Hi,
I have a patch that enables 32bit secondary DPDK processes to attach with
64bit primary DPDK process and share the mempool and ring datastructures. I
developed this patch to enable our product(from Brocade) which uses a 32bit
userspace network stack to be able to work with PMDs like ivshmem an
2014-12-05 19:03, Jean-Mickael Guerin:
> On 05/12/2014 18:07, Ananyev, Konstantin wrote:
> > From: Jean-Mickael Guerin [mailto:jean-mickael.guerin at 6wind.com]
> >> On 05/12/2014 16:20, Konstantin Ananyev wrote:
> >>> That's an alternative way to fix the problem described in the patch:
> >>> http:
> > Add a compiler barrier to make sure all fields covered by
> > the marker rearm_data are assigned before the read.
> >
> > Signed-off-by: Jean-Mickael Guerin
> > Acked-by: David Marchand
> > Fixes: 0ff3324da2 ("ixgbe: rework vector pmd following mbuf changes")
>
> Acked-by: Bruce Richardson
2014-12-05 17:37, Thomas Monjalon:
> 2014-12-05 11:24, John W. Linville:
> > On Fri, Dec 05, 2014 at 10:30:53AM +0100, Thomas Monjalon wrote:
> > > Hi John,
> > >
> > > 2014-12-04 16:42, John W. Linville:
> > > > This allows the PMD to compile with kernels that don't support the
> > > > options in
> Compiling latest DPDK with clang 3.3 on FreeBSD 10 shows up a number of
> compilation
> errors in the enic driver. These two small patches fix those errors.
>
> Bruce Richardson (2):
> enic: fix initialization error with clang
> enic: fix error with uninitialized variable.
Acked and applie
Hi,
Please don't reply to a previous email to start a different topic.
You can check how it looks here:
http://thread.gmane.org/gmane.comp.networking.dpdk.devel
> "DISCLAIMER: This message is proprietary to Aricent and is intended solely
> for the use of the individual to whom it is addr
On 05/12/2014 18:07, Ananyev, Konstantin wrote:
>
>
>> -Original Message-
>> From: Jean-Mickael Guerin [mailto:jean-mickael.guerin at 6wind.com]
>> Sent: Friday, December 05, 2014 4:59 PM
>> To: Ananyev, Konstantin
>> Cc: dev at dpdk.org
>> Subject: Re: [dpdk-dev] [RFC PATCH] ixgbe: ixgbe_r
Konstantin,
It would be easier if you sent a v2 of the patchset from Jean-Mickael.
So, the first patch would be included with your rework of the second patch.
Please use --in-reply-to to make tracking from emails or archives easier.
Thanks
--
Thomas
On 05/12/2014 16:20, Konstantin Ananyev wrote:
> That's an alternative way to fix the problem described in the patch:
> http://dpdk.org/ml/archives/dev/2014-December/009394.html.
> The main difference is:
> - move buf_len fields out of rearm_data marker.
> - make ixgbe_recv_pkts_vec() not touch buf
2014-12-05 11:24, John W. Linville:
> On Fri, Dec 05, 2014 at 10:30:53AM +0100, Thomas Monjalon wrote:
> > Hi John,
> >
> > 2014-12-04 16:42, John W. Linville:
> > > This allows the PMD to compile with kernels that don't support the
> > > options in question. The "#if defined(...)" lines are a bi
Adding call to mac_address_slaves_update from the lsc handler when the
first slave become active to propagate any mac changes made while
devices are inactive
Changed removing slave logic to use memmove instead of memcpy to move
data within the same array, as this was corrupting the slave array.
A
Hi,
I was trying to access the pci config space(BAR) of the virtual function device
visible in the virtual machine, similar to what DPDK does on host via both UIO
and VFIO. Did the following steps.
1. Bound PF Drivers to ixgbe and spawned virtual function drivers , bound them
to vfio-pci and se
> These changes are DPDK 1.8 modifications and some corrections to the
> Linux Getting Started Guide, the FreeBSD Getting Started Guide,
> the Programmers Guide, the Sample Application User Guide,
> the TestPMD Application User Guide and the Release Notes.
>
> The following changes since commit b6
2014-12-03 16:56, Bruce Richardson:
> The symmetric_mp example app is set up to allow two processes to
> share a NIC port, with each pulling packets from one queue. In order
> to have the app continue working when one of the process dies, the
> drop_en bit should be set in the NIC configuration. Wi
> When a lookup was done on a table_array structure with an incomplete
> bitmask, the results was always zero hits. This was because the
> pkts_mask value was cleared as we process each entry, and the result
> was assigned at the end of the loop, when pkts_mask was zero.
> Changing the assignment t
> > In order not to affect the FVL's performance by default setting, this patch
> > moves
> > the flow director initialization from i40e_pf_setup to i40e_dev_configure
> > according to the mode in fdir configure info.
> > Then the resources used for flow director will be only setup if it is
> > e
On 2014/12/5 4:05, Neil Horman wrote:
> On Thu, Dec 04, 2014 at 04:59:59PM +0100, Thomas Monjalon wrote:
>> 2014-12-04 10:32, Neil Horman:
>>> On Thu, Dec 04, 2014 at 02:47:03PM +0100, Thomas Monjalon wrote:
2014-12-04 08:29, Neil Horman:
> On Thu, Dec 04, 2014 at 12:59:31PM +0100, Thomas
> -Original Message-
> From: Jean-Mickael Guerin [mailto:jean-mickael.guerin at 6wind.com]
> Sent: Friday, December 05, 2014 4:59 PM
> To: Ananyev, Konstantin
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [RFC PATCH] ixgbe: ixgbe_recv_pkts_vec shouldn't
> override mbuf buffer length
>
> > We have got some feedback about backward compatibility of VXLAN TX checksum
> > offload API with 1G/10G NIC after the i40e VXLAN
> > TX checksum codes were applied, so we have to rework the APIs on i40e,
> > including the changes of mbuf, i40e PMD and csum forward
> > engine.
> >
> > The mai
> > It can be hard to implement a reliable detection of endianness.
> > Previous trials in testpmd failed.
> > The IBM Power patchset introduced a config option.
> >
> > This patchset try to improve the situation by having a detection
> > in EAL headers.
> >
> > Please test it (especially on IBM Po
> > When using test-pmd with flow director in FreeBSD, the application will
> > segfault/Bus error while parsing the command-line. This is due to how
> > each commands result structure is represented during parsing, where the
> > offsets
> > for each tokens value is stored in a character array(cha
On 2014/12/4 17:12, Michael Qiu wrote:
> lib/librte_eal/linuxapp/eal/eal_memory.c:324:4: error: comparison
> is always false due to limited range of data type [-Werror=type-limits]
> || (hugepage_sz == RTE_PGSIZE_16G)) {
> ^
> cc1: all warnings being treated as errors
>
> lib/librte_eal/
corrected couple of typos in distributor application
Signed-off-by: Reshma Pattan
---
doc/guides/sample_app_ug/dist_app.rst |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/doc/guides/sample_app_ug/dist_app.rst
b/doc/guides/sample_app_ug/dist_app.rst
index 76e3924..57
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Thursday, December 04, 2014 1:21 PM
> To: De Lara Guarch, Pablo
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] app/test: fix memory needs after
> RTE_MAX_LCORE was increased to 128
>
> > S
The variable notify_pa is only initialized inside one branch of
an if statement, triggering a compiler error with clang 3.3 on FreeBSD.
CC vnic/vnic_dev.o
/usr/home/bruce/dpdk.org/lib/librte_pmd_enic/vnic/vnic_dev.c:777:6: fatal
error: variable 'notify_pa'
is used uninitialized whenever '
This patch fixes the following compiler error raised by clang 3.3
on FreeBSD 10:
CC enic_clsf.o
/usr/home/bruce/dpdk.org/lib/librte_pmd_enic/enic_clsf.c:99:25: fatal error:
missing field 'u' initializer [-Wmissing-field-initializers]
struct filter fltr = {0};
It fixes it by changing th
Compiling latest DPDK with clang 3.3 on FreeBSD 10 shows up a number of
compilation
errors in the enic driver. These two small patches fix those errors.
Bruce Richardson (2):
enic: fix initialization error with clang
enic: fix error with uninitialized variable.
lib/librte_pmd_enic/enic_clsf
2014-12-05 09:42, Neil Horman:
> On Fri, Dec 05, 2014 at 02:10:27PM +0100, Thomas Monjalon wrote:
> > 2014-12-04 15:05, Neil Horman:
> > > On Thu, Dec 04, 2014 at 04:59:59PM +0100, Thomas Monjalon wrote:
> > > > I'm convinced we need to work on the build system but it's another
> > > > discussion
On Fri, Dec 05, 2014 at 03:19:07PM +0100, Olivier Matz wrote:
> From: Alan Carew
>
> When using test-pmd with flow director in FreeBSD, the application will
> segfault/Bus error while parsing the command-line. This is due to how
> each commands result structure is represented during parsing, wher
> -Original Message-
> From: Richardson, Bruce
> Sent: Friday, December 05, 2014 12:11 PM
> To: Ananyev, Konstantin
> Cc: Jean-Mickael Guerin; dev at dpdk.org
> Subject: Re: [PATCH v2] ixgbe: don't override mbuf buffer length
>
> On Fri, Dec 05, 2014 at 11:59:50AM +, Ananyev, Konstan
That's an alternative way to fix the problem described in the patch:
http://dpdk.org/ml/archives/dev/2014-December/009394.html.
The main difference is:
- move buf_len fields out of rearm_data marker.
- make ixgbe_recv_pkts_vec() not touch buf_len field at all
(as all other RX functions behave).
Si
From: Alan Carew
When using test-pmd with flow director in FreeBSD, the application will
segfault/Bus error while parsing the command-line. This is due to how
each commands result structure is represented during parsing, where the offsets
for each tokens value is stored in a character array(char
Hi Alan,
On 11/10/2014 10:19 AM, Alan Carew wrote:
> When using test-pmd with flow director in FreeBSD, the application will
> segfault/Bus error while parsing the command-line. This is due to how
> each commands result structure is represented during parsing, where the
> offsets
> for each tokens
Michael,
I'm looking at it. I'll give you feedback soon.
On 2014/12/5 14:56, Qiu, Michael wrote:
> Hi Chao
>
> Would you please take a look at this patch?
>
> It's solved issue introduce by Power Arch support patch.
>
> Your comments are very precious :)
>
> Thanks,
> Michael
> On 12/5/2014 2:03
On Fri, Dec 05, 2014 at 09:22:05AM -0500, Neil Horman wrote:
> On Fri, Dec 05, 2014 at 04:31:47PM +0800, Chao Zhu wrote:
> >
> > On 2014/12/4 17:12, Michael Qiu wrote:
> > >lib/librte_eal/linuxapp/eal/eal_memory.c:324:4: error: comparison
> > >is always false due to limited range of data type [-We
On Fri, Dec 05, 2014 at 11:30:41AM -0800, Malveeka Tewari wrote:
> Hi all
>
> I have a very basic DPDK app that runs in two modes: sender and sniffer.
>
> In the sender mode, the app just reads data from a shared memory buffer and
> sends packets out on a port using rte_eth_tx_burst.
> In the sni
2014-12-04 15:05, Neil Horman:
> On Thu, Dec 04, 2014 at 04:59:59PM +0100, Thomas Monjalon wrote:
> > 2014-12-04 10:32, Neil Horman:
> > > On Thu, Dec 04, 2014 at 02:47:03PM +0100, Thomas Monjalon wrote:
> > > > 2014-12-04 08:29, Neil Horman:
> > > > > On Thu, Dec 04, 2014 at 12:59:31PM +0100, Thom
Hi,
Two machines are connected back-to-back. On one machine we ran the DPDK &
bound the Ethernet port to the DPDK (back-to-back connected port).
We loaded the igb_uio and KNI modules and ran the KNI app with the
following command.
*kni -c 0xff -n 4 -- -p 0x1 -P --config="(0,
When using multiple processes, the TX function used in all processes
should be the same, otherwise the secondary processes cannot transmit
more than tx-ring-size - 1 packets.
To achieve this, we extract out the code to select the ixgbe TX function
to be used into a separate function inside the ixgb
On Fri, Dec 05, 2014 at 07:08:55PM +0100, Thomas Monjalon wrote:
> Hi,
>
> Please don't reply to a previous email to start a different topic.
> You can check how it looks here:
> http://thread.gmane.org/gmane.comp.networking.dpdk.devel
>
> > "DISCLAIMER: This message is proprietary to Arice
Hi,
On 12/02/2014 04:06 PM, Jijiang Liu wrote:
> Replace the inner_l2_len and the inner_l3_len field with the outer_l2_len and
> outer_l3_len field, and rework csum forward engine and i40e PMD due to these
> changes.
>
> Signed-off-by: Jijiang Liu
> ---
> app/test-pmd/csumonly.c |
Hi,
On 12/02/2014 04:06 PM, Jijiang Liu wrote:
> Replace PKT_TX_VXLAN_CKSUM with PKT_TX_UDP_TUNNEL_PKT in order to indicate a
> packet is an UDP tunneling packet, and introduce 3 TX offload flags for outer
> IP TX checksum, which are PKT_TX_OUTER_IP_CKSUM, PKT_TX_OUTER_IPV4 and
> PKT_TX_OUTER_I
On Fri, Dec 05, 2014 at 11:59:50AM +, Ananyev, Konstantin wrote:
>
>
> > -Original Message-
> > From: Richardson, Bruce
> > Sent: Friday, December 05, 2014 11:29 AM
> > To: Ananyev, Konstantin
> > Cc: Jean-Mickael Guerin; dev at dpdk.org
> > Subject: Re: [PATCH v2] ixgbe: don't overri
> -Original Message-
> From: Richardson, Bruce
> Sent: Friday, December 05, 2014 11:29 AM
> To: Ananyev, Konstantin
> Cc: Jean-Mickael Guerin; dev at dpdk.org
> Subject: Re: [PATCH v2] ixgbe: don't override mbuf buffer length
>
> On Fri, Dec 05, 2014 at 11:19:11AM +, Ananyev, Konstan
Hi all
I have a very basic DPDK app that runs in two modes: sender and sniffer.
In the sender mode, the app just reads data from a shared memory buffer and
sends packets out on a port using rte_eth_tx_burst.
In the sniffer mode, the app polls for incoming packets on a port using
rte_eth_rx_burst
On Fri, Dec 05, 2014 at 11:19:11AM +, Ananyev, Konstantin wrote:
>
>
> > -Original Message-
> > From: Richardson, Bruce
> > Sent: Friday, December 05, 2014 10:39 AM
> > To: Ananyev, Konstantin
> > Cc: Jean-Mickael Guerin; dev at dpdk.org
> > Subject: Re: [PATCH v2] ixgbe: don't overri
On Fri, Dec 05, 2014 at 10:30:53AM +0100, Thomas Monjalon wrote:
> Hi John,
>
> 2014-12-04 16:42, John W. Linville:
> > This allows the PMD to compile with kernels that don't support the
> > options in question. The "#if defined(...)" lines are a bit ugly,
> > but I don't know of any better way t
> -Original Message-
> From: Richardson, Bruce
> Sent: Friday, December 05, 2014 10:39 AM
> To: Ananyev, Konstantin
> Cc: Jean-Mickael Guerin; dev at dpdk.org
> Subject: Re: [PATCH v2] ixgbe: don't override mbuf buffer length
>
> On Fri, Dec 05, 2014 at 01:10:28AM +, Ananyev, Konstan
As we validate dpdk 1.8.0 rc2 from org, we report defect: Work_Request
IXA00388707
The defect is about vhost virtio based 4 x 10 G NIC, and vhost-switch cannot
startup, all cases for one copy on Fortville cannot validate. This defect can
be fixed by this patches. Below are details for validatio
Hi Helin,
> -Original Message-
> From: Zhang, Helin
> Sent: Friday, December 05, 2014 1:46 AM
> To: dev at dpdk.org
> Cc: Cao, Waterman; Cao, Min; Wu, Jingjing; Liu, Jijiang; Ananyev, Konstantin;
> olivier.matz at 6wind.com; Zhang, Helin
> Subject: [PATCH v2 2/2] mbuf: assign valid bit va
On Thu, Dec 04, 2014 at 07:09:13PM +0100, Jean-Mickael Guerin wrote:
> The template mbuf_initializer is hard coded with a buflen which
> might have been set differently by the application at the time of
> mbuf pool creation.
>
> Switch to a mbuf allocation, to fetch the correct default values.
> T
On Fri, Dec 05, 2014 at 01:10:28AM +, Ananyev, Konstantin wrote:
>
>
> > -Original Message-
> > From: Jean-Mickael Guerin [mailto:jean-mickael.guerin at 6wind.com]
> > Sent: Thursday, December 04, 2014 6:09 PM
> > To: dev at dpdk.org
> > Cc: Richardson, Bruce; Ananyev, Konstantin
> >
Hi John,
2014-12-04 16:42, John W. Linville:
> This allows the PMD to compile with kernels that don't support the
> options in question. The "#if defined(...)" lines are a bit ugly,
> but I don't know of any better way to accomplish the task.
>
> Signed-off-by: John W. Linville
> ---
> lib/lib
2014-12-05 05:34, Qiu, Michael:
> Any comments about this version? a new workaround solution :)
Yes, one comment: I think it's ugly :)
These aliasing errors are not reliable so I think we can disable it (like
Linux does).
But in case you don't want to disable the warning, please add a comment to y
On Fri, Dec 05, 2014 at 03:02:33PM +, Bruce Richardson wrote:
> On Fri, Dec 05, 2014 at 09:22:05AM -0500, Neil Horman wrote:
> > On Fri, Dec 05, 2014 at 04:31:47PM +0800, Chao Zhu wrote:
> > >
> > > On 2014/12/4 17:12, Michael Qiu wrote:
> > > >lib/librte_eal/linuxapp/eal/eal_memory.c:324:4: e
Hi Huawei,
2014-12-05 04:56, Xie, Huawei:
> Hi Thomas:
> I will continue work on this fix.
> Do you have comments to the v4 patch?
> For Bruce's comment, I add some descriptive commit message for the commit.
> For the constant number, I define a macro as the wrapper for the VFA array
> index and
Hi Michael,
Few nitpicks :-) (wording of the log message I guess is up to Thomas, I won't
comment on that)
> lib/librte_eal/common/eal_private.h| 36
> ++
> lib/librte_eal/linuxapp/eal/eal_pci_vfio.c | 24 +---
> 2 files changed, 57 insertions
Before redefining mbuf structure, there was lack of free bits in
'ol_flags' (32 bits in total) for new RX or TX flags. So it tried
to reuse existant bits as most as possible, or even assigning 0 to
some of bit flags. After new mbuf structure defined, there are
quite a lot of free bits. So those new
According to the latest datasheet, RX descriptor error flag of
'RECIPE' is not for public use, so its checks should be removed.
Signed-off-by: Helin Zhang
---
lib/librte_pmd_i40e/i40e_rxtx.c | 6 --
1 file changed, 6 deletions(-)
v2 changes:
* Removed error flag of 'ECIPE' processing only i
Before redefining mbuf structure, there was lack of free bits in 'ol_flags'
(32 bits in total) for new RX or TX flags. So it tried to reuse existant bits
as most as possible, or even assigning 0 to some of bit flags. After new mbuf
structure defined, there are quite a lot of free bits. So those new
On Fri, Dec 05, 2014 at 02:10:27PM +0100, Thomas Monjalon wrote:
> 2014-12-04 15:05, Neil Horman:
> > On Thu, Dec 04, 2014 at 04:59:59PM +0100, Thomas Monjalon wrote:
> > > 2014-12-04 10:32, Neil Horman:
> > > > On Thu, Dec 04, 2014 at 02:47:03PM +0100, Thomas Monjalon wrote:
> > > > > 2014-12-04 0
On Fri, Dec 05, 2014 at 04:31:47PM +0800, Chao Zhu wrote:
>
> On 2014/12/4 17:12, Michael Qiu wrote:
> >lib/librte_eal/linuxapp/eal/eal_memory.c:324:4: error: comparison
> >is always false due to limited range of data type [-Werror=type-limits]
> > || (hugepage_sz == RTE_PGSIZE_16G)) {
> >
On 12/4/2014 7:49 PM, Bruce Richardson wrote:
> When using multiple processes, the TX function used in all processes
> should be the same, otherwise the secondary processes cannot transmit
> more than tx-ring-size - 1 packets.
> To achieve this, we extract out the code to select the ixgbe TX functi
Hi,
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Thursday, December 4, 2014 11:33 PM
> To: Chen, Jing D
> Cc: dev at dpdk.org; Qiu, Michael
> Subject: Re: [dpdk-dev] [PATCH] i40e: Fix a vlan bug
>
> 2014-12-04 14:29, Chen, Jing D:
> > From: Th
Hi Chao
Would you please take a look at this patch?
It's solved issue introduce by Power Arch support patch.
Your comments are very precious :)
Thanks,
Michael
On 12/5/2014 2:03 PM, Michael Qiu wrote:
> lib/librte_eal/linuxapp/eal/eal_memory.c:324:4: error: comparison
> is always false due to l
Any comments about this version? a new workaround solution :)
Thanks,
Michael
On 12/4/2014 9:35 PM, Michael Qiu wrote:
> app/test-pmd/csumonly.c: In function ?get_psd_sum?:
> build/include/rte_ip.h:161: error: dereferencing pointer ?u16?
> does break strict-aliasing rules
> build/include/r
Hi Thomas:
I will continue work on this fix.
Do you have comments to the v4 patch?
For Bruce's comment, I add some descriptive commit message for the commit.
For the constant number, I define a macro as the wrapper for the VFA array
index and value.
One question is it isn't based on latest commit
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Friday, December 5, 2014 6:56 AM
> To: Olivier MATZ; Thomas Monjalon
> Cc: dev at dpdk.org; Liu, Jijiang
> Subject: RE: [dpdk-dev] [PATCH v5 2/3] mbuf:add three TX ol_flags and repalce
> PKT_TX_VXLAN_CKSUM
>
>
>
> > -Original
On 12/5/2014 12:31 AM, Burakov, Anatoly wrote:
> Hi Michael
>
>> But indeed, when try to unload both vfio and vfio_iommu_type1,
>> /dev/vfio/vfio still there, I'm also surprise.
>>
>> My ENV is fedora20, kernel version 3.6.7-200 X86_64.
>>
>> Believe or not, you can have a try, it seems a kernel is
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Thursday, December 4, 2014 6:20 PM
> To: Zhang, Helin; Olivier MATZ; Liu, Jijiang; dev at dpdk.org
> Subject: RE: [dpdk-dev] [PATCH v5 2/3] mbuf:add three TX ol_flags and repalce
> PKT_TX_VXLAN_CKSUM
>
> Hi Helin,
>
> > -Origi
> -Original Message-
> From: Jean-Mickael Guerin [mailto:jean-mickael.guerin at 6wind.com]
> Sent: Thursday, December 04, 2014 6:09 PM
> To: dev at dpdk.org
> Cc: Richardson, Bruce; Ananyev, Konstantin
> Subject: [PATCH v2] ixgbe: don't override mbuf buffer length
>
> The template mbuf_i
> -Original Message-
> From: Wu, Jingjing
> Sent: Thursday, December 4, 2014 11:40 PM
> To: dev at dpdk.org
> Cc: Wu, Jingjing; Zhang, Helin; Ananyev, Konstantin
> Subject: [PATCH]i40e: move the fdir_setup from dev_init to dev_configure
>
> In order not to affect the FVL's performance by
72 matches
Mail list logo
