On 24/07/14 07:59, Matthew Hall wrote:
> Hello,
>
> I ran into some weird symbol conflicts between system netinet/in.h and DPDK
> rte_ip.h. They have a lot of duplicated definitions for stuff like IPPROTO_IP
> and so on. This breaks when you want to use inet_pton from arpa/inet.h,
> because it
On Thu, Jul 24, 2014 at 08:41:33PM +, Venkatesan, Venky wrote:
> Neil,
>
> Nice patch! One question - what gcc versions did you try this out on? We'll
> round out with checking the other versions.
>
Thanks! This was built using gcc 4.8.3. I had hoped to do something simmilar
for the ACL
Neil,
Nice patch! One question - what gcc versions did you try this out on? We'll
round out with checking the other versions.
Regards,
-Venky
-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Neil Horman
Sent: Thursday, July 24, 2014 11:24 AM
To: dev at
Hi Chris,
2014-07-24 08:20, Chris Wright:
> * Pablo de Lara (pablo.de.lara.guarch at intel.com) wrote:
> > Signed-off-by: Patrice Buriez
>
> Just a mechanical nitpick on DCO. Pablo, this patch appears to be
> written by Patrice. If so, it should begin with "From: Patrice Buriez
> " and should
2014-07-24 16:54, Thomas Monjalon:
> > Unlike RHEL_RELEASE_CODE, there is no such UBUNTU_RELEASE_CODE available
> > out of
> > the box, so it needs to be crafted from the Makefile
> > Similarly, UBUNTU_KERNEL_CODE is generated with ABI and upload numbers.
>
> It's quite amazing to see that Linux
> Unlike RHEL_RELEASE_CODE, there is no such UBUNTU_RELEASE_CODE available out
> of
> the box, so it needs to be crafted from the Makefile
> Similarly, UBUNTU_KERNEL_CODE is generated with ABI and upload numbers.
It's quite amazing to see that Linux distributions do backports and do not
provide
Thanks and it is working and help me a lot.
On Thu, Jul 24, 2014 at 3:05 PM, Thomas Monjalon
wrote:
> 2014-07-24 15:00, sothy shan:
> > I have problem when I try dpdk ovs with DPDK 1.7. DPDK 1.7 alone working
> > well.
> [...]
> > I am not sure when I using DPDK ovs, how to debug at DPDK 1.7.0
On Thu, Jul 24, 2014 at 10:55:59PM +, Antti Kantee wrote:
> In my experience from years of fighting with more or less this exact same
> problem -- the fight is now thankfully over but the scars remain -- you
> either want to expose a complete set of types and provide support for
> everything,
I also noticed this problem. It is a serious one. I would like it solved.
-- Niraj Sharma
Principal Eng.
Cisco System, Inc.
On 7/24/14 12:59 AM, "Matthew Hall" wrote:
>Hello,
>
>I ran into some weird symbol conflicts between system netinet/in.h and
>DPDK
>rte_ip.h. They have a lot of
Recent Ubuntu kernel 3.13.0-30.54, although based on Linux kernel 3.13.11,
already provides skb_set_hash() inline function, slightly different than
the one provided by lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h
Ubuntu kernel 3.13.0-30.54 provides:
* i40e/i40evf: i40e implementation for
Hi Franck:
The natural wrapper for vhost library would be like port in OVDK, which you can
receive from or send to. In DPDK, we have a vhost example, which demonstrates
simple switching functionality between guest virtual machines.
BR
-huawei
> -Original Message-
> From: Franck BAUDIN
2014-07-24 15:00, sothy shan:
> I have problem when I try dpdk ovs with DPDK 1.7. DPDK 1.7 alone working
> well.
[...]
> I am not sure when I using DPDK ovs, how to debug at DPDK 1.7.0 code. I
> guess I need to enable to print PMD_DEBUG_TRACE message. At this point,
> DPDK ovs message is only
2014-07-24 14:44, Thomas Monjalon:
> 2014-07-24 14:34, sothy shan:
> > Second, any clue why rte_eth_dev_configure give an error with code -22?
>
> It depends of your driver.
> Example: #define IXGBE_ERR_NO_SAN_ADDR_PTR -22
It's more probably -EINVAL without driver dependency.
It means there is a
Hello,
2014-07-24 14:34, sothy shan:
> I am trying to use DPDK ovs (development branch) with DPDK1.7.
Could you try with testpmd?
> Cannot initi NIC port 0 (-22).
>
> Based on my search within code, rte_eth_dev_configure gives an error with
> code -22.
Which driver are you using?
> My
Six commands are added in testpmd to support configuring
hash functions. They are,
* i40e_get_sym_hash_ena_per_port
* i40e_set_sym_hash_ena_per_port
* i40e_get_sym_hash_ena_per_pctype
* i40e_set_sym_hash_ena_per_pctype
* i40e_get_filter_swap
* i40e_set_filter_swap
Signed-off-by: Helin Zhang
---
Symmetric hash function can be configured:
* Set symmetric hash enable per port.
* Set symmetric hash enable per pctype (packet classification type).
* Set the swap configurations.
Commands are implemented for 'rx_classification_filter_ctl'
to support above operations which can be called in
Toeplitz and simple XOR hash functions are supported by
hardware, code changes are to tell the hardware which hash
function is selected according to the configuration.
Signed-off-by: Helin Zhang
---
config/common_bsdapp | 1 +
config/common_linuxapp| 1 +
Two ops of 'check_command_supported' and 'rx_classification_filter_ctl'
are added.
* 'check_command_supported' is for capability discovery. In anothoer
word, it is to check if specific feature/command is supported by
the specific port.
* 'rx_classification_filter_ctl' is for receive
For better understanding, 'PCTYPE' was added to the name of i40e
RSS shift macros.
Signed-off-by: Helin Zhang
---
lib/librte_ether/rte_ethdev.h | 74 +--
1 file changed, 37 insertions(+), 37 deletions(-)
diff --git a/lib/librte_ether/rte_ethdev.h
These pathches mainly support configuring hash functions.
In detail, it can select Toeplitz and simple XOR hash
functions. It can configure symmetric hash function. Also
'ethdev' level interfaces are implemented in i40e, which
provides commands for application invoking, and to check
if specific
Hi Luke,
From: lukego at gmail.com [mailto:luk...@gmail.com] On Behalf Of Luke Gorrie
Sent: Thursday, July 24, 2014 11:58 AM
To: De Lara Guarch, Pablo
Cc: dev at dpdk.org; Putteneers, SvenX
Subject: Re: [dpdk-dev] Pktgen-DPDK compile error on Ubuntu 14.04
Hi Pablo,
On 24 July 2014 12:33, De
Hello!
I am trying to use DPDK ovs (development branch) with DPDK1.7.
There is an error when running ovs_dpdk command.
Error is
PANIC in ovdk_vport_phy_port_init.
Cannot initi NIC port 0 (-22).
Based on my search within code, rte_eth_dev_configure gives an error with
code -22.
My question
The ixgbe pmd currently can't be built without enabling sse instructions at
compile time. While sse extensions provide better performance, theres no reason
that we can't still create builds to run on systems that don't support sse. If
we modify the ixgbe code to use the __builtin_shuffle and
Many thanks to all for your help.
-Original Message-
From: Matthew Hall [mailto:mh...@mhcomputing.net]
Sent: Wednesday, July 23, 2014 10:56 PM
To: Kavanagh, Mark B
Cc: Antti Kantee; dev at dpdk.org
Subject: Re: [dpdk-dev] Performance - linking against DPDK shared vs static
libraries
On
Hi Pablo,
On 24 July 2014 12:33, De Lara Guarch, Pablo wrote:
> I think you are seeing the same error as other people are seeing for
> DPDK-1.7 on Ubuntu 14.04.
> Are you using kernel 3.13.0-24 or 3.13.0-30/32?
>
Thanks for the quick response. I'm currently using kernel 3.13.0-24.
Fix 2 compilation issues in virtio PMD when dump option is enabled.
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_virtio/virtio_ethdev.c | 2 +-
lib/librte_pmd_virtio/virtqueue.h | 4 ++--
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git
Refer to DPDK getting started guide paragraph 5.4. It might help.
Also it might be easier to do write simple application that send a packet on
port 1 and rx it on port 2 in separate threads on separate cores (simple
physical loop). You can then add timestamp and send packet back and see how
Kai, the latency depends both on what you do and how much you send.
A bigger packet will take longer time to transmit.
Now that thats out of the way I propose you use perf to see how busy
is the cpu and with what.
FYI, ~10us is something that can be achieved with netperf with a
kernel driver
Hi Luke,
I think you are seeing the same error as other people are seeing for DPDK-1.7
on Ubuntu 14.04.
Are you using kernel 3.13.0-24 or 3.13.0-30/32?
Thanks,
Pablo
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Luke Gorrie
> Sent: Thursday, July 24,
2014-07-24 08:14, Zhang, Helin:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > 2014-07-24 14:42, Helin Zhang:
> > > For better understanding, 'PCTYPE' was added to the name of i40e RSS
> > > shift macros.
> > >
> > > Signed-off-by: Helin Zhang
> >
> > > -#define
Howdy!
I am having trouble building Pktgen-DPDK from Github on Ubuntu 14.04. Is
this supported? If so, would anybody tell me how to get the build working?
I have tried to be faithful to the instructions in README.md and I have
tested with both master and the latest release tag (pktgen-2.7.1).
2014-07-24 01:01, Matthew Hall:
> On Thu, Jul 24, 2014 at 09:59:23AM +0200, Thomas Monjalon wrote:
> > Is it really a good idea to configure this kind of thing at build time?
> > Maybe yes, I'm not sure.
>
> Whether it's safe to set at runtime probably depends what happens to the card
> if it
2014-07-24 14:42, Helin Zhang:
> Toeplitz and simple XOR hash functions are supported by
> hardware, code changes are to tell the hardware which hash
> function is selected according to the configuration.
>
> Signed-off-by: Helin Zhang
> +CONFIG_RTE_LIBRTE_I40E_HASH_FUNC_TOEPLITZ=y
Is it
Hi Bruce,
Thank you for the response. That's a great Idea!
But I do not understand the last four parameters of this function. (vaddr,
paddr, pg_num, pg_shift)
I guess vaddr is the virtual address of the previously allocated mempool,
paddr is calculated using function call rte_mem_virt2phy(vaddr),
Hi Helin,
2014-07-24 14:42, Helin Zhang:
> For better understanding, 'PCTYPE' was added to the name of i40e
> RSS shift macros.
>
> Signed-off-by: Helin Zhang
> -#define ETH_RSS_NONF_IPV4_UDP_SHIFT 31
> +#define ETH_PCTYPE_NONF_IPV4_UDP 31
Why is it clearer? I don't
2014-07-24 05:41, Harish Patil:
> Gentle second request.
No need to do second request few hours after the first one.
You probably would have more answers if you better described the context
of what you are speaking.
If nobody has time to explain how it works, you can still read the code.
>
Hello,
My understanding is that this is a helper library, but not a PMD driver. Is my
understanding correct? If so, do you plan to write a vhost PMD driver?
Regards,
Franck.
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Xie, Huawei
> Sent: vendredi 18
24.07.2014 0:09, Stephen Hemminger ?:
>> Rewritten IRQ mode handling code introduced in commit 399a3f0d
>> (igb_uio: fix IRQ mode handling) renders some faulty NICs (VMware
>> e1000, for example) unusable if INTX mode is not supported.
>>
>> This patch gets these NICs up and running, but
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Thursday, July 24, 2014 3:57 PM
> To: Zhang, Helin
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 2/5] ethdev: add new ops of
> 'check_command_supported' and 'rx_classification_filter_ctl'
* Thomas Monjalon (thomas.monjalon at 6wind.com) wrote:
> 2014-07-24 08:20, Chris Wright:
> > * Pablo de Lara (pablo.de.lara.guarch at intel.com) wrote:
> > > Signed-off-by: Patrice Buriez
> >
> > Just a mechanical nitpick on DCO. Pablo, this patch appears to be
> > written by Patrice. If so,
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Thursday, July 24, 2014 4:20 PM
> To: Zhang, Helin
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 1/5] ethdev: Rename macros of packet
> classification type
>
> 2014-07-24 08:14, Zhang,
* Pablo de Lara (pablo.de.lara.guarch at intel.com) wrote:
> Signed-off-by: Patrice Buriez
Just a mechanical nitpick on DCO. Pablo, this patch appears to be
written by Patrice. If so, it should begin with "From: Patrice Buriez
" and should include your own Signed-off-by.
thanks,
-chris
> ---
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Thursday, July 24, 2014 3:48 PM
> To: Zhang, Helin
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 1/5] ethdev: Rename macros of packet
> classification type
>
> Hi Helin,
>
> 2014-07-24
On Thu, 24 Jul 2014 09:11:36 +0600
Yerden Zhumabekov wrote:
> 24.07.2014 0:09, Stephen Hemminger ?:
> >> Rewritten IRQ mode handling code introduced in commit 399a3f0d
> >> (igb_uio: fix IRQ mode handling) renders some faulty NICs (VMware
> >> e1000, for example) unusable if INTX mode is not
Gentle second request.
>DPDK-dev,
>From the documentation, its not very clear to me how would the inter-VM
>communication packet path (traffic crossing the physical port), especially
>the way macaddr table is populated in each PF port and role of the PF poll
>mode driver in this context. Could
If no reboot of the card is needed then it's probably better to add it to one
of the ethtool style APIs...
--
Sent from my mobile device.
On July 24, 2014 1:07:37 AM PDT, Thomas Monjalon
wrote:
>2014-07-24 01:01, Matthew Hall:
>> On Thu, Jul 24, 2014 at 09:59:23AM +0200, Thomas Monjalon
On Thu, Jul 24, 2014 at 09:59:23AM +0200, Thomas Monjalon wrote:
> Is it really a good idea to configure this kind of thing at build time?
> Maybe yes, I'm not sure.
Whether it's safe to set at runtime probably depends what happens to the card
if it gets changed. Do you have to reset the card or
Hello,
I ran into some weird symbol conflicts between system netinet/in.h and DPDK
rte_ip.h. They have a lot of duplicated definitions for stuff like IPPROTO_IP
and so on. This breaks when you want to use inet_pton from arpa/inet.h,
because it includes netinet/in.h to define struct in_addr.
48 matches
Mail list logo
