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
---
doc/guides/nics/index.rst| 1 +
doc/guides/nics/overview
On Tue, Jan 26, 2016 at 7:58 PM, Thomas Monjalon
wrote:
> 2016-01-26 19:35, Santosh Shukla:
>> On Tue, Jan 26, 2016 at 6:30 PM, Thomas Monjalon
>> wrote:
>> > 2016-01-26 15:56, Santosh Shukla:
>> >> In my observation, currently virtio work for vfio-noiommu, that's why
>> >> said drv->kdrv need to
On Tue, Jan 26, 2016 at 9:14 PM, Thomas Monjalon
wrote:
> 2015-12-11 11:20, Aaron Conole:
>> DPDK ports are only detected during the EAL initialization. After that, any
>> new DPDK ports which are bound will not be visible to the application.
>>
>> The dpdk_nic_bind.py can be a bit more helpful to
2015-12-11 11:20, Aaron Conole:
> DPDK ports are only detected during the EAL initialization. After that, any
> new DPDK ports which are bound will not be visible to the application.
>
> The dpdk_nic_bind.py can be a bit more helpful to let users know that DPDK
> enabled applications will not find
On Tue, Jan 26, 2016 at 6:30 PM, Thomas Monjalon
wrote:
> 2016-01-26 15:56, Santosh Shukla:
>> On Mon, Jan 25, 2016 at 8:59 PM, Thomas Monjalon
>> wrote:
>> > 2016-01-21 22:47, Santosh Shukla:
>> >> On Thu, Jan 21, 2016 at 8:16 PM, Thomas Monjalon
>> >> wrote:
>> >> > 2016-01-21 17:34, Santosh S
I have to practically identical systems, same hypervisor on each (Centos
7.x). ?In one, I have a 57800 card which works fine with DPDK with
SRIOV. ?In the other, I have a 57810 card which doesn't work with SRIOV.
For the 57810 I have tracked this down to the status block in the VF
failing to be up
if the user wants to have rte_pktmbuf_pool_create() use an external mempool
handler, they simply define MEMPOOL_HANDLER_NAME to be the name of the
mempool handler they wish to use. May move this to config
Signed-off-by: David Hunt
---
lib/librte_mbuf/rte_mbuf.c | 11 +++
1 file changed,
Signed-off-by: David Hunt
---
app/test/Makefile | 1 +
app/test/test_ext_mempool.c | 474
2 files changed, 475 insertions(+)
create mode 100644 app/test/test_ext_mempool.c
diff --git a/app/test/Makefile b/app/test/Makefile
index ec33e1a..
adds a simple ring-based mempool handler using mallocs for each object
Signed-off-by: David Hunt
---
lib/librte_mempool/Makefile | 1 +
lib/librte_mempool/custom_mempool.c | 160
2 files changed, 161 insertions(+)
create mode 100644 lib/librte_memp
adds a simple stack based mempool handler
Signed-off-by: David Hunt
---
app/test/test_mempool_perf.c | 1 -
lib/librte_mempool/Makefile| 1 +
lib/librte_mempool/rte_mempool_stack.c | 167 +
3 files changed, 168 insertions(+), 1 deletion(-
Adds the new rte_mempool_create_ext api and callback mechanism for
external mempool handlers
Modifies the existing rte_mempool_create to set up the handler_idx to
the relevant mempool handler based on the handler name:
ring_sp_sc
ring_mp_mc
ring_sp_mc
ring_mp_sc
Si
Hi all on the list.
Here's a proposed patch for an external mempool manager
The External Mempool Manager is an extension to the mempool API that allows
users to add and use an external mempool manager, which allows external memory
subsystems such as external hardware memory management systems and
This patch adds new function rte_jobstats_abort. It marks *job* as finished
and time of this work will be add to management time instead of execution time.
This function should be used instead of rte_jobstats_finish if condition occure,
condition is defined by the application for example when rece
What about the RTE_VERIFY? I think, it's more appropriate here.
Otherwise, feel free to add:
Reviewed-by: Jan Viktorin
On Fri, 22 Jan 2016 15:06:57 +0100
David Marchand wrote:
> We are in static functions and those passed arguments can't be NULL.
>
> Signed-off-by: David Marchand
> ---
> li
On Fri, 22 Jan 2016 15:06:58 +0100
David Marchand wrote:
> Move these error logs and checks on detach capabilities in a common place.
>
> Signed-off-by: David Marchand
Reviewed-by: Jan Viktorin
2016-01-20 10:48, krytarowski at caviumnetworks.com:
> --- a/tools/dpdk_nic_bind.py
> +++ b/tools/dpdk_nic_bind.py
> -for line in loaded_mods:
> +try:
> +# Get list of syfs modules, some of them might be builtin and merge
> with mods
Please could you explain this comment?
Is it re
2016-01-19 17:35, Kamil Rytarowski:
>
> W dniu 18.01.2016 o 15:32, Thomas Monjalon pisze:
> > Hi Kamil,
> >
> > 2015-12-09 14:19, Kamil Rytarowski:
> >> Currently dpdk_nic_bind.py detects Linux kernel modules via reading
> >> /proc/modules. Built-in ones aren't listed there and therefore they are
On Mon, Jan 25, 2016 at 8:59 PM, Thomas Monjalon
wrote:
> 2016-01-21 22:47, Santosh Shukla:
>> On Thu, Jan 21, 2016 at 8:16 PM, Thomas Monjalon
>> wrote:
>> > 2016-01-21 17:34, Santosh Shukla:
>> >> On Thu, Jan 21, 2016 at 4:58 PM, Thomas Monjalon
>> >> wrote:
>> >> > 2016-01-21 16:43, Santosh S
2016-01-26 19:35, Santosh Shukla:
> On Tue, Jan 26, 2016 at 6:30 PM, Thomas Monjalon
> wrote:
> > 2016-01-26 15:56, Santosh Shukla:
> >> In my observation, currently virtio work for vfio-noiommu, that's why
> >> said drv->kdrv need to know vfio mode.
> >
> > It is your observation. It may change i
This patch extended commands for filter's input set changing.
It added vlan as filter's input fields.
Signed-off-by: Jingjing Wu
---
app/test-pmd/cmdline.c | 6 +++---
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 2 +-
2 files changed, 4 insertions(+), 4 deletions(-)
diff
This patch extended flow director to select vlan id
as filter's input set and program the filter rule with vlan id.
Signed-off-by: Jingjing Wu
---
doc/guides/rel_notes/release_2_3.rst | 1 +
drivers/net/i40e/i40e_ethdev.c | 11 +++
drivers/net/i40e/i40e_fdir.c | 9 +++
From: Andrey Chilikin
This patch adds missing VLAN bitmask for inner frame in case of
tunneling and fixes VLAN tags bitmasks for single or outer frame
in case of tunneling.
Signed-off-by: Andrey Chilikin
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/i40e_ethdev.c | 12 +++-
1 file c
This patch extended commands for filter's input set changing.
It added GRE/Vxlan Tunnel as filter's input fields.
Signed-off-by: Jingjing Wu
---
app/test-pmd/cmdline.c | 27 +--
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 22 --
This patch extended flow director to select Vxlan/GRE tunnel ID
as filter's input set and program the filter rule with the defined
tunnel type.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/i40e_ethdev.c | 11
drivers/net/i40e/i40e_fdir.c | 125 ++---
This patch changed rte_eth_fdir_flow from union to struct to
support more packets formats, for example, Vxlan and GRE tunnel
packets with IP inner frame.
This patch also add new RTE_FDIR_TUNNEL_TYPE_GRE enum.
Signed-off-by: Jingjing Wu
---
doc/guides/rel_notes/release_2_3.rst | 4
lib/lib
This patch extended commands for filter's input set changing.
It added tos, protocol and ttl as filter's input fields, and
remove the words selection from flex payloads for flow director.
Signed-off-by: Jingjing Wu
---
app/test-pmd/cmdline.c | 100 ++-
This patch extended flow director to select more IP Header fields
as filter input set.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/i40e_ethdev.c | 69 ++
drivers/net/i40e/i40e_fdir.c | 26 +++-
2 files changed, 75 insertions(+), 20 deleti
This patch added a new function to set the input set to default
when initialization.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/i40e_ethdev.c | 56 ++
1 file changed, 56 insertions(+)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40
In this patch, flex payload is removed from valid fdir input set
values. It is because all flex payload configuration can be set
in struct rte_fdir_conf during device configure phase.
And it is a more flexible configuration including flexpayload's
selection, input set selection by word and mask set
This patch split function for input set changing of hash
and fdir to avoid multiple check on different situation.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/i40e_ethdev.c | 233 +
drivers/net/i40e/i40e_ethdev.h | 11 +-
drivers/net/i40e/i40e_fdir.c
This patch added RTE_ETH_INPUT_SET_L3_IP4_TTL,
RTE_ETH_INPUT_SET_L3_IP6_HOP_LIMITS input field type and extended
struct rte_eth_ipv4_flow and rte_eth_ipv6_flow to support filtering
by tos, protocol and ttl.
Signed-off-by: Jingjing Wu
---
lib/librte_ether/rte_eth_ctrl.h | 8
1 file chang
This patch set extends flow director to support filtering by
additional fields below in i40e driver:
- TOS, Protocol and TTL in IP header
- Tunnel id if NVGRE/GRE/VxLAN packets
- single vlan or inner vlan
Jingjing Wu (12):
ethdev: extend flow director to support input set selection
i40e: sp
Ping... This patch has been sitting^Hrotting for a bit over a month.
> DPDK ports are only detected during the EAL initialization. After that, any
> new DPDK ports which are bound will not be visible to the application.
>
> The dpdk_nic_bind.py can be a bit more helpful to let users know that DPDK
This patch extended commands for filter's input set changing.
It added vlan as filter's input fields.
Signed-off-by: Jingjing Wu
---
app/test-pmd/cmdline.c | 6 +++---
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 2 +-
2 files changed, 4 insertions(+), 4 deletions(-)
diff
This patch extended flow director to select vlan id
as filter's input set and program the filter rule with vlan id.
Signed-off-by: Jingjing Wu
---
doc/guides/rel_notes/release_2_3.rst | 1 +
drivers/net/i40e/i40e_ethdev.c | 11 +++
drivers/net/i40e/i40e_fdir.c | 9 +++
From: Andrey Chilikin
This patch adds missing VLAN bitmask for inner frame in case of
tunneling and fixes VLAN tags bitmasks for single or outer frame
in case of tunneling.
Signed-off-by: Andrey Chilikin
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/i40e_ethdev.c | 12 +++-
1 file c
This patch extended commands for filter's input set changing.
It added GRE/Vxlan Tunnel as filter's input fields.
Signed-off-by: Jingjing Wu
---
app/test-pmd/cmdline.c | 27 +--
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 22 --
This patch extended flow director to select Vxlan/GRE tunnel ID
as filter's input set and program the filter rule with the defined
tunnel type.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/i40e_ethdev.c | 11
drivers/net/i40e/i40e_fdir.c | 125 ++---
This patch changed rte_eth_fdir_flow from union to struct to
support more packets formats, for example, Vxlan and GRE tunnel
packets with IP inner frame.
This patch also add new RTE_FDIR_TUNNEL_TYPE_GRE enum.
Signed-off-by: Jingjing Wu
---
doc/guides/rel_notes/release_2_3.rst | 4
lib/lib
This patch extended commands for filter's input set changing.
It added tos, protocol and ttl as filter's input fields, and
remove the words selection from flex payloads for flow director.
Signed-off-by: Jingjing Wu
---
app/test-pmd/cmdline.c | 100 ++-
This patch extended flow director to select more IP Header fields
as filter input set.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/i40e_ethdev.c | 69 ++
drivers/net/i40e/i40e_fdir.c | 26 +++-
2 files changed, 75 insertions(+), 20 deleti
This patch added a new function to set the input set to default
when initialization.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/i40e_ethdev.c | 56 ++
1 file changed, 56 insertions(+)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40
In this patch, flex payload is removed from valid fdir input set
values. It is because all flex payload configuration can be set
in struct rte_fdir_conf during device configure phase.
And it is a more flexible configuration including flexpayload's
selection, input set selection by word and mask set
This patch splited function for input set changing of hash
and fdir to avoid multiple check on different situation.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/i40e_ethdev.c | 233 +
drivers/net/i40e/i40e_ethdev.h | 11 +-
drivers/net/i40e/i40e_fdir.c
This patch added RTE_ETH_INPUT_SET_L3_IP4_TTL,
RTE_ETH_INPUT_SET_L3_IP6_HOP_LIMITS input field type and extended
struct rte_eth_ipv4_flow and rte_eth_ipv6_flow to support filtering
by tos, protocol and ttl.
Signed-off-by: Jingjing Wu
---
lib/librte_ether/rte_eth_ctrl.h | 8
1 file chang
This patch set extends flow director to support filtering by
additional fields below in i40e driver:
- TOS, Protocol and TTL in IP header
- Tunnel id if NVGRE/GRE/VxLAN packets
- single vlan or inner vlan
Jingjing Wu (12):
ethdev: extend flow director to support input set selection
i40e: sp
Hi Michael,
On 1/26/2016 2:02 PM, Qiu, Michael wrote:
> On 1/11/2016 2:43 AM, Tan, Jianfeng wrote:
...
>>
>> f. Used with vhost-net
>> $: modprobe vhost
>> $: modprobe vhost-net
>> $: docker run -i -t --privileged \
>> -v /dev/vhost-net:/dev/vhost-net \
>> -v /dev/net/tun:/dev/net/tun \
Jumbo frames are generally handled by link lists (but called something else) of
mbufs.
Enabling jumbo frames for the device driver should enable the right portion of
the driver which handles the linked lists.
Don't make the mbufs huge.
Mike
-Original Message-
From: dev [mailto:dev-bou
2016-01-26 15:56, Santosh Shukla:
> On Mon, Jan 25, 2016 at 8:59 PM, Thomas Monjalon
> wrote:
> > 2016-01-21 22:47, Santosh Shukla:
> >> On Thu, Jan 21, 2016 at 8:16 PM, Thomas Monjalon
> >> wrote:
> >> > 2016-01-21 17:34, Santosh Shukla:
> >> >> On Thu, Jan 21, 2016 at 4:58 PM, Thomas Monjalon
>
On Mon, Jan 25, 2016 at 04:15:28PM -0800, Saurabh Mishra wrote:
> Hi Bruce --
>
> >The sharing of the mbuf pool is not an issue, but sharing of rx/tx queues
> is.
> >The ethdev queues are not multi-thread safe, so to share a queue between
> processes
> >or threads, you need to put in locks or othe
On Mon, Jan 25, 2016 at 02:31:05AM +, Wang, Xiao W wrote:
> Hi Bruce,
>
> > -Original Message-
> > From: Richardson, Bruce
> > Sent: Saturday, January 23, 2016 5:32 AM
> > To: Wang, Xiao W
> > Cc: Chen, Jing D ; dev at dpdk.org
> > Subject: Re: [dpdk-dev] [PATCH 15/16] fm10k: use defa
On 2016/01/25 19:29, Xie, Huawei wrote:
> On 1/21/2016 7:09 PM, Tetsuya Mukawa wrote:
>> +#define PCI_CONFIG_ADDR(_bus, _device, _function, _offset) ( \
>> +(1 << 31) | ((_bus) & 0xff) << 16 | ((_device) & 0x1f) << 11 | \
>> +((_function) & 0xf) << 8 | ((_offset) & 0xfc))
> (_function) & 0x
On 2016/01/25 19:17, Xie, Huawei wrote:
> On 1/21/2016 7:09 PM, Tetsuya Mukawa wrote:
>> +static void
>> +qtest_handle_one_message(struct qtest_session *s, char *buf)
>> +{
>> +int ret;
>> +
>> +if (strncmp(buf, interrupt_message, strlen(interrupt_message)) == 0) {
>> +if (rte_a
On 2016/01/25 19:15, Xie, Huawei wrote:
> On 1/22/2016 6:38 PM, Tetsuya Mukawa wrote:
>> On 2016/01/22 17:14, Xie, Huawei wrote:
>>> On 1/21/2016 7:09 PM, Tetsuya Mukawa wrote:
virtio: Extend virtio-net PMD to support container environment
The patch adds a new virtio-net PMD configur
Saurabh:
It sounds like you benchmarked Apache using Jumbo Packets, but not the
DPDK app using large mbufs. Those are two entirely different issues.
You should be able to write your packet inspection routines to work with
the mbuf chains, rather than copying them into a larger buffer (although
> From: Qiu, Michael
> > Whatever work the secondary was performing (in its own address space)
> > won't be directly changed by the primary being killed, because the
> > shared config and hugepages stay (EAL "cleans up" when the primary
> > is re-launched, not on quit).
>
> OK, when primary quit
On Mon, Jan 25, 2016 at 11:44:59AM +, Van Haaren, Harry wrote:
> > From: Richardson, Bruce
> > The details of what the config file is should largely be hidden from the
> > user
> > IMHO.
>
> Agreed, however hiding it totally removes the flexibility of waiting for a
> primary
> that is starti
On 1/26/2016 5:04 PM, Van Haaren, Harry wrote:
>> From: Qiu, Michael
>> On 1/25/2016 7:51 PM, Van Haaren, Harry wrote:
>>> Not really, the secondary process will need some CPU,
>>> however it can sleep so it doesn't have to use 100% of it.
>>> It shouldn't be run on a core that is used by the prima
Hi Tim,
On 01/22/2016 07:19 PM, O'Driscoll, Tim wrote:
> At the community call we held on governance in December, we agreed that a few
> of us would work with the Linux Foundation on a proposal for a light-weight
> governance of DPDK. This would include things like management of DPDK events,
>
ping?
W dniu 20.01.2016 o 10:48, krytarowski at caviumnetworks.com pisze:
> From: Kamil Rytarowski
>
> Currently dpdk_nic_bind.py 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
On 12/3/2015 2:03 PM, Yuanhan Liu wrote:
> Signed-off-by: Yuanhan Liu
> ---
> lib/librte_vhost/vhost_rxtx.c | 287
> +-
> 1 file changed, 113 insertions(+), 174 deletions(-)
Prefer to unroll copy_mbuf_to_desc and your COPY macro. It prevents us
processing
Hi Neil,
On 1/25/2016 9:46 PM, Neil Horman wrote:
> On Mon, Jan 25, 2016 at 02:49:53AM +0800, Jianfeng Tan wrote:
...
>> --
>> 2.1.4
>>
>>
>
> This doesn't make a whole lot of sense, for several reasons:
>
> 1) Applications, as a general rule shouldn't be interrogating the cgroups
> interface at
Saurabh:
Raising the mbuf size will make the packet handling for large packets
slightly more efficient, but it will use much more memory unless the
great majority of the packets you are handling are of the jumbo size.
Using more memory has its own costs. In order to evaluate this design
choice
On Tue, Jan 26, 2016 at 10:22:18AM +0800, Tan, Jianfeng wrote:
>
> Hi Neil,
>
> On 1/25/2016 9:46 PM, Neil Horman wrote:
> >On Mon, Jan 25, 2016 at 02:49:53AM +0800, Jianfeng Tan wrote:
> ...
> >>--
> >>2.1.4
> >>
> >>
> >
> >This doesn't make a whole lot of sense, for several reasons:
> >
> >1)
Hi Lawrence --
>It sounds like you benchmarked Apache using Jumbo Packets, but not the
DPDK app using large mbufs.
>Those are two entirely different issues.
I meant I ran Apache benchmark between two guest VMs through our
data-processing VM which is using DPDK.
I saw 3x better performance with 1
> From: Qiu, Michael
> On 1/25/2016 7:51 PM, Van Haaren, Harry wrote:
> > Not really, the secondary process will need some CPU,
> > however it can sleep so it doesn't have to use 100% of it.
> > It shouldn't be run on a core that is used by the primary
> > for packet-forwarding though - that will i
Hi,
1. Take care of unit size of mempool for mbuf.
2. Call rte_eth_dev_set_mtu() for each interface.
Note that some PMDs does not supported change MTU.
On 2016/01/26 6:02, Saurabh Mishra wrote:
> Hi,
>
> We wanted to use 10400 bytes size of each rte_mbuf to enable Jumbo frames.
> Do you guys
Hi,
Since we do full content inspection, we will end up coalescing mbuf chains
into one before inspecting the packet which would require allocating
another buffer of larger size.
I am inclined towards larger size mbuf for this reason.
I have benchmarked a bit using apache benchmark and we see 3x
On 1/11/2016 2:43 AM, Tan, Jianfeng wrote:
> This patchset is to provide high performance networking interface (virtio)
> for container-based DPDK applications. The way of starting DPDK apps in
> containers with ownership of NIC devices exclusively is beyond the scope.
> The basic idea here is to p
Tested-by: Yong Liu
- Tested Commit: a38e5ec15e3fe615b94f3cc5edca5974dab325ab
- OS: Fedora20 3.11.10-301.fc20.x86_64
- GCC: gcc version 4.8.3 20140911
- CPU: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz
- NIC: Intel Corporation Device RedrockCanyou [8086:15a4]
- Default x86_64-native-linuxapp-gcc co
On 1/25/2016 7:51 PM, Van Haaren, Harry wrote:
>> From: Qiu, Michael
>> Subject: Re: [dpdk-dev] [PATCH] eal: add function to check if primary proc
>> alive
>>
>> So secondary will waste a whole lcore to do such polling?
> Not really, the secondary process will need some CPU,
> however it can sleep
> >
> > Thanks for your comments. You are correct, I removed the
> > I40E_INSET_FLEX_PAYLOAD from valid fdir input set values, and this is
> > one reason why I splited function for input set change of hash and and
> > it is because all flex payload configuration can be set in struct
> > rte_fdir_c
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, January 25, 2016 5:17 PM
> To: Zhang, Helin
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 1/3] i40e: enable extended tag
>
> > From: Thomas Monjalon [mailto:thomas.monjalon at 6win
73 matches
Mail list logo
