This patch updates the base driver version to 8.30.8.0.
Signed-off-by: Rasesh Mody
---
drivers/net/qede/base/ecore.h |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/net/qede/base/ecore.h b/drivers/net/qede/base/ecore.h
index eaf2604..ce5f3a9 100644
--- a/driv
IOV regression testing led to discovery of a minor issue + possibly race in
IOV flows:
a. Malicious indications in VF-database on PF-side get cleared during FLR
flows - but not when disabling SRIOV. At least in linux if you disable
IOV while having a malicious VF you wouldn't be able to cl
This change aligns the PF behavior with that of the L2-fw, so that flows
(to remove an incorrect MAC) for PF and VF could remain the same.
Fixes: 86a2265e59d7 ("qede: add SRIOV support")
Cc: sta...@dpdk.org
Signed-off-by: Rasesh Mody
---
drivers/net/qede/base/ecore_sriov.c |9 +
1 f
- Add non-L2 personality macros and add a personality check before setting
the HW features
Signed-off-by: Rasesh Mody
---
drivers/net/qede/base/ecore.h | 14 +-
drivers/net/qede/base/ecore_dev.c | 15 +--
2 files changed, 18 insertions(+), 11 deletions(-)
dif
- Move define CRC8_TABLE_SIZE to header file
- Move/relocate ecore_vf_start()
- Other semantic changes
Signed-off-by: Rasesh Mody
---
drivers/net/qede/base/ecore.h |2 ++
drivers/net/qede/base/ecore_dcbx.c |8 +++
drivers/net/qede/base/ecore_dev.c
Fix an access to an uninitialized list when the management FW is not
initialized by simply doing the list initialization always, at a previous
step, before ecore_mcp_cmd_init() can stop in the middle and return.
Fixes: c1796ac8da2d ("net/qede/base: revise management FW mbox access scheme")
Cc: sta
Change the default page size of internal lookup table (ILT) clients to
64K
Signed-off-by: Rasesh Mody
---
drivers/net/qede/base/ecore_cxt.c |7 ++-
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/drivers/net/qede/base/ecore_cxt.c
b/drivers/net/qede/base/ecore_cxt.c
index f
- Remove some dead definitions, function declarations and unused
variables
- Remove an obsolete workaround from ecore_int_igu_enable()
- Remove set variables that are not used
- Remove needless check in ecore_init_wfq_param() when configuring
minimum vport BW. We already check whether to
Changes include:
- adding new log to print Dcbx version
- reformatting log in ecore_dmae_operation_wait() in case of DMA engine
failure
- changing verbosity of some log messages such as:
VFs incorrect behavior should be logged on PF with DP_VERBOSE(), not
DP_NOTICE(). In general keep IO
Fix few comments and add a new one describing SW channel implementation
limitation
Signed-off-by: Rasesh Mody
---
drivers/net/qede/base/ecore_dcbx.c |2 +-
drivers/net/qede/base/ecore_mcp.c |2 +-
drivers/net/qede/base/ecore_sriov.h |2 +-
drivers/net/qede/base/ecore_vf.c|
Check DMA engine state to verify it is not called before it is initialized
Signed-off-by: Rasesh Mody
---
drivers/net/qede/base/ecore_hw.c |9 +
1 file changed, 9 insertions(+)
diff --git a/drivers/net/qede/base/ecore_hw.c b/drivers/net/qede/base/ecore_hw.c
index d6217d8..bc0138f 10
- Introduce OSAL_IOV_VF_VPORT_STOP to allow VF to carry out required
operations and prevent a potential assert before closing vport
- Add OSAL_DIV_S64() for 64-bit division on 32-bit platforms.
- Add OSAL for transceiver update OSAL_TRANSCEIVER_UPDATE()
- Add OSAL for MFW command preemption
Use the correct size value while copying chassis/port id values.
Signed-off-by: Rasesh Mody
---
drivers/net/qede/base/ecore_dcbx.c |8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/net/qede/base/ecore_dcbx.c
b/drivers/net/qede/base/ecore_dcbx.c
index 97fd484
This patch adds base driver lldp protocol support and handle lldp received
TLV update events.
Signed-off-by: Rasesh Mody
---
drivers/net/qede/base/bcm_osal.h |2 +
drivers/net/qede/base/ecore_dcbx.c | 233
drivers/net/qede/base/ecore_dcbx.h |
Introduce 2 new API functions, one for the VF and the other for PF [per-VF]
which allows to decide whether to use HW/SW channel for PF<->VF
communication(a per-VF configuration). A HyperV might have different VMs
with different requirements.
Signed-off-by: Rasesh Mody
---
drivers/net/qede/base/e
Add NVRAM config options:
- Rename caution temperature threshold
- New field EXT_PHY_MDI_PAIR_SWAP
- BCM54220 1GBT PHY
Signed-off-by: Rasesh Mody
---
drivers/net/qede/base/nvm_cfg.h | 16 ++--
1 file changed, 10 insertions(+), 6 deletions(-)
diff --git a/drivers/net/qede/base/n
Add support to firmware for:
- New SFP type 1000BaseT
- DON (Diag Over Network). This feature implements a server side for
process data access commands over Ethernet.
Signed-off-by: Rasesh Mody
---
drivers/net/qede/base/mcp_public.h |4
1 file changed, 4 insertions(+)
diff --git a
Hi,
The changes in this patch set include few enhancements, semantic changes,
code cleanup and fixes. It updates base driver to version 8.30.8.0.
Thanks!
-Rasesh
Rasesh Mody (17):
net/qede/base: add xcvr type and DON FW defines
net/qede/base: add NVRAM config options
net/qede/base: introdu
On Saturday 07 October 2017 09:56 AM, Ferruh Yigit wrote:
> On 10/7/2017 4:51 AM, santosh wrote:
>> On Saturday 07 October 2017 02:25 AM, Thomas Monjalon wrote:
>>> 19/09/2017 10:29, santosh:
Ping?
>>> Pong
>>>
>>> It seems you forgot this series.
>> This series review is pending and planned
On 10/7/2017 4:51 AM, santosh wrote:
>
> On Saturday 07 October 2017 02:25 AM, Thomas Monjalon wrote:
>> 19/09/2017 10:29, santosh:
>>> Ping?
>> Pong
>>
>> It seems you forgot this series.
>
> This series review is pending and planned for -rc2 window.
Hi Santosh,
octeontx net pmd has dependency
On Saturday 07 October 2017 02:25 AM, Thomas Monjalon wrote:
> 19/09/2017 10:29, santosh:
>> Ping?
> Pong
>
> It seems you forgot this series.
This series review is pending and planned for -rc2 window.
> There is a compilation error and some doc changes requested.
Yes, I have renamed few mempoo
On Saturday 07 October 2017 02:21 AM, Thomas Monjalon wrote:
> 31/08/2017 08:37, Santosh Shukla:
>> Upon pool allocation request by application, Octeontx FPA alloc
>> does following:
>> - Gets free pool from pci fpavf array.
>> - Uses mbox to communicate fpapf driver about,
>> * gpool-id
>> *
Hi Ferruh,
> -Original Message-
> From: Yigit, Ferruh
> Sent: Saturday, October 7, 2017 7:26 AM
> To: Yao, Lei A ; Hu, Jiayu ;
> dev@dpdk.org
> Cc: Tan, Jianfeng ; Ananyev, Konstantin
> ; tho...@monjalon.net; Wu, Jingjing
>
> Subject: Re: [dpdk-dev] [PATCH v4] app/testpmd: enable the heav
On 10/6/2017 7:03 PM, John Miller wrote:
> Change loop counter that should be based on the number
> of rx queues, not tx queues. This only affects debug
> output.
>
> Signed-off-by: John Miller
Applied to dpdk-next-net/master, thanks.
On 10/7/2017 12:32 AM, Ferruh Yigit wrote:
> On 10/5/2017 9:36 PM, Mark Kavanagh wrote:
>> Generic Segmentation Offload (GSO) is a SW technique to split large
>> packets into small ones. Akin to TSO, GSO enables applications to
>> operate on large packets, thus reducing per-packet processing overhe
When copying VLAN tags from the RX descriptor to the vlan_tci field in the
mbuf header, igb_rxtx.c:eth_igb_recv_pkts() and
eth_igb_recv_scattered_pkts() both assume that the VLAN tag is always
little endian. While i350 VLAN non-loopback packets are stored
little endian, VLAN tags for i350 VLAN lo
On 10/5/2017 9:36 PM, Mark Kavanagh wrote:
> Generic Segmentation Offload (GSO) is a SW technique to split large
> packets into small ones. Akin to TSO, GSO enables applications to
> operate on large packets, thus reducing per-packet processing overhead.
>
> To enable more flexibility to applicati
> -Original Message-
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Thursday 5 October 2017 23:23
> To: Mokhtar, Amr
> Cc: dev@dpdk.org; f...@redhat.com; acon...@redhat.com; bl...@debian.org
> Subject: Re: [dpdk-dev] [RFC] Wireless Base Band Device (bbdev)
>
> 05/10/2017 2
On 10/7/2017 12:04 AM, Ferruh Yigit wrote:
> On 9/27/2017 8:23 AM, Yao, Lei A wrote:
> <...>
>
>>> The GRO library provides two modes to reassemble packets. Currently, the
>>> csum forwarding engine has supported to use the lightweight mode to
>>> reassemble TCP/IPv4 packets. This patch introduces
On 9/27/2017 8:23 AM, Yao, Lei A wrote:
<...>
>> The GRO library provides two modes to reassemble packets. Currently, the
>> csum forwarding engine has supported to use the lightweight mode to
>> reassemble TCP/IPv4 packets. This patch introduces the heavyweight mode
>> for TCP/IPv4 GRO in the csu
On Fri, Oct 06, 2017 at 10:52:38AM +0200, Nélio Laranjeiro wrote:
> On Thu, Oct 05, 2017 at 08:26:38PM -0700, Yongseok Koh wrote:
> > On Thu, Oct 05, 2017 at 02:49:42PM +0200, Nelio Laranjeiro wrote:
> > [...]
> > > +struct mlx5_rxq_ibv*
> > > +mlx5_priv_rxq_ibv_get(struct priv *priv, uint16_t idx)
On 10/6/2017 11:51 PM, Ferruh Yigit wrote:
> On 9/18/2017 8:46 AM, nic at opencloud.tech (Tonghao Zhang) wrote:
>> From: Tonghao Zhang
>>
>> This is not bugfix, but it's convenient to help developer
>> to review and maintain the igbuio codes.
>>
>> Signed-off-by: Tonghao Zhang
>
> Acked-by: Ferr
On 9/18/2017 8:46 AM, nic at opencloud.tech (Tonghao Zhang) wrote:
> From: Tonghao Zhang
>
> This is not bugfix, but it's convenient to help developer
> to review and maintain the igbuio codes.
>
> Signed-off-by: Tonghao Zhang
Acked-by: Ferruh Yigit
On Fri, Oct 06, 2017 at 09:03:25AM +0200, Nélio Laranjeiro wrote:
> On Thu, Oct 05, 2017 at 09:59:58PM -0700, Yongseok Koh wrote:
> > On Thu, Oct 05, 2017 at 02:49:47PM +0200, Nelio Laranjeiro wrote:
> > [...]
> > > +struct mlx5_hrxq*
> > > +mlx5_priv_hrxq_get(struct priv *priv, uint8_t *rss_key, u
Hi, Martin
Even though I had done quite serious tests before sending out the patch,
I figured out deadlock could happen if the Rx queue size is smaller. It is 128
by default in testpmd while I usually use 256.
I've fixed the bug and submitted a new patch [1], which actually reverts the
previous p
22/09/2017 17:33, De Lara Guarch, Pablo:
> From: Vladimir Medvedkin
> > 2017-08-22 15:02 GMT+03:00 Yangchao Zhou :
> >
> > > Use rte_bsf32 and fast bit unset operation to optimize the softrss
> > > computation.
> > > The following measurements shows improvement over the default softrss
> > > compu
When replenishing Rx ring, there're always buffered slots reserved between
consumed entries and HW owned entries. These have to be filled with fake
mbufs to protect from possible overflow rather than optimistically
expecting successful replenishment which can cause deadlock with
small-sized queue.
22/09/2017 15:46, Bruce Richardson:
> On Fri, Sep 22, 2017 at 05:25:43AM +0100, Pablo de Lara wrote:
> > When adding a new entry in a hash table, there is
> > a maximum number of evictions that can be
> > performed. When the counter of these evictions reaches
> > this maximum, the entry cannot be a
On 9/14/2017 1:04 PM, Markus Theil wrote:
<...>
> On 11.09.2017 19:48, Ferruh Yigit wrote:
>> On 9/5/2017 1:04 PM, Markus Theil wrote:
>>> Interrupt setup code in igb_uio has to deal with multiple
>>> types of interrupts and kernel versions. This patch moves
>>> the setup and teardown code into own
04/10/2017 14:59, Burakov, Anatoly:
> On 15-Aug-17 2:53 PM, Keith Wiles wrote:
> > When using EXTRA_CFLAGS="-g -O3" in the build the -O3 causes
> > compiler warnings for these two applications. Using Ubuntu 17.04
> > gcc compiler.
> >
> > Signed-off-by: Keith Wiles
>
> Acked-by: Anatoly Burakov
On 8/18/2017 5:03 PM, Ferruh Yigit wrote:
> On 8/11/2017 5:33 PM, Nirmoy Das wrote:
>> detect SLE version reverse chronologically as ">=" is being used.
>>
>> Signed-off-by: Nirmoy Das
>
> Fixes: 2972254ce163 ("kni: fix build on Suse 12 SP3")
>
> Acked-by: Ferruh Yigit
Applied to dpdk/master,
On 9/5/2017 10:55 AM, Nirmoy Das wrote:
>
>
> On 09/04/2017 11:46 AM, Ferruh Yigit wrote:
>> On 8/29/2017 4:06 PM, Nirmoy Das wrote:
>>> compilation error:
>>> build/lib/librte_eal/linuxapp/kni/kni_net.c:215:5: error:
>>> ‘struct net_device’ has no member named ‘trans_start’
>>> dev->trans_star
14/09/2017 13:31, Hemant Agrawal:
> On 9/14/2017 9:23 AM, Jianbo Liu wrote:
> > The kernel patch was merged to support pci resource mapping.
> > https://patchwork.kernel.org/patch/9677441/
> >
> > So enable igu_uio in the default arm64 configuration.
> >
> > v2:
> > - keep headline format
> >
> >
On 10/6/2017 10:21 AM, Declan Doherty wrote:
> During a link down event of a port participating in a LACP 802.3ad
> bond the current behaviour can cause all ports to be unselected
> and temporarily stop all traffic on the bond, causing unexpected
> traffic loss across all ports and not just the por
When using Python 3, dpdk-devbind.py fails to detect modules other than
igb_uio.
Signed-off-by: Omri Mor
---
usertools/dpdk-devbind.py | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/usertools/dpdk-devbind.py b/usertools/dpdk-devbind.py
index da6e40cbe..a539995cd 100755
---
On 6/6/2017 7:03 AM, Gregory Etelson wrote:
> re-bind PCI function to igb_uio
> after process termination
> tested adapters: Intel 10G, Intel 40G
> known issues: ENA adapters on AWS I3 instances fail to reset PCI function
>
> Signed-off-by: Gregory Etelson
Shijith's patch [1] applied in 17.08 sup
On 9/18/2017 7:22 PM, Mcnamara, John wrote:
>
>
>> -Original Message-
>> From: Yigit, Ferruh
>> Sent: Monday, September 11, 2017 5:34 PM
>> To: Mcnamara, John
>> Cc: dev@dpdk.org; Yigit, Ferruh ; Stephen Hurd
>> ; Ajit Khaparde ;
>> Wu, Jingjing ; Zhang, Helin
>> ; Zhang, Helin ; Ananyev
19/09/2017 10:29, santosh:
> Ping?
Pong
It seems you forgot this series.
There is a compilation error and some doc changes requested.
About commit titles, you might probably reword starting
without "implement pool" which is redundant with "mempool/octeontx:".
31/08/2017 08:37, Santosh Shukla:
> Upon pool allocation request by application, Octeontx FPA alloc
> does following:
> - Gets free pool from pci fpavf array.
> - Uses mbox to communicate fpapf driver about,
> * gpool-id
> * pool block_sz
> * alignemnt
> - Programs fpavf pool boundary.
>
> S
Hi,
We need a Signed-off-by to process your contribution.
Thank you
+Cc sergio.gonzalez.mon...@intel.com
21/09/2017 05:19, liucheng (J):
> There is no lock in function malloc_heap_get_stats. When we call this
> function, the elem may free by other thread.
>
> the call stack:
> Core was generat
> Santosh Shukla (8):
> mempool: remove unused flags argument
> mempool: change flags from int to unsigned int
> mempool: add flags arg in xmem size and usage
> doc: remove mempool notice
> mempool: get the mempool capability
> mempool: detect physical contiguous object in pool
> memp
> Santosh Shukla (2):
> eal: allow user to override default pool handle
> ethdev: get the supported pool for a port
I had not reviewed it before yesterday.
It seems functionnally ready.
I have some reserves about naming and placing but I could try
to address them later.
Applied, thanks
On 10/6/2017 5:59 PM, Jasvinder Singh wrote:
> The SoftNIC PMD is intended to provide SW fall-back options for specific
> ethdev APIs in a generic way to the NICs not supporting those features.
>
> Currently, the only implemented ethdev API is Traffic Management (TM),
> but other ethdev APIs such
On 10/6/2017 7:43 AM, Maxime Coquelin wrote:
>
>
> On 09/11/2017 05:13 PM, Olivier Matz wrote:
>> The compilation with gcc-6.3.0 and EXTRA_CFLAGS=-Og gives the following
>> error:
>>
>>CC virtio_rxtx.o
>>virtio_rxtx.c: In function ‘virtio_rx_offload’:
>>virtio_rxtx.c:680:10: error: ‘c
> Santosh Shukla (9):
> eal/pci: export match function
> eal/pci: get iommu class
> linuxapp/eal_pci: get iommu class
> bus: get iommu class
> eal: introduce helper API for iova mode
> eal: auto detect iova mode
> linuxapp/eal_vfio: honor iova mode before mapping
> linuxapp/eal_memo
> -Original Message-
> From: Nicolau, Radu
> Sent: Friday, October 6, 2017 10:18 AM
> To: Ananyev, Konstantin ; Akhil Goyal
> ; dev@dpdk.org
> Cc: Doherty, Declan ; De Lara Guarch, Pablo
> ; hemant.agra...@nxp.com;
> bor...@mellanox.com; avia...@mellanox.com; tho...@monjalon.net;
> san
On 10/6/2017 4:45 PM, Xueming Li wrote:
>
> This patchset enhances Mellanox multi-process by supporting all multi-process
> examples, also support reading ethdev (x)stats in secondary process.
>
> V5:
> * rebase on latest upstream code, patch 2/5 mlx5_ethdev.c updated.
>
> V4:
> * remove forked
On 10/6/2017 4:45 PM, Xueming Li wrote:
> Use a unix socket to get back the communication channel with the Kernel
> driver from the primary process, this is necessary to remap those pages
> in the secondary process memory space and thus use the same Tx queues.
>
> This is only supported from rdma-
Ipsec-secgw application is modified so that it can support
following type of actions for crypto operations
1. full protocol offload using crypto devices.
2. inline ipsec using ethernet devices to perform crypto operations
3. full protocol offload using ethernet devices.
4. non protocol offload
Sig
From: Radu Nicolau
Signed-off-by: Radu Nicolau
Signed-off-by: Declan Doherty
---
config/common_base | 1 +
drivers/net/Makefile | 2 +-
drivers/net/ixgbe/Makefile | 4 +-
drivers/net/ixgbe/ixgbe_ethdev.c | 17 +
drivers/net/ixgbe/i
Driver implementation to support rte_security APIs
Signed-off-by: Akhil Goyal
---
doc/guides/cryptodevs/features/dpaa2_sec.ini | 1 +
drivers/crypto/Makefile | 2 +-
drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c | 411 ++-
drivers/crypto/dpaa2_sec/
Signed-off-by: Akhil Goyal
Signed-off-by: Radu Nicolau
---
MAINTAINERS| 6 ++
config/common_base | 6 ++
lib/Makefile | 5 +
mk/rte.app.mk | 1 +
4 files changed, 18 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index a0cd75e..f5b790e 100644
--- a/MAINTAINE
From: Boris Pismenny
Signed-off-by: Boris Pismenny
Reviewed-by: John McNamara
---
doc/guides/prog_guide/rte_flow.rst | 84 +-
1 file changed, 82 insertions(+), 2 deletions(-)
diff --git a/doc/guides/prog_guide/rte_flow.rst
b/doc/guides/prog_guide/rte_flow.
From: Boris Pismenny
The crypto action is specified by an application to request
crypto offload for a flow.
Signed-off-by: Boris Pismenny
Signed-off-by: Aviad Yehezkel
---
lib/librte_ether/rte_flow.h | 38 ++
1 file changed, 38 insertions(+)
diff --git a/l
From: Declan Doherty
rte_flow_action type and ethdev updated to support rte_security
sessions for crypto offload to ethernet device.
Signed-off-by: Boris Pismenny
Signed-off-by: Aviad Yehezkel
Signed-off-by: Radu Nicolau
Signed-off-by: Declan Doherty
---
lib/librte_ether/rte_ethdev.c
From: Boris Pismenny
Add security crypto flags and update mbuf fields to support
IPsec crypto offload for transmitted packets, and to indicate
crypto result for received packets.
Signed-off-by: Aviad Yehezkel
Signed-off-by: Boris Pismenny
Signed-off-by: Radu Nicolau
---
lib/librte_mbuf/rte_m
Signed-off-by: Hemant Agrawal
Signed-off-by: Akhil Goyal
Acked-by: John McNamara
---
doc/api/doxy-api-index.md | 3 +-
doc/api/doxy-api.conf | 1 +
doc/guides/prog_guide/index.rst| 1 +
doc/guides/prog_guide/rte_security.rst | 553
From: Boris Pismenny
The ESP header is required for IPsec crypto actions.
Signed-off-by: Boris Pismenny
Signed-off-by: Aviad Yehezkel
---
doc/api/doxy-api-index.md | 1 +
lib/librte_ether/rte_flow.h | 26
lib/librte_net/Makefile | 2 +-
lib/librte_net/rte_esp.h
Security ops are added to crypto device to support
protocol offloaded security operations.
Signed-off-by: Akhil Goyal
Signed-off-by: Declan Doherty
---
doc/guides/cryptodevs/features/default.ini | 1 +
lib/librte_cryptodev/rte_crypto.h | 3 ++-
lib/librte_cryptodev/rte_crypto
rte_security library provides APIs for security session
create/free for protocol offload or offloaded crypto
operation to ethernet device.
Signed-off-by: Akhil Goyal
Signed-off-by: Boris Pismenny
Signed-off-by: Radu Nicolau
Signed-off-by: Declan Doherty
---
lib/librte_security/Makefile
This patchset introduce the rte_security library in DPDK.
This also includes the sample implementation of drivers and
changes in ipsec gateway application to demonstrate its usage.
rte_security library is implemented on the idea proposed earlier [1],[2],[3]
to support IPsec Inline and look aside
Removed legacy writes to ORT/PIT registers from
i40e_GLQF_reg_init(struct i40e_hw *hw) function.
Latest NVM versions contain all relevant values
and these values should not be overwritten by SW to
maintain driver/firmware compatibility and to avoid
conflicts with dynamic device personalization prof
Hi Konstantin,
Thanks for your comments.
On 10/5/2017 10:00 PM, Ananyev, Konstantin wrote:
Hi lads,
rte_security library provides APIs for security session
create/free for protocol offload or offloaded crypto
operation to ethernet device.
Signed-off-by: Akhil Goyal
Signed-off-by: Boris Pism
Change loop counter that should be based on the number
of rx queues, not tx queues. This only affects debug
output.
Signed-off-by: John Miller
---
drivers/net/ark/ark_ethdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ark/ark_ethdev.c b/drivers/net/ark/ark_
On Fri, Oct 6, 2017 at 3:37 PM, Thomas Monjalon wrote:
> 06/10/2017 15:12, Shreyansh Jain:
>> On Friday 06 October 2017 04:51 AM, Thomas Monjalon wrote:
>> > 19/09/2017 20:51, Jan Blunck:
>> >> On Mon, Sep 18, 2017 at 1:36 PM, Hemant Agrawal
>> >> wrote:
>> >>> Tested-by: Hemant Agrawal
>> >>>
On Thu, Oct 05, 2017 at 02:50:00PM +0200, Nelio Laranjeiro wrote:
> Add RSS support according to the RSS configuration.
>
> A special case is handled, when the pattern does not cover the RSS hash
> configuration request such as:
>
> flow create 0 ingress pattern eth / end actions rss queues 0 1
06/10/2017 18:21, Mcnamara, John:
> Hi,
>
> At the recent DPDK 2017 Userspace in Dublin we had a discussion about API/ABI
> stability with representation from Intel, OVS-DPDK, Debian, Red Hat and
> plenty of input from the floor.
>
> The main consensus after looking at different proposals was t
> > Fixes: 866c306ff03f ("mempool/dpaa: support NXP DPAA mempool")
> > Fixes: 37f9b54bd3cf ("net/dpaa: support Tx and Rx queue setup")
> >
> > Signed-off-by: Thomas Monjalon
>
> Acked-by: Hemant Agrawal
Applied
06/10/2017 18:25, Luca Boccassi:
> On Fri, 2017-10-06 at 16:05 +, Mcnamara, John wrote:
> > Hi,
> >
> > At the recent DPDK 2017 Userspace in Dublin it was agreed that we
> > should have a DPDK 17.11 LTS with 2 year support that would run in
> > parallel with the DPDK 16.11 LTS.
> >
> > Howeve
06/10/2017 18:22, Hemant Agrawal:
> On 10/6/2017 5:11 AM, Thomas Monjalon wrote:
> > 13/07/2017 13:48, Hemant Agrawal:
> >> In case no_pci is configured, fslmc bus will still need the
> >> the vfio to be enabled.
> >>
> >> Signed-off-by: Hemant Agrawal
> >> ---
> >> --- a/lib/librte_eal/linuxapp/e
This commit extends the testpmd application with new forwarding engine
that demonstrates the use of ethdev traffic management APIs and softnic
PMD for QoS traffic management.
In this mode, 5-level hierarchical tree of the QoS scheduler is built
with the help of ethdev TM APIs such as shaper profil
Implement ethdev TM hierarchy related APIs in SoftNIC PMD.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Jasvinder Singh
Acked-by: Lu, Wenzhuo
Acked-by: Cristian Dumitrescu
Acked-by: Thomas Monjalon
---
v5 change:
- add macro for the tc period
- add more comments
drivers/net/softnic/r
Implement ethdev TM capability APIs in SoftNIC PMD.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Jasvinder Singh
Acked-by: Lu, Wenzhuo
Acked-by: Cristian Dumitrescu
Acked-by: Thomas Monjalon
---
drivers/net/softnic/rte_eth_softnic.c | 12 +-
drivers/net/softnic/rte_eth_soft
Add SoftNIC PMD to provide SW fall-back for ethdev APIs.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Jasvinder Singh
Acked-by: Cristian Dumitrescu
Acked-by: Thomas Monjalon
---
v5 changes:
- change function name rte_pmd_softnic_run_default() to run_default()
v4 changes:
- Implemented
Add ethdev Traffic Management API support to SoftNIC PMD.
Signed-off-by: Cristian Dumitrescu
Signed-off-by: Jasvinder Singh
Acked-by: Lu, Wenzhuo
Acked-by: Cristian Dumitrescu
Acked-by: Thomas Monjalon
---
v5 changes:
- change function name rte_pmd_softnic_run_tm() to run_tm()
v3 changes:
The SoftNIC PMD is intended to provide SW fall-back options for specific
ethdev APIs in a generic way to the NICs not supporting those features.
Currently, the only implemented ethdev API is Traffic Management (TM),
but other ethdev APIs such as rte_flow, traffic metering & policing, etc
can be ea
Hi,
Comments inline, thanks for reviewing!
On 10/4/2017 11:52 AM, Shahaf Shuler wrote:
Tuesday, October 3, 2017 4:14 PM, Akhil Goyal:
From: Declan Doherty
rte_flow_action type and ethdev updated to support rte_security sessions
for crypto offload to ethernet device.
Signed-off-by: Boris Pi
On 10/6/2017 5:21 PM, Mcnamara, John wrote:
> Hi,
>
> At the recent DPDK 2017 Userspace in Dublin we had a discussion about API/ABI
> stability with representation from Intel, OVS-DPDK, Debian, Red Hat and
> plenty of input from the floor.
>
>
> The main consensus after looking at different pr
Acked-by: Hemant Agrawal
On 10/6/2017 9:40 PM, Thomas Monjalon wrote:
Fixes: 866c306ff03f ("mempool/dpaa: support NXP DPAA mempool")
Fixes: 37f9b54bd3cf ("net/dpaa: support Tx and Rx queue setup")
Signed-off-by: Thomas Monjalon
---
drivers/mempool/dpaa/dpaa_mempool.h | 1 -
drivers/net/dpaa
On Fri, 2017-10-06 at 16:05 +, Mcnamara, John wrote:
> Hi,
>
> At the recent DPDK 2017 Userspace in Dublin it was agreed that we
> should have a DPDK 17.11 LTS with 2 year support that would run in
> parallel with the DPDK 16.11 LTS.
>
> However, if order for this to happen the community will
On 10/6/2017 5:11 AM, Thomas Monjalon wrote:
13/07/2017 13:48, Hemant Agrawal:
In case no_pci is configured, fslmc bus will still need the
the vfio to be enabled.
Signed-off-by: Hemant Agrawal
---
--- a/lib/librte_eal/linuxapp/eal/eal.c
+++ b/lib/librte_eal/linuxapp/eal/eal.c
+#ifdef RTE_LIBRT
Hi,
At the recent DPDK 2017 Userspace in Dublin we had a discussion about API/ABI
stability with representation from Intel, OVS-DPDK, Debian, Red Hat and plenty
of input from the floor.
The main consensus after looking at different proposals was to maintain the
current process (http://dpdk.or
Fixes: 866c306ff03f ("mempool/dpaa: support NXP DPAA mempool")
Fixes: 37f9b54bd3cf ("net/dpaa: support Tx and Rx queue setup")
Signed-off-by: Thomas Monjalon
---
drivers/mempool/dpaa/dpaa_mempool.h | 1 -
drivers/net/dpaa/dpaa_rxtx.c| 2 --
2 files changed, 3 deletions(-)
diff --git a/d
Hi,
At the recent DPDK 2017 Userspace in Dublin it was agreed that we should have a
DPDK 17.11 LTS with 2 year support that would run in parallel with the DPDK
16.11 LTS.
However, if order for this to happen the community will have to provide a
maintainer, either for 17.11 LTS or for 16.11 LTS
06/10/2017 04:15, Ferruh Yigit:
> On 9/29/2017 8:17 AM, Zhiyong Yang wrote:
> > Zhiyong Yang (5):
> > net/bonding: remove bonding APIs using ABI versioning
> > ethdev: increase port_id range
> > examples: increase port_id range
> > test: increase port_id range
> > librte_mbuf: modify port
> -Original Message-
> From: Ferruh Yigit [mailto:ferruh.yi...@intel.com]
> Sent: Thursday, October 5, 2017 8:18 AM
> To: Xueming(Steven) Li ; Nélio Laranjeiro
> ; Adrien Mazarguil
>
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v4 0/5] net/mlx5 multi-process support
>
> On 9/19/2
This patch update the feature list and NIC guide to be multi-process
enabled.
Signed-off-by: Xueming Li
Acked-by: Nelio Laranjeiro
---
doc/guides/nics/features/mlx5.ini | 1 +
doc/guides/nics/mlx5.rst | 4 ++--
2 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/doc/guides/n
rte_eth_dev created by primary process were not available in secondary
process, it was not possible to use the primary process local memory
object from a secondary process.
This patch modify the reference of primary rte_eth_dev object, use
local rte_eth_dev secondary process instead.
Signed-off-b
PMD uses Verbs object which were not available in the shared memory.
This patch modify the location where Verbs objects are allocated (from
process memory address space to shared memory address space) and thus
allow a secondary process to use those object by mapping this shared
memory space its ow
Add operations that are safe for secondary processes:
* (x)stats
* device info get
* rx/tx descriptor status
Signed-off-by: Xueming Li
Acked-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5.c| 16 +++-
drivers/net/mlx5/mlx5.h| 1 +
drivers/net/mlx5/mlx5_ethdev.c | 11
1 - 100 of 192 matches
Mail list logo
