2016-05-19 15:35, Jasvinder Singh:
> Removes rte_acl_dump() call from rte_table_acl_build () as it invokes
> number of printf messages.
>
> Signed-off-by: Jasvinder Singh
> Acked-by: Cristian Dumitrescu
Applied, thanks
Dear All,
I implemented a program that uses flow director to forward packets to a
specific virtual function, however I faced the problem that I did not know
which queue belongs to a VF. I found in [1] that in the case of Intel
82599, queues 0-7 belongs to VF0, 8-15 to VF1 and so on, I tested it bu
2016-05-17 17:02, Fan Zhang:
> This patch fixes the NULL packet processing problem. Originally,
> pipeline's write attempt to NULL packets generated by source
> port may crash the application.
I still don't understand what is *the* NULL packet processing problem.
Anyway applied, thanks
> Piotr Azarewicz (7):
> examples/ip_pipeline: add helper functions for parsing string
> examples/ip_pipeline: modifies common pipeline CLI
> examples/ip_pipeline: modifies firewall pipeline CLI
> examples/ip_pipeline: modifies flow classifications pipeline CLI
> examples/ip_pipeline: mod
2016-05-25 15:34, Piotr Azarewicz:
> This patch improve generate_random_key() function by replacing rand()
> function with reading from /dev/urandom.
>
> CID 120136 : Calling risky function (DC.WEAK_CRYPTO)
> dont_call: rand should not be used for security related applications, as
> linear congrue
> When initializing crypto devices within the app,
> the provided key sizes are checked against the supported
> sizes from the crypto device capabilities.
>
> When the supported sizes are not a range, but a single value,
> the check may become an infinite loop (when size is not supported).
>
> Fi
2016-05-25 10:00, Fan Zhang:
> This patch enables AES-XCBC-MAC authentication algorithm support to
> l2fwd-crypto sample application.
>
> Signed-off-by: Fan Zhang
Applied, thanks
2016-05-25 10:00, Fan Zhang:
> Fixes ba7b86b1 ("doc: add l2fwd-crypto sample app guide")
>
> Corrected a typo in application name.
>
> Corrected authentication algorithm to fit the sample 16-byte
> authentication key.
>
> Signed-off-by: Fan Zhang
Applied, thanks
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
---
MAINTAINERS | 6 ++
1 file changed, 6 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 3e8558f..625423f 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -336,6 +336,12 @@ M: Sony Chacko
F: drivers/net/qede/
F: doc/guides/
Updated doc/guides/nics/overview.rst, doc/guides/nics/thunderx.rst
and release notes
Changed "*" to "P" in overview.rst to capture the partially supported
feature as "*" creating alignment issues with Sphinx table
Signed-off-by: Jerin Jacob
Signed-off-by: Slawomir Rosek
Acked-by: John McNamara
Signed-off-by: Jerin Jacob
---
config/defconfig_arm64-thunderx-linuxapp-gcc | 1 +
1 file changed, 1 insertion(+)
diff --git a/config/defconfig_arm64-thunderx-linuxapp-gcc
b/config/defconfig_arm64-thunderx-linuxapp-gcc
index 7940bbd..cc12cee 100644
--- a/config/defconfig_arm64-thunderx-linuxapp
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed-off-by: Radoslaw Biernacki
---
drivers/net/thunderx/nicvf_ethdev.c | 468
1 file changed, 468 insertions
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed-off-by: Radoslaw Biernacki
---
drivers/net/thunderx/nicvf_ethdev.c | 68 +
1 file changed, 68 insertions(
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed-off-by: Radoslaw Biernacki
---
drivers/net/thunderx/nicvf_ethdev.c | 175
drivers/net/thunderx/nicvf_rxt
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed-off-by: Radoslaw Biernacki
---
drivers/net/thunderx/nicvf_ethdev.c | 41 +
drivers/net/thunderx/nicvf_rxt
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed-off-by: Radoslaw Biernacki
---
drivers/net/thunderx/nicvf_ethdev.h | 33
drivers/net/thunderx/nicvf_rxtx.c | 317
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed-off-by: Radoslaw Biernacki
---
drivers/net/thunderx/Makefile | 2 +
drivers/net/thunderx/nicvf_ethdev.c | 5 +-
drivers/net/thunder
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed-off-by: Radoslaw Biernacki
---
drivers/net/thunderx/nicvf_ethdev.c | 69 +
1 file changed, 69 insertions(
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed-off-by: Radoslaw Biernacki
---
drivers/net/thunderx/nicvf_ethdev.c | 53 +
drivers/net/thunderx/nicvf_eth
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed-off-by: Radoslaw Biernacki
---
drivers/net/thunderx/nicvf_ethdev.c | 182
1 file changed, 182 insertions
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed-off-by: Radoslaw Biernacki
---
drivers/net/thunderx/nicvf_ethdev.c | 179
1 file changed, 179 insertions
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed-off-by: Radoslaw Biernacki
---
drivers/net/thunderx/nicvf_ethdev.c | 141
drivers/net/thunderx/nicvf_eth
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed-off-by: Radoslaw Biernacki
---
drivers/net/thunderx/nicvf_ethdev.c | 47 +
drivers/net/thunderx/nicvf_eth
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed-off-by: Radoslaw Biernacki
---
drivers/net/thunderx/nicvf_ethdev.c | 79 +
1 file changed, 79 insertions(
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed-off-by: Radoslaw Biernacki
---
drivers/net/thunderx/nicvf_ethdev.c | 30 ++
1 file changed, 30 insertions(+)
dif
Extended the nicvf_interrupt function to respond
NIC_MBOX_MSG_BGX_LINK_CHANGE mbox message from PF and update
struct rte_eth_link accordingly.
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed
Introduce driver initialization and enable build infrastructure for
nicvf pmd driver.
By default, It is enabled only for defconfig_arm64-thunderx-*
config as it is an inbuilt NIC device.
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta
Adds hardware specific API for ThunderX nicvf inbuilt NIC device under
drivers/net/thunderx/nicvf/base directory.
Signed-off-by: Jerin Jacob
Signed-off-by: Maciej Czekaj
Signed-off-by: Kamil Rytarowski
Signed-off-by: Zyta Szpak
Signed-off-by: Slawomir Rosek
Signed-off-by: Radoslaw Biernacki
This patch set provides the initial version of DPDK PMD for the
built-in NIC device in Cavium ThunderX SoC family.
Implemented features and ThunderX nicvf PMD documentation added
in doc/guides/nics/overview.rst and doc/guides/nics/thunderx.rst
respectively in this patch set.
These patches are che
2016-05-31 10:09, Fan Zhang:
> Fan Zhang (5):
> app/test: categorize crypto AES test vectors into new file
> app/test: add unified crypto aes test
> app/test: utilize new unified crypto AES test function
The first three patches have no real meaning separately.
As they move and rework some co
clang raises this error:
app/test/test_cryptodev_aes.c:318:7: fatal error: variable 'op' is used
uninitialized whenever 'if' condition is true [-Wsometimes-uninitialized]
if (!buf_p) {
^~
> > This patchset adds counter mode support to AES-NI multi-buffer library
> > and appropriate test cases.
> >
> > This patchset depends on the following patch/patchsets
> > "doc: fix supported AES-CBC key lengths"
> > (http://dpdk.org/dev/patchwork/patch/12398/)
> > and "Added AES counter mode ca
> > This patch changes to -O3 optimization flag in Intel QuickAssist Technology
> > driver Makefile
> >
> > Arek Kusztal (2):
> > qat: add aad_len variable to avoid GCC break strict-aliasing rules
> > warning
> > qat: change optimization flag for Intel QuickAssist Technology
> > Makefi
On Tue, Jun 07, 2016 at 02:42:58PM +0300, Panu Matilainen wrote:
> On 06/07/2016 06:51 AM, Yuanhan Liu wrote:
> >Introduce a new API rte_vhost_get_numa_node() to get the numa node
> >from which the virtio_net struct is allocated.
> >
> >Signed-off-by: Yuanhan Liu
> >Tested-by: Rich Lane
> >Acked-
Hello,
I have a question regarding QoS grinder implementation, specifically,
about the way queue WRR tokens are copied from pipe to grinder and back.
First, rte_sched_grinder uses uint16_t and rte_sched_pipe uses uint8_t
to represent wrr_tokens. Second, instead of just copying the tokens, we
s
On Mon, 6 Jun 2016 13:40:47 +0800
Wenzhuo Lu wrote:
> Define lock mode for RX/TX queue. Because when resetting
> the device we want the resetting thread to get the lock
> of the RX/TX queue to make sure the RX/TX is stopped.
>
> Using next ABI macro for this ABI change as it has too
> much impa
> From: Kusztal, ArkadiuszX
> This patchset adds AES counter mode capability for Intel QuickAssist
> Technology crypto driver.
> It adds six test cases for 16B, 24B, 32B key size.
> Series-Acked-by: Deepak Kumar JAIN
Applied, thanks
> From: Bruce Richardson [mailto:bruce.richardson at intel.com]
> Sent: Tuesday, June 07, 2016 5:47 AM
> To: Rasesh Mody
> Cc: dev at dpdk.org; Dept-Eng DPDK Dev
> Subject: Re: [PATCH 0/9] qede: update qede PMD to 1.1.0.1
>
> On Mon, May 23, 2016 at 12:56:50AM +, Rasesh Mody wrote:
> > Hi Br
Changed to -O3 optimization flag in Intel QuickAssist Technology Makefile
Signed-off-by: Arek Kusztal
---
drivers/crypto/qat/Makefile | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/crypto/qat/Makefile b/drivers/crypto/qat/Makefile
index 258c2d5..ee72a61 100644
--- a/drivers/crypto/q
To avoid GCC warning about "dereferencing type-punned pointer will break
strict-aliasing rules" aad_len pointer is dereferenced instead of direct
dereferencing of uint32_t* cast of the middle of an array.
Signed-off-by: Arek Kusztal
---
drivers/crypto/qat/qat_adf/qat_algs_build_desc.c | 8 +-
This patch changes to -O3 optimization flag in Intel QuickAssist Technology
driver Makefile
Arek Kusztal (2):
qat: add aad_len variable to avoid GCC break strict-aliasing rules
warning
qat: change optimization flag for Intel QuickAssist Technology
Makefile
drivers/crypto/qat/Makefil
> -Original Message-
> From: Kusztal, ArkadiuszX
> Sent: Tuesday, June 7, 2016 5:58 PM
> To: dev at dpdk.org
> Cc: Trahe, Fiona ; Griffin, John
> ; Jain, Deepak K ;
> Kusztal, ArkadiuszX
> Subject: [PATCH 0/2] Change optimization flag for Intel QuickAssist
> Technology driver Makefile
>
2016-05-31 11:29, De Lara Guarch, Pablo:
> Hi Deepak/Arek,
>
> > Hi Thomas,
> >
> > This change was made to fix the compilation issues arising when optimization
> > flag was changed. Hence its related to same optimization flag patch.
> > Do you want us to separate the Makefile and .c file change?
> > Fix an error with computation of physical address of content descriptor in
> > the
> > symmetric operations session
> >
> > Fixes: 1703e94ac5ce ("qat: add driver for QuickAssist devices")
> >
> > Signed-off-by: Arkadiusz Kusztal
> > ---
> > v2: Added fixes line to commit message
>
> Acked-
2016-05-18 17:12, Deepak Kumar Jain:
> Fix null pointer dereferencing while
> clearing session
>
> Fixes: 1703e94ac5ce ("qat: add driver for QuickAssist devices")
> Coverity ID 126586
>
> Signed-off-by: Deepak kumar JAIN
Applied, thanks
> > Fix null pointer dereferencing by reporing if null and
> > exiting the function.
> >
> > Coverity issue: 126584
> > Fixes: c0f87eb5252b ("cryptodev: change burst API to be crypto op oriented")
> >
> > Signed-off-by: Deepak Kumar Jain
>
> Acked-by: Pablo de Lara
Applied, thanks
> > Fix wrong indentation for return value
> >
> > Coverity ID: 126585
> > Fixes: 924e84f87306 ("aesni_mb: add driver for multi buffer based crypto")
> >
> > Signed-off-by: Deepak Kumar Jain
>
> Acked-by: Pablo de Lara
Applied, thanks
2016-06-03 11:11, Fan Zhang:
> This patchset adds counter mode support to AES-NI multi-buffer library
> and appropriate test cases.
>
> Fan Zhang (3):
> aesni_mb: add counter mode support
> app/test: add aes-ni multi-buffer pmd test cases for AES CTR
> examples/l2fwd-crypto: enable AES coun
2016-05-05 11:00, Pablo de Lara:
> AES-NI MB PMD supports 128, 192 and 256-bit keys,
> not 128, 256 and 512-bit keys.
>
> Fixes: 924e84f87306 ("aesni_mb: add driver for multi buffer based crypto")
>
> Signed-off-by: Pablo de Lara
Applied, thanks
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Neil Horman
> Sent: Monday, June 6, 2016 2:48 PM
> To: Thomas Monjalon
> Cc: dev at dpdk.org; Mcnamara, John ; Christian
> Ehrhardt ; Markos Chandras
> ; Panu Matilainen
> Subject: Re: [dpdk-dev] RFC: DPDK Lo
2016-06-07 15:07, Bruce Richardson:
> On Tue, Jun 07, 2016 at 03:00:45PM +0200, Thomas Monjalon wrote:
> > 2016-06-07 14:36, Christian Ehrhardt:
> > > But I still struggle to see how to fix the circular dependency between
> > > librte_eal and librte_mempool.
> >
> > Why is there a circular depende
On 2016/06/06 19:50, Tan, Jianfeng wrote:
>> Please let me make sure how we can invoke many DPDK applications in
>> hundreds containers.
>> (Do we have a way to do? Or, will we have it in the future?)
>
> Just to add some option here, we cannot say no to that kind of use case.
> To have many insta
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Sergio Gonzalez
> Monroy
> Sent: Wednesday, May 18, 2016 1:42 PM
> To: dev at dpdk.org
> Cc: Mcnamara, John
> Subject: [dpdk-dev] [PATCH v2 7/9] examples/ipsec-secgw: ipv6 support
>
> Support IPSec IPv6 allowi
2016-06-07 09:49, Neil Horman:
> On Tue, Jun 07, 2016 at 03:24:55PM +0200, Thomas Monjalon wrote:
> > 2016-06-07 09:03, Neil Horman:
> > > On Tue, Jun 07, 2016 at 02:53:36PM +0200, Thomas Monjalon wrote:
> > > > 2016-06-07 08:04, Neil Horman:
> > > > > On Tue, Jun 07, 2016 at 11:57:42AM +0200, Thom
Hi Tomasz,
Thanks for splitting tests.
2016-06-07 14:49, Tomasz Kantecki:
> 'red_autotest' changed to run only functional tests without test #4 which was
> taking ~53 seconds. 'red_autotest' takes ~12[s] now.
12s is long for a functional test. We are used to have tests of less than 1s.
Do you th
> -Original Message-
> From: Neil Horman [mailto:nhorman at tuxdriver.com]
> Sent: Sunday, June 5, 2016 7:15 PM
> To: Mcnamara, John
> Cc: dev ; Christian Ehrhardt
> ; Markos Chandras ;
> Panu Matilainen
> Subject: Re: [dpdk-dev] RFC: DPDK Long Term Support
>
> >
> I'm not opposed to a
On 7 June 2016 at 08:36, Christian Ehrhardt <
christian.ehrhardt at canonical.com> wrote:
> On Fri, Jun 3, 2016 at 5:07 PM, Mcnamara, John
> wrote:
> [...]
> >
> > LTS Version
> >
> >
> > The proposed initial LTS version will be DPDK 16.07. The next versions,
> > based
> > on a 2 yea
On Tue, Jun 07, 2016 at 04:19:47PM +0200, Thomas Monjalon wrote:
> 2016-06-07 15:07, Bruce Richardson:
> > On Tue, Jun 07, 2016 at 03:00:45PM +0200, Thomas Monjalon wrote:
> > > 2016-06-07 14:36, Christian Ehrhardt:
> > > > But I still struggle to see how to fix the circular dependency between
> >
On Tue, Jun 07, 2016 at 04:12:28PM +0900, Tetsuya Mukawa wrote:
> On 2016/06/06 19:50, Tan, Jianfeng wrote:
> >> Please let me make sure how we can invoke many DPDK applications in
> >> hundreds containers.
> >> (Do we have a way to do? Or, will we have it in the future?)
> >
> > Just to add some
2016-06-07 09:03, Neil Horman:
> On Tue, Jun 07, 2016 at 02:53:36PM +0200, Thomas Monjalon wrote:
> > 2016-06-07 08:04, Neil Horman:
> > > On Tue, Jun 07, 2016 at 11:57:42AM +0200, Thomas Monjalon wrote:
> > > > 2016-05-31 09:57, Neil Horman:
> > > > > +++ b/buildtools/Makefile
> > > > > @@ -0,0 +1
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Fan Zhang
> Sent: Friday, June 03, 2016 11:12 AM
> To: dev at dpdk.org
> Cc: Doherty, Declan
> Subject: [dpdk-dev] [PATCH v2 0/3] Add AES Counter mode support for AES-
> NI MB PMD
>
> This patchset adds counte
Hi Thomas,
> 12s is long for a functional test. We are used to have tests of less than 1s.
> Do you think this magnitude order can be achieved for red tests?
This is a real challenge :) Let me take another look into it.
Regards,
Tomasz
On Tue, Jun 07, 2016 at 03:00:45PM +0200, Thomas Monjalon wrote:
> 2016-06-07 14:36, Christian Ehrhardt:
> > But I still struggle to see how to fix the circular dependency between
> > librte_eal and librte_mempool.
>
> Why is there a circular dependency?
> Only because of logs using mempool?
>
>
2016-06-07 14:36, Christian Ehrhardt:
> But I still struggle to see how to fix the circular dependency between
> librte_eal and librte_mempool.
Why is there a circular dependency?
Only because of logs using mempool?
> Maybe now is a time to look at this part of the original threads again to
> eve
2016-06-07 08:04, Neil Horman:
> On Tue, Jun 07, 2016 at 11:57:42AM +0200, Thomas Monjalon wrote:
> > 2016-05-31 09:57, Neil Horman:
> > > +++ b/buildtools/Makefile
> > > @@ -0,0 +1,36 @@
> > > +# BSD LICENSE
> > > +#
> > > +# Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
> > >
Implement the device reset function.
1, Add the fake RX/TX functions.
2, The reset function tries to stop RX/TX by replacing
the RX/TX functions with the fake ones and getting the
locks to make sure the regular RX/TX finished.
3, After the RX/TX stopped, reset the VF port, and then
release
Add RX/TX paths with lock for VF. It's used when
the function of link reset on VF is needed.
When the lock for RX/TX is added, the RX/TX can be
stopped. Then we have a chance to reset the VF link.
Please be aware there's performence drop if the lock
path is chosen.
Signed-off-by: Zhe Tao
---
dr
Implement the device reset function.
1, Add the fake RX/TX functions.
2, The reset function tries to stop RX/TX by replacing
the RX/TX functions with the fake ones and getting the
locks to make sure the regular RX/TX finished.
3, After the RX/TX stopped, reset the VF port, and then
release
Add RX/TX paths with lock for VF. It's used when
the function of link reset on VF is needed.
When the lock for RX/TX is added, the RX/TX can be
stopped. Then we have a chance to reset the VF link.
Please be aware there's performence drop if the lock
path is chosen.
Signed-off-by: Wenzhuo Lu
---
Implement the device reset function.
1, Add the fake RX/TX functions.
2, The reset function tries to stop RX/TX by replacing
the RX/TX functions with the fake ones and getting the
locks to make sure the regular RX/TX finished.
3, After the RX/TX stopped, reset the VF port, and then
release
Add RX/TX paths with lock for VF. It's used when
the function of link reset on VF is needed.
When the lock for RX/TX is added, the RX/TX can be
stopped. Then we have a chance to reset the VF link.
Please be aware there's performence drop if the lock
path is chosen.
Signed-off-by: Wenzhuo Lu
---
Define lock mode for RX/TX queue. Because when resetting
the device we want the resetting thread to get the lock
of the RX/TX queue to make sure the RX/TX is stopped.
Using next ABI macro for this ABI change as it has too
much impact. 7 APIs and 1 global variable are impacted.
Signed-off-by: Wenz
Add an API to reset the device.
It's for VF device in this scenario, kernel PF + DPDK VF.
When the PF port down/up, APP should call this API to
reset VF port. Most likely, APP should call it in its
management thread and guarantee the thread safe.
Signed-off-by: Wenzhuo Lu
---
lib/librte_ether/rt
If the PF link is down and up, VF link will not work
accordingly.
This patch set addes the support of VF link reset. So, when VF
receices the messges of physical link down/up. APP can reset the
VF link and let it recover.
PS: This patch set is splitted from a previous patch set, *automatic
link re
'red_autotest' changed to run only functional tests without test #4 which was
taking ~53 seconds. 'red_autotest' takes ~12[s] now.
'red_perf' has been added to run performance tests only (~26[s]).
'red_all' has been added to run all functional tests (including #4) and
perfromance tests. This reflec
On 06/07/2016 06:51 AM, Yuanhan Liu wrote:
> Introduce a new API rte_vhost_get_numa_node() to get the numa node
> from which the virtio_net struct is allocated.
>
> Signed-off-by: Yuanhan Liu
> Tested-by: Rich Lane
> Acked-by: Rich Lane
> ---
> drivers/net/vhost/rte_eth_vhost.c | 13 --
On 6/7/16, 9:19 AM, "dev on behalf of Thomas Monjalon" wrote:
>2016-06-07 15:07, Bruce Richardson:
>> On Tue, Jun 07, 2016 at 03:00:45PM +0200, Thomas Monjalon wrote:
>> > 2016-06-07 14:36, Christian Ehrhardt:
>> > > But I still struggle to see how to fix the circular dependency between
>> > > li
Hi Panu,
thanks for refreshing that - I'm much more happy with ldd output of most
libraries now after build with that.
And it comes much slimmer than my initial approach that touched all lib
makefiles.
Acked-by: Christian Ehrhardt
FYI - for anyone else also trying this in a backported way to 16.
On Fri, Jun 3, 2016 at 5:07 PM, Mcnamara, John
wrote:
[...]
>
> LTS Version
>
>
> The proposed initial LTS version will be DPDK 16.07. The next versions,
> based
> on a 2 year cycle, will be DPDK 18.08, 20.08, etc.
>
I can see on the discussions that much more things around this have
On Tue, Jun 7, 2016 at 1:49 PM, Mauricio V?squez
wrote:
> Dear All,
>
> I implemented a program that uses flow director to forward packets to a
> specific virtual function, however I faced the problem that I did not know
> which queue belongs to a VF. I found in [1] that in the case of Intel
> 825
2016-06-07 08:08, Neil Horman:
> On Tue, Jun 07, 2016 at 11:34:56AM +0200, Thomas Monjalon wrote:
> > Hi Neil,
> >
> > You are creating a new directory buildtools/ which depends on mk/
> I don't know what this means. Every part of the dpdk that builds code depends
> on the infrastructure in mk/ (
In ipsec-secgw wrong SPI number is read from incoming ESP packet.
The problem exist inside function inbound_sa_lookup().
The SPI is read from mbuf where the information is stored in big-endian.
In low-endian environment the value is erroneous.
Fixed by add conversion rte_be_to_cpu_32().
Fixes: d29
Implement the device reset function.
1, Add the fake RX/TX functions.
2, The reset function tries to stop RX/TX by replacing
the RX/TX functions with the fake ones and getting the
locks to make sure the regular RX/TX finished.
3, After the RX/TX stopped, reset the VF port, and then
release
Add RX/TX paths with lock for VF. It's used when
the function of link reset on VF is needed.
When the lock for RX/TX is added, the RX/TX can be
stopped. Then we have a chance to reset the VF link.
Please be aware there's performence drop if the lock
path is chosen.
Signed-off-by: Zhe Tao
---
dr
Implement the device reset function.
1, Add the fake RX/TX functions.
2, The reset function tries to stop RX/TX by replacing
the RX/TX functions with the fake ones and getting the
locks to make sure the regular RX/TX finished.
3, After the RX/TX stopped, reset the VF port, and then
release
Add RX/TX paths with lock for VF. It's used when
the function of link reset on VF is needed.
When the lock for RX/TX is added, the RX/TX can be
stopped. Then we have a chance to reset the VF link.
Please be aware there's performence drop if the lock
path is chosen.
Signed-off-by: Wenzhuo Lu
---
Implement the device reset function.
1, Add the fake RX/TX functions.
2, The reset function tries to stop RX/TX by replacing
the RX/TX functions with the fake ones and getting the
locks to make sure the regular RX/TX finished.
3, After the RX/TX stopped, reset the VF port, and then
release
Add RX/TX paths with lock for VF. It's used when
the function of link reset on VF is needed.
When the lock for RX/TX is added, the RX/TX can be
stopped. Then we have a chance to reset the VF link.
Please be aware there's performence drop if the lock
path is chosen.
Signed-off-by: Wenzhuo Lu
---
Define lock mode for RX/TX queue. Because when resetting
the device we want the resetting thread to get the lock
of the RX/TX queue to make sure the RX/TX is stopped.
Using next ABI macro for this ABI change as it has too
much impact. 7 APIs and 1 global variable are impacted.
Signed-off-by: Wenz
Add an API to reset the device.
It's for VF device in this scenario, kernel PF + DPDK VF.
When the PF port down/up, APP should call this API to
reset VF port. Most likely, APP should call it in its
management thread and guarantee the thread safe.
Signed-off-by: Wenzhuo Lu
---
lib/librte_ether/rt
If the PF link is down and up, VF link will not work
accordingly.
This patch set addes the support of VF link reset. So, when VF
receices the messges of physical link down/up. APP can reset the
VF link and let it recover.
PS: This patch set is splitted from a previous patch set, *automatic
link re
On 07/06/2016 13:17, Slawomir Mrozowicz wrote:
> In ipsec-secgw wrong SPI number is read from incoming ESP packet.
> The problem exist inside function inbound_sa_lookup().
> The SPI is read from mbuf where the information is stored in big-endian.
> In low-endian environment the value is erroneous.
On Mon, May 23, 2016 at 12:56:50AM +, Rasesh Mody wrote:
> Hi Bruce,
>
> > From: Rasesh Mody [mailto:rasesh.mody at qlogic.com]
> > Sent: Friday, May 06, 2016 9:30 PM
> >
> > This patch set adds support for enabling 100G mode for QEDE PMD.
> > It also adds support for APIs like
> > - mtu_set
From: "zhe.tao"
Implement the device reset function.
1, Add the fake RX/TX functions.
2, The reset function tries to stop RX/TX by replacing
the RX/TX functions with the fake ones and getting the
locks to make sure the regular RX/TX finished.
3, After the RX/TX stopped, reset the VF port, a
From: "zhe.tao"
Add RX/TX paths with lock for VF. It's used when
the function of link reset on VF is needed.
When the lock for RX/TX is added, the RX/TX can be
stopped. Then we have a chance to reset the VF link.
Please be aware there's performence drop if the lock
path is chosen.
Signed-off-by
From: Wenzhuo Lu
Implement the device reset function.
1, Add the fake RX/TX functions.
2, The reset function tries to stop RX/TX by replacing
the RX/TX functions with the fake ones and getting the
locks to make sure the regular RX/TX finished.
3, After the RX/TX stopped, reset the VF port,
From: Wenzhuo Lu
Add RX/TX paths with lock for VF. It's used when
the function of link reset on VF is needed.
When the lock for RX/TX is added, the RX/TX can be
stopped. Then we have a chance to reset the VF link.
Please be aware there's performence drop if the lock
path is chosen.
Signed-off-b
From: Wenzhuo Lu
Implement the device reset function.
1, Add the fake RX/TX functions.
2, The reset function tries to stop RX/TX by replacing
the RX/TX functions with the fake ones and getting the
locks to make sure the regular RX/TX finished.
3, After the RX/TX stopped, reset the VF port,
From: Wenzhuo Lu
Add RX/TX paths with lock for VF. It's used when
the function of link reset on VF is needed.
When the lock for RX/TX is added, the RX/TX can be
stopped. Then we have a chance to reset the VF link.
Please be aware there's performence drop if the lock
path is chosen.
Signed-off-b
From: Wenzhuo Lu
Define lock mode for RX/TX queue. Because when resetting
the device we want the resetting thread to get the lock
of the RX/TX queue to make sure the RX/TX is stopped.
Using next ABI macro for this ABI change as it has too
much impact. 7 APIs and 1 global variable are impacted.
1 - 100 of 182 matches
Mail list logo
