-Original Message-
> Date: Thu, 11 May 2017 11:07:18 +0530
> From: "Sekhar, Ashwin"
> To: "Jacob, Jerin"
> CC: "bruce.richard...@intel.com" ,
> "tho...@monjalon.net" ,
> "konstantin.anan...@intel.com" ,
> "pablo.de.lara.gua...@intel.com" ,
> "Czekaj, Maciej" , "vikto...@rehivetech.co
Hi, Adrien
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Adrien Mazarguil
> Sent: Wednesday, December 21, 2016 10:52 PM
> To: dev@dpdk.org
> Subject: [dpdk-dev] [PATCH v5 19/26] app/testpmd: add item raw to flow
> command
>
> Matches arbitrary byte strings wi
guest_pages is being allocated in vhost_setup_mem_table(), reallocated
in add_one_guest_page(), but never freed. This patch fixes a memory leak.
Change-Id: I6edff00298a668054f4d106ffdaea828f41912fe
Signed-off-by: Dariusz Stojaczyk
---
lib/librte_vhost/vhost_user.c | 4
1 file changed, 4 ins
From: Daniel Verkamp
Fixes unaligned access to fields.
Change-Id: I2f8246481dc96d7e73134866f55a4227fc39031d
Signed-off-by: Daniel Verkamp
Signed-off-by: Dariusz Stojaczyk
---
lib/librte_vhost/vhost_user.c | 56 +--
1 file changed, 28 insertions(+), 28 d
Amount of allocated memory was too small, causing buffer overflow.
Change-Id: I02293cd4bcddef14d2bbf314c79d8c65fc44e555
Signed-off-by: Dariusz Stojaczyk
---
lib/librte_vhost/vhost.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vh
Hi,
Status quo:
Almost none of vdev supports primary/secondary model. Two exceptions are
rte_ring, virtio-user (limited support). Two problems facing this issue.
P1: How to attach vdev in secondary process?
Previous discussion: http://dpdk.org/ml/archives/dev/2017-April/063370.html
According
On 11 May 2017 at 12:27, Sekhar, Ashwin wrote:
>
> On Thu, 2017-05-11 at 04:14 +, Sekhar, Ashwin wrote:
> ...
>> > > Combining all the above comments, I made some changes on top of
>> > > your
>> > > patch. These changes are giving 3-4% improvement over your
>> > > version.
>> > >
>> > > You m
-Original Message-
> Date: Wed, 10 May 2017 03:16:37 -0700
> From: Ashwin Sekhar T K
> To: tho...@monjalon.net, jerin.ja...@caviumnetworks.com,
> maciej.cze...@caviumnetworks.com, vikto...@rehivetech.com,
> jianbo@linaro.org, bruce.richard...@intel.com,
> pablo.de.lara.gua...@intel.
-Original Message-
> Date: Wed, 10 May 2017 03:16:43 -0700
> From: Ashwin Sekhar T K
> To: tho...@monjalon.net, jerin.ja...@caviumnetworks.com,
> maciej.cze...@caviumnetworks.com, vikto...@rehivetech.com,
> jianbo@linaro.org, bruce.richard...@intel.com,
> pablo.de.lara.gua...@intel.
On Thu, 2017-05-11 at 10:54 +0530, Jerin Jacob wrote:
> -Original Message-
> >
> > Date: Wed, 10 May 2017 03:16:39 -0700
> > From: Ashwin Sekhar T K
> > To: tho...@monjalon.net, jerin.ja...@caviumnetworks.com,
> > maciej.cze...@caviumnetworks.com, vikto...@rehivetech.com,
> > jianbo
-Original Message-
> Date: Wed, 10 May 2017 03:16:42 -0700
> From: Ashwin Sekhar T K
> To: tho...@monjalon.net, jerin.ja...@caviumnetworks.com,
> maciej.cze...@caviumnetworks.com, vikto...@rehivetech.com,
> jianbo@linaro.org, bruce.richard...@intel.com,
> pablo.de.lara.gua...@intel.
-Original Message-
> Date: Wed, 10 May 2017 03:16:41 -0700
> From: Ashwin Sekhar T K
> To: tho...@monjalon.net, jerin.ja...@caviumnetworks.com,
> maciej.cze...@caviumnetworks.com, vikto...@rehivetech.com,
> jianbo@linaro.org, bruce.richard...@intel.com,
> pablo.de.lara.gua...@intel.
-Original Message-
> Date: Wed, 10 May 2017 03:16:40 -0700
> From: Ashwin Sekhar T K
> To: tho...@monjalon.net, jerin.ja...@caviumnetworks.com,
> maciej.cze...@caviumnetworks.com, vikto...@rehivetech.com,
> jianbo@linaro.org, bruce.richard...@intel.com,
> pablo.de.lara.gua...@intel.
> >> 5. The addition of inline crypto metadata into the rte_mbuf structure to
> allow the required egress metadata to be given to the NIC PMD to build the
> necessary transmit descriptors in tx_burst processing when the
> PKT_TX_IPSEC_INLINE_CRYPTO is set. We are looking for feedback on a
> better
-Original Message-
> Date: Wed, 10 May 2017 03:16:39 -0700
> From: Ashwin Sekhar T K
> To: tho...@monjalon.net, jerin.ja...@caviumnetworks.com,
> maciej.cze...@caviumnetworks.com, vikto...@rehivetech.com,
> jianbo@linaro.org, bruce.richard...@intel.com,
> pablo.de.lara.gua...@intel.
-Original Message-
> Date: Wed, 10 May 2017 03:16:38 -0700
> From: Ashwin Sekhar T K
> To: tho...@monjalon.net, jerin.ja...@caviumnetworks.com,
> maciej.cze...@caviumnetworks.com, vikto...@rehivetech.com,
> jianbo@linaro.org, bruce.richard...@intel.com,
> pablo.de.lara.gua...@intel.
On Thu, 2017-05-11 at 04:14 +, Sekhar, Ashwin wrote:
...
> > > Combining all the above comments, I made some changes on top of
> > > your
> > > patch. These changes are giving 3-4% improvement over your
> > > version.
> > >
> > > You may find the changes at
> > > https://gist.github.com/ashwi
On Thu, 2017-05-11 at 11:16 +0800, Jianbo Liu wrote:
> Hi Ashwin,
>
> On 10 May 2017 at 23:00, Sekhar, Ashwin
> wrote:
> >
> > Hi Jianbo,
> >
> > Thanks for version v2. Addition of the prefetch instructions is
> > definitely helping performance on ThunderX. But still performance
> > is
> > slig
Hi Ashwin,
On 10 May 2017 at 23:00, Sekhar, Ashwin wrote:
> Hi Jianbo,
>
> Thanks for version v2. Addition of the prefetch instructions is
> definitely helping performance on ThunderX. But still performance is
> slightly less than that of scalar.
>
> I tried few small tweaks which helped improve
A new major release is available:
http://fast.dpdk.org/rel/dpdk-17.05.tar.xz
It is the biggest release ever!
1263 patches from 128 authors
1218 files changed, 121233 insertions(+), 32610 deletions(-)
There are 60 new contributors
(including authors, reviewers and testers):
Add more explanations about vring size changes and different virtio_header
size.
Signed-off-by: Zhiyong Yang
---
Changes in v2:
1. Remove two lines and add "" to constants and struct names.
2. Modify the title.
doc/guides/nics/virtio.rst | 13 -
1 file changed, 8 insertions(+), 5
The NUMA node information for PCI devices provided through
sysfs is invalid for AMD Opteron(TM) Processor 62xx and 63xx
on Red Hat Enterprise Linux 6, and VMs on some hypervisors.
It is good to see more checking for valid values.
Signed-off-by: Tonghao Zhang
---
lib/librte_eal/linuxapp/eal/eal_p
On Wed, May 10, 2017 at 3:14 AM, Mcnamara, John
wrote:
>
>
> > -Original Message-
> > From: Ravi Kerur [mailto:rke...@gmail.com]
> > Sent: Monday, May 8, 2017 10:00 PM
> > To: dev@dpdk.org
> > Cc: Mcnamara, John ; tho...@monjalon.net; Ravi
> > Kerur
> > Subject: [PATCH v2] doc: Merge l3f
Merge l3fwd and l3fwd-acl documentation to reflect the code.
Add examples of LPM and Exact Match rules as text.
Signed-off-by: Ravi Kerur
---
v3:
Incorporate John's comments.
Remove svg files, instead include text format.
v2:
Remove binary PNG from .svg files.
v1:
> > Signed-off-by: Cristian Dumitrescu
>
> It is best not to use “smart quotes” in the text since the renderers do
> that automatically and these characters might interfere with it.
Fixed
> Apart from that:
>
> Acked-by: John McNamara
Applied, thanks
> -Original Message-
> From: Lipiec, Herakliusz
> Sent: Tuesday, May 9, 2017 10:38 PM
> To: Zhang, Helin; Wu, Jingjing
> Cc: dev@dpdk.org
> Subject: [DPDK] net/i40e: add return value checks
>
> Coverity issue: 1379362
> Coverity issue: 1379365
> Fixes: 71d35259ff67 ("i40e: tear down flow
On Wed, May 10, 2017 at 10:05:19PM +0800, Richardson, Bruce wrote:
>
> > -Original Message-
> > From: Bie, Tiwei
> > Sent: Tuesday, May 9, 2017 9:47 AM
> > To: dev@dpdk.org
> > Cc: Richardson, Bruce
> > Subject: [PATCH 0/2] Various fixes for contigmem
> >
> > Tiwei Bie (2):
> > contigm
> >>> The current crypto operation and symmetric crypto operation
> >>> structures will be reworked for correctness and improvement,
> >>> reducing also their sizes, to fit into less cache lines,
> >>> as stated in the following RFC:
> >>>
> >>> http://dpdk.org/dev/patchwork/patch/24011/
> >>>
> >>
05/05/2017 17:32, Declan Doherty:
> On 04/05/2017 4:37 PM, Tomasz Kulasek wrote:
> > API changes are planned for 17.08 to made sessions agnostic to the
> > underlaying devices, removing coupling with crypto PMDs, so a single
> > session can be used on multiple devices.
> >
> > It requires to change
> > Refer to RFC patch - cryptodev: remove crypto device type enumeration
> >
> > It is planned to remove device type enumeration rte_cryptodev_type from
> > library to remove the coupling between crypto PMD and the
> > librte_cryptodev.
> >
> > In this case following stuctures will be changed: r
> >>> The following PMD names definitions will be moved to the individual PMDs
> >>> to remove the coupling between crypto PMDs and the librte_cryptodev:
> >>> CRYPTODEV_NAME_NULL_PMD
> >>> CRYPTODEV_NAME_AESNI_MB_PMD
> >>> CRYPTODEV_NAME_AESNI_GCM_PMD
> >>> CRYPTODEV_NAME_OPENSSL_PMD
> >>> CRYPTOD
11/05/2017 02:25, Thomas Monjalon:
> Some work remains for VDEV bus move.
> Not sure if there will be some API or ABI changes.
> The notice is kept and postponed until the end of this rework.
>
> The PCI fields must be removed from cryptodev and the newly
> introduced eventdev, in order to complet
Some work remains for VDEV bus move.
Not sure if there will be some API or ABI changes.
The notice is kept and postponed until the end of this rework.
The PCI fields must be removed from cryptodev and the newly
introduced eventdev, in order to complete the bus rework.
The VLAN flags rework should
Some VFIO functions have been exported without any rename,
thus no breakage.
The struct eth_driver has been removed, but rte_pci_driver
is still used in rte_cryptodev_driver.
Fixes: a016873eb3c2 ("vfio: export utility functions in map file")
Fixes: 9dca21fb80b6 ("ethdev: remove ethdev driver")
S
> > The deprecation of the bypass functions in the ethdev has not been done in
> > 17.05. Let's postpone to 17.08.
> >
> > Signed-off-by: Bernard Iremonger
>
> Acked-by: John McNamara
Applied, thanks
> The change of _rte_eth_dev_callback_process has not been done in 17.05.
> > Let's postpone to 17.08.
> >
> > Signed-off-by: Bernard Iremonger
>
> Acked-by: John McNamara
Applied, thanks
> > > This is an ABI change notice for DPDK 17.08 in librte_ether
> > > about changes in rte_eth_txmode structure.
> > >
> > > Currently Tx offloads are enabled by default, and can be disabled
> > > using ETH_TXQ_FLAGS_NO* flags. This behaviour is not consistent with
> > > the Rx side where the Rx
10/05/2017 17:46, Gaetan Rivet:
> The PCI and virtual bus are planned to be moved to the generic
> drivers/bus directory in v17.08. For this change to be possible, the EAL
> must be made completely independent.
>
> The rte_devargs structure currently holds device representation internal
> to those
On 05/05/2017 12:54 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
---
v2:
* Convert RTE_DEV_TO_PCI(eth_dev->devi
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
---
v2:
* Convert RTE_DEV_TO_PCI(eth_dev->device) to RTE_ETH_DEV_TO_PCI(eth_dev)
---
drivers/net/ark/ark_ethdev.c
On Wed, May 10, 2017 at 10:54:07AM -0700, Stephen Hemminger wrote:
On Wed, 10 May 2017 17:46:10 +0200
Gaetan Rivet wrote:
The PCI and virtual bus are planned to be moved to the generic
drivers/bus directory in v17.08. For this change to be possible, the EAL
must be made completely independent.
10/05/2017 16:50, De Lara Guarch, Pablo:
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ferruh Yigit
> >
> > Signed-off-by: Ferruh Yigit
>
> Acked-by: Pablo de Lara
Applied, thanks
> > Signed-off-by: Ferruh Yigit
>
> Acked-by: Pablo de Lara
Applied, thanks
> > Update release notes with Mellanox tested platforms.
> >
> > Signed-off-by: Shahaf Shuler
>
> Acked-by: John McNamara
Applied, thanks
09/05/2017 18:31, Mcnamara, John:
> From: Pei, Yulong
> >
> > Add tested Intel platforms with Intel NICs to the release note.
> >
> > Signed-off-by: Yulong Pei
>
> Acked-by: John McNamara
Applied, thanks
10/05/2017 21:01, John McNamara:
> Fix grammar, spelling and formatting of DPDK 17.05 release notes.
>
> Signed-off-by: John McNamara
> Acked-By: Shreyansh Jain
Applied, thanks
10/05/2017 15:51, Mcnamara, John:
> From: On Behalf Of Herakliusz Lipiec
> >
> > Fixes: 35b09d76f89e ("doc: use corelist instead of coremask")
> > Fixes: 58c82067f1ae ("app/testpmd: use consistent vdev names")
> >
> > Signed-off-by: Herakliusz Lipiec
>
> Acked-by: John McNamara
Applied, thank
> -Original Message-
> From: Eads, Gage
> Sent: Wednesday, May 10, 2017 11:41 AM
> To: 'Jerin Jacob' ; Van Haaren, Harry
>
> Cc: dev@dpdk.org; Richardson, Bruce
> Subject: RE: [PATCH 1/3] examples/eventdev_pipeline: added sample app
>
>
>
> > -Original Message-
>
10/05/2017 11:36, Mcnamara, John:
> From: Lu, Wenzhuo
> >
> > Add a better explanation about how to disable MDD (Malicious Driver
> > Detection) in a kernel PF.
> >
> > Signed-off-by: Wenzhuo Lu
>
> Acked-by: John McNamara
Applied, thanks
On 05/10/2017 05:46 PM, Gaetan Rivet wrote:
The PCI and virtual bus are planned to be moved to the generic
drivers/bus directory in v17.08. For this change to be possible, the EAL
must be made completely independent.
The rte_devargs structure currently holds device representation internal
to t
10/05/2017 21:29, Adrien Mazarguil:
> On Wed, May 10, 2017 at 10:30:16AM -0700, Yongseok Koh wrote:
> > In mlx5_dev_start(), the spinlock isn't released when returning error.
> >
> > Fixes: c8d4ee50cc85 ("net/mlx5: fix startup when flow cannot be applied")
> > Cc: sta...@dpdk.org
> >
> > Signed-o
On Wed, May 10, 2017 at 10:30:16AM -0700, Yongseok Koh wrote:
> In mlx5_dev_start(), the spinlock isn't released when returning error.
>
> Fixes: c8d4ee50cc85 ("net/mlx5: fix startup when flow cannot be applied")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Yongseok Koh
Acked-by: Adrien Mazarguil
-sta...@dpdk.org per Thomas
I know this patch has already been applied but to complete the loop, I
have applied the patch to 17.05-rc4 and verified within my application
that the NULL function pointer issue for ixgbe_vf has been resolved.
Regards,
Roger
On 5/10/17 3:00 AM, Wei Dai wrote:
h
> -Original Message-
> From: Shahaf Shuler [mailto:shah...@mellanox.com]
> Sent: Wednesday, May 10, 2017 2:57 PM
> To: Mcnamara, John
> Cc: dev@dpdk.org; tho...@monjalon.net; ol...@mellanox.com
> Subject: [PATCH] doc: update tested platforms with Mellanox NICs
>
> Update release notes w
Fix grammar, spelling and formatting of DPDK 17.05 release notes.
Signed-off-by: John McNamara
Acked-By: Shreyansh Jain
---
V3: Fix mailing list comments.
doc/guides/rel_notes/release_17_05.rst | 208 +++--
1 file changed, 97 insertions(+), 111 deletions(-)
diff -
> -Original Message-
> From: Shreyansh Jain [mailto:shreyansh.j...@nxp.com]
> Sent: Wednesday, May 10, 2017 6:00 AM
> To: Mcnamara, John
> Cc: dev@dpdk.org
> Subject: Re: [PATCH v2] doc: update release notes for 17.05
>
> Hello John,
>
> Some super trivial comments inline:
>
Thanks.
Hey Gaetan,
On Wed, May 10, 2017 at 5:46 PM, Gaetan Rivet wrote:
> The PCI and virtual bus are planned to be moved to the generic
> drivers/bus directory in v17.08. For this change to be possible, the EAL
> must be made completely independent.
>
> The rte_devargs structure currently holds device
On Wed, 10 May 2017 17:46:10 +0200
Gaetan Rivet wrote:
> The PCI and virtual bus are planned to be moved to the generic
> drivers/bus directory in v17.08. For this change to be possible, the EAL
> must be made completely independent.
>
> The rte_devargs structure currently holds device represent
In mlx5_dev_start(), the spinlock isn't released when returning error.
Fixes: c8d4ee50cc85 ("net/mlx5: fix startup when flow cannot be applied")
Cc: sta...@dpdk.org
Signed-off-by: Yongseok Koh
---
drivers/net/mlx5/mlx5_trigger.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/ml
-Original Message-
> Date: Wed, 10 May 2017 17:46:10 +0200
> From: Gaetan Rivet
> To: dev@dpdk.org
> Subject: [dpdk-dev] [PATCH v2] devargs: announce ABI change for device
> parameters
> X-Mailer: git-send-email 2.1.4
>
> The PCI and virtual bus are planned to be moved to the generic
> d
On 10/05/2017 5:07 PM, Boris Pismenny wrote:
5. The addition of inline crypto metadata into the rte_mbuf structure to allow
the required egress metadata to be given to the NIC PMD to build the necessary
transmit descriptors in tx_burst processing when the PKT_TX_IPSEC_INLINE_CRYPTO
is set.
05/05/2017 15:59, Gang Jiang:
> My name is Gang Jiang.
>
> 发件人: "Thomas Monjalon"
> > 16/04/2017 13:39, jiangg:
> > > The return value of strnlen(s, maxlen) is never bigger than maxlen.
> > >
> > > Signed-off-by: jiangg
> >
> > Please, could you provide your full name in SoB?
> >
> > We could
09/05/2017 10:44, Van Haaren, Harry:
> > From: Michal Jastrzebski
> >
> > Coverity reported that an argument for sizeof was used improperly.
> > We should allocate memory for value size that pointer points to,
> > instead of pointer size itself.
> >
> > Coverity issue: 144523, 144521
> > Fixes:
08/05/2017 02:55, Wu, Jingjing:
> From: Gaetan Rivet [mailto:gaetan.ri...@6wind.com]
> >
> > Adds the "all" option to the print-event and mask-event parameters. This
> > option
> > will enable or disable all event notifications from being displayed.
> >
> > Signed-off-by: Gaetan Rivet
> > Cc: J
09/05/2017 10:54, Wu, Jingjing:
> From: Shahaf Shuler [mailto:shah...@mellanox.com]
> >
> > The test assumes the socket ids are contiguous. This is not necessarily the
> > case
> > on all servers and may cause mempool creation to fail.
> >
> > Fixing it by detecting the list of valid socket ids
> -Original Message-
> From: Jerin Jacob [mailto:jerin.ja...@caviumnetworks.com]
> Sent: Wednesday, May 10, 2017 9:12 AM
> To: Van Haaren, Harry
> Cc: dev@dpdk.org; Eads, Gage ; Richardson, Bruce
>
> Subject: Re: [PATCH 1/3] examples/eventdev_pipeline: added sample app
>
> ---
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Bernard Iremonger
> Sent: Tuesday, April 18, 2017 4:49 PM
> To: dev@dpdk.org
> Cc: Mcnamara, John ; Yigit, Ferruh
> ; Iremonger, Bernard
>
> Subject: [dpdk-dev] [PATCH] doc: postpone ABI change in ethdev
>
> Th
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Bernard Iremonger
> Sent: Tuesday, April 18, 2017 4:49 PM
> To: dev@dpdk.org
> Cc: Mcnamara, John ; Yigit, Ferruh
> ; Iremonger, Bernard
>
> Subject: [dpdk-dev] [PATCH] doc: postpone deprecation of functions in
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Bernard Iremonger
> Sent: Tuesday, April 18, 2017 4:49 PM
> To: dev@dpdk.org
> Cc: Mcnamara, John ; Yigit, Ferruh
> ; Iremonger, Bernard
>
> Subject: [dpdk-dev] [PATCH] doc: postpone deprecation of functions in
>
10/05/2017 17:13, Olivier Matz:
> From: Gregory Etelson
>
> The debug assertions when allocating a raw mbuf are not correct since
> commit 8f094a9ac5d7 ("mbuf: set mbuf fields while in pool"),
> which triggers a panic when using this function in debug mode
>
> Change the expected number of segme
Something I have gathered reading through code if it helps, ip_pipeline is
designed such that user side cli (_fe) runs in master pipeline and sends
conf requests to various other packet processing pipelines (_be) which
handle such requests at intervals. These can run on same or different
threads(co
08/05/2017 21:03, Mody, Rasesh:
> Hi Thomas,
>
> > From: Rasesh Mody [mailto:rasesh.m...@cavium.com]
> > Sent: Sunday, May 07, 2017 3:53 PM
> >
> > With the change in base APIs the logic for 100G handling needs to be
> > adjusted to pass cid values instead for queue ids. The current API works
> >
> 5. The addition of inline crypto metadata into the rte_mbuf structure to
> allow the required egress metadata to be given to the NIC PMD to build the
> necessary transmit descriptors in tx_burst processing when the
> PKT_TX_IPSEC_INLINE_CRYPTO is set. We are looking for feedback on a better
10/05/2017 00:10, Adrien Mazarguil:
> On Tue, May 09, 2017 at 01:49:29PM -0700, Yongseok Koh wrote:
> > This patchset is to fix a critical bug which can occur in case of resource
> > deficiency on Tx path. Flowing multi-segment packets can accelerate the
> > occurrence rate.
> >
> > v2:
> > * Spli
On Wed, 2017-05-10 at 17:16 +0200, Olivier Matz wrote:
> It seems it is planned for 17.08, see:
> http://dpdk.org/ml/archives/dev/2017-April/064082.html
indeed
I was mislead by rte_eth_dev_get_supported_ptypes() returning a non NULL
value
there is some logic to return NULL when vector mode is
On Wed, May 10, 2017 at 03:02:51PM +0200, Thomas Monjalon wrote:
10/05/2017 14:52, Gaetan Rivet:
This patch deprecates the function rte_eal_pci_detach in 17.05,
which will be removed in 17.08.
The generic detach bus method should be used instead.
Can we wait to have a new generic detach metho
The PCI and virtual bus are planned to be moved to the generic
drivers/bus directory in v17.08. For this change to be possible, the EAL
must be made completely independent.
The rte_devargs structure currently holds device representation internal
to those two busses. It must be made generic before
09/05/2017 03:15, Lu, Wenzhuo:
> Hi Qi,
>
> From: Zhang, Qi Z
> >
> > There is a bug in previous fix for lsc interrupt.
> > lsc interrupt is not disabled before delayed handler, that cause the delayed
> > handler be re-entered.
> >
> > Fixes: 9b667210700e ("net/ixgbe: fix blocked interrupts")
>
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Bernard Iremonger
> Sent: Tuesday, April 18, 2017 4:49 PM
> To: dev@dpdk.org
> Cc: Mcnamara, John ; Yigit, Ferruh
> ; Iremonger, Bernard
>
> Subject: [dpdk-dev] [PATCH] doc: postpone ABI change in ethdev
>
> The
10/05/2017 14:33, Laurent Hardy:
> Hi Ferruh, Wei, Roger,
> PSB
> I also redo test plan provided through
> http://dpdk.org/dev/patchwork/patch/20502/
>
> Thanks & regards,
> Laurent
>
> On 05/10/2017 02:20 PM, Roger B. Melton wrote:
> >
> >
> > On 5/10/17 5:50 AM, Ferruh Yigit wrote:
> >> On 5/1
Hi Maxime,
On Wed, 10 May 2017 16:27:06 +0200, Maxime Bizon wrote:
> Hello,
>
> With ixgbe driver, it seems mbuf packet_type is always 0 when vPMD code
> (SSE) is enabled (tried 16.11 & master).
>
> doc/guides/nics/ixgbe.rst seems a bit out of date, so I don't know if
> this is expected or not.
From: Gregory Etelson
The debug assertions when allocating a raw mbuf are not correct since
commit 8f094a9ac5d7 ("mbuf: set mbuf fields while in pool"),
which triggers a panic when using this function in debug mode
Change the expected number of segments to 1 instead of 0, and
factorize these san
Hi list,
I'm trying to use the ETH_MQ_RX_VMDQ_RSS mode. The vmdq example is working fine
with VMDQ_ONLY, that is if I change the MAC address and VLAN of my crafted
packets, the RX queue also changes according to the matching pool.
But if I change the mode to VMDQ_RSS, the packet always end up i
Hi Gregory,
On Fri, 5 May 2017 20:23:15 +0300, Gregory Etelson wrote:
> Hello Oliver,
>
> Our application writes data from incoming MBUFs to storage device.
> For performance considerations we use O_DIRECT storage access
> and work in 'zero copy' data mode.
> To achieve the 'zero copy' we MUST a
Hi Jianbo,
Thanks for version v2. Addition of the prefetch instructions is
definitely helping performance on ThunderX. But still performance is
slightly less than that of scalar.
I tried few small tweaks which helped improve performance on my
Thunderx setup. For details see comments inline.
On
10/05/2017 16:42, Olivier Matz:
> Hi,
>
> On Sat, 06 May 2017 10:26:49 +0300, Gregory Etelson wrote:
> > The debug assertions when allocating a raw mbuf are not correct since
> > commit 8f094a9ac5d7 ("mbuf: set mbuf fields while in pool"),
> > which triggers a panic when using this function in de
09/05/2017 10:46, Van Haaren, Harry:
> > From: Kozak, KubaX
> > Sent: Tuesday, May 9, 2017 6:22 AM
> > To: dev@dpdk.org
> > Cc: Van Haaren, Harry ; Jain, Deepak K
> > ;
> > Jastrzebski, MichalX K ; Kozak, KubaX
> >
> > Subject: [PATCH] ethdev: fix wrong sizeof argument in malloc function
> >
>
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ferruh Yigit
> Sent: Wednesday, May 10, 2017 2:34 PM
> To: Thomas Monjalon
> Cc: dev@dpdk.org; Yigit, Ferruh
> Subject: [dpdk-dev] [PATCH v2] maintainers: add drivers features files
>
> Signed-off-by: Ferruh Yigi
Hi,
On Sat, 06 May 2017 10:26:49 +0300, Gregory Etelson wrote:
> The debug assertions when allocating a raw mbuf are not correct since
> commit 8f094a9ac5d7 ("mbuf: set mbuf fields while in pool"),
> which triggers a panic when using this function in debug mode
>
Fixes: 8f094a9ac5d7 ("mbuf: set
09/05/2017 11:45, Jerin Jacob:
> This fixes compiler warnings with GCC 7 for arm64 build.
> Fixes: da8dcc27f644 ("hash: use armv8-a CRC32 instructions")
>
> Signed-off-by: Jerin Jacob
Applied, thanks
On Mon, May 01, 2017 at 09:58:12AM +0300, Shahaf Shuler wrote:
> This is an ABI change notice for DPDK 17.08 in librte_ether
> about changes in rte_eth_txmode structure.
>
> Currently Tx offloads are enabled by default, and can be disabled
> using ETH_TXQ_FLAGS_NO* flags. This behaviour is not con
Hello,
With ixgbe driver, it seems mbuf packet_type is always 0 when vPMD code
(SSE) is enabled (tried 16.11 & master).
doc/guides/nics/ixgbe.rst seems a bit out of date, so I don't know if
this is expected or not.
When vPMD is disabled, either at compile time or runtime, the field is
correctly
Thanks for your review.
> On May 10, 2017, at 8:45 PM, Thomas Monjalon wrote:
>
>> The NUMA node information for PCI devices provided through
>> sysfs is invalid for AMD Opteron(TM) Processor 62xx and 63xx
>> on Red Hat Enterprise Linux 6, and VMs on some hypervisors.
>
> Sorry I don't understa
-Original Message-
> Date: Fri, 21 Apr 2017 10:51:37 +0100
> From: Harry van Haaren
> To: dev@dpdk.org
> CC: jerin.ja...@caviumnetworks.com, Harry van Haaren
> , Gage Eads , Bruce
> Richardson
> Subject: [PATCH 1/3] examples/eventdev_pipeline: added sample app
> X-Mailer: git-send-email
> -Original Message-
> From: Bie, Tiwei
> Sent: Tuesday, May 9, 2017 9:47 AM
> To: dev@dpdk.org
> Cc: Richardson, Bruce
> Subject: [PATCH 0/2] Various fixes for contigmem
>
> Tiwei Bie (2):
> contigmem: free the allocated memory when error occurs
> contigmem: don't zero the pages du
07/05/2017 15:33, Tiwei Bie:
> Tiwei Bie (2):
> eal/bsd: fix ioport write operation
> eal/bsd: fix the read operation on PCI configuration space
>
> lib/librte_eal/bsdapp/eal/eal_pci.c | 28
> 1 file changed, 16 insertions(+), 12 deletions(-)
Applied, thanks
Update release notes with Mellanox tested platforms.
Signed-off-by: Shahaf Shuler
---
doc/guides/rel_notes/release_17_05.rst | 197 +
1 file changed, 197 insertions(+)
diff --git a/doc/guides/rel_notes/release_17_05.rst
b/doc/guides/rel_notes/release_17_05.rst
i
> -Original Message-
> From: stable [mailto:stable-boun...@dpdk.org] On Behalf Of Herakliusz
> Lipiec
> Sent: Wednesday, May 10, 2017 2:39 PM
> To: Doherty, Declan
> Cc: dev@dpdk.org; sta...@dpdk.org
> Subject: [dpdk-stable] [PATCH] doc: update bonding device name in prog
> guide
>
> Fi
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 default.ini for the full list of available PMD features.
> ;
> [Features]
> +Speed capabilities = Y
We should validate this feature only if the
> -Original Message-
> From: Yigit, Ferruh
> Sent: Wednesday, May 10, 2017 2:11 PM
> To: Shepard Siegel ; Ed Czeck
> ; John Miller ;
> Mcnamara, John ; Harish Patil
> ; Rasesh Mody ; Rahul
> Lakkireddy ; Hemant Agrawal
> ; Shreyansh Jain ; Lu,
> Wenzhuo ; Marcin Wojtas ; Michal
> Krawczyk
1 - 100 of 149 matches
Mail list logo
