As HW vlan strip will reduce the packet length by minus length of vlan tag,
so it need restore the packet length by plus it.
Signed-off-by: Changchun Ouyang
---
v2 change:
Update data length by plus offset in first segment instead of last segment.
v1 change:
Update the packet length by plus
Hello,
I am running "testpmd" in DPDK 1.7.1, using two ixgbevf devices in my host OS.
I am finding that "testpmd" reports link status with speed 100 Mbps and
half-duplex:
# $RTE_SDK/$RTE_TARGET/app/testpmd -c f -n 4 -- -i
...
EAL: PCI device :03:10.2 on NUMA socket 0
EAL: probe driver:
Hello,
I am trying to bring up DPDK on Xen Dom-U.
Have followed all the steps in DPDK programmers guide in section 14.3.
However, when I try to run testpmd or any other example application, it is
getting stuck during hugepage setup.
It hangs after the print "EAL: Setting up memory..."
Has
Thank you Danny. This is helpful info.
Regards,
Shivapriya Hiremath
On Thu, Oct 30, 2014 at 4:22 PM, Zhou, Danny wrote:
> Shivapriya, Alex
>
> There is obvious performance (up to 40% perf. drop) impact to DPDK 64B
> small packet performance when ?iommu = on? on Sandybridge, due to known
>
As HW vlan strip will reduce the packet length by minus length of vlan tag,
so it need restore the packet length by plus it.
Signed-off-by: Changchun Ouyang
---
examples/vhost/main.c | 17 ++---
1 file changed, 14 insertions(+), 3 deletions(-)
diff --git a/examples/vhost/main.c
When building shared libs (for both GCC and CLANG targets), -fPIC flag
has been added to CFLAGS and leaks to BSD module build system causing
the following error:
fatal error: error in backend: Cannot select: 0x802ad8010: i64 =
X86ISD::WrapperRIP 0x802ade110
[ID=13]
0x802ade110: i64 =
Hi
I`m reading the ixgbe code especially about H/W ring and S/W ring. Is
the relationship between H/W ring and S/W ring one-to-one mapping?
As far as I know, H/W ring size is determined in the code(hard coded)
while S/W ring size is determined in port configuration time.
In the
2014-10-29 13:47, linhaifeng:
> rte_eal_hugepage_free() is used for unlink all hugepages.If you want to
> free all hugepages you must make sure that you have stop to use it,and you
> must call this function before exit process.
>
> Signed-off-by: linhaifeng
Thanks for raising the need.
There is
(2014/10/30 0:14), Bruce Richardson wrote:
> Given that apps may regularly use eth_dev_count, might it not be
> worthwhile to retain the nb_ports local variable to make this API call
> just be a variable read. The other APIs to enable/disable individual
> ports could easily just inc/dec this var
> > Basically, Ubuntu GCC is always passing --as-needed to the linker
> > which causes some Linking issues for us.
> Can you elaborate here?
> Neil
>
Sorry, probably I could have given more info about the issue.
Currently if we build DPDK with shared libs on Ubuntu, build process fails when
it
There is a logical error in igb_uio/compat.h when define pci_num_vf().
Only CONFIG_PCI_IOV enabled, this function could be defined.
This patch fix the following error for linux-2.6.32.63:
CC [M] /root/dpdk-source/build/build/lib/librte_eal/linuxapp/igb_uio/igb_uio.o
Hi,
I'm not sure this is the right place to post a question like this, but I've
been stuck with the same problem for days now.
I'm trying to use flow director perfect match filters and so far I haven't
been able to get it working. I have tried writing my own simple app (based
on given examples)
2014-10-29 17:48, Yan Freedland:
> I am interested in working with VMXNET3 driver in a NIC - vSwitch
> fashion on VMware.
>
> For that purpose I did the following:
>
>
> 1. On my VMware VM I compiled the driver and loaded it as described
> in http://dpdk.org/doc/vmxnet3-usermap
>
> 2.
I am sorry, I missed it during the redaction of the email. I executed it
correcly. When I do "cat /sys/kernel/mm/dom0-mm/memsize-mB/memsize" it
prints 2048.
Thank you for your answer and your time.
El 30/10/14 a las 13:59, Liu, Jijiang escribi?:
> ">" was missed in the following command.
>
>
To support flow director, this patch sets up fortville resources. It includes
- Queue 0 pair allocated and set up for flow director
- Create vsi
- Reserve memzone for flow director programming packet
Signed-off-by: Jingjing Wu
---
lib/librte_pmd_i40e/Makefile | 2 +
Test command is added to configure flexible masks for each flow type
Signed-off-by: Jingjing Wu
---
app/test-pmd/cmdline.c | 173 +
1 file changed, 173 insertions(+)
diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index
Implement operation to flexible masks for each flow type in i40e pmd driver
Signed-off-by: Jingjing Wu
---
lib/librte_pmd_i40e/i40e_fdir.c | 124 +++-
1 file changed, 123 insertions(+), 1 deletion(-)
diff --git a/lib/librte_pmd_i40e/i40e_fdir.c
New structures are defined for configuring flexible masks
Signed-off-by: Jingjing Wu
---
lib/librte_ether/rte_eth_ctrl.h | 25 +
1 file changed, 25 insertions(+)
diff --git a/lib/librte_ether/rte_eth_ctrl.h b/lib/librte_ether/rte_eth_ctrl.h
index 4a51c22..e8e47fd 100644
Test command is added to configure flexible payload
Signed-off-by: Jingjing Wu
---
app/test-pmd/cmdline.c | 144 +
1 file changed, 144 insertions(+)
diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index f8960d8..5a84a6c 100644
---
Implement operation to flexible payload in i40e pmd driver
Signed-off-by: Jingjing Wu
---
lib/librte_pmd_i40e/i40e_fdir.c | 106
1 file changed, 106 insertions(+)
diff --git a/lib/librte_pmd_i40e/i40e_fdir.c b/lib/librte_pmd_i40e/i40e_fdir.c
index
New Structures are defined for configuring flexible payload
Signed-off-by: Jingjing Wu
---
lib/librte_ether/rte_eth_ctrl.h | 43 +
1 file changed, 43 insertions(+)
diff --git a/lib/librte_ether/rte_eth_ctrl.h b/lib/librte_ether/rte_eth_ctrl.h
index
Implement operation to flush flow director table
Signed-off-by: Jingjing Wu
---
lib/librte_pmd_i40e/i40e_fdir.c | 47 +
1 file changed, 47 insertions(+)
diff --git a/lib/librte_pmd_i40e/i40e_fdir.c b/lib/librte_pmd_i40e/i40e_fdir.c
index 85df220..f59eb24
Display flow director's statistics information
Signed-off-by: Jingjing Wu
---
app/test-pmd/config.c | 42 ++
1 file changed, 38 insertions(+), 4 deletions(-)
diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
index 9bc08f4..c516be0 100644
---
Implement operation to get flow director information in i40e pmd driver
Signed-off-by: Jingjing Wu
---
lib/librte_pmd_i40e/i40e_fdir.c | 34 ++
1 file changed, 34 insertions(+)
diff --git a/lib/librte_pmd_i40e/i40e_fdir.c b/lib/librte_pmd_i40e/i40e_fdir.c
index
Structure is defined for getting flow director information
Signed-off-by: Jingjing Wu
---
lib/librte_ether/rte_eth_ctrl.h | 22 ++
1 file changed, 22 insertions(+)
diff --git a/lib/librte_ether/rte_eth_ctrl.h b/lib/librte_ether/rte_eth_ctrl.h
index 77b784e..74a9c59 100644
Extended fdir info is printed in rxonly fwd engine when fdir match.
Signed-off-by: Jingjing Wu
---
app/test-pmd/rxonly.c | 14 +++---
1 file changed, 11 insertions(+), 3 deletions(-)
diff --git a/app/test-pmd/rxonly.c b/app/test-pmd/rxonly.c
index 4410c3d..281bee1 100644
---
Set the FDIR information in mbuf if match
Signed-off-by: Jingjing Wu
---
lib/librte_pmd_i40e/i40e_rxtx.c | 99 +++--
1 file changed, 95 insertions(+), 4 deletions(-)
diff --git a/lib/librte_pmd_i40e/i40e_rxtx.c b/lib/librte_pmd_i40e/i40e_rxtx.c
index
Fdir field is extened to support flex bytes reported when fdir match.
8 flex bytes can be reported in maximum.
The reported flex bytes are part of flexible payload.
Signed-off-by: Jingjing Wu
---
lib/librte_mbuf/rte_mbuf.h | 14 --
1 file changed, 12 insertions(+), 2 deletions(-)
Commands are added which can be used to test adding or deleting 8 flow
types of the flow director filters: ipv4, tcpv4, udpv4, sctpv4,
ipv6, tcpv6, udpv6, sctpv6.
Signed-off-by: Jingjing Wu
---
app/test-pmd/cmdline.c | 447 +
1 file changed, 447
Deal with two operations for flow director
- RTE_ETH_FILTER_ADD
- RTE_ETH_FILTER_DELETE
Encode the flow inputs to programming packet.
Sent the packet to filter programming queue and check status
on the status report queue.
Signed-off-by: Jingjing Wu
---
lib/librte_pmd_i40e/i40e_ethdev.c | 3
To support flow director tear down, this patch includes
- queue 0 pair release
- release vsi
Signed-off-by: Jingjing Wu
---
lib/librte_pmd_i40e/i40e_ethdev.c | 4 +++-
lib/librte_pmd_i40e/i40e_ethdev.h | 1 +
lib/librte_pmd_i40e/i40e_fdir.c | 19 +++
3 files changed, 23
The patch set supports flow director on fortville.
It includes:
- set up/tear down fortville resources to support flow director, such as queue
and vsi.
- support operation to add or delete 8 flow types of the flow director
filters, they are ipv4, tcpv4, udpv4, sctpv4, ipv6, tcpv6, udpv6,
On Thu, Oct 30, 2014 at 04:20:17PM +, Gonzalez Monroy, Sergio wrote:
> > > Basically, Ubuntu GCC is always passing --as-needed to the linker
> > > which causes some Linking issues for us.
> > Can you elaborate here?
> > Neil
> >
> Sorry, probably I could have given more info about the issue.
Hi ,
> -Original Message-
> From: Richardson, Bruce
> Sent: Thursday, October 30, 2014 6:10 PM
> To: Ouyang, Changchun
> Cc: Xie, Huawei; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 3/5] vhost: enable promisc mode and
> config VMDQ offload register for multicast feature
>
> On
For posterity.
1.When using MTU larger then 2K its advised to provide the value
to rte_pktmbuf_pool_init.
2.ixgbevf rounds down the ("MBUF size" - RTE_PKTMBUF_HEADROOM) to the
nearest 1K multiple when deciding on the receiving capabilities [buffer
size]of the Buffers in the pool.
The function
On Thu, Oct 30, 2014 at 09:18:23AM +, Gonzalez Monroy, Sergio wrote:
> I would say that D) is a good balance, although not being the simplest.
A, or D. Depending on things such as, "If you run the DPDK on Random Platform
X," where X could be something like Power CPUs or other weird stuff,
On Thu, Oct 30, 2014 at 02:48:42PM +0200, Alex Markuze wrote:
> For posterity.
>
> 1.When using MTU larger then 2K its advised to provide the value
> to rte_pktmbuf_pool_init.
> 2.ixgbevf rounds down the ("MBUF size" - RTE_PKTMBUF_HEADROOM) to the
> nearest 1K multiple when deciding on the
">" was missed in the following command.
echo 2048 > /sys/kernel/mm/dom0-mm/memsize-mB/memsize
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Murillo
> Sent: Thursday, October 30, 2014 6:27 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] DPDK xen-dom0 not
On FreeBSD, when initializing a secondary process,
EAL was complaining if there were ports not bound
to nic_uio module, exiting the application, which
should not happen, as this is expected behaviour,
and not an error
Signed-off-by: Pablo de Lara
---
lib/librte_eal/bsdapp/eal/eal_pci.c |2
Hi,
I'm seeing an unwanted behaviour in the receive flow of ixgbevf. While
using Jumbo frames and sending 4k+ bytes , the receive side breaks up the
packets into 2K buffers, and I receive 3 mbuffs per packet.
Im setting the .max_rx_pkt_len to 4.5K and the mempoll has 5K sized
elements?
Anything
Hi,
Apologies for the previous email format, I seems like I misconfigured something
on my email-client.
In case people find it annoying to read with extra line spaces, here is the RFC
with fixed formatting:
I would like to propose an update on the way the libraries are being built on
DPDK.
English is not my native language; please excuse typing errors.
I am new to DPDK with Xen, but I am able to run DPDK applications
without xen using hugepages. But when I try to do it in xen-dom0 it does
not work. Let me explain:
I have followed the intructions on the Getting Started Guide,
On Thu, Oct 30, 2014 at 12:23:09PM +0200, Alex Markuze wrote:
> Hi,
> I'm seeing an unwanted behaviour in the receive flow of ixgbevf. While
> using Jumbo frames and sending 4k+ bytes , the receive side breaks up the
> packets into 2K buffers, and I receive 3 mbuffs per packet.
>
> Im setting the
Some info about the issue for this patch:
https://wiki.ubuntu.com/ToolChain/CompilerFlags#Flags_passed_to_the_linker
https://wiki.debian.org/ToolChain/DSOLinking#Only_link_with_needed_libraries
Basically, Ubuntu GCC is always passing --as-needed to the linker which causes
some
Linking issues for
Ubuntu/Debian toolchain passes --as-needed flag to the linker by default.
Add --no-as-needed flag by default in linuxapp exec-env to ensure correct
linking.
Signed-off-by: Sergio Gonzalez Monroy
---
mk/exec-env/linuxapp/rte.vars.mk | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
On 10/30/2014 08:04 AM, Thomas Monjalon wrote:
> Hi Jincheng,
>
> 2014-10-24 15:07, Jincheng Miao:
>> From upstream kernel commit 3db2e9cd, strict_strto* serial functions
>> are removed. So that we should directly used kstrtoul instead.
> kstrtoul appeared in version 2.6.39 in commit
Since the commit 33e79bed3edc2bcf59 has fixed the issue in vector PMD,
and then it can receive jumbo frame by scatter-gather mode, so remove this
check.
Signed-off-by: Changchun Ouyang
---
examples/vhost/main.c | 7 ---
1 file changed, 7 deletions(-)
diff --git a/examples/vhost/main.c
On Thu, Oct 30, 2014 at 12:49:13AM +, Ouyang, Changchun wrote:
> Hi,
>
> > -Original Message-
> > From: Xie, Huawei
> > Sent: Thursday, October 30, 2014 7:26 AM
> > To: Ouyang, Changchun; dev at dpdk.org
> > Subject: RE: [dpdk-dev] [PATCH v2 3/5] vhost: enable promisc mode and
> >
On Thu, Oct 30, 2014 at 11:05:24AM +, Gonzalez Monroy, Sergio wrote:
> Some info about the issue for this patch:
> https://wiki.ubuntu.com/ToolChain/CompilerFlags#Flags_passed_to_the_linker
> https://wiki.debian.org/ToolChain/DSOLinking#Only_link_with_needed_libraries
>
> Basically, Ubuntu
On Thu, Oct 30, 2014 at 03:40:08AM +, Matthew Hall wrote:
> Signed-off-by: Matthew Hall
Acked-by: Bruce Richardson
> ---
> lib/librte_eal/linuxapp/eal/eal_pci.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/lib/librte_eal/linuxapp/eal/eal_pci.c
>
> From: Ouyang Changchun
> > Since the commit 33e79bed3edc2bcf59 has fixed the issue in vector PMD,
> > and then it can receive jumbo frame by scatter-gather mode, so remove this
> > check.
> >
> > Signed-off-by: Changchun Ouyang
>
> Acked-by: Huawei Xie
Applied
Thanks
--
Thomas
Hi,
I would like to propose an update on the way the libraries are being built on
DPDK.
Motivation/Issues:
- No agreement on libraries to build (separated, combined, different grouping).
- People having issues building their applications cannot rely on the
information given by ldd when
Hello All ,
I am currently trying to implement QOS scheduler using DPDK 1.6 . I have
configured 1 subport , 4096 pipes for the sub port and 4 TC's and 4 Queues .
Currently i am trying to send packets destined to single Queue of the
available 16 queues of one of the pipe .
Could some body
Please ignore this patch sent by mistake. It should be replaced by another
patch with the same name.
http://www.dpdk.org/ml/archives/dev/2014-October/007374.html
Sorry for the inconvenience.
> -Original Message-
> From: Wu, Jingjing
> Sent: Thursday, October 30, 2014 3:26 PM
> To: dev
Yes. I once had to write a cron job in Perl that monitored for unused regions
and deleted them before the SHM subsystem ran out of memory. Possible, yes, but
irritating.
--
Sent from my mobile device.
On October 30, 2014 3:18:56 AM PDT, Neil Horman
wrote:
>On Wed, Oct 29, 2014 at 08:23:25PM
Tested-by: Min Cao
Patch name: Support flow director programming on Fortville
Brief description: add flexible payload
Test Flag: Tested-by
Tester name:min.cao at intel.com
Result summary: total 3 cases, 3 passed, 0 failed
Test Case 1:
Tested-by: Min Cao
Patch name: Support flow director programming on Fortville
Brief description: add flexible payload
Test Flag: Tested-by
Tester name:min.cao at intel.com
Result summary: total 3 cases, 3 passed, 0 failed
Test Case 1:
On Wed, Oct 29, 2014 at 08:23:25PM -0700, Matthew Hall wrote:
> On Wed, Oct 29, 2014 at 11:32:12AM -0400, Neil Horman wrote:
> > >
> > Well, abnormal termination results in abnormal consequences. You expect
> > garbage to get left behind of a program crashes, so I wouldn't really worry
> > about
Acked-by Huawei Xie
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Ouyang Changchun
> Sent: Wednesday, October 29, 2014 7:11 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH] vhost: Remove the check between mergeable and
> vector pmd
>
> Since the
Signed-off-by: Matthew Hall
---
lib/librte_eal/linuxapp/eal/eal_pci.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/librte_eal/linuxapp/eal/eal_pci.c
b/lib/librte_eal/linuxapp/eal/eal_pci.c
index 5fe3961..ddb0535 100644
--- a/lib/librte_eal/linuxapp/eal/eal_pci.c
+++
Hi Jincheng,
2014-10-24 15:07, Jincheng Miao:
> From upstream kernel commit 3db2e9cd, strict_strto* serial functions
> are removed. So that we should directly used kstrtoul instead.
kstrtoul appeared in version 2.6.39 in commit 33ee3b2e2eb9b
("kstrto*: converting strings to integers done
Hi,
> -Original Message-
> From: Xie, Huawei
> Sent: Thursday, October 30, 2014 7:37 AM
> To: Ouyang, Changchun; Thomas Monjalon
> Cc: dev at dpdk.org
> Subject: RE: [dpdk-dev] [PATCH] vhost: Check descriptor number for vector
> Rx
>
>
>
> > -Original Message-
> > From: dev
> In the case where a userspace reports itself as Ubuntu, but the
> kernel isn't providing the expected version signature interface,
> turn off Ubuntu specializations.
>
> This situation happens often enough in development environments,
> and with multi-distribution build servers (e.g. chroot,
> > rte_ipv4_fragment_packet() and rte_ipv6_fragment packet()
> > call rte_pktmbuf_attach() to attach the segment of the original
> > packet to the segment of the new fragmented one. Such function
> > is not declared if RTE_MBUF_REFCNT is disabled, as it needs to
> > call rte_mbuf_refcnt_update,
Hi,
> -Original Message-
> From: Xie, Huawei
> Sent: Thursday, October 30, 2014 7:26 AM
> To: Ouyang, Changchun; dev at dpdk.org
> Subject: RE: [dpdk-dev] [PATCH v2 3/5] vhost: enable promisc mode and
> config VMDQ offload register for multicast feature
>
>
>
> > -Original
Hi:
I just change my DPDK version from DPDK 1.3 to DPDK 1.7.1.
After that I rebuild the kni module, and load it.
I can create KNI interface. But it time outs, when I try to turn it up.
sudo /sbin/ifconfig vEth_0 up
SIOCSIFFLAGS: Timer expired
Does Anybody have similar problem?
Thanks.
Hi Thomas,
Thanks for applying it.
This issue is reported by a customer.
We can reproduce it in Ubuntu14, on that environment, gcc will check any unused
result of function.
Thanks and regards,
Changchun
-Original Message-
From: Thomas Monjalon [mailto:thomas.monja...@6wind.com]
> > This patch set contains small fixes for issues with the vector PMD.
> > The issues and the fixes for them are described in each patch individually.
> >
> > Bruce Richardson (2):
> > ixgbe: remove static qualifier for thread safety
> > ixgbe: always perform vec RX setup if vpmd enabled
>
2014-10-29 14:39, Ouyang Changchun:
> It fixes this compilation complain: "error: ignoring return value of
> 'realpath',
> declared with attribute warn_unused_result [-Werror=unused-result]"
>
> Signed-off-by: Changchun Ouyang
Applied
Curiosity: how did you see this issue?
Thanks
--
Thomas
Please check with Yong Liu for the template of test report.
We discussed it together.
Thanks
2014-10-29 08:48, Fu, JingguoX:
> Sorry, not excepted, those are expected!
>
> Patch name: librte_vhost: Fix compilation issue
> Brief description:to fix compilation error on ubuntu14.04
2014-10-29 08:53, Wu, Jingjing:
> From: Zhang, Helin
> > For those registers and their relevant for early hardware only, they should
> > neither be defined nor be used. Thus i40e_register_x710_int.h should be
> > removed. As I40E_GLINT_CTL and its relevant is really needed, just define it
> >
2014-10-29 05:00, Chen, Jing D:
> From: Zhang, Helin
> > Inside NIC RX interrupt is needed for single RX descriptor
> > write back. The fix is to correct the wrong configuration
> > of register 'I40E_QINT_RQCTL'. In addition, several code
> > style fixes are added.
> > Note that interrupt will be
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Ouyang, Changchun
> Sent: Monday, October 27, 2014 6:56 AM
> To: Thomas Monjalon
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] vhost: Check descriptor number for vector Rx
>
> Hi Thomas,
>
> >
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Ouyang Changchun
> Sent: Sunday, October 26, 2014 8:46 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v2 3/5] vhost: enable promisc mode and config
> VMDQ offload register for multicast feature
>
>
74 matches
Mail list logo