On 4/9/15, 6:53 AM, "Neil Horman" wrote:
>On Wed, Apr 08, 2015 at 03:58:38PM -0500, Keith Wiles wrote:
>> Move a number of device specific define, structures and functions
>> into a generic device base set of files for all device not just
>>Ethernet.
>>
>> Signed-off-by: Keith Wiles
>> ---
>>
Any comments on this RFC Patch? Must be perfect right :-)
On 4/8/15, 3:58 PM, "Keith Wiles" wrote:
>Hi All,
>
>Here is a set of RFC patches to update DPDK to support a generic set of
>devices. The patches that follow create two files eal_common_device.h
>and rte_common_device.c and then a few it
Converted the Glossary table in the Programmer's Guide
to a definition list to improve rendering.
Signed-off-by: John McNamara
---
doc/guides/prog_guide/glossary.rst | 261 -
1 file changed, 168 insertions(+), 93 deletions(-)
diff --git a/doc/guides/prog_guid
2015-04-10 15:20, John McNamara:
> Added explicit field widths to Programmer's Guide
> Glossary table to control the PDF output. Previously
> the table overflowed the right margin.
Long tables should be avoided in RST as they are not easy to read or update.
Why not using definition lists?
Fixed two verbatim text sections in vmxnet3 docs.
Signed-off-by: John McNamara
---
doc/guides/nics/vmxnet3.rst | 27 ---
1 file changed, 20 insertions(+), 7 deletions(-)
diff --git a/doc/guides/nics/vmxnet3.rst b/doc/guides/nics/vmxnet3.rst
index 3aa5b40..830530d 100644
Added explicit field widths to Programmer's Guide
Glossary table to control the PDF output. Previously
the table overflowed the right margin.
This doesn't affect the Html output.
Also, added bold format to colunm headers for clarity.
Signed-off-by: John McNamara
---
doc/guides/prog_guide/gloss
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Friday, April 10, 2015 3:34 PM
> To: Mcnamara, John
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] doc: add field widths to prog guide
> glossary
>
> Long tables should be avoided in RST as
On 4/10/15, 6:41 AM, "Neil Horman" wrote:
>On Fri, Apr 10, 2015 at 01:49:33AM +, Wiles, Keith wrote:
>>
>>
>> On 4/9/15, 7:26 PM, "Neil Horman" wrote:
>>
>> >On Thu, Apr 09, 2015 at 09:10:19PM +, Wiles, Keith wrote:
>> >>
>> >>
>> >> On 4/9/15, 2:38 PM, "Jay Rolette" wrote:
>> >>
The issue is with my application logic, which calculates the SIP/DIP MASK
values.
I am using "__builtin_ctz(x)" to calculate the SIP/DIP MASK.
mask = 32 - __builtin_ctz(value).
If the value is ZERO, the outcome is undefined.
In my case, we are hitting the UNDEFINED case here.
On MACHINE 1, __bu
Hi, all
I'am trying to use valgrind to check memory leak with my dpdk application but
dpdk always failed to mmap hugepages.
Without valgrind it works well.How to run dpdk applications with valgrind?Is
there any other way to check memory leak
with dpdk applications?
On 2015-04-06 19:01, Eric Kinzie wrote:
>
> +static void
> +lacp_recv_cb(uint8_t slave_id, struct rte_mbuf *lacp_pkt)
> +{
> + lacpdu_rx_count[slave_id]++;
> + RTE_VERIFY(lacp_pkt != NULL);
> + rte_pktmbuf_free(lacp_pkt);
> +}
> +
Would be nice to check here if it is valid LACP packet
Hi Konstantin,
Thanks a lot for looking in to this.
In my case, I am using the ACL Pipeline Infra [My application logic is
built on the ACL pipeline].
After your mail, I did verify the same with "l3fwd-acl" example, it just
worked fine.
Seems like, the issue is observed only with the Pipeline imp
Hi Eric
Please see my comments.
On 2015-04-06 19:01, Eric Kinzie wrote:
>Provide functions to allow an external 802.3ad state machine to transmit
>and recieve LACPDUs and to set the collection/distribution flags on
>slave interfaces.
>
> Signed-off-by: Eric Kinzie
> ---
> lib/librte
On 2015-04-06 19:01, Eric Kinzie wrote:
> if (unlikely(hdr->ether_type == ether_type_slow_be ||
> !collecting || (!promisc &&
> - !is_same_ether_addr(&bond_mac,
> &hdr->d_addr {
> +
2015-04-10 09:47, Pawel Wodkowski:
> On 2015-04-06 19:01, Eric Kinzie wrote:
> >Copy all needed fields from the mode8023ad_private structure in
> >bond_mode_8023ad_conf_get().
> >
> > Signed-off-by: Eric Kinzie
> > ---
> > lib/librte_pmd_bond/rte_eth_bond_8023ad.c |1 +
> > 1 file c
This script ease testing of basic initializations and Rx/Tx bursts.
It may help to check obvious regressions.
In order to run it on a standard development machine, it doesn't use
neither hugepages nor real interfaces.
The optional parameters are:
- build directory (default: build)
- coremask (defa
Null PMD was not found when using a statically linked application:
EAL: no driver found for eth_null1
EAL: failed to initialize eth_null1 device
Fixes: c743e50c475f ("null: new poll mode driver")
Signed-off-by: Thomas Monjalon
---
mk/rte.app.mk | 4
1 file changed, 4 insert
The virtio pmd is not restricted to uio anymore.
Fixes: da978dfdc43b ("virtio: use port IO to get PCI resource")
Signed-off-by: Thomas Monjalon
---
lib/librte_pmd_virtio/Makefile | 2 +-
mk/rte.app.mk | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/li
On 2015-04-06 19:01, Eric Kinzie wrote:
>Copy all needed fields from the mode8023ad_private structure in
>bond_mode_8023ad_conf_get().
>
> Signed-off-by: Eric Kinzie
> ---
> lib/librte_pmd_bond/rte_eth_bond_8023ad.c |1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/lib/librte_
On 2015-04-06 19:01, Eric Kinzie wrote:
> If a device to be enslaved already has transmit and/or receive queues
> allocated, use those and then create any additional queues that are
> necessary.
>
> Signed-off-by: Eric Kinzie
> ---
> lib/librte_pmd_bond/rte_eth_bond_pmd.c |8 ++--
> 1 f
Same as ixgbe. Should not be as chatty in log and should be
possible to compile away all the debug messages.
Signed-off-by: Stephen Hemminger
---
lib/librte_pmd_e1000/e1000_logs.h | 3 +--
lib/librte_pmd_e1000/em_ethdev.c | 10 --
lib/librte_pmd_e1000/igb_ethdev.c | 16 ++--
The ixgbe driver likes to be far to chatty in the system log
which is good for the original developer but not good for a production
product.
Reduce the log spam by doing:
* All the normal messages should be changed from INFO to DEBUG.
* The log messages should be done with RTE_LOG so that they c
The Intel Poll Mode Drivers generate a lot of log messages and
they are mostly at the INFO logging level. This should be changed
to the common syslog conventions, ie. put unimportant stuff for
production users at DEBUG level and things of interest but not
errors at NOTICE level.
Also should use RT
On Fri, Apr 10, 2015 at 01:49:33AM +, Wiles, Keith wrote:
>
>
> On 4/9/15, 7:26 PM, "Neil Horman" wrote:
>
> >On Thu, Apr 09, 2015 at 09:10:19PM +, Wiles, Keith wrote:
> >>
> >>
> >> On 4/9/15, 2:38 PM, "Jay Rolette" wrote:
> >>
> >> >On Thu, Apr 9, 2015 at 2:16 PM, Neil Horman
> >
Paul Emmerich wrote:
> Stephen Hemminger wrote:
>
>> Your excess polling consumes PCI bandwidth which is a fixed resource.
>
> I doubt that this is the problem for three reasons:
>
4th: polling should not cause a PCIe access as all the required information is
written
to memory by the NIC [1
Stephen Hemminger wrote:
> Your excess polling consumes PCI bandwidth which is a fixed resource.
I doubt that this is the problem for three reasons:
* The poll rate would regulate itself if the PCIe bus was the bottleneck
* This problem only occurs with 82599 chips, not with X540 chips
(which
On 4/9/15, 7:26 PM, "Neil Horman" wrote:
>On Thu, Apr 09, 2015 at 09:10:19PM +, Wiles, Keith wrote:
>>
>>
>> On 4/9/15, 2:38 PM, "Jay Rolette" wrote:
>>
>> >On Thu, Apr 9, 2015 at 2:16 PM, Neil Horman
>>wrote:
>> >
>> >> On Thu, Apr 09, 2015 at 11:31:39AM -0500, Jay Rolette wrote:
>> >
thanks a lot .. great help .
On Wed, Apr 8, 2015 at 8:10 PM, Xie, Huawei wrote:
> On 4/6/2015 6:42 PM, Srinivasreddy R wrote:
> > Hi,
> > I have observed packet loss with usvhost dpdk interfaces even at a very
> > small rate .
> > This is because virtqueue [dev->virtqueue] is found to be fu
28 matches
Mail list logo
