> -Original Message-
> From: Kulasek, TomaszX
> Sent: Tuesday, February 09, 2016 5:03 PM
> To: Ananyev, Konstantin; dev at dpdk.org
> Subject: RE: [dpdk-dev] [PATCH 1/2] ethdev: add buffered tx api
>
>
>
> > -Original Message-
> > From: Ananyev, Konstantin
> > Sent: Tuesday,
2016-02-09 21:15, Liming Sun:
> Looks like this patch serie has been merged into dpdk-next-net/rel_16_04.
> What would be the usual way to submit changes for new comments? Would it be
> incremental changes (new commit) based on previous one? Thanks.
Good question.
I think it's better if Bruce
Hi DPDK community,
I'd like to have DPDK NIC IO operations in (primary) process and
execution logic in (secondary) processes.
Primary process pushes NIC Rx mbufs to Secondary process through S/W ring
Seconary process allocates mbuf for Tx path and pushes down to Primary
process for NIC Tx
I
This commit includes two small code optimization.
1. use __insn_fetchadd4() instead of rte_atomic32_xxx() in
mpipe_dp_enter/mpipe_dp_exit to avoid the nnecessary memory
fence.
2. replace the calling of rte_pktmbuf_reset() in mpipe_recv_mbuf()
with specific code to avoid some duplicated
rte_vect.h was missing earlier thus LPM was disabled and l3fwd is
not able to compile. This commit implements the vector api and
enable LPM in the tilegx configuration by default.
Signed-off-by: Liming Sun
Acked-by: Zhigang Lu
---
config/defconfig_tile-tilegx-linuxapp-gcc | 2 +-
This commit includes two small code optimization.
1. use __insn_fetchadd4() instead of rte_atomic32_xxx() in
mpipe_dp_enter/mpipe_dp_exit to avoid the nnecessary memory
fence.
2. replace the calling of rte_pktmbuf_reset() in mpipe_recv_mbuf()
with specific code to avoid some duplicated
rte_vect.h was missing earlier thus LPM was disabled and l3fwd is
not able to compile. This commit implements the vector api and
enable LPM in the tilegx configuration by default.
Signed-off-by: Liming Sun
Acked-by: Zhigang Lu
---
config/defconfig_tile-tilegx-linuxapp-gcc | 2 +-
2016-02-09 18:37, Liming Sun:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> A comment about the TILE-Mx would be welcome.
> Is it supported currently?
> Isn't it an ARM arch?
>
> [lsun] Yes, it's ARM arch, but with similar mpipe driver.
> It's not fully supported yet. I'll
Looks like this patch serie has been merged into dpdk-next-net/rel_16_04.
What would be the usual way to submit changes for new comments? Would it be
incremental changes (new commit) based on previous one? Thanks.
From: Thomas Monjalon
Helin is still in Chinese New Year Vacation. Will the below command option help
?
4.5.9. port config - max-pkt-len
Set the maximum packet length:
testpmd> port config all max-pkt-len (value)
This is equivalent to the --max-pkt-len command-line option.
-Original Message-
From: dev
Make sense. I'll split them into different patches.
Thanks!
Liming
-Original Message-
From: Bruce Richardson [mailto:bruce.richard...@intel.com]
Sent: Tuesday, February 09, 2016 10:56 AM
To: Liming Sun
Cc: dev at dpdk.org
Subject: Re: [dpdk-dev] [PATCH v2 1/2] driver/net/mpipe: add
The patch introduces a new PMD. This PMD is implemented as thin wrapper
of librte_vhost. It means librte_vhost is also needed to compile the PMD.
The vhost messages will be handled only when a port is started. So start
a port first, then invoke QEMU.
The PMD has 2 parameters.
- iface: The
This patch adds a below event type.
- RTE_ETH_EVENT_QUEUE_STATE_CHANGE
This event is used for notifying a queue state changed event.
Signed-off-by: Tetsuya Mukawa
Acked-by: Ferruh Yigit
---
lib/librte_ether/rte_ethdev.h | 2 ++
1 file changed, 2 insertions(+)
diff --git
The patch introduces a new PMD. This PMD is implemented as thin wrapper
of librte_vhost.
PATCH v9 changes:
- Fix a null pointer access issue implemented in v8 patch.
PATCH v8 changes:
- Manage ether devices list instead of internal structures list.
- Remove needless NULL checking.
- Replace
Thanks Thomas for the comments.
Please see the response inline.
Thanks,
Liming
-Original Message-
From: Thomas Monjalon [mailto:thomas.monja...@6wind.com]
Sent: Tuesday, February 09, 2016 11:16 AM
To: Liming Sun
Cc: dev at dpdk.org; bruce.richardson at intel.com
Subject: Re: [dpdk-dev]
On Fri, 29 Jan 2016 15:08:32 +0100
David Marchand wrote:
> Now that we only have vdev drivers, there is no need for a pmd type in
> rte_driver.
> rte_driver is now purely a vdev thing, and could be renamed later
> (then .init would become .probe, .uninit would become .remove).
>
>
Maybe, a better subject?
drivers: init pdev drivers in constructors
On Fri, 29 Jan 2016 15:08:30 +0100
David Marchand wrote:
> Now that pdev drivers have been converted to pci drivers, there is nothing
> left in their init functions that can't go in a constructor.
> pdev / vdev drivers init
On Tue, Feb 09, 2016 at 05:33:55PM +, Reshma Pattan wrote:
> Hi Ferruh,
>
Hi Reshma,
> On 1/27/2016 4:32 PM, Ferruh Yigit wrote:
>> This patch provides slow data path communication to the Linux kernel.
>> Patch is based on librte_kni, and heavily re-uses it.
>>
>> The main difference is
Hello Helin,
I tried to send jumbo frames to a i40e card. However, I observed that
all frames are dropped. Moreover, set_mtu function is not implemented on
i40e PMD.
> testpmd --log-level 8 --huge-dir=/mnt/huge -n 4 -l 2,18 --socket-mem
1024,1024 -w :02:00.0 -w :02:00.2 -- -i
Hi Ferruh,
On 1/27/2016 4:32 PM, Ferruh Yigit wrote:
> This patch provides slow data path communication to the Linux kernel.
> Patch is based on librte_kni, and heavily re-uses it.
>
> The main difference is librte_kni library converted into a PMD, to
> provide ease of use for applications.
>
>
Hi Qian --
Any suggestions? This is bit urgent.
/Saurabh
On Sat, Feb 6, 2016 at 9:22 AM, Saurabh Mishra
wrote:
> Hi Qian --
>
>
> Here's the data from Host:
>
> [root at oscompute3 ~]# ethtool -i p3p1
>
> driver: i40e
>
> version: 1.0.11-k
>
> firmware-version: f4.40 a1.4 n04.53 e80001dc0
>
>
On Mon, Feb 08, 2016 at 02:39:02PM +0100, Thomas Monjalon wrote:
> It was requested by Intel, more than one year ago, to replace the name
> "Intel DPDK" by "DPDK".
> Some references to the old name were still in some docs and code comments,
> leading to confusion.
>
> Fixes: ac8ada004c12 ("doc:
Hi,
Sorry for being late for commenting.
2016-01-08 09:30, Liming Sun:
> -EZchip TILE-Gx
> +EZchip TILE-Gx/Mx
A comment about the TILE-Mx would be welcome.
Is it supported currently?
Isn't it an ARM arch?
> M: Zhigang Lu
> +M: Liming Sun
> F: lib/librte_eal/common/include/arch/tile/
> F:
Implement rte_pmd_af_packet_devuninit() exposed through struct
rte_driver.uninit() and set dev_flags to RTE_ETH_DEV_DETACHABLE,
to allow af_packet device deinitialization with API function
rte_eth_dev_detach(). This fixes memory leak by freeing memory
allocated during initialization.
During device
Thank you!
W dniu 09.02.2016 o 15:56, Thomas Monjalon pisze:
> 2016-01-28 14:13, krytarowski at caviumnetworks.com:
>> From: Kamil Rytarowski
>>
>> Currently rte_eal_check_module() detects Linux kernel modules via reading
>> /proc/modules. Built-in ones aren't listed there and therefore they are
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Tuesday, February 2, 2016 14:50
> To: Kulasek, TomaszX ; dev at dpdk.org
> Subject: RE: [dpdk-dev] [PATCH 1/2] ethdev: add buffered tx api
>
> Hi Tomasz,
>
> > -Original Message-
> > From: Kulasek, TomaszX
> > Sent:
On Tue, Feb 09, 2016 at 05:09:06PM +0100, Wojciech Zmuda wrote:
> Implement rte_pmd_af_packet_devuninit() exposed through struct
> rte_driver.uninit() and set dev_flags to RTE_ETH_DEV_DETACHABLE,
> to allow af_packet device deinitialization with API function
> rte_eth_dev_detach(). This fixes
On Fri, Jan 08, 2016 at 09:39:08AM -0500, Liming Sun wrote:
> 1. Fixed the compiling issue of the ethtool example on tilegx
>platform.
> 2. Fixed the hung/crash issue when quitting testpmd under high
>traffic rate. The buffer error bit needs to be checked before
>processing the idesc
2016-01-28 14:13, krytarowski at caviumnetworks.com:
> From: Kamil Rytarowski
>
> Currently rte_eal_check_module() detects Linux kernel modules via reading
> /proc/modules. Built-in ones aren't listed there and therefore they are not
> being found by the script.
>
> Add support for checking
On Fri, Jan 08, 2016 at 09:39:07AM -0500, Liming Sun wrote:
> rte_vect.h was missing earlier thus LPM was disabled and l3fwd is
> not able to compile. This commit implements the vector api and
> enable LPM in the tilegx configuration by default. It also includes
> a minor optimization to use
On Fri, Jan 08, 2016 at 09:30:36AM -0500, Liming Sun wrote:
> This submit updates the CROSS setting to support native build on
> TileGx platform. It also enable the combined library by default.
>
> Signed-off-by: Liming Sun
> Acked-by: Zhigang Lu
Series applied to dpdk-next-net/rel_16_04.
This patch documents that the statistics of fm10k based NICs must be
read regularly in order to avoid an undetected 32 bit integer-overflow.
Signed-off-by: Harry van Haaren
---
v2:
-Moved notes to Known Issues section of Release Notes.
doc/guides/nics/fm10k.rst| 16
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
> Sent: Monday, February 8, 2016 4:01 PM
> To: Van Haaren, Harry
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] docs: add statistics read frequency to
> fm10k guide
>
> 2016-02-08
2016-01-27 20:59, yurai:
> This patch fixes syntax errors from tools/setup.sh during binding ethernet
> device
> on systems where Python 3 is default. Backward compability with Python 2 is
> preserved.
>
> Signed-off-by: Dawid Jurczak
Applied, thanks
On 2016/02/07 16:48, David Marchand wrote:
> Move all os / arch specifics to eal.
>
> Signed-off-by: David Marchand
> ---
> drivers/net/virtio/virtio_pci.c | 339
> +++-
> drivers/net/virtio/virtio_pci.h | 38 +
> 2 files changed, 55 insertions(+), 322
2016-02-09 11:10, Bruce Richardson:
> On Mon, Feb 08, 2016 at 04:33:46PM -0800, Jeff Shaw wrote:
> > We should call sys.exit(), not divide sys by exit().
Obviously :)
> > Signed-off-by: Jeff Shaw
>
> Acked-by: Bruce Richardson
Applied, thanks
2016-01-27 21:07, Thomas Monjalon:
> In order to better compare the drivers and check what is missing
> for a common baseline, we need to fill a matrix.
>
> A CSS trick is used to fit the HTML page.
> The PDF output needs some LaTeX wizardry.
>
> Signed-off-by: Thomas Monjalon
> ---
> v2: add
2016-02-02 13:11, Ferruh Yigit:
> * remove outdated chapter reference to Multi-process support.
> Fixes: fc1f2750a3ec ("doc: programmers guide")
>
> * html output converts "--" to "-", this is wrong when explaining the
> command arguments, used fixed width quotes for them.
>
> v2:
> * for
Enable promiscuous and allmulticast mode control from the VF using
rte_eth_promiscuous_enable()/rte_eth_promiscuous_disable() and
rte_eth_allmulticast_enable()/rte_eth_allmulticast_disable().
For promiscuous mode host/PF igb driver should be built with
IGB_ENABLE_VF_PROMISC.
For allmulticast
> > Add a further ACL example where the elements of the search key are not
> > entirely fitting into the 4 consecutive bytes of all input fields.
> >
> > Signed-off-by: Antonio Fischetti
>
> Acked-by: John McNamara
Applied, thanks
Adding support to the build system to allow for Makefile.XXX
extention to a subtree, which already has Makefiles. These
Makefiles could be from the autotools and others places. Using
the Makefile extention RTE_MKFILE_SUFFIX in a makefile subtree
using 'export RTE_MKFILE_SUFFIX=.XXX' to use
Patch v2 to add some comments and setup for RTE_NEXT_ABI changes.
The rte_mempool structure is changed, which will cause an ABI change
for this structure. Providing backward compat is not reasonable
here as this structure is used in multiple defines/inlines.
Allow mempool cache support to be
Is there any substantive feedback on this? It seems like silence =
consensus (or that the people who might care more are not on this list).
My suggestion is to go to the next stage, start drafting a membership
agreement and identify potential members to recruit. Will the Linux
Foundation take
On Mon, Feb 08, 2016 at 04:33:46PM -0800, Jeff Shaw wrote:
> We should call sys.exit(), not divide sys by exit().
>
> Signed-off-by: Jeff Shaw
Acked-by: Bruce Richardson
On Tue, Feb 09, 2016 at 09:39:26AM +0100, Adrien Mazarguil wrote:
> On Mon, Feb 08, 2016 at 10:11:49PM +, Bruce Richardson wrote:
> > When compiling with clang 3.6, the mlx4 driver gives the following error
> > message about an unneeded function.
> >
> > CC mlx4.o
> >
On 2016/02/08 18:42, Ferruh Yigit wrote:
> On Fri, Feb 05, 2016 at 03:28:37PM +0900, Tetsuya Mukawa wrote:
>> On 2016/02/04 20:17, Ferruh Yigit wrote:
>>> On Thu, Feb 04, 2016 at 04:26:31PM +0900, Tetsuya Mukawa wrote:
>>>
>>> Hi Tetsuya,
>>>
The patch introduces a new PMD. This PMD is
On Mon, Feb 08, 2016 at 05:14:54PM +, Reshma Pattan wrote:
Hi Reshma,
>
>
> On 1/27/2016 4:32 PM, Ferruh Yigit wrote:
>> This kernel module is based on KNI module, but this one is stripped
>> version of it and only for data messages, no control functionality
>> provided.
>>
>> FIFO
As demonstrated by the following code, CRC32c computation is not valid
when buffer length is not a multiple of 4 bytes:
(Output obtained by code below)
CRC of 1 NULL bytes expected: 0x527d5351
soft: 527d5351
rte accelerated: 48674bc7
rte soft: 48674bc7
CRC of 2 NULL bytes expected:
On Mon, Feb 8, 2016 at 3:13 PM, Burakov, Anatoly
wrote:
>> Include vfio map/rd/wr support for pci ioport.
>>
>> Signed-off-by: Santosh Shukla
>> ---
>> v7->v8:
>> - Remove rte_pci_ioport malloc and rte_free()/unmap() func from v7.
>> - removed umap from git header.
>>
>>
On Mon, Feb 8, 2016 at 12:03 PM, Jan Viktorin
wrote:
> On Fri, 29 Jan 2016 15:08:29 +0100
> David Marchand wrote:
>
>> pdev drivers are actually pci drivers.
>> Wrappers for ethdev and crypto pci drivers, that assume a 1 to 1
>> association between pci device and upper device, have been added
On Mon, Feb 08, 2016 at 10:11:49PM +, Bruce Richardson wrote:
> When compiling with clang 3.6, the mlx4 driver gives the following error
> message about an unneeded function.
>
> CC mlx4.o
> .../drivers/net/mlx4/mlx4.c:136:20: fatal error: function
> 'wr_id_t_check' is not needed and
Hello Jan,
On Mon, Feb 8, 2016 at 2:31 PM, Jan Viktorin wrote:
> I am confused a bit. I started to review the "[PATCH 0/9] prepare for
> rte_device
> / rte_driver" series and then I've noticed there are 2 patch series having
> "pci:
> no need for dynamic tailq init" patch there. But then,
Hi Wojciech
> > -Original Message-
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Wojciech Zmuda
> > Sent: Tuesday, January 5, 2016 2:05 PM
> > To: dev at dpdk.org
> > Subject: [dpdk-dev] [PATCH] af_packet: make the device detachable
> >
> > Fix memory leak when detaching
On Tue, Feb 9, 2016 at 4:52 AM, Tetsuya Mukawa wrote:
> On 2016/02/07 16:48, David Marchand wrote:
> One more legacy_virtio_resource_init() is defined like below.
>
> #ifdef RTE_EXEC_ENV_LINUXAPP
>
> #else
>
> static int
> legacy_virtio_resource_init(struct rte_pci_device *pci_dev __rte_unused)
54 matches
Mail list logo
