Flow director API does not provide a layer 2 configuration when the filter
is for layer 3 and 4 causing the translation to generic flow API to be
wrong, as not providing a mask for layer ends by using the default one. In
this case, the Ethernet mask layer is full whereas it must be empty.
Fixes:
On Thu, Oct 26, 2017 at 08:41:57PM +0300, Shahaf Shuler wrote:
> The flow engine in mlx5 searches for the most specific layer in the
> pattern in order to set the flow rule priority properly.
>
> Since the RSS can be currently performed only for the outer headers, avoid
> updating the layer for th
Move the vdev bus from lib/librte_eal to drivers/bus.
As the crypto vdev helper function refers to data structure
in rte_vdev.h, so we move those helper function into drivers/bus
too.
Signed-off-by: Jianfeng Tan
---
config/common_base| 5 +
doc/guides/rel_notes/release
Use specialized dynamic log type for vdev bus logging.
Suggested-by: Gaetan Rivet
Suggested-by: Shreyansh Jain
Signed-off-by: Jianfeng Tan
---
drivers/bus/vdev/vdev.c | 20
drivers/bus/vdev/vdev_logs.h | 45
2 files changed
We can call bus->plug() to avoid calling rte_vdev_init() explicitly.
Signed-off-by: Jianfeng Tan
---
lib/librte_eal/common/eal_common_dev.c | 22 ++
1 file changed, 6 insertions(+), 16 deletions(-)
diff --git a/lib/librte_eal/common/eal_common_dev.c
b/lib/librte_eal/common/
Remove rte_cryptodev_create_vdev() for duplication.
Signed-off-by: Jianfeng Tan
Acked-by: Pablo de Lara
---
doc/guides/rel_notes/deprecation.rst | 5 -
doc/guides/rel_notes/release_17_11.rst | 2 ++
lib/librte_cryptodev/rte_cryptodev.c | 6 --
lib/librte_c
v10:
- Rebase on 17.11-rc2.
v9:
- Avoid bumping library version (eal, crypto) as other commits in this
release already do that as suggested by Pablo.
- Move vdev's dir dependency to its own folder as suggested by Gaetan.
v8: (Suggested by Pablo)
- As we bump the library version and ad
In function eth_virtio_dev_init(), dynamic memory stored
in "eth_dev->data->mac_addrs" variable and it is not freed
when function return,
this is a possible memory leak.
Fixes: 8ced1542f7a3 ("net/virtio: eth_dev->data->mac_addrs is not freed")
Cc: sta...@dpdk.org
Signed-off-by: Pengzhen Liu
---
Hash table function will check if the input bucket size is power of 2,
so the parameter should be rounded up before sending to the creating function.
Signed-off-by: Jianbo Liu
---
examples/ip_pipeline/pipeline/pipeline_flow_classification_be.c | 2 +-
examples/ip_pipeline/pipeline/pipeline_routi
We are wrongly "OR-ing" the vnic->hash_type instead of assigning
the new hash type thereby wrongly configuring hash settings.
Fixing it.
Fixes: fcc0aa1edc10 ("net/bnxt: add RSS hash configuration")
Signed-off-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_rxq.c | 2 +-
1 file changed, 1 insertion(
> -Original Message-
> From: users [mailto:users-boun...@dpdk.org] On Behalf Of Iain Barker
> Sent: Wednesday, October 11, 2017 9:20 PM
> To: us...@dpdk.org
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-users] VLAN tags always stripped on i40evf [VMware SR-
> IOV]
>
> On Tuesday, October 10, 20
Use specialized dynamic log type for vdev bus logging.
Suggested-by: Gaetan Rivet
Suggested-by: Shreyansh Jain
Signed-off-by: Jianfeng Tan
---
drivers/bus/vdev/vdev.c | 20
drivers/bus/vdev/vdev_logs.h | 45
2 files changed
Move the vdev bus from lib/librte_eal to drivers/bus.
As the crypto vdev helper function refers to data structure
in rte_vdev.h, so we move those helper function into drivers/bus
too.
Signed-off-by: Jianfeng Tan
---
config/common_base| 5 +
doc/guides/rel_notes/release
v9:
- Avoid bumping library version (eal, crypto) as other commits in this
release already do that as suggested by Pablo.
- Move vdev's dir dependency to its own folder as suggested by Gaetan.
v8: (Suggested by Pablo)
- As we bump the library version and add new library (bus_vdev), we
We can call bus->plug() to avoid calling rte_vdev_init() explicitly.
Signed-off-by: Jianfeng Tan
---
lib/librte_eal/common/eal_common_dev.c | 22 ++
1 file changed, 6 insertions(+), 16 deletions(-)
diff --git a/lib/librte_eal/common/eal_common_dev.c
b/lib/librte_eal/common/
Remove rte_cryptodev_create_vdev() for duplication.
Signed-off-by: Jianfeng Tan
Acked-by: Pablo de Lara
---
doc/guides/rel_notes/deprecation.rst | 5 -
doc/guides/rel_notes/release_17_11.rst | 2 ++
lib/librte_cryptodev/rte_cryptodev.c | 6 --
lib/librte_c
On 10/26/2017 2:28 AM, Tan, Jianfeng wrote:
> Hi Rasesh,
>
>
> On 10/26/2017 7:43 AM, Mody, Rasesh wrote:
>> Hi Ferruh,
>>
>>> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ferruh Yigit
>>> Sent: Friday, October 20, 2017 9:58 AM
>>>
>>> On 10/20/2017 9:55 AM, Ferruh Yigit wrote:
Remov
SDXcentral is running a survey to gather industry insights and market research
on networking and open source. If you’re interested in participating and
giving your input here’s a link:
https://www.sdxcentral.com/articles/announcements/open-source-networking-survey/2017/10/
Note: for the purpose
A new DPDK release candidate is ready for testing:
http://dpdk.org/browse/dpdk/tag/?id=v17.11-rc2
What's new in RC2?
- igb_uio reset fixed
- EAL PCI moved to driver and lib
- Rx offload API for traffic metering and policing (rte_mtr)
- IPsec offload API (rte
On 10/26/2017 10:44 AM, Stefan Baranoff wrote:
> All,
>
> I think I've found a (very minor) memory leak in writing to a PCAP file. In
> drivers/net/pcap/rte_eth_pcap.c in open_single_tx_pcap around line 405 a
> pcap_t is allocated by pcap_open_dead but is never freed.
You are right.
> I see two
25/10/2017 17:50, Marko Kovacevic:
> From: Herakliusz Lipiec
>
> Moved duplicated, and occasionally outdated, doc sections from each
> of the sample app guides chapters to a common chapter at the start.
>
> This reduces the duplication in the docs and provides a single
> point of reference for c
25/10/2017 17:27, Mcnamara, John:
>
> > -Original Message-
> > From: Ori Kam [mailto:or...@mellanox.com]
> > Sent: Wednesday, October 25, 2017 3:51 PM
> > To: adrien.mazarg...@6wind.com; Mcnamara, John
> > Cc: dev@dpdk.org; or...@mellanox.com
> > Subject: [PATCH v2] doc: add basic howto f
Few comments below, Ori:
25/10/2017 19:58, Ori Kam:
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -978,3 +978,7 @@ F: examples/tep_termination/
> F: examples/vmdq/
> F: examples/vmdq_dcb/
> F: doc/guides/sample_app_ug/vmdq_dcb_forwarding.rst
> +
> +M: Ori Kam
> +F: examples/flow_filtering/
> +F
26/10/2017 10:29, Xueming Li:
> Simple_mp example receives message size less than 64 chars while send
> side accepts chars less than 128, this leads to different result when
> sending text lenght larger than 64.
> This patch uses same buffer length on both message pool and command
> line.
>
> Fixe
> > Signed-off-by: Nikhil Rao
> > Acked-by: Jerin Jacob
>
> Acked-by: John McNamara
Applied, thanks
25/10/2017 16:50, Pavan Nikhilesh:
> In case of sw event device the scheduling can be done on a service core
> using the service registered at the time of probe.
> This patch adds a helper function to get the service id that can be used
> by the application to assign a lcore for the service to run
25/10/2017 16:21, Pavan Nikhilesh:
> With the current scheme of event queue configuration the cfg schedule
> type macros (RTE_EVENT_QUEUE_CFG_*_ONLY) are inconsistent with the
> event schedule type (RTE_SCHED_TYPE_*) this requires unnecessary
> conversion between the fastpath and slowpath API's whi
On 10/26/2017 1:27 AM, Nélio Laranjeiro wrote:
> On Wed, Oct 25, 2017 at 04:30:40PM -0700, Yongseok Koh wrote:
>> When "net/mlx5: replace network to host macros" is rebased for v4, changes
>> of "net/mlx5: fix calculating TSO inline size" have been reverted.
>>
>> Fixes: 6b30a6a8552a ("net/mlx5: re
On 10/26/2017 1:12 AM, Nélio Laranjeiro wrote:
> On Wed, Oct 25, 2017 at 06:37:26PM +0300, Moti Haimovsky wrote:
>> This commit defines MLX4_CQ_DB_CI_MASK which is used when updating
>> the consumer index of the completion queue instead of the hardcoded
>> 0xff used until now.
>>
>> Signed-off-
> > Set log_level to RTE_LOG_INFO.
> > The RTE_LIBRTE_CLASSIFY_DEBUG macro has been removed from the config
> > file, use the log_level instead.
> >
> > Fixes: be41ac2a330f ("flow_classify: introduce flow classify library")
> > Signed-off-by: Bernard Iremonger
>
> Acked-by: Jasvinder Singh
App
04/09/2017 15:02, Bruce Richardson:
> Couple of other comments on this code area that should be taken into
> account:
> 1. the indentation of the asm code looks wrong, and should probably be
>fixed to make it more readable.
> 2. the comment on the "D" register is wrong as it refers to ebx
> 3.
Cc Ferruh, maintainer of this driver.
26/10/2017 19:44, Stefan Baranoff:
> All,
>
> I think I've found a (very minor) memory leak in writing to a PCAP file. In
> drivers/net/pcap/rte_eth_pcap.c in open_single_tx_pcap around line 405 a
> pcap_t is allocated by pcap_open_dead but is never freed.
>
26/10/2017 17:23, Burakov, Anatoly:
> On 26-Oct-17 4:19 PM, Jonas Pfefferle wrote:
> > Check return value of device reset ioctl
> >
> > Fixes: 33604c31354a ("vfio: refactor PCI BAR mapping")
> > Coverity issue: 195003
> >
> > Signed-off-by: Jonas Pfefferle
>
> Looks fine to me, so
>
> Acked-by
24/10/2017 10:25, Gaëtan Rivet:
> On Tue, Oct 24, 2017 at 07:44:53AM +, Jianfeng Tan wrote:
> > When checking if any devices bound to uio, we did not exclud
> > those which are blacklisted (or in the case that a whitelist
> > is specified).
> >
> > This patch fixes it by only checking whitelis
26/10/2017 12:05, Gaetan Rivet:
> Here is a new version of the PCI bus move out of the EAL.
>
> The EAL PCI implementation is divided in two parts:
>
> - librte_pci: library offering helpers to handle PCI objects
> - librte_bus_pci: bus driver for PCI devices
Applied with few fixes, thanks f
26/10/2017 16:05, Gonzalez Monroy, Sergio:
> On 25/10/2017 09:36, Ferruh Yigit wrote:
> > build error:
> >CC rte_cycles.o
> >cc1: warnings being treated as errors
> >...dpdk/lib/librte_eal/common/arch/x86/rte_cycles.c: In function
> >‘rdmsr’:
> >...dpdk/lib/librte_eal/common/arc
26/10/2017 16:10, Akhil Goyal:
> On 10/26/2017 7:39 PM, Shreyansh Jain wrote:
> > Various symbols are being used by DPAA Crypto driver which were not exposed
> > from DPAA bus during initial version. This breaks the shared build.
> >
> > This patch also adds the LDLIBS line required after (cbc12b0
26/10/2017 17:27, David Marchand:
> On Thu, Oct 26, 2017 at 4:15 PM, Radu Nicolau wrote:
> > v2: Added ifdefs in the ixgbe code, updated dependecies for ipsec_secgw,
> > dpaa2_sec kept as in v1, with a dependency check
> >
> > Radu Nicolau (3):
> > net/ixgbe: fix build issue
> > examples/ipsec
On 10/23/2017 9:42 PM, Srisivasubramanian S wrote:
> Signed-off-by: Srisivasubramanian S
> Signed-off-by: Shijith Thotton
Applied to dpdk-next-net/master, thanks.
On 10/26/2017 9:20 AM, Gaëtan Rivet wrote:
> Hello Matan,
>
> I think the commit log could be shorter.
> Proposing this, feel free to expand it if you prefer.
>
> ---8<---
>
> When removing a device, the fail-safe checks that it is not within its
> datapath before cleaning it.
>
> When checking
On 9/11/2017 8:13 AM, Olivier Matz wrote:
> The compilation with gcc-6.3.0 and EXTRA_CFLAGS=-Og gives the following
> error:
>
> CC rte_lpm6.o
> rte_lpm6.c: In function ‘rte_lpm6_add_v1705’:
> rte_lpm6.c:442:11: error: ‘tbl_next’ may be used uninitialized in
> th
On 10/25/2017 7:23 PM, Srisivasubramanian Srinivasan wrote:
> On Wed, 2017-10-25 at 11:23 -0700, Ferruh Yigit wrote:
>> On 10/25/2017 11:06 AM, Srisivasubramanian Srinivasan wrote:
>>> On Tue, 2017-10-24 at 10:49 -0700, Ferruh Yigit wrote:
On 10/23/2017 9:42 PM, Srisivasubramanian S wrote:
>>>
On 10/26/2017 10:18 AM, Ryan Yang wrote:
> Hello,
>
> I am really interested in contributing to DPDK by fixing the bug and
> development work. Besides cloning the code first, what are some good ways
> to start the work or is there any place to check the bug list? Your help is
> appreciated.
Hi Ry
All,
I think I've found a (very minor) memory leak in writing to a PCAP file. In
drivers/net/pcap/rte_eth_pcap.c in open_single_tx_pcap around line 405 a
pcap_t is allocated by pcap_open_dead but is never freed.
I see two obvious fixes:
1) Free the pcap_t immediately after either on pcap_dump_o
26/10/2017 12:06, Gaetan Rivet:
> lib/librte_pci/Makefile | 49 ++
> lib/librte_pci/include/rte_pci.h | 279 +
> lib/librte_pci/rte_pci.c | 220 +++
> lib/librte_pci/rte_pci_version.map | 15 +
The include file is usually
The flow engine in mlx5 searches for the most specific layer in the
pattern in order to set the flow rule priority properly.
Since the RSS can be currently performed only for the outer headers, avoid
updating the layer for the inner headers.
Fixes: 8086cf08b2f0 ("net/mlx5: handle RSS hash configu
Hello,
I am really interested in contributing to DPDK by fixing the bug and
development work. Besides cloning the code first, what are some good ways
to start the work or is there any place to check the bug list? Your help is
appreciated.
Ryan
Hi Nelio
> -Original Message-
> From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com]
> Sent: Thursday, October 26, 2017 4:44 PM
> To: Matan Azrad
> Cc: Ophir Munk ; Adrien Mazarguil
> ; dev@dpdk.org; Thomas Monjalon
> ; Olga Shern ; Mordechay
> Haimovsky
> Subject: Re: [dpdk-dev] [
Hello Matan,
I think the commit log could be shorter.
Proposing this, feel free to expand it if you prefer.
---8<---
When removing a device, the fail-safe checks that it is not within its
datapath before cleaning it.
When checking whether an Rx burst should be performed on a device, the
remove
26/10/2017 12:06, Gaetan Rivet:
> Libraries, drivers, unit tests and applications are updated to use the
> new rte_bus_pci.h header when necessary.
Few compilation fixes:
--- a/drivers/crypto/qat/qat_crypto.c
+++ b/drivers/crypto/qat/qat_crypto.c
@@ -60,6 +60,7 @@
#include
#include
#include
On Wed, Oct 18, 2017 at 09:48:04PM +, Carrillo, Erik G wrote:
>
>
> > -Original Message-
> > From: Pavan Nikhilesh Bhagavatula
> > [mailto:pbhagavat...@caviumnetworks.com]
> > Sent: Monday, October 16, 2017 7:37 AM
> > To: Carrillo, Erik G
> > Cc: dev@dpdk.org
> > Subject: Re: [RFC PAT
On Thu, Oct 26, 2017 at 4:15 PM, Radu Nicolau wrote:
> v2: Added ifdefs in the ixgbe code, updated dependecies for ipsec_secgw,
> dpaa2_sec kept as in v1, with a dependency check
>
> Radu Nicolau (3):
> net/ixgbe: fix build issue
> examples/ipsec-secgw: fix build issue
> crypto/dpaa2_sec: fi
On 26-Oct-17 4:19 PM, Jonas Pfefferle wrote:
Check return value of device reset ioctl
Fixes: 33604c31354a ("vfio: refactor PCI BAR mapping")
Coverity issue: 195003
Signed-off-by: Jonas Pfefferle
---
v2:
* fix whitespace/tab issue
lib/librte_eal/linuxapp/eal/eal_pci_vfio.c | 6 +-
1 fil
Check return value of device reset ioctl
Fixes: 33604c31354a ("vfio: refactor PCI BAR mapping")
Coverity issue: 195003
Signed-off-by: Jonas Pfefferle
---
v2:
* fix whitespace/tab issue
lib/librte_eal/linuxapp/eal/eal_pci_vfio.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff -
Check return value of device reset ioctl
Fixes: 33604c31354a ("vfio: refactor PCI BAR mapping")
Coverity issue: 195003
Signed-off-by: Jonas Pfefferle
---
lib/librte_eal/linuxapp/eal/eal_pci_vfio.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/lib/librte_eal/linuxapp/e
Hi,
Please see inside
On Thursday, October 26, 2017 10:49 AM Adrien Mazarguil wrote:
> To: Ophir Munk
> Cc: dev@dpdk.org; Thomas Monjalon ; Olga Shern
> ; Matan Azrad
> Subject: Re: [PATCH v2 2/7] net/mlx4: inline more Tx functions
>
> Hi Ophir,
>
> Please see below.
>
> On Wed, Oct 25, 2017
Build fails when rte_security is disabled; make rte_security mandatory
Fixes: 0a23d4b6f4c2 ("crypto/dpaa2_sec: support protocol offload IPsec")
Signed-off-by: Radu Nicolau
---
drivers/crypto/dpaa2_sec/Makefile | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/crypto/dpaa2_sec/Mak
v2: Added ifdefs in the ixgbe code, updated dependecies for ipsec_secgw,
dpaa2_sec kept as in v1, with a dependency check
Radu Nicolau (3):
net/ixgbe: fix build issue
examples/ipsec-secgw: fix build issue
crypto/dpaa2_sec: fix build issue
drivers/crypto/dpaa2_sec/Makefile | 6 ++
Fixes: 9a0752f498d2 ("net/ixgbe: enable inline IPsec")
Signed-off-by: Radu Nicolau
---
drivers/net/ixgbe/Makefile | 2 ++
drivers/net/ixgbe/ixgbe_ethdev.c | 8 +++
drivers/net/ixgbe/ixgbe_ethdev.h | 4
drivers/net/ixgbe/ixgbe_flow.c | 6 ++
driver
Build fails when rte_security is disabled; make rte_security mandatory
Fixes: ec17993a145a ("examples/ipsec-secgw: support security offload")
Signed-off-by: Radu Nicolau
---
examples/Makefile | 2 +-
examples/ipsec-secgw/Makefile | 6 ++
2 files changed, 7 insertions(+), 1 deleti
> -Original Message-
> From: Shreyansh Jain [mailto:shreyansh.j...@nxp.com]
> Sent: Thursday, October 26, 2017 19:56
> To: dev@dpdk.org
> Cc: tho...@monjalon.net; john.mcnam...@intel.com; Shreyansh Jain
>
> Subject: [PATCH] doc: update release note for DPAA2 Event PMD
>
> Signed-off-by:
Signed-off-by: Shreyansh Jain
---
doc/guides/rel_notes/release_17_11.rst | 4
1 file changed, 4 insertions(+)
diff --git a/doc/guides/rel_notes/release_17_11.rst
b/doc/guides/rel_notes/release_17_11.rst
index 93f94fc..d0ca2cf 100644
--- a/doc/guides/rel_notes/release_17_11.rst
+++ b/doc/gu
On 10/26/2017 7:39 PM, Shreyansh Jain wrote:
Various symbols are being used by DPAA Crypto driver which were not exposed
from DPAA bus during initial version. This breaks the shared build.
This patch also adds the LDLIBS line required after (cbc12b0a9) patch.
Fixes: c3e85bdcc6e6 ("crypto/dpaa_s
In function eth_virtio_dev_init(), dynamic memory stored
in "eth_dev->data->mac_addrs" variable and it is not freed
when function return,
this is a possible memory leak.
Fixes: 289ba0c0f52a ("net/virtio: eth_dev->data->mac_addrs is not freed")
Cc: sta...@dpdk.org
Signed-off-by: Pengzhen Liu
---
On 25/10/2017 09:36, Ferruh Yigit wrote:
build error:
CC rte_cycles.o
cc1: warnings being treated as errors
...dpdk/lib/librte_eal/common/arch/x86/rte_cycles.c: In function
‘rdmsr’:
...dpdk/lib/librte_eal/common/arch/x86/rte_cycles.c:67:2: error:
implicit declaration of function
From: Tonghao Zhang
If dpdk APPs call the rte_eth_dev_rx_intr_enable or
rte_eth_dev_rx_intr_disable frequently, and ixgbe vf will read the
IXGBE_VTEIMS register everytime. We can optimize the driver function.
The patch save the IXGBE_VTEIMS to mask to avoid read frequently.
Signed-off-by: Tongha
From: Tonghao Zhang
When we bind the ixgbevf to vfio and call the rte_eth_dev_rx_intr_enable
and rte_eth_dev_rx_intr_disable frequently, the interrupt setting
(msi_set_mask_bit) will take more CPU as show below. rte_intr_enable
call the ioctl to map the fd to interrupts frequently.
perf top:
5.4
From: Tonghao Zhang
Set EITR interval as default. This patch can improve the performance
when we enable the rx-interrupt to process the packets because we hope
rx-interrupt reduce CPU.
Signed-off-by: Tonghao Zhang
---
drivers/net/ixgbe/ixgbe_ethdev.c | 3 +++
1 file changed, 3 insertions(+)
d
From: Tonghao Zhang
When we bind the ixgbe VF (e.g 82599 card) to igb_uio and enable the
rx-interrupt, there will be more than one epoll_wait on intr_handle.fd.
One is in "eal-intr-thread" thread, and the others are in the thread
which call the "rte_epoll_wait". The problem is that sometiems
"ea
26/10/2017 15:16, Akhil Goyal:
> On 10/26/2017 6:37 PM, Thomas Monjalon wrote:
> > 26/10/2017 14:59, Akhil Goyal:
> >> Hi Thomas,
> >>
> >> On 10/26/2017 6:03 PM, Thomas Monjalon wrote:
> >>> 26/10/2017 14:28, Radu Nicolau:
>
> On 10/26/2017 12:39 PM, Thomas Monjalon wrote:
> > 26/10/
Various symbols are being used by DPAA Crypto driver which were not exposed
from DPAA bus during initial version. This breaks the shared build.
This patch also adds the LDLIBS line required after (cbc12b0a9) patch.
Fixes: c3e85bdcc6e6 ("crypto/dpaa_sec: add crypto driver for NXP DPAA platform")
C
Fixes: cbc12b0a96f5 ("mk: do not generate LDLIBS from directory dependencies")
Cc: olivier.m...@6wind.com
Fixes: b677d4c6d281 ("net/dpaa2: add API for event Rx adapter")
Cc: nipun.gu...@nxp.com
Signed-off-by: Shreyansh Jain
---
* This change should have been originally part of cbc12b0a96 but was
From: Tonghao Zhang
When we bind the ixgbevf to vfio and call the rte_eth_dev_rx_intr_enable
and rte_eth_dev_rx_intr_disable frequently, the interrupt setting
(msi_set_mask_bit) will take more CPU as show below. rte_intr_enable
call the ioctl to map the fd to interrupts frequently.
perf top:
5.4
From: Tonghao Zhang
Set EITR interval as default. This patch can improve the performance
when we enable the rx-intrrupt to process the packets because we hope
rx-intrrupt reduce CPU.
Signed-off-by: Tonghao Zhang
Signed-off-by: Tonghao Zhang
---
drivers/net/ixgbe/ixgbe_ethdev.c | 3 +++
1 file
From: Tonghao Zhang
If dpdk APPs call the rte_eth_dev_rx_intr_enable or rte_eth_dev_rx_intr_disable
frequently, and ixgbe vf will read the IXGBE_VTEIMS register everytime. We can
optimize the driver function. The patch save the IXGBE_VTEIMS to mask to avoid
read frequently.
Signed-off-by: Tongha
From: Tonghao Zhang
When we bind the ixgbe VF (e.g 82599 card) to igb_uio and enable the
rx-interrupt,
there will be more than one epoll_wait on intr_handle.fd. One is in
"eal-intr-thread"
thread, and the others are in the thread which call the "rte_epoll_wait". The
problem
is that sometiems
On Thu, Oct 26, 2017 at 12:30:54PM +, Matan Azrad wrote:
> Hi Nelio
> Please see my comments below (3).
>
>
> > -Original Message-
> > From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com]
> > Sent: Thursday, October 26, 2017 3:12 PM
> > To: Matan Azrad
> > Cc: Ophir Munk ; Adri
On 10/26/2017 6:37 PM, Thomas Monjalon wrote:
26/10/2017 14:59, Akhil Goyal:
Hi Thomas,
On 10/26/2017 6:03 PM, Thomas Monjalon wrote:
26/10/2017 14:28, Radu Nicolau:
On 10/26/2017 12:39 PM, Thomas Monjalon wrote:
26/10/2017 13:27, David Marchand:
On Thu, Oct 26, 2017 at 1:01 PM, Radu Nicol
26/10/2017 14:59, Akhil Goyal:
> Hi Thomas,
>
> On 10/26/2017 6:03 PM, Thomas Monjalon wrote:
> > 26/10/2017 14:28, Radu Nicolau:
> >>
> >> On 10/26/2017 12:39 PM, Thomas Monjalon wrote:
> >>> 26/10/2017 13:27, David Marchand:
> On Thu, Oct 26, 2017 at 1:01 PM, Radu Nicolau
> wrote:
>
Hi Thomas,
On 10/26/2017 6:03 PM, Thomas Monjalon wrote:
26/10/2017 14:28, Radu Nicolau:
On 10/26/2017 12:39 PM, Thomas Monjalon wrote:
26/10/2017 13:27, David Marchand:
On Thu, Oct 26, 2017 at 1:01 PM, Radu Nicolau wrote:
On 10/26/2017 11:36 AM, David Marchand wrote:
On Thu, Oct 26, 2017
Hi @all
I just stumbled upon this patch while testing on POWER. RTE_IOVA_VA will
not work for the sPAPR code since the dma window size is currently
determined by the physical address only. I'm preparing a patch to address
this.
Thanks,
Jonas
"dev" wrote on 08/31/2017 05:26:16 AM:
> From: Sant
26/10/2017 14:28, Radu Nicolau:
>
> On 10/26/2017 12:39 PM, Thomas Monjalon wrote:
> > 26/10/2017 13:27, David Marchand:
> >> On Thu, Oct 26, 2017 at 1:01 PM, Radu Nicolau
> >> wrote:
> >>> On 10/26/2017 11:36 AM, David Marchand wrote:
> On Thu, Oct 26, 2017 at 12:22 PM, Radu Nicolau
> >>>
> -Original Message-
> From: David Marchand [mailto:david.march...@6wind.com]
> Sent: Thursday, October 26, 2017 12:39 PM
> To: Ananyev, Konstantin
> Cc: Nicolau, Radu ; dev@dpdk.org; Doherty, Declan
> ; De Lara Guarch, Pablo
> ; Hemant Agrawal ;
> bor...@mellanox.com; avia...@mellanox
Hi Nelio
Please see my comments below (3).
> -Original Message-
> From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com]
> Sent: Thursday, October 26, 2017 3:12 PM
> To: Matan Azrad
> Cc: Ophir Munk ; Adrien Mazarguil
> ; dev@dpdk.org; Thomas Monjalon
> ; Olga Shern ; Mordechay
> Hai
On 10/26/2017 12:39 PM, David Marchand wrote:
On Thu, Oct 26, 2017 at 1:30 PM, Ananyev, Konstantin
wrote:
-Original Message-
From: Nicolau, Radu
It can be disabled in the code, but as far as I know there is a general
push back against having conditionally compiled code. I originally h
On 10/26/2017 12:39 PM, Thomas Monjalon wrote:
26/10/2017 13:27, David Marchand:
On Thu, Oct 26, 2017 at 1:01 PM, Radu Nicolau wrote:
On 10/26/2017 11:36 AM, David Marchand wrote:
On Thu, Oct 26, 2017 at 12:22 PM, Radu Nicolau
wrote:
--- a/drivers/net/ixgbe/Makefile
+++ b/drivers/net/ixgbe
On Thu, Oct 26, 2017 at 10:31:06AM +, Matan Azrad wrote:
> Hi Nelio
>
> I think the memory barrier discussion is not relevant for this patch
> (if it will be relevant I will create new one).
> Please see my comments inline.
It was not my single comment. There is also useless code like having
> -Original Message-
> From: Iremonger, Bernard
> Sent: Thursday, October 26, 2017 11:48 AM
> To: dev@dpdk.org; Singh, Jasvinder
> Cc: Iremonger, Bernard
> Subject: [PATCH] flow_classify: fix logging
>
> Set log_level to RTE_LOG_INFO.
> The RTE_LIBRTE_CLASSIFY_DEBUG macro has been remo
26/10/2017 13:27, David Marchand:
> On Thu, Oct 26, 2017 at 1:01 PM, Radu Nicolau wrote:
> > On 10/26/2017 11:36 AM, David Marchand wrote:
> >> On Thu, Oct 26, 2017 at 12:22 PM, Radu Nicolau
> >> wrote:
> >>> --- a/drivers/net/ixgbe/Makefile
> >>> +++ b/drivers/net/ixgbe/Makefile
> >>> +ifneq ($(
On Thu, Oct 26, 2017 at 1:30 PM, Ananyev, Konstantin
wrote:
>> -Original Message-
>> From: Nicolau, Radu
>> It can be disabled in the code, but as far as I know there is a general
>> push back against having conditionally compiled code. I originally had
>> the security sections in ixgbe PM
> -Original Message-
> From: Nicolau, Radu
> Sent: Thursday, October 26, 2017 12:01 PM
> To: David Marchand
> Cc: dev@dpdk.org; Doherty, Declan ; De Lara Guarch,
> Pablo ; Hemant
> Agrawal ; bor...@mellanox.com; avia...@mellanox.com;
> Thomas Monjalon ;
> sandeep.ma...@nxp.com; Jerin J
On Thu, Oct 26, 2017 at 1:01 PM, Radu Nicolau wrote:
> On 10/26/2017 11:36 AM, David Marchand wrote:
>> On Thu, Oct 26, 2017 at 12:22 PM, Radu Nicolau
>> wrote:
>>>
>>> Build fails when rte_security is disabled; make rte_security mandatory
>>> Fixes: 9a0752f498d2 ("net/ixgbe: enable inline IPsec"
On 10/26/2017 11:36 AM, David Marchand wrote:
On Thu, Oct 26, 2017 at 12:22 PM, Radu Nicolau wrote:
Build fails when rte_security is disabled; make rte_security mandatory
Fixes: 9a0752f498d2 ("net/ixgbe: enable inline IPsec")
Signed-off-by: Radu Nicolau
---
drivers/net/ixgbe/Makefile | 6
Fixes: 5b590fbe09b6 ("app/testpmd: add traffic management forwarding mode")
Signed-off-by: Jasvinder Singh
---
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 15 ---
1 file changed, 8 insertions(+), 7 deletions(-)
diff --git a/doc/guides/testpmd_app_ug/testpmd_funcs.rst
b/doc/guides
Set log_level to RTE_LOG_INFO.
The RTE_LIBRTE_CLASSIFY_DEBUG macro has been removed from the
config file, use the log_level instead.
Fixes: be41ac2a330f ("flow_classify: introduce flow classify library")
Signed-off-by: Bernard Iremonger
---
lib/librte_flow_classify/rte_flow_classify.c | 19 +
On Thu, 2017-10-26 at 08:08 +, Dai, Wei wrote:
> Current rte_eth_dev_reset( ) from my patches are just to reset the
> NIC port, NOT recovery its traffic.
> All my patches for NIC reset have already been accepted.
>
> It is the duty of user application to recovery any settings and
> traffic by
On Fri, Oct 06, 2017 at 01:18:00AM +0100, Ferruh Yigit wrote:
> On 9/11/2017 4:13 PM, Olivier Matz wrote:
> > The compilation with gcc-6.3.0 and EXTRA_CFLAGS=-Og gives the following
> > error:
> >
> > CC rte_lpm6.o
> > rte_lpm6.c: In function ‘rte_lpm6_add_v1705’:
> > rte_lpm6.c:442:11: erro
On Thu, Oct 26, 2017 at 12:22 PM, Radu Nicolau wrote:
> Build fails when rte_security is disabled; make rte_security mandatory
> Fixes: 9a0752f498d2 ("net/ixgbe: enable inline IPsec")
>
> Signed-off-by: Radu Nicolau
> ---
> drivers/net/ixgbe/Makefile | 6 ++
> 1 file changed, 6 insertions(+)
On 10/26/2017 3:52 PM, Radu Nicolau wrote:
Disabling CONFIG_RTE_LIBRTE_SECURITY results in build failure for
crypto/dpaa2_sec and net/ixgbe PMDs and IPsec sample app.
This patchset makes CONFIG_RTE_LIBRTE_SECURITY mandatory for all
of the above.
Radu Nicolau (3):
crypto/dpaa2_sec: fix build i
Hi Nelio
I think the memory barrier discussion is not relevant for this patch (if it
will be relevant I will create new one).
Please see my comments inline.
Regarding this specific patch, I didn't see any comment from you, Are you agree
with it?
> -Original Message-
> From: Nélio Lar
1 - 100 of 146 matches
Mail list logo
