On 2/19/2016 2:42 PM, Yuanhan Liu wrote:
> On Fri, Feb 19, 2016 at 10:16:42AM +0530, Santosh Shukla wrote:
>> On Tue, Feb 16, 2016 at 8:35 AM, Yuanhan Liu
>> wrote:
>>> On Mon, Feb 15, 2016 at 04:48:36PM +0530, Santosh Shukla wrote:
Hi Yuanhan,
On Mon, Feb 15, 2016 at 4:27 PM, Yuanh
On 2/19/2016 5:05 PM, Ilya Maximets wrote:
> On 19.02.2016 11:36, Xie, Huawei wrote:
>> On 2/19/2016 3:10 PM, Yuanhan Liu wrote:
>>> On Fri, Feb 19, 2016 at 09:32:43AM +0300, Ilya Maximets wrote:
Signed-off-by: Ilya Maximets
---
doc/guides/prog_guide/thread_safety_dpdk_functions.rs
Hi Stephen
I don't think so. I'd ask internally to see if somebody can help on that.
Thanks and sorry!
Regards,
Helin
> -Original Message-
> From: Stephen Hemminger [mailto:stephen at networkplumber.org]
> Sent: Sunday, February 21, 2016 9:07 AM
> To: Zhang, Helin
> Cc: dev at dpdk.org
PCIe feature of 'Extended Tag' is important for 40G performance.
It adds its enabling during each port initialization, to ensure
the high performance.
Signed-off-by: Helin Zhang
---
doc/guides/rel_notes/release_16_04.rst | 6
drivers/net/i40e/i40e_ethdev.c | 65
It enables 'extended tag' for i40e devices only during its port
initialization, which is key for 40G performance. It also deprecates
the similar in igb_uio, and eal lib.
v2:
- Changed the type of return value of i40e_enable_extended_tag() to 'void'.
- Fixed the compile warnings.
- Kept the sys
Remove pci configuration of 'extended tag' and 'max read request
size', as they are not required by all devices and it lets PMD to
configure them if neccessary.
In addition, 'pci_config_space_set()' is deprecated.
Signed-off-by: Helin Zhang
---
config/common_linuxapp | 1 +
lib
It deprecated sys files of 'extended_tag' and
'max_read_request_size', and announced the planned ABI changes of
them.
Signed-off-by: Helin Zhang
---
doc/guides/linux_gsg/enable_func.rst | 47 ---
doc/guides/nics/i40e.rst | 76 +++
On Mon, Feb 22, 2016 at 7:33 AM, Xie, Huawei wrote:
> On 2/19/2016 2:42 PM, Yuanhan Liu wrote:
>> On Fri, Feb 19, 2016 at 10:16:42AM +0530, Santosh Shukla wrote:
>>> On Tue, Feb 16, 2016 at 8:35 AM, Yuanhan Liu
>>> wrote:
On Mon, Feb 15, 2016 at 04:48:36PM +0530, Santosh Shukla wrote:
>
Hi Konstantin,
Many thanks for your review and comments. Sorry for the late response. Please
see inline :)
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Thursday, February 18, 2016 12:45 AM
> To: Lu, Wenzhuo ; dev at dpdk.org
> Subject: RE: [dpdk-dev] [PATCH] ixgbe: fix x550 VF
On Sun, Feb 21, 2016 at 07:47:58PM +0530, Santosh Shukla wrote:
> v9 patchset to support vfio infrasture for ioport, required for archs example
> arm64/arm and x86.
>
>
> For virtio inc_vector patch which is not part of v9..its under review, refer
> [2].
>
> Follow on patch history summary, ref
On Sat, Feb 20, 2016 at 12:12:50PM +0100, Mauricio Vasquez B wrote:
> Section numbers were "hard-coded". This patch adds them as hyperlinks.
Good catch. Thanks.
Acked-by: Yuanhan Liu
--yliu
Hi,
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Jingjing Wu
> Sent: Thursday, January 28, 2016 4:51 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH] doc: add doc for i40e pmd driver introduction
>
> A new doc "i40e.rst" is added to introduce i40e p
> -Original Message-
> From: Zhang, Helin
> Sent: Monday, February 22, 2016 12:00 PM
> To: dev at dpdk.org
> Cc: thomas.monjalon at 6wind.com; Wu, Jingjing; zhe.tag at intel.com; Zhang,
> Helin
> Subject: [PATCH v2 0/3] enable extended tag for i40e
>
> It enables 'extended tag' for i40e
Hi Neil,
Sorry that for my previous misconfiguration of email agent, I missed
this email.
> This looks alot better. One minor comment, the sigbus handler, you should
> probably store the previous bus handler and restore it after you map
> all the
> hugepages you want (lest you overwrite someth
It enables 'extended tag' for i40e devices only during its port
initialization, which is key for 40G performance. It also deprecates
the similar in igb_uio, and eal lib.
v2:
- Changed the type of return value of i40e_enable_extended_tag() to 'void'.
- Fixed the compile warnings.
- Kept the sys
PCIe feature of 'Extended Tag' is important for 40G performance.
It adds its enabling during each port initialization, to ensure
the high performance.
Signed-off-by: Helin Zhang
Acked-by: Jingjing Wu
---
doc/guides/rel_notes/release_16_04.rst | 6
drivers/net/i40e/i40e_ethdev.c |
Remove pci configuration of 'extended tag' and 'max read request
size', as they are not required by all devices and it lets PMD to
configure them if neccessary.
In addition, 'pci_config_space_set()' is deprecated.
Signed-off-by: Helin Zhang
Acked-by: Jingjing Wu
---
config/common_linuxapp
It deprecated sys files of 'extended_tag' and
'max_read_request_size', and announced the planned ABI changes of
them.
Signed-off-by: Helin Zhang
Acked-by: Jingjing Wu
---
doc/guides/linux_gsg/enable_func.rst | 47 ---
doc/guides/nics/i40e.rst | 76 +
On Fri, Feb 19, 2016 at 03:26:36PM +0100, Thomas Monjalon wrote:
> 2016-01-29 12:57, Yuanhan Liu:
> > Introduce vhost_log_write() helper function to log the dirty pages we
> > touched. Page size is harded code to 4096 (VHOST_LOG_PAGE), and each
> > log is presented by 1 bit.
> >
> > Therefore, vho
On Fri, Feb 19, 2016 at 04:01:25PM +0100, Thomas Monjalon wrote:
> 2016-01-29 12:57, Yuanhan Liu:
> > This patch set adds the vhost-user live migration support.
> >
> > The major task behind that is to log pages we touched during
> > live migration, including used vring and desc buffer. So, this
>
> -Original Message-
> From: Wu, Jingjing
> Sent: Wednesday, January 27, 2016 9:50 AM
> To: dev at dpdk.org
> Cc: Wu, Jingjing; Zhang, Helin; Lu, Wenzhuo; Pei, Yulong
> Subject: [PATCH v2 1/2] i40evf: allocate virtchnl cmd buffer for each vf
>
> Currently, i40evf PMD uses a global static
Currently, virtio_dev_info_get() retrieves driver name from pci_drv.
If the driver is virtual PMD, pci_drv will be invalid.
So retrieves the name from eth_dev.
Signed-off-by: Tetsuya Mukawa
---
drivers/net/virtio/virtio_ethdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a
The patch adds below function to cleanup virtio code.
- virtio_dev_check()
Signed-off-by: Tetsuya Mukawa
---
drivers/net/virtio/virtio_ethdev.c | 52 ++
drivers/net/virtio/virtio_ethdev.h | 32 +++
2 files changed, 57 insertions(+), 27 del
The option specifies how to mmap EAL memory.
If the option is specified like '--range-virtaddr=-',
EAL will check /proc/maps, then tries to find free region between addr1
and addr2. If a region is found, EAL will treat it as if 'base-virtaddr'
is specified. Because of this, the option will not work
The option will work with "--range-virtaddr", and if the option is
specified, mapped address will be align by EAL memory size.
Such an alignment is required for using virtio-net PMD extension
on container that uses QEMU QTest framework.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/common/eal
The patch adds a new virtio-net PMD configuration that allows the PMD to
work on host as if the PMD is in VM.
Here is new configuration for virtio-net PMD.
- CONFIG_RTE_VIRTIO_VDEV_QTEST
To use this mode, EAL needs map all hugepages as one file. Also the file
should be mapped between (1 << 31) and
Signed-off-by: Tetsuya Mukawa
---
doc/guides/rel_notes/release_16_04.rst | 3 +++
1 file changed, 3 insertions(+)
diff --git a/doc/guides/rel_notes/release_16_04.rst
b/doc/guides/rel_notes/release_16_04.rst
index 197d4e1..4c6d528 100644
--- a/doc/guides/rel_notes/release_16_04.rst
+++ b/doc/gui
> -Original Message-
> From: Wu, Jingjing
> Sent: Wednesday, January 27, 2016 9:50 AM
> To: dev at dpdk.org
> Cc: Wu, Jingjing; Zhang, Helin; Lu, Wenzhuo; Pei, Yulong
> Subject: [PATCH v2 2/2] i40evf: support interrupt based pf reset request
>
> Interrupt based request of PF reset from P
Hi,
Patch looks good to me although I haven't tested it. Adding Acked-by line.
Nico.
On 18/02/16 11:26, Ferruh Yigit wrote:
> 1- Remove duplicate nb_rx/tx_queues fields from internals
> 2- Move duplicate code into a common function
>
> Signed-off-by: Ferruh Yigit
> Acked-by: Nicolas Pernas Mara
Adding Acked-by line.
On 18/02/16 11:26, Ferruh Yigit wrote:
> 1- Remove duplicate nb_rx/tx_queues fields from internals
>
> Signed-off-by: Ferruh Yigit
> Acked-by: Nicolas Pernas Maradei
> ---
> drivers/net/ring/rte_eth_ring.c | 57
> ++---
> 1 file changed
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Mauricio Vasquez B
> Sent: Saturday, February 20, 2016 11:13 AM
> To: Xie, Huawei ; yuanhan.liu at linux.intel.com
> Cc: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH] doc: fix vhost bad section number references
Adding Acked-by line
On 18/02/16 11:26, Ferruh Yigit wrote:
> 1- remove duplicate nb_rx/tx_queues fields from internals
> 2- remove duplicate numa_node field from internals
>
> Signed-off-by: Ferruh Yigit
> Tested-by: Tetsuya Mukawa
> Acked-by: Tetsuya Mukawa
> Acked-by: Nicolas Pernas Maradei
2016-02-22 15:08, Yuanhan Liu:
> On Fri, Feb 19, 2016 at 04:01:25PM +0100, Thomas Monjalon wrote:
> > 2016-01-29 12:57, Yuanhan Liu:
> > > This patch set adds the vhost-user live migration support.
> > >
> > > The major task behind that is to log pages we touched during
> > > live migration, inclu
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Reshma Pattan
> Sent: Friday, February 12, 2016 2:57 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v2 5/5] doc: update doc for tcpdump feature
>
> Added tcpdump design changes to proc_info section of sam
2016-02-22 02:07, Xie, Huawei:
> On 2/19/2016 5:05 PM, Ilya Maximets wrote:
> > On 19.02.2016 11:36, Xie, Huawei wrote:
> >> On 2/19/2016 3:10 PM, Yuanhan Liu wrote:
> >>> On Fri, Feb 19, 2016 at 09:32:43AM +0300, Ilya Maximets wrote:
> Signed-off-by: Ilya Maximets
> ---
> doc/guid
2016-02-22 09:44, Santosh Shukla:
> On Mon, Feb 22, 2016 at 7:33 AM, Xie, Huawei wrote:
> > On 2/19/2016 2:42 PM, Yuanhan Liu wrote:
> >> On Fri, Feb 19, 2016 at 10:16:42AM +0530, Santosh Shukla wrote:
> >>> On Tue, Feb 16, 2016 at 8:35 AM, Yuanhan Liu
> >>> wrote:
> On Mon, Feb 15, 2016 at
Hi John,
I'll check for other hard-coded references like this, so, I think it is
better to drop this patch.
I'll include the doc name in the reference and send it in a patch series.
On 22 February 2016 at 10:56, Mcnamara, John
wrote:
> > -Original Message-
> > From: dev [mailto:dev-bou
Hi All,
On Wednesday, February 02/03/16, 2016 at 14:02:21 +0530, Rahul Lakkireddy wrote:
> This series of patches extend the flow director filter and add support
> for Chelsio T5 hardware filtering capabilities.
>
> Chelsio T5 supports carrying out filtering in hardware which supports 3
> actions
Hi Declan,
Build fails after make clean.
Fix is following change to include path.
And one unused include file can be removed.
diff --git a/lib/librte_cryptodev/rte_crypto.h
b/lib/librte_cryptodev/rte_crypto.h
index 489314b..5bded98 100644
--- a/lib/librte_cryptodev/rte_crypto.h
+++ b/lib/librte
This patchset contains:
1. Fix variable naming consistency in sample guide
2. Set last_seen time on core when it gets registered
3. An xstats implementation for last-seen and current core status
v2:
-Refactor to remove rte_ethdev.h include
v3:
-Rebase to git
Harry van Haaren (3):
doc: fix keep
This patch fixes some mismatches between the keepalive code
and the docs. Struct names, and descriptions are not in line
with the codebase.
Fixes: e64833f2273a ("examples/l2fwd-keepalive: add sample application")
Signed-off-by: Harry van Haaren
---
doc/guides/sample_app_ug/keep_alive.rst | 19 +
This patch sets a timestamp on each lcore when it is registered
for keepalive. This causes the first values read by the monitor
to show time since the core was registered, instead of the delta
between 0 and the timestamp counter.
Signed-off-by: Harry van Haaren
---
lib/librte_eal/common/rte_keep
This patch adds a function that exposes keepalive statistics
using a rte_keepalive_xstats struct. The function provides
the client API the opportunity to read last-seen and status of
each core.
Signed-off-by: Harry van Haaren
---
doc/guides/rel_notes/release_16_04.rst | 7
doc/gui
The patches will work on below patch series.
- [PATCH v2 0/5] virtio support for container
[Changes]
v3 changes:
- Rebase on latest master.
- remove "-qtest-virtio" option, then add "--range-virtaddr" and
"--align-memsize" options.
- Fix typos in qtest.c
v2 changes:
- Rebase on above patc
On 02/19/2016 09:17 PM, Eric Kinzie wrote:
> From: Eric Kinzie
>
> Provide functions to allow an external 802.3ad state machine to transmit
> and recieve LACPDUs and to set the collection/distribution flags on
> slave interfaces.
>
> Signed-off-by: Eric Kinzie
> Signed-off-by: Stephen Hemminger
For some reason git apply choked on the patches but patch -p1 applied
them fine. Guessing Thunderbird was adding a few rogue newlines..
On 22/02/2016 11:25, Harry van Haaren wrote:
> Harry van Haaren (3):
>doc: fix keepalive sample app guide
>eal: add keepalive core register timestamp
>
> EAL should not depend on the PQoS library.
> Please could you hook the command line parsing and help to add these new
> options from outside?
Thanks for the feedback - we heard about dependency problem from a few sources.
> The command line options may be convenient in some cases but not
> stra
On Mon, Feb 22, 2016 at 02:08:51PM +0800, Tan, Jianfeng wrote:
> Hi Neil,
>
> Sorry that for my previous misconfiguration of email agent, I missed this
> email.
>
> >This looks alot better. One minor comment, the sigbus handler, you should
> >probably store the previous bus handler and restore i
> -Original Message-
> From: Iremonger, Bernard
> Sent: Wednesday, February 10, 2016 10:14 AM
> To: dev at dpdk.org
> Cc: Doherty, Declan ; Iremonger, Bernard
>
> Subject: [PATCH 0/2] bonding fixes
>
> These patches fix segmentation faults which were occurring when
> slave devices were de
2016-02-22 13:12, Kantecki, Tomasz:
> Off the mail list, we received feedback that extending EAL commands is not an
> option due to Intel Architecture nature of CAT.
> Consequently, we changed direction when working on V2 PQoS patches in roder
> to:
> - reduce DPDK PQoS code print and reuse exis
> -Original Message-
> From: Chen, Jing D
> Sent: Saturday, February 6, 2016 6:48 AM
> To: dev at dpdk.org
> Cc: Mcnamara, John ; Chen, Jing D
>
> Subject: [PATCH v2] doc: add Vector FM10K introductions
>
> From: "Chen Jing D(Mark)"
>
> Add introductions on how to enable Vector FM10K Rx
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Wang Xiao W
> Sent: Thursday, February 4, 2016 3:39 AM
> To: Chen, Jing D
> Cc: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v3 3/3] doc: update release note for fm10k FTAG
> support
>
> Update the release note
In order to cleanup the configuration files some and reduce
the number of duplicate configuration information. Add a new
file called common_base which contains just about all of the
configuration lines in one place. Then have the common_bsdapp,
common_linuxapp files include this one file. Then in t
On Sat, Feb 20, 2016 at 05:07:08PM -0800, Stephen Hemminger wrote:
> Is anyone working on making KNI conform to current kernel best practices
> and attempt to get it upstream? After 2 minute code review I already
> see lots and lots of things that need work.
Hi Stephen,
I am working on not KNI,
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Wang Xiao W
> Sent: Thursday, February 4, 2016 3:39 AM
> To: Chen, Jing D
> Cc: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v3 2/3] doc: add introduction for fm10k FTAG
> based forwarding
>
> Add a brief intro
This patch adds CPU utilization measurement and idle cycle rate
computation to packet framework. The measurement is done by measuring
the cycles spent while a thread pulls zero packet from RX queue. These
cycles are treated as idle cycles (or headroom). A CLI command is added
to display idle cycle
On Mon, Feb 22, 2016 at 10:56:42AM +0100, Thomas Monjalon wrote:
> > > Patches 2 and 3 have been merged to avoid a compilation error.
> > > Applied, thanks
> >
> > Actually, there was a ongoing discussion about patch 6, the handling
> > of VHOST_USER_SEND_RARP request:
> >
> > http://dpdk.org/m
Broadcast RARP packet by injecting it to receiving mbuf array at
rte_vhost_dequeue_burst().
Commit 33226236a35e ("vhost: handle request to send RARP") iterates
all host interfaces and then broadcast it by all of them. It did
notify the switches about the new location of the migrated VM, however,
On 2/4/2016 1:24 AM, Olivier MATZ wrote:
> Hi,
>
> On 01/27/2016 02:56 PM, Panu Matilainen wrote:
>>
>> Since rte_pktmbuf_alloc_bulk() is an inline function, it is not part of
>> the library ABI and should not be listed in the version map.
>>
>> I assume its inline for performance reasons, but then
Hi Keith,
What makes a param common?
e.g. cryptodev QAT PMD is supported in linux, but currently not supported in
bsd.
So typically I disable it in the bsd file and enable it in the linux file.
Couldn't the same apply to any other parameter, i.e. there may be users who
want to have difference
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Jingjing Wu
> Sent: Thursday, January 28, 2016 8:51 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH] doc: add doc for i40e pmd driver introduction
>
> A new doc "i40e.rst" is added to introduce i40e pmd dr
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Tetsuya Mukawa
> Sent: Monday, February 22, 2016 8:18 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v3 6/6] docs: add release note for qtest virtio
> container support
> ...
>
> +* **Virtio support for c
>Hi Keith,
>
>What makes a param common?
>
>e.g. cryptodev QAT PMD is supported in linux, but currently not supported in
>bsd.
>So typically I disable it in the bsd file and enable it in the linux file.
>
>Couldn't the same apply to any other parameter, i.e. there may be users who
>want to have
Hi,
Sorry I missed the last few messages until now. I'm happy with just
removing the "if". Kyle, when you say you fixed it, do you mean that you
will push the patch or have already done so ?
Thanks,
Tom
On 02/18/2016 02:03 PM, Kyle Larose wrote:
> On Tue, Jan 5, 2016 at 2:13 AM,
On 02/22/2016 06:02 PM, Wiles, Keith wrote:
>> Hi Keith,
>>
>> What makes a param common?
>>
>> e.g. cryptodev QAT PMD is supported in linux, but currently not supported
>> in bsd.
>> So typically I disable it in the bsd file and enable it in the linux file.
>>
>> Couldn't the same apply to any o
>
>2016-02-20 07:40, Harish Patil:
>> This patch set introduces DPDK poll mode driver for new QLogic FastLinQ
>>QL4
>> 25G/40G capable family of CNAs as well as their SR-IOV Virtual
>>Functions (VF).
>>
>> The overall PMD driver design includes a common module called ecore
>>that deals
>> with
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Harish Patil
> Sent: Saturday, February 20, 2016 3:40 PM
> To: dev at dpdk.org
> Cc: Sony Chacko
> Subject: [dpdk-dev] [PATCH 2/6] qede: add documentation
>
> Signed-off-by: Harish Patil
> Signed-off-by: Rases
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Harry van Haaren
> Sent: Monday, February 22, 2016 11:26 AM
> To: Horton, Remy
> Cc: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v3 1/3] doc: fix keepalive sample app guide
>
> This patch fixes some mismatches
> -Original Message-
> From: Van Haaren, Harry
> Sent: Tuesday, February 16, 2016 1:40 PM
> To: Mcnamara, John
> Cc: dev at dpdk.org; Van Haaren, Harry
> Subject: [PATCH] doc/linux gsg: add gcc-multilib as package hint
>
> When compiling for i686 targets compilation could fail if the 32b
> -Original Message-
> From: Van Haaren, Harry
> Sent: Tuesday, February 9, 2016 3:20 PM
> To: Chen, Jing D
> Cc: Mcnamara, John ; dev at dpdk.org; Van Haaren,
> Harry
> Subject: [PATCH v2] docs: add statistics read frequency to fm10k guide
>
> This patch documents that the statistics of
> Known Issues
>
>
> +* **FM10K: Statistics Polling Frequency**
> +
> + A section has been added to the NIC guide for fm10k about the maximum
> + time between reading statistics and 32 bit packet counter overflows.
Hi Thomas,
We should probably call this section "New Known Issues
2016-02-22 16:47, Harish Patil:
> >Please could you share some performance numbers?
>
> We have measured ~68 MPPS @ 64B with zero drop for the 4x25G adapter
> running bi-di RFC traffic.
How many queues/cores to achieve this performance?
> >What is the status about the integration of this driver
> +
> + Assign MAC address to the VF using iproute2 utility. The syntax is:
> + ip link set vf mac
Also, make this a code/console section by adding :: to the previous line
and indenting 3-4 spaces.
2016-02-22 17:18, Mcnamara, John:
> > Known Issues
> >
> >
> > +* **FM10K: Statistics Polling Frequency**
> > +
> > + A section has been added to the NIC guide for fm10k about the maximum
> > + time between reading statistics and 32 bit packet counter overflows.
>
> Hi Thomas,
>
To preserve compatibility with Mellanox OFED 3.1, flow director and RX VLAN
stripping code is only enabled if compiled with 3.2.
Changes in v2:
- Rebased patchset on top of dpdk-next-net/rel_16_04.
- Fixed trivial compilation warnings (positive errnos are left on purpose).
- Updated documentation
From: Yaacov Hazan
Merge redundant code by adding a static initialization table to manage
promiscuous and allmulticast (special) flows.
New function priv_rehash_flows() implements the logic to enable/disable
relevant flows in one place from any context.
Signed-off-by: Yaacov Hazan
Signed-off-b
From: Yaacov Hazan
Until now, broadcast frames were handled like unicast. Moving the related
flow to the special flows table frees up the related unicast MAC entry.
The same method is used to handle IPv6 multicast frames.
Signed-off-by: Yaacov Hazan
Signed-off-by: Adrien Mazarguil
---
driver
From: Yaacov Hazan
Upcoming flow director support will reuse this function to generate filter
rules.
Signed-off-by: Yaacov Hazan
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx5/mlx5_mac.c| 4 ++--
drivers/net/mlx5/mlx5_rxmode.c | 5 +++--
drivers/net/mlx5/mlx5_rxq.c| 16 +++
From: Yaacov Hazan
Add support for flow director filters (RTE_FDIR_MODE_PERFECT and
RTE_FDIR_MODE_PERFECT_MAC_VLAN modes).
This feature requires MLNX_OFED >= 3.2.
Signed-off-by: Yaacov Hazan
Signed-off-by: Adrien Mazarguil
Signed-off-by: Raslan Darawsheh
---
doc/guides/nics/mlx5.rst
From: Yaacov Hazan
Allows HW to strip the 802.1Q header from incoming frames and report it
through the mbuf structure.
This feature requires MLNX_OFED >= 3.2.
Signed-off-by: Yaacov Hazan
Signed-off-by: Adrien Mazarguil
---
doc/guides/nics/mlx5.rst | 2 +
doc/guides/rel_notes/
This patchset improves the mlx5 PMD performance by doing better prefetching,
by reordering internal structure fields and by removing a few unnecessary
operations.
Note: should be applied after "Add flow director and RX VLAN stripping
support" to avoid conflicts.
Changes in v2:
- Rebased patchset
From: Nelio Laranjeiro
This change improves performance noticeably.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_rxtx.c | 16 +++-
1 file changed, 15 insertions(+), 1 deletion(-)
diff --git a/drivers/net/mlx5/mlx5_rxtx.c b/drivers/net/mlx5/mlx5_rxtx.c
index 7585570..b
From: Nelio Laranjeiro
Remove padding and move important fields to the beginning for better
performance.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_rxtx.h | 31 ---
1 file changed, 16 insertions(+), 15 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_
From: Nelio Laranjeiro
Avoid dereferencing pointers twice to get to fast Verbs functions by
storing them directly in RX/TX queue structures.
Signed-off-by: Nelio Laranjeiro
Signed-off-by: Yaacov Hazan
---
drivers/net/mlx5/Makefile| 1 +
drivers/net/mlx5/mlx5_rxq.c | 16
From: Nelio Laranjeiro
Improve performance by processing offloads only when requested by the
application.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_rxtx.c | 29 -
1 file changed, 16 insertions(+), 13 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_r
From: Nelio Laranjeiro
Improves performance as the lkey is not needed by hardware in this case.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_rxtx.c | 23 +--
1 file changed, 13 insertions(+), 10 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_rxtx.c b/drivers/
From: Nelio Laranjeiro
This lowers the amount of cache misses.
Signed-off-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_rxtx.c | 35 ---
1 file changed, 20 insertions(+), 15 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_rxtx.c b/drivers/net/mlx5/mlx5_rxtx.c
The first and last memory pool elements are usually cache-aligned but not
page-aligned, particularly when using huge pages.
Hardware performance can be improved significantly by registering memory
regions starting and ending on page boundaries.
Signed-off-by: Adrien Mazarguil
---
drivers/net/ml
This patchset addresses several minor issues, release notes are updated
accordingly.
Note: should be applied after "Performance optimizations for mlx5 and mlx4".
Adrien Mazarguil (3):
mlx5: manage all special flow types at once
mlx5: remove redundant debugging message
mlx5: apply VLAN filte
From: Or Ami
RSS configuration should not be freed when priv is NULL.
Fixes: 2f97422e7759 ("mlx5: support RSS hash update and get")
Signed-off-by: Or Ami
---
doc/guides/rel_notes/release_16_04.rst | 4
drivers/net/mlx5/mlx5.c| 6 --
2 files changed, 8 insertions(+), 2
From: Or Ami
If the port link layer is not Ethernet, notify the user.
Signed-off-by: Or Ami
---
doc/guides/rel_notes/release_16_04.rst | 5 +
drivers/net/mlx5/mlx5.c| 7 +++
2 files changed, 12 insertions(+)
diff --git a/doc/guides/rel_notes/release_16_04.rst
b/doc/gu
This commit adds helpers to remove redundant code.
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx5/mlx5.c | 5 +
drivers/net/mlx5/mlx5.h | 2 ++
drivers/net/mlx5/mlx5_rxmode.c | 40
drivers/net/mlx5/mlx5_trigger.c | 10 ++--
Signed-off-by: Adrien Mazarguil
---
drivers/net/mlx5/mlx5_rxmode.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_rxmode.c b/drivers/net/mlx5/mlx5_rxmode.c
index bcf4231..730527e 100644
--- a/drivers/net/mlx5/mlx5_rxmode.c
+++ b/drivers/net/mlx5/mlx5_rxmode.c
@@ -204,8
Unlike promiscuous and allmulticast flows, those should remain
VLAN-specific.
Signed-off-by: Adrien Mazarguil
---
doc/guides/rel_notes/release_16_04.rst | 4 ++
drivers/net/mlx5/mlx5_rxmode.c | 105 +
drivers/net/mlx5/mlx5_rxq.c| 5 +-
driv
From: Yaacov Hazan
Change rxq_cq_to_ol_flags() to set checksum flags according to packet type,
so for non L3/L4 packets the mbuf chksum_bad flags will not be set.
Fixes: 67fa62bc672d ("mlx5: support checksum offload")
Signed-off-by: Yaacov Hazan
---
doc/guides/rel_notes/release_16_04.rst | 4
From: Robin Jarry
In the documentation it is specified that the hardware only supports a
number of RX queues if it is a power of 2.
Since ibv_exp_create_qp may not return an error when the number of
queues is unsupported by hardware, sanitize the value in dev_configure.
Signed-off-by: Robin Jar
This patchset adds to mlx5 a few features available in mlx4 (TX from
secondary processes) or provided by Verbs (support for HW packet padding,
TX VLAN insertion).
Release notes and documentation are updated accordingly.
Note: should be applied after "Assorted fixes for mlx4 and mlx5".
Olga Shern
From: Or Ami
Burst functions are updated to make sure applications cannot attempt to
send/receive after link is brought down.
Signed-off-by: Or Ami
---
doc/guides/rel_notes/release_16_04.rst | 4 ++
drivers/net/mlx5/mlx5.c| 2 +
drivers/net/mlx5/mlx5.h| 2 +
From: Or Ami
Secondary processes are expected to use queues and other resources
allocated by the primary, however Verbs resources can only be shared
between processes when inherited through fork().
This limitation can be worked around for TX by configuring separate queues
from secondary processe
From: Olga Shern
Environment variable MLX5_PMD_ENABLE_PADDING enables HW packet padding
in PCI bus transactions.
When packet size is cache aligned and CRC stripping is enabled, 4 fewer
bytes are written to the PCI bus. Enabling padding makes such packets
aligned again.
In cases where PCI bandwi
1 - 100 of 116 matches
Mail list logo
