Hi Thomas,
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Wednesday, November 12, 2014 1:41 PM
> To: Olivier MATZ; Ananyev, Konstantin; dev at dpdk.org
> Cc: Yong Wang
> Subject: Re: [dpdk-dev] [PATCH v8 10/10] app/testpmd:test VxLAN Tx checksum
This patch depends on the vlan filter set fix.
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Huawei Xie
> Sent: Wednesday, November 12, 2014 3:34 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH 0/2] examples/vhost: support new VMDQ api and
> new nic
Hi,
Some libraries need AVX instruction, but if you don't use, it can disable.
e.g. CONFIG_RTE_LIBRTE_IXGBE_PMD=n in config file.
2014-11-12 21:45 GMT+09:00 Yaron Illouz :
> Hi
>
>
>
> I use dpdk in my product. I can't change my dpdk version.
>
> I compile it on a server that have avx.
>
>
Hi Bruce,
The issue seems to be triggered by a combination of using the
CONFIG_RTE_LIBRTE_MEMPOOL_DEBUG option and contention in the mempool(small
mempool size).
I disabled that and the header panics went away.
Thank you for your help!
--Kam
On Fri, Nov 7, 2014 at 4:48 AM, Bruce Richardson
Hi Bruce,
Thanks for your reply.
I agree that to logically divide the distributor functionality is the best
solution.
Meantime I tried some tricks and the result looks good: For same amount of
pkts (1M), the LLC stores and loads decrease 90% percent, and the miss
rates for both decrease to 25%.
Hi Konstantin,
On 11/12/2014 06:21 PM, Ananyev, Konstantin wrote:
>>> This patch introduces 2 new functions rte_get_rx_ol_flag_name()
>>> and rte_get_tx_ol_flag_name() that returns the name of a flag from
>>> its mask. It also fixes rxonly.c to use this new functions and to
>>> display the proper
2014-11-11 05:29, Liu, Jijiang:
> From: Olivier MATZ
> > On 11/10/2014 07:03 AM, Liu, Jijiang wrote:
> > > > - if PKT_TX_VXLAN_CKSUM is not set (legacy use case), then the
> > > >driver use l2_len and l3_len to offload inner IP/UDP/TCP checksums.
> > >
> > > If the flag is not set, and imply
> -Original Message-
> From: Richardson, Bruce
> Sent: Monday, November 10, 2014 5:30 PM
> To: Olivier Matz
> Cc: dev at dpdk.org; Walukiewicz, Miroslaw; Liu, Jijiang; Liu, Yong; jigsaw
> at gmail.com; Ananyev, Konstantin
> Subject: Re: [PATCH 06/12] mbuf: add functions to get the name
On 12/11/14 15:49, Burakov, Anatoly wrote:
> Hi Morten
>
> I'm not that familiar with valgrind, but as far as I know it's using some
> kind of CPU emulation. Most likely the processor being emulated by valgrind
> doesn't have the capabilities your actual CPU has (sort of like trying to run
> a
2014-11-12 14:31, Zhang, Helin:
> > 2014-10-23 02:23, Zhang, Helin:
> > > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
> > > > 2014-10-21 14:14, Liu, Jijiang:
> > > > > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > > > > > 2014-10-21 16:46, Jijiang Liu:
On Wed, Nov 12, 2014 at 10:23:49AM -0500, Neil Horman wrote:
> On Wed, Nov 12, 2014 at 02:24:10PM +, De Lara Guarch, Pablo wrote:
> >
> >
> > > -Original Message-
> > > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Masaru Oki
> > > Sent: Wednesday, November 12, 2014 1:48 PM
On Wed, Nov 12, 2014 at 10:37:33AM +0200, jigsaw wrote:
> Hi,
>
> OK it is now very clear it is due to memory transactions between different
> nodes.
>
> The test program is here:
> https://gist.github.com/jigsawecho/6a2e78d65f0fe67adf1b
>
> The test machine topology is:
>
> NUMA node0 CPU(s):
Hi Konstantin,
On 11/12/2014 03:39 PM, Ananyev, Konstantin wrote:
>> I'm not sure having get_ipv4_udptcp_checksum() in librte_net would
>> help. The value we have to set in the TCP checksum field depends on the
>> PMD (altought only ixgbe is supported now). So, it would require
>> another
On Tue, Nov 11, 2014 at 10:53:40AM +0200, jigsaw wrote:
> Hi Bruce,
>
> This patch has little, if any, performance impact.
> See the perf stat -d for original and patched version
> of test_distributor_perf.
Thanks for running the test. Results look ok to me. I confirm my previous ack.
/Bruce
>
Hi Morten
I'm not that familiar with valgrind, but as far as I know it's using some kind
of CPU emulation. Most likely the processor being emulated by valgrind doesn't
have the capabilities your actual CPU has (sort of like trying to run a
AVX2-compiled code on a QEMU machine emulating
Hi
I use dpdk in my product. I can't change my dpdk version.
I compile it on a server that have avx.
Most of the time, it run on machine that have avx, but not always.
When running the product on a machine with no avx, it crash on illegal
instruction.
The current solution is to build it
2014-11-12 14:05, Olivier MATZ:
> On 11/12/2014 10:55 AM, Ananyev, Konstantin wrote:
> >> From an API perspective, it looks a bit more complex to have to call
> >> dev_prep_tx() before sending the packets if they have been flagged
> >> for offload processing. But I admit I have no other argument.
> -Original Message-
> From: Olivier MATZ [mailto:olivier.matz at 6wind.com]
> Sent: Wednesday, November 12, 2014 1:06 PM
> To: Ananyev, Konstantin; Yong Wang; Liu, Jijiang
> Cc: dev at dpdk.org; Thomas Monjalon
> Subject: Re: [dpdk-dev] [PATCH v8 10/10] app/testpmd:test VxLAN Tx
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Wednesday, November 12, 2014 9:26 PM
> To: Liu, Jijiang
> Cc: Zhang, Helin; dev at dpdk.org; Richardson, Bruce
> Subject: Re: [dpdk-dev] [PATCH v6 1/9] librte_mbuf:the rte_mbuf structure
> changes
On Wed, Nov 12, 2014 at 04:16:12PM +, Bruce Richardson wrote:
> On Wed, Nov 12, 2014 at 10:23:49AM -0500, Neil Horman wrote:
> > On Wed, Nov 12, 2014 at 02:24:10PM +, De Lara Guarch, Pablo wrote:
> > >
> > >
> > > > -Original Message-
> > > > From: dev [mailto:dev-bounces at
Hi guys,
We still have some problems with the mbuf changes introduced for VXLAN.
I want to raise the packet type issue here.
2014-10-23 02:23, Zhang, Helin:
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
> > 2014-10-21 14:14, Liu, Jijiang:
> > > From: Thomas Monjalon
Per definition, rte_eth_rx_burst/rte_eth_tx_burst/rte_eth_rx_queue_count
returns the packet number
When RTE_LIBRTE_ETHDEV_DEBUG turns on, retval of FUNC_PTR_OR_ERR_RTE was set to
-ENOTSUP.
It makes confusing.
The patch always return 0 no matter no packet or there's error.
Signed-off-by: Cunming
The unit test can be used to measure cycles per packet in different rx/tx
rouines.
The NIC works in loopback mode. So it doesn't require test equipment to measure
throughput.
As result, the unit test shows the average cycles per packet consuming.
When doing the test, make sure the link is UP.
Signed-off-by: Cunming Liang
---
examples/dpdk_qat/main.c | 10 +++---
examples/ip_fragmentation/main.c | 10 +++---
examples/ip_reassembly/main.c| 10 +++---
examples/ipv4_multicast/main.c | 10 +++---
examples/l3fwd-acl/main.c| 10 +++---
Signed-off-by: Cunming Liang
---
app/test-pmd/config.c | 10 +++---
app/test-pmd/icmpecho.c | 16 ++--
app/test-pmd/rxonly.c | 11 +++
3 files changed, 8 insertions(+), 29 deletions(-)
diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
index
Signed-off-by: Cunming Liang
---
lib/librte_ether/rte_ether.h | 25 +
1 files changed, 25 insertions(+), 0 deletions(-)
diff --git a/lib/librte_ether/rte_ether.h b/lib/librte_ether/rte_ether.h
index f173509..187608d 100644
--- a/lib/librte_ether/rte_ether.h
+++
Signed-off-by: Cunming Liang
---
lib/librte_pmd_ixgbe/ixgbe_ethdev.c |6 ++
1 files changed, 6 insertions(+), 0 deletions(-)
diff --git a/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
b/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
index 9c73a30..05490dc 100644
--- a/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
adding support to allow packet burst generator to create packets in differenct
sizes
Signed-off-by: Cunming Liang
---
app/test/packet_burst_generator.c | 26
app/test/packet_burst_generator.h | 11 +++--
app/test/test_link_bonding.c | 39
v7 update:
# patch split and re-orginize
v6 update:
# leave FUNC_PTR_OR_*_RET unmodified
v5 update:
# fix the confusing of retval in some API of rte_ethdev
v4 ignore
v3 update:
# Codes refine according to the feedback.
1. add ether_format_addr to rte_ether.h
2. fix typo in code comments.
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Masaru Oki
> Sent: Wednesday, November 12, 2014 1:48 PM
> To: Yaron Illouz
> Cc:
> Subject: Re: [dpdk-dev] compile dpdk with no avx
>
> Hi,
>
> Some libraries need AVX instruction, but if you don't use, it
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Wednesday, November 12, 2014 5:31 PM
> To: Zhang, Helin
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v5 2/5] ethdev: add enum type and relevant
> structures for hash filter control
>
>
On Tue, Nov 11, 2014 at 07:18:32PM +0530, Manoj Viswanath wrote:
> Bruce,
>
> Thanks for the input.
> Sure, will figure out the offending file behind this error and update this
> thread.
>
> Meanwhile, wanted to share one more observation regarding this issue:
> The "file descriptor error" is
Hi Konstantin,
On 11/12/2014 10:55 AM, Ananyev, Konstantin wrote:
>> From an API perspective, it looks a bit more complex to have to call
>> dev_prep_tx() before sending the packets if they have been flagged
>> for offload processing. But I admit I have no other argument. I'll be
>> happy to
Hi Xie,
(2014/11/12 6:37), Xie, Huawei wrote:
> Hi Tetsuya:
> There are two major technical issues in my mind for vhost-user implementation.
>
> 1) memory region map
> Vhost-user passes us file fd and offset for each memory region. Unfortunately
> the mmap offset is "very" wrong. I discovered
> -Original Message-
> From: Olivier Matz [mailto:olivier.matz at 6wind.com]
> Sent: Monday, November 10, 2014 3:59 PM
> To: dev at dpdk.org
> Cc: olivier.matz at 6wind.com; Walukiewicz, Miroslaw; Liu, Jijiang; Liu,
> Yong; jigsaw at gmail.com; Richardson, Bruce; Ananyev, Konstantin
>
2014-11-12 10:29, Liang, Cunming:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > 2014-11-12 08:21, Liang, Cunming:
> > > For em, igb and i40e, we haven't provide so much rx/tx pair for switching.
> >
> > Sorry, I don't understand. Which pair are you telling about?
> > em, igb
scatter/non-scatter always be checked during dev_start.
For others, it's only have the two. Won't do additional check during
rx/tx_queue_setup(before dev_start).
So they won't have problem, I think.
But for ixgbe, it will check it meets vector condition or not, then choose the
best performance
Hi,
OK it is now very clear it is due to memory transactions between different
nodes.
The test program is here:
https://gist.github.com/jigsawecho/6a2e78d65f0fe67adf1b
The test machine topology is:
NUMA node0 CPU(s): 0-7,16-23
NUMA node1 CPU(s): 8-15,24-31
Change the 3rd param from 0
2014-11-11 23:48, Jia Yu:
> The two patches are minor fixes. No new API is introduced.
You are changing the API of rte_eth_stats_get().
It's a tricky corner case because it's a light change which should have
no impact. Anyone against changing the void return into int? Impact in mind?
I can
2014-11-12 05:52, Zhang, Helin:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > Actually you use ETH_FLOW_TYPE_IPV4 for ixgbe and
> > ETH_FLOW_TYPE_FRAG_IPV4 or ETH_FLOW_TYPE_NONFRAG_IPV4_* for i40e.
> > It's not consistent and clearly shows that you stick to the hardware
> >
Maybe pair is not accurate, I means the different rx/tx register function, like:
Ixgbe_recv_bulk_alloc/ixgbe_recv_(scattered_)pkts_vec/ixgbe_recv_scattered_pkts
ixgbe_xmit_pkts_simple/Ixgbe_xmit_pkts_vec/ixgbe_xmit_pkts
-Liang Cunming
> -Original Message-
> From: Thomas Monjalon
2014-11-12 08:21, Liang, Cunming:
> For em, igb and i40e, we haven't provide so much rx/tx pair for switching.
Sorry, I don't understand. Which pair are you telling about?
em, igb and i40e have scattered Rx functions.
--
Thomas
On Wed, Nov 12, 2014 at 02:24:10PM +, De Lara Guarch, Pablo wrote:
>
>
> > -Original Message-
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Masaru Oki
> > Sent: Wednesday, November 12, 2014 1:48 PM
> > To: Yaron Illouz
> > Cc:
> > Subject: Re: [dpdk-dev] compile dpdk
Hi Oliver,
> From: Olivier MATZ [mailto:olivier.matz at 6wind.com]
> Sent: Monday, November 10, 2014 3:58 PM
> To: Ananyev, Konstantin; Yong Wang; Liu, Jijiang
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v8 10/10] app/testpmd:test VxLAN Tx checksum
> offload
>
> Hello Konstantin,
>
Hi Cunming,
Please, could you provide an explanation for the commit log?
It should answer to the question "what was the issue?"
If it's a fix, the title should start with "fix".
Maybe that the same kind of fix is needed for em, igb and i40e?
Thanks
--
Thomas
The scattered_rx is update in dev_start.
In this unit test, we will re-configure and change the scatter mode.
When we stop, re-configure and then re-start, it expect using the new configure.
But during re-configure, the stored data may still old.
The patch clean the configure anyway in dev_stop.
Hi Thomas,
I've split the patch in v7 and also do cleanup by the new API.
Thanks.
-Liang Cunming
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Wednesday, November 12, 2014 7:29 AM
> To: Liang, Cunming
> Cc: dev at dpdk.org
> Subject: Re:
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Wednesday, November 12, 2014 5:09 AM
> To: Zhang, Helin
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v5 2/5] ethdev: add enum type and relevant
> structures for hash filter control
>
>
Hi Jia
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Jia Yu
> Sent: Saturday, November 8, 2014 1:32 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH 1/2] rte_ethdev: update link status (speed, duplex,
> link_up) after rte_eth_dev_start
>
> Since LSR
Hi,
Background:
What we are doing now is port make telecom network element to be cloud based.
For one of our product, DPDK is applied not only for fastpath/dataplane
processing, but also for Distributed Message eXchange (DMX) between different
processes/applications which may located in
Hi Thomas,
Thanks very much for applying this patch!
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Wednesday, November 12, 2014 7:17 AM
> To: Ouyang, Changchun
> Cc: dev at dpdk.org; Xie, Huawei
> Subject: Re: [dpdk-dev] [PATCH v4 0/5] Support
Hi Cunming,
2014-10-27 09:20, Cunming Liang:
> --- a/lib/librte_ether/rte_ether.h
> +++ b/lib/librte_ether/rte_ether.h
> @@ -45,6 +45,7 @@ extern "C" {
> #endif
>
> #include
> +#include
>
> #include
> #include
> @@ -266,6 +267,30 @@ static inline void ether_addr_copy(const struct
>
> -V1 change:
> This patch series support multicast feature in virtio and vhost.
> The vhost backend enables the promiscuous mode and config
> ETH_VMDQ_ACCEPT_BROADCAST
> and ETH_VMDQ_ACCEPT_MULTICAST in VMDQ offload register to receive the
> multicast and broadcast packets.
> The virtio
53 matches
Mail list logo