The ACL library is used to perform an N-tuple search over a set of rules with
multiple categories and find the best match for each category.
Signed-off-by: Konstantin Ananyev
---
config/common_linuxapp |6 +
lib/librte_acl/Makefile | 60 +
lib/librte_acl/acl.h
The ACL library is used to perform an N-tuple search over a set of rules
with multiple categories and find the best match (highest priority)
for each category.
This code was previously released under a proprietary license,
but is now being released under a BSD license to allow its
integration
From: Ivan Boule
Split input error stats to have a better understanding of why packets have been
dropped.
Keep ierrors field untouched for backward compatibility.
Signed-off-by: Ivan Boule
Signed-off-by: David Marchand
---
app/test-pmd/config.c | 24
Hi all,
I measured a roundtrip latency (using Spirent traffic generator) of sending
64B packets over a 10GbE to DPDK, and DPDK does nothing but simply forward
back to the incoming port (l3fwd without any lookup code, i.e., dstport =
port_id).
However, to my surprise, the average latency was
Signed-off-by: Konstantin Ananyev
---
examples/l3fwd/main.c | 467 ++---
1 files changed, 445 insertions(+), 22 deletions(-)
diff --git a/examples/l3fwd/main.c b/examples/l3fwd/main.c
index 8ee1af9..29f115a 100755
--- a/examples/l3fwd/main.c
+++
Introduce rte_lpm_lookupx4():
- Allows to lookup four IP addresses in an LPM table.
- Uses SSE instrincts.
Signed-off-by: Konstantin Ananyev
---
app/test/test_lpm.c | 70 ++
lib/librte_eal/common/Makefile |1 +
With latest HW and optimised RX/TX path there is a huge gap between
tespmd iofwd and l3fwd performance results.
So there is an attempt to optimise l3fwd LPM code path and reduce the gap:
- Instead of processing each input packet up to completion -
divide packet processing into several
2014-04-17 17:00, Thomas Monjalon:
> Hi Konstantin,
>
> 2014-04-07 08:53, Ananyev, Konstantin:
> > Yep indeed, there is a bug in eth_pcap_tx() that can cause mbuf
> > corruption.
> > I think it should be something like that instead:
> >
> > --- a/lib/librte_pmd_pcap/rte_eth_pcap.c
> > +++
2014-05-22 15:39, Ananyev, Konstantin:
> Hi Thomas,
> I believe I already resent it with signed-off-by.
Oh yes, you're right.
Sorry for the noise.
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Thursday, May 22, 2014 4:33 PM
> To: Ananyev,
Hi Konstantin,
2014-05-06 11:48, Konstantin Ananyev:
> When latest Linux ixgbe PF is used, and DPDK VF is used in DPDK application,
> jumbo frames are not received.
> Also - if Linux ixgbe PF has MTU set to 1500 (default),
> then normal sized packets can be received by DPDK VF.
> However, if
Hi Konstantin,
2014-05-06 12:28, Konstantin Ananyev:
> e1000_vfadapt type corresponds to 82576 VF devices,
> check e1000_set_mac_type() for more details.
You forgot "Signed-off-by:" line.
Please resend.
--
Thomas
Hi Changchun,
On 05/22/2014 04:44 PM, Ouyang, Changchun wrote:
> Hi Ivan
> For this one, it seems long story for that...
> In short,
> Some customer have such kind of requirement,
> they want to repeatedly start(rte_dev_start) and stop(rte_dev_stop) the port
> for RX and TX, but they find
>
Hi Changchun,
Please, it is preferred to answer below the question.
2014-05-20 01:14, Ouyang, Changchun:
> So "[PMD] [VHOST]" in the title should be removed in the cover letter,
> right? And in each separate patch letter, it could use "ixgbe:" or
> "examples/vhost:", instead of "[PMD] [VHOST]"
Hi Jijiang,
Some cosmetic comments:
Referring to previous patch about Xen Dom0, prefix "xen:" should be used
in commit title. So it should be something like that:
xen: reserve memory at init
Maybe it could need to be splitted in 2 patches. So the other one would be:
xen: reduce
Hello,
> I measured a roundtrip latency (using Spirent traffic generator) of sending
> 64B packets over a 10GbE to DPDK, and DPDK does nothing but simply forward
> back to the incoming port (l3fwd without any lookup code, i.e., dstport =
> port_id).
> However, to my surprise, the average
There were many comments on the first version of this serie,
and there is no comment on this second version.
So I understand everyone is OK with it and it can be applied in the git tree.
I'll wait 1 day more. Feel free to comment or acknowledge.
--
Thomas
2014-05-22 11:35, Helin Zhang:
> Error of "implicit-function-declaration" can be seen when building KNI
> kernel module on Linux kernel 3.6.10 platform, as follows.
> "x86_64-native-linuxapp-gcc/build/lib/librte_eal/linuxapp/kni/igb_ethtool.c:
> In function \u2018igb_get_eee\u2019:
>
Hi Prashant,
2014-05-15 20:14, Prashant Upadhyaya:
> Hi,
>
> I recently picked the 1.6.0r2 from dpdk.org and tried to compile it the
> usual way and ran into the following compilation error.
>
> I am aware I can sidestep these by getting the compiler to treat them as
> warnings, but these did
2014-05-14 09:25, Stephen Hemminger:
> These patches address a number of issues found during development
> and testing of QoS in Vyatta 5600 router.
Acked-by: Thomas Monjalon
Applied for version 1.7.0.
Thanks
--
Thomas
Thomas,
It's because we haven't gotten to testing the patch yet, and figuring out all
the problems. Putting it in and modifying MBUF needs a bit of time - one other
option that I've looked at is to let the transmit offload parts (except for the
VLAN) flow onto the second cache line. That
Hi Thomas,
I planned to, but very unlikely that it would happen that week.
Thanks
Konstantin
-Original Message-
From: Thomas Monjalon [mailto:thomas.monja...@6wind.com]
Sent: Thursday, May 22, 2014 4:47 PM
To: Ananyev, Konstantin
Cc: dev at dpdk.org
Subject: Re: [dpdk-dev]
Hi Stephen,
2014-05-14 09:34, Stephen Hemminger:
> The igb_uio driver does not handle interrupt mode correctly.
> It doesn't do the right thing and start with the most desired
> mode and fall back to legacy modes as needed.
I'm not sure to understand what you describe.
> It also has a custom
Hi, Thomas
Thanks very much for your guiding!
Best regards,
Changchun
-Original Message-
From: Thomas Monjalon [mailto:thomas.monja...@6wind.com]
Sent: Thursday, May 22, 2014 11:29 PM
To: Ouyang, Changchun
Cc: dev at dpdk.org
Subject: Re: [dpdk-dev] [PATCH] [PMD] [VHOST] Revert
Hi Thomas,
I believe I already resent it with signed-off-by.
Thanks
Konstantin
-Original Message-
From: Thomas Monjalon [mailto:thomas.monja...@6wind.com]
Sent: Thursday, May 22, 2014 4:33 PM
To: Ananyev, Konstantin
Cc: dev at dpdk.org
Subject: Re: [dpdk-dev] [PATCH] Use proper mac type
2014-05-22 13:24, Burakov, Anatoly:
> Hi Thomas,
>
> > I wonder if we could replace igb_uio by uio_pci_generic?
>
> Can it do DMA or IOMMU support? Not even VFIO does everything we need, you
> may have noticed that I have to go to PCI config space to enable bus
> mastering. I don't think that
2014-05-19 16:51, Anatoly Burakov:
> Support for loading/unloading VFIO drivers, binding/unbinding
> devices to/from VFIO, also setting up correct userspace permissions.
>
> Signed-off-by: Anatoly Burakov
> - ${RTE_SDK}/tools/igb_uio_bind.py --status
> +
Please, could you remove trailing whitespaces from your patch?
Thanks
--
Thomas
2014-05-19 16:51, Anatoly Burakov:
> This makes it possible to run DPDK without hugepage memory when VFIO
> is used, as VFIO uses virtual addresses to set up DMA mappings.
>
> Signed-off-by: Anatoly Burakov
> - addr = malloc(internal_config.memory);
> + addr = mmap(NULL,
2014-05-22 12:37, Burakov, Anatoly:
> > Yes, in some environments, it could be easier to be able to configure
> > devices directly on application command line instead of having to call a
> > python script. I think having a clear and extendable syntax to configure
> > devices in command line could
Hi Ivan
For this one, it seems long story for that...
In short,
Some customer have such kind of requirement,
they want to repeatedly start(rte_dev_start) and stop(rte_dev_stop) the port
for RX and TX, but they find
after several times start and stop, the RX and TX can't work well even the port
2014-05-19 16:51, Anatoly Burakov:
> Unlike igb_uio, VFIO interrupt type is not set by kernel module
> parameters but is set up via ioctl() calls at runtime. This warrants
> a new EAL command-line parameter. It will have no effect if VFIO is
> not compiled, but will set VFIO interrupt type to
2014-05-22 12:06, Burakov, Anatoly:
> > We should discuss a way to request igb_uio or VFIO binding of a device.
>
> Why? The device can either be bound to VFIO or igb_uio. So unless you want
> binding code in DPDK EAL (to avoid which the
> pci_unbind/igb_uio_bind/dpdk_bind script was created in
This patch adds commands to test the functionality of administrative link up
and down of PMD in testpmd.
Signed-off-by: Ouyang Changchun
---
app/test-pmd/cmdline.c | 78 ++
app/test-pmd/testpmd.c | 14 +
app/test-pmd/testpmd.h | 2 ++
3
This patch implements the functionality of administrative link up and down in
IXGBE PMD.
Signed-off-by: Ouyang Changchun
---
lib/librte_pmd_ixgbe/ixgbe_ethdev.c | 58 +
1 file changed, 58 insertions(+)
diff --git a/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
This patch addes API to support administrative link up and down.
Signed-off-by: Ouyang Changchun
---
lib/librte_ether/rte_ethdev.c | 38 ++
lib/librte_ether/rte_ethdev.h | 34 ++
2 files changed, 72 insertions(+)
diff --git
Hi Anatoly,
It seems to be the main patch, so I have many comments.
2014-05-19 16:51, Anatoly Burakov:
> VFIO is kernel 3.6+ only, and so is only compiled when DPDK config
> option CONFIG_RTE_EAL_VFIO is enabled, and kernel 3.6 or higher is
> detected, thus preventing compile failures on older
Hi Thomas,
> Yes I agree. But I stopped on the name for another thing: it's not really
> specific to vfio. Actually, vfio uses it for synchronization. But wouldn't it
> be
> more generic?
OK, _mp_sync it is then.
Best regards,
Anatoly Burakov
DPDK SW Engineer
Hi Thomas,
> I don't know if there is something defined in a Linux header which could help
> to check if VFIO is supported. But in general, it's better to check for a
> macro
> belonging to the feature instead of checking kernel version.
Not sure VFIO defines any macros anywhere, but I'll look
Hi Thomas,
> In case VFIO is backported on older kernel, it would be better to check a
> related macro instead of Linux version.
Not sure I follow. What is the "related macro" you're referring to?
Best regards,
Anatoly Burakov
DPDK SW Engineer
Hi Thomas,
> Yes, in some environments, it could be easier to be able to configure devices
> directly on application command line instead of having to call a python
> script.
> I think having a clear and extendable syntax to configure devices in command
> line could greatly improve usability.
Hi Thomas,
> How did you test this feature?
> Did you see some performance differences with igb_uio?
The same way everything else is tested - bind a NIC to the driver and see if it
works :-)
As for performance differences, potentially it can be degraded a bit because of
mandatory IOMMU
Error of "implicit-function-declaration" can be seen when building KNI
kernel module on Linux kernel 3.6.10 platform, as follows.
"x86_64-native-linuxapp-gcc/build/lib/librte_eal/linuxapp/kni/igb_ethtool.c:
In function \u2018igb_get_eee\u2019:
Hi Mr. Kim,
Please review version v1.0, which is going to be release soon. This version
need patch up with openstack patch.
https://01.org/packet-processing/intel%C2%AE-dpdk-vswitch-and-openstack-patches
Regards,
Choi, Sy Jong
Platform Application Engineer
-Original Message-
From:
Bug: When a timer is running
- if rte_timer_stop is called, the pending decrement is
skipped (decremented only if the timer is pending) and due
to the update flag the future processing is skipped so the
timer is counted as pending while it is stopped. - the same
applies when rte_timer_reset
In ethdev, it will ignore setting maximum packet length to less than 1518.
The changes is to fix it and let less than 1518 can be really set for
maximum packet length.
Signed-off-by: Helin Zhang
Signed-off-by: Mark Chen
---
lib/librte_ether/rte_ethdev.c | 10 +++---
1 file changed, 7
Sys files of 'extended_tag' and 'max_read_request_size' have been
supported in igb_uio, and can be changed during probing PCI. Three
items in configuration files are needed to support them at compile
time. Those three items are,
- CONFIG_RTE_PCI_CONFIG
- CONFIG_RTE_PCI_EXTENDED_TAG
-
Sys files of 'extended_tag' and 'max_read_request_size' are supported by
igb_uio. Reading or writing them to enable/disable 'Extended Tag' or
reset 'Max Read Request Size' automatically according to the configurations
are added.
Signed-off-by: Helin Zhang
Signed-off-by: Mark Chen
---
Enabling 'Extended Tag' and resetting 'Max Read Request Size' in PCI config
space
have big impacts to i40e performance. They cannot be changed on some BIOS
implementations, though can on others. Two sys files of 'extended_tag' and
'max_read_request_size' are added to support changing them by
As i40e PMD need to know the L3 packet type for TX checksum offloading,
modifications in testpmd should be done for that.
Signed-off-by: Helin Zhang
Signed-off-by: Mark Chen
---
app/test-pmd/csumonly.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/app/test-pmd/csumonly.c
As hash flags of RSS have been enlarged from 16 bits to 64 bits to support
more possible types in all PMDs, and new macro of ETH_RSS_IP has been defined
to cover all IP hash flags of RSS. That macro should be used in dpdk_qat
example application to support all PMDs.
Signed-off-by: Helin Zhang
As hash flags of RSS have been enlarged from 16 bits to 64 bits to support
more possible types in all PMDs, and new macro of ETH_RSS_IP has been defined
to cover all IP hash flags of RSS. That macro should be used in ip_reassembly
example application to support all PMDs.
Signed-off-by: Helin
As hash flags of RSS have been enlarged from 16 bits to 64 bits to support
more possible types in all PMDs, and new macro of ETH_RSS_IP has been defined
to cover all IP hash flags of RSS. That macro should be used in l3fwd-power
example application to support all PMDs.
Signed-off-by: Helin Zhang
As hash flags of RSS have been enlarged from 16 bits to 64 bits to support
more possible types in all PMDs, and new macro of ETH_RSS_IP has been defined
to cover all IP hash flags of RSS. That macro should be used in l3fwd example
application to support all PMDs.
Signed-off-by: Helin Zhang
As hash flags of RSS have been enlarged from 16 bits to 64 bits to support
more possible types in all PMDs, and new macro of ETH_RSS_IP has been defined
to cover all IP hash flags of RSS. That macro should be used in qos_meter
example application to support all PMDs.
Signed-off-by: Helin Zhang
As the hash flags of RSS has been enlarged from 16 bits to 64 bits in
'struct rte_eth_rss_conf' in rte_ethdev.h, the size of it in testpmd
should be enlarged as well. In addition, macro of ETH_RSS_IP is used
to replace all IP hash flags of RSS.
Signed-off-by: Helin Zhang
Signed-off-by: Mark Chen
As the hash flags of RSS has been enlarged from 16 bits to 64 bits in
'struct rte_eth_rss_conf' in rte_ethdev.h, the size of it in vmxnet3 should
be enlarged as well. In addition, the flags for vmxnet3 only should be masked,
as there are flags for others in that 64 bits.
Signed-off-by: Helin
As the hash flags of RSS has been enlarged from 16 bits to 64 bits in
'struct rte_eth_rss_conf' in rte_ethdev.h, the size of it in ixgbe should be
enlarged as well. In addition, the flags for ixgbe only should be masked,
as there are flags for others in that 64 bits.
Signed-off-by: Helin Zhang
As the hash flags of RSS has been enlarged from 16 bits to 64 bits in
'struct rte_eth_rss_conf' in rte_ethdev.h, the size of it in e1000 should be
enlarged as well. In addition, the flags for e1000 only should be masked,
as there are flags for others in that 64 bits.
Signed-off-by: Helin Zhang
Add PMD source files to support Intel(R) 40G Ethernet Controllers.
The new files are,
i40e_osdep.h
i40e_ethdev.c
i40e_ethdev.h
i40e_ethdev_vf.c
i40e_logs.h
i40e_pf.c
i40e_pf.h
i40e_rxtx.c
i40e_rxtx.h
Makefile
Signed-off-by: Helin Zhang
Signed-off-by: Mark Chen
---
lib/librte_pmd_i40e/Makefile
Add shared code source files to support basic operations to be called in
poll mode driver. The new files are,
i40e_adminq.c
i40e_adminq.h
i40e_adminq_cmd.h
i40e_alloc.h
i40e_common.c
i40e_dcb.c
i40e_dcb.h
i40e_diag.c
i40e_diag.h
i40e_hmc.c
i40e_hmc.h
i40e_lan_hmc.c
i40e_lan_hmc.h
i40e_nvm.c
60 matches
Mail list logo