On 01/16/2018 09:37 PM, Stephen Hemminger wrote:
While reviewing drivers, noticed a lot of unnecessary
duplication of code in drivers for handling the eth_dev link status
information. While consolidating this, it also became obvious that
some drivers behave differently for no good reason.
It als
Hi Olivier,
On 1/16/2018 7:21 PM, Olivier Matz wrote:
On Fri, Dec 22, 2017 at 09:48:01PM +0530, Hemant Agrawal wrote:
On 12/22/2017 7:29 PM, Olivier MATZ wrote:
On Wed, Dec 20, 2017 at 05:29:59PM +0530, Hemant Agrawal wrote:
On 12/19/2017 6:38 PM, Hemant Agrawal wrote:
That's true, I comme
On 01/16/2018 09:37 PM, Stephen Hemminger wrote:
Many drivers are all doing copy/paste of the same code to atomically
update the link status. Reduce duplication, and allow for future
changes by having common function for this.
Signed-off-by: Stephen Hemminger
---
lib/librte_ether/rte_ethdev.c
Hi, ALL:
1. for api "rte_vhost_dequeue_burst", is it safely called from muitiple
thread/process on 17.11?, just as single-produce multiple consume
2. for api "rte_vhost_enqueue_burst", is it safely called from muitiple
thread/process on 17.11? just as multi-produce single consume
3. I find t
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Junjie Chen
> Sent: Wednesday, January 17, 2018 6:42 PM
> To: y...@fridaylinux.org; maxime.coque...@redhat.com
> Cc: dev@dpdk.org; Chen, Junjie J
> Subject: [dpdk-dev] [PATCH] vhost: dequeue zero copy should resto
Tuesday, January 16, 2018 1:53 PM, Rafal Kozik:
> Ethdev Rx offloads API has changed since:
>
> commit ce17eddefc20 ("ethdev: introduce Rx queue offloads API")
>
> This commit support the new Rx offloads API.
>
> Signed-off-by: Rafal Kozik
> ---
> drivers/net/ena/ena_ethdev.c | 36
> ++
Tuesday, January 16, 2018 1:53 PM, Rafal Kozik:
> Subject: [dpdk-dev] [PATCH 1/2] net/ena: convert to new Tx offloads API
>
> Ethdev Tx offloads API has changed since:
>
> commit cba7f53b717d ("ethdev: introduce Tx queue offloads API")
>
> This commit support the new Tx offloads API. Queue confi
Hi Akhil,
> -Original Message-
> From: Akhil Goyal [mailto:akhil.go...@nxp.com]
> Sent: Tuesday, January 16, 2018 6:32 PM
> To: Gujjar, Abhinandan S ; Doherty, Declan
> ; Jacob, Jerin
>
> Cc: dev@dpdk.org; Vangati, Narender ; Rao,
> Nikhil
> Subject: Re: [PATCH 1/2] lib/cryptodev: add su
Cc: John, Marko
From: Bhagavatula, Pavan
Sent: 16 January 2018 23:16
To: Jacob, Jerin; Shukla, Santosh; harry.van.haa...@intel.com;
gage.e...@intel.com; hemant.agra...@nxp.com; nipun.gu...@nxp.com;
liang.j...@intel.com
Cc: dev@dpdk.org; Bhagavatula, Pavan
Subjec
On 1/16/2018 8:22 PM, Olivier Matz wrote:
On Fri, Dec 22, 2017 at 05:00:11PM +0530, Hemant Agrawal wrote:
Introduce a new helper for pktmbuf pool, which will allow
the application to optionally specify the mempool ops name
as well.
Signed-off-by: Hemant Agrawal
---
Hi Hemant,
It looks this
A new API qbman_swp_dqrr_idx_consume is defined which takes
input as DQRR index to consume corresponding DQRR entry.
Signed-off-by: Nipun Gupta
Acked-by: Hemant Agrawal
---
drivers/bus/fslmc/qbman/include/fsl_qbman_portal.h | 7 +++
drivers/bus/fslmc/qbman/qbman_portal.c | 7 +++
Signed-off-by: Nipun Gupta
Acked-by: Hemant Agrawal
---
drivers/net/dpaa2/Makefile | 1 +
drivers/net/dpaa2/dpaa2_ethdev.c | 7 +++
drivers/net/dpaa2/dpaa2_ethdev.h | 5 +
drivers/net/dpaa2/dpaa2_rxtx.c | 43 +---
4 files changed, 53 inserti
With the current QBMAN multi-tx API, we need to create separate
enqueue descriptors for each of the packet which is required to
be enqueued to the hardware, once we support Atomic Queues
(with DCA) in dpaa2 drivers. Creating enqueue descriptor for
each packet is costly and have significant performa
This patch provides cleaner approach to store the DQRR entries,
which are yet to be consumed in case of atomic queues.
Also, this patch changes the storage of the DQRR entry index
into the mbuf->seqn instead of ev->opaque
Signed-off-by: Nipun Gupta
Acked-by: Hemant Agrawal
---
drivers/bus/fslm
Signed-off-by: Nipun Gupta
Acked-by: Hemant Agrawal
---
drivers/event/dpaa2/dpaa2_eventdev.c | 83 ---
drivers/event/dpaa2/dpaa2_eventdev.h | 12 -
drivers/event/dpaa2/dpaa2_eventdev_logs.h | 37 ++
drivers/event/dpaa2/dpaa2_hw_dpcon.c |
Signed-off-by: Nipun Gupta
Acked-by: Hemant Agrawal
---
drivers/event/dpaa2/dpaa2_eventdev.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/event/dpaa2/dpaa2_eventdev.c
b/drivers/event/dpaa2/dpaa2_eventdev.c
index 542bccc..65c2c7a 100644
--- a/drivers/event/dpaa2/
These patches assume that DPAA2 series is already applied [1].
This patch series supports atomic queues in DPAA2 Ethernet driver
with eventdev. Also other changes such as dynamic logging is
supported with these changes.
Patch 1 - support dynamic logging in eventdev
Patch 2,3,4 - Cleanups and perf
On Wednesday 17 January 2018 12:07 AM, Stephen Hemminger wrote:
While reviewing drivers, noticed a lot of unnecessary
duplication of code in drivers for handling the eth_dev link status
information. While consolidating this, it also became obvious that
some drivers behave differently for no good
To support C11 memory model barrier, 2 options are suggested by Jerin:
1. use rte_smp_rmb
2. use load_acquire/store_release(refer to [1]).
CONFIG_RTE_RING_USE_C11_MEM_MODEL is provided, and by default it is "n"
on any architectures so far.
The reason why providing 2 options is due to the performan
Move the common part of rte_ring.h into rte_ring_generic.h.
Move the memory barrier part into update_tail().
No functional changes here.
Signed-off-by: Jia He
Suggested-by: Jerin Jacob
Suggested-by: Ananyev Konstantin
Acked-by: Jerin Jacob
Acked-by: Olivier Matz
---
lib/librte_eventdev/rte_
for the code as follows:
if (condition)
rte_smp_rmb();
else
rte_smp_wmb();
Without this patch, compiler will report this error:
error: 'else' without a previous 'if'
Fixes: 84733fd0d75e ("eal/arm64: fix memory barrier definition")
Signed-off-by: Jia He
Acked-by: Jerin Jacob
---
To support C11 memory model barrier, 2 options are suggested by Jerin:
1. use rte_smp_rmb
2. use load_acquire/store_release
CONFIG_RTE_RING_USE_C11_MEM_MODEL is provided, and by default it is "n"
on any architectures so far.
The reason why providing 2 options is due to the performance benchmark
di
Hi Jerin,
Please do not apply this version on dpdk-eventdev-next as there is some
compilation issue
when both the DPAA series for eventdev support and this series are applied
(though
they are compileable independently).
I will fix this and send a new version by today EOD.
Thanks,
Nipun
> ---
Yes done :)
> -Original Message-
> From: Kovacevic, Marko [mailto:marko.kovace...@intel.com]
> Sent: Tuesday, January 16, 2018 21:03
> To: Nipun Gupta ; jerin.ja...@caviumnetworks.com;
> dev@dpdk.org
> Cc: Sunil Kumar Kori ; Hemant Agrawal
>
> Subject: RE: [dpdk-dev] [PATCH 12/12 v3] doc:
dequeue zero copy change buf_addr and buf_iova of mbuf, and return
to mbuf pool without restore them, it breaks vm memory if others allocate
mbuf from same pool since mbuf reset doesn't reset buf_addr and buf_iova.
Signed-off-by: Junjie Chen
---
lib/librte_vhost/virtio_net.c | 21 +++
On Mon, Jan 15, 2018 at 4:57 PM, Tonghao Zhang wrote:
> On Mon, Jan 15, 2018 at 12:57 PM, Dai, Wei wrote:
>>> -Original Message-
>>> From: xiangxia.m@gmail.com [mailto:xiangxia.m@gmail.com]
>>> Sent: Sunday, January 14, 2018 6:04 PM
>>> To: Dai, Wei ; Xing, Beilei
>>> Cc: dev@dpd
> -Original Message-
> From: Chen, Junjie J
> Sent: Tuesday, January 16, 2018 3:39 PM
> To: Yang, Zhiyong ; y...@fridaylinux.org;
> maxime.coque...@redhat.com
> Cc: dev@dpdk.org
> Subject: RE: [dpdk-dev] [PATCH] vhost: do deep copy while reallocate vq
>
> Hi
> > > > > @@ -227,6 +227,7 @@
> On Jan 9, 2018, at 6:11 AM, Xueming Li wrote:
>
> This patch introduce new DEV_TX_OFFLOAD_GENERIC_TNL_CKSUM_TSO flag for
> devices that support tunnel agnostic TX checksum and tso offloading.
>
> Checksum offset and TSO header length are calculated based on mbuf
> inner length l*_len, outer_l
Hi Matan,
> -Original Message-
> From: Matan Azrad [mailto:ma...@mellanox.com]
> Sent: Tuesday, January 16, 2018 4:16 PM
> To: Lu, Wenzhuo ; Thomas Monjalon
> ; Gaetan Rivet ; Wu,
> Jingjing
> Cc: dev@dpdk.org; Neil Horman ; Richardson,
> Bruce ; Ananyev, Konstantin
>
> Subject: RE: [dpd
17/01/2018 00:46, Gaëtan Rivet:
> On Wed, Jan 17, 2018 at 12:22:43AM +0100, Thomas Monjalon wrote:
> > 17/01/2018 00:19, Gaëtan Rivet:
> > > It might be a nitpick, but the driver specific properties might not
> > > follow the key/value pair syntax. At least for the fail-safe, a custom
> > > parsing
12/01/2018 18:24, Burakov, Anatoly:
> On 29-Dec-17 7:58 AM, Pavan Nikhilesh wrote:
> > Make max vfio groups compile-time configurable so that platforms can
> > choose vfio group limit.
> >
> > Signed-off-by: Pavan Nikhilesh
>
> Acked-by: Anatoly Burakov
Series applied, thanks
On Wed, Jan 17, 2018 at 12:22:43AM +0100, Thomas Monjalon wrote:
> 17/01/2018 00:19, Gaëtan Rivet:
> > Hi Yuanhan,
> >
> > On Tue, Jan 16, 2018 at 10:50:18PM +0800, Yuanhan Liu wrote:
> > > +The ``devargs`` can be used for whitelisting/blacklisting devices,
> > > identifying
> > > +DPDK ports and
15/01/2018 06:11, Hemant Agrawal:
> other vfio based module e.g. fslmc will also need to use
> the clear_group call.
> So, exposing it and renaming it to *rte_vfio_clear_group*
>
> Signed-off-by: Hemant Agrawal
> Acked-by: Anatoly Burakov
Series applied, thanks
16/01/2018 20:43, Ferruh Yigit:
> On 1/1/2018 10:00 PM, Xiao Wang wrote:
> > In some case, one device are accessed by different processes via
> > different BARs, so one uio device may be opened by more than one
> > process, for this case we just need to enable interrupt once, and
> > pci_clear_mast
17/01/2018 00:19, Gaëtan Rivet:
> Hi Yuanhan,
>
> On Tue, Jan 16, 2018 at 10:50:18PM +0800, Yuanhan Liu wrote:
> > +The ``devargs`` can be used for whitelisting/blacklisting devices,
> > identifying
> > +DPDK ports and attaching/deatching devices. They all share the same syntax.
> > +
> > +It is
21/12/2017 13:59, Adrien Mazarguil:
> Adrien Mazarguil (6):
> devtools: update check-includes exceptions
> net/i40e: fix issue in exported header
> flow_classify: fix issue in exported header
> member: fix issue in exported header
> fix missing includes in exported headers
> net: fix rt
Hi Yuanhan,
On Tue, Jan 16, 2018 at 10:50:18PM +0800, Yuanhan Liu wrote:
> This patch documents the new devargs syntax, which is going to be
> implemented in DPDK v18.05.
>
> The new devargs proposal is introduced for having a consistent
> interface for:
>
> - whitelisting/blacklisting devices
>
22/12/2017 15:25, Adrien Mazarguil:
> Such a change is unlikely to be accepted for 18.02 in any case, therefore if
> the proposed workaround is deemed too risky, I offer to remove this patch
> from the series. What do you suggest?
Series applied without this patch.
Please let's fix it in 18.05.
Hi Stephen,
On Tue, Jan 16, 2018 at 10:29:06AM -0800, Stephen Hemminger wrote:
> While doing link status refactoring, I noticed that the eth_dev_link_update
> function
> return value is unused by common code (in rte_eth_dev_link_get); but is being
> expected
> by failsafe driver.
>
> The return
Hi Matan,
On Tue, Jan 16, 2018 at 05:20:27PM +, Matan Azrad wrote:
> Hi Gaetan
>
> > > > In 18.05, or 18.08 there should be an EAL function that would be
> > > > able to identify a device given a specific ID string (very close to an
> > rte_devargs).
> > > > Currently, this API does not ex
10/01/2018 03:01, Beilei Xing:
> --- a/doc/guides/rel_notes/release_18_02.rst
> +++ b/doc/guides/rel_notes/release_18_02.rst
> +* **Updated i40e driver to support MPLSoUDP/MPLSoGRE.**
> +
> + Updated i40e PMD to support MPLSoUDP/MPLSoGRE with MPLSoUDP/MPLSoGRE
> + supporting profiles which can be
> > > This patch add inforation about i40e queue region realted to release
> > > notes, it has been missed before in v17.11 release notes. This feature
> > > has been implemented in v17.11.
> > >
> > > Signed-off-by: Wei Zhao
> > Acked-by: Helin Zhang
> Applied to dpdk-next-net-intel, with minor
Hi Konstantin
From: Ananyev, Konstantin, Tuesday, January 16, 2018 9:11 PM
> Hi Matan,
>
> >
> > Hi Konstantin
> > From: Ananyev, Konstantin, Monday, January 15, 2018 8:44 PM
> > > Hi Matan,
> > > > Hi Konstantin
> > > > From: Ananyev, Konstantin, Monday, January 15, 2018 1:45 PM
> > > > > Hi M
On 11/30/2017 7:35 AM, Yuanhan Liu wrote:
> The ethdev name is taken from the "name" parameter from the helper
> function rte_eth_dev_allocate(name). The name is given by the caller,
> thus, there is no way to guarantee all the callers will follow the
> same syntax, leaving us the port name is not
On 1/1/2018 10:00 PM, Xiao Wang wrote:
> In some case, one device are accessed by different processes via
> different BARs, so one uio device may be opened by more than one
> process, for this case we just need to enable interrupt once, and
> pci_clear_master only when the last process closed.
>
>
On 1/16/2018 7:06 PM, so...@rbbn.com wrote:
> From: Souvik Dey
>
> We need to add the pci_dev info to the rte_eth_dev structure
> during the eth_ena_dev_init. Informantions like driver_name
> and numa_node will not be populated otherwise.
stacktrace is like:
eth_ena_pci_probe
rte_eth_dev_pci_
Hi Matan,
>
> Hi Konstantin
> From: Ananyev, Konstantin, Monday, January 15, 2018 8:44 PM
> > Hi Matan,
> > > Hi Konstantin
> > > From: Ananyev, Konstantin, Monday, January 15, 2018 1:45 PM
> > > > Hi Matan,
> > > > > Hi Konstantin
> > > > > From: Ananyev, Konstantin, Friday, January 12, 2018 2:
From: Souvik Dey
We need to add the pci_dev info to the rte_eth_dev structure
during the eth_ena_dev_init. Informantions like driver_name
and numa_node will not be populated otherwise.
Signed-off-by: Souvik Dey
---
drivers/net/ena/ena_ethdev.c | 2 ++
1 file changed, 2 insertions(+)
diff --
Hi Oliver, Xueming,
Tuesday, January 16, 2018 7:29 PM, Xueming(Steven) Li:
> > Hi Xueming,
> >
> > On Tue, Jan 09, 2018 at 10:11:08PM +0800, Xueming Li wrote:
> > > */
> > > #define DEV_TX_OFFLOAD_SECURITY 0x0002
> > > +/**< Device supports arbitrary tunnel chksum and tso offloading
Add note about PMD expects the network interfaces provided to be up,
documented behavior to set expectations right.
Also added minor fix.
Signed-off-by: Ferruh Yigit
---
Cc: Vipin Varghese
Cc: Michael Glynn
---
doc/guides/nics/pcap_ring.rst | 8 +++-
1 file changed, 7 insertions(+), 1 del
Common function matches this drivers usage.
Signed-off-by: Stephen Hemminger
---
drivers/net/octeontx/octeontx_ethdev.c | 17 ++---
1 file changed, 2 insertions(+), 15 deletions(-)
diff --git a/drivers/net/octeontx/octeontx_ethdev.c
b/drivers/net/octeontx/octeontx_ethdev.c
index 4f
This driver was not doing atomic update of link status information.
And the return value was different than others.
The hardware also does not do autonegotiation (at least on Linux).
Signed-off-by: Stephen Hemminger
---
drivers/net/enic/enic_ethdev.c | 5 ++---
drivers/net/enic/enic_main.c |
Yet another driver which was not returing correct value on
link change.
Since this driver can't be built on x86 could not even
do a compile test.
Signed-off-by: Stephen Hemminger
---
drivers/net/szedata2/rte_eth_szedata2.c | 19 ++-
1 file changed, 6 insertions(+), 13 deletions(
Use new helper function.
Signed-off-by: Stephen Hemminger
---
drivers/net/thunderx/nicvf_ethdev.c | 18 +++---
1 file changed, 3 insertions(+), 15 deletions(-)
diff --git a/drivers/net/thunderx/nicvf_ethdev.c
b/drivers/net/thunderx/nicvf_ethdev.c
index 72dc8ae24ba9..94a2a9b7683d 10
Use the new link update API, and cleanup the logic in the the
link update routine.
Tested-by: Shijith Thotton
Signed-off-by: Stephen Hemminger
---
drivers/net/liquidio/lio_ethdev.c | 53 ++-
1 file changed, 8 insertions(+), 45 deletions(-)
diff --git a/drive
Use new rte_linkstatus update API
Signed-off-by: Stephen Hemminger
---
drivers/net/i40e/i40e_ethdev.c| 43 +--
drivers/net/i40e/i40e_ethdev_vf.c | 18 ++--
2 files changed, 7 insertions(+), 54 deletions(-)
diff --git a/drivers/net/i40e/i40e_et
Use the new API (_rte_eth_linkstatus_set) to handle link status update.
Signed-off-by: Stephen Hemminger
---
drivers/net/sfc/sfc_ethdev.c | 21 +++--
drivers/net/sfc/sfc_ev.c | 20 ++--
2 files changed, 5 insertions(+), 36 deletions(-)
diff --git a/drivers/ne
Use the new helper functions from eth_dev for
handling atomic link_info update.
Signed-off-by: Stephen Hemminger
---
drivers/net/ixgbe/ixgbe_ethdev.c | 96 +++-
1 file changed, 17 insertions(+), 79 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/
Use new rte_eth_linkstatus_get/set API.
Signed-off-by: Stephen Hemminger
---
drivers/net/e1000/em_ethdev.c | 69 +++--
drivers/net/e1000/igb_ethdev.c | 70 +++---
2 files changed, 10 insertions(+), 129 deletions(-)
diff --
Use new rte_eth_linkstatus_get/set helper function.
Signed-off-by: Stephen Hemminger
---
drivers/net/nfp/nfp_net.c | 77 +--
1 file changed, 7 insertions(+), 70 deletions(-)
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index 0501
Use new helper function to update the link status.
Signed-off-by: Stephen Hemminger
---
drivers/net/dpaa2/dpaa2_ethdev.c | 75
1 file changed, 7 insertions(+), 68 deletions(-)
diff --git a/drivers/net/dpaa2/dpaa2_ethdev.c b/drivers/net/dpaa2/dpaa2_ethdev
Use new rte_eth_link_update helper.
Also remove no longer necessary include of rte_atomic.h
Signed-off-by: Stephen Hemminger
---
drivers/net/vmxnet3/vmxnet3_ethdev.c | 86 +++-
1 file changed, 16 insertions(+), 70 deletions(-)
diff --git a/drivers/net/vmxnet3/vmx
Use the new comon code in ethdev to handle link status update.
Signed-off-by: Stephen Hemminger
---
drivers/net/virtio/virtio_ethdev.c | 65 +++---
1 file changed, 12 insertions(+), 53 deletions(-)
diff --git a/drivers/net/virtio/virtio_ethdev.c
b/drivers/net/vi
To handle atomic update of link status (64 bit), every driver
was doing its own version using cmpset.
Atomic exchange is a useful primitive in its own right;
therefore make it a EAL routine.
Signed-off-by: Stephen Hemminger
---
.../common/include/arch/x86/rte_atomic.h | 24 +++
...
Many drivers are all doing copy/paste of the same code to atomically
update the link status. Reduce duplication, and allow for future
changes by having common function for this.
Signed-off-by: Stephen Hemminger
---
lib/librte_ether/rte_ethdev.c | 22 +++
lib/librte_ether/rte_ethdev.h
While reviewing drivers, noticed a lot of unnecessary
duplication of code in drivers for handling the eth_dev link status
information. While consolidating this, it also became obvious that
some drivers behave differently for no good reason.
It also was a good chance to introduce atomic exchange pr
While doing link status refactoring, I noticed that the eth_dev_link_update
function
return value is unused by common code (in rte_eth_dev_link_get); but is being
expected
by failsafe driver.
The return values are not consistent across drivers, and link update function
is only
valid if device i
Use new helper function.
Signed-off-by: Stephen Hemminger
---
drivers/net/thunderx/nicvf_ethdev.c | 18 +++---
1 file changed, 3 insertions(+), 15 deletions(-)
diff --git a/drivers/net/thunderx/nicvf_ethdev.c
b/drivers/net/thunderx/nicvf_ethdev.c
index 72dc8ae24ba9..94a2a9b7683d 10
16/01/2018 18:38, Burakov, Anatoly:
> On 16-Jan-18 5:34 PM, Thomas Monjalon wrote:
> > 16/01/2018 16:05, Burakov, Anatoly:
> >> On 16-Jan-18 12:20 PM, Thomas Monjalon wrote:
> >>> 16/01/2018 12:56, Burakov, Anatoly:
> On 12-Jan-18 11:50 AM, Thomas Monjalon wrote:
> > 12/01/2018 12:44, Bura
This driver was not doing atomic update of link status information.
And the return value was different than others.
The hardware also does not do autonegotiation (at least on Linux).
Signed-off-by: Stephen Hemminger
---
drivers/net/enic/enic_ethdev.c | 5 ++---
drivers/net/enic/enic_main.c |
Yet another driver which was not returing correct value on
link change.
Since this driver can't be built on x86 could not even
do a compile test.
Signed-off-by: Stephen Hemminger
---
drivers/net/szedata2/rte_eth_szedata2.c | 19 ++-
1 file changed, 6 insertions(+), 13 deletions(
Common function matches this drivers usage.
Signed-off-by: Stephen Hemminger
---
drivers/net/octeontx/octeontx_ethdev.c | 17 ++---
1 file changed, 2 insertions(+), 15 deletions(-)
diff --git a/drivers/net/octeontx/octeontx_ethdev.c
b/drivers/net/octeontx/octeontx_ethdev.c
index 4f
Use the new API (_rte_eth_linkstatus_set) to handle link status update.
Signed-off-by: Stephen Hemminger
---
drivers/net/sfc/sfc_ethdev.c | 21 +++--
drivers/net/sfc/sfc_ev.c | 20 ++--
2 files changed, 5 insertions(+), 36 deletions(-)
diff --git a/drivers/ne
Use new rte_linkstatus update API
Signed-off-by: Stephen Hemminger
---
drivers/net/i40e/i40e_ethdev.c| 43 +--
drivers/net/i40e/i40e_ethdev_vf.c | 18 ++--
2 files changed, 7 insertions(+), 54 deletions(-)
diff --git a/drivers/net/i40e/i40e_et
Use the new link update API, and cleanup the logic in the the
link update routine.
Tested-by: Shijith Thotton
Signed-off-by: Stephen Hemminger
---
drivers/net/liquidio/lio_ethdev.c | 53 ++-
1 file changed, 8 insertions(+), 45 deletions(-)
diff --git a/drive
Use the new helper functions from eth_dev for
handling atomic link_info update.
Signed-off-by: Stephen Hemminger
---
drivers/net/ixgbe/ixgbe_ethdev.c | 96 +++-
1 file changed, 17 insertions(+), 79 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/
Use new rte_eth_linkstatus_get/set API.
Signed-off-by: Stephen Hemminger
---
drivers/net/e1000/em_ethdev.c | 71 --
drivers/net/e1000/igb_ethdev.c | 70 +++--
2 files changed, 11 insertions(+), 130 deletions(-)
diff --
Use new rte_eth_linkstatus_get/set helper function.
Signed-off-by: Stephen Hemminger
---
drivers/net/nfp/nfp_net.c | 77 +--
1 file changed, 7 insertions(+), 70 deletions(-)
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index 0501
Use new helper function to update the link status.
Signed-off-by: Stephen Hemminger
---
drivers/net/dpaa2/dpaa2_ethdev.c | 75
1 file changed, 7 insertions(+), 68 deletions(-)
diff --git a/drivers/net/dpaa2/dpaa2_ethdev.c b/drivers/net/dpaa2/dpaa2_ethdev
Use new rte_eth_link_update helper.
Also remove no longer necessary include of rte_atomic.h
Signed-off-by: Stephen Hemminger
---
drivers/net/vmxnet3/vmxnet3_ethdev.c | 86 +++-
1 file changed, 16 insertions(+), 70 deletions(-)
diff --git a/drivers/net/vmxnet3/vmx
Use the new comon code in ethdev to handle link status update.
Signed-off-by: Stephen Hemminger
---
drivers/net/virtio/virtio_ethdev.c | 67 --
1 file changed, 13 insertions(+), 54 deletions(-)
diff --git a/drivers/net/virtio/virtio_ethdev.c
b/drivers/net/vi
Many drivers are all doing copy/paste of the same code to atomically
update the link status. Reduce duplication, and allow for future
changes by having common function for this.
Signed-off-by: Stephen Hemminger
---
lib/librte_ether/rte_ethdev.c | 22 +++
lib/librte_ether/rte_ethdev.h
To handle atomic update of link status (64 bit), every driver
was doing its own version using cmpset.
Atomic exchange is a useful primitive in its own right;
therefore make it a EAL routine.
Signed-off-by: Stephen Hemminger
---
.../common/include/arch/x86/rte_atomic.h | 24 +++
...
While reviewing drivers, noticed a lot of unnecessary
duplication of code in drivers for handling the eth_dev link status
information. While consolidating this, it also became obvious that
some drivers behave differently for no good reason.
It also was a good chance to introduce atomic exchange pr
Recently, some additional patches were added to allow for programmatic
marking of C symbols as experimental. The addition of these markers is
dependent on the manual addition of exported symbols to the EXPERIMENTAL
section of the corresponding libraries version map file. The consensus
on review i
Signed-off-by: Pavan Nikhilesh
---
This patch depends on http://dpdk.org/dev/patchwork/patch/33429 patchset.
doc/guides/rel_notes/release_18_02.rst | 7 +++
1 file changed, 7 insertions(+)
diff --git a/doc/guides/rel_notes/release_18_02.rst
b/doc/guides/rel_notes/release_18_02.rst
index
Hi Kirill,
On Tue, Jan 16, 2018 at 05:23:05PM +, Rybalchenko, Kirill wrote:
> Hi Adrien,
> after some discussion we found that change I've done
> in Mellanox PMD is not really necessary: size of array
> flow_types_mask[] is still 1 and the loop in patch
>
> for (i = 0; i < RTE_FLOW_MASK_AR
On Tue, Jan 16, 2018 at 03:54:57PM +0100, george@gmail.com wrote:
> Hi Adrien,
>
> Thanks for your insights and sorry for omitting the cover letter (this is
> my first patch in DPDK).
No problem, don't worry about that. Remember to put as much context as close
as possible to the related chang
On 1/9/2018 11:47 AM, De Lara Guarch, Pablo wrote:
>
>
>> -Original Message-
>> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Tomasz Duszynski
>> Sent: Tuesday, January 9, 2018 9:09 AM
>> To: dev@dpdk.org
>> Cc: m...@semihalf.com; d...@marvell.com; nsams...@marvell.com;
>> jianbo..
There will be a new function added in v18.05 that will return
number of detected sockets, which will change the ABI.
Signed-off-by: Anatoly Burakov
---
doc/guides/rel_notes/deprecation.rst | 2 ++
1 file changed, 2 insertions(+)
diff --git a/doc/guides/rel_notes/deprecation.rst
b/doc/guides/re
On Tue, Jan 16, 2018 at 04:04:32PM +, Mcnamara, John wrote:
>
>
> > -Original Message-
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Pavan Nikhilesh
> > Sent: Tuesday, January 16, 2018 3:40 PM
> > To: jerin.ja...@caviumnetworks.com; santosh.shu...@caviumnetworks.com; Van
> >
Signed-off-by: Pavan Nikhilesh
---
doc/guides/rel_notes/release_18_02.rst | 10 ++
1 file changed, 10 insertions(+)
diff --git a/doc/guides/rel_notes/release_18_02.rst
b/doc/guides/rel_notes/release_18_02.rst
index b2dc39cfe..2a243cb47 100644
--- a/doc/guides/rel_notes/release_18_02.rst
Signed-off-by: Pavan Nikhilesh
Acked-by: Harry van Haaren
---
app/test-eventdev/test_pipeline_atq.c | 280 +-
1 file changed, 279 insertions(+), 1 deletion(-)
diff --git a/app/test-eventdev/test_pipeline_atq.c
b/app/test-eventdev/test_pipeline_atq.c
index 6c9ac6
This is a pipeline test case that aims at testing the following with
``all types queue`` eventdev scheme.
1. Measure the end-to-end performance of an event dev with a ethernet dev.
2. Maintain packet ordering from Rx to Tx.
The atq queue test functions as same as ``pipeline_queue`` test.
The diffe
Signed-off-by: Pavan Nikhilesh
Acked-by: Harry van Haaren
---
app/test-eventdev/test_pipeline_common.h | 80 +
app/test-eventdev/test_pipeline_queue.c | 288 ++-
2 files changed, 367 insertions(+), 1 deletion(-)
diff --git a/app/test-eventdev/test_pipeline_
This is a pipeline queue test case that aims at testing the following:
1. Measure the end-to-end performance of an event dev with a ethernet dev.
2. Maintain packet ordering from Rx to Tx.
The pipeline queue test configures the eventdev with Q queues and P ports,
where Q is (nb_ethdev * nb_stages)
Setup one port event port for Tx and link the respective event queue.
Register the Tx function as a service to be called from a service core.
The Tx function dequeues the events from the event queue and transmits
the packet to its respective ethernet port.
Signed-off-by: Pavan Nikhilesh
Acked-by:
The event master lcore's test termination and the logic to print the mpps
are common for the queue and all types queue test.
Move them as the common function.
Signed-off-by: Pavan Nikhilesh
Acked-by: Harry van Haaren
---
app/test-eventdev/test_pipeline_common.c | 67 +++
Setup one port per worker and link to all queues and setup producer port
based on Rx adapter capabilities.
Signed-off-by: Pavan Nikhilesh
Acked-by: Harry van Haaren
---
app/test-eventdev/test_pipeline_common.c | 114 +++
app/test-eventdev/test_pipeline_common.h | 5
Add common ethdev port setup and destroy along with event dev destroy.
Signed-off-by: Pavan Nikhilesh
Acked-by: Harry van Haaren
---
app/test-eventdev/test_pipeline_common.c | 91
app/test-eventdev/test_pipeline_common.h | 3 ++
2 files changed, 94 insertions(+
1 - 100 of 314 matches
Mail list logo
