From: Amruta Zende
Signed-off-by: Amruta Zende
Signed-off-by: David Hunt
---
MAINTAINERS|5 +
config/defconfig_arm-native-linuxapp-gcc | 56
.../common/include/arch/arm/rte_atomic.h | 269
.../common/i
This patch provides EAL support for the ARMv7 architecture. We hope that this
will encourage the ARM community to contribute PMDs for their SoCs to DPDK.
For now, we've added Intel engineers to the MAINTAINERS file. We would like to
encourage the ARM community to take over maintenance of this ar
On Fri, Oct 02, 2015 at 05:00:14PM +0300, Michael S. Tsirkin wrote:
> On Thu, Oct 01, 2015 at 02:02:24PM -0700, Alexander Duyck wrote:
> > validation and translation would add 10s if not 100s of nanoseconds to the
> > time needed to process each packet. In addition we are talking about doing
> > t
Hi Rahul,
Rahul Lakkireddy writes:
> Update sge initialization with respect to free-list manager configuration
> and ingress arbiter. Also update refill logic to refill mbufs only after
> a certain threshold for rx. Optimize tx packet prefetch and free.
<>
> for (i = 0; i
Add a sample application that acts as a PTP slave using the
DPDK ieee1588 functions.
Signed-off-by: Daniel Mrzyglod
---
MAINTAINERS| 3 +
examples/Makefile | 1 +
examples/ptpclient/Makefile| 59 +
examples/ptpclient/ptpclient.c | 525 +++
Add additional functions to support the existing IEEE1588
functionality and to enable getting, setting and adjusting
the device time.
Signed-off-by: Daniel Mrzyglod
---
drivers/net/ixgbe/ixgbe_ethdev.c | 250 +--
drivers/net/ixgbe/ixgbe_ethdev.h | 24
2
Add additional functions to support the existing IEEE1588
functionality.
* rte_eth_timesync_settime(), function to set the device clock time.
* rte_eth_timesync_gettime, function to get the device clock time.
* rte_eth_timesync_adjust, function to adjust the device clock time.
Signed-off-by: Dani
Add a sample application that acts as a PTP slave using the DPDK IEEE1588
functions.
Also add some additional IEEE1588 support functions to enable getting,
setting and adjusting the device time.
Some V1 limitations of the app:
* The mater clock sequence id and clock id are not verified fully.
*
Since freeing a ring is now possible, then when freeing
a hash table, its internal ring can be freed as well.
Therefore when a new table, with the same name as a previously
freed table, is created, there is no need to look up
the already allocated ring.
Signed-off-by: Pablo de Lara
---
This patc
On Thu, Oct 01, 2015 at 02:02:24PM -0700, Alexander Duyck wrote:
> validation and translation would add 10s if not 100s of nanoseconds to the
> time needed to process each packet. In addition we are talking about doing
> this in kernel space which means we wouldn't really be able to take
> advanta
From: "De Lara Guarch, Pablo"
When creating a ring, a memzone is created to allocate it in memory,
but the ring could not be freed, as memzones could not be.
Since memzones can be freed now, then rings can be as well,
taking into account if they were initialized using pre-allocated memory
(in wh
On Thu, Oct 01, 2015 at 02:17:49PM -0700, Alexander Duyck wrote:
> On 10/01/2015 02:42 AM, Michael S. Tsirkin wrote:
> >On Thu, Oct 01, 2015 at 12:22:46PM +0300, Avi Kivity wrote:
> >>even when they are some users
> >>prefer to avoid the performance penalty.
> >I don't think there's a measureable p
- Add a missed step to mount huge pages in Linux.
- Re-structure Sample Application Notes.
- Add Jumbo Frame support to list of supported features and instructions
on how to enable it via testpmd.
- Update release notes.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Kumar Sanghvi
---
doc/gui
Add a mtu_set() eth_dev_ops to allow DPDK apps to modify device mtu.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Kumar Sanghvi
---
drivers/net/cxgbe/cxgbe_ethdev.c | 29 +
1 file changed, 29 insertions(+)
diff --git a/drivers/net/cxgbe/cxgbe_ethdev.c b/drivers/ne
Ensure jumbo mode is enabled and that the mbuf data room size can
accommodate jumbo size. If the mbuf data room size can't accommodate
jumbo size, chain mbufs to jumbo size.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Kumar Sanghvi
---
drivers/net/cxgbe/sge.c | 58 ++
Add a non-coalesce path. Skip coalescing for Jumbo Frames, and send the
packet through non-coalesced path if there are enough credits. Also,
free these non-coalesced packets while reclaiming credits.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Kumar Sanghvi
---
drivers/net/cxgbe/sge.c | 96
Increase max_rx_pktlen to accommodate jumbo frame size. Perform sanity
checks and enable jumbo mode in rx queue setup. Set link mtu based on
max_rx_pktlen.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Kumar Sanghvi
---
drivers/net/cxgbe/cxgbe.h| 3 +++
drivers/net/cxgbe/cxgbe_ethdev.
Update sge initialization with respect to free-list manager configuration
and ingress arbiter. Also update refill logic to refill mbufs only after
a certain threshold for rx. Optimize tx packet prefetch and free.
Approx. 4 MPPS improvement seen in forwarding performance after the
optimization.
S
This series of patches improve forwarding performance for Chelsio T5 40GbE
cards and add Jumbo Frame support for cxgbe pmd. Also update documentation
and release notes.
Rahul Lakkireddy (6):
cxgbe: Optimize forwarding performance for 40G
cxgbe: Update device info and perform sanity checks to e
> -Original Message-
> From: Charles (Chas) Williams [mailto:3chas3 at gmail.com]
>
> On Fri, 2015-10-02 at 16:18 +0100, Bruce Richardson wrote:
> > On Fri, Oct 02, 2015 at 11:00:07AM -0400, Chas Williams wrote:
> > > If a system is using deterministic interface names, it may be easier
> >
Hello,
I was wondering if anybody has tested and managed to get working the following
scenario:
Host with 2 NIC: 1 Niantic 82599 and 1 Fortville i40e, both NICs configured
with 8 VFs. There are two VMs running on this host, each VM has two SRIOV
interfaces 1 for each type (Niantic based VF an
Hi,
Working with the patchset to include new features to make install,
some questions I missed to ask before:
for example if you use only "make install":
"T" variable is going to get "*" value and the
makefiles are going to build all TARGETS,
Is it an expected behavior?,
and if you use "make i
Anwser for [PATCH v3 8/8] mk: Add rule for installing runtime files
Hi Panu,
Thank you for taking time in this revision :),. In this patchset
I?ve tried to keep current behavior (make install) untouched, I mean
thye don't affect the current makefile rules and they work like "new feature
Hi, Panu and Bruce, sounds good your suggestion :) , then I'm going
to change the patch for installing dpdk_nic_bind.py and cpu_layout.py in
/usr/bin
Thanks.
Mario.
From: Richardson, Bruce
Sent: Friday, October 02, 2015 3:54 AM
To: Panu Matilainen; Areva
Hi, Panu, perfect thank you for your feedback, I going to change that
path by /lib/modules/$(uname -r)/extra/ in my patch.
Thanks.
Mario.
From: Panu Matilainen [pmati...@redhat.com]
Sent: Friday, October 02, 2015 3:38 AM
To: Arevalo, Mario Alfredo C; d
On Fri, Oct 02, 2015 at 11:00:07AM -0400, Chas Williams wrote:
> If a system is using deterministic interface names, it may be easier in
> some cases to use the interface name to blacklist an interface.
>
Is it possible to do this using the existing arguments, i.e. have the -b flag
detect if it's
On Fri, Oct 02, 2015 at 03:01:25PM +0100, Pablo de Lara wrote:
> From: "Pablo de Lara"
>
> When creating a ring, a memzone is created to allocate it in memory,
> but the ring could not be freed, as memzones could not be.
>
> Since memzones can be freed now, then rings can be as well,
> taking in
On Fri, Oct 02, 2015 at 05:00:14PM +0300, Michael S. Tsirkin wrote:
> On Thu, Oct 01, 2015 at 02:02:24PM -0700, Alexander Duyck wrote:
> > validation and translation would add 10s if not 100s of nanoseconds to the
> > time needed to process each packet. In addition we are talking about doing
> > t
From: "Pablo de Lara"
When creating a ring, a memzone is created to allocate it in memory,
but the ring could not be freed, as memzones could not be.
Since memzones can be freed now, then rings can be as well,
taking into account if they were initialized using pre-allocated memory
(in which case
On Fri, 2015-10-02 at 16:44 +, Richardson, Bruce wrote:
> > -Original Message-
> > From: Charles (Chas) Williams [mailto:3chas3 at gmail.com]
> >
> > On Fri, 2015-10-02 at 16:18 +0100, Bruce Richardson wrote:
> > > On Fri, Oct 02, 2015 at 11:00:07AM -0400, Chas Williams wrote:
> > > >
Hi Bruce,
> -Original Message-
> From: Richardson, Bruce
> Sent: Friday, October 02, 2015 3:10 PM
> To: De Lara Guarch, Pablo
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3] ring: add function to free a ring
>
> On Fri, Oct 02, 2015 at 03:01:25PM +0100, Pablo de Lara wrote:
> >
On 10/01/2015 03:11 AM, Mario Carrillo wrote:
> Add hierarchy-file support to the DPDK libraries, modules,
> binary files, nic bind files and documentation,
> when invoking "make install-fhs" (filesystem hierarchy standard)
> runtime files will be by default installed in:
> $(DESTDIR)/$(BIN_DIR) wh
On 10/01/2015 03:11 AM, Mario Carrillo wrote:
> Add hierarchy-file support to the DPDK nic bind files,
> when invoking "make install-sbin" nic bind files will
> be installed by default in: $(DESTDIR)/$(SBIN_DIR)
> where SBIN_DIR=/usr/sbin/dpdk_nic_bind by default,
> you can override SBIN_DIR var.
>
Make a distniction between dropped packets and error statistics to allow
a higher level fault management entity to interact with DPDK and take
appropriate measures when errors are detected. It will also provide
valuable information for any applications that collects/extracts DPDK
stats, such applic
On 10/01/2015 03:11 AM, Mario Carrillo wrote:
> Add hierarchy-file support to the DPDK modules,
> when invoking "make install-mod" modules will be
> installed in: $(DESTDIR)/$(KERNEL_DIR)
> if RTE_EXEC_ENV=linuxapp then
> KERNEL_DIR=/lib/modules/$(uname -r)/build
/lib/modules/$(uname -r)/build is
On Fri, 2015-10-02 at 16:18 +0100, Bruce Richardson wrote:
> On Fri, Oct 02, 2015 at 11:00:07AM -0400, Chas Williams wrote:
> > If a system is using deterministic interface names, it may be easier in
> > some cases to use the interface name to blacklist an interface.
> >
>
> Is it possible to do t
On Fri, Oct 02, 2015 at 02:15:29PM +0300, Panu Matilainen wrote:
> On 10/01/2015 03:11 AM, Mario Carrillo wrote:
> >Add hierarchy-file support to the DPDK libraries, modules,
> >binary files, nic bind files and documentation,
> >when invoking "make install-fhs" (filesystem hierarchy standard)
> >ru
From: "Alejandro.Lucero"
Signed-off-by: Alejandro.Lucero
Signed-off-by: Rolf.Neugebauer
---
tools/dpdk_nic_bind.py |8 ++--
tools/setup.sh | 122 ++--
2 files changed, 101 insertions(+), 29 deletions(-)
diff --git a/tools/dpdk_nic_bind.
From: "Alejandro.Lucero"
Current Netronome's PMD just supports Virtual Functions. Future Physical
Function support will require specific Netronome code here.
Signed-off-by: Alejandro.Lucero
Signed-off-by: Rolf.Neugebauer
---
lib/librte_eal/common/include/rte_pci.h |1 +
lib/librte_eal/l
From: "Alejandro.Lucero"
Signed-off-by: Alejandro.Lucero
Signed-off-by: Rolf.Neugebauer
---
config/common_linuxapp |6 +
doc/guides/nics/nfp.rst | 248
drivers/net/Makefile |1 +
drivers/net/nfp/Makefile | 88 ++
drivers/net/nfp/nfp_net.c
From: "Alejandro.Lucero"
Alejandro.Lucero (3):
This patch adds a PMD driver for Netronome NFP PCI cards.
This patch adds a new UIO driver for Netronome NFP PCI cards.
Modifying configuration scripts for Netronome's nfp_uio driver.
config/common_linuxapp|6 +
doc/gu
Ping.
On 17.09.2015 08:55, Ilya Maximets wrote:
> Ok. Thank you. I'll wait.
>
> On 16.09.2015 18:37, Wiles, Keith wrote:
>> Thanks the patch looks fine, but I have not had a lot of time to review it
>> detail. I hope to get to it next week after I return back home.
>>
>> On 9/16/15, 2:09 AM, "Ily
I looked at the code and everything looks good. I will try to merge the
code next week as I am traveling again :-(
Thanks for the patch, I am glad you found this problem as I believe
someone else reported something odd in that area, but was not able to give
me many details.
?
Regards,
++Keith Wil
If a system is using deterministic interface names, it may be easier in
some cases to use the interface name to blacklist an interface.
Signed-off-by: Chas Williams <3chas3 at gmail.com>
---
app/test/test_devargs.c | 2 ++
lib/librte_eal/common/eal_common_devargs.c | 8
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Panu Matilainen
> Sent: Friday, October 2, 2015 11:50 AM
> To: Arevalo, Mario Alfredo C; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 6/8] mk: Add rule for installing nic
> bind files
>
> On 10/01/2015 03
2015-10-01 15:28, St Leger, Jim:
> When can we expect the main website (including home page http://dpdk.org/) to
> be updated?
As there was no comment about the demo page, it is now applied to every pages.
> Have we opened up the website to allow the community to edit it? (I think
> this has be
add PCI Port Hotplug support to the xenvirt PMD
This patch depends on 4 patches from the following patch set:
-remove-pci-driver-from-vdevs.patch
0001-librte_eal-add-RTE_KDRV_NONE-for-vdevs.patch
0002-librte_ether-add-fields-from-rte_pci_driver-to-rte_e.patch
0003-librte_ether-add-function
Signed-off-by: Bernard Iremonger
---
examples/vhost_xen/main.c | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/examples/vhost_xen/main.c b/examples/vhost_xen/main.c
index 5d20700..d124be1 100644
--- a/examples/vhost_xen/main.c
+++ b/examples/vhost_xen/main.c
@@ -1,7 +1
Signed-off-by: Bernard Iremonger
---
drivers/net/xenvirt/rte_eth_xenvirt.c | 24 ++--
1 file changed, 22 insertions(+), 2 deletions(-)
diff --git a/drivers/net/xenvirt/rte_eth_xenvirt.c
b/drivers/net/xenvirt/rte_eth_xenvirt.c
index 8923826..1bf35b7 100644
--- a/drivers/net/x
Signed-off-by: Bernard Iremonger
---
drivers/net/xenvirt/rte_eth_xenvirt.c | 63 +++
drivers/net/xenvirt/rte_xen_lib.c | 26 ---
drivers/net/xenvirt/rte_xen_lib.h | 5 ++-
3 files changed, 83 insertions(+), 11 deletions(-)
diff --git a/drivers
On 10/02/2015 07:00 AM, Michael S. Tsirkin wrote:
> On Thu, Oct 01, 2015 at 02:02:24PM -0700, Alexander Duyck wrote:
>> validation and translation would add 10s if not 100s of nanoseconds to the
>> time needed to process each packet. In addition we are talking about doing
>> this in kernel space w
All of this information is in shared memory, is it not? For example,
you could patch the ring library to give a programmable interface to
the following function:
http://dpdk.org/doc/api/rte__ring_8h.html#a7bfcef0ad324fcc4c03bcb59cd7e867f.
This would allow you to see the full set of rings in a proce
Can you improve the comments on these counters? If you didn't happen to
follow this thread, there's no way to reasonably figure out what the
difference is from looking at the code without chasing it all the way down
and cross-referencing the NIC datasheet.
Thanks,
Jay
On Fri, Oct 2, 2015 at 7:47
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
> Sent: Friday, October 2, 2015 12:39 AM
> To: Wu, Jingjing
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 00/52] update i40e base driver
>
> 2015-09-09 01:53, Zhang, Helin:
> > Acked-by:
2015-09-09 01:53, Zhang, Helin:
> Acked-by: Helin Zhang
Applied, thanks.
Some titles were fixed and some patches are squashed or reordered.
Maybe it deserves an entry in the release notes.
55 matches
Mail list logo