Hi Thomas,
I think Larry explains things pretty clearly below.
A new application facing API is bring provided that avoids reusing kernel
specific types and structs or dipping into the kernel itself. It also clearly
separates the user space API from driver related functions/types.
We do want
I believe that their is value in this interface for software stacks not
based on Linux being moved toward DPDK that need simple operations like
getting the mac address. Some of these stacks have a dearth of resources
available and dedicating a core/thread to KNI to get/set a mac address
is
On 5/21/2015 3:51 PM, Ouyang Changchun wrote:
> Each virtio device could have multiple queues, say 2 or 4, at most 8.
> Enabling this feature allows virtio device/port on guest has the ability to
> use different vCPU to receive/transmit packets from/to each queue.
>
> In multiple queues mode,
Fix enqueue/dequeue can't handle chained vring descriptors;
Remove unnecessary vring descriptor length updating;
Add support copying scattered mbuf to vring;
Changchun Ouyang (4):
lib_vhost: Fix enqueue/dequeue can't handle chained vring descriptors
lib_vhost: Refine code style
lib_vhost:
Vring enqueue need consider the 2 cases:
1. use separate descriptors to contain virtio header and actual data, e.g. the
first descriptor
is for virtio header, and then followed by descriptors for actual data.
2. virtio header and some data are put together in one descriptor, e.g. the
first
Extract codes into 2 common functions:
update_secure_len which is used to accumulate the buffer len in the vring
descriptors.
and fill_buf_vec which is used to fill struct buf_vec.
Changes in v5
- merge fill_buf_vec into update_secure_len
- do both tasks in one-time loop
Signed-off-by:
Remove these unnecessary vring descriptor length updating, vhost should not
change them.
virtio in front end should assign value to desc.len for both rx and tx.
Signed-off-by: Changchun Ouyang
---
lib/librte_vhost/vhost_rxtx.c | 18 +-
1 file changed, 1 insertion(+), 17
Remove unnecessary new line.
Signed-off-by: Changchun Ouyang
---
lib/librte_vhost/vhost_rxtx.c | 9 +++--
1 file changed, 3 insertions(+), 6 deletions(-)
diff --git a/lib/librte_vhost/vhost_rxtx.c b/lib/librte_vhost/vhost_rxtx.c
index a11d007..5824ffc 100644
---
Hi,
I found that, in dpdk 2.0, rte_eal_alarm_set() is affected by
discontinuous jumps in the system time because eal_alarm_callback()
and rte_eal_alarm_set() use gettimeofday() to get the current time.
Here is what I encountered.
I set up a rte eal alarm as below, and I like it to be triggered
Acked-by: Jingjing Wu
> -Original Message-
> From: Zhang, Helin
> Sent: Thursday, May 07, 2015 9:41 AM
> To: dev at dpdk.org
> Cc: Cao, Min; Wu, Jingjing; Liu, Jijiang; Xu, Qian Q; Zhang, Helin
> Subject: [PATCH v2] i40evf: fix of supporting jumbo frame
>
> It wouldn't check the
On 6/3/2015 12:21 AM, Stephen Hemminger wrote:
> On Tue, 2 Jun 2015 14:53:15 +0800
> Cunming Liang wrote:
>
>> The patch adds 'rte_epoll_wait' and 'rte_epoll_ctl' for async event wakeup.
>> It defines 'struct rte_epoll_event' as the event param.
>> The 'op' uses the same enum as epoll_wait/ctl
On 6/3/2015 12:27 AM, Stephen Hemminger wrote:
> Minor spelling and stuff. I am a terrible speller...
>
>> The patch exposes intr event fd create and release for PMD.
>> The device driver can assign the number of event associated with interrupt
>> vector.
>> It also provides misc funtions to
On 6/3/2015 12:24 AM, Stephen Hemminger wrote:
> On Tue, 2 Jun 2015 14:53:16 +0800
> Cunming Liang wrote:
>
>> +if (!rc)
>> +RTE_LOG(DEBUG, EAL, "eventfd %d associated with vec %d"
>> +" is added on epfd %d\n", rev->fd, vec, epfd);
>
Tested-by: Qian Xu
Signed-off-by: Qian Xu
-Tested commit: 1a1109404e702d3ad1ccc1033df55c59bec1f89a
-Host OS/Kernel: FC21/3.19
-Guest OS/Kernel: FC21/3.19
-NIC: Intel 82599 10G
-Default x86_64-native-linuxapp-gcc configuration
-Total 2 cases, 2 passed.
Test Case 1:
2015-05-15 17:08, Andrey Chilikin:
> Added optional ip version check to l3fwd app to allow to detect
> the ip version if mbuf ol_flags are not set in case of running
> in a VM with emulated network controllers
>
> Signed-off-by: Andrey Chilikin
[...]
> +#define DO_IP_VERSION_CHECK 0
2015-05-18 13:13, Andrey Chilikin:
> Add a command-line parameter to l3fwd, to allow the user to specify the
> destination mac address for each ethernet port used.
>
> v2 changes:
> - apply command-line parameter to fast path as well (val_eth)
>
> Signed-off-by: Andrey Chilikin
>
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Wednesday, June 3, 2015 9:17 AM
> To: Chilikin, Andrey
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] examples: add ip version check for l3fwd app
>
> 2015-05-15 17:08, Andrey Chilikin:
> >
2015-06-03 08:24, Chilikin, Andrey:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > 2015-05-15 17:08, Andrey Chilikin:
> > > Added optional ip version check to l3fwd app to allow to detect the ip
> > > version if mbuf ol_flags are not set in case of running in a VM with
> > >
Tested-by: Yong Liu
- Tested Commit: 7c4c66bf666b8059ed0ad2f2478ef349b3272f51
- OS: Fedora20 3.15.5
- GCC: gcc version 4.8.3 20140911
- CPU: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz
- NIC: Intel Corporation Device XL710 [8086:1584] Firmware 4.33
- Default x86_64-native-linuxapp-gcc
Hello,
I see several PTP processing functions in "test-pmd/ieee1588fwd.c" for
Intel 82576. Are there any equivalences for Intel I-350 and I-210?
Thanks,
Chui-hui
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Huawei Xie
> Sent: Tuesday, June 02, 2015 2:50 AM
> To: dev at dpdk.org
> Cc: Sun, Peng A
> Subject: [dpdk-dev] [PATCH] vhost: provide vhost API to unregister vhost
> unix domain socket
>
>
On Tue, Jun 02, 2015 at 02:59:24PM +0200, Thomas Monjalon wrote:
> 2015-05-13 16:30, Bruce Richardson:
> > Add coding standards document to guides directory. This document
> > codifies the current DPDK C coding conventions, to make it easier for
> > contributors to see the format their code should
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Chui Hui Chiu
> Sent: Wednesday, June 3, 2015 10:31 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] IEEE 1588 PTP processing functions for I-350 and I-210
>
> Hello,
>
> I see several PTP processing functions
On Mon, Jun 01, 2015 at 11:32:25AM +0200, Olivier Matz wrote:
> As it's done in __rte_pktmbuf_prefree_seg(), we can avoid using an
> atomic increment in rte_pktmbuf_attach() by checking if we are the
> only owner of the mbuf first.
>
> Signed-off-by: Olivier Matz
> ---
>
OVERVIEW
1) Setting .rxmode.mq_mode for bonding device to ETH_MQ_RX_RSS makes bonding
device fully RSS-capable, so all slaves are synchronized with its configuration.
This mode is intended to provide RSS configuration as known from "dynamic RSS
configuration for one port" and made slaves
Bonding device implements independent management of RSS settings. It stores its
own copies of settings i.e. RETA, RSS hash function and RSS key. It?s required
to ensure consistency.
1) RSS hash function set for bonding device is maximal set of RSS hash functions
supported by all bonded devices.
This implementation allows to set and read RSS configuration for ring device,
and is used to validate right values propagation over the slaves, in test units
for dynamic RSS configuration for bonding.
It have no impact on packet processing by ring device.
Signed-off-by: Tomasz Kulasek
---
This test module uses ring device to check right RSS configuration propagation.
1) Propagation test
a) Set RSS hash function for bonding, fetch RSS hash function from bonded
slave, check if same. Do it for all slaves.
b) Repeat above for RSS key and RETA.
2)Dynamic adding slave to the
Queue stats mapping is used in example application. This patch adds propagation
of mapping over the slaves, and fills bonding port's stats with a sum of
corresponding values taken from bonded slaves, when stats are requested for
bonding port.
Signed-off-by: Tomasz Kulasek
---
Per queue statistics are already implemented for ring device, but with static
mapping (stat_idx == queue_id).
This fix is required, if you want to use ring device in test application and is
used only to point that per queue statistics are provided for this device.
Signed-off-by: Tomasz Kulasek
This application allows you to test RSS configuration for bonded devices
changing configuration dynamically, during receiving packets on slaves and
observe the changes in its distribution over queues.
After initialization process, all accessible ports are attached to one bonding
as slaves.
Signed-off-by: Tomasz Kulasek
---
.../prog_guide/link_bonding_poll_mode_drv_lib.rst |8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/doc/guides/prog_guide/link_bonding_poll_mode_drv_lib.rst
b/doc/guides/prog_guide/link_bonding_poll_mode_drv_lib.rst
index
Documentation update about implementation details and requirements for Dynamic
RSS Configuration for Bonding.
Signed-off-by: Tomasz Kulasek
---
.../prog_guide/link_bonding_poll_mode_drv_lib.rst | 32 ++--
1 file changed, 30 insertions(+), 2 deletions(-)
diff --git
On Wed, Jun 03, 2015 at 02:31:47PM +0800, Selmon Yang wrote:
> Hi,
>
> I found that, in dpdk 2.0, rte_eal_alarm_set() is affected by
> discontinuous jumps in the system time because eal_alarm_callback()
> and rte_eal_alarm_set() use gettimeofday() to get the current time.
>
> Here is what I
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Huawei Xie
> Sent: Tuesday, June 2, 2015 9:50 AM
> To: dev at dpdk.org
> Cc: Sun, Peng A
> Subject: [dpdk-dev] [PATCH] vhost: provide vhost API to unregister vhost
> unix domain socket
>
>
On Wed, Jun 3, 2015 at 7:54 AM, Bruce Richardson wrote:
> On Wed, Jun 03, 2015 at 02:31:47PM +0800, Selmon Yang wrote:
> > Hi,
> >
> > I found that, in dpdk 2.0, rte_eal_alarm_set() is affected by
> > discontinuous jumps in the system time because eal_alarm_callback()
> > and rte_eal_alarm_set()
On Tue, Jun 02, 2015 at 04:38:48PM +0200, Thomas Monjalon wrote:
> GCC 5.1 detects more out-of-bounds accesses:
> error: array subscript is above array bounds [-Werror=array-bounds]
>
> Signed-off-by: Thomas Monjalon
> ---
Acked-by: Bruce Richardson
> examples/load_balancer/config.c |
Add coding standards document to guides directory. This document
codifies the current DPDK C coding conventions, to make it easier for
contributors to see the format their code should be in.
Signed-off-by: Siobhan Butler
Signed-off-by: Bruce Richardson
---
Updates in V2:
* Fixed file creation
This patch depends on the Port Hotplug Framework.
It implements the eth_dev_uninit functions for rte_em_pmd,
rte_igb_pmd and rte_igbvf_pmd.
Changes in V3:
Add igb_adapter_stopped and em_adapter_stopped booleans.
Release rx and tx queues
Changes in V2:
Call dev_close() from dev_uninit()
2015-05-08 10:58, Vladimir Medvedkin:
> Software implementation of the Toeplitz hash function used by RSS.
> Can be used either for packet distribution on single queue NIC
> or for simulating of RSS computation on specific NIC (for example
> after GRE header decapsulating).
>
> v3 changes
> -
2015-05-29 12:57, Neil Horman:
> On Fri, May 29, 2015 at 03:34:12PM +0100, Bruce Richardson wrote:
> > V2 Changes:
> > * use RTE_MIN instead of additional comparison checks
> > in some loops, as suggested by Neil.
> > * For stats reset in null PMD, just use the size of the
> > array [which is
2015-06-03 14:06, Bruce Richardson:
> On Tue, Jun 02, 2015 at 04:38:48PM +0200, Thomas Monjalon wrote:
> > GCC 5.1 detects more out-of-bounds accesses:
> > error: array subscript is above array bounds [-Werror=array-bounds]
> >
> > Signed-off-by: Thomas Monjalon
>
> Acked-by: Bruce
On Fri, May 29, 2015 at 02:46:36PM +0100, Bernard Iremonger wrote:
> This patch depends on the Port Hotplug Framework.
> It implements the rte_dev_uninit_t() function for the ring pmd.
>
> Changes in V3:
> Rebase to use drivers/net/ring directory
> Handle no parameters case
>
> Changes in V2:
>
2015-04-22 08:44, Stephen Hemminger:
> On Wed, 22 Apr 2015 14:31:55 +0100
> Bruce Richardson wrote:
>
> > Using the "physical_package_id" as a fallback for determining the
> > numa node of a core tends to be unreliable. Fix this by using a
> > detection routine which reads the numa information
2015-06-03 14:58, Bruce Richardson:
> Add coding standards document to guides directory. This document
> codifies the current DPDK C coding conventions, to make it easier for
> contributors to see the format their code should be in.
>
> Signed-off-by: Siobhan Butler
> Signed-off-by: Bruce
Reported by Zi Hu :
"...
cat test_data/rule1
@192.168.0.0/24 192.168.0.0/24 400 : 500 0 : 52 6/0xff
@192.168.0.0/24 192.168.0.0/24 400 : 500 54 : 65280 6/0xff
@192.168.0.0/24 192.168.0.0/24 400 : 500 0 : 65535 6/0xff
cat test_data/trace1
0xc0a80005 0xc0a80009 450 53 0x06
I run the test by:
sudo
v2:
- reorder code a bit to avoid gcc 5.1 warnings.
Konstantin Ananyev (3):
ACL: fix a problem in acl_merge_trie
ACL: add new test case for ranges build
ACL: remove subtree_id calculations at build stage
app/test/test_acl.c | 147 +--
Signed-off-by: Konstantin Ananyev
---
app/test/test_acl.c | 147 ++--
1 file changed, 143 insertions(+), 4 deletions(-)
diff --git a/app/test/test_acl.c b/app/test/test_acl.c
index 7119ad3..6a032f9 100644
--- a/app/test/test_acl.c
+++
v2:
- reorder code a bit to avoid gcc 5.1 warnings.
As now subtree_id is not used acl_merge_trie() any more,
there is no point to calculate and maintain that information.
Signed-off-by: Konstantin Ananyev
---
lib/librte_acl/acl.h | 7 ---
lib/librte_acl/acl_bld.c | 121
On 02/06/15 18:35, Ananyev, Konstantin wrote:
>
>
>> -Original Message-
>> From: Zoltan Kiss [mailto:zoltan.kiss at linaro.org]
>> Sent: Tuesday, June 02, 2015 4:08 PM
>> To: Ananyev, Konstantin; dev at dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH] ixgbe: fix checking for tx_free_thresh
>>
Hi,
I am trying run the packet generation application on DPDK. When I run the
application it is not configuring any ports. When I try to attach a port it
was not working. Below is the screen shot
Lua 5.2.4 Copyright (C) 1994-2015 Lua.org, PUC-Rio
>>> Packet Burst 32, RX Desc 512, TX Desc 512,
Hi,
I use mempool APIs to allocate DMA-able descriptor ring and buffers so that
I get both physical and virtual addresses of allocated memory. I buffers I
get from the mempool APIs are 64 byte aligned. My requirement is 256byte
alignment. How to achieve this? mempool APIs do not seem to take
On Wed, 3 Jun 2015 11:30:17 -0700
Hi Dax,
Dax Rawal wrote:
> Hi,
> I use mempool APIs to allocate DMA-able descriptor ring and buffers
> so that I get both physical and virtual addresses of allocated
> memory. I buffers I get from the mempool APIs are 64 byte aligned. My
> requirement is
On Tue, 2 Jun 2015 15:11:30 +0200
Roman Dementiev wrote:
>
> This series of patches adds methods that use hardware memory transactions
> (HTM)
> on fast-path for DPDK locks (a.k.a. lock elision). Here the methods are
> implemented
> for x86 using Restricted Transactional Memory instructions
Signed-off-by: Keith Wiles
---
lib/librte_eal/bsdapp/eal/eal.c | 20
lib/librte_eal/common/include/rte_eal.h | 32
2 files changed, 52 insertions(+)
diff --git a/lib/librte_eal/bsdapp/eal/eal.c b/lib/librte_eal/bsdapp/eal/eal.c
index
Hi Cyril,
Definitely, yes.
Thanks.
-Dax
On Wed, Jun 3, 2015 at 11:36 AM, Cyril Chemparathy
wrote:
> On Wed, 3 Jun 2015 11:30:17 -0700
> Hi Dax,
>
> Dax Rawal wrote:
>
> > Hi,
> > I use mempool APIs to allocate DMA-able descriptor ring and buffers
> > so that I get both physical and virtual
On 6/3/2015 8:55 PM, Ouyang, Changchun wrote:
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Huawei Xie
>> Sent: Tuesday, June 2, 2015 9:50 AM
>> To: dev at dpdk.org
>> Cc: Sun, Peng A
>> Subject: [dpdk-dev] [PATCH] vhost: provide vhost API to unregister
Don't need the 'safe' version of list_for_each_entry() if you aren't deleting
from the list as you iterate over it
Signed-off-by: Jay Rolette
---
lib/librte_eal/linuxapp/kni/kni_misc.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git
No reason to check out many entries are in kni->rx_q prior to
actually pulling them from the fifo. You can't dequeue more than
are there anyway. Max entries to dequeue is either the max batch
size or however much space is available on kni->free_q (lesser of the two)
Signed-off-by: Jay Rolette
Loop processing packets dequeued from rx_q was using the number of
packets requested, not how many it actually received.
Variable rename to make code a little more clear
Signed-off-by: Jay Rolette
---
lib/librte_eal/linuxapp/kni/kni_net.c | 14 +++---
1 file changed, 7 insertions(+), 7
Don't need the 'safe' version of list_for_each_entry() if you aren't deleting
from the list as you iterate over it
Signed-off-by: Jay Rolette
---
lib/librte_eal/linuxapp/kni/kni_misc.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git
No reason to check out many entries are in kni->rx_q prior to
actually pulling them from the fifo. You can't dequeue more than
are there anyway. Max entries to dequeue is either the max batch
size or however much space is available on kni->free_q (lesser of the two)
Signed-off-by: Jay Rolette
Loop processing packets dequeued from rx_q was using the number of
packets requested, not how many it actually received.
Variable rename to make code a little more clear
Signed-off-by: Jay Rolette
---
lib/librte_eal/linuxapp/kni/kni_net.c | 14 +++---
1 file changed, 7 insertions(+), 7
Hi Keith,
2015-06-03 13:49, Keith Wiles:
> Signed-off-by: Keith Wiles
> ---
> lib/librte_eal/bsdapp/eal/eal.c | 20
> lib/librte_eal/common/include/rte_eal.h | 32
> 2 files changed, 52 insertions(+)
These comments would be useful:
On 6/3/15, 2:24 PM, "Thomas Monjalon" wrote:
>Hi Keith,
>
>2015-06-03 13:49, Keith Wiles:
>> Signed-off-by: Keith Wiles
>> ---
>> lib/librte_eal/bsdapp/eal/eal.c | 20
>> lib/librte_eal/common/include/rte_eal.h | 32
>>
>> 2 files
While looking at code is seems lots of drivers do not know the convention
that info_get and stats_get both clear the buffer for the caller.
Stephen Hemminger (3):
bonding: remove useless memset
ethdev: remove useless memset
null: remove unnecessary memset
Remove 2 unneeded memset's. The info area is already cleared
rte_eth_dev_info_get
and the statistics buffer is cleared by rte_eth_stats_get
Signed-off-by: Stephen Hemminger
---
drivers/net/bonding/rte_eth_bond_api.c | 1 -
drivers/net/bonding/rte_eth_bond_pmd.c | 4
2 files changed, 5
eth_stats is already cleared by rte_eth_stats_get
Signed-off-by: Stephen Hemminger
---
lib/librte_ether/rte_ethdev.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
index 024fe8b..5a94654 100644
--- a/lib/librte_ether/rte_ethdev.c
The stats area is already cleared before calling eth_stats_get().
Signed-off-by: Stephen Hemminger
---
drivers/net/null/rte_eth_null.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/net/null/rte_eth_null.c b/drivers/net/null/rte_eth_null.c
index 7792315..e244595 100644
---
On Wed, 3 Jun 2015 13:49:53 -0500
Keith Wiles wrote:
> +/* Launch threads, called at application init() and parse app args. */
> +int
> +rte_eal_init_parse(int argc, char **argv,
> + int (*parse)(int, char **))
> +{
> + int ret;
> +
> + ret = rte_eal_init(argc, argv);
>
I have noticed a number of emails from this list are going to spam. It looks
like it might be gmail filtering based on the fact that most of the list has a
valid SPF based on an IPv4 address that reports out like below:
Received: from dpdk.org (dpdk.org. [92.243.14.124])
by
On Wed, Jun 03, 2015 at 07:54:11PM -0700, Alexander Duyck wrote:
> However the ones that are going straight into my spam folder list an IPv6
> address that is rated neutral by the SPF:
> Received: from dpdk.org ([2001:4b98:dc0:41:216:3eff:fe72:dd13])
> by mx.google.com with ESMTP id
On Wed, Jun 03, 2015 at 08:07:57PM -0700, Matthew Hall wrote:
> Hello,
>
> I can confirm your claims. When I enabled IPv6 on my mail server Google began
> dropping or severely graylisting 100% of the mail until the IPv6 subnet was
> added to the SPF record. This needs to get fixed or the IPv6
73 matches
Mail list logo
