> s/unrequired/"not required"/
> s/consme/consume/ .two different places
> s/accros/across/
>
> Signed-off-by: Bhaskar Chowdhury
Acked-by: Igor Russkikh
Thanks
Igor
>> v3: change us to ns in docs
>> v2: Review comments from Jesper
>>
>> CC: Jesper Dangaard Brouer
>
> Did a quick review and everything looks okay.
> The patches are already applied, but you will still get my ACK,
> even-though it will not make it to the commit log.
>
> Acked-by: Jesper Danga
or that are:
- using multiple devices. We have to generate full linerate on
all physical functions (ports) of our multiport device.
- pushing multiple flows (with different packet options)
Signed-off-by: Igor Russkikh
---
samples/pktgen/README.rst | 17 +++
samples
Hello netdev community,
Please consider small improvements to pktgen scripts we use in our environment.
Adding delay parameter through command line,
Adding new -a (append) parameter to make flex runs
v3: change us to ns in docs
v2: Review comments from Jesper
CC: Jesper Dangaard Brouer
Igor
DELAY may now be explicitly specified via common parameter -w
Signed-off-by: Igor Russkikh
---
samples/pktgen/README.rst | 1 +
samples/pktgen/parameters.sh | 10 +-
samples/pktgen/pktgen_bench_xmit_mode_netif_receive.sh | 3
>> +echo " -w : (\$DELAY) Tx Delay value (us)"
>This is not in "us" it is in "ns" (nanosec). (Like I pointed out last time...)
Ah, sorry lost that. Will fix.
One extra thing I wanted to raise is "set -o errexit" in functions.sh.
It basically contradicts with the usecase I'm using (doing
DELAY may now be explicitly specified via common parameter -w
Signed-off-by: Igor Russkikh
---
samples/pktgen/parameters.sh | 10 +-
samples/pktgen/pktgen_bench_xmit_mode_netif_receive.sh | 3 ---
samples/pktgen/pktgen_bench_xmit_mode_queue_xmit.sh| 3
or that are:
- using multiple devices. We have to generate full linerate on
all physical functions (ports) of our multiport device.
- pushing multiple flows (with different packet options)
Signed-off-by: Igor Russkikh
---
samples/pktgen/README.rst | 18 +++
samples
Hello netdev community,
Please consider small improvements to pktgen scripts we use in our environment.
Adding delay parameter through command line,
Adding new -a (append) parameter to make flex runs
v2: Review comments from Jesper
CC: Jesper Dangaard Brouer
Igor Russkikh (2):
samples
> In order to correctly clean up the failed attempt to start a device call
> aq_nic_stop.
>
> Signed-off-by: Nathan Rossi
Looks good to me, thanks!
Reviewed-by: Igor Russkikh
Igor
Hi Jesper,
Thanks for reviewing this.
>> DELAY may now be explicitly specified via common parameter -w
>
> What are you actually using this for?
Basically, for the second patch.
When running multidev pktgen (using that -a option) with large amount of clones
and
bursts (-c and -b) I saw that s
Hello netdev community,
Please consider small improvements to pktgen scripts we use in our environment.
Adding delay parameter through command line,
Adding new -a (append) parameter to make flex runs
Igor Russkikh (2):
samples: pktgen: allow to specify delay parameter via new opt
samples
or that are:
- using multiple devices. We have to generate full linerate on
all physical functions (ports) of our multiport device.
- pushing multiple flows (with different packet options)
Signed-off-by: Igor Russkikh
---
samples/pktgen/README.rst | 18 ++
samples
DELAY may now be explicitly specified via common parameter -w
Signed-off-by: Igor Russkikh
---
samples/pktgen/parameters.sh | 10 +-
samples/pktgen/pktgen_bench_xmit_mode_netif_receive.sh | 3 ---
samples/pktgen/pktgen_bench_xmit_mode_queue_xmit.sh| 3
>> Here we do the initialization of coalescing values on load.
>> Although the default device values are the same - explicit
>> config is better visible.
>
> Can you also make the driver store the settings across ifdown / ifup
> and allow the settings to be checked while the interface is down,
> On Fri, 15 Jan 2021 01:06:08 -0800 Bhaskar Upadhaya wrote:
>> Add net poll controller support to transmit kernel printks
>> over UDP
>
> Why do you need this patch? Couple years back netpoll was taught
> how to pull NAPIs by itself, and all you do is schedule NAPIs.
>
> All the driver should
>> I've been trying out the latest CentOS 8 Stream kernel and found that I
>> get kernel panic
> (https://urldefense.proofpoint.com/v2/url?u=https-3A__bugzilla.redhat.com_
> show-5Fbug.cgi-3Fid-3D1913481&d=DwICAg&c=nKjWec2b6R0mOyPaz7xtfQ&r=3kUjVPjr
> PMvlbd3rzgP63W0eewvCq4D-kzQRqaXHOqU&m=5qgG2X2
>>
>> Since normal layout is 1400 packets - we do use 2K (half page) for each
> packet.
>
> What is 'normal layout is 1400 packets' ? Didn't you mean the 1500 byte
> standard MTU? So this is what you've been trying to tell me - that for
> 1500 byte mtu and 1k HW granularity you need to provide t
On 23/11/2020 7:20 am, Ramsay, Lincoln wrote:
>> Yep, that could be the only way to fix this for now.
>> Have you tried to estimate any performance drops from this?
>
> Unfortunately, I am not in a very good position to do this. The 10G
> interfaces on our device don't actually have enough raw
GvqcEGj8wEOVoN1BySZhGUVECVTBJCmNiRsHUw&s=J1SWrfEL
> erJOzUlJdD_S5afGaZosmVP8lyKsu9DTULw&e=
> Signed-off-by: Gustavo A. R. Silva
Reviewed-by: Igor Russkikh
Thanks,
Igor
On 20/11/2020 1:49 am, Maciej Fijalkowski wrote:
> External Email
>
> --
> On Thu, Nov 19, 2020 at 10:34:48PM +, Ramsay, Lincoln wrote:
>> When performing IPv6 forwarding, there is an expectation that SKBs
>> will have some
> Just always use the slower codepath that copies memory into an
> allocated SKB.
>
> Signed-off-by: Lincoln Ramsay
Acked-by: Igor Russkikh
Yep, that could be the only way to fix this for now.
Have you tried to estimate any performance drops from this?
The most harm may be here on
Hi Ramsay,
> When performing IPv6 forwarding, there is an expectation that SKBs
> will have some headroom. When forwarding a packet from the aquantia
> driver, this does not always happen, triggering a kernel warning.
>
> It was observed that napi_alloc_skb and other ethernet drivers
> reserve
Hi Anant,
>>
>> +static int aqc111_set_mac_addr(struct net_device *net, void *p)
>> +{
>> +struct usbnet *dev = netdev_priv(net);
>> +int ret = 0;
>> +
>> +ret = eth_mac_addr(net, p);
>> +if (ret < 0)
>> +return ret;
>> +
>
> When eth_mac_addr() fails, from what I can
feature may increase linkup time.
Signed-off-by: Igor Russkikh
---
.../ethernet/aquantia/atlantic/aq_ethtool.c | 12 +
.../net/ethernet/aquantia/atlantic/aq_hw.h| 4 +++
.../net/ethernet/aquantia/atlantic/aq_nic.c | 27 +++
.../net/ethernet/aquantia/atlantic
Define get/set phy tunable callbacks in ethtool ops.
This will allow MAC drivers with integrated PHY still to implement
these tunables.
Reviewed-by: Andrew Lunn
Signed-off-by: Igor Russkikh
---
include/linux/ethtool.h | 4
net/ethtool/ioctl.c | 37
EDPC tunables.
v3: fixed flaw in EDPD logic, from Andrew
v2: comments from Andrew
Igor Russkikh (3):
ethtool: allow netdev driver to define phy tunables
net: atlantic: implement phy downshift feature
net: atlantic: implement media detect feature via phy tunables
.../ethernet/aquantia
PHY downshift allows phy to try renegotiate if link is unstable
and can carry higher speed.
AQC devices has integrated PHY which is controlled by MAC firmware.
Thus, driver defines new ethtool callbacks to implement phy tunables
via netdev.
Reviewed-by: Andrew Lunn
Signed-off-by: Igor Russkikh
>> +if (val > 0 && val != AQ_HW_MEDIA_DETECT_CNT) {
>> +netdev_err(self->ndev, "EDPD on this device could have
> only fixed value of %d\n",
>> + AQ_HW_MEDIA_DETECT_CNT);
>> +return -EINVAL;
>> +}
>> +
>> +/* msecs plays no role - configur
EDPC tunables.
v2: comments from Andrew
Igor Russkikh (3):
ethtool: allow netdev driver to define phy tunables
net: atlantic: implement phy downshift feature
net: atlantic: implement media detect feature via phy tunables
.../ethernet/aquantia/atlantic/aq_ethtool.c | 53
PHY downshift allows phy to try renegotiate if link is unstable
and can carry higher speed.
AQC devices has integrated PHY which is controlled by MAC firmware.
Thus, driver defines new ethtool callbacks to implement phy tunables
via netdev.
Signed-off-by: Igor Russkikh
---
.../ethernet
feature may increase linkup time.
Signed-off-by: Igor Russkikh
---
.../ethernet/aquantia/atlantic/aq_ethtool.c | 12 +
.../net/ethernet/aquantia/atlantic/aq_hw.h| 4 +++
.../net/ethernet/aquantia/atlantic/aq_nic.c | 26 +++
.../net/ethernet/aquantia/atlantic
Define get/set phy tunable callbacks in ethtool ops.
This will allow MAC drivers with integrated PHY still to implement
these tunables.
Signed-off-by: Igor Russkikh
---
include/linux/ethtool.h | 4
net/ethtool/ioctl.c | 37 -
2 files changed, 28
Hi Andrew,
> Since this is your own PHY, not some magical black box, i assume you
> actually know what value it is using? It probably even lists it in the
> data sheet.
>
> So just hard code that value in the driver. That has got to be better
> than saying the incorrect value of 1ms.
You mean al
>> switch (tuna->id) {
>> +case ETHTOOL_PHY_EDPD: {
>> +const u16 *val = data;
>> +
>> +/* msecs plays no role - configuration is always fixed in
> PHY */
>> +aq_nic->aq_nic_cfg.is_media_detect = *val ? 1 : 0;
>
> This is the wrong usage of the API:
>
>
> Hi Igor
>
> I think all other implementations return -EINVAL or -E2BIG or similar
> when the value is not supported.
>
> Also, given that a u8 is being passed, is cfg->downshift_counter > 255
> possible? I'm not even sure 255 makes any sense. Autoneg takes around
> 1.5s, maybe longer. Do
Define get/set phy tunable callbacks in ethtool ops.
This will allow MAC drivers with integrated PHY still to implement
these tunables.
Signed-off-by: Igor Russkikh
---
include/linux/ethtool.h | 4
net/ethtool/ioctl.c | 37 -
2 files changed, 28
PHY downshift allows phy to try renegotiate if link is unstable
and can carry higher speed.
AQC devices has integrated PHY which is controlled by MAC firmware.
Thus, driver defines new ethtool callbacks to implement phy tunables
via netdev.
Signed-off-by: Igor Russkikh
---
.../ethernet
EDPC tunables.
Igor Russkikh (3):
ethtool: allow netdev driver to define phy tunables
net: atlantic: implement phy downshift feature
net: atlantic: implement media detect feature via phy tunables
.../ethernet/aquantia/atlantic/aq_ethtool.c | 56 +++
.../net/ethernet
: Igor Russkikh
---
.../net/ethernet/aquantia/atlantic/aq_ethtool.c | 14 ++
drivers/net/ethernet/aquantia/atlantic/aq_hw.h | 2 ++
drivers/net/ethernet/aquantia/atlantic/aq_nic.c | 16
drivers/net/ethernet/aquantia/atlantic/aq_nic.h | 2 ++
.../aquantia/atlantic
Driver subfolder files refer parent folder includes in an
absolute manner.
Makefile contains a -I for this, but apparently that does not
work if object tree is separated.
Adding srctree to fix that.
Signed-off-by: Igor Russkikh
---
drivers/net/ethernet/aquantia/atlantic/Makefile | 4 ++--
1
From: Dmitry Bogdanov
Fix the assert during VF driver installation when the personality is iWARP
Fixes: 1fe614d10f45 ("qed: Relax VF firmware requirements")
Signed-off-by: Igor Russkikh
Signed-off-by: Michal Kalderon
Signed-off-by: Dmitry Bogdanov
---
drivers/net/ethernet/
From: Dmitry Bogdanov
In CMT and NPAR the PF is unknown when the GFS block processes the
packet. Therefore cannot use searcher as it has a per PF database,
and thus ARFS must be disabled.
Fixes: d51e4af5c209 ("qed: aRFS infrastructure support")
Signed-off-by: Manish Chopra
Signed-of
From: Dmitry Bogdanov
In some configurations ARFS cannot be used, so disable it if device
is not capable.
Fixes: e4917d46a653 ("qede: Add aRFS support")
Signed-off-by: Manish Chopra
Signed-off-by: Igor Russkikh
Signed-off-by: Michal Kalderon
Signed-off-by: Dmitry Bogdanov
---
d
This patchset fixes some recent issues found by customers.
v3:
resending on Dmitry's behalf
v2:
correct hash in Fixes tag
Dmitry Bogdanov (3):
net: qed: Disable aRFS for NPAR and 100G
net: qede: Disable aRFS for NPAR and 100G
net: qed: RDMA personality shouldn't fail VF load
drivers/
mitry Bogdanov wrote:
> In CMT and NPAR the PF is unknown when the GFS block processes the
> packet. Therefore cannot use searcher as it has a per PF database,
> and thus ARFS must be disabled.
>
> Fixes: d51e4af5c209 ("qed: aRFS infrastructure support")
> Signed-off-by: Man
> drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c | 5 ++---
> 1 file changed, 2 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
> b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
> index 1426c691c7c4..0346771396ce 100644
> --- a/drivers/net/e
In case recovery was not successful, netdev still should be
present. But we should clear cdev if something bad happens
on recovery.
We also check cdev for null on dev close. That could be a case
if recovery was not successful.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Remove forcible recovery trigger and put it as a normal devlink
callback.
This allows user to enable/disable it via
devlink health set pci/:03:00.0 reporter fw_fatal auto_recover false
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
Acked
Remove extra indent on some of adjacent declarations.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
include/linux/qed/qed_if.h | 69 ++
1 file changed, 33 insertions(+), 36 deletions(-)
diff --git a
Gather and push out full device dump to devlink.
Device dump is the same as with `ethtool -d`, but now its generated
exactly at the moment bad thing happens.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/qed
We are extending devlink infrastructure, thus move the existing
stuff into a new file qed_devlink.c
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
Reviewed-by: Jesse Brandeburg
---
drivers/net/ethernet/qlogic/qed/Makefile | 1 +
drivers
object, devlink private data contains a linkage structure,
qed_devlink.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/qed/qed.h | 1 -
drivers/net/ethernet/qlogic/qed/qed_devlink.c | 40
nging serial to board.serial and fw to fw.app
v5: improved patch 4 description
v4:
- commit message and other fixes after Jiri's comments
- removed one patch (will send to net)
v3: fix uninit var usage in patch 11
v2: fix #include issue from kbuild test robot.
Igor Russkikh (10):
qed: move o
Use devlink_health_report to push error indications.
We implement this in qede via callback function to make it possible
to reuse the same for other drivers sitting on top of qed in future.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
Acked-by
ng loaded
on each start by the driver from FW binary blob.
The base device specific structure (qed_dev_info) was not directly
available to the base driver before. Thus, here we create and store
a private copy of this structure in qed_dev root object to
access the data.
Signed-off-by: Igor Russkik
Here we declare health reporter ops (empty for now)
and register these in qed probe and remove callbacks.
This way we get devlink attached to all kind of qed* PCI
device entities: networking or storage offload entity.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by
This patch replaces stubs in kconfig help entries with an actual description.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/Kconfig | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers
>> ~$ sudo ~/iproute2/devlink/devlink dev info
>> pci/:01:00.1:
>> driver qed
>> board.serial_number REE1915E44552
>> versions:
>> running:
>> fw.app 8.42.2.0
>> stored:
>> fw.mgmt 8.52.10.0
>
> Are you not able to report the running version of the stored f
> > v6: patch 4: changing serial to board.serial and fw to fw.app
> > v5: improved patch 4 description
> > v4:
> > - commit message and other fixes after Jiri's comments
> > - removed one patch (will send to net)
> > v3: fix uninit var usage in patch 11
> > v2: fix #include issue from kbuild t
Here we declare health reporter ops (empty for now)
and register these in qed probe and remove callbacks.
This way we get devlink attached to all kind of qed* PCI
device entities: networking or storage offload entity.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by
Gather and push out full device dump to devlink.
Device dump is the same as with `ethtool -d`, but now its generated
exactly at the moment bad thing happens.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/qed
This patch replaces stubs in kconfig help entries with an actual description.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/Kconfig | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers
ommit message and other fixes after Jiri's comments
- removed one patch (will send to net)
v3: fix uninit var usage in patch 11
v2: fix #include issue from kbuild test robot.
Igor Russkikh (10):
qed: move out devlink logic into a new file
qed/qede: make devlink survive recovery
qed: fix kc
ng loaded
on each start by the driver from FW binary blob.
The base device specific structure (qed_dev_info) was not directly
available to the base driver before. Thus, here we create and store
a private copy of this structure in qed_dev root object to
access the data.
Signed-off-by: Igor Russkik
In case recovery was not successful, netdev still should be
present. But we should clear cdev if something bad happens
on recovery.
We also check cdev for null on dev close. That could be a case
if recovery was not successful.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Fix indent on some of adjacent declarations.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
include/linux/qed/qed_if.h | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/include/linux/qed/qed_if.h b/include/linux
We are extending devlink infrastructure, thus move the existing
stuff into a new file qed_devlink.c
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/qed/Makefile | 1 +
drivers/net/ethernet/qlogic/qed
object, devlink private data contains a linkage structure,
qed_devlink.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/qed/qed.h | 1 -
drivers/net/ethernet/qlogic/qed/qed_devlink.c | 40
Remove forcible recovery trigger and put it as a normal devlink
callback.
This allows user to enable/disable it via
devlink health set pci/:03:00.0 reporter fw_fatal auto_recover false
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
Use devlink_health_report to push error indications.
We implement this in qede via callback function to make it possible
to reuse the same for other drivers sitting on top of qed in future.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
> using
>> usleep() should work and avoid busy spinning.
>>
>> Use readx_poll_timeout() in the poll loop.
>>
>> Cc: Mark Starovoytov
>> Cc: Igor Russkikh
>> Signed-off-by: Sebastian Andrzej Siewior
>
> Fixes: 8dcf2ad39fdb2 ("net: atlantic
Hi Jakub,
On 04/08/2020 12:29 am, Jakub Kicinski wrote:
>> ~$ sudo ~/iproute2/devlink/devlink dev info
>> pci/:01:00.0:
>> driver qed
>> serial_number REE1915E44552
>
> Just to confirm - this is not:
>
>* - ``board.serial_number``
> - Board serial number of the device.
>
>
In case recovery was not successful, netdev still should be
present. But we should clear cdev if something bad happens
on recovery.
We also check cdev for null on dev close. That could be a case
if recovery was not successful.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Fix indent on some of adjacent declarations.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
include/linux/qed/qed_if.h | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/include/linux/qed/qed_if.h b/include/linux
Gather and push out full device dump to devlink.
Device dump is the same as with `ethtool -d`, but now its generated
exactly at the moment bad thing happens.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/qed
ore. Thus, here we create and store
a private copy of this structure in qed_dev root object to
access the data.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/qed/qed.h | 1 +
drivers/net/ethernet/qlogic/qed/
Here we declare health reporter ops (empty for now)
and register these in qed probe and remove callbacks.
This way we get devlink attached to all kind of qed* PCI
device entities: networking or storage offload entity.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by
object, devlink private data contains a linkage structure,
qed_devlink.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/qed/qed.h | 1 -
drivers/net/ethernet/qlogic/qed/qed_devlink.c | 40
We are extending devlink infrastructure, thus move the existing
stuff into a new file qed_devlink.c
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/qed/Makefile | 1 +
drivers/net/ethernet/qlogic/qed
This patch replaces stubs in kconfig help entries with an actual description.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/Kconfig | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers
Use devlink_health_report to push error indications.
We implement this in qede via callback function to make it possible
to reuse the same for other drivers sitting on top of qed in future.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
Remove forcible recovery trigger and put it as a normal devlink
callback.
This allows user to enable/disable it via
devlink health set pci/:03:00.0 reporter fw_fatal auto_recover false
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
one patch (will send to net)
v3: fix uninit var usage in patch 11
v2: fix #include issue from kbuild test robot.
Igor Russkikh (10):
qed: move out devlink logic into a new file
qed/qede: make devlink survive recovery
qed: fix kconfig help entries
qed: implement devlink info request
q
>> Thus, here we create and store a private copy of this structure
>> in qed_dev root object.
>
> Please include example output of devlink info on you device.
Hi Jakub, will do. Here is an example:
~$ sudo ~/iproute2/devlink/devlink dev info
pci/:01:00.0:
driver qed
serial_number REE1
Fix indent on some of adjacent declarations.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
include/linux/qed/qed_if.h | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/include/linux/qed/qed_if.h b/include/linux
We are extending devlink infrastructure, thus move the existing
stuff into a new file qed_devlink.c
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/qed/Makefile | 1 +
drivers/net/ethernet/qlogic/qed
t)
v3: fix uninit var usage in patch 11
v2: fix #include issue from kbuild test robot.
Igor Russkikh (10):
qed: move out devlink logic into a new file
qed/qede: make devlink survive recovery
qed: fix kconfig help entries
qed: implement devlink info request
qed: health reporter init deinit seq
This patch replaces stubs in kconfig help entries with an actual description.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/Kconfig | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers
Here we declare health reporter ops (empty for now)
and register these in qed probe and remove callbacks.
This way we get devlink attached to all kind of qed* PCI
device entities: networking or storage offload entity.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by
In case recovery was not successful, netdev still should be
present. But we should clear cdev if something bad happens
on recovery.
We also check cdev for null on dev close. That could be a case
if recovery was not successful.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Gather and push out full device dump to devlink.
Device dump is the same as with `ethtool -d`, but now its generated
exactly at the moment bad thing happens.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/qed
Remove forcible recovery trigger and put it as a normal devlink
callback.
This allows user to enable/disable it via
devlink health set pci/:03:00.0 reporter fw_fatal auto_recover false
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
Use devlink_health_report to push error indications.
We implement this in qede via callback function to make it possible
to reuse the same for other drivers sitting on top of qed in future.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
object, devlink private data contains a linkage structure,
qed_devlink.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/qed/qed.h | 1 -
drivers/net/ethernet/qlogic/qed/qed_devlink.c | 40
Here we return existing fw & mfw versions, we also fetch device's
serial number.
The base device specific structure (qed_dev_info) was not directly
available to the base driver before.
Thus, here we create and store a private copy of this structure
in qed_dev root object.
Signed-off
On 29/07/2020 4:08 pm, Jiri Pirko wrote:
>
> --
> Wed, Jul 29, 2020 at 01:38:42PM CEST, irussk...@marvell.com wrote:
>> Use devlink_health_report to push error indications.
>> We implement this in qede via callback function to
Fix indent on some of adjacent declarations.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
include/linux/qed/qed_if.h | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/include/linux/qed/qed_if.h b/include/linux
Gather and push out full device dump to devlink.
Device dump is the same as with `ethtool -d`, but now its generated
exactly at the moment bad thing happens.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/qed
In case recovery was not successful, netdev still should be
present. But we should clear cdev if something bad happens
on recovery.
We also check cdev for null on dev close. That could be a case
if recovery was not successful.
Signed-off-by: Igor Russkikh
Signed-off-by: Alexander Lobakin
1 - 100 of 741 matches
Mail list logo
