Re: [dpdk-dev] [dpdk-stable] [PATCH] mbuf: fix comment about vxlan destination port used

[Thomas Monjalon]

10/07/2017 15:33, Olivier Matz:
> On Fri,  7 Jul 2017 15:17:40 +0100, Cian Ferriter  
> wrote:
> > IANA assigns a destination port of 4789 for the VXLAN in the Service
> > Name and Transport Protocol Port Number Registry. This is mentioned in
> > RFC 7348.
> > 
> > Signed-off-by: Cian Ferriter 
> 
> Fixes: f295a00a2b44 ("mbuf: add definitions of unified packet types")
> CC: sta...@dpdk.org
> 
> Acked-by: Olivier Matz 

Applied, thanks

Re: [dpdk-dev] [PATCH] examples/load_balancer: enable the build for lesser lcores

[Hemant Agrawal]

On 7/19/2017 10:51 AM, Thomas Monjalon wrote:

17/07/2017 11:41, Hemant Agrawal:

--- a/examples/load_balancer/main.h
+++ b/examples/load_balancer/main.h
 #ifndef APP_MAX_IO_LCORES
+#if (APP_MAX_LCORES > 16)
 #define APP_MAX_IO_LCORES 16
+#else
+#define APP_MAX_IO_LCORES APP_MAX_LCORES
 #endif
-#if (APP_MAX_IO_LCORES > APP_MAX_LCORES)
-#error "APP_MAX_IO_LCORES is too big"
 #endif

[...]

 #ifndef APP_MAX_WORKER_LCORES
+#if (APP_MAX_LCORES > 16)
 #define APP_MAX_WORKER_LCORES 16
+#else
+#define APP_MAX_WORKER_LCORES APP_MAX_LCORES
 #endif
-#if (APP_MAX_WORKER_LCORES > APP_MAX_LCORES)
-#error "APP_MAX_WORKER_LCORES is too big"
 #endif


Why removing the checks > APP_MAX_LCORES ?


It is not going to be needed unless someone defines it.
I will send v2 with reverting them back.

Re: [dpdk-dev] [dpdk-stable] [PATCH v2] examples/qos_sched: fix core limit for lower num of lcore

[Hemant Agrawal]

On 7/19/2017 10:54 AM, Thomas Monjalon wrote:

17/07/2017 11:37, Hemant Agrawal:

--- a/examples/qos_sched/main.h
+++ b/examples/qos_sched/main.h
@@ -69,8 +69,13 @@ extern "C" {
 #define BURST_TX_DRAIN_US 100

 #ifndef APP_MAX_LCORE
+#if (APP_MAX_LCORE > 64)
 #define APP_MAX_LCORE 64


If APP_MAX_LCORE is not defined, it cannot be > 64, right?
Or what am I missing?


It should be RTE_MAX_LCORE


+#else
+#define APP_MAX_LCORE RTE_MAX_LCORE
+#endif
 #endif

Re: [dpdk-dev] [PATCH 0/3] L2fwd-crypto fixes

[De Lara Guarch, Pablo]

> -Original Message-
> From: De Lara Guarch, Pablo
> Sent: Tuesday, July 18, 2017 8:58 AM
> To: Doherty, Declan 
> Cc: dev@dpdk.org; De Lara Guarch, Pablo
> 
> Subject: [PATCH 0/3] L2fwd-crypto fixes
> 
> After the changes added to include AEAD algorithm specific parameters,
> there were some parameters that were being set wrong or not even set.
> 
> Pablo de Lara (3):
>   examples/l2fwd-crypto: fix digest length
>   examples/l2fwd-crypto: fix AEAD IV setting
>   examples/l2fwd-crypto: fix AEAD key setting
> 
>  examples/l2fwd-crypto/main.c | 11 +--
>  1 file changed, 9 insertions(+), 2 deletions(-)
> 
> --
> 2.9.4

Applied to dpdk-next-crypto.

Pablo

Re: [dpdk-dev] Occasional instability in RSS Hashes/Queues from X540 NIC

[Li, Xiaoyun]

Hi, Yuanhan !
I did the same tests on DPDK 16.11 this morning and cannot reproduce the 
problem on 16.11. It seems that the problem has been fixed on version 16.11.
So could you reproduce the problem on 16.11 LTS release? I think maybe it has 
been fixed already.


Best Regards,
Xiaoyun Li



-Original Message-
From: Yuanhan Liu [mailto:y...@fridaylinux.org] 
Sent: Tuesday, July 18, 2017 22:01
To: Matt Laswell 
Cc: Yang, Qiming ; dev@dpdk.org; Li, Xiaoyun 

Subject: Re: [dpdk-dev] Occasional instability in RSS Hashes/Queues from X540 
NIC

On Tue, Jul 18, 2017 at 08:43:42AM -0500, Matt Laswell wrote:
> Hi Qiming,
> 
> That's fantastic news.  Thank you very much for taking the time to 
> figure the issue out.
> 
> Would it be possible to backport the fix to the 16.11 LTS release?   This
> kind of problem seems tailor-made for LTS.

Agreed. You might want to find the fix commit with git bisect.

--yliu

[dpdk-dev] [PATCH v3] examples/qos_sched: fix core limit for lower num of lcore

[Hemant Agrawal]

APP_MAX_LCORES is hardcoded as 64.
This will cause build err when RTE_MAX_LCORE is less then 64.

"args.c:127:22: error: iteration 8 invokes undefined behavior
[-Werror=aggressive-loop-optimizations]
   if (cfg->lcore_role[i] == ROLE_RTE)
   ~~~^~~
args.c:126:2: note: within this loop
  for (i = 0; i < APP_MAX_LCORE; i++) {"

Fixes: d52b5e735aa3 ("examples/qos_sched: fix lcore limit")
Cc: sta...@dpdk.org

Signed-off-by: Hemant Agrawal 
---
v3: fix as per review comment of Thomas

 examples/qos_sched/main.h | 5 +
 1 file changed, 5 insertions(+)

diff --git a/examples/qos_sched/main.h b/examples/qos_sched/main.h
index c7490c6..8d02e1a 100644
--- a/examples/qos_sched/main.h
+++ b/examples/qos_sched/main.h
@@ -69,8 +69,13 @@ extern "C" {
 #define BURST_TX_DRAIN_US 100
 
 #ifndef APP_MAX_LCORE
+#if (RTE_MAX_LCORE > 64)
 #define APP_MAX_LCORE 64
+#else
+#define APP_MAX_LCORE RTE_MAX_LCORE
+#endif
 #endif
+
 #define MAX_DATA_STREAMS (APP_MAX_LCORE/2)
 #define MAX_SCHED_SUBPORTS 8
 #define MAX_SCHED_PIPES4096
-- 
2.7.4

[dpdk-dev] [PATCH v2] examples/load_balancer: enable the build for lesser lcores

[Hemant Agrawal]

load_balancer app can also work for lower number of cores.
Limit the cores Worker and IO cores to 16 as defined in original
App. Otherwise use the actual number of lcores as MAX.

Signed-off-by: Hemant Agrawal 
---
v2: Revert back the condition check

 examples/load_balancer/main.h | 8 
 1 file changed, 8 insertions(+)

diff --git a/examples/load_balancer/main.h b/examples/load_balancer/main.h
index d98468a..dc40755 100644
--- a/examples/load_balancer/main.h
+++ b/examples/load_balancer/main.h
@@ -56,7 +56,11 @@
 #endif
 
 #ifndef APP_MAX_IO_LCORES
+#if (APP_MAX_LCORES > 16)
 #define APP_MAX_IO_LCORES 16
+#else
+#define APP_MAX_IO_LCORES APP_MAX_LCORES
+#endif
 #endif
 #if (APP_MAX_IO_LCORES > APP_MAX_LCORES)
 #error "APP_MAX_IO_LCORES is too big"
@@ -74,7 +78,11 @@
 #endif
 
 #ifndef APP_MAX_WORKER_LCORES
+#if (APP_MAX_LCORES > 16)
 #define APP_MAX_WORKER_LCORES 16
+#else
+#define APP_MAX_WORKER_LCORES APP_MAX_LCORES
+#endif
 #endif
 #if (APP_MAX_WORKER_LCORES > APP_MAX_LCORES)
 #error "APP_MAX_WORKER_LCORES is too big"
-- 
2.7.4

Re: [dpdk-dev] [PATCH v2] eal/armv8: fix poly64/128 compile issue in old GCC(<4.9.0)

[Herbert Guan]

Thomas,

Thanks a lot for your review and comment.  But I have some concern in this 
approach.  "poly128_t" is for ARM64 platform only and in fact it's more likely 
that rte_v128u8_t (generic DPDK data type) could be defined from poly128_t (ARM 
data type) which seems more reasonable.

Best regards,
Herbert

-Original Message-
From: Thomas Monjalon [mailto:tho...@monjalon.net]
Sent: Tuesday, July 18, 2017 22:44
To: Herbert Guan 
Cc: dev@dpdk.org; jianbo@linaro.org; jerin.ja...@caviumnetworks.com; 
nelio.laranje...@6wind.com
Subject: Re: [dpdk-dev] [PATCH v2] eal/armv8: fix poly64/128 compile issue in 
old GCC(<4.9.0)

13/07/2017 05:16, Herbert Guan:
> --- a/lib/librte_eal/common/include/arch/arm/rte_vect.h
> +++ b/lib/librte_eal/common/include/arch/arm/rte_vect.h
> +#if (GCC_VERSION < 40900)
> +typedef uint64_t poly64_t;
> +typedef uint64x2_t poly64x2_t;
> +typedef uint8_t poly128_t __attribute__((vector_size(16), aligned(16)));
> +#endif

I think a better fix would be to switch to DPDK types
like rte_v128u8_t.

IMPORTANT NOTICE: The contents of this email and any attachments are 
confidential and may also be privileged. If you are not the intended recipient, 
please notify the sender immediately and do not disclose the contents to any 
other person, use it for any purpose, or store or copy the information in any 
medium. Thank you.

[dpdk-dev] [PATCH v5] ether: add support for vtune task tracing

[ilia . kurakin]

From: Ilia Kurakin 

The patch adds tracing of loop iterations that yielded no packets in a DPDK
application. It is using ITT task API:
https://software.intel.com/en-us/node/544206

We suppose the flow of using this tracing would assume the user has ITT lib
and header on machine and re-build DPDK with additional make parameters:

make EXTRA_CFLAGS=-I
 EXTRA_LDLIBS="-L -littnotify"

Signed-off-by: Ilia Kurakin 

---

-V2 change:
ITT tasks collection is moved to rx callback

-V3 change:
rte_ethdev_profile.c created, all profile specific code moved there.

Added generic profile function

-V4 change:
checkpatch issues fixed

Added documentation topic

-V5 change:
Documentation fixes


 config/common_base|   1 +
 doc/guides/prog_guide/profile_app.rst |  34 
 lib/librte_ether/Makefile |   1 +
 lib/librte_ether/rte_ethdev.c |   4 +
 lib/librte_ether/rte_ethdev_profile.c | 156 ++
 lib/librte_ether/rte_ethdev_profile.h |  52 
 6 files changed, 248 insertions(+)
 create mode 100644 lib/librte_ether/rte_ethdev_profile.c
 create mode 100644 lib/librte_ether/rte_ethdev_profile.h

diff --git a/config/common_base b/config/common_base
index 8ae6e92..dda51db 100644
--- a/config/common_base
+++ b/config/common_base
@@ -136,6 +136,7 @@ CONFIG_RTE_MAX_QUEUES_PER_PORT=1024
 CONFIG_RTE_LIBRTE_IEEE1588=n
 CONFIG_RTE_ETHDEV_QUEUE_STAT_CNTRS=16
 CONFIG_RTE_ETHDEV_RXTX_CALLBACKS=y
+CONFIG_RTE_ETHDEV_PROFILE_ITT_WASTED_RX_ITERATIONS=n
 
 #
 # Turn off Tx preparation stage
diff --git a/doc/guides/prog_guide/profile_app.rst 
b/doc/guides/prog_guide/profile_app.rst
index 54b546a..590cb72 100644
--- a/doc/guides/prog_guide/profile_app.rst
+++ b/doc/guides/prog_guide/profile_app.rst
@@ -59,6 +59,40 @@ Refer to the
 for details about application profiling.
 
 
+Profiling wasted iterations with ITT
+
+
+Iterations that yielded no RX packets (wasted loop iterations) can be analyzed
+using Intel® VTune\ :sup:`TM` Amplifier. This profiling employs the
+`Instrumentation and Tracing Technology (ITT) API
+`_
+feature of VTune Amplifier and requires only reconfiguring of DPDK library,
+no changes in a DPDK application are needed.
+
+To trace wasted iterations on RX queues, first reconfigure DPDK with
+``CONFIG_RTE_ETHDEV_RXTX_CALLBACKS`` and
+``CONFIG_RTE_ETHDEV_PROFILE_ITT_WASTED_RX_ITERATIONS`` enabled.
+
+Then rebuild DPDK, specifying paths to the ITT header and library, which can
+be found in any VTune Amplifier distribution in the *include* and *lib*
+directories respectively:
+
+.. code-block:: console
+
+make EXTRA_CFLAGS=-I \
+ EXTRA_LDLIBS="-L -littnotify"
+
+Finally, to see wasted iterations in your performance analysis results,
+select the *"Analyze user tasks, events, and counters"* checkbox in the
+*"Analysis Type"* tab when configuring analysis via VTune Amplifier GUI.
+Alternatively, when running VTune Amplifier via command line, specify
+``-knob enable-user-tasks=true`` option.
+
+Collected regions of wasted iterations will be marked on VTune Amplifier's
+timeline as ITT tasks. These ITT tasks have predefined names, containing
+Ethernet device and RX queue identifiers.
+
+
 Profiling on ARM64
 --
 
diff --git a/lib/librte_ether/Makefile b/lib/librte_ether/Makefile
index db692ae..7224a11 100644
--- a/lib/librte_ether/Makefile
+++ b/lib/librte_ether/Makefile
@@ -46,6 +46,7 @@ LIBABIVER := 6
 SRCS-y += rte_ethdev.c
 SRCS-y += rte_flow.c
 SRCS-y += rte_tm.c
+SRCS-y += rte_ethdev_profile.c
 
 #
 # Export include files
diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
index a1b7447..2eba36e 100644
--- a/lib/librte_ether/rte_ethdev.c
+++ b/lib/librte_ether/rte_ethdev.c
@@ -67,6 +67,7 @@
 
 #include "rte_ether.h"
 #include "rte_ethdev.h"
+#include "rte_ethdev_profile.h"
 
 static const char *MZ_RTE_ETH_DEV_DATA = "rte_eth_dev_data";
 struct rte_eth_dev rte_eth_devices[RTE_MAX_ETHPORTS];
@@ -825,6 +826,9 @@ rte_eth_dev_configure(uint8_t port_id, uint16_t nb_rx_q, 
uint16_t nb_tx_q,
return diag;
}
 
+   /* See rte_ethdev_profile.h to find comments on code below. */
+   rte_eth_profile_rx_init(port_id, dev);
+
return 0;
 }
 
diff --git a/lib/librte_ether/rte_ethdev_profile.c 
b/lib/librte_ether/rte_ethdev_profile.c
new file mode 100644
index 000..bb83c8a
--- /dev/null
+++ b/lib/librte_ether/rte_ethdev_profile.c
@@ -0,0 +1,156 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright(c) 2010-2017 Intel Corporation. All rights reserved.
+ *   All rights reserved.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ * * Redistributions of source code must retain the above copyright
+ *   notice, this list of conditions and the 

Re: [dpdk-dev] [PATCH v2] net/ixgbe: add queue index check in filter

[Ferruh Yigit]

On 7/19/2017 4:34 AM, Wei Zhao wrote:
> Add queue index check when create filter rule, or
> filter with invalid queue id can be created successfully.
> 
> Signed-off-by: Wei Zhao 
Applied to dpdk-next-net/master, thanks.

[dpdk-dev] [PATCH v5] examples/vhost: introduce a new vhost-user-scsi sample application

[Changpeng Liu]

vhost-user protocol is common to many virtio devices, such as
virtio_net/virtio_scsi/virtio_blk. Since DPDK vhost library
removed the NET specific data structures, the vhost library
is common to other virtio devices, such as virtio-scsi.

Here we introduce a simple memory based block device that
can be presented to Guest VM through vhost-user-scsi-pci
controller. Similar with vhost-net, the sample application
will process the I/Os sent via virt rings.

Signed-off-by: Changpeng Liu 
---
 MAINTAINERS |   2 +
 doc/guides/sample_app_ug/index.rst  |   1 +
 doc/guides/sample_app_ug/vhost_scsi.rst | 115 +++
 examples/Makefile   |   2 +-
 examples/vhost_scsi/Makefile|  59 
 examples/vhost_scsi/scsi.c  | 539 
 examples/vhost_scsi/scsi_spec.h | 493 +
 examples/vhost_scsi/vhost_scsi.c| 474 
 examples/vhost_scsi/vhost_scsi.h| 108 +++
 9 files changed, 1792 insertions(+), 1 deletion(-)
 create mode 100644 doc/guides/sample_app_ug/vhost_scsi.rst
 create mode 100644 examples/vhost_scsi/Makefile
 create mode 100644 examples/vhost_scsi/scsi.c
 create mode 100644 examples/vhost_scsi/scsi_spec.h
 create mode 100644 examples/vhost_scsi/vhost_scsi.c
 create mode 100644 examples/vhost_scsi/vhost_scsi.h

diff --git a/MAINTAINERS b/MAINTAINERS
index 875bee0..c6ab9a7 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -447,6 +447,8 @@ F: lib/librte_vhost/
 F: doc/guides/prog_guide/vhost_lib.rst
 F: examples/vhost/
 F: doc/guides/sample_app_ug/vhost.rst
+F: examples/vhost_scsi/
+F: doc/guides/sample_app_ug/vhost_scsi.rst
 
 Vhost PMD
 M: Tetsuya Mukawa 
diff --git a/doc/guides/sample_app_ug/index.rst 
b/doc/guides/sample_app_ug/index.rst
index f9239e3..069d4f1 100644
--- a/doc/guides/sample_app_ug/index.rst
+++ b/doc/guides/sample_app_ug/index.rst
@@ -66,6 +66,7 @@ Sample Applications User Guides
 packet_ordering
 vmdq_dcb_forwarding
 vhost
+vhost_scsi
 netmap_compatibility
 ip_pipeline
 test_pipeline
diff --git a/doc/guides/sample_app_ug/vhost_scsi.rst 
b/doc/guides/sample_app_ug/vhost_scsi.rst
new file mode 100644
index 000..8be069e
--- /dev/null
+++ b/doc/guides/sample_app_ug/vhost_scsi.rst
@@ -0,0 +1,115 @@
+
+..  BSD LICENSE
+Copyright(c) 2010-2017 Intel Corporation. All rights reserved.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions
+are met:
+
+* Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
+* Redistributions in binary form must reproduce the above copyright
+notice, this list of conditions and the following disclaimer in
+the documentation and/or other materials provided with the
+distribution.
+* Neither the name of Intel Corporation nor the names of its
+contributors may be used to endorse or promote products derived
+from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+
+Vhost_scsi Sample Application
+=
+
+The vhost_scsi sample application implemented a simple SCSI block device,
+which used as the  backend of Qemu vhost-user-scsi device. Users can extend
+the exist example to use other type of block device(e.g. AIO) besides
+memory based block device. Similar with vhost-user-net device, the sample
+application used domain socket to communicate with Qemu, and the virtio
+ring was processed by vhost_scsi sample application.
+
+The sample application reuse lots codes from SPDK(Storage Performance
+Development Kit, https://github.com/spdk/spdk) vhost-user-scsi target,
+for DPDK vhost library used in storage area, user can take SPDK as
+reference as well.
+
+Testing steps
+-
+
+This section shows the steps how to start a VM with the block device as
+fast data path for critical application.
+
+Build
+~
+
+Follow the *Getting Started Guide for Linux* on generic info about
+environment setup and building DPDK f

[dpdk-dev] [PATCH v2] net/i40e: vf add/del mac error log issue

[Jeff Guo]

when i40e vf close, it would stop vf at first, if vf had been stopped,
that would result of duplicating to add/del mac address, then the failed
of executing admin queue command info would exposure. The patch fix that
by add vf stop status check and sync up the vf mac number when add/del.

Signed-off-by: Jeff Guo 
---
v2->v1: fix complie error issue and coding style issue
---
 drivers/net/i40e/i40e_ethdev.c|  6 +-
 drivers/net/i40e/i40e_ethdev_vf.c | 17 +++--
 2 files changed, 20 insertions(+), 3 deletions(-)

diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 97a73e1..4ed9619 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -2051,12 +2051,15 @@ static void
 i40e_dev_stop(struct rte_eth_dev *dev)
 {
struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
+   struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct i40e_vsi *main_vsi = pf->main_vsi;
struct i40e_mirror_rule *p_mirror;
struct rte_pci_device *pci_dev = RTE_ETH_DEV_TO_PCI(dev);
struct rte_intr_handle *intr_handle = &pci_dev->intr_handle;
int i;
 
+   if (hw->adapter_stopped == 1)
+   return;
/* Disable all queues */
i40e_dev_switch_queues(pf, FALSE);
 
@@ -2101,6 +2104,8 @@ i40e_dev_stop(struct rte_eth_dev *dev)
 
/* reset hierarchy commit */
pf->tm_conf.committed = false;
+
+   hw->adapter_stopped = 1;
 }
 
 static void
@@ -2116,7 +2121,6 @@ i40e_dev_close(struct rte_eth_dev *dev)
PMD_INIT_FUNC_TRACE();
 
i40e_dev_stop(dev);
-   hw->adapter_stopped = 1;
i40e_dev_free_queues(dev);
 
/* Disable interrupt */
diff --git a/drivers/net/i40e/i40e_ethdev_vf.c 
b/drivers/net/i40e/i40e_ethdev_vf.c
index dc6c794..a21cfdb 100644
--- a/drivers/net/i40e/i40e_ethdev_vf.c
+++ b/drivers/net/i40e/i40e_ethdev_vf.c
@@ -906,6 +906,8 @@ i40evf_add_mac_addr(struct rte_eth_dev *dev,
if (err)
PMD_DRV_LOG(ERR, "fail to execute command "
"OP_ADD_ETHER_ADDRESS");
+   else
+   vf->vsi.mac_num++;
 
return err;
 }
@@ -944,6 +946,8 @@ i40evf_del_mac_addr_by_addr(struct rte_eth_dev *dev,
if (err)
PMD_DRV_LOG(ERR, "fail to execute command "
"OP_DEL_ETHER_ADDRESS");
+   else
+   vf->vsi.mac_num--;
return;
 }
 
@@ -2058,10 +2062,16 @@ i40evf_add_del_all_mac_addr(struct rte_eth_dev *dev, 
bool add)
args.out_buffer = vf->aq_resp;
args.out_size = I40E_AQ_BUF_SZ;
err = i40evf_execute_vf_cmd(dev, &args);
-   if (err)
+   if (err) {
PMD_DRV_LOG(ERR, "fail to execute command %s",
add ? "OP_ADD_ETHER_ADDRESS" :
"OP_DEL_ETHER_ADDRESS");
+   } else {
+   if (add)
+   vf->vsi.mac_num++;
+   else
+   vf->vsi.mac_num--;
+   }
rte_free(list);
begin = next_begin;
} while (begin < I40E_NUM_MACADDR_MAX);
@@ -2140,9 +2150,12 @@ i40evf_dev_stop(struct rte_eth_dev *dev)
 {
struct rte_pci_device *pci_dev = RTE_ETH_DEV_TO_PCI(dev);
struct rte_intr_handle *intr_handle = &pci_dev->intr_handle;
+   struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev);
 
PMD_INIT_FUNC_TRACE();
 
+   if (hw->adapter_stopped == 1)
+   return;
i40evf_stop_queues(dev);
i40evf_disable_queues_intr(dev);
i40e_dev_clear_queues(dev);
@@ -2155,6 +2168,7 @@ i40evf_dev_stop(struct rte_eth_dev *dev)
}
/* remove all mac addrs */
i40evf_add_del_all_mac_addr(dev, FALSE);
+   hw->adapter_stopped = 1;
 
 }
 
@@ -2342,7 +2356,6 @@ i40evf_dev_close(struct rte_eth_dev *dev)
struct rte_intr_handle *intr_handle = &pci_dev->intr_handle;
 
i40evf_dev_stop(dev);
-   hw->adapter_stopped = 1;
i40e_dev_free_queues(dev);
i40evf_reset_vf(hw);
i40e_shutdown_adminq(hw);
-- 
2.7.4

[dpdk-dev] [PATCH] all: refactor coding style

[Tiwei Bie]

Remove the unwanted spaces before `;' across DPDK source code
by below one-liner with some minor manual refinements.

find . -name '*.[ch]' | xargs sed -i 's/\([^;(]\) \+;/\1;/g'

The fixes for cmdline library are skipped, because it has a
different coding style. It deserves a separate cleanup if
necessary. The fixes for drivers' base code are also skipped
to keep the base code intact.

Signed-off-by: Tiwei Bie 
---
 app/test-pmd/testpmd.h |  4 ++--
 drivers/crypto/qat/qat_adf/icp_qat_fw.h|  2 +-
 drivers/event/dpaa2/dpaa2_eventdev.c   |  2 +-
 drivers/mempool/dpaa2/dpaa2_hw_mempool.c   |  2 +-
 drivers/net/bnx2x/bnx2x.c  |  3 ++-
 drivers/net/bnx2x/elink.h  |  2 +-
 drivers/net/e1000/igb_pf.c |  2 +-
 drivers/net/ena/ena_ethdev.c   |  4 ++--
 drivers/net/qede/qede_ethdev.c |  2 +-
 drivers/net/vhost/rte_eth_vhost.c  |  2 +-
 drivers/net/virtio/virtio_rxtx.c   |  4 ++--
 drivers/net/xenvirt/rte_eth_xenvirt.c  |  4 ++--
 drivers/net/xenvirt/rte_xen_lib.c  |  2 +-
 drivers/net/xenvirt/virtqueue.h|  2 +-
 examples/ip_pipeline/cpu_core_map.c|  4 ++--
 examples/multi_process/l2fwd_fork/main.c   |  2 +-
 examples/netmap_compat/lib/compat_netmap.c |  2 +-
 examples/performance-thread/l3fwd-thread/main.c|  2 +-
 examples/qos_sched/app_thread.c|  2 +-
 examples/quota_watermark/qw/main.c |  2 +-
 examples/vhost_xen/xenstore_parse.c|  3 +--
 lib/librte_distributor/rte_distributor.c   | 12 +-
 lib/librte_eal/linuxapp/eal/eal_memory.c   |  2 +-
 lib/librte_eal/linuxapp/eal/eal_xen_memory.c   |  2 +-
 lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c |  4 ++--
 .../linuxapp/kni/ethtool/ixgbe/ixgbe_main.c|  2 +-
 .../linuxapp/kni/ethtool/ixgbe/kcompat.c   |  2 +-
 lib/librte_ether/rte_ethdev.c  |  2 +-
 lib/librte_sched/rte_approx.c  |  8 +++
 lib/librte_sched/rte_bitmap.h  |  3 ++-
 test/test/test_cryptodev.c |  2 +-
 test/test/test_cryptodev_perf.c| 26 +++---
 test/test/test_eventdev_sw.c   |  2 +-
 test/test/test_malloc.c|  4 ++--
 test/test/test_memory.c|  2 +-
 test/test/test_mempool.c   |  2 +-
 test/test/test_ring.c  |  6 ++---
 test/test/test_table_acl.c |  2 +-
 test/test/test_table_pipeline.c|  2 +-
 39 files changed, 69 insertions(+), 70 deletions(-)

diff --git a/app/test-pmd/testpmd.h b/app/test-pmd/testpmd.h
index c9d7739..8f88d70 100644
--- a/app/test-pmd/testpmd.h
+++ b/app/test-pmd/testpmd.h
@@ -118,8 +118,8 @@ struct fwd_stream {
unsigned int rx_packets;  /**< received packets */
unsigned int tx_packets;  /**< received packets transmitted */
unsigned int fwd_dropped; /**< received packets not forwarded */
-   unsigned int rx_bad_ip_csum ; /**< received packets has bad ip checksum 
*/
-   unsigned int rx_bad_l4_csum ; /**< received packets has bad l4 checksum 
*/
+   unsigned int rx_bad_ip_csum; /**< received packets has bad ip checksum 
*/
+   unsigned int rx_bad_l4_csum; /**< received packets has bad l4 checksum 
*/
 #ifdef RTE_TEST_PMD_RECORD_CORE_CYCLES
uint64_t core_cycles; /**< used for RX and TX processing */
 #endif
diff --git a/drivers/crypto/qat/qat_adf/icp_qat_fw.h 
b/drivers/crypto/qat/qat_adf/icp_qat_fw.h
index 5de34d5..c80989b 100644
--- a/drivers/crypto/qat/qat_adf/icp_qat_fw.h
+++ b/drivers/crypto/qat/qat_adf/icp_qat_fw.h
@@ -51,7 +51,7 @@
 
 #define QAT_FIELD_SET(flags, val, bitpos, mask) \
 { (flags) = (((flags) & (~((mask) << (bitpos | \
-   (((val) & (mask)) << (bitpos))) ; }
+   (((val) & (mask)) << (bitpos))); }
 
 #define QAT_FIELD_GET(flags, bitpos, mask) \
(((flags) >> (bitpos)) & (mask))
diff --git a/drivers/event/dpaa2/dpaa2_eventdev.c 
b/drivers/event/dpaa2/dpaa2_eventdev.c
index ed57376..89bc1ce 100644
--- a/drivers/event/dpaa2/dpaa2_eventdev.c
+++ b/drivers/event/dpaa2/dpaa2_eventdev.c
@@ -587,7 +587,7 @@ dpaa2_eventdev_setup_dpci(struct dpaa2_dpci_dev *dpci_dev,
dpci_dev->queue[DPAA2_EVENT_DPCI_ATOMIC_QUEUE].cb =
dpaa2_eventdev_process_atomic;
 
-   for (i = 0 ; i < DPAA2_EVENT_DPCI_MAX_QUEUES; i++) {
+   for (i = 0; i < DPAA2_EVENT_DPCI_MAX_QUEUES; i++) {
rx_queue_cfg.user_ctx = (uint64_t)(&dpci_dev->queue[i]);
ret = dpci_set_rx_queue(&dpci_dev->dpci,
CMD_PRI_LOW,
diff --git a/drivers/mempool/dpaa2/dpaa2_hw_mempool.c

Re: [dpdk-dev] [PATCH] net/ixgbe: fix NIC 82599ES type check error

[Ferruh Yigit]

On 7/19/2017 4:33 AM, Wei Zhao wrote:
> This NIC type check is specific for 82599ES.
> 
> Fixes: 16f534e508d ("net/ixgbe: add support 82599ES SCTP packet drop action")
> 
> Signed-off-by: Wei Zhao 

Squashed into relevant commit in next-net, thanks.

Re: [dpdk-dev] [PATCH v2] net/ixgbe: fix Rx/Tx queue interrupt for x550 devices

[Dai, Wei]

> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Qiming Yang
> Sent: Tuesday, July 18, 2017 10:29 AM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Lu, Wenzhuo ; Yang, Qiming
> 
> Subject: [dpdk-dev] [PATCH v2] net/ixgbe: fix Rx/Tx queue interrupt for x550
> devices
> 
> x550 devices not do interrupt vector mapping before enable Rx/Tx queue
> interrupt, makes interrupt mode can't work neither with igb_uio or VFIO.
> 
> Fixes: d2e72774e58c ("ixgbe/base: support X550")
> 
> Signed-off-by: Qiming Yang 
Have looked through datasheet of 82599, X540 and X550.
Acked-by: Wei Dai 

> ---
> v2 changes:
> * fixed other cause interrupt vector map and typo issue
> ---
>  drivers/net/ixgbe/ixgbe_ethdev.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c
> b/drivers/net/ixgbe/ixgbe_ethdev.c
> index 9b06ac1..0caafd5 100644
> --- a/drivers/net/ixgbe/ixgbe_ethdev.c
> +++ b/drivers/net/ixgbe/ixgbe_ethdev.c
> @@ -5611,7 +5611,8 @@ ixgbe_set_ivar_map(struct ixgbe_hw *hw, int8_t
> direction,
>   tmp |= (msix_vector << (8 * (queue & 0x3)));
>   IXGBE_WRITE_REG(hw, IXGBE_IVAR(idx), tmp);
>   } else if ((hw->mac.type == ixgbe_mac_82599EB) ||
> - (hw->mac.type == ixgbe_mac_X540)) {
> + (hw->mac.type == ixgbe_mac_X540) ||
> + (hw->mac.type == ixgbe_mac_X550)) {
>   if (direction == -1) {
>   /* other causes */
>   idx = ((queue & 1) * 8);
> @@ -5719,6 +5720,7 @@ ixgbe_configure_msix(struct rte_eth_dev *dev)
>   break;
>   case ixgbe_mac_82599EB:
>   case ixgbe_mac_X540:
> + case ixgbe_mac_X550:
>   ixgbe_set_ivar_map(hw, -1, 1, IXGBE_MISC_VEC_ID);
>   break;
>   default:
> --
> 2.7.4

Re: [dpdk-dev] [PATCH] all: refactor coding style

[Van Haaren, Harry]

> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Tiwei Bie
> Sent: Wednesday, July 19, 2017 10:07 AM
> To: dev@dpdk.org
> Cc: tho...@monjalon.net
> Subject: [dpdk-dev] [PATCH] all: refactor coding style
> 
> Remove the unwanted spaces before `;' across DPDK source code
> by below one-liner with some minor manual refinements.
> 
> find . -name '*.[ch]' | xargs sed -i 's/\([^;(]\) \+;/\1;/g'
> 
> The fixes for cmdline library are skipped, because it has a
> different coding style. It deserves a separate cleanup if
> necessary. The fixes for drivers' base code are also skipped
> to keep the base code intact.
> 
> Signed-off-by: Tiwei Bie 

Hi Tiwei,

Although the idea and motivation for code-cleanup are good, performing
large cleanup across a code-base is not a good solution. The reason that
these types of cleanups (or even re-formatting the entire codebase) are not
performed often is that it "invalidates" any currently-in-progress patch-sets.
As a result, more work is required from many contributors to rebase useful
features due to across-the-board white-space cleanups.

Just expressing concern that we need to think carefully about the impacts
of such a patch.

Regards, -Harry

Re: [dpdk-dev] [dpdk-stable] [PATCH v2] net/ixgbe: fix Rx/Tx queue interrupt for x550 devices

[Ferruh Yigit]

On 7/19/2017 10:22 AM, Dai, Wei wrote:
> 
> 
>> -Original Message-
>> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Qiming Yang
>> Sent: Tuesday, July 18, 2017 10:29 AM
>> To: dev@dpdk.org
>> Cc: sta...@dpdk.org; Lu, Wenzhuo ; Yang, Qiming
>> 
>> Subject: [dpdk-dev] [PATCH v2] net/ixgbe: fix Rx/Tx queue interrupt for x550
>> devices
>>
>> x550 devices not do interrupt vector mapping before enable Rx/Tx queue
>> interrupt, makes interrupt mode can't work neither with igb_uio or VFIO.
>>
>> Fixes: d2e72774e58c ("ixgbe/base: support X550")
>>
>> Signed-off-by: Qiming Yang 

> Have looked through datasheet of 82599, X540 and X550.
> Acked-by: Wei Dai 

Applied to dpdk-next-net/master, thanks.

Re: [dpdk-dev] [PATCH] all: refactor coding style

[Tiwei Bie]

On Wed, Jul 19, 2017 at 05:24:38PM +0800, Van Haaren, Harry wrote:
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Tiwei Bie
> > Sent: Wednesday, July 19, 2017 10:07 AM
> > To: dev@dpdk.org
> > Cc: tho...@monjalon.net
> > Subject: [dpdk-dev] [PATCH] all: refactor coding style
> > 
> > Remove the unwanted spaces before `;' across DPDK source code
> > by below one-liner with some minor manual refinements.
> > 
> > find . -name '*.[ch]' | xargs sed -i 's/\([^;(]\) \+;/\1;/g'
> > 
> > The fixes for cmdline library are skipped, because it has a
> > different coding style. It deserves a separate cleanup if
> > necessary. The fixes for drivers' base code are also skipped
> > to keep the base code intact.
> > 
> > Signed-off-by: Tiwei Bie 
> 
> Hi Tiwei,
> 
> Although the idea and motivation for code-cleanup are good, performing
> large cleanup across a code-base is not a good solution. The reason that
> these types of cleanups (or even re-formatting the entire codebase) are not
> performed often is that it "invalidates" any currently-in-progress patch-sets.
> As a result, more work is required from many contributors to rebase useful
> features due to across-the-board white-space cleanups.
> 
> Just expressing concern that we need to think carefully about the impacts
> of such a patch.
> 

Yeah, I agree. Such patch may cause many conflicts. But this patch
is almost generated automatically, that is to say, it's a quick work.
And it's more like some fixes (for the bad coding style) rather than
silly re-formatting done by `indent'. So I just want to share it with
the community, and see the potential feedbacks. Thank you for your
comments! :)

Best regards,
Tiwei Bie

Re: [dpdk-dev] [PATCH v2] drivers/net: add support for IF-MIB and EtherLike-MIB for e1000

[Ferruh Yigit]

On 6/26/2017 10:42 AM, Radu Nicolau wrote:
> From: Michal Jastrzebski 
> 
> If-MIB xstats:
> ifNumber
> ifIndex
> ifType
> ifMtu
> ifSpeed
> ifPhysAddress
> ifOperStatus
> ifLastChange
> ifHighSpeed
> ifConnectorPresent
> ifCounterDiscontinuityTime
> 
> EtherLike-MIB xstats:
> dot3PauseOperMode
> dot3StatsDuplexStatus
> dot3StatsRateControlAbility
> dot3StatsRateControlStatus
> dot3ControlFunctionsSupported
> 
> -updated in v2: coding style
> 
> Signed-off-by: Piotr Azarewicz 
> Signed-off-by: Michal Jastrzebski 
> Signed-off-by: Radu Nicolau 

This patchset won't able to make for this release. There is already a
new RFC [1] for next release.

[1]
http://dpdk.org/dev/patchwork/patch/27032/

Re: [dpdk-dev] [PATCH] all: refactor coding style

[Trahe, Fiona]

> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Tiwei Bie
> Sent: Wednesday, July 19, 2017 10:07 AM
> To: dev@dpdk.org
> Cc: tho...@monjalon.net
> Subject: [dpdk-dev] [PATCH] all: refactor coding style
> 
> Remove the unwanted spaces before `;' across DPDK source code
> by below one-liner with some minor manual refinements.
> 
> find . -name '*.[ch]' | xargs sed -i 's/\([^;(]\) \+;/\1;/g'
> 
> The fixes for cmdline library are skipped, because it has a
> different coding style. It deserves a separate cleanup if
> necessary. The fixes for drivers' base code are also skipped
> to keep the base code intact.
> 
> Signed-off-by: Tiwei Bie 
Acked-by: Fiona Trahe 

Re: [dpdk-dev] Occasional instability in RSS Hashes/Queues from X540 NIC

[Yuanhan Liu]

On Wed, Jul 19, 2017 at 08:13:02AM +, Li, Xiaoyun wrote:
> Hi, Yuanhan !
> I did the same tests on DPDK 16.11 this morning and cannot reproduce the 
> problem on 16.11. It seems that the problem has been fixed on version 16.11.

Good to know.

--yliu

> So could you reproduce the problem on 16.11 LTS release? I think maybe it has 
> been fixed already.
> 
> 
> Best Regards,
> Xiaoyun Li
> 
> 
> 
> -Original Message-
> From: Yuanhan Liu [mailto:y...@fridaylinux.org] 
> Sent: Tuesday, July 18, 2017 22:01
> To: Matt Laswell 
> Cc: Yang, Qiming ; dev@dpdk.org; Li, Xiaoyun 
> 
> Subject: Re: [dpdk-dev] Occasional instability in RSS Hashes/Queues from X540 
> NIC
> 
> On Tue, Jul 18, 2017 at 08:43:42AM -0500, Matt Laswell wrote:
> > Hi Qiming,
> > 
> > That's fantastic news.  Thank you very much for taking the time to 
> > figure the issue out.
> > 
> > Would it be possible to backport the fix to the 16.11 LTS release?   This
> > kind of problem seems tailor-made for LTS.
> 
> Agreed. You might want to find the fix commit with git bisect.
> 
>   --yliu

Re: [dpdk-dev] [dpdk-stable] [PATCH] app/crypto-perf: stop crypto devices after test is finished

[De Lara Guarch, Pablo]

> -Original Message-
> From: stable [mailto:stable-boun...@dpdk.org] On Behalf Of Kirill
> Rybalchenko
> Sent: Wednesday, June 14, 2017 9:44 AM
> To: Doherty, Declan 
> Cc: dev@dpdk.org; Rybalchenko, Kirill ;
> sta...@dpdk.org
> Subject: [dpdk-stable] [PATCH] app/crypto-perf: stop crypto devices after
> test is finished
> 
> Call of rte_cryptodev_stop() function from test destructors is added.
> 
> Fixes: f8be1786b1b8 ("app/crypto-perf: introduce performance test
> application")
> Cc: sta...@dpdk.org
> 
> Signed-off-by: Kirill Rybalchenko 

Acked-by: Pablo de Lara 

Re: [dpdk-dev] [dpdk-stable] [PATCH] app/crypto-perf: stop crypto devices after test is finished

[De Lara Guarch, Pablo]

> -Original Message-
> From: stable [mailto:stable-boun...@dpdk.org] On Behalf Of De Lara
> Guarch, Pablo
> Sent: Wednesday, July 19, 2017 11:58 AM
> To: Rybalchenko, Kirill ; Doherty, Declan
> 
> Cc: dev@dpdk.org; Rybalchenko, Kirill ;
> sta...@dpdk.org
> Subject: Re: [dpdk-stable] [PATCH] app/crypto-perf: stop crypto devices
> after test is finished
> 
> 
> 
> > -Original Message-
> > From: stable [mailto:stable-boun...@dpdk.org] On Behalf Of Kirill
> > Rybalchenko
> > Sent: Wednesday, June 14, 2017 9:44 AM
> > To: Doherty, Declan 
> > Cc: dev@dpdk.org; Rybalchenko, Kirill ;
> > sta...@dpdk.org
> > Subject: [dpdk-stable] [PATCH] app/crypto-perf: stop crypto devices
> > after test is finished
> >
> > Call of rte_cryptodev_stop() function from test destructors is added.
> >
> > Fixes: f8be1786b1b8 ("app/crypto-perf: introduce performance test
> > application")
> > Cc: sta...@dpdk.org
> >
> > Signed-off-by: Kirill Rybalchenko 
> 
> Acked-by: Pablo de Lara 

Applied to dpdk-next-crypto.
Thanks,

Pablo

Re: [dpdk-dev] [PATCH] net/virtio: fix fail to configure rxq interrupt

[Yuanhan Liu]

On Wed, Jul 19, 2017 at 11:18:23AM +0800, Jiayu Hu wrote:
> When use rte_eth_dev_configure() to enable rx queue interrupt for virtio
> devices, virtio_configure_intr() isn't called to set up the interrupt
> environment, which causes rx queue interrupt setup failed. This patch is
> to fix this issue.

Hmm, how was this supposed to work in the begining when this feature
was introduced? Jianfeng?

--yliu


> 
> Fixes: 26b683b4f7d0 ("net/virtio: setup Rx queue interrupts")
> Cc: sta...@dpdk.org
> 
> Signed-off-by: Jiayu Hu 
> ---
>  drivers/net/virtio/virtio_ethdev.c | 7 +++
>  1 file changed, 7 insertions(+)
> 
> diff --git a/drivers/net/virtio/virtio_ethdev.c 
> b/drivers/net/virtio/virtio_ethdev.c
> index 00a3122..66656ed 100644
> --- a/drivers/net/virtio/virtio_ethdev.c
> +++ b/drivers/net/virtio/virtio_ethdev.c
> @@ -1688,6 +1688,13 @@ virtio_dev_configure(struct rte_eth_dev *dev)
>   return -ENOTSUP;
>   }
>  
> + if (dev->data->dev_conf.intr_conf.rxq) {
> + if (virtio_configure_intr(dev) < 0) {
> + PMD_DRV_LOG(ERR, "failed to configure interrupt");
> + return -ENOTSUP;
> + }
> + }
> +
>   if (dev->data->dev_flags & RTE_ETH_DEV_INTR_LSC)
>   /* Enable vector (0) for Link State Intrerrupt */
>   if (VTPCI_OPS(hw)->set_config_irq(hw, 0) ==
> -- 
> 2.7.4

Re: [dpdk-dev] [PATCH v2] net/i40e: vf add/del mac error log issue

[Xing, Beilei]

> -Original Message-
> From: Guo, Jia
> Sent: Wednesday, July 19, 2017 5:01 PM
> To: Xing, Beilei ; Wu, Jingjing 
> Cc: dev@dpdk.org; Guo, Jia 
> Subject: [PATCH v2] net/i40e: vf add/del mac error log issue
> 
> when i40e vf close, it would stop vf at first, if vf had been stopped, that
> would result of duplicating to add/del mac address, then the failed of
> executing admin queue command info would exposure. The patch fix that by
> add vf stop status check and sync up the vf mac number when add/del.
> 
> Signed-off-by: Jeff Guo 

Acked-by: Beilei Xing 

Re: [dpdk-dev] [PATCH v2] net/i40e: vf add/del mac error log issue

[Ferruh Yigit]

On 7/19/2017 12:11 PM, Xing, Beilei wrote:
> 
> 
>> -Original Message-
>> From: Guo, Jia
>> Sent: Wednesday, July 19, 2017 5:01 PM
>> To: Xing, Beilei ; Wu, Jingjing 
>> 
>> Cc: dev@dpdk.org; Guo, Jia 
>> Subject: [PATCH v2] net/i40e: vf add/del mac error log issue
>>
>> when i40e vf close, it would stop vf at first, if vf had been stopped, that
>> would result of duplicating to add/del mac address, then the failed of
>> executing admin queue command info would exposure. The patch fix that by
>> add vf stop status check and sync up the vf mac number when add/del.
>>
>> Signed-off-by: Jeff Guo 
> 
> Acked-by: Beilei Xing 

Applied to dpdk-next-net/master, thanks.

Re: [dpdk-dev] [PATCH v2] crypto/dpaa2_sec: fix the return of supported API

[Yuanhan Liu]

On Tue, Jul 18, 2017 at 05:32:44PM +, Hemant Agrawal wrote:
> > > > > Subject: [PATCH v2] crypto/dpaa2_sec: fix the return of supported
> > > > > API
> > > > >
> > > > > call to dpaa2_sec_dev_configure() is made mandatory, but
> > > > > dpaa2_sec_pmd returns a ENOTSUP which results in device not
> > > > > getting
> > > configured.
> > > > >
> > > > > dpaa2_sec PMD does not need any further configuration to be done
> > > > > in dpaa2_sec_dev_configure, hence returning 0
> > > > >
> > > > > Fixes: e5cbdfc53765 ("crypto/dpaa2_sec: add basic operations")
> > > > >
> > > > > Cc: sta...@dpdk.org
> > > > >
> > > > > Signed-off-by: Akhil Goyal 
> > > >
> > > > Looks ok to me, but this is only applicable in the stable branch, so
> > > > no need to send it to dev@dpdk.org.
> > >
> > > Why? We already have such fix in upstream? Normally, we just pick
> > > upstream commits (but not patches: the emails) to stable release.
> > 
> > It looks like this fix was included in
> > 13273250eec5 ("crypto/dpaa2_sec: support AES-GCM and CTR").
> > Unfortunately, this patch should have been split into two different patches.
> > Since this has already been merged, I think our only way to integrate this 
> > In
> > 17.05.1 is by getting it separately.
> 
> In general, there may be other incidents, where a patch is only applicable 
> for the stable tree. It may not be applicable for upstream tree due to 
> architecture changes or other reasons.
> How do you want to handle such patches? 
> 
> e.g. in OVS, we can do it by marking the patch header with "[branch-2.6]"


Yes, you are right, it might happen. Then you need cook a standalone
patch and send it to stable ml only. Since I don't usually pick stable
patches directly from stable ML, you probably need add some marks in
the commit log. Something like "this is for stable tree only and add
a bit explanation".

Normally, every time I saw a patch sent only to stable ML I will ask
the same question I have asked in this email. But I could just miss
it. So you are suggested to do above.

For this case, just as Pablo said, the patch should be split in the
beginning, then only the (small) bug fixing patch will be picked to a
specific stable release. And since it already happened, you could just
send it to stable ML only, and better, with me cc-ed.

Thanks.

--yliu

[dpdk-dev] [PATCH] examples/l2fwd-crypto: add option --[no-]mac-updating

[Kuba Kozak]

This patch adds a new option to enable/disable the
MAC addresses updating done at forwarding time: --[no-]mac-updating

By default, MAC address updating remains enabled, to keep consistency
with previous usage.

Signed-off-by: Kuba Kozak 
---
 doc/guides/sample_app_ug/l2_forward_crypto.rst |  7 ++--
 examples/l2fwd-crypto/main.c   | 50 ++
 2 files changed, 48 insertions(+), 9 deletions(-)

diff --git a/doc/guides/sample_app_ug/l2_forward_crypto.rst 
b/doc/guides/sample_app_ug/l2_forward_crypto.rst
index 2a61af7..91ce278 100644
--- a/doc/guides/sample_app_ug/l2_forward_crypto.rst
+++ b/doc/guides/sample_app_ug/l2_forward_crypto.rst
@@ -46,7 +46,7 @@ for each packet that is received on a RX_PORT and performs L2 
forwarding.
 The destination port is the adjacent port from the enabled portmask, that is,
 if the first four ports are enabled (portmask 0xf),
 ports 0 and 1 forward into each other, and ports 2 and 3 forward into each 
other.
-Also, the MAC addresses are affected as follows:
+Also, if MAC addresses updating is enabled, the MAC addresses are affected as 
follows:
 
 *   The source MAC address is replaced by the TX_PORT MAC address
 
@@ -90,7 +90,8 @@ The application requires a number of command line options:
 [--auth_algo ALGO] [--auth_op GENERATE/VERIFY] [--auth_key KEY] /
 [--auth_key_random_size SIZE] [--auth_iv IV] [--auth_iv_random_size SIZE] /
 [--aad AAD] [--aad_random_size SIZE] /
-[--digest size SIZE] [--sessionless] [--cryptodev_mask MASK]
+[--digest size SIZE] [--sessionless] [--cryptodev_mask MASK] /
+[--mac-updating] [--no-mac-updating]
 
 where,
 
@@ -191,6 +192,8 @@ where,
 
 (default is all cryptodevs).
 
+*   [no-]mac-updating: Enable or disable MAC addresses updating (enabled by 
default).
+
 
 The application requires that crypto devices capable of performing
 the specified crypto operation are available on application initialization.
diff --git a/examples/l2fwd-crypto/main.c b/examples/l2fwd-crypto/main.c
index 71cb133..a347ad6 100644
--- a/examples/l2fwd-crypto/main.c
+++ b/examples/l2fwd-crypto/main.c
@@ -193,6 +193,8 @@ struct l2fwd_crypto_options {
char string_type[MAX_STR_LEN];
 
uint64_t cryptodev_mask;
+
+   unsigned int mac_updating;
 };
 
 /** l2fwd crypto lcore params */
@@ -608,21 +610,31 @@ l2fwd_send_packet(struct rte_mbuf *m, uint8_t port)
 }
 
 static void
-l2fwd_simple_forward(struct rte_mbuf *m, unsigned portid)
+l2fwd_mac_updating(struct rte_mbuf *m, unsigned int dest_portid)
 {
struct ether_hdr *eth;
void *tmp;
-   unsigned dst_port;
 
-   dst_port = l2fwd_dst_ports[portid];
eth = rte_pktmbuf_mtod(m, struct ether_hdr *);
 
/* 02:00:00:00:00:xx */
tmp = ð->d_addr.addr_bytes[0];
-   *((uint64_t *)tmp) = 0x0002 + ((uint64_t)dst_port << 40);
+   *((uint64_t *)tmp) = 0x0002 + ((uint64_t)dest_portid << 40);
 
/* src addr */
-   ether_addr_copy(&l2fwd_ports_eth_addr[dst_port], ð->s_addr);
+   ether_addr_copy(&l2fwd_ports_eth_addr[dest_portid], ð->s_addr);
+}
+
+static void
+l2fwd_simple_forward(struct rte_mbuf *m, unsigned int portid,
+   struct l2fwd_crypto_options *options)
+{
+   unsigned int dst_port;
+
+   dst_port = l2fwd_dst_ports[portid];
+
+   if (options->mac_updating)
+   l2fwd_mac_updating(m, dst_port);
 
l2fwd_send_packet(m, (uint8_t) dst_port);
 }
@@ -920,7 +932,8 @@ l2fwd_main_loop(struct l2fwd_crypto_options *options)
m = ops_burst[j]->sym->m_src;
 
rte_crypto_op_free(ops_burst[j]);
-   l2fwd_simple_forward(m, portid);
+   l2fwd_simple_forward(m, portid,
+   options);
}
} while (nb_rx == MAX_PKT_BURST);
}
@@ -975,7 +988,12 @@ l2fwd_crypto_usage(const char *prgname)
"  --digest_size SIZE: size of digest to be 
generated/verified\n"
 
"  --sessionless\n"
-   "  --cryptodev_mask MASK: hexadecimal bitmask of crypto devices 
to configure\n",
+   "  --cryptodev_mask MASK: hexadecimal bitmask of crypto devices 
to configure\n"
+
+   "  --[no-]mac-updating: Enable or disable MAC addresses 
updating (enabled by default)\n"
+   "  When enabled:\n"
+   "   - The source MAC address is replaced by the TX port MAC 
address\n"
+   "   - The destination MAC address is replaced by 
02:00:00:00:00:TX_PORT_ID\n",
   prgname);
 }
 
@@ -1322,6 +1340,16 @@ l2fwd_crypto_parse_args_long_options(struct 
l2fwd_crypto_options *options,
else if (strcmp(lgopts[option_index].name, "cryptodev_mask") == 0)
return parse_cryptodev_mask(options, optar

Re: [dpdk-dev] [PATCH v2] eal/armv8: fix poly64/128 compile issue in old GCC(<4.9.0)

[Thomas Monjalon]

19/07/2017 11:21, Herbert Guan:
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> > 13/07/2017 05:16, Herbert Guan:
> > > --- a/lib/librte_eal/common/include/arch/arm/rte_vect.h
> > > +++ b/lib/librte_eal/common/include/arch/arm/rte_vect.h
> > > +#if (GCC_VERSION < 40900)
> > > +typedef uint64_t poly64_t;
> > > +typedef uint64x2_t poly64x2_t;
> > > +typedef uint8_t poly128_t __attribute__((vector_size(16),
> > > aligned(16)));
> > > +#endif
> > 
> > I think a better fix would be to switch to DPDK types
> > like rte_v128u8_t.
> 
> Thanks a lot for your review and comment.
> But I have some concern in this approach.  "poly128_t" is for
> ARM64 platform only and in fact it's more likely that rte_v128u8_t
> (generic DPDK data type) could be defined from poly128_t
> (ARM data type) which seems more reasonable.

How poly128_t is different from rte_v128u8_t?
You are defining poly128_t as (vector_size(16),aligned(16))
and rte_v128u8_t is exactly that.
Is it interpreted differently with newer compilers?
In that case, you could at least fallback on rte_v128u8_t.

[dpdk-dev] [PATCH] crypto/qat: fix session initialization

[Pablo de Lara]

When creating a session, if there is a failure when
setting some of the parameters, QAT was not propagating
the error to the session initialization function.
Therefore, it was reporting a success, when it should
be report a failure.

Fixes: b3bbd9e5f265 ("cryptodev: support device independent sessions")

Signed-off-by: Pablo de Lara 
---
 drivers/crypto/qat/qat_crypto.c | 83 +
 drivers/crypto/qat/qat_crypto.h | 11 --
 2 files changed, 50 insertions(+), 44 deletions(-)

diff --git a/drivers/crypto/qat/qat_crypto.c b/drivers/crypto/qat/qat_crypto.c
index ec481ae..d92de35 100644
--- a/drivers/crypto/qat/qat_crypto.c
+++ b/drivers/crypto/qat/qat_crypto.c
@@ -295,11 +295,12 @@ qat_get_cipher_xform(struct rte_crypto_sym_xform *xform)
 
return NULL;
 }
-void *
+
+int
 qat_crypto_sym_configure_session_cipher(struct rte_cryptodev *dev,
-   struct rte_crypto_sym_xform *xform, void *session_private)
+   struct rte_crypto_sym_xform *xform,
+   struct qat_session *session)
 {
-   struct qat_session *session = session_private;
struct qat_pmd_private *internals = dev->data->dev_private;
struct rte_crypto_cipher_xform *cipher_xform = NULL;
 
@@ -440,14 +441,14 @@ qat_crypto_sym_configure_session_cipher(struct 
rte_cryptodev *dev,
cipher_xform->key.length))
goto error_out;
 
-   return session;
+   return 0;
 
 error_out:
if (session->bpi_ctx) {
bpi_cipher_ctx_free(session->bpi_ctx);
session->bpi_ctx = NULL;
}
-   return NULL;
+   return -1;
 }
 
 int
@@ -498,36 +499,44 @@ qat_crypto_set_session_parameters(struct rte_cryptodev 
*dev,
qat_cmd_id = qat_get_cmd_id(xform);
if (qat_cmd_id < 0 || qat_cmd_id >= ICP_QAT_FW_LA_CMD_DELIMITER) {
PMD_DRV_LOG(ERR, "Unsupported xform chain requested");
-   goto error_out;
+   return -1;
}
session->qat_cmd = (enum icp_qat_fw_la_cmd_id)qat_cmd_id;
switch (session->qat_cmd) {
case ICP_QAT_FW_LA_CMD_CIPHER:
-   session = qat_crypto_sym_configure_session_cipher(dev, xform, session);
+   if (qat_crypto_sym_configure_session_cipher(dev, xform, 
session) < 0)
+   return -1;
break;
case ICP_QAT_FW_LA_CMD_AUTH:
-   session = qat_crypto_sym_configure_session_auth(dev, xform, session);
+   if (qat_crypto_sym_configure_session_auth(dev, xform, session) 
< 0)
+   return -1;
break;
case ICP_QAT_FW_LA_CMD_CIPHER_HASH:
-   if (xform->type == RTE_CRYPTO_SYM_XFORM_AEAD)
-   session = qat_crypto_sym_configure_session_aead(xform,
-   session);
-   else {
-   session = qat_crypto_sym_configure_session_cipher(dev,
-   xform, session);
-   session = qat_crypto_sym_configure_session_auth(dev,
-   xform, session);
+   if (xform->type == RTE_CRYPTO_SYM_XFORM_AEAD) {
+   if (qat_crypto_sym_configure_session_aead(xform,
+   session) < 0)
+   return -1;
+   } else {
+   if (qat_crypto_sym_configure_session_cipher(dev,
+   xform, session) < 0)
+   return -1;
+   if (qat_crypto_sym_configure_session_auth(dev,
+   xform, session) < 0)
+   return -1;
}
break;
case ICP_QAT_FW_LA_CMD_HASH_CIPHER:
-   if (xform->type == RTE_CRYPTO_SYM_XFORM_AEAD)
-   session = qat_crypto_sym_configure_session_aead(xform,
-   session);
-   else {
-   session = qat_crypto_sym_configure_session_auth(dev,
-   xform, session);
-   session = qat_crypto_sym_configure_session_cipher(dev,
-   xform, session);
+   if (xform->type == RTE_CRYPTO_SYM_XFORM_AEAD) {
+   if (qat_crypto_sym_configure_session_aead(xform,
+   session) < 0)
+   return -1;
+   } else {
+   if (qat_crypto_sym_configure_session_auth(dev,
+   xform, session) < 0)
+   return -1;
+   if (qat_crypto_sym_configure_session_cipher(dev,
+   xform, session) < 0)
+   return -1;
}

Re: [dpdk-dev] [PATCH] examples/l2fwd-crypto: add option --[no-]mac-updating

[De Lara Guarch, Pablo]

> -Original Message-
> From: Kozak, KubaX
> Sent: Wednesday, July 19, 2017 1:20 PM
> To: dev@dpdk.org
> Cc: Jain, Deepak K ; De Lara Guarch, Pablo
> ; Jastrzebski, MichalX K
> ; Kozak, KubaX
> 
> Subject: [PATCH] examples/l2fwd-crypto: add option --[no-]mac-updating
> 
> This patch adds a new option to enable/disable the MAC addresses
> updating done at forwarding time: --[no-]mac-updating
> 
> By default, MAC address updating remains enabled, to keep consistency
> with previous usage.
> 
> Signed-off-by: Kuba Kozak 

Acked-by: Pablo de Lara 

[dpdk-dev] [PATCH] net/failsafe: fix compilation issues on GCC 4.8.5

[Gaetan Rivet]

Reintroduce the -pedantic flag to highlight those mistakes.

Fixes: 57a089a5020f ("net/failsafe: add fail-safe PMD")

Signed-off-by: Gaetan Rivet 
---

 drivers/net/failsafe/Makefile | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/failsafe/Makefile b/drivers/net/failsafe/Makefile
index cec33df..322 100644
--- a/drivers/net/failsafe/Makefile
+++ b/drivers/net/failsafe/Makefile
@@ -54,7 +54,9 @@ CFLAGS += -std=c11 -Wextra
 CFLAGS += -O3
 CFLAGS += -I.
 CFLAGS += -D_DEFAULT_SOURCE
+CFLAGS += -D_XOPEN_SOURCE=700
 CFLAGS += $(WERROR_FLAGS)
 CFLAGS += -Wno-strict-prototypes
+CFLAGS += -pedantic
 
 include $(RTE_SDK)/mk/rte.lib.mk
-- 
2.1.4

Re: [dpdk-dev] [PATCH] net/failsafe: fix compilation issues on GCC 4.8.5

[Ferruh Yigit]

On 7/19/2017 2:18 PM, Gaetan Rivet wrote:
> Reintroduce the -pedantic flag to highlight those mistakes.
> 
> Fixes: 57a089a5020f ("net/failsafe: add fail-safe PMD")
> 
> Signed-off-by: Gaetan Rivet 

Squashed into relevant commit in next-net, thanks.

Re: [dpdk-dev] [PATCH v2 0/4] cryptodev vdev changes for -rc2

[De Lara Guarch, Pablo]

> -Original Message-
> From: Jan Blunck [mailto:jblu...@gmail.com] On Behalf Of Jan Blunck
> Sent: Wednesday, July 12, 2017 8:59 PM
> To: dev@dpdk.org
> Cc: Doherty, Declan ; De Lara Guarch, Pablo
> 
> Subject: [PATCH v2 0/4] cryptodev vdev changes for -rc2
> 
> This series is a preparation to move the vdev bus out of librte_eal. For that
> the newly added cryptodev vdev functions need to change signature to not
> require the rte_vdev.h header.
> 
> Changes since v1:
> - move params parsing into new header
> - make rte_cryptodev_vdev_pmd_init() static inline
> 
> Jan Blunck (4):
>   cryptodev: remove obsolete include
>   cryptodev: move initialization
>   cryptodev: rework PMD init to not require rte_vdev.h
>   cryptodev: move parameter parsing to its own header
> 
>  lib/librte_cryptodev/Makefile|  1 +
>  lib/librte_cryptodev/rte_cryptodev.c |  3 +
>  lib/librte_cryptodev/rte_cryptodev.h |  1 -
>  lib/librte_cryptodev/rte_cryptodev_pmd.c | 37 +-
>  lib/librte_cryptodev/rte_cryptodev_vdev.h| 71 ---
>  lib/librte_cryptodev/rte_cryptodev_vdev_params.h | 89
> 
>  lib/librte_cryptodev/rte_cryptodev_version.map   |  1 -
>  7 files changed, 122 insertions(+), 81 deletions(-)  create mode 100644
> lib/librte_cryptodev/rte_cryptodev_vdev_params.h
> 
> --
> 2.13.2

Hi Jan,

Since there are some issues in this patchset and knowing that these changes are 
required
for work to be done in 17.11 (vdev moving out of EAL), let's postpone this for 
the next release,
as also there is no API breakage in this (correct me if I am wrong).

Thanks,
Pablo

Re: [dpdk-dev] [PATCH v2] igb_uio: issue FLR during open and release of device file

[Ferruh Yigit]

On 7/12/2017 4:40 AM, Tan, Jianfeng wrote:
> 
> 
>> -Original Message-
>> From: Shijith Thotton [mailto:shijith.thot...@caviumnetworks.com]
>> Sent: Friday, July 7, 2017 7:14 PM
>> To: dev@dpdk.org
>> Cc: Yigit, Ferruh; Gregory Etelson; Thomas Monjalon; Stephen Hemminger;
>> Tan, Jianfeng; Lu, Wenzhuo
>> Subject: [PATCH v2] igb_uio: issue FLR during open and release of device file
>>
>> Set UIO info device file operations open and release. Call pci reset
>> function inside open and release to clear device state at start and end.
>> Copied this behaviour from vfio_pci kernel module code. With this patch,
>> it is not mandatory to issue FLR by PMD's during init and close.
>>
>> Bus master enable and disable are added in open and release respectively
>> to take care of device DMA.
>>
>> Signed-off-by: Shijith Thotton 
> 
> Reviewed-by: Jianfeng Tan 

Acked-by: Ferruh Yigit 

Re: [dpdk-dev] [PATCH] cryptodev: remove crypto operation status value

[Declan Doherty]

On 17/07/2017 4:25 PM, Kirill Rybalchenko wrote:

Crypto operation status RTE_CRYPTO_OP_STATUS_ENQUEUED is removed
from rte_crypto.h as it is not needed for crypto operation processing.
This status value is redundant to RTE_CRYPTO_OP_STATUS_NOT_PROCESSED value
and it was not intended to be part of public API.

Signed-off-by: Kirill Rybalchenko 
---

...




Acked-by: Declan Doherty 

Re: [dpdk-dev] [PATCH v1] crypto/scheduler: fix multicore scheduler reordering

[Declan Doherty]

On 18/07/2017 11:35 AM, Kirill Rybalchenko wrote:

Operations can be dequeued from the reordering ring only after they
were dequeued from the crypto pmd with rte_cryptodev_dequeue_burst()
function. It is not correct to dequeue them when status just changed
from RTE_CRYPTO_OP_STATUS_NOT_PROCESSED to any other value, as the
operations still can be processed by crypto pmd internally.
Now multicore scheduler workers mark status of all dequeued from
crypto pmd operations with CRYPTO_OP_STATUS_BIT_COMPLETE bit set.
Scheduler will dequeue crypto operations from reordering ring only
when this status bit is set. Prior to put this operation to output
buffer, scheduler clears this bit, so the application gets
unmodified status from crypto pmd.

Fixes: 4c07e0552f0a ("crypto/scheduler: add multicore scheduling mode")

Signed-off-by: Kirill Rybalchenko 
---

...



Acked-by: Declan Doherty 

Re: [dpdk-dev] [PATCH] crypto/qat: fix session initialization

[Trahe, Fiona]

> -Original Message-
> From: De Lara Guarch, Pablo
> Sent: Wednesday, July 19, 2017 5:59 AM
> To: Trahe, Fiona ; Jain, Deepak K 
> ; Griffin, John
> 
> Cc: dev@dpdk.org; De Lara Guarch, Pablo 
> Subject: [PATCH] crypto/qat: fix session initialization
> 
> When creating a session, if there is a failure when
> setting some of the parameters, QAT was not propagating
> the error to the session initialization function.
> Therefore, it was reporting a success, when it should
> be report a failure.
> 
> Fixes: b3bbd9e5f265 ("cryptodev: support device independent sessions")
> 
> Signed-off-by: Pablo de Lara 
In the 3 fn prototypes in qat_crypto.h it would be nice to remove the _private 
from 
struct qat_session *session_private. Apart from that it's fine.

Acked-by: Fiona Trahe 

[dpdk-dev] [PATCH] event/dpaa2: advertise the burst mode capability

[Nipun Gupta]

Burst mode capability flag was introduced in 73e6b8c9 for event drivers.
DPAA2 event driver supports burst mode so this patch adds this capability
flag in DPAA2 event driver

Signed-off-by: Nipun Gupta 
---
 drivers/event/dpaa2/dpaa2_eventdev.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/event/dpaa2/dpaa2_eventdev.c 
b/drivers/event/dpaa2/dpaa2_eventdev.c
index ed57376..cf2d274 100644
--- a/drivers/event/dpaa2/dpaa2_eventdev.c
+++ b/drivers/event/dpaa2/dpaa2_eventdev.c
@@ -315,7 +315,8 @@ static void dpaa2_eventdev_process_atomic(struct qbman_swp 
*swp,
dev_info->max_event_port_enqueue_depth =
DPAA2_EVENT_MAX_PORT_ENQUEUE_DEPTH;
dev_info->max_num_events = DPAA2_EVENT_MAX_NUM_EVENTS;
-   dev_info->event_dev_cap = RTE_EVENT_DEV_CAP_DISTRIBUTED_SCHED;
+   dev_info->event_dev_cap = RTE_EVENT_DEV_CAP_DISTRIBUTED_SCHED |
+   RTE_EVENT_DEV_CAP_BURST_MODE;
 }
 
 static int
-- 
1.9.1

[dpdk-dev] [PATCH v2 0/3] Bonding add additional aggregators mode for 802.3AD

[Daniel Mrzyglod]

This patchseries add support for other aggregators in similar manner
that is provided in Linux kernel.

Modes added in patches:
stable - this is default mode from IEEE802.11AX/IEEE802.3AD documentation
bandwidth - takes aggregator with the biggest available speed
count - takes aggregators with the biggest number of slaves

V2:
-fix eal argument parsing
-add cmdline help in testpmd
-add unit test
-fix checkpatch warnings
-update device name in unit tests

Daniel Mrzyglod (3):
  drivers/bonding: add other agg selection modes
  testpmd: add cmndlines to support different aggregation modes
  test/bonding: add test case for agg selection in mode4

 app/test-pmd/cmdline.c|  94 ++-
 drivers/net/bonding/rte_eth_bond_8023ad.c | 193 --
 drivers/net/bonding/rte_eth_bond_8023ad.h |  32 
 drivers/net/bonding/rte_eth_bond_8023ad_private.h |   1 +
 drivers/net/bonding/rte_eth_bond_args.c   |  33 
 drivers/net/bonding/rte_eth_bond_pmd.c|  19 ++-
 drivers/net/bonding/rte_eth_bond_private.h|   5 +
 drivers/net/bonding/rte_eth_bond_version.map  |   5 +
 test/test/test_link_bonding_mode4.c   |  83 +-
 9 files changed, 446 insertions(+), 19 deletions(-)

-- 
2.13.3

[dpdk-dev] [PATCH v2 1/3] drivers/bonding: add other agg selection modes

[Daniel Mrzyglod]

This patch add support for setting additional aggregator modes for IEEE802.3AD
in similar manner that are supported in kernel mode.

This will add support for other manner:
stable - default mode taken from IEEE802.11AX this is default aggregator mode
bandwidth - takes aggregator with highest bandwidth
count - takes aggregator with biggest number of slaves

Signed-off-by: Daniel Mrzyglod 
---
 drivers/net/bonding/rte_eth_bond_8023ad.c | 193 --
 drivers/net/bonding/rte_eth_bond_8023ad.h |  32 
 drivers/net/bonding/rte_eth_bond_8023ad_private.h |   1 +
 drivers/net/bonding/rte_eth_bond_args.c   |  33 
 drivers/net/bonding/rte_eth_bond_pmd.c|  19 ++-
 drivers/net/bonding/rte_eth_bond_private.h|   5 +
 drivers/net/bonding/rte_eth_bond_version.map  |   5 +
 7 files changed, 274 insertions(+), 14 deletions(-)

diff --git a/drivers/net/bonding/rte_eth_bond_8023ad.c 
b/drivers/net/bonding/rte_eth_bond_8023ad.c
index a2313b327..eb273988a 100644
--- a/drivers/net/bonding/rte_eth_bond_8023ad.c
+++ b/drivers/net/bonding/rte_eth_bond_8023ad.c
@@ -44,7 +44,6 @@
 #include "rte_eth_bond_private.h"
 
 static void bond_mode_8023ad_ext_periodic_cb(void *arg);
-
 #ifdef RTE_LIBRTE_BOND_DEBUG_8023AD
 #define MODE4_DEBUG(fmt, ...) RTE_LOG(DEBUG, PMD, "%6u [Port %u: %s] " fmt, \
bond_dbg_get_time_diff_ms(), slave_id, \
@@ -660,6 +659,25 @@ tx_machine(struct bond_dev_private *internals, uint8_t 
slave_id)
SM_FLAG_CLR(port, NTT);
 }
 
+static uint8_t
+max_index(uint64_t *a, int n)
+{
+   if (n <= 0)
+   return -1;
+
+   int i, max_i = 0;
+   uint64_t max = a[0];
+
+   for (i = 1; i < n; ++i) {
+   if (a[i] > max) {
+   max = a[i];
+   max_i = i;
+   }
+   }
+
+   return max_i;
+}
+
 /**
  * Function assigns port to aggregator.
  *
@@ -670,8 +688,13 @@ static void
 selection_logic(struct bond_dev_private *internals, uint8_t slave_id)
 {
struct port *agg, *port;
-   uint8_t slaves_count, new_agg_id, i;
+   uint8_t slaves_count, new_agg_id, i, j = 0;
uint8_t *slaves;
+   uint64_t agg_bandwidth[8] = {0};
+   uint64_t agg_count[8] = {0};
+   uint8_t default_slave = 0;
+   uint8_t mode_count_id, mode_band_id;
+   struct rte_eth_link link_info;
 
slaves = internals->active_slaves;
slaves_count = internals->active_slave_count;
@@ -684,6 +707,10 @@ selection_logic(struct bond_dev_private *internals, 
uint8_t slave_id)
if (agg->aggregator_port_id != slaves[i])
continue;
 
+   agg_count[agg->aggregator_port_id] += 1;
+   rte_eth_link_get_nowait(slaves[i], &link_info);
+   agg_bandwidth[agg->aggregator_port_id] += link_info.link_speed;
+
/* Actors system ID is not checked since all slave device have 
the same
 * ID (MAC address). */
if ((agg->actor.key == port->actor.key &&
@@ -694,15 +721,36 @@ selection_logic(struct bond_dev_private *internals, 
uint8_t slave_id)
(agg->actor.key &
rte_cpu_to_be_16(BOND_LINK_FULL_DUPLEX_KEY)) != 
0) {
 
-   break;
+   if (j == 0)
+   default_slave = i;
+   j++;
}
}
 
-   /* By default, port uses it self as agregator */
-   if (i == slaves_count)
-   new_agg_id = slave_id;
-   else
-   new_agg_id = slaves[i];
+   switch (internals->mode4.agg_selection) {
+   case AGG_COUNT:
+   mode_count_id = max_index(
+   (uint64_t *)agg_count, slaves_count);
+   new_agg_id = mode_count_id;
+   break;
+   case AGG_BANDWIDTH:
+   mode_band_id = max_index(
+   (uint64_t *)agg_bandwidth, slaves_count);
+   new_agg_id = mode_band_id;
+   break;
+   case AGG_STABLE:
+   if (default_slave == slaves_count)
+   new_agg_id = slave_id;
+   else
+   new_agg_id = slaves[default_slave];
+   break;
+   default:
+   if (default_slave == slaves_count)
+   new_agg_id = slave_id;
+   else
+   new_agg_id = slaves[default_slave];
+   break;
+   }
 
if (new_agg_id != port->aggregator_port_id) {
port->aggregator_port_id = new_agg_id;
@@ -909,7 +957,7 @@ bond_mode_8023ad_activate_slave(struct rte_eth_dev 
*bond_dev, uint8_t slave_id)
 
/* default states */
port->actor_state = STATE_AGGREGATION | STATE_LACP_ACTIVE | 
STATE_DEFAULTED;
-   port->partner_state = STATE_LACP_ACTIVE;
+   port->partner_state = STATE_LACP_ACT

Re: [dpdk-dev] [PATCH v1] crypto/scheduler: fix multicore scheduler reordering

[De Lara Guarch, Pablo]

> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Declan Doherty
> Sent: Wednesday, July 19, 2017 3:03 PM
> To: Rybalchenko, Kirill ; Zhang, Roy Fan
> 
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v1] crypto/scheduler: fix multicore
> scheduler reordering
> 
> On 18/07/2017 11:35 AM, Kirill Rybalchenko wrote:
> > Operations can be dequeued from the reordering ring only after they
> > were dequeued from the crypto pmd with
> rte_cryptodev_dequeue_burst()
> > function. It is not correct to dequeue them when status just changed
> > from RTE_CRYPTO_OP_STATUS_NOT_PROCESSED to any other value, as
> the
> > operations still can be processed by crypto pmd internally.
> > Now multicore scheduler workers mark status of all dequeued from
> > crypto pmd operations with CRYPTO_OP_STATUS_BIT_COMPLETE bit set.
> > Scheduler will dequeue crypto operations from reordering ring only
> > when this status bit is set. Prior to put this operation to output
> > buffer, scheduler clears this bit, so the application gets unmodified
> > status from crypto pmd.
> >
> > Fixes: 4c07e0552f0a ("crypto/scheduler: add multicore scheduling
> > mode")
> >
> > Signed-off-by: Kirill Rybalchenko 
> > ---
> ...
> >
> Acked-by: Declan Doherty 

Applied to dpdk-next-crypto.
Thanks,

Pablo

Re: [dpdk-dev] [PATCH] cryptodev: remove crypto operation status value

[De Lara Guarch, Pablo]

> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Declan Doherty
> Sent: Wednesday, July 19, 2017 3:00 PM
> To: Rybalchenko, Kirill ; Zhang, Roy Fan
> 
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] cryptodev: remove crypto operation status
> value
> 
> On 17/07/2017 4:25 PM, Kirill Rybalchenko wrote:
> > Crypto operation status RTE_CRYPTO_OP_STATUS_ENQUEUED is
> removed from
> > rte_crypto.h as it is not needed for crypto operation processing.
> > This status value is redundant to
> RTE_CRYPTO_OP_STATUS_NOT_PROCESSED
> > value and it was not intended to be part of public API.
> >
> > Signed-off-by: Kirill Rybalchenko 
> > ---
> ...
> >
> 
> Acked-by: Declan Doherty 

Applied to dpdk-next-crypto.
Thanks,

Pablo

Re: [dpdk-dev] [PATCH v7 0/5] Support NIC reset and keep same port id

[Dai, Wei]

Remy is a Native English speaker.
He will help me review my v8 patch set.
Thank you, Remy.

> -Original Message-
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Monday, July 17, 2017 3:26 AM
> To: Dai, Wei 
> Cc: dev@dpdk.org; Lu, Wenzhuo ; Ananyev,
> Konstantin ; Wu, Jingjing
> ; Xing, Beilei ; Mcnamara,
> John 
> Subject: Re: [dpdk-dev] [PATCH v7 0/5] Support NIC reset and keep same
> port id
> 
> 13/07/2017 17:53, Wei Dai:
> > chagnes:
> > v7:
> >   add description of NIC reset in doc/poll_mode_drv.rst
> 
> Thanks for bringing more explanations.
> The texts have to be reviewed by a native english speaker.
> 
> About the organization of the doc, it is good to explain the VF use case in 
> the
> prog guide as you do. But you should not add it to the doxygen.
> The doxygen must be concise enough and explains only the API:
>   - when it can be called
>   - what a driver is doing when calling reset
>   - what is the expected state of the device after reset

Re: [dpdk-dev] [PATCH] crypto/qat: fix session initialization

[De Lara Guarch, Pablo]

> -Original Message-
> From: Trahe, Fiona
> Sent: Wednesday, July 19, 2017 3:06 PM
> To: De Lara Guarch, Pablo ; Jain, Deepak K
> ; Griffin, John 
> Cc: dev@dpdk.org; Trahe, Fiona 
> Subject: RE: [PATCH] crypto/qat: fix session initialization
> 
> 
> 
> > -Original Message-
> > From: De Lara Guarch, Pablo
> > Sent: Wednesday, July 19, 2017 5:59 AM
> > To: Trahe, Fiona ; Jain, Deepak K
> > ; Griffin, John 
> > Cc: dev@dpdk.org; De Lara Guarch, Pablo
> > 
> > Subject: [PATCH] crypto/qat: fix session initialization
> >
> > When creating a session, if there is a failure when setting some of
> > the parameters, QAT was not propagating the error to the session
> > initialization function.
> > Therefore, it was reporting a success, when it should be report a
> > failure.
> >
> > Fixes: b3bbd9e5f265 ("cryptodev: support device independent sessions")
> >
> > Signed-off-by: Pablo de Lara 
> In the 3 fn prototypes in qat_crypto.h it would be nice to remove the
> _private from struct qat_session *session_private. Apart from that it's fine.
> 
> Acked-by: Fiona Trahe 

I fixed that when merging the patch :) Thanks!

Applied to dpdk-next-crypto.

Pablo

Re: [dpdk-dev] [PATCH] examples/l2fwd-crypto: add option --[no-]mac-updating

[De Lara Guarch, Pablo]

> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of De Lara Guarch,
> Pablo
> Sent: Wednesday, July 19, 2017 2:17 PM
> To: Kozak, KubaX ; dev@dpdk.org
> Cc: Jain, Deepak K ; Jastrzebski, MichalX K
> 
> Subject: Re: [dpdk-dev] [PATCH] examples/l2fwd-crypto: add option --[no-
> ]mac-updating
> 
> 
> 
> > -Original Message-
> > From: Kozak, KubaX
> > Sent: Wednesday, July 19, 2017 1:20 PM
> > To: dev@dpdk.org
> > Cc: Jain, Deepak K ; De Lara Guarch, Pablo
> > ; Jastrzebski, MichalX K
> > ; Kozak, KubaX
> > 
> > Subject: [PATCH] examples/l2fwd-crypto: add option --[no-]mac-updating
> >
> > This patch adds a new option to enable/disable the MAC addresses
> > updating done at forwarding time: --[no-]mac-updating
> >
> > By default, MAC address updating remains enabled, to keep consistency
> > with previous usage.
> >
> > Signed-off-by: Kuba Kozak 
> 
> Acked-by: Pablo de Lara 

Applied to dpdk-next-crypto.
Thanks,

Pablo

Re: [dpdk-dev] [PATCH] event/dpaa2: advertise the burst mode capability

[Jerin Jacob]

-Original Message-
> Date: Wed, 19 Jul 2017 19:48:32 +0530
> From: Nipun Gupta 
> To: dev@dpdk.org
> CC: hemant.agar...@nxp.com, jerin.ja...@caviumnetworks.com,
>  tho...@monjalon.net, Nipun Gupta 
> Subject: [PATCH] event/dpaa2: advertise the burst mode capability
> X-Mailer: git-send-email 1.9.1
> 
> Burst mode capability flag was introduced in 73e6b8c9 for event drivers.
> DPAA2 event driver supports burst mode so this patch adds this capability
> flag in DPAA2 event driver
> 
> Signed-off-by: Nipun Gupta 

Reviewed-by: Jerin Jacob 

> ---
>  drivers/event/dpaa2/dpaa2_eventdev.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/event/dpaa2/dpaa2_eventdev.c 
> b/drivers/event/dpaa2/dpaa2_eventdev.c
> index ed57376..cf2d274 100644
> --- a/drivers/event/dpaa2/dpaa2_eventdev.c
> +++ b/drivers/event/dpaa2/dpaa2_eventdev.c
> @@ -315,7 +315,8 @@ static void dpaa2_eventdev_process_atomic(struct 
> qbman_swp *swp,
>   dev_info->max_event_port_enqueue_depth =
>   DPAA2_EVENT_MAX_PORT_ENQUEUE_DEPTH;
>   dev_info->max_num_events = DPAA2_EVENT_MAX_NUM_EVENTS;
> - dev_info->event_dev_cap = RTE_EVENT_DEV_CAP_DISTRIBUTED_SCHED;
> + dev_info->event_dev_cap = RTE_EVENT_DEV_CAP_DISTRIBUTED_SCHED |
> + RTE_EVENT_DEV_CAP_BURST_MODE;
>  }
>  
>  static int
> -- 
> 1.9.1
> 

[dpdk-dev] [pull-request] next-crypto 17.08 rc2

[Pablo de Lara]

The following changes since commit 838d6b775bfab436859764ea734cddaf95c7b5fd:

  mbuf: fix VXLAN port in comment (2017-07-19 09:17:35 +0300)

are available in the git repository at:

  http://dpdk.org/git/next/dpdk-next-crypto 

for you to fetch changes up to 488587bb9a327f195379ced5650043189d05584d:

  examples/l2fwd-crypto: add option to update MAC address (2017-07-19 15:33:07 
+0100)


Akhil Goyal (3):
  crypto/dpaa2_sec: fix the incorrect free usage for dpsec
  crypto/dpaa2_sec: fix typo in PMD name
  doc: add missing devices in test-crypto-perf

Arek Kusztal (4):
  crypto/qat: remove unused ablkcipher struct and functions
  crypto/qat: fix NULL auth hang issue
  crypto/qat: fix authentication offset and length for GMAC
  crypto/qat: fix handle device-agnostic session

Fan Zhang (1):
  crypto/scheduler: fix slave name parsing

Kirill Rybalchenko (4):
  doc: extend info on multi-core scheduler
  app/crypto-perf: stop crypto devices after test
  cryptodev: remove crypto operation status value
  crypto/scheduler: fix multicore scheduler reordering

Kuba Kozak (1):
  examples/l2fwd-crypto: add option to update MAC address

Pablo de Lara (19):
  doc: fix crypto scheduler command line examples
  doc: use new crypto driver names
  crypto/aesni_mb: fix possible crypto job leak
  cryptodev: remove AAD size in auth capabilities
  doc: fix supported algorithm table for AESNI GCM PMD
  doc: remove incorrect limitation on QAT PMD
  doc: remove incorrect limitation on AESNI MB PMD
  doc: add missing algorithm in limitations for QAT
  doc: extend installation section on SNOW3G/KASUMI PMDs
  cryptodev: fix KASUMI F9 expected parameters
  crypto/aesni_mb: fix zero burst dequeue
  cryptodev: modify some field sizes
  cryptodev: rename field name
  cryptodev: reorder auth transform
  crypto/scheduler: fix buffer not null terminated
  examples/l2fwd-crypto: fix digest length
  examples/l2fwd-crypto: fix AEAD IV setting
  examples/l2fwd-crypto: fix AEAD key setting
  crypto/qat: fix session initialization

Srisivasubramanian S (1):
  test/crypto-perf: fix ARMV8 session creation

 app/test-crypto-perf/cperf_ops.c   |   2 +-
 app/test-crypto-perf/cperf_test_latency.c  |   3 +-
 app/test-crypto-perf/cperf_test_throughput.c   |   2 +
 app/test-crypto-perf/cperf_test_verify.c   |   2 +
 app/test-crypto-perf/main.c|   1 -
 doc/guides/cryptodevs/aesni_mb.rst |   2 +-
 doc/guides/cryptodevs/features/aesni_gcm.ini   |   1 +
 doc/guides/cryptodevs/kasumi.rst   |  31 +++-
 doc/guides/cryptodevs/qat.rst  |  51 --
 doc/guides/cryptodevs/scheduler.rst|   8 +-
 doc/guides/cryptodevs/snow3g.rst   |  11 ++
 doc/guides/prog_guide/cryptodev_lib.rst|   4 +-
 doc/guides/rel_notes/release_17_08.rst |   5 +
 doc/guides/sample_app_ug/ipsec_secgw.rst   |   6 +-
 doc/guides/sample_app_ug/l2_forward_crypto.rst |  11 +-
 doc/guides/tools/cryptoperf.rst|   9 +-
 drivers/crypto/aesni_gcm/aesni_gcm_pmd.c   |   2 +-
 drivers/crypto/aesni_gcm/aesni_gcm_pmd_ops.c   |   1 -
 drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c |  10 +-
 drivers/crypto/aesni_mb/rte_aesni_mb_pmd_ops.c |   7 -
 drivers/crypto/armv8/rte_armv8_pmd_ops.c   |   2 -
 drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c|   4 +-
 drivers/crypto/dpaa2_sec/dpaa2_sec_priv.h  |  10 +-
 drivers/crypto/kasumi/rte_kasumi_pmd.c |  27 +--
 drivers/crypto/kasumi/rte_kasumi_pmd_ops.c |   7 +-
 drivers/crypto/kasumi/rte_kasumi_pmd_private.h |   1 -
 drivers/crypto/null/null_crypto_pmd_ops.c  |   1 -
 drivers/crypto/openssl/rte_openssl_pmd.c   |   2 +-
 drivers/crypto/openssl/rte_openssl_pmd_ops.c   |  12 --
 drivers/crypto/qat/qat_adf/qat_algs.h  |  20 +--
 drivers/crypto/qat/qat_adf/qat_algs_build_desc.c   |  16 +-
 drivers/crypto/qat/qat_crypto.c| 168 +-
 drivers/crypto/qat/qat_crypto.h|  23 ++-
 drivers/crypto/qat/qat_crypto_capabilities.h   |  24 +--
 drivers/crypto/qat/qat_qp.c|   5 +
 drivers/crypto/qat/rte_qat_cryptodev.c |  16 +-
 drivers/crypto/scheduler/rte_cryptodev_scheduler.c |  14 ++
 drivers/crypto/scheduler/rte_cryptodev_scheduler.h |   1 +
 drivers/crypto/scheduler/scheduler_multicore.c |  94 +++---
 drivers/crypto/scheduler/scheduler_pmd.c   |   2 +-
 drivers/crypto/scheduler/scheduler_pmd_private.h   |   3 +-
 drivers/crypto/snow3g/rte_snow3g_pmd_ops.c |   3 +-
 drivers/crypto/zuc/rte_zuc_pmd_ops.c   |   3 +-
 examples/ipsec-secgw/sa.c  |   4 +-

[dpdk-dev] [PATCH v2 2/3] testpmd: add cmndlines to support different aggregation modes

[Daniel Mrzyglod]

This patch add support for different aggregator modes in similar manner
that is provided in linux kernel.

testpmd> set bonding agg_mode  
testpmd> show bonding config 

Signed-off-by: Daniel Mrzyglod 
---
 app/test-pmd/cmdline.c | 94 +-
 1 file changed, 93 insertions(+), 1 deletion(-)

diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index 7decb96c8..7e08674b9 100644
--- a/app/test-pmd/cmdline.c
+++ b/app/test-pmd/cmdline.c
@@ -579,6 +579,9 @@ static void cmd_help_long_parsed(void *parsed_result,
"set bonding mac_addr (port_id) (address)\n"
"   Set the MAC address of a bonded device.\n\n"
 
+   "set bonding mode IEEE802.3AD aggregator policy 
(port_id) (agg_name)"
+   "   Set Aggregation mode for IEEE802.3AD (mode 4)"
+
"set bonding xmit_balance_policy (port_id) 
(l2|l23|l34)\n"
"   Set the transmit balance policy for bonded 
device running in balance mode.\n\n"
 
@@ -4589,7 +4592,7 @@ static void cmd_show_bonding_config_parsed(void 
*parsed_result,
__attribute__((unused)) void *data)
 {
struct cmd_show_bonding_config_result *res = parsed_result;
-   int bonding_mode;
+   int bonding_mode, agg_mode;
uint8_t slaves[RTE_MAX_ETHPORTS];
int num_slaves, num_active_slaves;
int primary_id;
@@ -4630,6 +4633,23 @@ static void cmd_show_bonding_config_parsed(void 
*parsed_result,
}
}
 
+   if (bonding_mode == BONDING_MODE_8023AD) {
+   agg_mode = rte_eth_bond_8023ad_agg_selection_get(port_id);
+   printf("\tIEEE802.3AD Aggregator Mode: ");
+   switch (agg_mode) {
+   case AGG_BANDWIDTH:
+   printf("bandwidth");
+   break;
+   case AGG_STABLE:
+   printf("stable");
+   break;
+   case AGG_COUNT:
+   printf("count");
+   break;
+   }
+   printf("\n");
+   }
+
num_slaves = rte_eth_bond_slaves_get(port_id, slaves, RTE_MAX_ETHPORTS);
 
if (num_slaves < 0) {
@@ -5062,6 +5082,77 @@ cmdline_parse_inst_t cmd_set_bond_mon_period = {
}
 };
 
+
+
+struct cmd_set_bonding_agg_mode_policy_result {
+   cmdline_fixed_string_t set;
+   cmdline_fixed_string_t bonding;
+   cmdline_fixed_string_t agg_mode;
+   uint8_t port_num;
+   cmdline_fixed_string_t policy;
+};
+
+
+static void
+cmd_set_bonding_agg_mode(void *parsed_result,
+   __attribute__((unused)) struct cmdline *cl,
+   __attribute__((unused)) void *data)
+{
+   struct cmd_set_bonding_agg_mode_policy_result *res = parsed_result;
+   uint8_t policy = AGG_BANDWIDTH;
+
+   if (res->port_num >= nb_ports) {
+   printf("Port id %d must be less than %d\n",
+   res->port_num, nb_ports);
+   return;
+   }
+
+   if (!strcmp(res->policy, "bandwidth"))
+   policy = AGG_BANDWIDTH;
+   else if (!strcmp(res->policy, "stable"))
+   policy = AGG_STABLE;
+   else if (!strcmp(res->policy, "count"))
+   policy = AGG_COUNT;
+
+   rte_eth_bond_8023ad_agg_selection_set(res->port_num, policy);
+}
+
+
+cmdline_parse_token_string_t cmd_set_bonding_agg_mode_set =
+   TOKEN_STRING_INITIALIZER(struct cmd_set_bonding_agg_mode_policy_result,
+   set, "set");
+cmdline_parse_token_string_t cmd_set_bonding_agg_mode_bonding =
+   TOKEN_STRING_INITIALIZER(struct cmd_set_bonding_agg_mode_policy_result,
+   bonding, "bonding");
+
+cmdline_parse_token_string_t cmd_set_bonding_agg_mode_agg_mode =
+   TOKEN_STRING_INITIALIZER(struct cmd_set_bonding_agg_mode_policy_result,
+   agg_mode, "agg_mode");
+
+cmdline_parse_token_num_t cmd_set_bonding_agg_mode_portnum =
+   TOKEN_NUM_INITIALIZER(struct cmd_set_bonding_agg_mode_policy_result,
+   port_num, UINT8);
+
+cmdline_parse_token_string_t cmd_set_bonding_agg_mode_policy_string =
+   TOKEN_STRING_INITIALIZER(
+   struct cmd_set_bonding_balance_xmit_policy_result,
+   policy, "stable#bandwidth#count");
+
+cmdline_parse_inst_t cmd_set_bonding_agg_mode_policy = {
+   .f = cmd_set_bonding_agg_mode,
+   .data = (void *) 0,
+   .help_str = "set bonding mode IEEE802.3AD aggregator policy  
",
+   .tokens = {
+   (void *)&cmd_set_bonding_agg_mode_set,
+   (void *)&cmd_set_bonding_agg_mode_bonding,
+   (void *)&cmd_set_bonding_agg_mode_agg_mode,
+   (void *)&cmd_set_bonding_agg_mode_portnum,
+   (void *)&cmd_se

[dpdk-dev] [PATCH v2 3/3] test/bonding: add test case for agg selection in mode4

[Daniel Mrzyglod]

Signed-off-by: Daniel Mrzyglod 
---
 test/test/test_link_bonding_mode4.c | 83 +++--
 1 file changed, 79 insertions(+), 4 deletions(-)

diff --git a/test/test/test_link_bonding_mode4.c 
b/test/test/test_link_bonding_mode4.c
index 8b64bbf71..8e9e23db5 100644
--- a/test/test/test_link_bonding_mode4.c
+++ b/test/test/test_link_bonding_mode4.c
@@ -73,11 +73,11 @@
 #define MAX_PKT_BURST   (32)
 #define DEF_PKT_BURST   (16)
 
-#define BONDED_DEV_NAME ("ut_mode4_bond_dev")
+#define BONDED_DEV_NAME ("net_bonding_m4_bond_dev")
 
-#define SLAVE_DEV_NAME_FMT  ("ut_mode4_slave_%d")
-#define SLAVE_RX_QUEUE_FMT  ("ut_mode4_slave_%d_rx")
-#define SLAVE_TX_QUEUE_FMT  ("ut_mode4_slave_%d_tx")
+#define SLAVE_DEV_NAME_FMT  ("net_virt_%d")
+#define SLAVE_RX_QUEUE_FMT  ("net_virt_%d_rx")
+#define SLAVE_TX_QUEUE_FMT  ("net_virt_%d_tx")
 
 #define INVALID_SOCKET_ID   (-1)
 #define INVALID_PORT_ID (0xFF)
@@ -682,6 +682,74 @@ test_mode4_lacp(void)
 
return TEST_SUCCESS;
 }
+static int
+test_mode4_agg_mode_selection(void)
+{
+   int retval;
+   /* Test and verify for Stable mode */
+   retval = initialize_bonded_device_with_slaves(TEST_LACP_SLAVE_COUT, 0);
+   TEST_ASSERT_SUCCESS(retval, "Failed to initialize bonded device");
+
+
+   retval = rte_eth_bond_8023ad_agg_selection_set(
+   test_params.bonded_port_id, AGG_STABLE);
+   TEST_ASSERT_SUCCESS(retval, "Failed to initialize bond aggregation 
mode");
+   retval = bond_handshake();
+   TEST_ASSERT_SUCCESS(retval, "Initial handshake failed");
+
+
+   retval = rte_eth_bond_8023ad_agg_selection_get(
+   test_params.bonded_port_id);
+   TEST_ASSERT_EQUAL(retval, AGG_STABLE,
+   "Wrong agg mode received from bonding device");
+
+   retval = remove_slaves_and_stop_bonded_device();
+   TEST_ASSERT_SUCCESS(retval, "Test cleanup failed.");
+
+
+   /* test and verify for Bandwidth mode */
+   retval = initialize_bonded_device_with_slaves(TEST_LACP_SLAVE_COUT, 0);
+   TEST_ASSERT_SUCCESS(retval, "Failed to initialize bonded device");
+
+
+   retval = rte_eth_bond_8023ad_agg_selection_set(
+   test_params.bonded_port_id,
+   AGG_BANDWIDTH);
+   TEST_ASSERT_SUCCESS(retval,
+   "Failed to initialize bond aggregation mode");
+   retval = bond_handshake();
+   TEST_ASSERT_SUCCESS(retval, "Initial handshake failed");
+
+   retval = rte_eth_bond_8023ad_agg_selection_get(
+   test_params.bonded_port_id);
+   TEST_ASSERT_EQUAL(retval, AGG_BANDWIDTH,
+   "Wrong agg mode received from bonding device");
+
+   retval = remove_slaves_and_stop_bonded_device();
+   TEST_ASSERT_SUCCESS(retval, "Test cleanup failed.");
+
+   /* test and verify selection for count mode */
+   retval = initialize_bonded_device_with_slaves(TEST_LACP_SLAVE_COUT, 0);
+   TEST_ASSERT_SUCCESS(retval, "Failed to initialize bonded device");
+
+
+   retval = rte_eth_bond_8023ad_agg_selection_set(
+   test_params.bonded_port_id, AGG_COUNT);
+   TEST_ASSERT_SUCCESS(retval,
+   "Failed to initialize bond aggregation mode");
+   retval = bond_handshake();
+   TEST_ASSERT_SUCCESS(retval, "Initial handshake failed");
+
+   retval = rte_eth_bond_8023ad_agg_selection_get(
+   test_params.bonded_port_id);
+   TEST_ASSERT_EQUAL(retval, AGG_COUNT,
+   "Wrong agg mode received from bonding device");
+
+   retval = remove_slaves_and_stop_bonded_device();
+   TEST_ASSERT_SUCCESS(retval, "Test cleanup failed.");
+
+   return TEST_SUCCESS;
+}
 
 static int
 generate_packets(struct ether_addr *src_mac,
@@ -1535,6 +1603,11 @@ test_mode4_executor(int (*test_func)(void))
 }
 
 static int
+test_mode4_agg_mode_selection_wrapper(void){
+   return test_mode4_executor(&test_mode4_agg_mode_selection);
+}
+
+static int
 test_mode4_lacp_wrapper(void)
 {
return test_mode4_executor(&test_mode4_lacp);
@@ -1581,6 +1654,8 @@ static struct unit_test_suite 
link_bonding_mode4_test_suite  = {
.setup = test_setup,
.teardown = testsuite_teardown,
.unit_test_cases = {
+   TEST_CASE_NAMED("test_mode4_agg_mode_selection",
+   test_mode4_agg_mode_selection_wrapper),
TEST_CASE_NAMED("test_mode4_lacp", test_mode4_lacp_wrapper),
TEST_CASE_NAMED("test_mode4_rx", test_mode4_rx_wrapper),
TEST_CASE_NAMED("test_mode4_tx_burst", 
test_mode4_tx_burst_wrapper),
-- 
2.13.3

Re: [dpdk-dev] [PATCH v2 1/3] drivers/bonding: add other agg selection modes

[Declan Doherty]

On 19/07/2017 3:31 PM, Daniel Mrzyglod wrote:

This patch add support for setting additional aggregator modes for IEEE802.3AD
in similar manner that are supported in kernel mode.

This will add support for other manner:
stable - default mode taken from IEEE802.11AX this is default aggregator mode
bandwidth - takes aggregator with highest bandwidth
count - takes aggregator with biggest number of slaves

Signed-off-by: Daniel Mrzyglod 
---
 drivers/net/bonding/rte_eth_bond_8023ad.c | 193 --
 drivers/net/bonding/rte_eth_bond_8023ad.h |  32 
 drivers/net/bonding/rte_eth_bond_8023ad_private.h |   1 +
 drivers/net/bonding/rte_eth_bond_args.c   |  33 
 drivers/net/bonding/rte_eth_bond_pmd.c|  19 ++-
 drivers/net/bonding/rte_eth_bond_private.h|   5 +
 drivers/net/bonding/rte_eth_bond_version.map  |   5 +
 7 files changed, 274 insertions(+), 14 deletions(-)

diff --git a/drivers/net/bonding/rte_eth_bond_8023ad.c 
b/drivers/net/bonding/rte_eth_bond_8023ad.c
index a2313b327..eb273988a 100644
--- a/drivers/net/bonding/rte_eth_bond_8023ad.c

...



Acked-by: Declan Doherty 

Re: [dpdk-dev] [PATCH v2 2/3] testpmd: add cmndlines to support different aggregation modes

[Declan Doherty]

On 19/07/2017 3:46 PM, Daniel Mrzyglod wrote:

This patch add support for different aggregator modes in similar manner
that is provided in linux kernel.

testpmd> set bonding agg_mode  
testpmd> show bonding config 

Signed-off-by: Daniel Mrzyglod 
---

...





Acked-by: Declan Doherty 

Re: [dpdk-dev] [PATCH v2 3/3] test/bonding: add test case for agg selection in mode4

[Declan Doherty]

On 19/07/2017 3:54 PM, Daniel Mrzyglod wrote:

Signed-off-by: Daniel Mrzyglod 
---

...




Acked-by: Declan Doherty 

Re: [dpdk-dev] [PATCH v2 0/3] Bonding add additional aggregators mode for 802.3AD

[Ferruh Yigit]

On 7/19/2017 3:31 PM, Daniel Mrzyglod wrote:
> This patchseries add support for other aggregators in similar manner
> that is provided in Linux kernel.
> 
> Modes added in patches:
> stable - this is default mode from IEEE802.11AX/IEEE802.3AD documentation
> bandwidth - takes aggregator with the biggest available speed
> count - takes aggregators with the biggest number of slaves
> 
> V2:
> -fix eal argument parsing
> -add cmdline help in testpmd
> -add unit test
> -fix checkpatch warnings
> -update device name in unit tests
> 
> Daniel Mrzyglod (3):
>   drivers/bonding: add other agg selection modes
>   testpmd: add cmndlines to support different aggregation modes
>   test/bonding: add test case for agg selection in mode4

Series applied to dpdk-next-net/master, thanks.

Re: [dpdk-dev] [PATCH v2 2/3] testpmd: add cmndlines to support different aggregation modes

[Ferruh Yigit]

On 7/19/2017 3:46 PM, Daniel Mrzyglod wrote:
> This patch add support for different aggregator modes in similar manner
> that is provided in linux kernel.
> 
> testpmd> set bonding agg_mode  
> testpmd> show bonding config 
> 
> Signed-off-by: Daniel Mrzyglod 

I got this as it is, to be able to include code in rc2, but please send
testpmd documentation as already requested by Jingjing as separate patch.

Thanks,
ferruh

[dpdk-dev] [pull-request] next-net 17.08 RC2

[Ferruh Yigit]

The following changes since commit 838d6b775bfab436859764ea734cddaf95c7b5fd:

  mbuf: fix VXLAN port in comment (2017-07-19 09:17:35 +0300)

are available in the git repository at:

  http://dpdk.org/git/next/dpdk-next-net 

for you to fetch changes up to 283fd29236f4e4fde25cc9a1540a224ce0326ea1:

  test/bonding: add test case for agg selection in mode4 (2017-07-19 16:14:52 
+0100)


Beilei Xing (3):
  net/i40e: check invalid VF queue id for FDIR
  net/i40e: fix virtchnl message response timeout
  net/i40e: fix ethertype filter issue for new FW

Daniel Mrzyglod (3):
  net/bonding: add other aggregator modes
  app/testpmd: support different aggregation modes
  test/bonding: add test case for agg selection in mode4

Gaetan Rivet (11):
  ethdev: add deferred intermediate device state
  ethdev: count devices consistently
  net/failsafe: add fail-safe PMD
  net/failsafe: add plug-in support
  net/failsafe: add flexible device definition
  net/failsafe: support flow API
  net/failsafe: support Rx offload capabilities
  net/failsafe: add fast burst functions
  net/failsafe: support device removal
  net/failsafe: support link status change event
  net/failsafe: support flow API isolation mode

Hemant Agrawal (1):
  net/dpaa2: add support for secondary process attach

Ivan Malov (2):
  net/sfc: request MAC stats upload immediately on port start
  net/sfc: add support for xstats retrieval by ID

Jeff Guo (1):
  net/i40e: fix VF add/del MAC

Jerin Jacob (1):
  net/thunderx: update sq config register field

Jingjing Wu (3):
  net/ixgbe: fix LSC interrupt
  net/i40e: fix LSC interrupt
  net/e1000: fix LSC interrupt

Qi Zhang (4):
  net/i40e: fix VF Tx bytes
  net/ixgbe: fix drop action for signature match
  net/ixgbe: fix ipv6 flow create limitation for x550
  doc: add known issue for i40e VF performance

Qiming Yang (1):
  net/ixgbe: fix Rx/Tx queue interrupt for x550 devices

Rasesh Mody (1):
  doc: update supported architectures for QEDE PMD

Raslan Darawsheh (1):
  net/tap: remove Linux version check

Thomas Monjalon (2):
  net/tap: add missing newlines in logs
  net/ring: add missing newlines in logs

Tiwei Bie (3):
  net/i40e: remove an unnecessary goto
  net/ixgbe: remove an unnecessary goto
  net/virtio: refactor coding style

Wei Zhao (2):
  net/ixgbe: add support 82599ES SCTP packet drop action
  net/ixgbe: add queue index check in filter

Yongseok Koh (1):
  net/mlx5: change start pointer of compressed completion

 MAINTAINERS   |   5 +
 app/test-pmd/cmdline.c|  94 ++-
 config/common_base|   5 +
 doc/guides/nics/fail_safe.rst | 221 ++
 doc/guides/nics/features/failsafe.ini |  26 +
 doc/guides/nics/features/qede.ini |   2 +
 doc/guides/nics/features/qede_vf.ini  |   2 +
 doc/guides/nics/i40e.rst  |  27 +
 doc/guides/nics/index.rst |   1 +
 drivers/net/Makefile  |   2 +
 drivers/net/bonding/rte_eth_bond_8023ad.c | 194 -
 drivers/net/bonding/rte_eth_bond_8023ad.h |  32 +
 drivers/net/bonding/rte_eth_bond_8023ad_private.h |   1 +
 drivers/net/bonding/rte_eth_bond_args.c   |  33 +
 drivers/net/bonding/rte_eth_bond_pmd.c|  19 +-
 drivers/net/bonding/rte_eth_bond_private.h|   5 +
 drivers/net/bonding/rte_eth_bond_version.map  |   5 +
 drivers/net/dpaa2/dpaa2_ethdev.c  |  14 +-
 drivers/net/e1000/igb_ethdev.c|  15 +-
 drivers/net/failsafe/Makefile |  62 ++
 drivers/net/failsafe/failsafe.c   | 299 
 drivers/net/failsafe/failsafe_args.c  | 468 
 drivers/net/failsafe/failsafe_eal.c   | 118 +++
 drivers/net/failsafe/failsafe_ether.c | 437 +++
 drivers/net/failsafe/failsafe_flow.c  | 244 ++
 drivers/net/failsafe/failsafe_ops.c   | 867 ++
 drivers/net/failsafe/failsafe_private.h   | 359 +
 drivers/net/failsafe/failsafe_rxtx.c  | 203 +
 drivers/net/failsafe/rte_pmd_failsafe_version.map |   4 +
 drivers/net/i40e/i40e_ethdev.c|  26 +-
 drivers/net/i40e/i40e_ethdev_vf.c |  22 +-
 drivers/net/i40e/i40e_flow.c  |   5 +-
 drivers/net/i40e/i40e_tm.c|   2 -
 drivers/net/ixgbe/ixgbe_ethdev.c  |  19 +-
 drivers/net/ixgbe/ixgbe_fdir.c|   4 +-
 drivers/net/ixgbe/ixgbe_flow.c|  72 +-
 drivers/net/ixgbe/ixgbe_tm.c  |   2 -
 drivers/net/mlx5/mlx5_rxtx.c  

Re: [dpdk-dev] [PATCH] net/virtio: fix fail to configure rxq interrupt

[Tan, Jianfeng]

On 7/19/2017 4:08 AM, Yuanhan Liu wrote:

On Wed, Jul 19, 2017 at 11:18:23AM +0800, Jiayu Hu wrote:

When use rte_eth_dev_configure() to enable rx queue interrupt for virtio
devices, virtio_configure_intr() isn't called to set up the interrupt
environment, which causes rx queue interrupt setup failed. This patch is
to fix this issue.

Hmm, how was this supposed to work in the begining when this feature
was introduced? Jianfeng?


In the v17.05, virtio_configure_intr() will be called by 
virtio_dev_configure() if the request feature is changed. That's not 
correct.





--yliu



Fixes: 26b683b4f7d0 ("net/virtio: setup Rx queue interrupts")
Cc: sta...@dpdk.org

Signed-off-by: Jiayu Hu 
---
  drivers/net/virtio/virtio_ethdev.c | 7 +++
  1 file changed, 7 insertions(+)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index 00a3122..66656ed 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1688,6 +1688,13 @@ virtio_dev_configure(struct rte_eth_dev *dev)
return -ENOTSUP;
}
  
+	if (dev->data->dev_conf.intr_conf.rxq) {

+   if (virtio_configure_intr(dev) < 0) {
+   PMD_DRV_LOG(ERR, "failed to configure interrupt");
+   return -ENOTSUP;
+   }
+   }
+


Hi Jiayu,

I would expect this will not work for virtio pci devices (QEMU might 
crash). Could you double check that?


Thanks,
Jianfeng


if (dev->data->dev_flags & RTE_ETH_DEV_INTR_LSC)
/* Enable vector (0) for Link State Intrerrupt */
if (VTPCI_OPS(hw)->set_config_irq(hw, 0) ==
--
2.7.4

Re: [dpdk-dev] [PATCH v1] metrics: fix missing NULL termination

[Thomas Monjalon]

18/07/2017 12:43, Remy Horton:
> The public API (struct rte_metric_name) includes the NULL terminator
> byte in RTE_METRICS_MAX_NAME_LENGTH but the library itself internally
> excludes it. This makes it possible for an application to receive an
> unterminated name string. Fix be enforcing the NULL termination of all
> name strings to the length that the public API expects.
> 
> Fixes: 349950ddb9c5 ("metrics: add information metrics library")
> 
> Signed-off-by: Remy Horton 

Applied, thanks

Re: [dpdk-dev] [PATCH v2] igb_uio: issue FLR during open and release of device file

[Gregory Etelson]

On Wednesday, 19 July 2017 16:32:34 IDT Ferruh Yigit wrote:
> On 7/12/2017 4:40 AM, Tan, Jianfeng wrote:
> > 
> > 
> >> -Original Message-
> >> From: Shijith Thotton [mailto:shijith.thot...@caviumnetworks.com]
> >> Sent: Friday, July 7, 2017 7:14 PM
> >> To: dev@dpdk.org
> >> Cc: Yigit, Ferruh; Gregory Etelson; Thomas Monjalon; Stephen Hemminger;
> >> Tan, Jianfeng; Lu, Wenzhuo
> >> Subject: [PATCH v2] igb_uio: issue FLR during open and release of device 
> >> file
> >>
> >> Set UIO info device file operations open and release. Call pci reset
> >> function inside open and release to clear device state at start and end.
> >> Copied this behaviour from vfio_pci kernel module code. With this patch,
> >> it is not mandatory to issue FLR by PMD's during init and close.
> >>
> >> Bus master enable and disable are added in open and release respectively
> >> to take care of device DMA.
> >>
> >> Signed-off-by: Shijith Thotton 
> > 
> > Reviewed-by: Jianfeng Tan 
> 
> Acked-by: Ferruh Yigit 
> 

Acked-by: Gregory Etelson 

Re: [dpdk-dev] Rx Can't receive anymore packet after received 1.5 billion packet.

[Dumitrescu, Cristian]

> -Original Message-
> From: vuon...@viettel.com.vn [mailto:vuon...@viettel.com.vn]
> Sent: Tuesday, July 18, 2017 2:37 AM
> To: Dumitrescu, Cristian 
> Cc: us...@dpdk.org; dev@dpdk.org
> Subject: Re: [dpdk-dev] Rx Can't receive anymore packet after received 1.5
> billion packet.
> 
> 
> 
> On 07/17/2017 05:31 PM, cristian.dumitre...@intel.com wrote:
> >
> >> -Original Message-
> >> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of
> >> vuon...@viettel.com.vn
> >> Sent: Monday, July 17, 2017 3:04 AM
> >> Cc: us...@dpdk.org; dev@dpdk.org
> >> Subject: [dpdk-dev] Rx Can't receive anymore packet after received 1.5
> >> billion packet.
> >>
> >> Hi DPDK team,
> >> Sorry when I send this email to both of group users and dev. But I have
> >> big problem: Rx core on my application can not receive anymore packet
> >> after I did the stress test to it (~1 day Rx core received ~ 1.5 billion
> >> packet). Rx core still alive but didn't receive any packet and didn't
> >> generate any log. Below is my system configuration:
> >> - OS: CentOS 7
> >> - Kernel: 3.10.0-514.16.1.el7.x86_64
> >> - Huge page: 32G: 16384 page 2M
> >> - NIC card: Intel 85299
> >> - DPDK version: 16.11
> >> - Architecture: Rx (lcore 1) received packet then queue to the ring
> >> - Worker (lcore 2) dequeue packet in the ring and free it (use
> >> rte_pktmbuf_free() function).
> >> - Mempool create: rte_pktmbuf_pool_create (
> >>"rx_pool",  /*
> >> name */
> >>8192, /*
> >> number of elemements in the mbuf pool */
> >> 256,/* Size of per-core
> >> object cache */
> >> 0, /* Size of
> >> application private are between rte_mbuf struct and data buffer */
> >>RTE_MBUF_DEFAULT_BUF_SIZE, /*
> >> Size of data buffer in each mbuf (2048 + 128)*/
> >> 0   /* socket id */
> >>   );
> >> If I change "number of elemements in the mbuf pool" from 8192 to 512,
> Rx
> >> have same problem after shorter time (~ 30s).
> >>
> >> Please tell me if you need more information. I am looking forward to
> >> hearing from you.
> >>
> >>
> >> Many thanks,
> >> Vuong Le
> > Hi Vuong,
> >
> > This is likely to be a buffer leakage problem. You might have a path in your
> code where you are not freeing a buffer and therefore this buffer gets
> "lost", as the application is not able to use this buffer any more since it 
> is not
> returned back to the pool, so the pool of free buffers shrinks over time up to
> the moment when it eventually becomes empty, so no more packets can be
> received.
> >
> > You might want to periodically monitor the numbers of free buffers in your
> pool; if this is the root cause, then you should be able to see this number
> constantly decreasing until it becomes flat zero, otherwise you should be
> able to the number of free buffers oscillating around an equilibrium point.
> >
> > Since it takes a relatively big number of packets to get to this issue, it 
> > is
> likely that the code path that has this problem is not executed very
> frequently: it might be a control plane packet that is not freed up, or an ARP
> request/reply pkt, etc.
> >
> > Regards,
> > Cristian
> Hi Cristian,
> Thanks for your response, I am doing your ideal. But let me show you
> another case i have tested before. I changed architecture of my
> application as below:
> - Architecture: Rx (lcore 1) received packet then queue to the ring
> - after that: Rx (lcore 1) dequeue packet in the ring and free it
> immediately.
> (old architecture as above)
> With new architecture Rx still receive packet after 2 day and everything
> look good.  Unfortunately, My application must run in old architecture.
> 
> Any ideal for me?
> 
> 
> Many thanks,
> Vuong Le

I am not sure I understand the old architecture and the new architecture you 
are referring to, can you please clarify them.

Regards,
Cristian

Re: [dpdk-dev] [PATCH v5] examples/vhost: introduce a new vhost-user-scsi sample application

[Yuanhan Liu]

On Thu, Jul 20, 2017 at 05:16:29PM +0800, Changpeng Liu wrote:
> vhost-user protocol is common to many virtio devices, such as
> virtio_net/virtio_scsi/virtio_blk. Since DPDK vhost library
> removed the NET specific data structures, the vhost library
> is common to other virtio devices, such as virtio-scsi.
> 
> Here we introduce a simple memory based block device that
> can be presented to Guest VM through vhost-user-scsi-pci
> controller. Similar with vhost-net, the sample application
> will process the I/Os sent via virt rings.
> 
> Signed-off-by: Changpeng Liu 

Applied to dpdk-next-virtio.

--yliu

[dpdk-dev] [PATCH] jobstats: fix a typo in rte_jobstats.h.

[Rami Rosen]

This patch fixes a trivial typo in rte_jobstats.h.

Signed-off-by: Rami Rosen 
---
 lib/librte_jobstats/rte_jobstats.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/librte_jobstats/rte_jobstats.h 
b/lib/librte_jobstats/rte_jobstats.h
index b3686030e820..7e76fd50c6fa 100644
--- a/lib/librte_jobstats/rte_jobstats.h
+++ b/lib/librte_jobstats/rte_jobstats.h
@@ -98,7 +98,7 @@ struct rte_jobstats {
 } __rte_cache_aligned;
 
 struct rte_jobstats_context {
-   /** Viariable holding time at different points:
+   /** Variable holding time at different points:
 * -# loop start time if loop was started but no job executed yet.
 * -# job start time if job is currently executing.
 * -# job finish time if job finished its execution.
-- 
2.7.4

Re: [dpdk-dev] Occasional instability in RSS Hashes/Queues from X540 NIC

[Li, Xiaoyun]

Hi, Yuanhan !
I did the same tests on DPDK 16.11 this morning and cannot reproduce the 
problem on 16.11. It seems that the problem has been fixed on version 16.11.
So could you reproduce the problem on 16.11 LTS release? I think maybe it has 
been fixed already.

Best regards
Xiaoyun Li

-Original Message-
From: Yuanhan Liu [mailto:y...@fridaylinux.org] 
Sent: Tuesday, July 18, 2017 22:01
To: Matt Laswell 
Cc: Yang, Qiming ; dev@dpdk.org; Li, Xiaoyun 

Subject: Re: [dpdk-dev] Occasional instability in RSS Hashes/Queues from X540 
NIC

On Tue, Jul 18, 2017 at 08:43:42AM -0500, Matt Laswell wrote:
> Hi Qiming,
> 
> That's fantastic news.  Thank you very much for taking the time to 
> figure the issue out.
> 
> Would it be possible to backport the fix to the 16.11 LTS release?   This
> kind of problem seems tailor-made for LTS.

Agreed. You might want to find the fix commit with git bisect.

--yliu

Re: [dpdk-dev] [PATCH] doc/metrics: fix variable undefined error

[Thomas Monjalon]

18/07/2017 16:10, Remy Horton:
> On 18/07/2017 13:03, Yong Wang wrote:
> > Signed-off-by: Yong Wang 
> 
> Acked-by: Remy Horton 

Applied, thanks

Re: [dpdk-dev] [PATCH v4 0/8] fix hotplug API

[Thomas Monjalon]

15/07/2017 20:56, Gaetan Rivet:
> Sending those fixes as separate patches as they stand on their own.
> This series improves usability of the hotplug API and fixes a few issues
> with existing implementations.
> 
> The hotplug API can be tested with the fail-safe PMD[1]. Its
> documentation describes how to declare slaves and how to use it.
[...]
> Gaetan Rivet (8):
>   vdev: implement plug operation
>   devargs: introduce removal function
>   devargs: introduce insert function
>   eal: fix hotplug add / remove
>   pci: use given name as generic name
>   pci: fix generic driver pointer on probe error
>   pci: fix hotplug operations
>   bus: remove useless plug parameter

Applied to fix the current bus state, thanks.
It is now clear that it can be done differently in 17.11 after
the deprecation of the device specification syntax.
Jan already sent some patches in this direction.

Re: [dpdk-dev] [PATCH] devargs: restore rte_devtype API

[Thomas Monjalon]

15/07/2017 20:59, Gaetan Rivet:
> Revert "devargs: make device types generic"
> 
> This commit broke the rte_devargs API by changing the meaning of
> the rte_devtype enum.
> 
> Restore the previous API, unit tests and function calls.
> Introduce parallel enum that acts as translation between previous API
> and current structures.
> 
> Restoring the previous API means that -w and -b are not usable anymore
> with any bus having implemented the "parse" operation. Only PCI devices
> can be used with -w and -b, virtual devices are declared using vdev.
> 
> This (partially) reverts commit bd279a79366f50a4893fb84db91bbf64b56f9fb1.
> 
> Signed-off-by: Gaetan Rivet 
> ---
> 
> As correctly pointed out by Jan Blunck in [1], the previous change breaks an 
> existing API.
> This commit only deals with this API. Doing so however, restore the previous 
> behavior as well.
> This means that the devargs behavior is iso-functional with that of the 
> v17.05.
> It is possible, however, to explicitly set the bus in a device declaration.
> 
> [1]: http://dpdk.org/ml/archives/dev/2017-July/071318.html

Applied, thanks

Re: [dpdk-dev] [PATCH] jobstats: fix a typo in rte_jobstats.h.

[Thomas Monjalon]

19/07/2017 23:06, Rami Rosen:
> This patch fixes a trivial typo in rte_jobstats.h.
> 
> Signed-off-by: Rami Rosen 

Applied, thanks

[dpdk-dev] [PATCH v2] net/ixgbe: net/ixgbe: fix sctp port support limitation

[Qi Zhang]

Only x550 family support sctp port in fdir filter, so
add this limiation when parse consistent API.

Fixes: 11777435c727 ("net/ixgbe: parse flow director filter")
Cc: sta...@dpdk.org

Signed-off-by: Qi Zhang 
---
v2:
- rebase to next-net

 drivers/net/ixgbe/ixgbe_flow.c | 57 --
 1 file changed, 44 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ixgbe/ixgbe_flow.c b/drivers/net/ixgbe/ixgbe_flow.c
index b34835a..7a5c3f9 100644
--- a/drivers/net/ixgbe/ixgbe_flow.c
+++ b/drivers/net/ixgbe/ixgbe_flow.c
@@ -1387,7 +1387,8 @@ static inline uint8_t signature_match(const struct 
rte_flow_item pattern[])
  * Item->last should be NULL.
  */
 static int
-ixgbe_parse_fdir_filter_normal(const struct rte_flow_attr *attr,
+ixgbe_parse_fdir_filter_normal(struct rte_eth_dev *dev,
+  const struct rte_flow_attr *attr,
   const struct rte_flow_item pattern[],
   const struct rte_flow_action actions[],
   struct ixgbe_fdir_rule *rule,
@@ -1410,9 +1411,10 @@ ixgbe_parse_fdir_filter_normal(const struct 
rte_flow_attr *attr,
const struct rte_flow_item_vlan *vlan_mask;
const struct rte_flow_item_raw *raw_mask;
const struct rte_flow_item_raw *raw_spec;
-
uint8_t j;
 
+   struct ixgbe_hw *hw = IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
if (!pattern) {
rte_flow_error_set(error, EINVAL,
RTE_FLOW_ERROR_TYPE_ITEM_NUM,
@@ -1910,7 +1912,21 @@ ixgbe_parse_fdir_filter_normal(const struct 
rte_flow_attr *attr,
return -rte_errno;
}
 
-   if (item->mask) {
+   /* only x550 family only support sctp port */
+   if (hw->mac.type == ixgbe_mac_X550 ||
+   hw->mac.type == ixgbe_mac_X550EM_x ||
+   hw->mac.type == ixgbe_mac_X550EM_a) {
+   /**
+* Only care about src & dst ports,
+* others should be masked.
+*/
+   if (!item->mask) {
+   memset(rule, 0, sizeof(struct ixgbe_fdir_rule));
+   rte_flow_error_set(error, EINVAL,
+   RTE_FLOW_ERROR_TYPE_ITEM,
+   item, "Not supported by fdir filter");
+   return -rte_errno;
+   }
rule->b_mask = TRUE;
sctp_mask =
(const struct rte_flow_item_sctp *)item->mask;
@@ -1924,21 +1940,36 @@ ixgbe_parse_fdir_filter_normal(const struct 
rte_flow_attr *attr,
}
rule->mask.src_port_mask = sctp_mask->hdr.src_port;
rule->mask.dst_port_mask = sctp_mask->hdr.dst_port;
-   }
 
-   if (item->spec) {
-   rule->b_spec = TRUE;
-   sctp_spec =
+   if (item->spec) {
+   rule->b_spec = TRUE;
+   sctp_spec =
(const struct rte_flow_item_sctp *)item->spec;
-   rule->ixgbe_fdir.formatted.src_port =
-   sctp_spec->hdr.src_port;
-   rule->ixgbe_fdir.formatted.dst_port =
-   sctp_spec->hdr.dst_port;
+   rule->ixgbe_fdir.formatted.src_port =
+   sctp_spec->hdr.src_port;
+   rule->ixgbe_fdir.formatted.dst_port =
+   sctp_spec->hdr.dst_port;
+   }
+   /* others even sctp port is not supported */
+   } else {
+   sctp_mask =
+   (const struct rte_flow_item_sctp *)item->mask;
+   if (sctp_mask &&
+   (sctp_mask->hdr.src_port ||
+sctp_mask->hdr.dst_port ||
+sctp_mask->hdr.tag ||
+sctp_mask->hdr.cksum)) {
+   memset(rule, 0, sizeof(struct ixgbe_fdir_rule));
+   rte_flow_error_set(error, EINVAL,
+   RTE_FLOW_ERROR_TYPE_ITEM,
+   item, "Not supported by fdir filter");
+   return -rte_errno;
+   }
}
 
item = next_no_fuzzy_pattern(pattern, item);
if (item->type != RTE_FLOW_ITEM_TYPE_RAW &&
-   item->type != RTE_FLOW_ITEM_TYPE_END) {
+   item->type != RTE_FLOW_ITEM_TYPE_END) {

[dpdk-dev] [pull-request] next-eventdev 17.08 rc2

[Jerin Jacob]

The following changes since commit da94a999d3eae6a4c565f2a23c3ab303909e9b53:

  examples/vhost_scsi: introduce a new sample app (2017-07-19 22:49:47 +0300)

are available in the git repository at:

  http://dpdk.org/git/next/dpdk-next-eventdev 

for you to fetch changes up to cb78ef9859ff68b9e260a844f151055ebd9ae78f:

  eventdev: fix memory realloc check in port config (2017-07-20 09:49:47 +0530)


Harry van Haaren (1):
  eventdev: fix memory realloc check in port config

Nipun Gupta (1):
  event/dpaa2: advertise the burst mode capability

 drivers/event/dpaa2/dpaa2_eventdev.c | 3 ++-
 lib/librte_eventdev/rte_eventdev.c   | 2 +-
 2 files changed, 3 insertions(+), 2 deletions(-)

[dpdk-dev] only two of the four queues work on the VF interface of Intel X553 nic.

[Bill Bonaparte]

Hi all,
I encounter a problem that there are two of the four queues working
for dpdk-apps when using the VF interface of intel x553 10G-ethernet NIC.
Straightly speaking, there are no packets coming into the last two queues,
no matter what packets I send to the dpdk-apps.
I checked the initialization of the queues,  it shows  that all
four queues are initialized successfully.
If I use the PF directly for dpdk-apps, there is no such problem.
If I missed some configuration when using the VF ?
if so, what should I do ?

I am appreciate to get some  discussion from you.

[dpdk-dev] [PATCH 4/8] net/bnxt: fix set link config

[Ajit Khaparde]

remove the unnecessary rte_delay in bnxt_set_hwrm_link_config

Fixes: 7bc8e9a227cc ("net/bnxt: support async link notification")

Signed-off-by: Ajit Khaparde 
---
 drivers/net/bnxt/bnxt_hwrm.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index 2c66092..7cd4978 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -1971,7 +1971,6 @@ int bnxt_set_hwrm_link_config(struct bnxt *bp, bool 
link_up)
"Set link config failed with rc %d\n", rc);
}
 
-   rte_delay_ms(BNXT_LINK_WAIT_INTERVAL);
 error:
return rc;
 }
-- 
2.10.1 (Apple Git-78)

[dpdk-dev] [PATCH 2/8] net/bnxt: fix to avoid a segfault

[Ajit Khaparde]

Fix use of local variable to avoid segfault.
cnt was incorrectly tested and decremented in the loop that removes
a VLAN from the table.

Fixes: 36735a932ca7 ("support set VF QOS and MAC anti spoof")

Signed-off-by: Stephen Hurd 
Signed-off-by: Ajit Khaparde 
---
 drivers/net/bnxt/rte_pmd_bnxt.c | 17 +
 1 file changed, 9 insertions(+), 8 deletions(-)

diff --git a/drivers/net/bnxt/rte_pmd_bnxt.c b/drivers/net/bnxt/rte_pmd_bnxt.c
index 0a8fb1e..0d48873 100644
--- a/drivers/net/bnxt/rte_pmd_bnxt.c
+++ b/drivers/net/bnxt/rte_pmd_bnxt.c
@@ -500,6 +500,7 @@ int rte_pmd_bnxt_set_vf_vlan_filter(uint8_t port, uint16_t 
vlan,
continue;
}
 
+   /* cnt is one less than vlan_count */
cnt = bp->pf.vf_info[i].vlan_count++;
/*
 * And finally, add to the
@@ -511,19 +512,19 @@ int rte_pmd_bnxt_set_vf_vlan_filter(uint8_t port, 
uint16_t vlan,
ve->vid = rte_cpu_to_be_16(vlan);
}
} else {
-   for (j = 0; cnt; j++) {
+   for (j = 0; j < cnt; j++) {
if (rte_be_to_cpu_16(
-   bp->pf.vf_info[i].vlan_table[j].vid) !=
-   vlan)
+bp->pf.vf_info[i].vlan_table[j].vid) !=
+vlan)
continue;
memmove(
-&bp->pf.vf_info[i].vlan_table[j],
-&bp->pf.vf_info[i].vlan_table[j + 1],
-getpagesize() -
-((j + 1) *
+   &bp->pf.vf_info[i].vlan_table[j],
+  &bp->pf.vf_info[i].vlan_table[j + 1],
+   getpagesize() -
+   ((j + 1) *
 sizeof(struct bnxt_vlan_table_entry)));
j--;
-   cnt = bp->pf.vf_info[i].vlan_count--;
+   cnt = --bp->pf.vf_info[i].vlan_count;
}
}
rte_pmd_bnxt_set_vf_vlan_anti_spoof(dev->data->port_id,
-- 
2.10.1 (Apple Git-78)

[dpdk-dev] [PATCH 0/8] bnxt patchset

[Ajit Khaparde]

Hi,
This patch set fixes some of the issues found during testing.
Please apply.

Thanks

  net/bnxt: fix log levels for non error conditions.
  net/bnxt: fix to avoid a segfault
  net/bnxt: fix vnic cleanup
  net/bnxt: fix set link config
  net/bnxt: reset VF stats during initialization
  net/bnxt: fix VLAN antispoof configuration code
  net/bnxt: check invalid l2_filter_id
  net/bnxt: fix to free a filter before reusing it

 drivers/net/bnxt/bnxt.h|   7 +++
 drivers/net/bnxt/bnxt_cpr.c|   2 +-
 drivers/net/bnxt/bnxt_hwrm.c   |  76 ++-
 drivers/net/bnxt/bnxt_hwrm.h   |   3 +
 drivers/net/bnxt/hsi_struct_def_dpdk.h |  81 +++-
 drivers/net/bnxt/rte_pmd_bnxt.c| 110 ++---

-- 
2.10.1 (Apple Git-78)

[dpdk-dev] [PATCH 1/8] net/bnxt: fix log levels for non error conditions.

[Ajit Khaparde]

1) handle_async_event is a DEBUG level log message.
2) Log "Unable to get default VNIC for VF %d" at INFO level.

Fixes: 36735a932ca7 ("support set VF QOS and MAC anti spoof")

Signed-off-by: Stephen Hurd 
Signed-off-by: Ajit Khaparde 
---
 drivers/net/bnxt/bnxt_cpr.c | 2 +-
 drivers/net/bnxt/rte_pmd_bnxt.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/bnxt/bnxt_cpr.c b/drivers/net/bnxt/bnxt_cpr.c
index f369cf6..68979bc 100644
--- a/drivers/net/bnxt/bnxt_cpr.c
+++ b/drivers/net/bnxt/bnxt_cpr.c
@@ -58,7 +58,7 @@ void bnxt_handle_async_event(struct bnxt *bp,
bnxt_link_update_op(bp->eth_dev, 0);
break;
default:
-   RTE_LOG(ERR, PMD, "handle_async_event id = 0x%x\n", event_id);
+   RTE_LOG(DEBUG, PMD, "handle_async_event id = 0x%x\n", event_id);
break;
}
 }
diff --git a/drivers/net/bnxt/rte_pmd_bnxt.c b/drivers/net/bnxt/rte_pmd_bnxt.c
index dd04f58..0a8fb1e 100644
--- a/drivers/net/bnxt/rte_pmd_bnxt.c
+++ b/drivers/net/bnxt/rte_pmd_bnxt.c
@@ -333,7 +333,7 @@ int rte_pmd_bnxt_set_vf_vlan_anti_spoof(uint8_t port, 
uint16_t vf, uint8_t on)
 * This simply indicates there's no driver
 * loaded.  This is not an error.
 */
-   RTE_LOG(ERR, PMD,
+   RTE_LOG(INFO, PMD,
  "Unable to get default VNIC for VF %d\n",
vf);
} else {
-- 
2.10.1 (Apple Git-78)

[dpdk-dev] [PATCH 3/8] net/bnxt: fix vnic cleanup

[Ajit Khaparde]

Check if the vnic_id and rss_rule is not invalid before passing it
to the firmware to cleanup the VNIC. Log a message if the vnic_id
is invalid.

Fixes: db678d5c2b54 ("net/bnxt: add HWRM VNIC configure")

Signed-off-by: Ajit Khaparde 
---
 drivers/net/bnxt/bnxt_hwrm.c | 19 ++-
 1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index 3583ec7..2c66092 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -980,6 +980,7 @@ int bnxt_hwrm_vnic_alloc(struct bnxt *bp, struct 
bnxt_vnic_info *vnic)
HWRM_CHECK_RESULT;
 
vnic->fw_vnic_id = rte_le_to_cpu_16(resp->vnic_id);
+   RTE_LOG(DEBUG, PMD, "VNIC ID %x\n", vnic->fw_vnic_id);
return rc;
 }
 
@@ -1045,6 +1046,11 @@ int bnxt_hwrm_vnic_cfg(struct bnxt *bp, struct 
bnxt_vnic_info *vnic)
uint32_t ctx_enable_flag = HWRM_VNIC_CFG_INPUT_ENABLES_RSS_RULE;
struct bnxt_plcmodes_cfg pmodes;
 
+   if (vnic->fw_vnic_id == INVALID_HW_RING_ID) {
+   RTE_LOG(DEBUG, PMD, "VNIC ID %x\n", vnic->fw_vnic_id);
+   return rc;
+   }
+
rc = bnxt_hwrm_vnic_plcmodes_qcfg(bp, vnic, &pmodes);
if (rc)
return rc;
@@ -1103,6 +1109,10 @@ int bnxt_hwrm_vnic_qcfg(struct bnxt *bp, struct 
bnxt_vnic_info *vnic,
struct hwrm_vnic_qcfg_input req = {.req_type = 0 };
struct hwrm_vnic_qcfg_output *resp = bp->hwrm_cmd_resp_addr;
 
+   if (vnic->fw_vnic_id == INVALID_HW_RING_ID) {
+   RTE_LOG(DEBUG, PMD, "VNIC QCFG ID %d\n", vnic->fw_vnic_id);
+   return rc;
+   }
HWRM_PREP(req, VNIC_QCFG, -1, resp);
 
req.enables =
@@ -1149,6 +1159,7 @@ int bnxt_hwrm_vnic_ctx_alloc(struct bnxt *bp, struct 
bnxt_vnic_info *vnic)
HWRM_CHECK_RESULT;
 
vnic->rss_rule = rte_le_to_cpu_16(resp->rss_cos_lb_ctx_id);
+   RTE_LOG(DEBUG, PMD, "VNIC RSS Rule %x\n", vnic->rss_rule);
 
return rc;
 }
@@ -1160,6 +1171,10 @@ int bnxt_hwrm_vnic_ctx_free(struct bnxt *bp, struct 
bnxt_vnic_info *vnic)
struct hwrm_vnic_rss_cos_lb_ctx_free_output *resp =
bp->hwrm_cmd_resp_addr;
 
+   if (vnic->rss_rule == 0x) {
+   RTE_LOG(DEBUG, PMD, "VNIC RSS Rule %x\n", vnic->rss_rule);
+   return rc;
+   }
HWRM_PREP(req, VNIC_RSS_COS_LB_CTX_FREE, -1, resp);
 
req.rss_cos_lb_ctx_id = rte_cpu_to_le_16(vnic->rss_rule);
@@ -1179,8 +1194,10 @@ int bnxt_hwrm_vnic_free(struct bnxt *bp, struct 
bnxt_vnic_info *vnic)
struct hwrm_vnic_free_input req = {.req_type = 0 };
struct hwrm_vnic_free_output *resp = bp->hwrm_cmd_resp_addr;
 
-   if (vnic->fw_vnic_id == INVALID_HW_RING_ID)
+   if (vnic->fw_vnic_id == INVALID_HW_RING_ID) {
+   RTE_LOG(DEBUG, PMD, "VNIC FREE ID %x\n", vnic->fw_vnic_id);
return rc;
+   }
 
HWRM_PREP(req, VNIC_FREE, -1, resp);
 
-- 
2.10.1 (Apple Git-78)

[dpdk-dev] [PATCH 5/8] net/bnxt: reset VF stats during initialization

[Ajit Khaparde]

This patch resets the VF stats during initialization

Fixes: b7778e8a1c00 ("net/bnxt: refactor to properly allocate resources for
 PF/VF")

Signed-off-by: Ajit Khaparde 
---
 drivers/net/bnxt/bnxt_hwrm.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index 7cd4978..231f06b 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -2306,6 +2306,7 @@ int bnxt_hwrm_allocate_vfs(struct bnxt *bp, int num_vfs)
 
reserve_resources_from_vf(bp, &req, i);
bp->pf.active_vfs++;
+   bnxt_hwrm_func_clr_stats(bp, bp->pf.vf_info[i].fid);
}
 
/*
-- 
2.10.1 (Apple Git-78)

[dpdk-dev] [PATCH 7/8] net/bnxt: check invalid l2_filter_id

[Ajit Khaparde]

Add code to check for invalid filter_id in bnxt_hwrm_clear_filter

Fixes: f92735db1e4c ("net/bnxt: add L2 filter alloc/init/free")

Signed-off-by: Ajit Khaparde 
---
 drivers/net/bnxt/bnxt_hwrm.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index a82cc81..e230b46 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -327,6 +327,9 @@ int bnxt_hwrm_clear_filter(struct bnxt *bp,
struct hwrm_cfa_l2_filter_free_input req = {.req_type = 0 };
struct hwrm_cfa_l2_filter_free_output *resp = bp->hwrm_cmd_resp_addr;
 
+   if (filter->fw_l2_filter_id == UINT64_MAX)
+   return 0;
+
HWRM_PREP(req, CFA_L2_FILTER_FREE, -1, resp);
 
req.l2_filter_id = rte_cpu_to_le_64(filter->fw_l2_filter_id);
-- 
2.10.1 (Apple Git-78)

[dpdk-dev] [PATCH 6/8] net/bnxt: fix VLAN antispoof configuration code

[Ajit Khaparde]

We are wrongly using a Rx side HWRM command set_rx_mask to configure
VLAN anti-spoof. This being a Tx side feature, this patch
tries to fix it.

Since the HWRM command to do it ringt is available only in
the newer firmware versions, the patch verifies the firmware
version before attempting to send the HWRM command to
the firmware.

Fixes: 36735a932ca7 ("support set VF QOS and MAC anti spoof")

Signed-off-by: Stephen Hurd 
Signed-off-by: Ajit Khaparde 
---
 drivers/net/bnxt/bnxt.h|  7 +++
 drivers/net/bnxt/bnxt_hwrm.c   | 49 ++
 drivers/net/bnxt/bnxt_hwrm.h   |  3 ++
 drivers/net/bnxt/hsi_struct_def_dpdk.h | 81 -
 drivers/net/bnxt/rte_pmd_bnxt.c| 93 +++---
 5 files changed, 203 insertions(+), 30 deletions(-)

diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index ec1aad9..405d94d 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -97,9 +97,16 @@ struct bnxt_vlan_table_entry {
uint16_tvid;
 } __attribute__((packed));
 
+struct bnxt_vlan_antispoof_table_entry {
+   uint16_ttpid;
+   uint16_tvid;
+   uint16_tmask;
+} __attribute__((packed));
+
 struct bnxt_child_vf_info {
void*req_buf;
struct bnxt_vlan_table_entry*vlan_table;
+   struct bnxt_vlan_antispoof_table_entry  *vlan_as_table;
STAILQ_HEAD(, bnxt_filter_info) filter;
uint32_tfunc_cfg_flags;
uint32_tl2_rx_mask;
diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index 231f06b..a82cc81 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -282,6 +282,44 @@ int bnxt_hwrm_cfa_l2_set_rx_mask(struct bnxt *bp,
return rc;
 }
 
+int bnxt_hwrm_cfa_vlan_antispoof_cfg(struct bnxt *bp, uint16_t fid,
+   uint16_t vlan_count,
+   struct bnxt_vlan_antispoof_table_entry *vlan_table)
+{
+   int rc = 0;
+   struct hwrm_cfa_vlan_antispoof_cfg_input req = {.req_type = 0 };
+   struct hwrm_cfa_vlan_antispoof_cfg_output *resp =
+   bp->hwrm_cmd_resp_addr;
+
+   /*
+* Older HWRM versions did not support this command, and the set_rx_mask
+* list was used for anti-spoof. In 1.8.0, the TX path configuration was
+* removed from set_rx_mask call, and this command was added.
+*
+* This command is also present from 1.7.8.11 and higher,
+* as well as 1.7.8.0
+*/
+   if (bp->fw_ver < ((1 << 24) | (8 << 16))) {
+   if (bp->fw_ver != ((1 << 24) | (7 << 16) | (8 << 8))) {
+   if (bp->fw_ver < ((1 << 24) | (7 << 16) | (8 << 8) |
+   (11)))
+   return 0;
+   }
+   }
+   HWRM_PREP(req, CFA_VLAN_ANTISPOOF_CFG, -1, resp);
+   req.fid = rte_cpu_to_le_16(fid);
+
+   req.vlan_tag_mask_tbl_addr =
+   rte_cpu_to_le_64(rte_mem_virt2phy(vlan_table));
+   req.num_vlan_entries = rte_cpu_to_le_32((uint32_t)vlan_count);
+
+   rc = bnxt_hwrm_send_message(bp, &req, sizeof(req));
+
+   HWRM_CHECK_RESULT;
+
+   return rc;
+}
+
 int bnxt_hwrm_clear_filter(struct bnxt *bp,
   struct bnxt_filter_info *filter)
 {
@@ -389,6 +427,17 @@ int bnxt_hwrm_func_qcaps(struct bnxt *bp)
else
rte_mem_lock_page(
bp->pf.vf_info[i].vlan_table);
+   bp->pf.vf_info[i].vlan_as_table =
+   rte_zmalloc("VF VLAN AS table",
+   getpagesize(),
+   getpagesize());
+   if (bp->pf.vf_info[i].vlan_as_table == NULL)
+   RTE_LOG(ERR, PMD,
+   "Alloc VLAN AS table for VF %d fail\n",
+   i);
+   else
+   rte_mem_lock_page(
+  bp->pf.vf_info[i].vlan_as_table);
STAILQ_INIT(&bp->pf.vf_info[i].filter);
}
}
diff --git a/drivers/net/bnxt/bnxt_hwrm.h b/drivers/net/bnxt/bnxt_hwrm.h
index dd0999a..51cd0dd 100644
--- a/drivers/net/bnxt/bnxt_hwrm.h
+++ b/drivers/net/bnxt/bnxt_hwrm.h
@@ -48,6 +48,9 @@ int bnxt_hwrm_cfa_l2_clear_rx_mask(struct bnxt *bp,
 int bnxt_hwrm_cfa_l2_set_rx_mask(struct bnxt *bp, struct bnxt_vnic_info *vnic,
 uint16_t vlan_count,
 struct bnxt_vlan_table_entry *vlan_table);
+int b

[dpdk-dev] [PATCH 8/8] net/bnxt: fix to free a filter before reusing it

[Ajit Khaparde]

This patch sends the HWRM command to free a filter in the hardware,
before using it again.

Fixes: f92735db1e4c ("net/bnxt: add L2 filter alloc/init/free")

Signed-off-by: Ajit Khaparde 
---
 drivers/net/bnxt/bnxt_hwrm.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index e230b46..4b1810c 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -352,6 +352,9 @@ int bnxt_hwrm_set_filter(struct bnxt *bp,
struct hwrm_cfa_l2_filter_alloc_output *resp = bp->hwrm_cmd_resp_addr;
uint32_t enables = 0;
 
+   if (filter->fw_l2_filter_id != UINT64_MAX)
+   bnxt_hwrm_clear_filter(bp, filter);
+
HWRM_PREP(req, CFA_L2_FILTER_ALLOC, -1, resp);
 
req.flags = rte_cpu_to_le_32(filter->flags);
-- 
2.10.1 (Apple Git-78)

Re: [dpdk-dev] [PATCH] all: refactor coding style

[Shreyansh Jain]

On Wednesday 19 July 2017 02:36 PM, Tiwei Bie wrote:

Remove the unwanted spaces before `;' across DPDK source code
by below one-liner with some minor manual refinements.

find . -name '*.[ch]' | xargs sed -i 's/\([^;(]\) \+;/\1;/g'

The fixes for cmdline library are skipped, because it has a
different coding style. It deserves a separate cleanup if
necessary. The fixes for drivers' base code are also skipped
to keep the base code intact.

Signed-off-by: Tiwei Bie 
---
  app/test-pmd/testpmd.h |  4 ++--
  drivers/crypto/qat/qat_adf/icp_qat_fw.h|  2 +-
  drivers/event/dpaa2/dpaa2_eventdev.c   |  2 +-
  drivers/mempool/dpaa2/dpaa2_hw_mempool.c   |  2 +-
  drivers/net/bnx2x/bnx2x.c  |  3 ++-
  drivers/net/bnx2x/elink.h  |  2 +-
  drivers/net/e1000/igb_pf.c |  2 +-
  drivers/net/ena/ena_ethdev.c   |  4 ++--
  drivers/net/qede/qede_ethdev.c |  2 +-
  drivers/net/vhost/rte_eth_vhost.c  |  2 +-
  drivers/net/virtio/virtio_rxtx.c   |  4 ++--
  drivers/net/xenvirt/rte_eth_xenvirt.c  |  4 ++--
  drivers/net/xenvirt/rte_xen_lib.c  |  2 +-
  drivers/net/xenvirt/virtqueue.h|  2 +-
  examples/ip_pipeline/cpu_core_map.c|  4 ++--
  examples/multi_process/l2fwd_fork/main.c   |  2 +-
  examples/netmap_compat/lib/compat_netmap.c |  2 +-
  examples/performance-thread/l3fwd-thread/main.c|  2 +-
  examples/qos_sched/app_thread.c|  2 +-
  examples/quota_watermark/qw/main.c |  2 +-
  examples/vhost_xen/xenstore_parse.c|  3 +--
  lib/librte_distributor/rte_distributor.c   | 12 +-
  lib/librte_eal/linuxapp/eal/eal_memory.c   |  2 +-
  lib/librte_eal/linuxapp/eal/eal_xen_memory.c   |  2 +-
  lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c |  4 ++--
  .../linuxapp/kni/ethtool/ixgbe/ixgbe_main.c|  2 +-
  .../linuxapp/kni/ethtool/ixgbe/kcompat.c   |  2 +-
  lib/librte_ether/rte_ethdev.c  |  2 +-
  lib/librte_sched/rte_approx.c  |  8 +++
  lib/librte_sched/rte_bitmap.h  |  3 ++-
  test/test/test_cryptodev.c |  2 +-
  test/test/test_cryptodev_perf.c| 26 +++---
  test/test/test_eventdev_sw.c   |  2 +-
  test/test/test_malloc.c|  4 ++--
  test/test/test_memory.c|  2 +-
  test/test/test_mempool.c   |  2 +-
  test/test/test_ring.c  |  6 ++---
  test/test/test_table_acl.c |  2 +-
  test/test/test_table_pipeline.c|  2 +-
  39 files changed, 69 insertions(+), 70 deletions(-)



[...]

I agree with Harry's comments (in another reply to this thread) that 
such patches might impact other pending patch series, quite late in 
merge window. But, this is a good-to-have set, probably early in 1711 
window.



/* compute the number of steps to the right */
diff --git a/lib/librte_sched/rte_bitmap.h b/lib/librte_sched/rte_bitmap.h
index 010d752..e487b58 100644
--- a/lib/librte_sched/rte_bitmap.h
+++ b/lib/librte_sched/rte_bitmap.h
@@ -500,7 +500,8 @@ __rte_bitmap_scan_read(struct rte_bitmap *bmp, uint32_t 
*pos, uint64_t *slab)
uint64_t *slab2;
  
  	slab2 = bmp->array2 + bmp->index2;

-   for ( ; bmp->go2 ; bmp->index2 ++, slab2 ++, bmp->go2 = bmp->index2 & 
RTE_BITMAP_CL_SLAB_MASK) {
+   for ( ; bmp->go2; bmp->index2++, slab2++,
+ bmp->go2 = bmp->index2 & RTE_BITMAP_CL_SLAB_MASK) {


   
Trivial: space before ';' in 'for' here should also be removed.

[...]

Other than the above (and also from dpaa2 code change perspective):

Acked-by: Shreyansh Jain 

Re: [dpdk-dev] [PATCH] all: refactor coding style

[Tiwei Bie]

On Thu, Jul 20, 2017 at 10:34:39AM +0530, Shreyansh Jain wrote:
> On Wednesday 19 July 2017 02:36 PM, Tiwei Bie wrote:
> > Remove the unwanted spaces before `;' across DPDK source code
> > by below one-liner with some minor manual refinements.
> > 
> > find . -name '*.[ch]' | xargs sed -i 's/\([^;(]\) \+;/\1;/g'
> > 
> > The fixes for cmdline library are skipped, because it has a
> > different coding style. It deserves a separate cleanup if
> > necessary. The fixes for drivers' base code are also skipped
> > to keep the base code intact.
> > 
> > Signed-off-by: Tiwei Bie 
> > ---
[...]
> > /* compute the number of steps to the right */
> > diff --git a/lib/librte_sched/rte_bitmap.h b/lib/librte_sched/rte_bitmap.h
> > index 010d752..e487b58 100644
> > --- a/lib/librte_sched/rte_bitmap.h
> > +++ b/lib/librte_sched/rte_bitmap.h
> > @@ -500,7 +500,8 @@ __rte_bitmap_scan_read(struct rte_bitmap *bmp, uint32_t 
> > *pos, uint64_t *slab)
> > uint64_t *slab2;
> > slab2 = bmp->array2 + bmp->index2;
> > -   for ( ; bmp->go2 ; bmp->index2 ++, slab2 ++, bmp->go2 = bmp->index2 & 
> > RTE_BITMAP_CL_SLAB_MASK) {
> > +   for ( ; bmp->go2; bmp->index2++, slab2++,
> > + bmp->go2 = bmp->index2 & RTE_BITMAP_CL_SLAB_MASK) {
> 
>
> Trivial: space before ';' in 'for' here should also be removed.
> 

Thank you for your feedbacks! :-)

Hmm.. Actually the space between `(' and `;' was kept intentionally
when I wrote this 's/\([^;(]\) \+;/\1;/g' sed script. There are many
other such cases. It's acceptable to me, and I thought we like it:

diff --git i/app/test-eventdev/parser.h w/app/test-eventdev/parser.h
index 75a5a3b..372b85f 100644
--- i/app/test-eventdev/parser.h
+++ w/app/test-eventdev/parser.h
@@ -41,7 +41,7 @@
 #define skip_white_spaces(pos) \
 ({ \
__typeof__(pos) _p = (pos); \
-   for ( ; isspace(*_p); _p++) \
+   for (; isspace(*_p); _p++)  \
;   \
_p; \
 })
diff --git i/app/test-eventdev/test_perf_common.c 
w/app/test-eventdev/test_perf_common.c
index a5b768c..36b78bf 100644
--- i/app/test-eventdev/test_perf_common.c
+++ w/app/test-eventdev/test_perf_common.c
@@ -284,7 +284,7 @@ perf_event_dev_port_setup(struct evt_test *test, struct 
evt_options *opt,
.new_event_threshold = 1200,
};
prod = 0;
-   for ( ; port < perf_nb_event_ports(opt); port++) {
+   for (; port < perf_nb_event_ports(opt); port++) {
struct prod_data *p = &t->prod[port];
 
p->dev_id = opt->dev_id;
diff --git i/drivers/net/bonding/rte_eth_bond_pmd.c 
w/drivers/net/bonding/rte_eth_bond_pmd.c
index 383e27c..5d8e068 100644
--- i/drivers/net/bonding/rte_eth_bond_pmd.c
+++ w/drivers/net/bonding/rte_eth_bond_pmd.c
@@ -371,7 +371,7 @@ bond_ethdev_tx_burst_8023ad_fast_queue(void *queue, struct 
rte_mbuf **bufs,
/* If tx burst fails move packets to end of bufs */
if (unlikely(num_tx_slave < slave_nb_pkts[i])) {
uint16_t j = nb_pkts - num_tx_fail_total;
-   for ( ; num_tx_slave < slave_nb_pkts[i]; j++,
+   for (; num_tx_slave < slave_nb_pkts[i]; j++,
num_tx_slave++)
bufs[j] = slave_bufs[i][num_tx_slave];
}
@@ -1308,7 +1308,7 @@ bond_ethdev_tx_burst_8023ad(void *queue, struct rte_mbuf 
**bufs,
slave_bufs[i], slave_nb_pkts[i]);
 
/* If tx burst fails drop slow packets */
-   for ( ; num_tx_slave < slave_slow_nb_pkts[i]; num_tx_slave++)
+   for (; num_tx_slave < slave_slow_nb_pkts[i]; num_tx_slave++)
rte_pktmbuf_free(slave_bufs[i][num_tx_slave]);
 
num_tx_total += num_tx_slave - slave_slow_nb_pkts[i];
@@ -1317,7 +1317,7 @@ bond_ethdev_tx_burst_8023ad(void *queue, struct rte_mbuf 
**bufs,
/* If tx burst fails move packets to end of bufs */
if (unlikely(num_tx_slave < slave_nb_pkts[i])) {
uint16_t j = nb_pkts - num_tx_fail_total;
-   for ( ; num_tx_slave < slave_nb_pkts[i]; j++, 
num_tx_slave++)
+   for (; num_tx_slave < slave_nb_pkts[i]; j++, 
num_tx_slave++)
bufs[j] = slave_bufs[i][num_tx_slave];
}
}
diff --git i/drivers/net/cxgbe/base/t4_hw.c w/drivers/net/cxgbe/base/t4_hw.c
index a8ccea0..ed26f71 100644
--- i/drivers/net/cxgbe/base/t4_hw.c
+++ w/drivers/net/cxgbe/base/t4_hw.c
@@ -277,7 +277,7 @@ static void t4_report_fw_error(struct adapter *adap)
 static void get_mbox_rpl(struct adapter *adap, __be64 *rpl, int nflit,
 u32 mbox_addr)
 {
-   for ( ; nflit; nflit--, mbox_a

Re: [dpdk-dev] [pull-request] next-net 17.08 RC2

[Thomas Monjalon]

19/07/2017 18:23, Ferruh Yigit:
>   http://dpdk.org/git/next/dpdk-next-net 

Pulled, thanks

Re: [dpdk-dev] [pull-request] next-eventdev 17.08 rc2

[Thomas Monjalon]

20/07/2017 07:22, Jerin Jacob:
>   http://dpdk.org/git/next/dpdk-next-eventdev 

Pulled, thanks