On 4/13/2017 10:21 AM, Yuanhan Liu wrote:
On Thu, Apr 13, 2017 at 10:18:43AM +0800, Tan, Jianfeng wrote:
On 4/13/2017 9:58 AM, Yuanhan Liu wrote:
On Fri, Apr 07, 2017 at 07:57:40AM +, Jianfeng Tan wrote:
After the introduction of vhost MTU, VIRTIO_NET_F_MTU is enabled
by default. Howeve
On 4/13/2017 10:36 AM, Yuanhan Liu wrote:
On Thu, Apr 13, 2017 at 10:18:43AM +0800, Tan, Jianfeng wrote:
On 4/13/2017 9:58 AM, Yuanhan Liu wrote:
On Fri, Apr 07, 2017 at 07:57:40AM +, Jianfeng Tan wrote:
After the introduction of vhost MTU, VIRTIO_NET_F_MTU is enabled
by default.
Actua
DPDK community has several emails discussion on this topic,
these mails name is "FW: Issues with ixgbe and rte_flow",
items like VLAN can already have several valid "types"
(0x88a8, 0x8100, 0x9100), and who knows what will come up
in the future.And ixgbe_flow just ignores the types when do
filter c
Fix a typo in of item type in ixgbe generic filter.
Fixes: 11777435c72 ("net/ixgbe: parse flow director filter")
Signed-off-by: Wei Zhao
---
drivers/net/ixgbe/ixgbe_flow.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_flow.c b/drivers/net/ixgbe/ix
On 06/04/2017 11:06, Chao Zhu wrote:
On IBM POWER platform, when mapping /dev/zero file to hugepage memory
space, mmap will not respect the requested address hint. This will cause
the memory initilization for the second process fails. This patch adds
the required mmap flags to make it work. Besid
According to HW implementation, the bandwidth of QoS
means the L2 bandwidth, not count the bytes added by
physical layer.
Signed-off-by: Wenzhuo Lu
---
drivers/net/i40e/rte_pmd_i40e.h | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/net/i40e/rte_pmd_i40e.h b/drivers/net/i40e/rte
some customers find adding mac addr to VF sometimes can fail,
but it is still stored in dev->data->mac_addrs[ ]. So this
can lead to some errors that assumes the non-zero entry in
dev->data->mac_addrs[ ] is valid.
Fixes: af75078fece3 ("first public release")
Cc: sta...@dpdk.org
Signed-off-by: Wei
Current ethdev always stores MAC address even it fails to be added.
Other function may regard the failed MAC address valid and lead to
some errors. So There is a need to check if the addr is added
successfully or not and discard it if it fails.
In 3rd patch, add a command "add_more_mac_addr port_i
Add following lines in section of API change in release note.
If a MAC address fails to be added without this change, it is still stored
and may be regarded as a valid one. This may lead to errors in application.
The type of return value of eth_mac_addr_add_t in rte_ethdev.h is changed.
Any specif
This patch is added to introduce a testpmd command which
is used to add more than one MAC addresses one time.
This command can simplify the test for the change where
the type of return value of adding MAC address.
Normally a MAC address may fails to be added only after
many MAC addresses have been
Yes, with the __rte_cache_aligned, the :8 should be useless anyway.
On Thu, Apr 13, 2017 at 08:29:27AM +0300, Shahaf Shuler wrote:
Compilation error seen while compiling mlx5 in debug mode
under RHEL 7.3:
rte_ethdev.h:1670:7: error: type of bit-field 'state' is a GCC extension
[-Werror=pedan
Hi Olivier,
> -Original Message-
> From: Olivier MATZ [mailto:olivier.m...@6wind.com]
> Sent: Wednesday, April 12, 2017 8:24 PM
> To: De Lara Guarch, Pablo
> Cc: thomas.monja...@6wind.com; dev@dpdk.org
> Subject: Re: [PATCH v2] eal: redefine logtype values
>
> Hi Pablo,
>
>
> On Wed, 12
On Thu, Apr 13, 2017 at 07:42:39AM +0100, Wang, Zhihong wrote:
> Hi Bruce,
>
> This patch changes the behavior and causes some existing code to
> malfunction, e.g. bond_ethdev_stop() will get stuck here:
>
> while (rte_ring_dequeue(port->rx_ring, &pkt) != -ENOENT)
> rte_pktmbuf_free
On Thu, Apr 13, 2017 at 04:21:04PM +0800, Wei Dai wrote:
> some customers find adding mac addr to VF sometimes can fail,
> but it is still stored in dev->data->mac_addrs[ ]. So this
> can lead to some errors that assumes the non-zero entry in
> dev->data->mac_addrs[ ] is valid.
>
> Fixes: af75078f
Hi, Konstantin
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Thursday, April 6, 2017 5:03 PM
> To: Zhao1, Wei ; Thomas Monjalon
>
> Cc: Mcnamara, John ; dev@dpdk.org; Lu,
> Wenzhuo
> Subject: RE: [dpdk-dev] [PATCH v4 1/3] lib/librte_ether: add support for port
> reset
>
>
>
Signed-off-by: Shreyansh Jain
---
Context:
From the discussion in [1], it was observed that application should
have a default pool already linked even in case of shared builds.
Ring is especially important because packet mbuf creation API refer to
ring_mp_mc as default handler.
Documentatio
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of De Lara Guarch,
> Pablo
> Sent: Thursday, April 13, 2017 9:32 AM
> To: Olivier MATZ
> Cc: thomas.monja...@6wind.com; dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2] eal: redefine logtype values
>
> Hi Olivier,
On 4/12/2017 11:02 AM, Bruce Richardson wrote:
> On Wed, Apr 12, 2017 at 04:12:18AM +0100, Lu, Wenzhuo wrote:
>> Hi,
>>
>>
>>> -Original Message-
>>> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Zhang, Helin
>>> Sent: Friday, April 7, 2017 10:03 AM
>>> To: Richardson, Bruce
>>> Cc:
On Wednesday 12 April 2017 08:58 PM, Ferruh Yigit wrote:
On 4/11/2017 2:49 PM, Hemant Agrawal wrote:
This patch adds the NXP dpaa2 architecture and pmd details
in the Network interfaces section.
Signed-off-by: Hemant Agrawal
Acked-by: John McNamara
---
MAINTAINERS
On 4/13/2017 10:22 AM, Shreyansh Jain wrote:
> On Wednesday 12 April 2017 08:58 PM, Ferruh Yigit wrote:
>> On 4/11/2017 2:49 PM, Hemant Agrawal wrote:
>>> This patch adds the NXP dpaa2 architecture and pmd details
>>> in the Network interfaces section.
>>>
>>> Signed-off-by: Hemant Agrawal
>>> Ack
Thanks, Nelio Laranjeiro.
What about other parts, dear maintainers ?
> -Original Message-
> From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com]
> Sent: Thursday, April 13, 2017 4:45 PM
> To: Dai, Wei
> Cc: thomas.monja...@6wind.com; harish.pa...@cavium.com;
> rasesh.m...@cavium.c
Signed-off-by: Shreyansh Jain
---
Note:
Targetted for net-next, including Shijith's [1] clean-up patch for
documentation
[1] http://dpdk.org/ml/archives/dev/2017-April/063822.html
doc/guides/nics/dpaa2.rst | 81 ++-
1 file changed, 30 insertions(+
This commit fixes AES-GCM length of authentication input data,
cipher length is used instead
Fixes: 1703e94ac5ce ("qat: add driver for QuickAssist devices")
Signed-off-by: Arek Kusztal
---
drivers/crypto/qat/qat_crypto.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/crypto/q
On 4/11/2017 3:33 PM, Ferruh Yigit wrote:
> Coverity issue: 1419721
> Fixes: 75e2bc54c018 ("net/kni: add KNI PMD")
>
> Signed-off-by: Ferruh Yigit
Applied to dpdk-next-net/master, thanks.
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Shahaf Shuler
> Sent: Thursday, April 13, 2017 6:29 AM
> To: thomas.monja...@6wind.com
> Cc: adrien.mazarg...@6wind.com; nelio.laranje...@6wind.com; dev@dpdk.org
> Subject: [dpdk-dev] [PATCH] ethdev: fix compilation issue with strict flags
>
>
The error return code for rte_ring_dequeue() function should be -ENOENT
rather than -ENOBUFS (which is the error value from the enqueue() fn).
Fixes: cfa7c9e6fc1f ("ring: make bulk and burst return values consistent")
Reported-by: Zhihong Wang
Signed-off-by: Bruce Richardson
---
lib/librte_rin
On 4/11/2017 4:20 PM, Ferruh Yigit wrote:
> Same check duplicated, updated check according what commend states.
>
> Coverity issue: 1407507
> Fixes: 11777435c727 ("net/ixgbe: parse flow director filter")
>
> Signed-off-by: Ferruh Yigit
Superseded by http://dpdk.org/dev/patchwork/patch/23606/
> -Original Message-
> From: Qi Zhang [mailto:qi.z.zh...@intel.com]
> Sent: 2017年4月12日 21:56
> To: ferruh.yi...@intel.com
> Cc: jingjing...@intel.com; helin.zh...@intel.com; jianbo@linaro.org;
> chao...@linux.vnet.ibm.com; dev@dpdk.org; Qi Zhang
> Subject: [PATCH v9 0/3] net/i40e: conf
In rx vector mode, QinQ is not supported.
When hw_vlan_extend is set for QinQ ensure that
rx vector mode is not selected.
Fixes: 5b2d37858d32 ("net/i40e: fix single VLAN tag to be outer VLAN tag")
Cc: sta...@dpdk.org
Signed-off-by: Bernard Iremonger
---
Changes in v3:
Rebased to latest dpdk-ne
2017-04-13 08:55, Zhao1, Wei:
> From: Ananyev, Konstantin
> > From: Zhao1, Wei
> > > From: Thomas Monjalon [mailto:thomas.monja...@6wind.com]
> > > > 2017-04-06 02:57, Zhao1, Wei:
> > > > > > /**
> > > > > > > + * Reset an ethernet device when it's not working. One
> > > > > > > + scenario is, af
The feature negotiation in virtio-user is proven to be broken,
which results in device initialization failure.
Originally, we get features from vhost backend, and remove those
that are not supported. But when new feature is added, for example,
VIRTIO_NET_F_MTU, we fail to remove this new feature.
On 3/28/2017 3:28 AM, Qiming Yang wrote:
> This patch moved i40e related known issues from doc/guides/rel_notes/
> known_issues.rst to doc/guides/nics/i40e.rst, makes us easy
> to track the issue about i40e.
>
> Signed-off-by: Qiming Yang
> ---
> doc/guides/nics/i40e.rst | 26 ++
On 4/13/2017 10:33 AM, Shreyansh Jain wrote:
> Signed-off-by: Shreyansh Jain
Applied to dpdk-next-net/master, thanks.
On 4/13/2017 8:49 AM, Wei Zhao wrote:
> DPDK community has several emails discussion on this topic,
> these mails name is "FW: Issues with ixgbe and rte_flow",
Can you please add reference to discussions by mail list archive links.
Like: http://dpdk.org/ml/archives/dev/2017-April/063942.html
> it
On 4/13/2017 9:21 AM, Wenzhuo Lu wrote:
> According to HW implementation, the bandwidth of QoS
> means the L2 bandwidth, not count the bytes added by
> physical layer.
>
> Signed-off-by: Wenzhuo Lu
Applied to dpdk-next-net/master, thanks.
On 4/12/2017 9:52 AM, Wei Zhao wrote:
> Fix generic filter return info is not readable
> when repeat to create a rule.
>
> Fixes: 72c135a89f8 ("net/ixgbe: create consistent filter")
>
> Signed-off-by: Wei Zhao
Applied to dpdk-next-net/master, thanks.
On 4/13/2017 10:50 AM, Chao Zhu wrote:
>> -Original Message-
>> From: Qi Zhang [mailto:qi.z.zh...@intel.com]
>> Sent: 2017年4月12日 21:56
>> To: ferruh.yi...@intel.com
>> Cc: jingjing...@intel.com; helin.zh...@intel.com; jianbo@linaro.org;
>> chao...@linux.vnet.ibm.com; dev@dpdk.org; Qi Zh
We only enabled LSC when using vhost-user as the backend, but it is
reported even when using vhost-kernel as the backend.
Fix it by only reportting LSC support when using vhost-user as the
backend.
Fixes: 37a7eb2ae816("net/virtio-user: support to report net status")
Signed-off-by: Jianfeng Tan
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Michal Jastrzebski
> Sent: Tuesday, April 11, 2017 5:37 PM
> To: dev@dpdk.org
> Cc: Jain, Deepak K ; Van Haaren, Harry
> ; Piasecki, JacekX
> ; Kozak, KubaX ;
> Kulasek, TomaszX
> Subject: [dpdk-dev] [PATCH v5 1/
On 4/12/2017 8:30 PM, Qi Zhang wrote:
> VF link status rely on PF's notification, so when PF link status
> be updated, it should notify VF to update link status also.
> Current implementation only cover part of the situation when PF's link
> status is updated, call i40e_notify_all_vfs_link_status i
On 4/12/2017 5:51 PM, Ferruh Yigit wrote:
> On 4/12/2017 4:02 PM, Ferruh Yigit wrote:
>> On 4/11/2017 4:21 PM, Nelio Laranjeiro wrote:
>>> Flow queues array offset is not properly computed causing all sorts of
>>> issues. Address it by making the rte_flow structure less complex.
>>>
>>> Fixes: 360
Hi Ferruh,
-Original Message-
From: Yigit, Ferruh
Sent: Thursday, April 13, 2017 5:14 PM
To: Richardson, Bruce ; Lu, Wenzhuo
Cc: Zhang, Helin ; Olivier Matz
; dev@dpdk.org; Wu, Jingjing
Subject: Re: [dpdk-dev] [PATCH] net/i40e: disable init and driver logs by
default
On 4/12/2017 11
On 4/3/2017 4:13 PM, Jerin Jacob wrote:
> On Fri, Mar 31, 2017 at 03:57:48PM +0200, Andriy Berestovskyy wrote:
>> Some applications and DPDK examples expect link up/down
>> functionality to be provided.
>>
>> Signed-off-by: Andriy Berestovskyy
>> ---
>> drivers/net/thunderx/nicvf_ethdev.c | 14 ++
Hi Charles,
On Thu, Apr 13, 2017 at 07:52:34AM +, Myers, Charles wrote:
> I ran into an issue where the mlx4 driver stops receiving packets when mbuf
> allocation fails in mlx4_rx_burst().
>
> This issue appears to be caused because the code doesn't recycle the existing
> mbuf to the sges a
On 4/13/2017 1:05 PM, Adrien Mazarguil wrote:
> Hi Charles,
>
> On Thu, Apr 13, 2017 at 07:52:34AM +, Myers, Charles wrote:
>> I ran into an issue where the mlx4 driver stops receiving packets when mbuf
>> allocation fails in mlx4_rx_burst().
>>
>> This issue appears to be caused because the
On Thu, Apr 13, 2017 at 01:22:39PM +0100, Ferruh Yigit wrote:
> On 4/13/2017 1:05 PM, Adrien Mazarguil wrote:
> > Hi Charles,
> >
> > On Thu, Apr 13, 2017 at 07:52:34AM +, Myers, Charles wrote:
> >> I ran into an issue where the mlx4 driver stops receiving packets when
> >> mbuf allocation fa
On Mon, Apr 10, 2017 at 04:56:43PM +0100, Harry van Haaren wrote:
> The flow id of packets was not being hashed on ingress
> on an ordered queue. Fix by applying same hashing as is
> applied in the atomic queue case. The hashing itself is
> broken out into a macro to avoid duplication of code.
>
>
Acked-by: Jan Medala
Jan
-Original Message-
From: Marcin Wojtas [mailto:m...@semihalf.com]
Sent: Wednesday, April 12, 2017 4:22 PM
To: dev@dpdk.org
Cc: thomas.monja...@6wind.com; ferruh.yi...@intel.com; jan.med...@outlook.com;
jpali...@gmail.com; neta...@amazon.com; evge...@amazon.com;
I ran into an issue where the mlx4 driver stops receiving packets when mbuf
allocation fails in mlx4_rx_burst().
This issue appears to be caused because the code doesn't recycle the existing
mbuf to the sges array
when mbuf allocation fails as is done in the code right above it which handles
th
Extended xstats API in ethdev library to allow grouping of stats logically
so they can be retrieved per logical grouping managed by the application.
Changed existing functions rte_eth_xstats_get_names and rte_eth_xstats_get
to use a new list of arguments: array of ids and array of values.
ABI vers
From: Jacek Piasecki
Extended xstats API in ethdev library to allow grouping of stats
logically so they can be retrieved per logical grouping managed
by the application.
Changed existing functions rte_eth_xstats_get_names and
rte_eth_xstats_get to use a new list of arguments: array of ids
and ar
Introduced new function: rte_eth_xstats_get_id_by_name
to retrieve xstats ids by its names.
doc: added release note
Signed-off-by: Kuba Kozak
---
doc/guides/rel_notes/release_17_05.rst | 2 ++
lib/librte_ether/rte_ethdev.c | 44 ++
lib/librte_ether/rte_
There is a new argument --xstats-ids and --xstats-name
in proc_info command line to retrieve statistics given by ids
and by name.
E.g. --xstats-ids="1,3,5,7,8"
E.g. --xstats-name rx_errors
ethdev: mark functions as deprecated
Functions rte_eth_xstats_get_all and rte_eth_xstats_get_names_all
are m
From: Jacek Piasecki
To achieve functionality of retrieving only specific statistics
given by application there are two new functions added:
eth_igb_xstats_get_by_ids() which retrieve
values of statistics specified by ids array
and eth_igb_xstats_get_names_by_ids() which retrieve
names of statist
From: Jacek Piasecki
To achieve functionality of retrieving only specific statistics
given by application there are two new functions added:
ixgbe_dev_xstats_get_by_ids() which retrieve
values of statistics specified by ids array
and ixgbe_dev_xstats_get_names_by_ids() which retrieve
names of sta
Use correct config option, without CONFIG prefix.
Fixes: b2b0f85182ef ("kni: add build option for ethtool support")
Signed-off-by: Ferruh Yigit
---
lib/librte_eal/linuxapp/kni/kni_misc.c | 14 +++---
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/lib/librte_eal/linuxapp/k
Hi,
I have a proposal for mbuf race condition detection and I would like to get
your opinions before
committing any patch.
Race condition is the worst bug I can think of; as it causes crashing long
after the first crime scene,
and the consequence is unpredictable and difficult to understand.
Bes
After the changes in commit c1b5fa94a46f
("eal: support dynamic log types"), logtype is not treated as a
bitmask, but a decimal value. Therefore, values have to be
converted.
Fixes: c1b5fa94a46f ("eal: support dynamic log types")
Signed-off-by: Pablo de Lara
---
Changes in v3:
- Created array o
> -Original Message-
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> Sent: Wednesday, April 12, 2017 10:41 PM
> To: De Lara Guarch, Pablo
> Cc: olivier.m...@6wind.com; thomas.monja...@6wind.com;
> dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2] eal: redefine logtype valu
>
> Current ethdev always stores MAC address even it fails to be added.
> Other function may regard the failed MAC address valid and lead to
> some errors. So There is a need to check if the addr is added
> successfully or not and discard it if it fails.
>
> In 3rd patch, add a command "add_mor
virtio-user cannot work on 32-bit system as higher 32-bit of the
addr field (64-bit) in the desc is filled with non-zero value
which should not happen for a 32-bit system.
This is a regression bug. For 32-bit system, the first 4 bytes
is the virtual address, with following 8 bytes pointing to
phys
We only enabled LSC when using vhost-user as the backend, but it is
reported even when using vhost-kernel as the backend.
Fix it by only reportting LSC support when using vhost-user as the
backend.
Fixes: 35c4f8554833 ("net/virtio-user: support to report net status")
Signed-off-by: Jianfeng Tan
pci_enable_msix() will be removed in kernel 4.12. The new API is available since
linux 4.8, thus let's use it.
Link:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=aff171641d18
Link:
https://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next.git/commit/?id=4244
Hi Qinglai,
On Thu, Apr 13, 2017 at 04:38:19PM +0300, jigsaw wrote:
> Hi,
>
> I have a proposal for mbuf race condition detection and I would like to get
> your opinions before
> committing any patch.
>
> Race condition is the worst bug I can think of; as it causes crashing long
> after the firs
2017-04-13 10:42, Bruce Richardson:
> The error return code for rte_ring_dequeue() function should be -ENOENT
> rather than -ENOBUFS (which is the error value from the enqueue() fn).
>
> Fixes: cfa7c9e6fc1f ("ring: make bulk and burst return values consistent")
>
> Reported-by: Zhihong Wang
> Si
A segfault when flushing packets has been revealed by further testing
when using distributor in single packet mode. This patch resolves that.
Fixes: 775003ad2f96 ("distributor: add new burst-capable library")
Reported-by: Yong Liu
Signed-off-by: David Hunt
---
lib/librte_distributor/rte_distri
> From: Kozak, KubaX
> Sent: Thursday, April 13, 2017 3:59 PM
> To: dev@dpdk.org
> Cc: Van Haaren, Harry ; Jain, Deepak K
> ;
> Kozak, KubaX
> Subject: [PATCH v6 0/5] Extended xstats API in ethdev library to allow
> grouping of stats
>
> Extended xstats API in ethdev library to allow grouping o
> From: Kozak, KubaX
> Sent: Thursday, April 13, 2017 3:59 PM
> To: dev@dpdk.org
> Cc: Van Haaren, Harry ; Jain, Deepak K
> ;
> Piasecki, JacekX ; Kozak, KubaX
> ; Kulasek,
> TomaszX
> Subject: [PATCH v6 1/5] ethdev: new xstats API add retrieving by ID
>
> From: Jacek Piasecki
>
> Extended xs
> From: Kozak, KubaX
> Sent: Thursday, April 13, 2017 3:59 PM
> To: dev@dpdk.org
> Cc: Van Haaren, Harry ; Jain, Deepak K
> ;
> Kozak, KubaX
> Subject: [PATCH v6 2/5] ethdev: added new function for xstats ID
>
> Introduced new function: rte_eth_xstats_get_id_by_name
> to retrieve xstats ids by i
> From: Kozak, KubaX
> Sent: Thursday, April 13, 2017 3:59 PM
> To: dev@dpdk.org
> Cc: Van Haaren, Harry ; Jain, Deepak K
> ;
> Kozak, KubaX
> Subject: [PATCH v6 3/5] proc-info: add support for new xstats API
>
> There is a new argument --xstats-ids and --xstats-name
> in proc_info command line
> From: Kozak, KubaX
> Sent: Thursday, April 13, 2017 3:59 PM
> To: dev@dpdk.org
> Cc: Van Haaren, Harry ; Jain, Deepak K
> ;
> Piasecki, JacekX ; Kozak, KubaX
>
> Subject: [PATCH v6 4/5] net/e1000: new xstats API add ID support for e1000
>
> From: Jacek Piasecki
>
> To achieve functionality
> From: Kozak, KubaX
> Sent: Thursday, April 13, 2017 3:59 PM
> To: dev@dpdk.org
> Cc: Van Haaren, Harry ; Jain, Deepak K
> ;
> Piasecki, JacekX ; Kozak, KubaX
>
> Subject: [PATCH v6 5/5] net/ixgbe: new xstats API add ID support for ixgbe
>
> From: Jacek Piasecki
>
> To achieve functionality
After the refactoring of the capabilities, AES DOCSIS BPI
algorithm was left out.
Fixes: 6a3c87bc6a6c ("crypto/qat: refactor capabilities infrastructure")
Signed-off-by: Pablo de Lara
---
drivers/crypto/qat/qat_crypto_capabilities.h | 20
1 file changed, 20 insertions(+)
d
Current implementation is error-prone if the max inline size
(txq->max_inilne) is decoupled from txq->inline_en and becomes zero. If it
becomes zero, HW can crash due to WQ overflow.
Signed-off-by: Yongseok Koh
Acked-by: Shahaf Shuler
---
drivers/net/mlx5/mlx5_rxtx.c | 12 +++-
1 file c
Hi Adrien,
Thanks for your comment.
The LOCK/UNLOCK may be called by user application only. There are several
reasons.
1. If the lib calls LOCK, user application may be punished unexpectedly.
Consider what if the Rx burst function calls the LOCK in core #1, and then
the mbuf is handed over from
From: Charles
Fixes issue where mlx4 driver stops receiving packets when mbuf
allocation fails in mlx4_rx_burst().
This issue appears to be caused because the code doesn't recycle the
existing mbuf to the sges array when mbuf allocation fails as is done
in the code right above it which handles (
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of jigsaw
> Sent: Thursday, April 13, 2017 9:59 PM
> To: Adrien Mazarguil
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] Proposal of mbuf Race Condition Detection
>
> Hi Adrien,
>
> Thanks for your comment.
>
> The
On Thu, 13 Apr 2017 23:19:45 +
"Ananyev, Konstantin" wrote:
> > -Original Message-
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of jigsaw
> > Sent: Thursday, April 13, 2017 9:59 PM
> > To: Adrien Mazarguil
> > Cc: dev@dpdk.org
> > Subject: Re: [dpdk-dev] Proposal of mbuf Rac
When receiving the unsupported AQ messages, it's taken as
an error. It's not appropriate and triggers too much
unnecessary print.
Signed-off-by: Wenzhuo Lu
---
drivers/net/i40e/i40e_ethdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/driv
Signed-off-by: Wenzhuo Lu
---
MAINTAINERS | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 05f2f5f..29683d1 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -333,7 +333,7 @@ F: doc/guides/nics/e1000em.rst
F: doc/guides/nics/intel_vf.rst
Intel ix
Hi, Thomas Monjalon
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monja...@6wind.com]
> Sent: Thursday, April 13, 2017 6:07 PM
> To: Zhao1, Wei
> Cc: Ananyev, Konstantin ; Mcnamara, John
> ; dev@dpdk.org; Lu, Wenzhuo
>
> Subject: Re: [dpdk-dev] [PATCH v4 1/3] lib/librte_eth
Hi, Ferruh
> -Original Message-
> From: Yigit, Ferruh
> Sent: Thursday, April 13, 2017 6:21 PM
> To: Zhao1, Wei ; dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] net/ixgbe: remove tpid check for fdir filter
>
> On 4/13/2017 8:49 AM, Wei Zhao wrote:
> > DPDK community has several emails dis
testpmd> set vf tx max-bandwidth (port_id) (vf_id) (bandwidth)
bandwidth(Mbps)
I think that Mbps usually refer to physical layers.
Bet Regards
Yulong Pei
-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ferruh Yigit
Sent: Thursday, April 13, 2017 6:26 PM
To: Lu,
DPDK community has several emails discussion on this topic,
these mails link is bellow:
http://dpdk.org/ml/archives/dev/2017-March/060379.html,
http://dpdk.org/ml/archives/dev/2017-March/060295.html,
items like VLAN can already have several valid "types"
(0x88a8, 0x8100, 0x9100), and who knows what
> -Original Message-
> From: Yigit, Ferruh
> Sent: Tuesday, April 11, 2017 11:21 PM
> To: Zhang, Helin ; Ananyev, Konstantin
>
> Cc: dev@dpdk.org; Yigit, Ferruh ; Lu, Wenzhuo
> ; Zhao1, Wei
> Subject: [PATCH] net/ixgbe: fix duplicated check
>
> Same check duplicated, updated check acco
On Thu, Apr 13, 2017 at 10:11:27AM +, Jianfeng Tan wrote:
> The feature negotiation in virtio-user is proven to be broken,
> which results in device initialization failure.
>
> Originally, we get features from vhost backend, and remove those
> that are not supported. But when new feature is ad
> -Original Message-
> From: Yuanhan Liu [mailto:yuanhan@linux.intel.com]
> Sent: Friday, April 14, 2017 12:24 PM
> To: Tan, Jianfeng
> Cc: dev@dpdk.org; maxime.coque...@redhat.com
> Subject: Re: [PATCH v2] net/virtio-user: fix cannot get initialized
>
> On Thu, Apr 13, 2017 at 10:11
On Thu, Apr 13, 2017 at 02:16:24PM +, Jianfeng Tan wrote:
> We only enabled LSC when using vhost-user as the backend, but it is
> reported even when using vhost-kernel as the backend.
>
> Fix it by only reportting LSC support when using vhost-user as the
> backend.
>
> Fixes: 35c4f8554833 ("n
Hi Yulong,
> -Original Message-
> From: Pei, Yulong
> Sent: Friday, April 14, 2017 11:07 AM
> To: Yigit, Ferruh; Lu, Wenzhuo; dev@dpdk.org
> Subject: RE: [dpdk-dev] [PATCH] net/i40e: add more explanation for QoS APIs
>
> testpmd> set vf tx max-bandwidth (port_id) (vf_id) (bandwidth)
>
>
Hi Olivier,
If I understand it correctly, this patch is to shift log level setting
earlier. But we did not remove the one in eal_parse_common_option(). So
we can see this parameter will be analyzed twice. Does it make sense to
remove analysis of log level in eal_parse_common_option()?
Thanks
On Thu, Apr 13, 2017 at 02:12:56PM +, Jianfeng Tan wrote:
> virtio-user cannot work on 32-bit system as higher 32-bit of the
> addr field (64-bit) in the desc is filled with non-zero value
> which should not happen for a 32-bit system.
>
> This is a regression bug. For 32-bit system, the first
Hi Olivier,
On 4/5/2017 12:40 AM, Olivier Matz wrote:
Introduce a function to set the log level of several log types that
match a regular expression.
Signed-off-by: Olivier Matz
---
lib/librte_eal/bsdapp/eal/rte_eal_version.map | 1 +
lib/librte_eal/common/eal_common_log.c | 21
On Fri, Apr 14, 2017 at 04:35:57AM +, Tan, Jianfeng wrote:
>
>
> > -Original Message-
> > From: Yuanhan Liu [mailto:yuanhan@linux.intel.com]
> > Sent: Friday, April 14, 2017 12:24 PM
> > To: Tan, Jianfeng
> > Cc: dev@dpdk.org; maxime.coque...@redhat.com
> > Subject: Re: [PATCH v2]
> -Original Message-
> From: Yuanhan Liu [mailto:yuanhan@linux.intel.com]
> Sent: Friday, April 14, 2017 1:32 PM
> To: Tan, Jianfeng
> Cc: dev@dpdk.org; olivier.m...@6wind.com; sta...@dpdk.org
> Subject: Re: [PATCH] net/virtio-user: fix not working on 32-bit system
>
> On Thu, Apr 13
> -Original Message-
> From: Yuanhan Liu [mailto:yuanhan@linux.intel.com]
> Sent: Friday, April 14, 2017 1:38 PM
> To: Tan, Jianfeng
> Cc: dev@dpdk.org; maxime.coque...@redhat.com
> Subject: Re: [PATCH v2] net/virtio-user: fix cannot get initialized
>
> On Fri, Apr 14, 2017 at 04:35:
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of David Hunt
> Sent: Thursday, April 13, 2017 4:36 PM
> To: dev@dpdk.org
> Cc: Hunt, David
> Subject: [dpdk-dev] [PATCH v1] lib/distributor: fix segfaults in flush
>
> A segfault when flushing packets has been revea
Previously, we miss to set intr_handle->fd which will be used as
target file for epoll to check LSC.
As a result, stdin (0) is used and intr thread keeps busy whenever
data comes from stdin.
To fix this, we use vhostfd as the target file for epoll to check
the link status change events. And we mo
On Fri, Apr 14, 2017 at 05:53:55AM +, Tan, Jianfeng wrote:
> > > diff --git a/drivers/net/virtio/virtqueue.h
> > > b/drivers/net/virtio/virtqueue.h
> > > index f9e3736..f43ea70 100644
> > > --- a/drivers/net/virtio/virtqueue.h
> > > +++ b/drivers/net/virtio/virtqueue.h
> > > @@ -72,7 +72,8 @@
2017-04-14 01:29, Zhao1, Wei:
> From: Thomas Monjalon [mailto:thomas.monja...@6wind.com]
> > 2017-04-13 08:55, Zhao1, Wei:
> > > From: Ananyev, Konstantin
> > > > From: Zhao1, Wei
> > > > > From: Thomas Monjalon [mailto:thomas.monja...@6wind.com]
> > > > > > 2017-04-06 02:57, Zhao1, Wei:
> > > > >
The virtio port link status will always be UP, even the port is stopped:
testpmd> port stop 0
Stopping ports...
Checking link statuses...
Port 0 Link Up - speed 1 Mbps - full-duplex
Done
The link status is queried by link_update callback when LSC is disabled.
Which in turn
1 - 100 of 102 matches
Mail list logo
