This patch changes the behavior for KNI interface shutdown.
Previously we would receive a real response from the driver, now we
always receive success.
I think this should be reflected in the docs/release notes.
Igor
On Wed, Apr 21, 2021 at 2:07 AM Thomas Monjalon wrote:
> 12/04/2021 16:35, Ela
wrote:
> On 4/23/2021 9:41 AM, Igor Ryzhov wrote:
> > This patch changes the behavior for KNI interface shutdown.
> > Previously we would receive a real response from the driver, now we
> > always receive success.
> > I think this should be reflected in the docs/rel
t settings...
On Fri, Apr 23, 2021 at 3:43 PM Igor Ryzhov wrote:
> Hi Ferruh,
>
> Thanks. I think it would be great to make this configurable, and maybe even
> make shutdown synchronous by default to preserve the old behavior.
>
> I would be grateful if you could spend time o
Hi Ferruh, all,
Let's please discuss another approach to setting KNI link status before
making this API stable:
http://patches.dpdk.org/project/dpdk/patch/20190925093623.18419-1-iryz...@nfware.com/
I explained the problem with the current implementation there.
More than that, using ioctl approach
Thanks Ferruh,
I'll send an update later this week.
I also want to add a "Suggested-by: Dan Gora " as it was his
idea.
Dan, please let me know if you don't want this tag to be added.
Thanks,
Igor
On Mon, Jun 28, 2021 at 3:55 PM Ferruh Yigit wrote:
> On 10/27/2019 8:1
interface right after we enabled it and we don't have to use any
additional thread to track link status.
Propagating speed/duplex/autoneg to the kernel module also allows us to
implement ethtool_ops.get_link_ksettings callback.
Suggested-by: Dan Gora
Signed-off-by: Igor Ryzhov
---
app/
Signed-off-by: Igor Ryzhov
---
kernel/linux/kni/kni_net.c | 25 +
1 file changed, 25 insertions(+)
diff --git a/kernel/linux/kni/kni_net.c b/kernel/linux/kni/kni_net.c
index 99da8d37dd6b..84357bec1341 100644
--- a/kernel/linux/kni/kni_net.c
+++ b/kernel/linux/kni
Hi Ferruh,
Thanks for the review. My comments inline. I'll send a new version later
this week.
On Mon, Jul 5, 2021 at 2:58 PM Ferruh Yigit wrote:
> On 7/4/2021 6:06 PM, Igor Ryzhov wrote:
> > Current implementation doesn't allow us to update KNI carrier if the
> > i
Looks correct.
Could you, please, also change the order of `list_del` and `kni_dev_remove`
in `kni_release`? It suffers from the same problem.
Igor
On Fri, Jan 28, 2022 at 5:43 AM Min Hu (Connor) wrote:
> From: Huisong Li
>
> The "kni_dev" is the private data of the "net_device" in kni, and al
Acked-by: Igor Ryzhov
On Wed, Feb 9, 2022 at 10:36 AM Min Hu (Connor) wrote:
> Hi, Igor,
> fixed in v2, please check it, thanks.
>
> 在 2022/2/8 20:41, Igor Ryzhov 写道:
> > Looks correct.
> > Could you, please, also change the order of `list_del` and
> > `kni_de
Acked-by: Igor Ryzhov
On Sat, Oct 9, 2021 at 2:58 AM Ferruh Yigit wrote:
> To enable bifurcated device support, rtnl_lock is released before calling
> userspace callbacks and asynchronous requests are enabled.
>
> But these changes caused more issues, like bug #809, #816. To reduce
This looks more like a hack than an actual fix to me.
After this commit:
"ip link set up" is sent to the userspace with unlocked rtnl_lock
"ip link set down" is sent to the userspace with locked rtnl_lock
How is this really fixing anything? IMHO it only complicates the code.
If talking with users
; then you will always have to return success, even on failure, as Linux
> kernel does not have a mechanism to asynchronously report on failure
> for such system calls.
>
> IMHO - weighting the non-reporting of failure versus how the code
> looks (as it functions perfectly OK), I decide
t; Having KNI deadlock frequently for real life applications is far worst,
> IMHO.
>
> FYI
>
> Elad.
>
> On Wed, Feb 24, 2021 at 4:04 PM Igor Ryzhov wrote:
> >
> > Elad,
> >
> > I understand your point.
> > But the fact that this fix works for you doesn'
y to lock rtnl, resulting in a deadlock.
> The remedy is to release rtnl before locking the KNI sync lock.
> Since in between nothing is accessing Linux network-wise,
> no rtnl locking is needed.
> "
>
> FYI,
>
> Elad.
>
> On Wed, Feb 24, 2021 at 4:41 PM Igor Ryzhov
Elad.
>
> On Wed, Feb 24, 2021 at 5:18 PM Igor Ryzhov wrote:
> >
> > Stephen's idea was to fix the deadlock when working with the bifurcated
> driver.
> > Your rework breaks this because you still send link down requests under
> rtnl_lock.
> > Did you test your
Hi Elad,
Thanks for the patch, but this is still NACK from me.
The only real advantage of KNI over other exceptional-path techniques
like virtio-user is the ability to configure DPDK-managed interfaces
directly
from the kernel using well-known utils like iproute2. A very important part
of this is
which I think we all
> agree exist (we just do not agree on the proper solution or patch)...
>
> That's my two cents,
>
> Elad.
>
> On Fri, Feb 26, 2021 at 5:49 PM Stephen Hemminger
> wrote:
> >
> > On Fri, 26 Feb 2021 00:01:01 +0300
> > Igor Ryzhov
interface right after we enabled it and we don't have to use any
additional thread to track link status.
Propagating speed/duplex/autoneg to the kernel module also allows us to
implement ethtool_ops.get_link_ksettings callback.
Suggested-by: Dan Gora
Signed-off-by: Igor Ryzhov
---
app/
It allows inspecting link parameters using ethtool.
Signed-off-by: Igor Ryzhov
---
kernel/linux/kni/kni_net.c | 25 +
1 file changed, 25 insertions(+)
diff --git a/kernel/linux/kni/kni_net.c b/kernel/linux/kni/kni_net.c
index 611719b5ee27..931b9daf104f 100644
--- a
If the child process returns -1, WEXITSTATUS converts to 255 because it
takes only 8 least significant bits. As a result, the test is not
considered failed when it should.
Signed-off-by: Igor Ryzhov
---
app/test/test_kni.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/app
Could you please clarify where exactly do I need to use rtnl lock?
>From what I understand, netif_carrier_on/off can be called without the lock.
On Thu, Aug 26, 2021 at 8:15 PM Stephen Hemminger <
step...@networkplumber.org> wrote:
> On Thu, 26 Aug 2021 18:19:09 +0300
> Igo
interface right after we enabled it and we don't have to use any
additional thread to track link status.
Propagating speed/duplex/autoneg to the kernel module also allows us to
implement ethtool_ops.get_link_ksettings callback.
Suggested-by: Dan Gora
Signed-off-by: Igor Ryzhov
---
app/
It allows inspecting link parameters using ethtool.
Signed-off-by: Igor Ryzhov
---
kernel/linux/kni/kni_net.c | 25 +
1 file changed, 25 insertions(+)
diff --git a/kernel/linux/kni/kni_net.c b/kernel/linux/kni/kni_net.c
index 611719b5ee27..931b9daf104f 100644
--- a
If the child process returns -1, WEXITSTATUS converts to 255 because it
takes only 8 least significant bits. As a result, the test is not
considered failed when it should.
Signed-off-by: Igor Ryzhov
---
app/test/test_kni.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/app
On Thu, Aug 26, 2021 at 9:06 PM Stephen Hemminger <
step...@networkplumber.org> wrote:
> On Thu, 26 Aug 2021 20:46:47 +0300
> Igor Ryzhov wrote:
>
> > Could you please clarify where exactly do I need to use rtnl lock?
> > From what I understand, netif_carrier_on/of
in use?
Best regards,
Igor Ryzhov
eady reserved memory
back to "free_memseg", and impossibility of unreserving memory is just
because there is no function for that, not because it is impossible in
principle.
Am I right? Or there are any restrictions?
Best regards,
Igor Ryzhov
06.05.2014 13:05, Igor Ryzhov ?:
> Hel
, not size.
Best regards,
Igor Ryzhov
07.05.2014 15:39, Olivier MATZ ?:
> Hi Igor,
>
> On 05/07/2014 09:54 AM, Igor Ryzhov wrote:
>> I noticed that in Memzone realization there is a special global variable
>> "free_memseg" containing pointers on free memory
Guide says that a configure parameter to choose between vhost cuse and vhost
user will be introduced in the future, but it?s already added by commit
28a1ccca41bf.
Signed-off-by: Igor Ryzhov
---
doc/guides/sample_app_ug/vhost.rst | 7 +++
1 file changed, 3 insertions(+), 4 deletions
Sorry, I used wrong email address to reply from. This one is correct.
On Mon, Apr 13, 2015 at 10:11 AM, Igor Ryzhov wrote:
> Hello, Changchun.
>
> Previous paragraph says ?To enable vhost, turn on vhost library in the
> configure file config/common_linuxapp?, but string in a c
already changed string. So I
used the same style.
Regards,
Igor
> 13 ???. 2015 ?., ? 7:52, Ouyang, Changchun
> ???(?):
>
> Hi Igor,
>
> Good catch, comments as below.
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Igo
Hello Steve,
Thank you for the patch.
I think a couple of improvements can be done:
1. Function existence check – if (*dev->dev_ops->mac_addr_add) – can be
taken out of the loop. We don't need to check it on each iteration.
2. I'm not completely sure, but I think loop can be started from 1, not
f
)(dev, addr);
>
>
>
> + /* Update pool bitmap in NIC data structure */
>
> + dev->data->mac_pool_sel[0] = 1;
>
> +
>
> return 0;
>
>
>
> Any thoughts would be appreciated.
>
>
>
> Regards,
>
> Steve
>
>
>
&
Hello Steve,
Thank you for all the fixes.
But I think I noticed another one issue in MAC replay process.
Pool number is extracted only once:
if (RTE_ETH_DEV_SRIOV(dev).active)
pool = RTE_ETH_DEV_SRIOV(dev).def_vmdq_idx;
But when MAC address is added using rte_eth_dev_mac_addr_add several
differ
Thank you Steve.
I never did it before and I don't know if I have rights for that, but:
Acked-by: Igor Ryzhov
On Tue, Jan 24, 2017 at 5:21 AM, Steve Shin wrote:
> This patch fixes a bug in replaying MAC address to the hardware
> in rte_eth_dev_config_restore() routine. Added
t can
be good to explain it in DPDK development guidelines.
Best regards,
Igor
On Tue, Jan 24, 2017 at 4:21 PM, Ferruh Yigit
wrote:
> On 1/24/2017 10:09 AM, Igor Ryzhov wrote:
> > Thank you Steve.
> >
>
> > I never did it before and I don't know if I have righ
Hello Keith,
comments inline below.
On Thu, Feb 9, 2017 at 9:14 PM, Keith Wiles wrote:
> Signed-off-by: Keith Wiles
> ---
> doc/guides/howto/flow_bifurcation.rst | 2 +-
> doc/guides/howto/lm_bond_virtio_sriov.rst | 6 +++---
> doc/guides/howto/lm_virtio_vhost_user.rst | 6 +++---
> 3 file
Hello everyone,
I think there is a bug in rte_eth_dev_config_restore function.
During restoration of MAC address configuration, all MAC addresses are
restored with mac_addr_add function, but as I think MAC address with index
0 shouldn't be restored in such way, because it is a default MAC address.
Thank you, Ferruh.
As we are staying on the existing implementation, I think we can do some
improvements:
1. Implement more commands for net_device_ops.
2. Implement ethtool support the same way as net_device_ops are implemented
? send commands to application.
3. Add ability to set default MAC add
On Fri, Oct 28, 2016 at 7:13 PM, Thomas Monjalon
wrote:
> 2016-10-28 15:51, Richardson, Bruce:
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
> > > 2016-10-28 15:31, Ferruh Yigit:
> > > > * virtio-user + vhost-net
> > > > This can be valid alternative, removes the out
On Fri, Oct 28, 2016 at 9:40 PM, Thomas Monjalon
wrote:
> 2016-10-28 20:29, Igor Ryzhov:
> > On Fri, Oct 28, 2016 Thomas Monjalon wrote:
> > > 2016-10-28 15:51, Richardson, Bruce:
> > > > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
>
Hello Wenzhuo,
How about also delete meaningless "ixgbe_remove_rar(dev, 0);"?
Best regards,
Igor
On Tue, May 2, 2017 at 11:34 AM, Wenzhuo Lu wrote:
> Pool 0 is not PF, it's VF 0. So the MAC is set for VF 0
> but not PF.
> The code introduced a weird issue. In the scenario PF + VF,
> when only
0e) we still use
kni_ethtool_ops with implemented .get_link function.
Signed-off-by: Igor Ryzhov
---
kernel/linux/kni/Makefile | 2 +-
kernel/linux/kni/kni_ethtool.c | 210 -
kernel/linux/kni/kni_misc.c| 6 +-
3 files changed, 6 insertions(+), 212 del
0e) we still use
kni_ethtool_ops with implemented .get_link function.
Signed-off-by: Igor Ryzhov
---
kernel/linux/kni/Makefile | 2 +-
kernel/linux/kni/kni_ethtool.c | 210 -
kernel/linux/kni/kni_misc.c| 9 +-
3 files changed, 7 insertions(+), 214 del
ckets to Linux.
Can virtio user be used the same way, as a mirror of physical port?
Best regards,
Igor
On Sat, Dec 1, 2018 at 2:38 AM Stephen Hemminger
wrote:
> On Fri, 30 Nov 2018 22:47:50 +0300
> Igor Ryzhov wrote:
>
> > Current implementation of kni_ethtool_ops jus
tif_running(dev) && dev->ethtool_ops->get_link(dev);
if (copy_to_user(useraddr, &edata, sizeof(edata)))
return -EFAULT;
return 0;
}
On Sat, Dec 1, 2018 at 8:31 PM Stephen Hemminger
wrote:
> On Sat, 1 Dec 2018 14:12:54 +0300
> Igor Ryzhov wrote:
>
> > Hi Stephen
other ethtool operations may be implemented in a driver-independent
way using
the same concept as for netdev_ops.
On Mon, Dec 3, 2018 at 4:09 PM Ferruh Yigit wrote:
> On 11/30/2018 11:38 PM, Stephen Hemminger wrote:
> > On Fri, 30 Nov 2018 22:47:50 +0300
> > Igor Ryzhov wro
Hi Ferruh,
Please, look at my patch http://patches.dpdk.org/patch/48454/ and consider
rebasing your patch over mine.
As we discussed with Stephen, KNI needs to supply ethtool_ops with
.get_link function, to properly support link status.
So we should save ethtool_ops and implement .get_link using
tu or
.ndo_set_mac_address.
I want to add more types of requests to rte_kni_req_id enum and implement
them.
It's not an urgent task, so I don't know yet when I'll find time to do it.
Best regards,
Igor
On Tue, Dec 18, 2018 at 9:13 PM Ferruh Yigit wrote:
> On 12/3/2018 2:0
Hi Ferruh,
I answered in another thread.
Regarding this patch – I have no objections now.
Best regards,
Igor
On Tue, Dec 18, 2018 at 9:17 PM Ferruh Yigit wrote:
> On 12/18/2018 9:20 AM, Ferruh Yigit wrote:
> > On 12/18/2018 8:20 AM, Igor Ryzhov wrote:
> >> Hi Ferruh,
>
It's just exact copy of eth_header_parse function from Linux kernel.
No problem, can do that with ether_addr_copy.
On Sat, Sep 29, 2018 at 10:22 AM Stephen Hemminger <
step...@networkplumber.org> wrote:
> On Thu, 27 Sep 2018 03:02:24 +0300
> Igor Ryzhov wrote:
>
>
at 10:19 PM Igor Ryzhov wrote:
> It's just exact copy of eth_header_parse function from Linux kernel.
>
> No problem, can do that with ether_addr_copy.
>
> On Sat, Sep 29, 2018 at 10:22 AM Stephen Hemminger <
> step...@networkplumber.org> wrote:
>
>> On T
ll be better to use default eth_header_ops.
Best regards,
Igor
On Tue, Oct 2, 2018 at 7:58 PM Ferruh Yigit wrote:
> On 9/27/2018 1:02 AM, Igor Ryzhov wrote:
> > Signed-off-by: Igor Ryzhov
>
> Hi Igor,
>
> What is the motivation to add this support? What is enabled by this?
&
-off-by: Igor Ryzhov
---
lib/librte_kni/rte_kni.c | 392 ---
lib/librte_kni/rte_kni.h | 6 +-
test/test/test_kni.c | 6 -
3 files changed, 128 insertions(+), 276 deletions(-)
diff --git a/lib/librte_kni/rte_kni.c b/lib/librte_kni/rte_kni.c
index
Hello Ferruh,
Thanks for the review, comments inline.
On Mon, Aug 27, 2018 at 8:06 PM, Ferruh Yigit
wrote:
> On 8/2/2018 3:25 PM, Igor Ryzhov wrote:
> > Long time ago preallocation of memory for KNI was introduced in commit
> > 0c6bc8e. It was done because of lack of ability to
Hi Dan,
We use KNI device exactly the same way you described – with IP addresses,
routing, etc.
And we also faced the same problem of having the actual link status in
Linux kernel.
There is a special callback for link state management in net_device_ops for
soft-devices like KNI called ndo_change_
_open.
Best regards,
Igor
On Thu, Aug 30, 2018 at 12:49 PM, Igor Ryzhov wrote:
> Hi Dan,
>
> We use KNI device exactly the same way you described – with IP addresses,
> routing, etc.
> And we also faced the same problem of having the actual link status in
> Linux kernel.
>
> Th
-off-by: Igor Ryzhov
---
v2:
* allocate KNI using rte_zmalloc
* swap reserve/release functions
* use "kni" as a variable name
* use macros for memzone names
lib/librte_kni/rte_kni.c | 502 +--
lib/librte_kni/rte_kni.h | 6 +-
test/test/
Hi Ferruh,
Will fix it today.
Igor
On Wed, Sep 26, 2018 at 1:41 PM Ferruh Yigit wrote:
> On 9/23/2018 8:12 PM, Igor Ryzhov wrote:
> > Long time ago preallocation of memory for KNI was introduced in commit
> > 0c6bc8e. It was done because of lack of ability to free previousl
-off-by: Igor Ryzhov
---
v3:
* style fixes
v2:
* allocate KNI using rte_zmalloc
* swap reserve/release functions
* use "kni" as a variable name
* use macros for memzone names
lib/librte_kni/rte_kni.c | 496 +--
lib/librte_kni/rte_kni.h | 6 +-
Signed-off-by: Igor Ryzhov
---
kernel/linux/kni/kni_net.c | 14 ++
1 file changed, 14 insertions(+)
diff --git a/kernel/linux/kni/kni_net.c b/kernel/linux/kni/kni_net.c
index 7fcfa106c..128a5477c 100644
--- a/kernel/linux/kni/kni_net.c
+++ b/kernel/linux/kni/kni_net.c
@@ -678,6
Hi Dan, Ferruh,
Why do we need "struct rte_eth_link" as a parameter at all?
Only link status is used in the function – let's use it only:
rte_kni_update_link(struct rte_kni *kni, int link_status) /* 0 – down, 1 –
up */
It will also solve your differences as we won't have any "redundant"
informat
Hi Ferruh,
What about this patch?
Can you merge it as-is, or should I change it to use relevant
eth_header_ops functions? Or maybe completely use eth_header_ops?
Best regards,
Igor
On Fri, Nov 30, 2018 at 10:07 PM Igor Ryzhov wrote:
> Hi Ferruh,
>
> header_ops.parse method is us
ld be investigated separately, and possibly
resolved by a separate patch.
Best regards,
Igor
On Thu, Jan 24, 2019 at 5:10 PM Ferruh Yigit wrote:
> On 1/24/2019 9:18 AM, Igor Ryzhov wrote:
> > Hi Ferruh,
> >
> > What about this patch?
> > Can you merge it as-is, or sh
header_ops usage should be investigated separately, and possibly
> resolved by
> > a separate patch.
>
> Agreed, eth_header_ops usage should be investigated separately.
>
> >
> > Best regards,
> > Igor
> >
> > On Thu, Jan 24, 2019 at 5:10 PM Ferruh Yigit > &
After read, file offset must be set to 0 before write.
Otherwise, the third byte will be overwritten instead of the first.
Fixes: c6fd54f28c24 ("kni: add function to set link state on kernel interface")
Cc: sta...@dpdk.org
Signed-off-by: Igor Ryzhov
---
lib/librte_kni/rte_kni.c | 6
Hi Ferruh,
Can you, please, take a look at this patch?
The current implementation is broken, I think the patch should be merged
into 19.02 and 18.11.1.
Best regards,
Igor
On Thu, Jan 24, 2019 at 11:47 PM Igor Ryzhov wrote:
> After read, file offset must be set to 0 before write.
> Oth
Hi again,
Sorry for bothering, I should have done more testing. It works as it is now.
So, self NACK on the patch.
Best regards,
Igor
On Mon, Jan 28, 2019 at 2:45 PM Igor Ryzhov wrote:
> Hi Ferruh,
>
> Can you, please, take a look at this patch?
> The current implementation is bro
Hello everyone,
We are currently testing i40e support for RSS calculation.
RSS is configured with all supported flags:
#define I40E_RSS_OFFLOAD_ALL ( \
ETH_RSS_FRAG_IPV4 | \
ETH_RSS_NONFRAG_IPV4_TCP | \
ETH_RSS_NONFRAG_IPV4_UDP | \
ETH_RSS_NONFRAG_IPV4_SCTP | \
ETH_RSS_NONFRAG_IPV4_OTHER | \
ETH
Hi Ferruh,
What's the plan with this patch?
Best regards,
Igor
On Sat, Jan 5, 2019 at 7:55 PM Igor Ryzhov wrote:
> Hi Ferruh,
>
> I answered in another thread.
>
> Regarding this patch – I have no objections now.
>
> Best regards,
> Igor
>
> On Tue, D
Hi Ferruh,
Thanks. Should I be a maintainer to ack the patch?
Best regards,
Igor
On Mon, Feb 18, 2019 at 3:33 PM Ferruh Yigit wrote:
> On 2/6/2019 1:12 PM, Igor Ryzhov wrote:
> > Hi Ferruh,
> >
> > What's the plan with this patch?
>
> Hi Igor,
>
> I j
Acked-by: Igor Ryzhov
On Mon, Feb 18, 2019 at 3:30 PM Ferruh Yigit wrote:
> Remove KNI ethtool support.
>
> Signed-off-by: Ferruh Yigit
> ---
> RFC Patch: https://patches.dpdk.org/patch/49025/
> ---
> doc/guides/rel_notes/deprecation.rst | 8
> 1 fi
Use the same value as in Linux driver.
Signed-off-by: Igor Ryzhov
---
drivers/net/i40e/i40e_ethdev.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 285d92b..93b9dd0 100644
--- a/drivers/net/i40e
Hello everyone.
It's sad that my comments were unanswered.
I'm ok with the first two – they were mostly style-related.
But I made an investigation on the third one and it is a bug.
This is a description (from X710 datasheet) of flags sent to
mac_address_write command:
By bits:
0-7 – Reserved
8 –
Use the same value as in Linux driver.
Fixes: e18e01e92c29 ("i40e: support default MAC address setting")
Cc: sta...@dpdk.org
Signed-off-by: Igor Ryzhov
---
v2:
* fix checkpatch warning (long line)
* fix commit subject
* add Fixes line
* CC to stable
---
drivers/net/i40e/i40e_et
omments (in another mail
> thread).
>
>
>
> Best Regards,
>
> Beilei
>
>
>
> *From:* Igor Ryzhov [mailto:iryz...@nfware.com]
> *Sent:* Thursday, January 11, 2018 6:47 AM
> *To:* Zhang, Helin
> *Cc:* Xing, Beilei ; Olivier Matz <
> olivier.m...@6
Current flag is in wrong byte order.
Use defined macro as it is done in Linux driver.
Fixes: e18e01e92c29 ("i40e: support default MAC address setting")
Cc: sta...@dpdk.org
Signed-off-by: Igor Ryzhov
---
v3:
* fix commit message
v2:
* fix checkpatch warning (long line)
* fix commit sub
Sure.
Sent v3.
On Fri, Jan 12, 2018 at 5:22 AM, Xing, Beilei wrote:
>
>
> > -Original Message-
> > From: Igor Ryzhov [mailto:iryz...@nfware.com]
> > Sent: Thursday, January 11, 2018 6:07 PM
> > To: dev@dpdk.org
> > Cc: Xing, Beilei ; sta...@dpdk.
Signed-off-by: Igor Ryzhov
---
lib/librte_lpm/rte_lpm.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/lib/librte_lpm/rte_lpm.c b/lib/librte_lpm/rte_lpm.c
index 64c074e..140dbb2 100644
--- a/lib/librte_lpm/rte_lpm.c
+++ b/lib/librte_lpm/rte_lpm.c
@@ -218,6 +218,7 @@ rte_lpm_create_v20
Signed-off-by: Igor Ryzhov
---
lib/librte_lpm/rte_lpm6.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/lib/librte_lpm/rte_lpm6.c b/lib/librte_lpm/rte_lpm6.c
index b4a7df3..5ca322c 100644
--- a/lib/librte_lpm/rte_lpm6.c
+++ b/lib/librte_lpm/rte_lpm6.c
@@ -191,6 +191,7 @@ rte_lpm6_create
Hello Andrew,
Don't you think that it's not correct that net/sfc works that way?
If we go further, dev->dev_ops->mac_addr_set not only should be called
before ether_addr_copy.
It should return status code, and in case of error ether_addr_copy
shouldn't be called at all.
Am I wrong?
Best regards,
Thank you for the patch! Comments inline.
On Wed, Jan 3, 2018 at 5:29 PM, Olivier Matz wrote:
>
> drivers/net/i40e/i40e_ethdev.c | 29 +
> 1 file changed, 29 insertions(+)
>
> diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_
> ethdev.c
> index 811c
Hello.
I found some problem with virtio xstats counters.
Example:
rx_good_packets: 3
rx_good_bytes: 180
rx_errors: 0
rx_q0_good_packets: 3
rx_q0_good_bytes: 180
rx_q0_errors: 0
rx_q0_multicast_packets: 3
rx_q0_broadcast_packets: 1
rx_q0_undersize_packets: 3
It means that undersize packets are
->broadcast++;
} else {
vq->multicast++;
}
}
Best regards,
Igor
> 26 . 2016 ?., ? 17:29, Van Haaren, Harry
> ???(?):
>
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Igor Ryzhov
>> I found so
Signed-off-by: Igor Ryzhov
---
drivers/net/virtio/virtio_rxtx.c | 9 +++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/drivers/net/virtio/virtio_rxtx.c b/drivers/net/virtio/virtio_rxtx.c
index 41a1366..fe18e1d 100644
--- a/drivers/net/virtio/virtio_rxtx.c
+++ b/drivers/net
Sent a patch: http://dpdk.org/dev/patchwork/patch/10887/
<http://dpdk.org/dev/patchwork/patch/10887/>.
> 26 . 2016 ?., ? 17:35, Igor Ryzhov ???(?):
>
> Hello, Harry.
>
> Understood about size of packets. It's a bit confusing, because in all other
> dr
Hello, everyone.
How about exposing stats according to IF-MIB?
Statistics to be exposed are - octets, unicast packets, multicast packets,
broadcast packets, errors and discards for both TX and RX.
These counters are basic and implemented by most of drivers.
All other driver-specific counters ca
LPM table overflow may occur if table is full and added rule has the biggest
depth that already have some rules.
Signed-off-by: Igor Ryzhov
---
lib/librte_lpm/rte_lpm.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/lib/librte_lpm/rte_lpm.c b/lib/librte_lpm/rte_lpm.c
index 983e04b
Hello again. Will anybody review this patch?
This is really critical issue, because it can lead to memory corruption and
break any program using LPM.
CCing this to Bruce Richardson, because he is maintainer of LPM.
Regards,
Igor Ryzhov
On Fri, Feb 20, 2015 at 4:16 PM, Igor Ryzhov wrote:
>
y I missed this Friday. I'll look at it shortly.
>
>
>
> On 21 Feb 2015, at 22:56, Igor Ryzhov
> <mailto:iryzhov at nfware.com >> wrote:
>
> Hello again. Will anybody review this patch?
> This is really critical issue, because it can lead to memory corruption
>
cards with other
firmware version, but the problem is that we can't find it. Can anybody
help with getting firmware version 4.2.4? Thank you.
Regards,
Igor Ryzhov
This is really useful, thank you!
Best regards,
Igor Ryzhov
> 12 . 2014 ?., ? 6:22, Chi, Xiaobo (NSN - CN/Hangzhou) nsn.com> ???(?):
>
> Hi,
> Background:
> What we are doing now is port make telecom network element to be cloud based.
> For one of our product,
Hello.
Are there any docs with detailed description of cmdline library?
I found only some information in ?DPDK Sample Apps? document, but it describes
only a couple of features.
Best regards,
Igor Ryzhov
gt;>>
>>> Hi Igor,
>>>
>>> On 11/14/2014 09:52 AM, Igor Ryzhov wrote:
>>>> Are there any docs with detailed description of cmdline library?
>>>> I found only some information in ?DPDK Sample Apps? document, but it
>>>> descri
Length Error)? Or is it a bug?
Another one question is about incompleteness of rte_eth_stats structure. IXGBE
and other drivers have a lot of counters but only a part of them is represented
in rte_eth_stats. Is there any valuable reasons for that or it's just not
implemented?
Best regards,
n, Harry
> ???(?):
>
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Igor Ryzhov
>> Hello everyone.
>
> Hi Igor,
>
>> Investigating IXGBE driver I found an mspdc counter (MAC Short Packet
>> Discard). And I am wondering why this counter is not used in the calc
Hello.
How about deleting rx_nombuf from rte_eth_stats?
Do you think this counter is necessary? It just shows enormous numbers in case
of a lack of processing speed.
But we already have imissed counter which shows real number of packets, dropped
for the same reason.
> 14 2016 ?., ? 16:37,
Hello.
I think good alternative is rte_ethtool library from ethtool sample
application.
But I am wondering why this code is only in app, not in lib.
Best regards,
Igor
On Thu, Jul 21, 2016 at 4:33 PM, Ferruh Yigit
wrote:
> On 7/20/2016 5:07 PM, Thomas Monjalon wrote:
> > The out-of-tree kernel
Hello, Marc!
You swapped values for X710 and XL710 - you use 1G and 10G for XL710, 10G and
40G for X710.
Best regards,
Igor
> 26 ??? 2015 ?., ? 22:50, Marc Sune ???(?):
>
> Added speed capabilities to all pmds supporting physical NICs:
>
> * e1000
> * ixgbe
> * i40
> * mlx4
> * fm10k
>
1 - 100 of 155 matches
Mail list logo
