not mandatory to issue FLR by PMD's during init an
10/07/2017 15:09, Adrien Mazarguil:
> Adding a cover letter to keep the same title for that thread, since I took
> over the original patch from Bernard, modified it somewhat and made a couple
> of additional fixes on top of it.
>
> Olivier's comment [1] and Bernard's feedback about dynamic tokens
10/07/2017 18:13, Thomas Monjalon:
> The maintainer of librte_cfgfile is Cristian (+Cc),
> please set him as recipient for next time.
>
> 10/07/2017 14:44, Jacek Piasecki:
> > New API for cfgfile library allows to create a cfgfile at runtime, add new
> > section, add entry in a section, update exi
mlx5_tx_complete() polls completion queue multiple times until it
encounters an invalid entry. As Tx completions are suppressed by
MLX5_TX_COMP_THRESH, it is waste of cycles to expect multiple completions
in a poll. And freeing too many buffers in a call can cause high jitter.
This patch improve
As I said, there are primitives which are designed to handle frequent reads
and rare mutations.
Even with such primitives, rarely lock is more than never lock.
You do realize that the cache mutation involves ibv_dereg_mr() right?
Any locking scheme for mutation is negligible compared to that
mlx5_tx_complete() polls completion queue multiple times until it
encounters an invalid entry. As Tx completions are suppressed by
MLX5_TX_COMP_THRESH, it is waste of cycles to expect multiple completions
in a poll. And freeing too many buffers in a call can cause high jitter.
This patch improves t
20/07/2017 15:27, Gaetan Rivet:
> ctype.h is not compilable in BSD 10 on GCC 4.8 in C11 mode.
>
> == Build drivers/net/failsafe
> CC failsafe.o
> In file included from /usr/include/_ctype.h:94:0,
> from /usr/include/ctype.h:46,
> from /root/dpdk.org/build/includ
19/07/2017 17:52, Pablo de Lara:
> http://dpdk.org/git/next/dpdk-next-crypto
Pulled, thanks
Hi Sagi,
Thursday, July 20, 2017 5:06 PM, Sagi Grimberg:
> >> Its worse than just a drop, without debug enabled the error
> >> completion is ignored, and the wqe_pi is taken from an invalid field,
> >> which leads to bogus mbufs free (elts_tail is not valid).
> >
> > Right
>
> A simple work-arou
On 7/8/2017 7:28 AM, Yuanhan Liu wrote:
> On Tue, Jul 04, 2017 at 05:13:37PM +0100, Ferruh Yigit wrote:
>> @@ -157,8 +164,12 @@ rte_eth_dev_pci_generic_probe(struct rte_pci_device
>> *pci_dev,
>>
>> RTE_FUNC_PTR_OR_ERR_RET(*dev_init, -EINVAL);
>> ret = dev_init(eth_dev);
>> -if (re
Its worse than just a drop, without debug enabled the error completion
is ignored, and the wqe_pi is taken from an invalid field, which leads
to bogus mbufs free (elts_tail is not valid).
Right
A simple work-around would be to simply fill a correct tail so that
error completions will still h
HW pool manager e.g. Octeontx SoC demands s/w to program start and end
address of pool. Currently, there is no such handle in external mempool.
Introducing rte_mempool_update_range handle which will let HW(pool
manager)
know when common layer selects hugepage:
For each hugepage - update its start/e
xmem_size and xmem_usage need to know the status of mp->flag.
Following patch will make use of that.
Signed-off-by: Santosh Shukla
---
v1 -- v2:
- added new mempool param in xmem_size/usage, Per deprecation notice [1]
and discussion based on thread [2]
[1] http://dpdk.org/dev/patchwork/patch/268
Some mempool hw like octeontx/fpa block, demands block size
(/total_elem_sz) aligned object start address.
Introducing an MEMPOOL_F_POOL_BLK_SZ_ALIGNED flag.
If this flag is set:
- Align object start address to a multiple of total_elt_sz.
- Allocate one additional object. Additional object is need
mp->flags is int and mempool API updates unsigned int
value in 'flags', so fix the 'flags' data type.
Patch also does mp->flags cleanup like:
* Remove redundant 'flags' API description from
- __rte_mempool_generic_put
- __rte_mempool_generic_get
* Remove unused 'flags' param from
- rte_memp
The memory area containing all the objects must be physically
contiguous.
Introducing MEMPOOL_F_CAPA_PHYS_CONTIG flag for such use-case.
The flag useful to detect whether pool area has sufficient space
to fit all objects. If not then return -ENOSPC.
This way, we make sure that all object within a
Allow mempool to advertise its capability.
A handler been introduced called rte_mempool_ops_get_capabilities.
- Upon ->get_capabilities call, mempool driver will advertise
capability by updating to 'mp->flags'.
Signed-off-by: Santosh Shukla
Signed-off-by: Jerin Jacob
---
v2 -- v3:
- Changed vers
v3:
(Note: v3 work is based on deprecation notice [1], It's for 17.11)
* Changed _version.map from 17.08 to 17.11.
* build fixes reported by stv_sys.
* Patchset rebased on upstream commit: da94a999.
v2:
In order to support octeontx HW mempool manager, the common mempool layer must
meet below con
On Wed, Jul 19, 2017 at 09:21:39AM +0300, Sagi Grimberg wrote:
>
> > There is none, if you send a burst of 9 packets each one coming from a
> > different mempool the first one will be dropped.
>
> Its worse than just a drop, without debug enabled the error completion
> is ignored, and the wqe_pi
Unless the plan is for this patch to be squashed, ought to have a brief
commit message..
On 17/07/2017 16:15, Wei Dai wrote:
Signed-off-by: Wei Dai
---
doc/guides/prog_guide/poll_mode_drv.rst | 40 +
1 file changed, 40 insertions(+)
Reviewed-by: Remy Horton
See suggested wording inline
On 17/07/2017 16:14, Wei Dai wrote:
This patch adds a new eth_dev layer API function rte_eth_dev_reset().
A DPDK application can call this function to reset a NIC and keep
its port id afterwards.
It means that all SW resources allocated in ethdev layer should be
kept
ctype.h is not compilable in BSD 10 on GCC 4.8 in C11 mode.
== Build drivers/net/failsafe
CC failsafe.o
In file included from /usr/include/_ctype.h:94:0,
from /usr/include/ctype.h:46,
from /root/dpdk.org/build/include/rte_common.h:50,
from /root
There were mising :: so file was parsed in wrong way
Fixes: c735b831b0c1 ("app/testpmd: add cmd for dedicated LACP Rx/Tx queues")
Signed-off-by: Daniel Mrzyglod
---
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/doc/guides/testpmd
Hi.
New info on my problem.
"Tx descriptor is not done" error messages and tx queues stopped
transmitting packets showed up after I upgraded dpdk from 16.07 to
17.02.1 version and switched from using standalone ports to using the
bonding driver. Since then I've been trying to find a cause of the
20/07/2017 12:27, santosh:
> On Thursday 20 July 2017 02:16 PM, Olivier Matz wrote:
>
> > On Thu, 13 Jul 2017 09:12:31 +, Santosh Shukla
> > wrote:
> >> [PATCH] mempool: add notice to change mempool API/ABI
> > I think the usual titles for these notices are more:
> >
> > doc: announce API/AB
On 7/20/2017 8:25 AM, Wei Zhao wrote:
> This NIC type port check should use port index
> not mask for 82599ES.
>
> Fixes: 16f534e508d ("net/ixgbe: add support 82599ES SCTP packet drop action")
>
> Signed-off-by: Wei Zhao
Applied to dpdk-next-net/master, thanks.
On 7/20/2017 9:19 AM, Qi Zhang wrote:
> Only x550 family support sctp port in fdir filter, so
> add this limiation when parse consistent API.
>
> Fixes: 11777435c727 ("net/ixgbe: parse flow director filter")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Qi Zhang
Applied to dpdk-next-net/master, than
An API/ABI change is planned for 17.11 to change following
* Remove unused flag param from rte_mempool_generic_get and _put.
* Change data type for mempool 'flag' from int to unsigned int.
Refer [1].
* Add struct rte_mempool * param into func rte_mempool_xmem_size,
rte_mempool_xmem_usage to ma
On 7/20/2017 5:48 AM, Ajit Khaparde wrote:
> Hi,
> This patch set fixes some of the issues found during testing.
> Please apply.
>
> Thanks
>
> net/bnxt: fix log levels for non error conditions.
> net/bnxt: fix to avoid a segfault
> net/bnxt: fix vnic cleanup
> net/bnxt: fix set link conf
Sounds fine to me. Are you planning to add this, or you want me to do it?
On 20/07/2017, 12:09 PM, "Olivier Matz" wrote:
>Hi Ido,
>
>On Tue, 18 Jul 2017 12:18:20 +, "Ido Barnea (ibarnea)"
>wrote:
>> Hi,
>> Is it intentional that rte_ipv4_phdr_cksum and others in this file assume
>> tha
On 7/20/2017 5:48 AM, Ajit Khaparde wrote:
> Fix use of local variable to avoid segfault.
> cnt was incorrectly tested and decremented in the loop that removes
> a VLAN from the table.
>
> Fixes: 36735a932ca7 ("support set VF QOS and MAC anti spoof")
>
> Signed-off-by: Stephen Hurd
> Signed-off-
On Thursday 20 July 2017 02:16 PM, Olivier Matz wrote:
> On Thu, 13 Jul 2017 09:12:31 +, Santosh Shukla
> wrote:
>> [PATCH] mempool: add notice to change mempool API/ABI
> I think the usual titles for these notices are more:
>
> doc: announce API/ABI changes for mempool
in v2.
> Ideally, t
Document NIC features, add more information about them and add more
implementation related support.
Signed-off-by: Ferruh Yigit
Signed-off-by: John McNamara
Reviewed-by: Andrew Rybchenko
---
Cc: Olivier Matz
v5:
* Add flow_type_rss_offloads and RTE_PMD_REGISTER_KMOD_DEP
v4:
* Apply review co
On 7/8/2017 10:47 AM, Andrew Rybchenko wrote:
> On 07/07/2017 08:21 PM, Ferruh Yigit wrote:
>> Document NIC features, add more information about them and add more
>> implementation related support.
>>
>> Signed-off-by: Ferruh Yigit
>> Signed-off-by: John McNamara
>
> Reviewed-by: Andrew Rybchenk
Hi Ido,
On Tue, 18 Jul 2017 12:18:20 +, "Ido Barnea (ibarnea)"
wrote:
> Hi,
> Is it intentional that rte_ipv4_phdr_cksum and others in this file assume
> that ipv4 header is sizeof(struct ipv4_hdr), actually assuming that there are
> no IPv4 options?
> If not, I would like to submit a patc
On Thu, Jul 20, 2017 at 11:32:18AM +0300, Thomas Monjalon wrote:
> 20/07/2017 10:56, Jens Freimann:
> > On Wed, Jul 19, 2017 at 06:23:21PM +0800, Tiwei Bie wrote:
> > >On Wed, Jul 19, 2017 at 05:24:38PM +0800, Van Haaren, Harry wrote:
> > [...]
> > > > Hi Tiwei,
> > > >
> > > > Although the idea an
Hi,
On Fri, 30 Jun 2017 16:26:09 +0200, Olivier Matz wrote:
> The initial objective of
> commit d9f0d3a1ffd4 ("ring: remove split cacheline build setting")
> was to add an empty cache line betwee, the producer and consumer
> data (on platform with cache line size = 64B), preventing from
> having
On Thu, 13 Jul 2017 09:12:31 +, Santosh Shukla
wrote:
> [PATCH] mempool: add notice to change mempool API/ABI
I think the usual titles for these notices are more:
doc: announce API/ABI changes for mempool
Ideally, the title should describe more precisely the kind of
changes. In that partic
Enable the functions set link down and set link up in i40e by check
phy_type, and fix the issue of auto negotiation failed in XXV710 when
bind kernel driver after unbind from dpdk driver by modify the speed
setting distinguish from set link up and down. With this fix, if unbind
dpdk to bind kernel
20/07/2017 10:56, Jens Freimann:
> On Wed, Jul 19, 2017 at 06:23:21PM +0800, Tiwei Bie wrote:
> >On Wed, Jul 19, 2017 at 05:24:38PM +0800, Van Haaren, Harry wrote:
> [...]
> >> Hi Tiwei,
> >>
> >> Although the idea and motivation for code-cleanup are good, performing
> >> large cleanup across a cod
On Wed, Jul 19, 2017 at 06:23:21PM +0800, Tiwei Bie wrote:
On Wed, Jul 19, 2017 at 05:24:38PM +0800, Van Haaren, Harry wrote:
[...]
Hi Tiwei,
Although the idea and motivation for code-cleanup are good, performing
large cleanup across a code-base is not a good solution. The reason that
these ty
This NIC type port check should use port index
not mask for 82599ES.
Fixes: 16f534e508d ("net/ixgbe: add support 82599ES SCTP packet drop action")
Signed-off-by: Wei Zhao
---
drivers/net/ixgbe/ixgbe_flow.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ixgbe
add support for ipv4-other type in ntuple filter.
Signed-off-by: Wei Zhao
---
drivers/net/ixgbe/ixgbe_flow.c | 12 +---
1 file changed, 9 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_flow.c b/drivers/net/ixgbe/ixgbe_flow.c
index 04540af..2bb5c46 100644
--- a/driver
DPDK has support for both sw and hw mempool and
currently user is limited to use ring_mp_mc pool.
In case user want to use other pool handle,
need to update config RTE_MEMPOOL_OPS_DEFAULT, then
build and run with desired pool handle.
Introducing eal option to override default pool handle.
Now use
Now that dpdk supports more than one mempool drivers and
each mempool driver works best for specific PMD, example:
- sw ring based mempool for Intel PMD drivers
- dpaa2 HW mempool manager for dpaa2 PMD driver.
- fpa HW mempool manager for Octeontx PMD driver.
Application like to know `preferred me
DPDK has support for hw and sw mempool. Those mempool
can work optimal for specific PMD's.
Example:
sw ring based PMD for Intel NICs.
HW mempool manager dpaa2 for dpaa2 PMD.
HW mempool manager fpa for octeontx PMD.
There could be a use-case where different vendor NIC's used
on the same platform a
On Thursday 20 July 2017 11:23 AM, Tiwei Bie wrote:
On Thu, Jul 20, 2017 at 10:34:39AM +0530, Shreyansh Jain wrote:
On Wednesday 19 July 2017 02:36 PM, Tiwei Bie wrote:
Remove the unwanted spaces before `;' across DPDK source code
by below one-liner with some minor manual refinements.
find . -
71 matches
Mail list logo
