Monday, May 15, 2017 5:26 PM, Thomas Monjalon:
> 15/05/2017 16:12, Richardson, Bruce:
> > From: Yigit, Ferruh
> > > On 5/15/2017 2:15 PM, Bruce Richardson wrote:
> > > > On Mon, May 15, 2017 at 02:35:55PM +0200, Thomas Monjalon wrote:
> > > >> Hi,
> > > >>
> > > >> I would like to open a discussion
Hi, Adrien Mazarguil
> -Original Message-
> From: Adrien Mazarguil [mailto:adrien.mazarg...@6wind.com]
> Sent: Friday, May 12, 2017 5:13 PM
> To: Zhao1, Wei
> Cc: dev@dpdk.org; Xing, Beilei ; Lu, Wenzhuo
>
> Subject: Re: [dpdk-dev] [PATCH v5 19/26] app/testpmd: add item raw to flow
> co
Hi, Srini
There is a bit confusion. Your patch shows that your code is added into the
function eth_ixgbevf_pci_remove( ).
But it is not. It is added into the fucntion ixgbe_dev_start( ), right ?
So would you please rebase it to R 17.05 ?
Which type of ixgbe device id did you tested ?
There are
-Original Message-
> Date: Mon, 15 May 2017 14:40:14 -0500
> From: Gage Eads
> To: dev@dpdk.org
> CC: jerin.ja...@caviumnetworks.com
> Subject: [PATCH v2] eventdev: clarify atomic and ordered queue config
> X-Mailer: git-send-email 2.7.4
>
> The nb_atomic_flows and nb_atomic_order_sequenc
Hi,
> -Original Message-
> From: Chen, Jing D
> Sent: Friday, May 5, 2017 2:29 PM
> To: Yang, Qiming ; dev@dpdk.org
> Cc: Wu, Jingjing
> Subject: RE: [PATCH v2] test: add delay time in test alarm
>
> Hi,
>
> > diff --git a/test/test/test_alarm.c b/test/test/test_alarm.c index
> > ecb2f6
> 15/05/2017 16:12, Richardson, Bruce:
> > From: Yigit, Ferruh
> > > On 5/15/2017 2:15 PM, Bruce Richardson wrote:
> > > > On Mon, May 15, 2017 at 02:35:55PM +0200, Thomas Monjalon wrote:
> > > >> Hi,
> > > >>
> > > >> I would like to open a discussion about SIMD code in drivers.
> > > >>
> > > >>
The nb_atomic_flows and nb_atomic_order_sequences fields are only inspected
if the queue is configured for atomic or ordered scheduling, respectively.
This commit updates the documentation to reflect that.
Signed-off-by: Gage Eads
---
v2: Fixed doxygen output issue and tweaked the ranges
lib/li
On 5/15/2017 1:37 PM, Hemant Agrawal wrote:
> Signed-off-by: Hemant Agrawal
> ---
> drivers/bus/fslmc/rte_bus_fslmc_version.map | 8
> 1 file changed, 8 insertions(+)
>
> diff --git a/drivers/bus/fslmc/rte_bus_fslmc_version.map
> b/drivers/bus/fslmc/rte_bus_fslmc_version.map
> index 2d
On 5/15/2017 1:37 PM, Hemant Agrawal wrote:
> Signed-off-by: Hemant Agrawal
Same comment here, can you please mention in the commit log why this
piece of code removed, what will be effect of the removal.
Thanks,
ferruh
On 5/15/2017 1:37 PM, Hemant Agrawal wrote:
> Signed-off-by: Hemant Agrawal
> ---
> config/defconfig_arm64-dpaa2-linuxapp-gcc | 2 +-
> mk/machine/dpaa2/rte.vars.mk | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/config/defconfig_arm64-dpaa2-linuxapp-gcc
Document NIC features, add more information about them and add more
implementation related support.
TODO:
- Not all features all fully documented, need help from community
- Instead of having a new file, it would be nice to auto generate this
file from defaults.ini, and perhaps add extra informat
15/05/2017 16:12, Richardson, Bruce:
> From: Yigit, Ferruh
> > On 5/15/2017 2:15 PM, Bruce Richardson wrote:
> > > On Mon, May 15, 2017 at 02:35:55PM +0200, Thomas Monjalon wrote:
> > >> Hi,
> > >>
> > >> I would like to open a discussion about SIMD code in drivers.
> > >>
> > >> I think we should
> -Original Message-
> From: Yigit, Ferruh
> Sent: Monday, May 15, 2017 2:36 PM
> To: Richardson, Bruce ; Thomas Monjalon
>
> Cc: dev@dpdk.org; Yuanhan Liu ; Maxime
> Coquelin ; Chen, Jing D
> ; Zhang, Helin ; Wu,
> Jingjing ; Lu, Wenzhuo ;
> Ananyev, Konstantin
> Subject: Re: SIMD Rx/T
On 5/15/2017 2:15 PM, Bruce Richardson wrote:
> On Mon, May 15, 2017 at 02:35:55PM +0200, Thomas Monjalon wrote:
>> Hi,
>>
>> I would like to open a discussion about SIMD code in drivers.
>>
>> I think we should not have different behaviours or features capabilities,
>> in the different code paths
On Mon, May 15, 2017 at 02:35:55PM +0200, Thomas Monjalon wrote:
> Hi,
>
> I would like to open a discussion about SIMD code in drivers.
>
> I think we should not have different behaviours or features capabilities,
> in the different code paths of a same driver.
> I suggest to consider such diffe
Instead of many PMD define their own macro, define a generic one in
ethdev and use that in PMDs.
Signed-off-by: Ferruh Yigit
Acked-by: Allain Legacy
Acked-by: Wenzhuo Lu
---
v3:
* rebased on top of latest next-net, there were updates in ixgbe and
virtio
v2:
* Convert RTE_DEV_TO_PCI(eth_dev->d
On 05/15/2017 01:24 PM, Ferruh Yigit wrote:
Instead of many PMD define their own macro, define a generic one in
ethdev and use that in PMDs.
Signed-off-by: Ferruh Yigit
Acked-by: Allain Legacy
Acked-by: Wenzhuo Lu
Acked-by: Andrew Rybchenko
---
v3:
* rebased on top of latest next-net, t
Signed-off-by: Hemant Agrawal
---
drivers/net/dpaa2/dpaa2_ethdev.c | 36 +++-
1 file changed, 35 insertions(+), 1 deletion(-)
diff --git a/drivers/net/dpaa2/dpaa2_ethdev.c b/drivers/net/dpaa2/dpaa2_ethdev.c
index 31a2358..d64ec48 100644
--- a/drivers/net/dpaa2/dpa
Signed-off-by: Hemant Agrawal
---
drivers/bus/fslmc/fslmc_vfio.c | 13 +++--
drivers/bus/fslmc/portal/dpaa2_hw_dpbp.c | 2 +-
drivers/bus/fslmc/portal/dpaa2_hw_dpio.c | 11 +--
3 files changed, 5 insertions(+), 21 deletions(-)
diff --git a/drivers/bus/fslmc/fslmc_vfio.
This is required to configure the stashing destination
based on the cluster size.
Signed-off-by: Hemant Agrawal
---
drivers/bus/fslmc/Makefile | 1 +
drivers/bus/fslmc/mc/dpmng.c| 81 +++
drivers/bus/fslmc/mc/fsl_dpmng.h| 99 +
Signed-off-by: Hemant Agrawal
---
drivers/net/dpaa2/dpaa2_ethdev.c | 118 +++
drivers/net/dpaa2/dpaa2_ethdev.h | 2 +-
drivers/net/dpaa2/dpaa2_rxtx.c | 25 +
3 files changed, 144 insertions(+), 1 deletion(-)
diff --git a/drivers/net/dpaa2/dpaa2_et
Signed-off-by: Hemant Agrawal
---
doc/guides/nics/features/dpaa2.ini | 1 +
drivers/net/dpaa2/dpaa2_ethdev.c| 142
drivers/net/dpaa2/mc/dpni.c | 18 +
drivers/net/dpaa2/mc/fsl_dpni.h | 24 ++
drivers/net/dpaa2/mc/fsl_dpni_cmd.h |
Signed-off-by: Hemant Agrawal
---
doc/guides/nics/features/dpaa2.ini | 1 +
drivers/net/dpaa2/dpaa2_ethdev.c | 19 +++
2 files changed, 20 insertions(+)
diff --git a/doc/guides/nics/features/dpaa2.ini
b/doc/guides/nics/features/dpaa2.ini
index 720c35b..48422f0 100644
--- a/do
Signed-off-by: Hemant Agrawal
---
drivers/net/dpaa2/dpaa2_rxtx.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/net/dpaa2/dpaa2_rxtx.c b/drivers/net/dpaa2/dpaa2_rxtx.c
index 0670ae3..d7e18ad 100644
--- a/drivers/net/dpaa2/dpaa2_rxtx.c
+++ b/drivers/net/dpaa2/dpaa2_rxtx.c
@@ -330,6
This patch updates the usages of malloc and TAILQ in
dpbp and dpio objects.
Signed-off-by: Hemant Agrawal
---
drivers/bus/fslmc/portal/dpaa2_hw_dpbp.c | 28 +---
drivers/bus/fslmc/portal/dpaa2_hw_dpio.c | 32
2 files changed, 21 insertions
Signed-off-by: Hemant Agrawal
---
doc/guides/nics/features/dpaa2.ini | 1 +
drivers/net/dpaa2/dpaa2_ethdev.c| 61 +++--
drivers/net/dpaa2/mc/dpni.c | 41 +
drivers/net/dpaa2/mc/fsl_dpni.h | 27
drivers/net/
Signed-off-by: Hemant Agrawal
---
doc/guides/nics/features/dpaa2.ini | 2 ++
drivers/net/dpaa2/dpaa2_ethdev.c| 55 +++
drivers/net/dpaa2/mc/dpni.c | 66 +
drivers/net/dpaa2/mc/fsl_dpni.h | 54 +++
Signed-off-by: Hemant Agrawal
---
doc/guides/nics/features/dpaa2.ini | 1 +
drivers/net/dpaa2/dpaa2_ethdev.c| 81 -
drivers/net/dpaa2/mc/dpni.c | 76 ++
drivers/net/dpaa2/mc/fsl_dpni.h | 45 +
Typically when the PMD issues a RX command to DPAA2 hardware,
the hw writes the available descriptors into the given memory.
The RX function then processes the frames and prepare them as
mbufs.
This patch adds support to issue another pull request to hardware
in another memory location, before we
This will help in limiting the size of queues and avoid
them growing practicaly infinite.
Signed-off-by: Hemant Agrawal
---
drivers/net/dpaa2/dpaa2_ethdev.c| 19 +
drivers/net/dpaa2/dpaa2_ethdev.h| 8
drivers/net/dpaa2/mc/dpni.c | 50 ++
drivers/
Typically when the PMD issues a RX command to DPAA2 hardware,
the hw writes the available descriptors into the given memory.
The RX function then processes the frames and prepare them as
mbufs.
This patch adds support to issue another pull request to hardware
in another memory location, before we
Signed-off-by: Hemant Agrawal
---
drivers/bus/fslmc/portal/dpaa2_hw_pvt.h | 5 +-
drivers/net/dpaa2/dpaa2_ethdev.c| 62 ++-
drivers/net/dpaa2/dpaa2_ethdev.h| 14
drivers/net/dpaa2/dpaa2_rxtx.c | 4 +
drivers/net/dpaa2/mc/dpni.c | 47 +++
Signed-off-by: Hemant Agrawal
---
drivers/bus/fslmc/rte_bus_fslmc_version.map | 8
1 file changed, 8 insertions(+)
diff --git a/drivers/bus/fslmc/rte_bus_fslmc_version.map
b/drivers/bus/fslmc/rte_bus_fslmc_version.map
index 2db0fce..6e7b94d 100644
--- a/drivers/bus/fslmc/rte_bus_fslmc_
DPAA2 hardware support the option to configure
multiple memories for Rx recv buffer (DPAA2-DQRR).
Each Rx request executing is called as 'DQ' request.
This patch adds routines to get information w.r.t each DQ request.
Signed-off-by: Hemant Agrawal
---
drivers/bus/fslmc/portal/dpaa2_hw_dpio.c |
Signed-off-by: Hemant Agrawal
---
drivers/net/dpaa2/dpaa2_ethdev.c | 25 -
1 file changed, 25 deletions(-)
diff --git a/drivers/net/dpaa2/dpaa2_ethdev.c b/drivers/net/dpaa2/dpaa2_ethdev.c
index 4576442..f4c73de 100644
--- a/drivers/net/dpaa2/dpaa2_ethdev.c
+++ b/drivers/n
Signed-off-by: Hemant Agrawal
---
drivers/net/dpaa2/dpaa2_ethdev.c | 69 +++-
1 file changed, 39 insertions(+), 30 deletions(-)
diff --git a/drivers/net/dpaa2/dpaa2_ethdev.c b/drivers/net/dpaa2/dpaa2_ethdev.c
index 007eda7..698fb84 100644
--- a/drivers/net/dpa
This patchset adds following features to DPAA2 PMD:
1. improved error handling
2. parallel RX DQ requests to improve performance
3. Support for various ethdev operations
4. enable jumbo frames
Hemant Agrawal (20):
mk: change to cortex-a72
net/dpaa2: remove port level buffer layout definition
Signed-off-by: Hemant Agrawal
---
config/defconfig_arm64-dpaa2-linuxapp-gcc | 2 +-
mk/machine/dpaa2/rte.vars.mk | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/config/defconfig_arm64-dpaa2-linuxapp-gcc
b/config/defconfig_arm64-dpaa2-linuxapp-gcc
index 314a0ec
The DPAA2 driver is not using the DPAA2 frame descriptor
software annotation area. This patch reduces the
PTA length to zero and adjust the RX Buffer Layout
accordingly.
Signed-off-by: Hemant Agrawal
---
drivers/bus/fslmc/portal/dpaa2_hw_pvt.h | 2 +-
drivers/net/dpaa2/base/dpaa2_hw_dpni.c | 1
Hi,
I would like to open a discussion about SIMD code in drivers.
I think we should not have different behaviours or features capabilities,
in the different code paths of a same driver.
I suggest to consider such differences as exceptions.
So we should merge features files (i.e. matrix columns),
Signed-off-by: Ferruh Yigit
---
Please fix partially support to fully supported speed capabilities.
Please check other features of PMD and document missing ones in features
file.
Thanks.
v3:
* Update tap PMD speed capability to "P"
v2:
* Mark PMDs that return speed capabilities without checki
On 5/15/2017 1:17 PM, Pascal Mazon wrote:
> Hi,
>
> In the case of tap PMD, I would say it's rather partially supported (P).
Hi Pascal,
Thanks, I will update it.
>
> Best regards,
>
> Pascal
>
Hi,
In the case of tap PMD, I would say it's rather partially supported (P).
Best regards,
Pascal
On Mon, 15 May 2017 11:59:46 +0100
Ferruh Yigit wrote:
> Signed-off-by: Ferruh Yigit
> ---
>
> Please fix partially support to fully supported speed capabilities.
>
> Please check other featur
Signed-off-by: Ferruh Yigit
---
Please fix partially support to fully supported speed capabilities.
Please check other features of PMD and document missing ones in features
file.
Thanks.
v2:
* Mark PMDs that return speed capabilities without checking real
hardware capabilities to Partially sup
On 5/15/2017 1:53 AM, Lu, Wenzhuo wrote:
> Hi,
>
>> -Original Message-
>> From: Yigit, Ferruh
>> Sent: Friday, May 12, 2017 6:33 PM
>> To: John W. Linville; Legacy, Allain (Wind River); Peters, Matt (Wind River);
>> Harish Patil; Rasesh Mody; Stephen Hurd; Ajit Khaparde; Doherty, Declan; L
On 5/12/2017 11:15 AM, Ashwin Sekhar T K wrote:
This series of patches adds the clang compilation support for armv8a linuxapp.
Patch 1 adds an rte_pause() to a tight while loop in rte_eal_wait_lcore().
It fixes warning -Wempty-body seen with armv8a clang compilation.
Patch 2 is basically for re
On 5/12/2017 11:40 AM, Jianbo Liu wrote:
On 12 May 2017 at 13:45, Ashwin Sekhar T K
wrote:
Moved all common defines from defconfig_arm64-armv8a-linuxapp-gcc
to common_armv8a_linuxapp.
Created new config arm64-armv8a-linuxapp-clang which adds the
clang support to armv8a.
Now defconfigs arm64-a
Signed-off-by: Thomas Monjalon
---
lib/librte_eal/common/include/rte_version.h | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/lib/librte_eal/common/include/rte_version.h
b/lib/librte_eal/common/include/rte_version.h
index 07a085eb8..e9534fb25 100644
--- a/lib/librte_ea
11/05/2017 14:57, John McNamara:
> Add template release notes for DPDK 17.08 with inline
> comments and explanations of the various sections.
>
> Signed-off-by: John McNamara
> ---
> doc/guides/rel_notes/release_17_08.rst | 198
> +
> 1 file changed, 198 insertio
Changes device type identification to be based on a unique
driver id replacing the current device type enumeration, which needed
library changes every time a new crypto driver was added.
The driver id is assigned dynamically during driver registration using
the new macro RTE_PMD_REGISTER_CRYPTO_DR
Hi Lijia,
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Michael Lilja
> Sent: Thursday, May 11, 2017 6:21 PM
> To: dev@dpdk.org
> Cc: Michael Lilja
> Subject: [dpdk-dev] [PATCH v2] drivers/net/i40e/i40e_fdir.c: Improved i40e
> FDIR programming times
>
> Duri
15/05/2017 10:40, Ferruh Yigit:
> On 5/12/2017 4:48 PM, Thomas Monjalon wrote:
> > 12/05/2017 12:49, Ferruh Yigit:
> >> On 5/10/2017 2:51 PM, Thomas Monjalon wrote:
> >>> 10/05/2017 15:10, Ferruh Yigit:
> --- a/doc/guides/nics/features/bnx2x.ini
> +++ b/doc/guides/nics/features/bnx2x.ini
On Mon, May 15, 2017 at 10:26:00AM +0100, Ferruh Yigit wrote:
> On 5/15/2017 10:19 AM, Yuanhan Liu wrote:
> > On Mon, May 15, 2017 at 10:17:43AM +0100, Ferruh Yigit wrote:
> >> On 5/12/2017 11:33 AM, Ferruh Yigit wrote:
> >>> remove __rte_unused instances that are not required.
> >>>
> >>
> >> Hi Y
On 5/15/2017 10:19 AM, Yuanhan Liu wrote:
> On Mon, May 15, 2017 at 10:17:43AM +0100, Ferruh Yigit wrote:
>> On 5/12/2017 11:33 AM, Ferruh Yigit wrote:
>>> remove __rte_unused instances that are not required.
>>>
>>
>> Hi Yuanhan,
>>
>> Does this kind of refactoring patches, specially the ones cove
On Mon, May 15, 2017 at 10:17:43AM +0100, Ferruh Yigit wrote:
> On 5/12/2017 11:33 AM, Ferruh Yigit wrote:
> > remove __rte_unused instances that are not required.
> >
>
> Hi Yuanhan,
>
> Does this kind of refactoring patches, specially the ones covering
> multiple drivers, cause trouble (more c
On 5/12/2017 11:33 AM, Ferruh Yigit wrote:
> remove __rte_unused instances that are not required.
>
Hi Yuanhan,
Does this kind of refactoring patches, specially the ones covering
multiple drivers, cause trouble (more conflicts) to you while getting
patches for stable trees?
If so, I can postpon
On 5/15/2017 3:31 AM, Yuanhan Liu wrote:
> On Fri, May 12, 2017 at 11:33:03AM +0100, Ferruh Yigit wrote:
>> remove __rte_unused instances that are not required.
>
> I'm wondering this is done by some scripts?
Manually, unfortunately..
>
> --yliu
>
On 5/12/2017 4:48 PM, Thomas Monjalon wrote:
> 12/05/2017 12:49, Ferruh Yigit:
>> On 5/10/2017 2:51 PM, Thomas Monjalon wrote:
>>> 10/05/2017 15:10, Ferruh Yigit:
--- a/doc/guides/nics/features/bnx2x.ini
+++ b/doc/guides/nics/features/bnx2x.ini
@@ -4,6 +4,7 @@
; Refer to defaul
On 5/12/2017 7:18 PM, Ferruh Yigit wrote:
On 4/19/2017 2:09 PM, Hemant Agrawal wrote:
Typically when the PMD issues a RX command to DPAA2 hardware,
the hw writes the available descriptors into the given memory.
The RX function then processes the frames and prepare them as
mbufs.
This patch adds
On Sat, May 13, 2017 at 02:57:25PM +0530, Jerin Jacob wrote:
> Different drivers use internal macros like force_inline for compiler
> always inline feature.
> Standardizing it through __rte_always_inline macro.
>
> Signed-off-by: Jerin Jacob
> ---
Good cleanup.
Series Acked-by: Bruce Richardson
60 matches
Mail list logo
