2016-03-04 12:11, Keith Wiles:
> In order to cleanup the configuration files some and reduce
> the number of duplicate configuration information. Add a new
> file called common_base which contains just about all of the
> configuration lines in one place. Then have the common_bsdapp,
> common_linuxa
2016-03-04 11:03, Keith Wiles:
> Signed-off-by: Keith Wiles
> ---
> config/defconfig_x86_64-native-bsdapp-clang | 1 +
> config/defconfig_x86_64-native-bsdapp-gcc | 1 +
Until now, the generic 64-bit flag was used only for ARM or Linux,
and was not defined for BSD environment.
Fixes: d05e7115f
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Chas Williams
> Sent: Wednesday, December 30, 2015 4:38 PM
>
> From: "Charles (Chas) Williams"
>
> The VF needs to determine the queues sizes before .dev_infos_get so that it
> can hint to the upper layer the proper sizes. Move
> bnx2x_v
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Chas Williams
> Sent: Wednesday, December 30, 2015 4:38 PM
>
> From: "Charles (Chas) Williams"
>
> bnx2x_loop_obtain_resources() returns a struct containing the status and
> the error message. If bnx2x_do_req4pf() fails, it shouldn't ret
On Fri, Mar 04, 2016 at 03:57:43PM +0100, Thomas Monjalon wrote:
> 2015-11-24 14:25, Bruce Richardson:
> > The tbl8 and tbl24 structures were essentially identical except for
> > slightly different names for one or two fields. Merge these two
> > structures into a single structure definition.
> >
On 3/4/16, 7:25 AM, "Remy Horton" wrote:
>Implements driver support for setting of MAC address.
>
>Signed-off-by: Remy Horton
Acked-by: Yong Wang
mac_hi is not initialized so the upper 16bits are random when calling
VMXNET3_WRITE_BAR1_REG(). The handler of the mac changing ignores
them so i
I'm looking into details...
2016-02-23 14:10, Harry van Haaren:
> EAL
> ~~~
>
> +* **Added rte_eal_primary_proc_alive() function**
> +
> + A new function ``rte_eal_primary_proc_alive()`` has been added
> + to allow the user to detect if a primary process is running.
> + Use cases for this f
On 2016/03/04 17:39, Yuanhan Liu wrote:
> On Fri, Mar 04, 2016 at 01:17:42PM +0900, Tetsuya Mukawa wrote:
>> The patch introduces a new PMD. This PMD is implemented as thin wrapper
>> of librte_vhost. It means librte_vhost is also needed to compile the PMD.
>> The vhost messages will be handled onl
This patch adds an option, --huge-trybest, to use a recover mechanism to
the case that there are not so many hugepages (declared in sysfs), which
can be used. It relys on a mem access to fault-in hugepages, and if fails
with SIGBUS, recover to previously saved stack environment with
siglongjmp().
On 2016/03/04 15:10, Tan, Jianfeng wrote:
> Hi Tetsuya,
>
> On 3/4/2016 1:05 PM, Tetsuya Mukawa wrote:
>> On 2016/03/04 11:18, Tan, Jianfeng wrote:
>>> Hi Tetsuya,
>>>
>>> Seems that this patch is too long. Is it possible to split into
>>> multiple commits?
>> Hi Jianfeng,
>>
>> Sure, will do.
>>
>
2016-03-04 17:43, Trahe, Fiona:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > 2016-03-04 17:17, Fiona Trahe:
> > > This patch set separates the symmetric crypto operations from generic
> > operations and then modifies the cryptodev burst API to accept bursts of
> > rte_crypto_op
2016-03-04 17:17, Fiona Trahe:
> This patch set separates the symmetric crypto operations from generic
> operations and then modifies the cryptodev burst API to accept bursts of
> rte_crypto_op rather than rte_mbufs.
>
> V5:
> - updates .map file
> - removes EXPERIMENTAL label from rte_cryptodev
This patch modifies the crypto burst enqueue/dequeue APIs to operate on bursts
rte_crypto_op's rather than the current implementation which operates on
rte_mbuf bursts, this simplifies the burst processing in the crypto PMDs and the
use of crypto operations in general.
The changes also continues t
This patch splits symmetric specific definitions and
functions away from the common crypto APIs to facilitate the future extension
and expansion of the cryptodev framework, in order to allow asymmetric
crypto operations to be introduced at a later date, as well as to clean the
logical structur
This patch set separates the symmetric crypto operations from generic
operations and then modifies the cryptodev burst API to accept bursts of
rte_crypto_op rather than rte_mbufs.
v6:
- restore EXPERIMENTAL label to cryptodev. Will handle removal in separate
thread.
(email subject was incorrect
2016-03-04 11:01, Keith Wiles:
> +#
> +# Combine to one single library
> +#
> +CONFIG_RTE_BUILD_COMBINE_LIBS=n
It has been removed.
>2016-03-04 11:01, Keith Wiles:
>> +#
>> +# Combine to one single library
>> +#
>> +CONFIG_RTE_BUILD_COMBINE_LIBS=n
>
>It has been removed.
>
OK, Will send a new patch. Not sure how I missed it in the diffs :-(
Regards,
Keith
This patch adds option, --avail-cores, to use lcores which are available
by calling pthread_getaffinity_np() to narrow down detected cores before
parsing coremask (-c), corelist (-l), and coremap (--lcores).
Test example:
$ taskset 0xc ./examples/helloworld/build/helloworld \
-
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Friday, March 04, 2016 5:46 PM
> To: Trahe, Fiona
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v5 0/2] cryptodev API changes
>
> 2016-03-04 17:43, Trahe, Fiona:
> > From: Thomas Monjalon
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Friday, March 04, 2016 5:39 PM
> To: Trahe, Fiona
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v5 0/2] cryptodev API changes
>
> 2016-03-04 17:17, Fiona Trahe:
> > This patch set separates
> -Original Message-
> From: Trahe, Fiona
> Sent: Friday, March 04, 2016 5:18 PM
> To: dev at dpdk.org
> Cc: Doherty, Declan; Trahe, Fiona
> Subject: [PATCH v5 0/2] cryptodev API changes
>
> This patch set separates the symmetric crypto operations from generic
> operations and then modif
Hi Ferruh,
> -Original Message-
> From: Yigit, Ferruh
> Sent: Friday, March 4, 2016 5:14 PM
> To: Iremonger, Bernard ; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] bonding: fix crash when no slave devices
>
> On 2/23/2016 12:13 PM, Bernard Iremonger wrote:
> > If a bonded device is cr
The changes also continues the separatation of the symmetric operation
parameters
from the more general operation parameters, this will simplify the integration
of
asymmetric crypto operations in the future.
As well as the changes to the crypto APIs this patch adds functions for managing
rte_cry
Signed-off-by: Fiona Trahe
---
app/test/test_cryptodev.c | 164 +++---
app/test/test_cryptodev_perf.c | 79 +--
drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c | 44 +-
drivers/crypto/aesni_mb/rte_aesni_mb_pmd_ops.c | 6 +-
drivers/crypto/ae
This patch set separates the symmetric crypto operations from generic
operations and then modifies the cryptodev burst API to accept bursts of
rte_crypto_op rather than rte_mbufs.
V5:
- updates .map file
- removes EXPERIMENTAL label from rte_cryptodev.h
V4:
- Fixes for issues introduced in __rt
On 2/23/2016 12:13 PM, Bernard Iremonger wrote:
> If a bonded device is created when there are no slave devices
> there is loop in bond_ethdev_promiscous_enable() which results
> in a segmentation fault.
> I have applied a similar fix to bond_ethdev_promiscous_disable()
> where a similar loop could
> > Fixed issue of byte order in ethdev library that the structure for setting
> > fdir's
> > mask and flow entry is inconsist and made inputs of mask be in big endian.
> >
> > Fixes: 76c6f89e80d4 ("ixgbe: support new flow director masks")
> > Fixes: 2d4c1a9ea2ac ("ethdev: add new flow director m
On Fri, Mar 04, 2016 at 01:17:42PM +0900, Tetsuya Mukawa wrote:
> The patch introduces a new PMD. This PMD is implemented as thin wrapper
> of librte_vhost. It means librte_vhost is also needed to compile the PMD.
> The vhost messages will be handled only when a port is started. So start
> a port f
This patch will work on below patch series.
- [PATCH v6 00/11] Add API to get packet type info
As a example to use ptype info, l3fwd needs firstly to use
rte_eth_dev_get_ptype_info() API to check if device and/or
its PMD driver will parse and fill the needed packet type;
if not, use the newly add
2016-03-04 14:44, Wiles, Keith:
> >>>
> >>>--- a/config/defconfig_x86_64-native-bsdapp-clang
> >>>+++ b/config/defconfig_x86_64-native-bsdapp-clang
> >>>@@ -37,6 +37,7 @@ CONFIG_RTE_MACHINE="native"
> >>> CONFIG_RTE_ARCH="x86_64"
> >>> CONFIG_RTE_ARCH_X86_64=y
> >>> CONFIG_RTE_ARCH_X86=y
> >>>+C
2016-02-11 00:05, Vincent JARDIN:
> When CONFIG_RTE_LIBEAL_USE_HPET=y is set, eal_timer.c does not compile
> anymore. Just add simple missing include.
>
> Signed-off-by: Vincent JARDIN
It has been fixed in another patch:
http://dpdk.org/browse/dpdk/commit/?id=3560681d68
2015-11-29 10:46, Stephen Hemminger:
> This is the last round of sched updates for 2.2. It is based
> on code changes (extensively) tested by QA and used in the vRouter.
>
> Stephen Hemminger (3):
> rte_sched: keep track of RED drops
> rte_sched: introduce reciprocal divide
> rte_sched: elim
On Fri, Mar 4, 2016 at 3:02 PM, Stephen Hemminger <
stephen at networkplumber.org> wrote:
>
> > New driver for Broadcom bnxt (NexXtreme C-series) devices.>
>
> Looks good, I just have a couple of functionality comments.
>
> 1. Driver does not appear to support Link State interrupt. Not a big
>
2015-11-24 14:25, Bruce Richardson:
> The tbl8 and tbl24 structures were essentially identical except for
> slightly different names for one or two fields. Merge these two
> structures into a single structure definition.
>
> Two fields have been renamed as part of this change: the
> "ext_entry" fi
2016-03-04 10:09, Olivier Matz:
> From: Ravi Kerur
>
> Macros RTE_MBUF_DATA_DMA_ADDR and RTE_MBUF_DATA_DMA_ADDR_DEFAULT
> are defined in each PMD driver file. Convert macros to inline
> functions and move them to common lib/librte_mbuf/rte_mbuf.h file.
> PMD drivers include rte_mbuf.h file direct
2016-02-29 16:52, Declan Doherty:
> From: Fiona Trahe
>
> This patch splits symmetric specific definitions and functions away from the
> common crypto APIs to facilitate the future extension and expansion of the
> cryptodev framework, in order to allow asymmetric crypto operations to be
> introd
2016-03-04 10:05, Olivier MATZ:
> Hi Marc,
>
> On 03/03/2016 12:52 AM, Marc Sune wrote:
> > cmdline_parse_*.h headers use struct cmdline_token_hdr /
> > cmdline_parse_token_hdr_t which is defined in cmdline_parse.h, but
> > do not include it, forcing manual inclusion.
> >
> > This commit includes
Implements driver support for setting of MAC address.
Signed-off-by: Remy Horton
---
doc/guides/rel_notes/release_16_04.rst | 4
drivers/net/vmxnet3/vmxnet3_ethdev.c | 19 +++
2 files changed, 23 insertions(+)
diff --git a/doc/guides/rel_notes/release_16_04.rst
b/doc/gu
Implements driver support for dumping of EEPROM and registers,
and the setting of MAC address.
Signed-off-by: Remy Horton
---
doc/guides/rel_notes/release_16_04.rst |5 +
drivers/net/i40e/i40e_ethdev.c | 116
drivers/net/i40e/i40e_regs.h | 1000 +++
Several rte_eth_dev_* functions are currently only supported
by the ixgbe NIC driver. This patchset adds driver support
for some of these functions to the i40e and vmxnet3 drivers.
--
This patchset depends on:
* http://dpdk.org/dev/patchwork/patch/10654/
v2 changes:
* Rebased to origin/master
*
> -Original Message-
> From: Jerin Jacob [mailto:jerin.jacob at caviumnetworks.com]
> Sent: Friday, March 4, 2016 4:04 PM
> To: Thomas Monjalon
> Cc: Richardson, Bruce ; dev at dpdk.org;
> Kobylinski, MichalX
> Subject: Re: [dpdk-dev] [PATCH] lpm: merge tbl24 and tbl8 structures
> Impor
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Ferruh Yigit
> Sent: Friday, March 4, 2016 11:07 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH] maintainer: claim responsibility for igb_uio
>
> igb_iuo has no maintainer, claim responsibility for igb_
igb_iuo has no maintainer, claim responsibility for igb_uio
Signed-off-by: Ferruh Yigit
---
MAINTAINERS | 1 +
1 file changed, 1 insertion(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 628bc05..e253bf7 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -164,6 +164,7 @@ F: lib/librte_eal/linuxapp/e
On Fri, 4 Mar 2016 13:05:24 -0800
Stephen Hurd wrote:
> New driver for Broadcom bnxt (NexXtreme C-series) devices.
>
> Standards-compliant 10/25/50G support with 30MPPS full-duplex throughput
> http://www.broadcom.com/press/release.php?id=s923886
>
> Signed-off-by: Stephen Hurd
> ---
> v3:
>
>>>
>>>--- a/config/defconfig_x86_64-native-bsdapp-clang
>>>+++ b/config/defconfig_x86_64-native-bsdapp-clang
>>>@@ -37,6 +37,7 @@ CONFIG_RTE_MACHINE="native"
>>> CONFIG_RTE_ARCH="x86_64"
>>> CONFIG_RTE_ARCH_X86_64=y
>>> CONFIG_RTE_ARCH_X86=y
>>>+CONFIG_RTE_ARCH_64=y
>>>
>>> CONFIG_RTE_TOOLCH
On Fri, 4 Mar 2016 11:31:20 +0100
Christian Ehrhardt wrote:
> In certain autotests lpm->max_rules turned out to be non initialized.
> That was caused by a failing allocation for lpm->rules_tbl in rte_lpm6_create.
> It then left the function via goto exit with lpm freed, but still a pointer
> val
We're delighted to announce that the first event in our DPDK Summit series for
2016, for the China/Asia-Pacific region, will be held on Wednesday May 18th at
the Renaissance Shanghai Yangtze Hotel, Shanghai, China.
The DPDK community in the China/Asia-Pacific region will meet face-to-face at
th
lpm6 autotests failed with the default alloc of 512M Memory.
While >=2500M was a workaround it became clear while debugging that it
had a leak.
One could see a lot of output like:
LPM Test tests6[i]: FAIL
LPM: LPM memory allocation failed
It turned out that in rte_lpm6_free
- lpm might not be
Hi everybody,
I created a fix for it which will hit the mailing list soon, but considered
it important to send this mail ahead.
All that analysis has no place in the patch description, but it helps to
understand why/what was going on.
The follow up patch will have title "lpm/lpm6: fix missing free
>On 03/03/2016 08:37 PM, Thomas Monjalon wrote:
>>> --- a/config/defconfig_x86_64-native-bsdapp-clang
>>> +++ b/config/defconfig_x86_64-native-bsdapp-clang
>>> @@ -37,6 +37,7 @@ CONFIG_RTE_MACHINE="native"
>>> CONFIG_RTE_ARCH="x86_64"
>>> CONFIG_RTE_ARCH_X86_64=y
>>> CONFIG_RTE_ARCH_X86=y
>>>
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
>> Sent: Thursday, March 3, 2016 6:38 PM
>> To: Wiles, Keith
>> Cc: dev at dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH] config: remove duplicate configuration
>> information
>>
>> 2016-02-22
Hi Tetsuya,
On 3/4/2016 1:05 PM, Tetsuya Mukawa wrote:
> On 2016/03/04 11:18, Tan, Jianfeng wrote:
>> Hi Tetsuya,
>>
>> Seems that this patch is too long. Is it possible to split into
>> multiple commits?
> Hi Jianfeng,
>
> Sure, will do.
>
>> On 2/22/2016 4:17 PM, Tetsuya Mukawa wrote:
>>> The pa
On 2016/03/04 11:18, Tan, Jianfeng wrote:
> Hi Tetsuya,
>
> Seems that this patch is too long. Is it possible to split into
> multiple commits?
Hi Jianfeng,
Sure, will do.
>
> On 2/22/2016 4:17 PM, Tetsuya Mukawa wrote:
>> The patch adds a new virtio-net PMD configuration that allows the PMD to
On 04/03/16 10:58, Ananyev, Konstantin wrote:
> Hi,
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Olivier MATZ
>> Sent: Friday, March 04, 2016 9:29 AM
>> To: Zoltan Kiss; dev at dpdk.org
>> Subject: Re: [dpdk-dev] rte_mbuf's packet_type field
>>
>> Hi
The patch introduces a new PMD. This PMD is implemented as thin wrapper
of librte_vhost. It means librte_vhost is also needed to compile the PMD.
The vhost messages will be handled only when a port is started. So start
a port first, then invoke QEMU.
The PMD has 2 parameters.
- iface: The parame
This patch adds a below event type.
- RTE_ETH_EVENT_QUEUE_STATE_CHANGE
This event is used for notifying a queue state changed event.
Signed-off-by: Tetsuya Mukawa
Acked-by: Ferruh Yigit
---
lib/librte_ether/rte_ethdev.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/lib/librte_ether/rte
> Subject: [dpdk-dev] [PATCH] igb_uio: cast private data to correct struct type
>
> Fixes: af75078fece3 ("first public release")
>
> This was working fine because addresses of two structs are same:
>
> struct A {
> struct B b;
> } a;
>
> As above sample "a" and "b" has same address.
>
>
The patch introduces a new PMD. This PMD is implemented as thin wrapper
of librte_vhost.
PATCH v10 changes:
- Rebase on latest master.
- Fix DPDK version number(2.3 to 16.04)
- Set port id to mbuf while receiving packets.
PATCH v9 changes:
- Fix a null pointer access issue implemented in v8
This is a wholesale replacement of the Enic PMD receive path in order
to improve performance and code clarity. The changes are:
- Simplify and reduce code path length of receive function.
- Put most of the fast-path receive funtions in one file.
- Reduce the number of posted_index updates (pay atte
Initial guide for bnxt driver, documents current limitations and
provides information link.
Signed-off-by: Stephen Hurd
---
doc/guides/nics/bnxt.rst | 49
1 file changed, 49 insertions(+)
create mode 100644 doc/guides/nics/bnxt.rst
diff --git a/
Signed-off-by: Stephen Hurd
---
doc/guides/nics/overview.rst | 64 ++--
1 file changed, 32 insertions(+), 32 deletions(-)
diff --git a/doc/guides/nics/overview.rst b/doc/guides/nics/overview.rst
index d4c6ff4..e606bdf 100644
--- a/doc/guides/nics/overview.
Signed-off-by: Stephen Hurd
---
config/common_bsdapp | 5 +
config/common_linuxapp | 5 +
drivers/net/Makefile | 1 +
mk/rte.app.mk | 1 +
4 files changed, 12 insertions(+)
diff --git a/config/common_bsdapp b/config/common_bsdapp
index 696382c..f37c7bb 100644
--- a/config/co
Claim ownership of new drivers/net/bnxt driver.
Signed-off-by: Stephen Hurd
---
MAINTAINERS | 4
1 file changed, 4 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 628bc05..6ee6c3c 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -359,6 +359,10 @@ F: drivers/crypto/aesni_mb/
Intel
New driver for Broadcom bnxt (NexXtreme C-series) devices.
Standards-compliant 10/25/50G support with 30MPPS full-duplex throughput
http://www.broadcom.com/press/release.php?id=s923886
Signed-off-by: Stephen Hurd
---
v3:
* Fix incorrect format specifier compilation error on i686
(PRIx64 instea
Add Broadcom Vendor ID and RTE_PCI_DEV_ID_DECL_BNXT() macro.
Add Device IDs for Broadcom bnxt devices.
Signed-off-by: Stephen Hurd
---
lib/librte_eal/common/include/rte_pci_dev_ids.h | 45 ++---
1 file changed, 40 insertions(+), 5 deletions(-)
diff --git a/lib/librte_eal/com
Add additional ETH_LINK_SPEED_* macros for 2, 2.5, 25, and 50 Gbps links
Signed-off-by: Stephen Hurd
---
lib/librte_ether/rte_ethdev.h | 4
1 file changed, 4 insertions(+)
diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
index 16da821..cb40bbb 100644
--- a/lib/lib
New driver for Broadcom NetXtreme-C family of controllers and cards
capable of up to 50Gbps link with 30Mpps throughput.
v2:
* Split into multiple patches
* Add nic guide
* Add features in overview.rst
v3:
* Fix incorrect format specifier compilation error on i686
(PRIx64 instead of lx for uint
On Fri, 4 Mar 2016 15:25:24 +
Remy Horton wrote:
> Implements driver support for setting of MAC address.
>
> Signed-off-by: Remy Horton
> ---
> doc/guides/rel_notes/release_16_04.rst | 4
> drivers/net/vmxnet3/vmxnet3_ethdev.c | 19 +++
> 2 files changed, 23 insert
On 03/03/2016 08:37 PM, Thomas Monjalon wrote:
>> --- a/config/defconfig_x86_64-native-bsdapp-clang
>> +++ b/config/defconfig_x86_64-native-bsdapp-clang
>> @@ -37,6 +37,7 @@ CONFIG_RTE_MACHINE="native"
>> CONFIG_RTE_ARCH="x86_64"
>> CONFIG_RTE_ARCH_X86_64=y
>> CONFIG_RTE_ARCH_X86=y
>> +CONFIG
In order to cleanup the configuration files some and reduce
the number of duplicate configuration information. Add a new
file called common_base which contains just about all of the
configuration lines in one place. Then have the common_bsdapp,
common_linuxapp files include this one file. Then in t
The videos of the conference FOSDEM 2016 are available.
This is my list of DPDK related talks:
* Saying again that "Userspace is not slow"
https://fosdem.org/2016/schedule/event/linux_petascale_storage/
* OPNFV
https://fosdem.org/2016/schedule/event/telco_on_free_software/
* DPDK
On 04/03/16 01:47, Lu, Wenzhuo wrote:
> Hi Zoltan,
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Zoltan Kiss
>> Sent: Wednesday, March 2, 2016 3:19 AM
>> To: dev at dpdk.org
>> Subject: [dpdk-dev] ixgbe TX function selection
>>
>> Hi,
>>
>> I've notice
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Ferruh Yigit
> Sent: Friday, March 04, 2016 11:21 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v2] igb_uio: use macros for array size calculation
>
> Minor code cleanup.
> Remove array size calculatio
On 03/04/2016 11:28 AM, Traynor, Kevin wrote:
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
>> Sent: Thursday, March 3, 2016 6:38 PM
>> To: Wiles, Keith
>> Cc: dev at dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH] config: remove duplicate c
On 03/03/2016 08:37 PM, Thomas Monjalon wrote:
> 2016-02-22 07:53, Keith Wiles:
>> --- /dev/null
>> +++ b/config/common_base
>> +CONFIG_RTE_EAL_IGB_UIO=y
>> +CONFIG_RTE_EAL_VFIO=y
>
> These options should be disabled in the base file
> and enabled in Linux.
>
>> +CONFIG_RTE_LIBRTE_PMD_AF_PACKET=y
>
In certain autotests lpm->max_rules turned out to be non initialized.
That was caused by a failing allocation for lpm->rules_tbl in rte_lpm6_create.
It then left the function via goto exit with lpm freed, but still a pointer
value being set.
In case of an allocation failure it resets lpm to NULL n
2016-03-04 09:58, Traynor, Kevin:
> From: Panu Matilainen [mailto:pmatilai at redhat.com]
> > On 03/04/2016 11:28 AM, Traynor, Kevin wrote:
> > > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
> > >> 2016-02-22 07:53, Keith Wiles:
> > >>> +CONFIG_RTE_LIBRTE_VHOST=y
> > >>
>
Minor code cleanup.
Remove array size calculations and remove unnecessary assignment.
Signed-off-by: Ferruh Yigit
---
lib/librte_eal/linuxapp/igb_uio/igb_uio.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/lib/librte_eal/linuxapp/igb_uio/igb_uio.c
b/lib/librte_eal/
Signed-off-by: Keith Wiles
---
config/defconfig_x86_64-native-bsdapp-clang | 1 +
config/defconfig_x86_64-native-bsdapp-gcc | 1 +
2 files changed, 2 insertions(+)
diff --git a/config/defconfig_x86_64-native-bsdapp-clang
b/config/defconfig_x86_64-native-bsdapp-clang
index d2baf2c..8b870b3 100
In order to cleanup the configuration files some and reduce
the number of duplicate configuration information. Add a new
file called common_base which contains just about all of the
configuration lines in one place. Then have the common_bsdapp,
common_linuxapp files include this one file. Then in t
Hi,
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Olivier MATZ
> Sent: Friday, March 04, 2016 9:29 AM
> To: Zoltan Kiss; dev at dpdk.org
> Subject: Re: [dpdk-dev] rte_mbuf's packet_type field
>
> Hi Zoltan,
>
> On 03/01/2016 06:15 PM, Zoltan Kiss wrote:
>
The patch add VxLAN & NVGRE TX checksum off-load. When the flag of
outer IP header checksum offload is set, we'll set the context
descriptor to enable this checksum off-load.
Also update release note for VxLAN & NVGRE checksum off-load support.
Signed-off-by: Wenzhuo Lu
---
doc/guides/rel_notes
X550 will do VxLAN & NVGRE RX checksum off-load automatically.
This patch exposes the result of the checksum off-load.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ixgbe/ixgbe_ethdev.c | 4
drivers/net/ixgbe/ixgbe_rxtx.c | 11 ++-
lib/librte_ether/rte_ethdev.h| 1 +
lib/librte_
Add UDP tunnel port add/del support on ixgbe. Now only
support VxLAN port configuration.
Although according to the specification the VxLAN port has
a default value 4789, it can be changed. We support VxLAN
port configuration to meet the change.
Note, the default value of VxLAN port in ixgbe NICs is
As the names of tunnel port config functions are not
accurate, change them from tunnel_add/del to
tunnel_port_add/del.
And support both the old and new rte ops.
Signed-off-by: Wenzhuo Lu
---
drivers/net/i40e/i40e_ethdev.c | 22 --
1 file changed, 12 insertions(+), 10 deletion
The names of function for tunnel port configuration are not
accurate. They're tunnel_add/del, better change them to
tunnel_port_add/del.
As it may be an ABI change if change the names directly, the
new functions are added but not remove the old ones. The old
ones will be removed in the next release
This patch set add the VxLAN & NVGRE checksum off-load support.
Both RX and TX checksum off-load can be used for VxLAN & NVGRE.
And the VxLAN port can be set, it's implemented in this patch
set either.
v2:
- Update release note.
v3:
- Update RX/TX offload capability.
- Reuse PKT_RX_EIP_CKSUM_BAD
On Thu, Mar 03, 2016 at 05:40:14PM +, Xie, Huawei wrote:
> On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> > The current rte_vhost_dequeue_burst() implementation is a bit messy
> [...]
> > +
> > uint16_t
> > rte_vhost_dequeue_burst(struct virtio_net *dev, uint16_t queue_id,
> > struct rte_memp
Hi Zoltan,
On 03/01/2016 06:15 PM, Zoltan Kiss wrote:
> I have a quick question about this field: how do I know if the
> underlying PMD supports a particular protocol parsing. Let's say I want
> to check for SCTP packets, looking at this field tells me EITHER the
> packet is SCTP (or not), OR that
On Thu, Mar 03, 2016 at 04:21:19PM +, Xie, Huawei wrote:
> On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> > The current rte_vhost_dequeue_burst() implementation is a bit messy
> > and logic twisted. And you could see repeat code here and there: it
> > invokes rte_pktmbuf_alloc() three times at thre
Hi Tetsuya,
On 2/22/2016 4:17 PM, Tetsuya Mukawa wrote:
> The option specifies how to mmap EAL memory.
> If the option is specified like '--range-virtaddr=-',
> EAL will check /proc/maps, then tries to find free region between addr1
> and addr2. If a region is found, EAL will treat it as if 'base-
All the error checks in virtqueue_enqueue_xmit are already done
by the caller. Therefore they can be removed to improve performance.
Signed-off-by: Stephen Hemminger
---
drivers/net/virtio/virtio_rxtx.c | 23 ++-
1 file changed, 2 insertions(+), 21 deletions(-)
diff --git a/
Virtio supports a feature that allows sender to put transmit
header prepended to data. It requires that the mbuf be writeable, correct
alignment, and the feature has been negotiatied. If all this works out,
then it will be the optimum way to transmit a single segment packet.
Signed-off-by: Steph
The virtio ring in QEMU/KVM is usually limited to 256 entries
and the normal way that virtio driver was queuing mbufs required
nsegs + 1 ring elements. By using the indirect ring element feature
if available, each packet will take only one ring slot even for
multi-segment packets.
Signed-off-by: S
This patch series uses virtio negotiated features to allow for
more packets to be queued to host even though the default QEMU/KVM
virtio queue is very small 256 elements.
Stephen Hemminger (3):
virtio: use indirect ring elements
virtio: use any layout on transmit
virtio: optimize transmit en
Hi Paul,
On 02/15/2016 05:55 PM, Paul Emmerich wrote:
> Signed-off-by: Paul Emmerich
> ---
> lib/librte_mbuf/rte_mbuf.c | 1 +
> lib/librte_mbuf/rte_mbuf.h | 6 ++
> 2 files changed, 7 insertions(+)
>
> diff --git a/lib/librte_mbuf/rte_mbuf.c b/lib/librte_mbuf/rte_mbuf.c
> index f506517..74
Hi Tetsuya,
Seems that this patch is too long. Is it possible to split into multiple
commits?
On 2/22/2016 4:17 PM, Tetsuya Mukawa wrote:
> The patch adds a new virtio-net PMD configuration that allows the PMD to
> work on host as if the PMD is in VM.
> Here is new configuration for virtio-net P
On Fri, 4 Mar 2016 06:18:17 +
"Xie, Huawei" wrote:
> On 1/14/2016 9:49 PM, Xie, Huawei wrote:
> > On 1/6/2016 8:04 PM, Thomas Monjalon wrote:
> >> 2016-01-05 08:10, Xie, Huawei:
> >>> On 10/26/2015 10:06 PM, Xie, Huawei wrote:
> On 10/19/2015 1:16 PM, Stephen Hemminger wrote:
> > Thi
On Thu, Mar 03, 2016 at 04:30:42PM +, Xie, Huawei wrote:
> On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> > + mbuf_avail = 0;
> > + mbuf_offset = 0;
> > + while (desc_avail || (desc->flags & VRING_DESC_F_NEXT) != 0) {
> > + /* This desc reachs to its end, get the next one */
> > +
On 2016/03/02 11:24, Qiu, Michael wrote:
> On 3/1/2016 10:19 AM, Tetsuya Mukawa wrote:
>> On 2016/03/01 11:00, Qiu, Michael wrote:
>>> On 2/26/2016 4:36 PM, Tetsuya Mukawa wrote:
On 2016/02/26 13:29, Tetsuya Mukawa wrote:
> [...]
>
> BTW, I have set the frontend mergeable off.
> I have
1 - 100 of 139 matches
Mail list logo
