31/07/2017 03:43, Jiayu Hu:
> When try to get GRO types, expression "1 << i" with type "int" may
> overflow. This patch is to fix this issue.
>
> Fixes: e996506a1c07 ("lib/gro: add Generic Receive Offload API framework")
> Coverity issue: 158664
>
> Signed-off-by: Jiayu Hu
Applied, thanks
13/07/2017 04:16, Herbert Guan:
> Use built-in data types for unsupported poly64/128 types
> for GCC version lower than 4.9.0
>
> Fixes: 3c4b4024c225 ("arch/arm: add vcopyq_laneq_u32 for old gcc")
>
> Signed-off-by: Herbert Guan
Applied, thanks
Hi Yuanhan,
> -Original Message-
> From: Yuanhan Liu [mailto:y...@fridaylinux.org]
> Sent: Monday, July 31, 2017 11:47 AM
> To: Hu, Jiayu
> Cc: Tan, Jianfeng ; dev@dpdk.org
> Subject: Re: [PATCH] net/virtio: fix fail to configure rxq interrupt
>
> On Mon, Jul 31, 2017 at 02:47:13AM +
22/07/2017 22:07, Charles (Chas) Williams:
> This function wasn't provided in the original commit.
>
> Fixes: 4f0981e6ec6c ("eal: deprecate log functions")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Chas Williams
Applied, thanks
26/07/2017 13:52, Gaetan Rivet:
> When rte_eal_hotplug_add() successfully probe a device, the return value
> is zero. The check afterward only returns on error different from
> -EINVAL. It should return also on success, as there is no need to
> attempt probing the device with vdev.
>
> Fixes: 0bba
When use rte_eth_dev_configure() to enable rx queue interrupt for virtio
devices, virtio_init_device() isn't called to set up the interrupt
environment, which causes rx queue interrupt setup failed. This patch is
to fix this issue.
Fixes: 26b683b4f7d0 ("net/virtio: setup Rx queue interrupts")
Cc:
On Wed, Jul 26, 2017 at 03:35:51PM +0200, Gaetan Rivet wrote:
> Device detach in librte_ether is rough right now.
>
> - Device hotplug capability is not properly checked
> - Device state should be set after a successful detach
> - MLX drivers are lacking the relevant flag
> - And this flag sho
> > Gaetan Rivet (6):
> > ethdev: fix device state on detach
> > ethdev: properly check detach capability
> > net/mlx4: advertize the detach capability
> > net/mlx5: advertize the detach capability
> > app/testpmd: let the user know device detach failed
> > doc: announce ethdev API chan
The flag RTE_ETH_DEV_DETACHABLE will disappear.
This flag is not needed anymore following the hotplug work done for
v17.08. It can be removed, its function is now implicitly made available
by the relevant EAL and rte_bus implementations.
Signed-off-by: Gaetan Rivet
Acked-by: John McNamara
Acked
On 7/31/2017 4:24 AM, Herbert Guan wrote:
> Fixes: 4c42498d916d ("net/bonding: allow slaves to also be bonded devices")
>
> Signed-off-by: Herbert Guan
Thanks for the patch, this looks like duplicate [1], I will get the
other one since it has been sent earlier.
[1]
http://dpdk.org/dev/patchwork
Coverity issue: 158660
Fixes: 4afd440ec132 ("app/testeventdev: update options through command line")
Cc: sta...@dpdk.org
Signed-off-by: Jerin Jacob
---
app/test-eventdev/evt_options.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/app/test-eventdev/evt_options.c b/app/test-e
pkts == 0 will result in divide by zero case.
Added a check to fix it.
Coverity issue: 158652
Fixes: 9d3aeb185eb4 ("app/testeventdev: launch perf lcores")
Cc: sta...@dpdk.org
Signed-off-by: Jerin Jacob
---
app/test-eventdev/test_perf_common.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-
> From: Jerin Jacob [mailto:jerin.ja...@caviumnetworks.com]
> Sent: Monday, July 31, 2017 11:05 AM
> To: dev@dpdk.org
> Cc: tho...@monjalon.net; Richardson, Bruce ; Van
> Haaren, Harry
> ; hemant.agra...@nxp.com; Eads, Gage
> ;
> nipun.gu...@nxp.com; gprathyu...@caviumnetworks.com; Jerin Jacob
>
> From: Jerin Jacob [mailto:jerin.ja...@caviumnetworks.com]
> Sent: Monday, July 31, 2017 11:05 AM
> To: dev@dpdk.org
> Cc: tho...@monjalon.net; Richardson, Bruce ; Van
> Haaren, Harry
> ; hemant.agra...@nxp.com; Eads, Gage
> ;
> nipun.gu...@nxp.com; gprathyu...@caviumnetworks.com; Jerin Jacob
>
The name of a device is copied in a provided buffer within
rte_eth_dev_detach(). The current sizeof is done on a pointer instead of
the intended array usually pointed to.
The name field of an rte_device is not assured however to point an
rte_devargs name field. The almost correct length to base th
When registering a crypto driver, if memory allocation
fails, application should exit and do not allow
a NULL pointer dereference.
Coverity issue: 158645
Fixes: 7a364faef185 ("cryptodev: remove crypto device type enumeration")
Signed-off-by: Pablo de Lara
---
lib/librte_cryptodev/rte_cryptodev
If the device cannot be parsed as a PCI device, the rte_devargs
function returns -EFAULT. This error code signifies that the address
given to the PCI bus is invalid. When it happens, the best course of
action is to try to plug the device using the vdev bus.
Fixes: 1c35f666df07 ("dev: fix attach pr
Checking against error values returned by rte_eal_hotplug_add is
inelegant and prone to mistakes. Additionally, the failed PCI probe
prints a useless error that would throw off unsuspecting users.
Use the relevant functions to infer the intended bus. The limitation to
PCI or vdev device is kept fo
Current mlx4 OFED version has bug which returns error to
ibv destroy functions when the device was plugged out, in
spite of the resources were destroyed correctly.
Hence, failsafe PMD was aborted, only in debug mode, when
it tries to remove the device in plug-out process.
The workaround removed t
On 7/25/2017 4:32 AM, Ajit Khaparde wrote:
> The callback arguments to _rte_eth_dev_callback_process() are swapped.
> Fix them.
>
> Fixes: d6af1a13d7a1 ("ethdev: add return values to callback process API")
>
> Signed-off-by: Ajit Khaparde
> ---
> drivers/net/bnxt/rte_pmd_bnxt.c | 2 +-
> 1 file
On 7/25/2017 4:32 AM, Ajit Khaparde wrote:
> If the current VLAN anti spoof setting is same as the new value,
> the firmware can return an error. Call the HWRM command to update
> the new setting if it is different from the current value.
>
> Fixes: 36735a932ca7 ("net/bnxt: support set VF QOS and
> -Original Message-
> From: Shahaf Shuler [mailto:shah...@mellanox.com]
> Sent: Friday, July 28, 2017 3:29 PM
> To: tho...@monjalon.net; Mcnamara, John
> Cc: dev@dpdk.org
> Subject: [PATCH v5 3/3] doc: update mlx guides
>
> Update the guides with:
>* New supported features.
>*
On 7/31/2017 12:19 PM, Ferruh Yigit wrote:
> On 7/25/2017 4:32 AM, Ajit Khaparde wrote:
>> The callback arguments to _rte_eth_dev_callback_process() are swapped.
>> Fix them.
>>
>> Fixes: d6af1a13d7a1 ("ethdev: add return values to callback process API")
>>
>> Signed-off-by: Ajit Khaparde
>> ---
>
On 7/25/2017 2:24 PM, Lu, Wenzhuo wrote:
> Hi Kuba,
>
>> -Original Message-
>> From: Kozak, KubaX
>> Sent: Tuesday, July 25, 2017 8:19 AM
>> To: Lu, Wenzhuo
>> Cc: dev@dpdk.org; Jain, Deepak K ; Jastrzebski,
>> MichalX K ; Kozak, KubaX
>> ; sta...@dpdk.org
>> Subject: [PATCH] net/e1000: f
When checking if a buggy commit was introduced in an old version,
the script compares last tag containing the bug and current version.
The non merged tags from non related branches must be ignored.
Signed-off-by: Thomas Monjalon
---
devtools/git-log-fixes.sh | 2 +-
1 file changed, 1 insertion(+
24/07/2017 15:47, Matan Azrad:
> some issue seen when trying to test failsafe plug out and in.
>
> The V2 fixes some checkpatch issues.
>
> Matan Azrad (2):
> app/testpmd: fix flow rule copy functions
> ethdev: fix flow rule copy functions
Acked-by: Adrien Mazarguil
Applied, thanks
On 31/07/2017 03:30, Pablo de Lara wrote:
When registering a crypto driver, if memory allocation
fails, application should exit and do not allow
a NULL pointer dereference.
Coverity issue: 158645
Fixes: 7a364faef185 ("cryptodev: remove crypto device type enumeration")
Signed-off-by: Pablo de L
On 7/25/2017 3:01 PM, Nélio Laranjeiro wrote:
> On Tue, Jul 25, 2017 at 07:08:07PM +0530, Gowrishankar wrote:
>> From: Gowrishankar Muthukrishnan
>>
>> ETHTOOL_GLINKSETTINGS ioctl call in mlx5 pmd returns inconsistent
>> link status due to which any application relying on it would not
>> function
rte_cryptodev_socket_id() returns the socket id of
a crypto device, unless the device id is not valid,
in which case, it returns -1.
This should not happen, as the device id is controlled
by the application, but just for safety, a check is added.
Coverity issue: 158628, 158638, 158656, 158662
Fixe
On 7/27/2017 1:28 AM, Wu, Jingjing wrote:
>
>
>> -Original Message-
>> From: Hunt, David
>> Sent: Monday, July 24, 2017 4:49 PM
>> To: dev@dpdk.org
>> Cc: Wu, Jingjing ; sta...@dpdk.org; Hunt, David
>>
>> Subject: [PATCH v2] net/i40e: fix sync phy type by adding retry
>>
>> Some phy's ta
On 7/26/2017 6:06 PM, Declan Doherty wrote:
> On 26/07/2017 4:50 PM, Tomasz Kulasek wrote:
>> rte_eth_bond_primary_set segfaults for invalid port. This patch moves
>> devices check before use of internal data.
>>
>> Fixes: 4c42498d916d ("net/bonding: allow slaves to also be bonded devices")
>>
>> S
From: Gowrishankar Muthukrishnan
Earlier bonding pmd was disabled in default config for ppc64le.
Hence, removing it as it has been verified.
Signed-off-by: Gowrishankar Muthukrishnan
---
v2:
- remove configuration variable for default enablement
config/defconfig_ppc_64-power8-linuxapp-gcc |
On 7/26/2017 6:08 PM, Declan Doherty wrote:
> On 26/07/2017 4:53 PM, Tomasz Kulasek wrote:
>> RTE_ETH_EVENT_INTR_LSC callbacks are not called when all slaves goes down
>> in bond_ethdev_lsc_event_callback. It causes that link status change of
>> bonded device is not propagated up.
>>
>> Fixes: deba
Checking against error values returned by rte_eal_hotplug_add is
inelegant and prone to mistakes. Additionally, the failed PCI probe
prints a useless error that would throw off unsuspecting users:
ERROR: failed to parse device "pci:net_ring0"
This error is printed when attempting to probe a vi
On 7/31/2017 1:54 PM, Gowrishankar wrote:
> From: Gowrishankar Muthukrishnan
>
> Earlier bonding pmd was disabled in default config for ppc64le.
> Hence, removing it as it has been verified.
>
> Signed-off-by: Gowrishankar Muthukrishnan
Applied to dpdk-next-net/master, thanks.
On 7/27/2017 2:02 AM, Wei Dai wrote:
> This patch adds support of 2.5G and 5G ethernet interface on X550.
Should speed capabilities (dev_info->speed_capa) also updated?
>
> Signed-off-by: Wei Dai
<...>
31/07/2017 14:58, Gaetan Rivet:
> Checking against error values returned by rte_eal_hotplug_add is
> inelegant and prone to mistakes. Additionally, the failed PCI probe
> prints a useless error that would throw off unsuspecting users:
>
>ERROR: failed to parse device "pci:net_ring0"
>
> This
On 7/28/2017 4:29 AM, Wu, Jingjing wrote:
>
>
>> -Original Message-
>> From: Kozak, KubaX
>> Sent: Thursday, July 27, 2017 3:28 PM
>> To: Wu, Jingjing
>> Cc: dev@dpdk.org; Jain, Deepak K ; Jastrzebski,
>> MichalX K
>> ; Kozak, KubaX ;
>> sta...@dpdk.org
>> Subject: [PATCH] net/i40e: fix
31/07/2017 12:29, Gaetan Rivet:
> The name of a device is copied in a provided buffer within
> rte_eth_dev_detach(). The current sizeof is done on a pointer instead of
> the intended array usually pointed to.
>
> The name field of an rte_device is not assured however to point an
> rte_devargs name
On Mon, Jul 31, 2017 at 03:27:29PM +0200, Thomas Monjalon wrote:
> 31/07/2017 12:29, Gaetan Rivet:
> > The name of a device is copied in a provided buffer within
> > rte_eth_dev_detach(). The current sizeof is done on a pointer instead of
> > the intended array usually pointed to.
> >
> > The name
The name of a device is copied in a provided buffer within
rte_eth_dev_detach(). The current sizeof is done on a pointer instead of
the intended array usually pointed to.
The name field of an rte_device is not assured however to point an
rte_devargs name field. The almost correct length to base th
Hi Matan,
On Mon, Jul 31, 2017 at 02:15:09PM +0300, Matan Azrad wrote:
> Current mlx4 OFED version has bug which returns error to
> ibv destroy functions when the device was plugged out, in
> spite of the resources were destroyed correctly.
>
> Hence, failsafe PMD was aborted, only in debug mode,
31/07/2017 15:40, Gaetan Rivet:
> The name of a device is copied in a provided buffer within
> rte_eth_dev_detach(). The current sizeof is done on a pointer instead of
> the intended array usually pointed to.
>
> The name field of an rte_device is not assured however to point an
> rte_devargs name
The corrupted code doesn't return error when probe function
fails due to error in device mac address getting.
By this way, the probe function may return success even if the
ETH dev is not allocated.
Hence, the probe caller, for example failsafe PMD, fails when it
tries to get ETH dev after the dev
Hi Gowrishankar, Declan,
On Mon, Jul 10, 2017 at 12:02:24PM +0530, gowrishankar muthukrishnan wrote:
> On Friday 07 July 2017 09:08 PM, Declan Doherty wrote:
> >On 04/07/2017 12:57 PM, Gowrishankar wrote:
> >>From: Gowrishankar Muthukrishnan
> >>
> >>At present, creating bonding devices using --v
For anybody planning to attend our DPDK Summit Userspace event in Dublin
(September 26/27), registration is now open at: https://dpdksummit.com/.
The CFP for this event is also now open. You can submit presentation proposals
at: https://dpdksummit.com/AbstractProposalSubmission.aspx. The deadlin
On 7/31/2017 3:50 AM, Wu, Jingjing wrote:
>
>
>> -Original Message-
>> From: Li, Xiaoyun
>> Sent: Friday, July 28, 2017 11:48 PM
>> To: Wu, Jingjing
>> Cc: dev@dpdk.org; Li, Xiaoyun ; sta...@dpdk.org
>> Subject: [PATCH v3] net/i40e: fix PF notify issue when VF not up
>>
>> This patch sto
Hi Zhiyong,
On Fri, Jul 28, 2017 at 04:10:14PM +0800, Zhiyong Yang wrote:
> The code eth_dev->intr_handle = &dev->intr_handle; has duplicate code
> in the function rte_eth_copy_pci_info(), remove it here.
>
> Signed-off-by: Zhiyong Yang
Acked-by: Gaetan Rivet
> ---
> lib/librte_ether/rte_eth
On 7/29/2017 9:54 PM, Ajit Khaparde wrote:
> Only set the vlanonly bit if vlan_nonvlan is clear. Also, allow the
> VLAN table to be cleared when vlanonly is set.
>
> Clearing the VLAN table when vlanonly is set will stop all traffic
> since it requires all frames to have a VLAN tag, and that tag t
31/07/2017 16:42, Gaëtan Rivet:
> Hi Zhiyong,
>
> On Fri, Jul 28, 2017 at 04:10:14PM +0800, Zhiyong Yang wrote:
> > The code eth_dev->intr_handle = &dev->intr_handle; has duplicate code
> > in the function rte_eth_copy_pci_info(), remove it here.
> >
> > Signed-off-by: Zhiyong Yang
>
> Acked-by
On Fri, Jul 28, 2017 at 3:28 AM, Bruce Richardson
wrote:
> On Thu, Jul 27, 2017 at 02:41:46PM -0600, asom...@gmail.com wrote:
>> From: Alan Somers
>>
>> "/bin/bash" is a Linuxism. "/usr/bin/env bash" is portable.
>>
>> Signed-off-by: Alan Somers
>> ---
>> examples/performance-thread/l3fwd-thre
26/07/2017 06:20, Pablo de Lara:
> When calling rte_pktmbuf_alloc_bulk, if there are
> not enough objects in the mempool, it returns
> a negative value, which should be reflected
> in the Doxygen comments.
>
> Fixes: 9ec201f5d6e7 ("mbuf: provide bulk allocation")
> Cc: sta...@dpdk.org
>
> Signed-
On 7/26/2017 6:06 PM, Declan Doherty wrote:
> On 26/07/2017 4:46 PM, Tomasz Kulasek wrote:
>> Device name in device structure in virtual device used in
>> link_bonding_autotest is not set what causes segmentation fault when
>> rte_eth_dev_allocated is called.
>>
>> Fixes: a1e7c17555e8 ("ethdev: use
On Mon, Jul 31, 2017 at 09:11:11AM -0600, alan somers wrote:
> On Fri, Jul 28, 2017 at 3:28 AM, Bruce Richardson
> wrote:
> > On Thu, Jul 27, 2017 at 02:41:46PM -0600, asom...@gmail.com wrote:
> >> From: Alan Somers
> >>
> >> "/bin/bash" is a Linuxism. "/usr/bin/env bash" is portable.
> >>
> >>
31/07/2017 14:32, Sergio Gonzalez Monroy:
> On 31/07/2017 03:30, Pablo de Lara wrote:
> > --- a/lib/librte_cryptodev/rte_cryptodev.c
> > +++ b/lib/librte_cryptodev/rte_cryptodev.c
> > @@ -1404,6 +1404,12 @@ rte_cryptodev_allocate_driver(const struct
> > rte_driver *drv)
> > struct cryptodev_dr
On 7/12/2017 11:23 AM, Declan Doherty wrote:
> On 10/07/17 12:13, Herbert Guan wrote:
>> Patch V2: fix build warnings by deleting unused variables.
>>
>> There were double-free problems in some test cases, which will cause
>> a duplicated mbuf will be added into mempool. After double-free,
>> some
> > - Theme is based on your original rx adapter proposal but with eventpmd
> > ops(not adapter ops).i.e Reuse as much of your existing Rx adapter
> > implementation as common code and add hooks for HW based adapters. For
> > example, before you add to "rx_poll" in
> > eth_poll_wrr_calc(),
> >
On 7/25/2017 10:46 AM, Declan Doherty wrote:
> On 13/07/2017 9:00 AM, Herbert Guan wrote:
>> When test case "test_balance_l23_tx_burst_ipv4_toggle_ip_addr" is
>> calling balance_l23_tx_burst(), the ip_addr instead of mac_addr
>> should be toggled according to the test name.
Fixes: 92073ef961ee ("b
On 7/26/2017 6:06 PM, Declan Doherty wrote:
> On 26/07/2017 4:44 PM, Daniel Mrzyglod wrote:
>> drivers are looking by name of the device so change namespace to proper one
>>
>> Fixes: 43b630244e7e ("app/test: add dynamic bonding RSS configuration")
>>
>> Signed-off-by: Daniel Mrzyglod
> Acked-by:
On 7/26/2017 6:06 PM, Declan Doherty wrote:
> On 26/07/2017 4:48 PM, Tomasz Kulasek wrote:
>> Bonding devices name must start with "net_bonding" prefix.
>>
>> Fixes: 9bf4901d1a11 ("bus/vdev: remove probe with driver name option")
>>
>> Signed-off-by: Tomasz Kulasek
> Acked-by: Declan Doherty
Ap
This commit fixes shifts to an integer (1 << shift) which
is assumed to be a 32-bit integer. In this case, the shift is
variable and expected to be valid for 64-bit integers. Given that
the expectation to work with 64 bits exists, we must ensure that
the (1 << shift) one in that formula is actually
On 7/28/2017 11:05 AM, Mcnamara, John wrote:
>
>
>> -Original Message-
>> From: Mrzyglod, DanielX T
>> Sent: Thursday, July 27, 2017 7:41 AM
>> To: Mcnamara, John ; Doherty, Declan
>>
>> Cc: dev@dpdk.org; Mrzyglod, DanielX T
>> Subject: [PATCH] doc: add testpmd bonding mode 4 aggregator
On 7/20/2017 4:48 PM, Yongseok Koh wrote:
> mlx5_tx_complete() polls completion queue multiple times until it
> encounters an invalid entry. As Tx completions are suppressed by
> MLX5_TX_COMP_THRESH, it is waste of cycles to expect multiple completions
> in a poll. And freeing too many buffers in a
Hi Harry,
On Mon, Jul 31, 2017 at 04:58:27PM +0100, Harry van Haaren wrote:
> This commit fixes shifts to an integer (1 << shift) which
> is assumed to be a 32-bit integer. In this case, the shift is
> variable and expected to be valid for 64-bit integers. Given that
> the expectation to work with
On Mon, Jul 31, 2017 at 9:22 AM, Bruce Richardson
wrote:
> On Mon, Jul 31, 2017 at 09:11:11AM -0600, alan somers wrote:
>> On Fri, Jul 28, 2017 at 3:28 AM, Bruce Richardson
>> wrote:
>> > On Thu, Jul 27, 2017 at 02:41:46PM -0600, asom...@gmail.com wrote:
>> >> From: Alan Somers
>> >>
>> >> "/bin
On 7/25/2017 5:42 PM, Yongseok Koh wrote:
> When advancing Tx ring index (txq->wqe_ci) in txq_scatter_v(), the title
> descriptor of multi-packet send isn't taken into account if it doesn't
> cross 64B boundary.
>
> Fixes: 6cb559d67b83 ("net/mlx5: add vectorized Rx/Tx burst for x86")
>
> Signed-o
> From: Gaëtan Rivet [mailto:gaetan.ri...@6wind.com]
> Sent: Monday, July 31, 2017 5:18 PM
> To: Van Haaren, Harry
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] service: fix shifts to operate on 64 bit
> integers
>
> Hi Harry,
>
> On Mon, Jul 31, 2017 at 04:58:27PM +0100, Harry van Haare
The callback arguments to _rte_eth_dev_callback_process() are swapped.
Fix them.
Fixes: d6af1a13d7a1 ("ethdev: add return values to callback process API")
Signed-off-by: Ajit Khaparde
--
v1->v2: rename a cb_param to ret_param as per review comment.
---
drivers/net/bnxt/rte_pmd_bnxt.c | 17
On 7/26/2017 8:29 PM, Yongseok Koh wrote:
> Even though HW supports more packet types, PMD doesn't parse it. Changed to
> have
> more packet_type flags in mbuf.
>
> Yongseok Koh (2):
> net/mlx5: fix missing packet type calculation
> net/mlx5: fix L4 packet type support
Series applied to dpdk
This commit fixes shifts to an integer (1 << shift) which
is assumed to be a 32-bit integer. In this case, the shift is
variable and expected to be valid for 64-bit integers. Given that
the expectation to work with 64 bits exists, we must ensure that
the (1 << shift) one in that formula is actually
On 7/27/2017 11:27 AM, Matan Azrad wrote:
> The corrupted code causes segmentation fault when user creates
> flow with drop action before device starting.
>
> For example, failsafe PMD recreates all the flows before calling
> dev_start in plug-in sequence and mlx4 allocated its flow drop
> queue i
> > If there's no Traffic Management node added,
> > not necessary to check if TM is committed.
> >
> > Fixes: 5713ade69776 ("net/ixgbe: support committing TM hierarchy")
> >
> > Signed-off-by: Wenzhuo Lu
>
> Acked-by: Cristian Dumitrescu
Added TM in the title and applied, thanks
On 07/31/2017 09:56 AM, Jiayu Hu wrote:
When use rte_eth_dev_configure() to enable rx queue interrupt for virtio
devices, virtio_init_device() isn't called to set up the interrupt
environment, which causes rx queue interrupt setup failed. This patch is
to fix this issue.
Fixes: 26b683b4f7d0 ("
> On Jul 31, 2017, at 11:18 AM, alan somers wrote:
>
> On Mon, Jul 31, 2017 at 9:22 AM, Bruce Richardson
> wrote:
>> On Mon, Jul 31, 2017 at 09:11:11AM -0600, alan somers wrote:
>>> On Fri, Jul 28, 2017 at 3:28 AM, Bruce Richardson
>>> wrote:
On Thu, Jul 27, 2017 at 02:41:46PM -0600, asom
Hi Adrien
> -Original Message-
> From: Adrien Mazarguil [mailto:adrien.mazarg...@6wind.com]
> Sent: Monday, July 31, 2017 5:17 PM
> To: Matan Azrad
> Cc: dev@dpdk.org; Thomas Monjalon ; Olga Shern
> ; sta...@dpdk.org
> Subject: Re: [PATCH] net/mlx4: workaround to verbs wrong error return
On 7/31/2017 3:30 PM, Matan Azrad wrote:
> The corrupted code doesn't return error when probe function
> fails due to error in device mac address getting.
> By this way, the probe function may return success even if the
> ETH dev is not allocated.
>
> Hence, the probe caller, for example failsafe
> > If there's no Traffic Management node added. Not
> > necessary to check if TM is committed.
> >
> > Fixes: cac29c3c00a4 ("net/i40e: support committing TM hierarchy")
> > Signed-off-by: Wenzhuo Lu
>
> Acked-by: Cristian Dumitrescu
Added TM in the title and applied, thanks
31/07/2017 19:01, Maxime Coquelin:
> On 07/31/2017 09:56 AM, Jiayu Hu wrote:
> > When use rte_eth_dev_configure() to enable rx queue interrupt for virtio
> > devices, virtio_init_device() isn't called to set up the interrupt
> > environment, which causes rx queue interrupt setup failed. This patch
When register a crypto driver, a cryptodev driver
structure was being allocated, using malloc.
Since this call may fail, it is safer to allocate
this memory statically in each PMD, so driver registration
will never fail.
Coverity issue: 158645
Fixes: 7a364faef185 ("cryptodev: remove crypto device
On 7/31/2017 5:24 PM, Ajit Khaparde wrote:
> The callback arguments to _rte_eth_dev_callback_process() are swapped.
> Fix them.
>
> Fixes: d6af1a13d7a1 ("ethdev: add return values to callback process API")
>
> Signed-off-by: Ajit Khaparde
Applied to dpdk-next-net/master, thanks.
Hi Ferruh,
> -Original Message-
> From: Yigit, Ferruh
> Sent: Monday, July 31, 2017 9:15 AM
> To: Dai, Wei ; Lu, Wenzhuo ;
> Ananyev, Konstantin
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v5] net/ixgbe: add support of 2.5G and 5G on
> X550
>
> On 7/27/2017 2:02 AM, Wei Dai wrote
On 7/24/2017 11:19 AM, Rasesh Mody wrote:
> This patch modifies the recovery flow to allow ongoing PCIe
> transactions to be completed. To achieve this, the load sequence is
> changed such that the "final_cleanup" notification is sent while the
> FID_enable is cleared.
> This change ensures that th
From: Wei Dai
This patch adds support of 2.5G and 5G ethernet interface on X550.
Signed-off-by: Wei Dai
Signed-off-by: Wenzhuo Lu
---
v6:
- update the link capability.
drivers/net/ixgbe/ixgbe_ethdev.c | 31 ---
drivers/net/ixgbe/ixgbe_ethdev.h | 8
2 fi
On 7/31/2017 7:12 PM, Wenzhuo Lu wrote:
> From: Wei Dai
>
> This patch adds support of 2.5G and 5G ethernet interface on X550.
>
> Signed-off-by: Wei Dai
> Signed-off-by: Wenzhuo Lu
Applied to dpdk-next-net/master, thanks.
The following changes since commit 41ab4ff5e5d9bb6c0ba43dce1eecbf62d322e19c:
net/i40e: fix TM hierarchy commit check (2017-07-31 19:16:23 +0200)
are available in the git repository at:
http://dpdk.org/git/next/dpdk-next-net
for you to fetch changes up to 766cc8607384ab26d60a1889a59e5afeb00
On Mon, Jul 31, 2017 at 11:06 AM, Wiles, Keith wrote:
>
>> On Jul 31, 2017, at 11:18 AM, alan somers wrote:
>>
>> On Mon, Jul 31, 2017 at 9:22 AM, Bruce Richardson
>> wrote:
>>> On Mon, Jul 31, 2017 at 09:11:11AM -0600, alan somers wrote:
On Fri, Jul 28, 2017 at 3:28 AM, Bruce Richardson
>>
Currently, on i40e, the max queue number per VF is set by a macro.
It means the value is decided when compiling. It's not friendly to
the users. Because every time the users want to change the value,
the code need to be re-compiled.
The reason of using a macro for the max queue number is that the
Add a global variable to set the max queue number per VF.
This variable is only valid on PF.
Signed-off-by: Wenzhuo Lu
---
lib/librte_ether/rte_ethdev.c | 3 +++
lib/librte_ether/rte_ethdev.h | 2 ++
2 files changed, 5 insertions(+)
diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/
Remove the compile macro for max queue number per
VF. Use the global variable to set the number.
Signed-off-by: Wenzhuo Lu
---
config/common_base | 1 -
drivers/net/i40e/i40e_ethdev.c | 4 ++--
2 files changed, 2 insertions(+), 3 deletions(-)
diff --git a/config/common_base b/config
Add a parameter in testpmd CLI. This parameter is used
to set the max queue number per VF.
Signed-off-by: Wenzhuo Lu
---
app/test-pmd/parameters.c | 33 -
app/test-pmd/testpmd.c| 2 ++
app/test-pmd/testpmd.h| 1 +
3 files changed, 35 insertions(+), 1 del
31/07/2017 11:18, Pablo de Lara:
> When register a crypto driver, a cryptodev driver
> structure was being allocated, using malloc.
> Since this call may fail, it is safer to allocate
> this memory statically in each PMD, so driver registration
> will never fail.
>
> Coverity issue: 158645
>
> Fi
31/07/2017 20:22, Ferruh Yigit:
> http://dpdk.org/git/next/dpdk-next-net
Pulled, thanks for your great help!
31/07/2017 18:38, Harry van Haaren:
> This commit fixes shifts to an integer (1 << shift) which
> is assumed to be a 32-bit integer. In this case, the shift is
> variable and expected to be valid for 64-bit integers. Given that
> the expectation to work with 64 bits exists, we must ensure that
> th
31/07/2017 16:58, Dirk-Holger Lenz:
> If dpdk is built with 'CONFIG_RTE_BUILD_SHARED_LIB=y' then
> using the vdev feature (args: e.g. -c 3 -n 4 --vdev="crypto_openssl")
> the rte_eal_init() returns 'ERROR: failed to parse device "crypto_openssl"'.
> It looks to me that rte_eal_devargs_add() calling
> > Coverity issue: 158660
> > Fixes: 4afd440ec132 ("app/testeventdev: update options through command
> > line")
> > Cc: sta...@dpdk.org
> >
> > Signed-off-by: Jerin Jacob
>
> Acked-by: Harry van Haaren
Series applied, thanks
31/07/2017 06:40, Pablo de Lara:
> rte_cryptodev_socket_id() returns the socket id of
> a crypto device, unless the device id is not valid,
> in which case, it returns -1.
> This should not happen, as the device id is controlled
> by the application, but just for safety, a check is added.
>
> Cove
27/04/2017 08:51, Guduri Prathyusha:
> rte_cfgfile_section_num_entries_by_index() is added to get the number of
> entries of a section when multiple sections of the same name are
> present.
>
> Signed-off-by: Guduri Prathyusha
> ---
> lib/librte_cfgfile/rte_cfgfile.c | 13 +
> lib/
Hi,
12/07/2017 16:12, Reshma Pattan:
> Describes how to use pdump library and dpdk-pdump tool
> to capture traffic on dpdk ports.
> --- /dev/null
> +++ b/doc/guides/howto/img/packet_capture_framework.svg
> @@ -0,0 +1,128 @@
> +
> + "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd";>
> +http://www
28/07/2017 23:35, Rami Rosen:
> This trivial patch removes wrong comments about
> the return value of the rte_eth_copy_pci_info() method.
> The method return value type is void.
>
> Signed-off-by: Rami Rosen
Applied, thanks
28/07/2017 20:55, Wenzhuo Lu:
> There're '\n's missing, so the error prompt
> looks bad.
>
> Fixes: 3af72783d1fd ("app/testpmd: configure event display")
> Fixes: 7ee3e9446290 ("app/testpmd: add --flow-isolate-all option")
> CC: sta...@dpdk.org
>
> Signed-off-by: Wenzhuo Lu
Applied, thanks
1 - 100 of 110 matches
Mail list logo
