The patch c344eab3ee has moved the hardware definition of CPU flags.
Now the functions checking these hardware flags are also moved.
The function rte_cpu_get_flag_enabled() is no more inline.
The benefits are:
- remove rte_cpu_feature_table from the ABI (recently added)
- hide hardware details
The structure feature_entry does not need leaf/subleaf
which were copied from x86 CPUID implementation.
On x86, a valid flag is detected with the non-zero leaf value.
This check is replaced by a check with a dummy "none" register.
Signed-off-by: Thomas Monjalon
---
Hello Thomas,
the patch set fails when compiling for ARMv7 with the following (full
build.log is attached):
/var/lib/jenkins/jobs/dpdk-armv7-testing/workspace/lib/librte_eal/common/arch/arm/rte_cpuflags.c:114:22:
error: ?REG_PLATFORM? undeclared here (not in a function)
FEAT_DEF(V7L,
Hi Bruce,
> -Original Message-
> From: Richardson, Bruce
> Sent: Tuesday, February 2, 2016 8:03 PM
> To: Lu, Wenzhuo
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 2/7] lib/librte_ether: support l2 tunnel
> config
>
> On Tue, Feb 02, 2016 at 02:57:00PM +0800, Wenzhuo Lu
> -Original Message-
> From: Julien Meunier [mailto:julien.meunier at 6wind.com]
> Sent: Tuesday, February 2, 2016 9:51 PM
> To: Zhang, Helin
> Cc: dev at dpdk.org
> Subject: [PATCH v2] i40e: fix vlan filtering
>
> VLAN filtering was always performed, even if hw_vlan_filter was
On 2016/02/03 8:43, Ferruh Yigit wrote:
> On Tue, Feb 02, 2016 at 08:18:42PM +0900, Tetsuya Mukawa wrote:
>> diff --git a/doc/guides/nics/index.rst b/doc/guides/nics/index.rst
>> index 33c9cea..5819cdb 100644
>> --- a/doc/guides/nics/index.rst
>> +++ b/doc/guides/nics/index.rst
>> @@ -47,6 +47,7
On Tue, 2 Feb 2016 14:57:00 +0800
Wenzhuo Lu wrote:
> +/**
> + * l2 tunnel configuration.
> + */
> +struct rte_eth_l2_tunnel {
> + enum rte_eth_l2_tunnel_type l2_tunnel_type;
> + uint16_t ether_type;
> +};
Building an API with an indirect (structure) parameter with only two
elements
On 2016/01/30 2:16, Ferruh Yigit wrote:
> 1- remove duplicate nb_rx/tx_queues fields from internals
> 2- remove duplicate numa_node field from internals
>
> Signed-off-by: Ferruh Yigit
> ---
> drivers/net/null/rte_eth_null.c | 36
> 1 file changed, 12
Hi,
Best Regards,
Mark
> -Original Message-
> From: He, Shaopeng
> Sent: Thursday, January 28, 2016 1:49 PM
> To: dev at dpdk.org
> Cc: Chen, Jing D; Wang, Xiao W; He, Shaopeng
> Subject: [PATCH] fm10k: enable PCIe port level Loopback Suppression
>
> A PCIe port may represent within it
Hi,
Best Regards,
Mark
> -Original Message-
> From: Qiu, Michael
> Sent: Monday, January 11, 2016 3:28 PM
> To: dev at dpdk.org
> Cc: Chen, Jing D; Qiu, Michael
> Subject: [PATCH 2/2] fm10k: update doc for Atwood Channel
>
> Atwood Channel is 20GbE NIC and belongs to Intel FM10K
On 2016/02/03 8:43, Ferruh Yigit wrote:
> On Tue, Feb 02, 2016 at 08:18:42PM +0900, Tetsuya Mukawa wrote:
>> +
>> +/* find an ethdev entry */
>> +eth_dev = rte_eth_dev_allocated(name);
>> +if (eth_dev == NULL)
>> +return -ENODEV;
>> +
>> +internal =
Hi Stephen,
> -Original Message-
> From: Stephen Hemminger [mailto:stephen at networkplumber.org]
> Sent: Wednesday, February 3, 2016 11:36 AM
> To: Lu, Wenzhuo
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 2/7] lib/librte_ether: support l2 tunnel
> config
>
> On Tue, 2 Feb
This series of patches extend the flow director filter and add support
for Chelsio T5 hardware filtering capabilities.
Chelsio T5 supports carrying out filtering in hardware which supports 3
actions to carry out on a packet which hit a filter viz.
1. Action Pass - Packets hitting a filter rule
Add a new raw packet flow that allows specifying generic flow input.
Add the ability to provide masks for fields in flow to allow range of
values.
Add a new behavior switch.
Add the ability to provide behavior arguments to allow rewriting matched
fields with new values. Ex: allows to provide
Add a new test_cxgbe_filters command line example to test support for
Chelsio T5 hardware filtering. Shows how to pass the Chelsio input flow
and input masks. Also, shows how to pass extra behavior arguments to
rewrite fields in matched filter rules.
Also add documentation and update MAINTAINERS.
Update base to get info about supported filtering features.
Add TID table that holds filter state and add filter specification
that provide a skeleton to add LE-TCAM (Maskfull) and Hash (Maskless)
T5 filtering capabilities.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Kumar Sanghvi
---
Add a control txq with its own mempool to create control packets
containing filtering information to be communicated with firmware.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Kumar Sanghvi
---
drivers/net/cxgbe/base/adapter.h| 15 +++
drivers/net/cxgbe/base/common.h | 2
Add Compressed Local IP (CLIP) table that holds IPv6 addresses to be
matched against packets.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Kumar Sanghvi
---
drivers/net/cxgbe/Makefile | 1 +
drivers/net/cxgbe/base/adapter.h| 52
Add Layer 2 Table (L2T) that holds destination mac addresses and
vlan ids to be modified on a packet that matches a corresponding
set 'switch' action filter.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Kumar Sanghvi
---
drivers/net/cxgbe/Makefile | 1 +
Add Source MAC Table (SMT) that holds source mac addresses to be
modified on a packet that matches a corresponding set 'switch'
action filter.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Kumar Sanghvi
---
drivers/net/cxgbe/Makefile | 1 +
drivers/net/cxgbe/base/adapter.h | 1 +
Add support for setting LE-TCAM (Maskfull) filters. IPv4 filters
occupy one index per filter, but IPv6 filters occupy 4 indices per
filter and must be on boundary aligned by 4. Filters with lower
index have higher priority over filters with higher index. When
a filter is hit, the rest of the
Add flow director support for setting/deleting LE-TCAM (maskfull)
and HASH (maskless) filters.
Wait and poll firmware event queue for replies about the filter status.
Also, the firware event queue doesn't have any freelists. So, there is
no need to refill any.
Provide stats showing the number
Hello,
On Tue, Feb 2, 2016 at 3:27 PM, wrote:
> From: Kamil Rytarowski
>
> This change enables drivers needing custom pci (de)initialization functions
> through the standard callback overloading.
>
> For example:
>
> /*
> * virtual function driver struct
> */
> static struct eth_driver
If test the feature by testpmd, how to set the parameter? Need testpmd document
update also?
Thanks
Qian
-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Zhe Tao
Sent: Thursday, January 21, 2016 3:24 PM
To: dev at dpdk.org
Subject: [dpdk-dev] PATCH 1/2] i40e:
On Wed, Feb 03, 2016 at 04:48:17PM +0900, Tetsuya Mukawa wrote:
> On 2016/02/03 8:43, Ferruh Yigit wrote:
> > On Tue, Feb 02, 2016 at 08:18:42PM +0900, Tetsuya Mukawa wrote:
> >> +
> >> + /* find an ethdev entry */
> >> + eth_dev = rte_eth_dev_allocated(name);
> >> + if (eth_dev == NULL)
> >> +
On 01/29/2016 03:49 PM, David Marchand wrote:
> Reuse pci hook to implement automatic bind / unbind.
> The more I look at this, the more I think this should go to the PMDs
> themselves (with options per devices to control this), with EAL offering
> helpers to achieve this.
This sounds good to me.
On 2016/02/03 18:24, Ferruh Yigit wrote:
> On Wed, Feb 03, 2016 at 04:48:17PM +0900, Tetsuya Mukawa wrote:
>> On 2016/02/03 8:43, Ferruh Yigit wrote:
>>> On Tue, Feb 02, 2016 at 08:18:42PM +0900, Tetsuya Mukawa wrote:
+
+ /* find an ethdev entry */
+ eth_dev =
On Tue, Feb 2, 2016 at 9:48 PM, Santosh Shukla wrote:
> On Tue, Feb 2, 2016 at 9:21 PM, Santosh Shukla wrote:
>> On Tue, Feb 2, 2016 at 2:19 PM, Yuanhan Liu
>> wrote:
>>> On Tue, Feb 02, 2016 at 06:50:18AM +0100, David Marchand wrote:
On Tue, Feb 2, 2016 at 6:43 AM, Yuanhan Liu >>>
> -Original Message-
> From: De Lara Guarch, Pablo
> Sent: Monday, February 01, 2016 9:41 AM
> To: dev at dpdk.org
> Cc: De Lara Guarch, Pablo
> Subject: [PATCH] l2fwd_crypto: fix typo in command line help
>
> Signed-off-by: Pablo de Lara
> ---
> examples/l2fwd-crypto/main.c | 2 +-
>
On Thu, Dec 03, 2015 at 05:05:04PM -0800, Stephen Hemminger wrote:
> A couple of fixes for vlan offload, and the patch to support
> multi-segment frames.
>
> Charles (Chas) Williams (1):
> vmxnet3: don't clear vf_table on restart
>
> Stephen Hemminger (2):
> vmxnet3: support mult-segment
This patchset includes param handling, typo and
statistics fixes in the L2fwd-crypto sample app.
Pablo de Lara (4):
l2fwd-crypto: fix accumulated stats
l2fwd-crypto: fix incorrect params in command line help
l2fwd-crypto: fix auth params setting
l2fwd-crypto: fix typos
Total number of packets (adding up packets RX/TX/dropped in all ports),
where being increased by the accumulated number of packets per port
every time the stats were printed, instead of the resetting
them.
Signed-off-by: Pablo de Lara
---
examples/l2fwd-crypto/main.c | 7 +--
1 file
Signed-off-by: Pablo de Lara
---
examples/l2fwd-crypto/main.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/examples/l2fwd-crypto/main.c b/examples/l2fwd-crypto/main.c
index 928b285..37f59c4 100644
--- a/examples/l2fwd-crypto/main.c
+++ b/examples/l2fwd-crypto/main.c
@@
Signed-off-by: Pablo de Lara
---
examples/l2fwd-crypto/main.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/examples/l2fwd-crypto/main.c b/examples/l2fwd-crypto/main.c
index 37f59c4..17a7774 100644
--- a/examples/l2fwd-crypto/main.c
+++ b/examples/l2fwd-crypto/main.c
Signed-off-by: Pablo de Lara
---
examples/l2fwd-crypto/main.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/examples/l2fwd-crypto/main.c b/examples/l2fwd-crypto/main.c
index 17a7774..9145f1c 100644
--- a/examples/l2fwd-crypto/main.c
+++ b/examples/l2fwd-crypto/main.c
@@
On Fri, Dec 04, 2015 at 02:18:34PM -0500, Eric Kinzie wrote:
> On Fri Dec 04 19:36:09 +0100 2015, Andriy Berestovskyy wrote:
> > Hi guys,
> > I'm not quite sure if we can support less TX queues on a slave that easy:
> >
> > > queue_id = bond_slave_txqid(internals, i, bd_tx_q->queue_id);
> > >
Hello,
INFO log level is used in order to keep code homogeneity:
i40e_vsi_config_vlan_stripping or i40e_dev_init_vlan use this log level
during failure for example.
Tell me if ERR log level for VLAN filtering issue must be set.
On 02/03/2016 02:15 AM, Zhang, Helin wrote:
>
>> -Original
W dniu 03.02.2016 o 09:47, David Marchand pisze:
> Hello,
Hello,
> On Tue, Feb 2, 2016 at 3:27 PM, wrote:
>> From: Kamil Rytarowski
>>
>> This change enables drivers needing custom pci (de)initialization functions
>> through the standard callback overloading.
>>
>> For example:
>>
>> /*
>>
On Tue, Feb 02, 2016 at 09:48:44PM +0530, Santosh Shukla wrote:
> On Tue, Feb 2, 2016 at 9:21 PM, Santosh Shukla wrote:
> > On Tue, Feb 2, 2016 at 2:19 PM, Yuanhan Liu > linux.intel.com> wrote:
> >> On Tue, Feb 02, 2016 at 06:50:18AM +0100, David Marchand wrote:
> >>> On Tue, Feb 2, 2016 at 6:43
On Wed, Feb 03, 2016 at 03:20:09PM +0530, Santosh Shukla wrote:
> On Tue, Feb 2, 2016 at 9:48 PM, Santosh Shukla wrote:
> > On Tue, Feb 2, 2016 at 9:21 PM, Santosh Shukla
> > wrote:
> >> On Tue, Feb 2, 2016 at 2:19 PM, Yuanhan Liu >> linux.intel.com> wrote:
> >>> On Tue, Feb 02, 2016 at
On Tue, Feb 02, 2016 at 11:59:48PM +0100, Thomas Monjalon wrote:
> Following the work of Ferruh, I suggest this cleanup to remove as much
> as possible of the code from the cpuflags headers.
> The goal is to un-inline these functions (not performance sensitive)
> and remove the CPU flags table
On Wed, Feb 3, 2016 at 12:39 PM, Kamil Rytarowski
wrote:
> W dniu 03.02.2016 o 09:47, David Marchand pisze:
>> And do your custom things in its devinit function ?
>
> I'm requesting from PF the mode of the device to be initialized. This part
> is handled dynamically and depends of the current
2016-02-02 21:48, Yuanhan Liu:
> v7: - make checkpatch a bit happier; few (false) warnings still left.
>
> - rebase to latest code: fixed a conflict.
>
> v6: unfold IO_READ/WRITE_DEF macro
>
> v5: minor fixes:
>
> - fix wrong type of arg "offset" of read/write_dev_config(): patch 2
>
On 03/02/16 11:28, Bruce Richardson wrote:
> On Fri, Dec 04, 2015 at 02:18:34PM -0500, Eric Kinzie wrote:
>> On Fri Dec 04 19:36:09 +0100 2015, Andriy Berestovskyy wrote:
>>> Hi guys,
>>> I'm not quite sure if we can support less TX queues on a slave that easy:
>>>
queue_id =
2016-02-03 15:17, Declan Doherty:
> On 03/02/16 11:28, Bruce Richardson wrote:
> > Hi Eric, Stephen, Declan,
> >
> > all patches of the set apart from this one and the next (nos 6 & 7) have no
> > comments and have been acked. Is there a resolution on these two patches,
> > so they
> > can be
On Wed, Feb 03, 2016 at 03:01:26PM +0100, Thomas Monjalon wrote:
> 2016-02-03 19:08, Jerin Jacob:
> > On Tue, Feb 02, 2016 at 11:59:48PM +0100, Thomas Monjalon wrote:
> > compilation errors on arm64 too.
>
> Yes I forgot REG_PLATFORM.
>
> > arm64 toolchains are publicly available @
> >
W dniu 03.02.2016 o 16:49, Kamil Rytarowski pisze:
>
>
> W dniu 03.02.2016 o 15:08, David Marchand pisze:
>> On Wed, Feb 3, 2016 at 12:39 PM, Kamil Rytarowski
>> wrote:
>>> W dniu 03.02.2016 o 09:47, David Marchand pisze:
And do your custom things in its devinit function ?
>>> I'm
On Wed, Dec 23, 2015 at 01:32:59PM +0800, Helin Zhang wrote:
> Workaround for the issue of cannot processing adminq commands during
> initialization, when 2x40G or 4x10G is receiving packets in highest
> throughput. Register 0x002698a8 and 0x002698ac should be cleared at
> first, and restored with
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Keith Wiles
> Sent: Tuesday, February 02, 2016 11:03 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH] mempool: Reduce rte_mempool structure size
>
> The rte_mempool structure is changed, which will
Hi Stephen,
Please find some comments below.
On 01/22/2016 02:40 PM, Mrzyglod, DanielX T wrote:
>> +/*
>> + * Creates a copy of the given packet mbuf.
>> + *
>> + * Walks through all segments of the given packet mbuf, and for each of
>> them:
>> + * - Creates a new packet mbuf from the given
Hi,
On 01/27/2016 02:56 PM, Panu Matilainen wrote:
>
> Since rte_pktmbuf_alloc_bulk() is an inline function, it is not part of
> the library ABI and should not be listed in the version map.
>
> I assume its inline for performance reasons, but then you lose the
> benefits of dynamic linking such
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Tomasz Kulasek
> Sent: Friday, January 29, 2016 4:49 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH] examples: l3fwd exact-match path rework
>
> Current implementation of Exact-Match uses different
When checking build with clang and RTE_MACHINE=default,
3 errors appeared.
Thomas Monjalon (3):
eal/x86: fix build with clang for old AVX
examples/l3fwd: fix build without SSE4.1
examples/ip_pipeline: fix build for x86_64 without SSE4.2
examples/ip_pipeline/pipeline/hash_func.h
When configuring RTE_MACHINE to "default", rte_memcpy implementation
is the default one (old AVX).
In this code, clang raises a warning thanks to -Wsometimes-uninitialized:
rte_memcpy.h:838:6: error:
variable 'srcofs' is used uninitialized whenever 'if' condition is false
if (dstofss > 0)
clang reports this error:
examples/l3fwd/main.c:550:1: error: unused function 'send_packetsx4'
The function is used only when ENABLE_MULTI_BUFFER_OPTIMIZE is 1.
Fixes: 96ff445371e0 ("examples/l3fwd: reorganise and optimize LPM code path")
Fixes: 6f1c1e28d98e ("examples/l3fwd: fix build with
The compiler cannot use _mm_crc32_u64:
examples/ip_pipeline/pipeline/hash_func.h:165:9:
error: implicit declaration of function '_mm_crc32_u64' is invalid in C99
Fixes: 947024a26df7 ("examples/ip_pipeline: rework passthrough pipeline")
Signed-off-by: Thomas Monjalon
---
Is there any way to enable multi-queue for SR-IOV of ixgbe?
I've seen that PF driver automatically disables multi-queue when VFs are
created from host.
We want to use multiple queues with DPDK in case of ixgbevf too.
[781203.692378] ixgbe :06:00.0: Multiqueue Disabled: Rx Queue count =
1,
Hi Saurabh,
May I know the model number of your physical nic?
Regards,
Choi, Sy Jong
Platform Application Engineer
-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Saurabh Mishra
Sent: Thursday, February 04, 2016 3:47 AM
To: dev at dpdk.org; users at dpdk.org
It's X520-2. Note that we can see Multiple queues when we use PCI-pass
through of whole NIC but the moment we enable VFs, ixgbe disables
multiple-queues.
Here are more details:
06:00.0 *Ether*net controller: Intel Corporation 82599ES 10-Gigabit
SFI/SFP+ Network Connection (rev 01)
06:00.1
When configuring RTE_MACHINE to "default", rte_memcpy implementation
is the default one (old AVX).
In this code, clang raises a warning thanks to -Wsometimes-uninitialized:
rte_memcpy.h:838:6: error:
variable 'srcofs' is used uninitialized whenever 'if' condition is false
if (dstofss > 0)
60 matches
Mail list logo