On 10/17/22 10:10, Andrew Rybchenko wrote:
On 10/17/22 15:32, Chas Williams wrote:
This appears to be correct. A minor comment inline.
On 10/17/22 04:42, Andrew Rybchenko wrote:
Chas, Cornor, could you review the patch, please.
Thanks,
Andrew.
On 9/11/22 15:24, Ivan Malov wrote
Acked-by: Chas Williams <3ch...@gmail.com>
On 10/17/22 04:40, Andrew Rybchenko wrote:
Chas, Cornor, could you review the patch, please.
Thanks,
Andrew.
On 9/11/22 15:19, Ivan Malov wrote:
Commit 5be3b40fea60 ("net/bonding: fix values of descriptor limits")
breaks reporting
This appears to be correct, but it needs to be coordinated with the
proposed changes in net/bonding: make bonded device configure method re-entrant
On 10/17/22 04:41, Andrew Rybchenko wrote:
Chas, Cornor, could you review the patch, please.
Thanks,
Andrew.
On 9/11/22 15:22, Ivan Malov wrote:
This appears to be correct. A minor comment inline.
On 10/17/22 04:42, Andrew Rybchenko wrote:
Chas, Cornor, could you review the patch, please.
Thanks,
Andrew.
On 9/11/22 15:24, Ivan Malov wrote:
According to the documentation, rte_eth_dev_configure()
can be invoked repeatedly while in stopp
Thanks for making the change, IMHO this is much clearer.
Signed-off-by: Chas Williams <3ch...@gmail.com>
On 9/29/22 00:05, Usman Tanveer wrote:
when dedicated queues are enable with bonding mode 4 (mlx5), the
application sets the flow, which cannot be set if the device is not
started
This looks fine. Thanks for making the changes!
Signed-off-by: Chas Williams <3ch...@gmail.com>
On 10/11/22 09:20, Chengwen Feng wrote:
Normally, to use the HW offloads capability (e.g. checksum and TSO) in
the Tx direction, the application needs to call rte_eth_tx_prepare() to
d
On 10/8/22 23:36, Chengwen Feng wrote:
uint16_t slaves[RTE_MAX_ETHPORTS];
uint8_t tx_failed_flag = 0;
uint16_t num_of_slaves;
+ uint16_t num_tx_prep;
uint16_t max_nb_of_tx_pkts = 0;
@@ -1320,12 +1339,18 @@ bond_ethdev_tx_burst_broadcast(void *queue, struc
On 9/26/22 06:18, Konstantin Ananyev wrote:
Hi everyone,
Sorry for late reply.
The main problem is hard to design a tx_prepare for bonding device:
1. as Chas Williams said, there maybe twice hash calc to get target slave
devices.
2. also more important, if the slave devices have
It's probably cleaner to just move the bond_ethdev_8023ad_flow_set
until after the device start. For the reader, they don't need to
understand why you might not have started the device earlier.
On 9/24/22 10:19, Usman Tanveer wrote:
when dedicated queues are enable with bonding mode 4 (mlx5), th
Thanks for making the changes!
Signed-off-by: 3ch...@gmail.com
On 9/21/22 21:33, Huisong Li wrote:
This patch adds link speeds configuration.
---
-v3: add an intersection of the supported speeds to check 'link_speeds'.
-v2: resend due to CI compiling failure.
Signed-off-by: Huisong Li
--
On 9/21/22 22:12, fengchengwen wrote:
On 2022/9/20 7:02, Chas Williams wrote:
On 9/19/22 10:07, Konstantin Ananyev wrote:
On 9/16/22 22:35, fengchengwen wrote:
Hi Chas,
On 2022/9/15 0:59, Chas Williams wrote:
On 9/13/22 20:46, fengchengwen wrote:
The main problem is hard to
On 9/19/22 10:07, Konstantin Ananyev wrote:
On 9/16/22 22:35, fengchengwen wrote:
Hi Chas,
On 2022/9/15 0:59, Chas Williams wrote:
On 9/13/22 20:46, fengchengwen wrote:
The main problem is hard to design a tx_prepare for bonding device:
1. as Chas Williams said, there maybe twice hash
On 9/16/22 22:35, fengchengwen wrote:
Hi Chas,
On 2022/9/15 0:59, Chas Williams wrote:
On 9/13/22 20:46, fengchengwen wrote:
The main problem is hard to design a tx_prepare for bonding device:
1. as Chas Williams said, there maybe twice hash calc to get target slave
devices.
2. also
On 9/15/22 22:09, lihuisong (C) wrote:
在 2022/9/15 21:43, Chas Williams 写道:
On 9/15/22 09:14, Huisong Li wrote:
This patch adds link speeds configuration.
---
-v2: resend due to CI compiling failure.
Signed-off-by: Huisong Li
---
drivers/net/bonding/rte_eth_bond_pmd.c | 2 ++
1
On 9/15/22 09:14, Huisong Li wrote:
This patch adds link speeds configuration.
---
-v2: resend due to CI compiling failure.
Signed-off-by: Huisong Li
---
drivers/net/bonding/rte_eth_bond_pmd.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/bonding/rte_eth_bond_pmd.c
b/
On 9/13/22 20:46, fengchengwen wrote:
The main problem is hard to design a tx_prepare for bonding device:
1. as Chas Williams said, there maybe twice hash calc to get target slave
devices.
2. also more important, if the slave devices have changes(e.g. slave device
link down or remove
On 9/13/22 06:22, Ferruh Yigit wrote:
> On 7/25/2022 5:08 AM, Chengwen Feng wrote:
>
>
> I assume intention is to make this as transparent as possible to the
> user, that is why you are using a wrapper that combines
> `rte_eth_tx_prepare()` & `rte_eth_tx_burst()` APIs. But for other PMDs
> `rte_eth
rte_raw_cksum_mbuf can fail, so we should check to see if it
has. If so, return with an error.
Fixes: 96cb6711939e ("net/virtio: support Rx checksum offload")
Signed-off-by: Chas Williams <3ch...@gmail.com>
---
drivers/net/virtio/virtio_rxtx.c| 5 +++--
dr
If the overall pkt_len and segment lengths are out of agreement,
it is possible for the seg to be NULL after the loop. Add assert
to check this condition in debug builds. Otherwise, return failure.
Fixes: c442fed81bb9 ("net: add function to calculate checksum in mbuf")
Signed-of
On 9/28/20 11:01 PM, wangyunjian wrote:
>> -Original Message-
>> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Chas Williams
>> Sent: Monday, September 28, 2020 11:32 PM
>> To: dev@dpdk.org
>> Cc: olivier.m...@6wind.com; Chas Williams <3ch...@gmail.c
If the overall pkt_len and segment lengths are out of agreement,
it is possible for the seg to be NULL after the loop. Add assert
to check this condition in debug builds.
Fixes: c442fed81bb9 ("net: add function to calculate checksum in mbuf")
Signed-off-by: Chas Williams <3ch
ol.
Fixes: cc0251813277 ("net/netvsc: split send buffers from Tx descriptors")
Cc: sta...@dpdk.org
Signed-off-by: Chas Williams <3ch...@gmail.com>
---
drivers/net/netvsc/hn_rxtx.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/net/netvsc/hn_r
Yes, this does look like a typo on someone's part.
Please change use net/bonding: instead of pmd_bond:
Please add a Fixes: line
Please add a Cc: sta...@dpdk.org
On 6/17/20 3:02 PM, Sergey Lvov wrote:
> It seems to be a typo
>
> Signed-off-by: Sergey Lvov
> ---
> drivers/net/bonding/rte_eth_bo
On 6/16/20 11:45 AM, Stephen Hemminger wrote:
> On Tue, 16 Jun 2020 09:52:01 -0400
> Chas Williams <3ch...@gmail.com> wrote:
>
>> On 6/16/20 7:48 AM, Jay Rolette wrote:
>> > On Mon, Jun 15, 2020 at 5:52 PM Stephen Hemminger <
>> > step...@netw
On 6/16/20 7:48 AM, Jay Rolette wrote:
> On Mon, Jun 15, 2020 at 5:52 PM Stephen Hemminger <
> step...@networkplumber.org> wrote:
>
>> I am disturbed by the wide spread use of master/slave in Ethernet
bonding.
>> Asked the current IEEE chairs and it looks like it is already fixed
>> "upstream
Acked-by: Chas Williams
On 6/16/20 5:46 AM, David Marchand wrote:
Caught by code review, rte_eth_dev_socket_id() returns -1 on error.
The code should behave the same, but still, do not use LCORE_ID_ANY for
something that is not a lcore id.
Fixes: c15c5897340d ("net/bonding: avoid alloc
On 6/15/20 6:52 PM, Stephen Hemminger wrote:
> I am disturbed by the wide spread use of master/slave in Ethernet
bonding.
> Asked the current IEEE chairs and it looks like it is already fixed
"upstream".
>
> The proper terminology is for Ethernet link aggregation in the
> the current standard
100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -529,6 +529,7 @@ F: doc/guides/nics/features/default.ini
Link bonding
M: Chas Williams
+M: Wei Hu (Xavier)
F: drivers/net/bonding/
F: doc/guides/prog_guide/link_bonding_poll_mode_drv_lib.rst
F: app/test/test_link_bonding*
Acked-by
The behavior is probably going to be inconsistent. Only one of the calls
to mac_address_slaves_update() is checked for failure. The two calls
are in bond_ethdev_lsc_event_callback()
if (internals->active_slave_count < 1) {
mac_address_slaves_update(bonded_e
It looks like the cxgbe driver doesn't set the DEV_TX_OFFLOAD_MULTI_SEGS
offload capability. However, it looks like sge.c does have support for
sending mbuf chains, e.g. write_sgl(). Is this just an oversight or is
there something fundamentally broken here?
On 12/9/19 9:31 AM, Andrew Rybchenko wrote:
> On 12/9/19 4:36 PM, Chas Williams wrote:
>>
>>
>> On 12/9/19 2:16 AM, Andrew Rybchenko wrote:
>>> On 12/8/19 6:44 PM, Chas Williams wrote:
>>>> On 2019-11-19 07:40, Andrew Rybchenko wrote:
>>>>&
On 12/9/19 2:16 AM, Andrew Rybchenko wrote:
> On 12/8/19 6:44 PM, Chas Williams wrote:
>> On 2019-11-19 07:40, Andrew Rybchenko wrote:
>>> On 11/19/19 3:18 PM, Ferruh Yigit wrote:
>>>> On 11/19/2019 9:03 AM, Andrew Rybchenko wrote:
>>>>>
On 2019-11-19 07:40, Andrew Rybchenko wrote:
> On 11/19/19 3:18 PM, Ferruh Yigit wrote:
>> On 11/19/2019 9:03 AM, Andrew Rybchenko wrote:
>>> Bonding device should control bonded devices configuration.
>>>
>>> Also avoid usage of slave's data->dev_conf.
>>>
>>> Fixes: 2efb58cbab6e ("bond: new l
et/bonding: add dedicated HW queues for LACP control")
Cc: sta...@dpdk.org
Signed-off-by: Jiang JunyuX
Acked-by: Chas Williams
Patchwork says that there is a build error but it doesn't seem to
be related to this change.
---
drivers/net/bonding/rte_eth_bond_8023ad.c | 32 +++
This looks better. While reviewing this I noticed that a few lines:
case AGG_STABLE:
if (default_slave == slaves_count)
new_agg_id = slave_id; <
else
new_agg_id = slaves[default_slave];
On 10/10/19 4:27 AM, kka...@marvell.com wrote:
From: Krzysztof Kanas
Arrays agg_count and agg_bandwidth should be indexed by slave_id not by
aggregator port_id.
Fixes: 6d72657ce379 ("net/bonding: add other aggregator modes")
Cc: danielx.t.mrzyg...@intel.com
Signed-off-by: Krzysztof Kanas
Acked-by: Chas Williams
On 9/10/19 4:25 AM, Andrew Rybchenko wrote:
From: Igor Romanov
Fix the issue that the link speed of the bond device was set to the
link speed of the first active slave in broadcast mode.
Set the link speed of the bond device to the minimum value across
all of the
On 9/5/19 5:32 AM, Andrzej Ostruszka wrote:
During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized. This patch silences these
warnings.
Signed-off-by: Andrzej Ostruszka
---
app/test/test_hash_readwrite.c | 2 +-
app/test/tes
On 7/2/19 11:01 AM, Ferruh Yigit wrote:
On 5/16/2019 10:12 AM, David Marchand wrote:
Hello Chas,
On Fri, Apr 19, 2019 at 12:50 AM Chas Williams <3ch...@gmail.com> wrote:
On 4/18/19 3:11 AM, David Marchand wrote:
Hello Chas,
On Fri, Apr 12, 2019 at 4:02 PM Chas William
On 6/27/19 4:08 AM, Ferruh Yigit wrote:
On 4/10/2019 1:53 PM, David Marchand wrote:
Another series with focus on the fast/normal rx/tx handlers for 802.3ad.
The first two patches make sure that the rx (resp. tx) fast and normal
handlers are equivalent.
The third one will most likely have an
On 6/19/19 7:29 AM, lidejun wrote:
Why doesn’t DPDK support `DEFAULTED` state in Receive machine in 802.3ad bond
implementation？
Can you be a bit more specific about your question? This is some support
for a DEFAULTED state.
Seems fine.
Acked-by: Chas Williams
On 4/29/19 10:56 AM, Ferruh Yigit wrote:
On 4/26/2019 11:30 PM, kka...@marvell.com wrote:
From: Krzysztof Kanas
Fix test_set_bonded_port_initialization_mac_assignment so that it works
after 're run' test_link_bonding.
Fixes: f2ef6f21ee2e (&quo
a huge concern.
Acked-by: Chas Williams
---
v2: add fixes lines
v3: fix buffer end calculation
drivers/net/bonding/rte_eth_bond_pmd.c | 9 +++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/drivers/net/bonding/rte_eth_bond_pmd.c
b/drivers/net/bonding/rte_eth_bond_pmd.
On 4/18/19 3:11 AM, David Marchand wrote:
Hello Chas,
On Fri, Apr 12, 2019 at 4:02 PM Chas Williams <3ch...@gmail.com
<mailto:3ch...@gmail.com>> wrote:
I should have some time this weekend to run these patches through our
regression system.
Did you manage to run
erruh Yigit
---
Cc: Stephen Hemminger
Cc: Chas Williams
This is another approach to RFC to fix the vlan_insert:
https://patches.dpdk.org/patch/51870/
vlan_insert() mostly used by drivers to insert VLAN tag into packet
data in Tx path, drivers creating new copies of mbufs in Tx path may
r
On 4/16/19 11:51 AM, Ferruh Yigit wrote:
The vlan_insert() is buggy when it tires to handle the shared mbufs,
s/tries/tries/
instead don't support inserting VLAN tag into shared mbufs and return
an error for that case.
Signed-off-by: Ferruh Yigit
---
Cc: Stephen Hemminger
Cc:
I should have some time this weekend to run these patches through our
regression system.
On 4/10/19 8:53 AM, David Marchand wrote:
fast queue Rx burst function is missing checks on promisc and the
slave collecting state.
Define an inline wrapper to have a common base.
Fixes: 112891cd27e5 ("net/
On 4/4/19 7:54 PM, Stephen Hemminger wrote:
On Sat, 30 Mar 2019 08:41:33 -0400
Chas Williams <3ch...@gmail.com> wrote:
Unfortunately, I think the complete fix is more complicated than this.
Drivers that use rte_vlan_insert don't anticipate that the mbuf might
change and tha
Unfortunately, I think the complete fix is more complicated than this.
Drivers that use rte_vlan_insert don't anticipate that the mbuf might
change and that (hardware) transmit can fail.
They make a copy of the mbuf pointer from the incoming transmit list
and don't update the original if rte_vlan
On 3/27/19 11:18 AM, Stephen Hemminger wrote:
On Tue, 26 Mar 2019 18:38:57 -0400
Chas Williams <3ch...@gmail.com> wrote:
On 3/26/19 3:15 PM, Stephen Hemminger wrote:
If mbuf refcnt was > 1 then rte_vlan_insert() would incorrectly
modify the original copy. Original code was expect
On 3/26/19 3:15 PM, Stephen Hemminger wrote:
If mbuf refcnt was > 1 then rte_vlan_insert() would incorrectly
modify the original copy. Original code was expecting clone to make
a copy (it doesn't). Better to let the caller deal with making
a copy or setting up mbuf chain to allow for header to be
On 3/26/19 12:50 PM, Ferruh Yigit wrote:
On 3/25/2019 3:05 PM, Chas Williams wrote:
From: Bill Hong
A PMD might use rte_vlan_insert to implement Tx VLAN offload. Typically
the PMD will insert the VLAN header in the transmit path and then attempt
to send the packets. If this fails, the
PKT_TX_VLAN flag is not cleared, the transmit path may attempt to insert
the VLAN header again.
Fixes: 47aa48b969f8 ("net: fix stripped VLAN flag for offload emulation");
Cc: sta...@dpdk.org
Signed-off-by: Bill Hong
Signed-off-by: Chas Williams
---
lib/librte_net/rte_ether.h | 2
On 3/24/19 1:11 PM, David Marchand wrote:
On Sun, Mar 24, 2019 at 2:35 PM Chas Williams <3ch...@gmail.com
<mailto:3ch...@gmail.com>> wrote:
Have you ever experienced this problem in practice? I ask because I am
considering some fixes that would limit the number of slav
avid Marchand
Acked-by: Chas Williams
---
drivers/net/bonding/rte_eth_bond_8023ad.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/bonding/rte_eth_bond_8023ad.c
b/drivers/net/bonding/rte_eth_bond_8023ad.c
index 3943ec1..5004898 100644
--- a/driver
See inline.
On 3/21/19 4:28 PM, David Marchand wrote:
From: Zhaohui
mode_bond_id and mode_band_id are slave ids, stored on 16bits.
Please change mode_bond_id to mode_count_id in this commit message.
Otherwise, this looks fine.
Fixes: f8244c6399d9 ("ethdev: increase port id range")
Cc: sta.
On 3/19/19 1:37 PM, Ferruh Yigit wrote:
On 3/6/2019 3:42 AM, Zhaohui (zhaohui, Polestar) wrote:
When the number of slave slave devices exceeds 8, it will cause the array
subscript to cross the boundary.
---
drivers/net/bonding/rte_eth_bond_8023ad.c | 4 ++--
1 file changed, 2 insertions(+), 2 de
On 3/19/19 5:41 AM, h...@netitest.com wrote:
Hi Guys,
I found a bug in dpdk bond code, while one port has 2 more tx queues,
8023ad bond port will be crashed in tx burst.
Just analyzed the code below, if 2 more CPU cores send packets on a port
by different tx queue, the arrays like
slave_po
On 3/15/19 5:06 AM, Liang Zhang wrote:
When monitor(port-mirroring) traffic from other lacp port-channel,
rx_machine_update may recieving other lacp negotiation packets.
Thus bond mode 4 will negotiation failed.
Signed-off-by: Liang Zhang
Acked-by: Chas Williams
---
drivers/net/bonding
At a minimum you need to address the long lines in your patch. See
http://mails.dpdk.org/archives/test-report/2019-March/076663.html
otherwise this looks fine to me.
On 3/13/19 9:23 AM, Liang Zhang wrote:
When monitor(port-mirroring) traffic from other lacp port-channel,
rx_machine_update may
On 3/6/19 4:42 AM, Zhaohui (zhaohui, Polestar) wrote:
Hi Williams：
We have some questions that need your help. About “Segfault when eal
thread executing mlx5 nic‘s lsc event”.
In order to solve this problem(core dump), What should we do?( Looking forward
to your reply）
The link sta
du Nicolau
Acked-by: Chas Williams
<mailto:radu.nico...@intel.com>>
Hi Radu, Hari,
There is another bonding patch, can you please check how related are they and if
are these fixing same root cause:
net/bonding: avoid the next active slave going out of bound
https://patches.dpdk.org/pat
From: Chas Williams
Copying the link properties of the first slave added may copy an
invalid link status. The speed and duplex of the slave may not
be known at this time. Delay setting the properties until the
first slave reports as link up. Note that we are still ignoring
an error from
From: Chas Williams
The tx burst routine always needs to check for pending LACPDUs
and send them if available. Do this first to prioritize the
control traffic. We can still early exit, before calculating
the distribution slaves, if there isn't any data packets.
Fixes: 09150784a776
We can transmit if there is at least one distributing slave.
Fixes: 09150784a776 ("net/bonding: burst mode hash calculation")
Cc: sta...@dpdk.org
Signed-off-by: Chas Williams
---
drivers/net/bonding/rte_eth_bond_pmd.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
How strange. I was just looking at this issue this weekend. I think
we can just move the control packet handling before the data packet
handling. That avoids the goto and is a little more sensible -- we should
try to prioritize control traffic (even if we can't guarantee there will
be space it's b
g
Signed-off-by: Hyong Youb Kim
Acked-by: Chas Williams
---
drivers/net/bonding/rte_eth_bond_pmd.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/net/bonding/rte_eth_bond_pmd.c
b/drivers/net/bonding/rte_eth_bond_pmd.c
index 44deaf119..daf2440cd 100644
--- a/dri
d-off-by: Hyong Youb Kim
Acked-by: Chas Williams
---
drivers/net/bonding/rte_eth_bond_pmd.c | 14 ++
1 file changed, 14 insertions(+)
diff --git a/drivers/net/bonding/rte_eth_bond_pmd.c
b/drivers/net/bonding/rte_eth_bond_pmd.c
index daf2440cd..bc2405e54 100644
--- a/driv
Sort the options alphabetically and make a note of the preferred
order at the top of the file.
Signed-off-by: Chas Williams <3ch...@gmail.com>
---
meson_options.txt | 14 --
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/meson_options.txt b/meson_options.txt
Allow users and packagers to override the default RTE_MAX_ETHPORTS.
This adds a new meson option, max_ethports which defaults to the
current value.
Signed-off-by: Chas Williams <3ch...@gmail.com>
Acked-by: Bruce Richardson
---
config/meson.build | 1 +
config/rte_config
On 2/4/19 4:49 AM, Bruce Richardson wrote:
On Thu, Jan 31, 2019 at 06:53:25PM -0500, Chas Williams wrote:
Allow users and packagers to override the default RTE_MAX_ETHPORTS.
This adds a new meson option, max_ethports which defaults to the
current value.
Actually in this v2, you have
Sort the options alphabetically and make a note of the preferred
order at the top of the file.
Signed-off-by: Chas Williams <3ch...@gmail.com>
---
meson_options.txt | 14 --
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/meson_options.txt b/meson_options.txt
Allow users and packagers to override the default RTE_MAX_ETHPORTS.
This adds a new meson option, max_ethports which defaults to the
current value.
Signed-off-by: Chas Williams <3ch...@gmail.com>
Acked-by: Bruce Richardson
---
config/meson.build | 1 +
config/rte_config
library")
Cc: sta...@dpdk.org
Signed-off-by: Hari Kumar Vemula
Acked-by: Chas Williams
---
v2: bonded device desc_lim values are received from slave configuration
---
drivers/net/bonding/rte_eth_bond_pmd.c | 14 ++
1 file changed, 10 insertions(+), 4 deletions(-)
diff --gi
On 1/31/19 4:32 AM, Bruce Richardson wrote:
On Wed, Jan 30, 2019 at 09:18:58PM -0500, Chas Williams wrote:
Allow users and packagers to override the default RTE_MAX_ETHPORTS.
This adds a new meson option, max_ethports which defaults to the
current value.
Signed-off-by: Chas Williams <
Allow users and packagers to override the default RTE_MAX_ETHPORTS.
This adds a new meson option, max_ethports which defaults to the
current value.
Signed-off-by: Chas Williams <3ch...@gmail.com>
---
config/meson.build | 1 +
config/rte_config.h | 1 -
meson_options.txt | 2 ++
3
On 1/7/19 8:01 AM, Hari Kumar Vemula wrote:
Create bonded device test is failing due to improper initialisation in
bonded device configuration. which leads to crash while setting up queues.
The value of nb_rx_desc is checked if it is not in range of rx_desc_lim of
bonded device which fails.
T
AM, Linhaifeng wrote:
Hi, Chars
Thank you.
I use it for send pkts to the dedicated queue of slaves.
Maybe i should not use it. I would though another way.
-邮件原件-
发件人: Chas Williams [mailto:3ch...@gmail.com]
发送时间: 2018年11月30日 11:27
收件人: Linhaifeng ; dev@dpdk.org
抄送: ch...@att.com
主题
I guess this is slightly more correct. There is still a race here though.
After you make your copy of active_slave_count, the number of active
slaves could go to 0 and the memcpy() would copy an invalid element,
acitve_slaves[0]. There is no simple fix to this problem. Your patch
reduces the opp
On 11/28/18 11:04 AM, Radu Nicolau wrote:
On 11/28/2018 2:28 PM, Chas Williams wrote:
On 11/28/2018 08:48 AM, Radu Nicolau wrote:
Hi
On 11/28/2018 11:08 AM, Ferruh Yigit wrote:
On 11/14/2018 12:19 PM, Radu Nicolau wrote:
Do not start the packet processing threads until all
On 11/28/2018 08:48 AM, Radu Nicolau wrote:
Hi
On 11/28/2018 11:08 AM, Ferruh Yigit wrote:
On 11/14/2018 12:19 PM, Radu Nicolau wrote:
Do not start the packet processing threads until all configured
slaves become active.
Hi Radu,
What happens if packet processing threads started before a
On 11/14/2018 07:19 AM, Radu Nicolau wrote:
Do not start the packet processing threads until all configured
slaves become active.
Signed-off-by: Radu Nicolau
Acked-by: Chas Williams
---
examples/bond/main.c | 15 +++
1 file changed, 15 insertions(+)
diff --git a
On 11/13/2018 11:46 AM, Radu Nicolau wrote:
Queue setup will fail if called before adding slaves.
Fixes: 7a0665940fa8 ("net/bonding: inherit descriptor limits from slaves")
Cc: sta...@dpdk.org
Signed-off-by: Radu Nicolau
Acked-by: Chas Williams
---
examples/bond/m
From: Chas Williams
The same issue was fixed on for the ipv4 version of this routine in
commit 8d4d3a4f7337 ("ip_frag: handle MTU sizes not aligned to 8 bytes").
Briefly, the size of an ipv6 header is always 40 bytes. With an MTU of
1500, this will never produce a multiple of 8 byt
;)
Cc: sta...@dpdk.org
Signed-off-by: Radu Nicolau
Acked-by: Chas Williams
---
drivers/net/bonding/rte_eth_bond_pmd.c | 11 ---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/drivers/net/bonding/rte_eth_bond_pmd.c
b/drivers/net/bonding/rte_eth_bond_pmd.c
index 1a6d8e4..2661
On 11/07/2018 04:17 AM, Zhao1, Wei wrote:
Hi, Luca Boccassi
The purpose of this patch is to reduce the mailbox interrupt from vf to
pf, but there seem some point need for discussion in this patch.
First, I do not know why do you change code of function ixgbe_check_mac_link_vf(), beca
On 11/05/2018 12:41 PM, Zhang, Qi Z wrote:
-Original Message-
From: Luca Boccassi [mailto:bl...@debian.org]
Sent: Friday, November 2, 2018 8:19 AM
To: dev@dpdk.org
Cc: Lu, Wenzhuo ; Ananyev, Konstantin
; Zhang, Qi Z ;
3ch...@gmail.com; Luca Boccassi ; sta...@dpdk.org
Subject: [PATCH
On 11/02/2018 10:33 AM, Ferruh Yigit wrote:
On 11/1/2018 2:45 PM, Luca Boccassi wrote:
On Mon, 2017-07-17 at 19:05 -0400, Charles (Chas) Williams wrote:
.dev_uninit calls .dev_stop and .dev_close. The work that is done in
those routines doesn't need repeated. Use started and open
river assignment to end of probing")
Fixes: 1620175b400e ("net/bonding: fix invalid port id")
Signed-off-by: Radu Nicolau
Acked-by: Chas Williams
---
v2: reverted earlier patch
drivers/net/bonding/rte_eth_bond_pmd.c | 8 +++-
1 file changed, 3 insertions(+), 5 deletions(-)
diff
On 10/31/2018 11:06 AM, Thomas Monjalon wrote:
31/10/2018 14:59, Radu Nicolau:
After the patch below the call to rte_eth_bond_8023ad_agg_selection_set
from probe() segfaults; there is no need to call the function, just set
the mode directly.
Fixes: 391797f04208 ("drivers/bus: move driver ass
From: Chas Williams
Calling rte_eth_macaddr_get to get a copy of the MAC address causes
a hot spot according to profiling. We can easily get the current
MAC address by just examining the bonded device.
Signed-off-by: Chas Williams
---
drivers/net/bonding/rte_eth_bond_pmd.c | 14
Any other error messages? There's really only one path out of
slave_configure() that doesn't emit another error message and
that's slave_configure_slow_queue. You need to set dedicated_queues to
be enabled to see that happen.
On 10/07/18 01:12, chetan bhasin wrote:
Can anybody suggest? Stuc
What does (or should) rte_eth_dev_set_mtu() do? The documentation says
"Change the MTU of an Ethernet device." At least of few of the PMDs do
something similar to the following:
eth_em_infos_get(dev, &dev_info);
frame_size = mtu + ETHER_HDR_LEN + ETHER_CRC_LEN + ...
/*
On Wed, Sep 5, 2018 at 5:14 AM Andrew Rybchenko
wrote:
>
> From: Ivan Malov
>
> Descriptor limits are used by applications to take
> optimal decisions on queue sizing.
>
> Signed-off-by: Ivan Malov
> Signed-off-by: Andrew Rybchenko
Acked-by: Chas Williams
>
m exercising default settings in the same way.
>
> Restructure the code to collect various settings
> from slave ports and make it possible to combine
> default Rx/Tx configuration of these devices and
> report it to the callers of rte_eth_dev_info_get.
>
> Signed-off-by: Ivan
From: Chas Williams
Some PMDs, especially ones with vector receives, require a minimum number
of receive buffers in order to receive any packets. If the first slave
read leaves less than this number available, a read from the next slave
may return 0 implying that the slave doesn't hav
On Thu, Sep 20, 2018 at 2:28 AM Matan Azrad wrote:
>
> Hi Chas
> Please see small comments.
>
> > From: Chas Williams
> > Some PMDs, especially ones with vector receives, require a minimum
> > number of receive buffers in order to receive any packets. If the first
From: Chas Williams
Some PMDs, especially ones with vector receives, require a minimum number
of receive buffers in order to receive any packets. If the first slave
read leaves less than this number available, a read from the next slave
may return 0 implying that the slave doesn't hav
On Wed, Sep 19, 2018 at 8:58 AM Luca Boccassi wrote:
>
> The vmxnet3 driver can't call back into dev_close(), and possibly
> dev_stop(), in dev_uninit(). When dev_uninit() is called, anything
> that those routines would want to clean up has already been released.
> Further, for complete cleanup,
On Thu, Sep 13, 2018 at 11:40 AM Matan Azrad wrote:
>
> Hi Chas
>
> From: Chas Williams
> > On Wed, Sep 12, 2018 at 1:56 AM Matan Azrad
> > wrote:
> > >
> > > Hi Chas
> > >
> > > From: Chas Williams
>
1 - 100 of 353 matches
Mail list logo
