From: YueHaibing
drivers/net/dsa/mt7530.c:649:3-4: Unneeded semicolon
drivers/net/ethernet/cisco/enic/enic_clsf.c:35:2-3: Unneeded semicolon
drivers/net/ethernet/faraday/ftgmac100.c:1640:2-3: Unneeded semicolon
drivers/net/ethernet/mediatek/mtk_eth_soc.c:229:2-3: Unneeded semicolon
drivers/net/us
From: Jiri Pirko
Follow-up patch is going to need a devlink port instance according to
a netdev. Devlink port instance should be always available when devlink
is used. So change the recently introduced ndo_get_devlink to
ndo_get_devlink_port. With that, adjust the wrapper for the only
user to get
From: Jiri Pirko
phys_port_name may be assembled by a helper in devlink. It is currently
the case only for mlxsw driver. Benefit from the get_devlink_port ndo
and call into devlink directly from dev_get_phys_port_name(). That saves
the trip to the driver, simplifies the code and makes it similar
From: Jiri Pirko
Now it is unused, remove it.
Signed-off-by: Jiri Pirko
---
include/net/devlink.h | 9 -
net/core/devlink.c| 7 ---
2 files changed, 16 deletions(-)
diff --git a/include/net/devlink.h b/include/net/devlink.h
index ae2cd34da99e..227c453cc20e 100644
--- a/include
From: Jiri Pirko
Rely on the previously introduce fallback and let the core directly call
devlink in order to get the physical port name.
Signed-off-by: Jiri Pirko
---
drivers/net/ethernet/mellanox/mlxsw/core.c | 10 --
drivers/net/ethernet/mellanox/mlxsw/core.h | 2 --
driver
From: Jiri Pirko
Introduce devlink_compat_phys_port_name_get() helper that
gets the physical port name for specified netdevice
according to devlink port attributes.
Call this helper from dev_get_phys_port_name()
in case ndo_get_phys_port_name is not defined.
Signed-off-by: Jiri Pirko
---
inclu
From: Jiri Pirko
In order for devlink compat functions to work, implement
ndo_get_devlink_port.
Signed-off-by: Jiri Pirko
---
drivers/net/ethernet/mellanox/mlxsw/core.c | 12
drivers/net/ethernet/mellanox/mlxsw/core.h | 3 +++
drivers/net/ethernet/mellanox/mlxsw/spectrum.
From: Nicholas Nunley
Introduce a new ioctl for applying per-queue parameters.
Users can apply commands to specific queues by setting SUB_COMMAND and
queue_mask with the following ethtool command:
ethtool -Q|--per-queue DEVNAME [queue_mask %x] SUB_COMMAND
If queue_mask is not set, the SUB_COMM
From: Nicholas Nunley
Move option parsing code into find_option function.
No behavior changes.
Based on patch by Kan Liang
Signed-off-by: Nicholas Nunley
---
ethtool.c | 49 +++--
1 file changed, 31 insertions(+), 18 deletions(-)
diff --git a/eth
From: Nicholas Nunley
This patch adds the ability to configure the coalesce settings from
do_scoalesce on a per-queue basis.
For each masked queue the current settings are read, modified, and written
back to the kernel.
Example:
$ sudo ./ethtool --per-queue eth5 queue_mask 0x1 --coalesce
rx-
> >> > Hi Flavio
> >> >
> >> > u-boot should be able to write the MAC address in the correct part of
> >> > device tree. Boards have been doing this a long time.
> >> >
> >> > Module parameters are considered bad. You should only do it if you
> >> > have no other option. Here you do have another op
From: Nicholas Nunley
Move the definition of cmdline_coalesce out of do_scoalesce and into a
macro so it can be resused across functions.
No behavior change.
Based on patch by Kan Liang
Signed-off-by: Nicholas Nunley
---
ethtool.c | 142 --
From: Nicholas Nunley
Get all masked queues' coalesce settings from kernel and dump them one by
one.
Example:
$ sudo ./ethtool --per-queue eth5 queue_mask 0x11 --show-coalesce
Queue: 0
Adaptive RX: off TX: off
stats-block-usecs: 0
sample-interval: 0
pkt-rate-low: 0
pkt-rate-high: 0
rx
From: Nicholas Nunley
When the coalesce settings are printed with --show-coalesce a few of the
option names lack the pluralization that is present in the man page and
usage info, but are otherwise identical.
This inconsistency could lead to some confusion if a user attempts to set
the coalesce s
> Let us briefly describe our test setup to ensure everyone on this mailing
> list is one the same page.
>
> Our general setup looks like this:
> 1 $ iw wlp1s0 info
> Interface wlp1s0
> ifindex 5
> wdev 0x1
> addr 4c:5e:0c:11:43:ac
> type managed
> wiphy 0
> txp
Hello!
On 01.03.2019 8:56, Su Yanjun wrote:
From: Su Yanjun
Because nf_conntrack_helper_unregister maybe used in an unloadable module,
it uses 'synchronize_rcu' which may cause kernel panic.
According to the artical:
Article?
RCU and Unloadable Modules
https://lwn.net/Articles/217484/
On 03/01/2019 04:51 AM, Jakub Kicinski wrote:
> On Fri, 1 Mar 2019 00:18:24 +0100, Daniel Borkmann wrote:
>> This work adds two new map creation flags BPF_F_RDONLY_PROG
>> and BPF_F_WRONLY_PROG in order to allow for read-only or
>> write-only BPF maps from a BPF program side.
>>
>> Today we have B
On 03/01/2019 06:52 AM, Andrii Nakryiko wrote:
> On Thu, Feb 28, 2019 at 3:31 PM Daniel Borkmann wrote:
>>
>> Trivial addition to allow '.' aside from '_' as "special" characters
>> in the object name. Used to name maps from loader side as ".bss",
>> ".data", ".rodata".
>>
>> Signed-off-by: Daniel
On 2019/3/1 16:43, Sergei Shtylyov wrote:
Hello!
On 01.03.2019 8:56, Su Yanjun wrote:
From: Su Yanjun
Because nf_conntrack_helper_unregister maybe used in an unloadable
module,
it uses 'synchronize_rcu' which may cause kernel panic.
According to the artical:
Article?
I got it.
R
Hi Simon,
On 2/27/2019 7:02 PM, Simon Huelck wrote:
> Hi,
>
>
> the thing is , that im not a stmmac developer. Yes , maybe i can bissect
> it and yes you are lucky since im a C-developer since a long time for
> embedded systems.
>
> The problem is that i dont understand the structure of stmmac
On 03/01/2019 06:46 AM, Andrii Nakryiko wrote:
> On Thu, Feb 28, 2019 at 3:31 PM Daniel Borkmann wrote:
>>
>> This generic extension to BPF maps allows for directly loading an
>> address residing inside a BPF map value as a single BPF ldimm64
>> instruction.
>
> This is great! I'm going to review
Some of these macros were conflicting with global namespace,
hence prefixing them with CXGB4.
Signed-off-by: Arjun Vynipadath
Signed-off-by: Vishal Kulkarni
---
drivers/net/ethernet/chelsio/cxgb4/cudbg_lib.c | 2 +-
drivers/net/ethernet/chelsio/cxgb4/cxgb4.h | 24 ++---
drivers/n
Some of these macros were conflicting with global namespace,
hence prefixing them with CXGB4VF.
Signed-off-by: Arjun Vynipadath
Signed-off-by: Vishal Kulkarni
---
drivers/net/ethernet/chelsio/cxgb4vf/adapter.h | 12 ++--
.../net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c| 68 +++---
Use ndo_set_vf_link_state to control the link states associated
with the virtual interfaces.
Signed-off-by: Arjun Vynipadath
Signed-off-by: Vishal Kulkarni
---
v2:
- Using reverse christmas tree for variable declaration
- Renamed FW_LINK* macros for vf link states to make it more clean.
---
dri
This series of patches adds support for ndo_set_vf_link_state in
cxgb4 driver.
Patch 1 implements ndo_set_vf_link_state
Patch 2 reverts the existing force_link_up behaviour for cxgb4vf driver.
v2:
- Using reverse christmas tree for variable declaration in Patch 1
- Patch 2 has no change
Arjun Vy
Reverting force link up changes since this behaviour can be
achieved using VF link state feature.
Reverts:
commit 0913667ab3ad ("cxgb4vf: Forcefully link up virtual interfaces")
Signed-off-by: Arjun Vynipadath
Signed-off-by: Vishal Kulkarni
---
v2:
- No changes
---
drivers/net/ethernet/chelsio
On 01. 03. 19 0:49, Florian Fainelli wrote:
Because we skip the prepare phase, we would not get a chance to have the
port_vlan_prepare() callback return -EOPNOTSUPP and tell us about that.
This causes problems with mv88e6xxx which specifically checks for VLAN
ID = 0. Turns out we do not actually
netif_carrier_off() should be called only after register_netdev().
Signed-off-by: Arjun Vynipadath
Signed-off-by: Vishal Kulkarni
---
drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb
Do you see this Kernel warning while running selftests bpf tes_sock
on x86_64 running Linux kernel version 5.0.0-rc8-next-20190301 ?
selftests: bpf: test_sock
Test case: bind4 load with invalid access: src_ip6 .. [PASS]
Test case: bind4 load with invalid access: mark .. [PASS]
Test case: bind6
On 03/01/2019 07:53 AM, Andrii Nakryiko wrote:
> On Thu, Feb 28, 2019 at 3:31 PM Daniel Borkmann wrote:
>>
>> This work adds BPF loader support for global data sections
>> to libbpf. This allows to write BPF programs in more natural
>> C-like way by being able to define global variables and const
Replace the while loop in MDIO read/write functions with a timeout.
In addition, add a check for MDIO bus busy before initiating a new
operation as well to make sure there is no ongoing MDIO operation.
Signed-off-by: Shubhrajyoti Datta
Signed-off-by: Sai Pavan Boddu
Signed-off-by: Harini Katakam
TSU clock needs to be enabled/disabled as per support in devicetree
and it should also be controlled during suspend/resume (WOL has no
dependency on this clock).
Signed-off-by: Harini Katakam
---
v3 and v2:
No changes
drivers/net/ethernet/cadence/macb.h | 3 ++-
drivers/net/ethernet/caden
When macb device is suspended and system is powered down, the clocks
are removed and hence macb should be closed gracefully and restored
upon resume. This patch does the same by switching off the net device,
suspending phy and performing necessary cleanup of interrupts and BDs.
Upon resume, all the
Add runtime pm functions and move clock handling there.
Add runtime PM calls to mdio functions to allow for active mdio bus.
Signed-off-by: Shubhrajyoti Datta
Signed-off-by: Harini Katakam
---
v3 changes:
Fix exit path using goto
v2 changes:
Allow for mdio bus to be active
Changes from RFC:
Up
This series adds support for macb suspend/resume with system power down.
In relation to the above, this series also updates mdio_read/write
function for PM and adds tsu clock management.
Harini Katakam (4):
net: macb: Check MDIO state before read/write and use timeouts
net: macb: Support clock
When using a C45 PHY that is in PHY_FORCING state we are currently not
taking into account that this kind of PHY has different update_link
functions.
Use the C45 Helpers instead or the driver built-in read_status() helper,
if possible.
Signed-off-by: Jose Abreu
Cc: Andrew Lunn
Cc: Florian Faine
For -net. Please review!
Cc: Andrew Lunn
Cc: Florian Fainelli
Cc: "David S. Miller"
Cc: Joao Pinto
Jose Abreu (2):
net: phy: Use C45 Helpers in phy_read_status()
net: phy: Use C45 Helpers in PHY_FORCING state
drivers/net/phy/phy.c | 2 +-
include/linux/phy.h | 15 +++
2 f
The patch fixes the computation of RXQs being used by the PPv2 driver,
which is set depending on the PPv2 engine version and the queue mode
used. There are three cases:
- PPv2.1: 1 RXQ per CPU.
- PPV2.2 with MVPP2_QDIST_MULTI_MODE: 1 RXQ per CPU.
- PPv2.2 with MVPP2_QDIST_SINGLE_MODE: 1 RXQ is sha
Currently phy_read_status() considers that either the PHY driver has the
read_status() callback or uses the generic callback.
For C45 PHYs we need to use the gen10g_read_status() callback.
Signed-off-by: Jose Abreu
Cc: Andrew Lunn
Cc: Florian Fainelli
Cc: "David S. Miller"
Cc: Joao Pinto
---
Hello,
This series aims to improve the Marvell PPv2 driver and to fix various
issues we encountered while testing the ports in many different
configurations. The series is based on top of Russell PPv2 phylink
rework and improvement.
I'm not sending a v2 of the previous fixes series as half the pa
This patch makes sure both PPv2 MACs (GMAC + XLG MAC) are set in reset
while a port is reconfigured. This is done so that we make sure a MAC is
in a reset state when not used, as only one of the two will be set out
of reset after the port is configured properly.
Signed-off-by: Antoine Tenart
---
This patch modifies the port_disable() helper to always disable both the
GMAC and the XLG MAC when called. At boot time we do not know of a port
was enabled in the firmware/bootloader, and if so what mode was used
(hence which of the two MACs was used).
This also help in implementing a logic where
This patch adds calls in the stop() helper to ensure both MACs and
both PCS blocks are set in reset when the user manually sets a port
down. This is done so that we have the exact same block reset states at
boot time and when a port is set down.
Signed-off-by: Antoine Tenart
---
drivers/net/ethe
This patch reworks the way the XLG MAC is set in reset: the XLG MAC is
set in reset at probe time and taken out of this state only when used.
The idea is to move forward a situation where only the blocks used are
taken out of reset. This also has the effect to handle the GMAC and the
XLG MAC in a s
This patch sets both the XPCS and MPCS blocks in reset when they aren't
used. This is done both at boot time and when reconfiguring a port mode.
The advantage now is that only the PCS used is set out of reset when the
port is configured (10GKR uses the MCPS while RXAUI uses the XPCS).
Signed-off-b
We had a check in the mvpp2_mac_link_down() function (called by phylink)
to avoid disabling the port when link interrupts are used. It turned out
the interrupt can still be used with the port disabled. We can thus
remove this check.
Signed-off-by: Antoine Tenart
---
drivers/net/ethernet/marvell/
This patch improves the XLG configuration function, to only update the
XLG configuration register when a change is needed. This helps not
writing over and over the same XLG configuration each time phylink
request the MAC to be configured. This mimics the GMAC configuration
function.
Signed-off-by:
The GMAC configuration helper modifies values in the auto-negotiation
register. Some of its values require the port to be forced down when
modifying their values. This patches fixes the check made on the bit to
be updated in this register, so that the port is forced down when
needed. This fix cases
This patch force the XLG MAC link state in the phylink link_up() and
link_down() helpers when not using in-band auto-negotiation. This mimics
what's already done for the GMAC and follows what's advised in the
phylink documentation.
Signed-off-by: Antoine Tenart
---
drivers/net/ethernet/marvell/m
The Phylink validate function is the Marvell PPv2 driver makes a check
on the GoP id. This is valid an has to be done when using PPv2.2 engines
but makes no sense when using PPv2.1. The check done when using an RGMII
interface makes sure the GoP id is not 0, but this breaks PPv2.1. Fixes
it.
Fixes
This patch adds a check on the PPv2 version in-use not to reconfigure
the port mode when an interface is updated when using PPv2.1 as the
functions called are PPv2.2 specific.
Signed-off-by: Antoine Tenart
---
drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c | 14 ++
1 file changed, 6
The Marvell PPv2 port structure stores the GoP id of a given port. This
information is specific to PPv2.2, but cannot be used by PPv2.1. Update
its comment to denote this specificity.
Signed-off-by: Antoine Tenart
---
drivers/net/ethernet/marvell/mvpp2/mvpp2.h | 2 +-
1 file changed, 1 insertion
Cosmetic patch fix the alignment of the MVPP2_GMAC_CONFIG_MII_SPEED
macro definition.
Signed-off-by: Antoine Tenart
---
drivers/net/ethernet/marvell/mvpp2/mvpp2.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2.h
b/drivers/net/etherne
This cosmetic patch fixes a typo made in a comment in the Marvell PPv2
Ethernet driver header.
Signed-off-by: Antoine Tenart
---
drivers/net/ethernet/marvell/mvpp2/mvpp2.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2.h
b/drivers/ne
Hello,
This series implements the suspend/resume callbacks in the marvell10g
PHY driver:
- When the PHY isn't used, it is set in low power mode.
- At boot time we might now know the PHY status (as it's depending on
the hardware configuration, or on what the previous stages
configured), it is
This patch adds the suspend/resume callbacks for Marvell 10G PHYs. The
three PCS (base-t, base-r and 1000base-x) are set in low power (the PCS
are powered down) when the PHY isn't used.
Signed-off-by: Antoine Tenart
---
drivers/net/phy/marvell10g.c | 14 ++
1 file changed, 14 inserti
When the 88x2110 PHY support was added, the suspend and resume callbacks
were forgotten. This patch adds them to the 88x2110 PHY callback
definition.
Signed-off-by: Antoine Tenart
---
drivers/net/phy/marvell10g.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/phy/marvell10g.c
When the Marvell 10G PHYs are set out of reset, the LPOWER bit is set
depending on an hardware configuration choice. We also do not know what
is the PHY state at boot time. Hence, set the PHY in low power by
default when this driver probes.
Signed-off-by: Antoine Tenart
---
drivers/net/phy/marve
Hi Russell,
I've been using this patch on my local branch for the past 3 weeks, and
I saw no obvious issue so far. (I also used a similar patch before).
Do you have pending questions regarding this patch? Otherwise I think
you can sent it again without the RFC tag, with my Tested-by and/or
Acked-
Hi all,
We encountered a problem that if there are two devices in kernel v5.0 with
marvell phy 88E1510 connect with each other directly, one with autoneg on and
the other off. The one who has disabled auto-negotiation will failed on setting
speed and duplex mode. Their speed and duplex mode will
On Thu, 28 Feb 2019 19:06:45 -0800
Jakub Kicinski wrote:
> On Thu, 28 Feb 2019 13:54:38 -0800, Stephen Hemminger wrote:
> > @@ -12140,12 +12144,14 @@ static int i40e_xdp(struct net_device *dev,
> > struct i40e_netdev_priv *np = netdev_priv(dev);
> > struct i40e_vsi *vsi = np->vsi;
> >
>
Hello Mark,
I am sorry to bother you.
Do you think you can take this patch?
Thanks,
Fab
> From: Fabrizio Castro
> Sent: 13 December 2018 20:20
> Subject: [PATCH] dt-bindings: can: rcar_can: Add r8a774c0 support
>
> Document RZ/G2E (r8a774c0) SoC specific bindings.
>
> Signed-off-by: Fabrizio
I can reproduce this on 4.19.0-3-amd64 both with, and without the JIT enabled.
Dumping the "root" and "non-root" programs with bpftool,
the subtraction instructions differ:
"non-root":
0: (85) call bpf_ktime_get_ns#74944
1: (bf) r7 = r0
2: (85) call bpf_ktime_get_ns#74944
3: (bf) r6 =
Hello Marc,
I am sorry to bother you.
Do you think you can take this patch?
Thanks,
Fab
> From: Fabrizio Castro
> Sent: 16 January 2019 18:38
> Subject: [PATCH 09/11] dt-bindings: can: rcar_can: Fix RZ/G2 CAN clocks
>
> According to the latest information, the clock options for CAN on RZ/G2
>
Hello Marc,
I am sorry to bother you.
Do you think you can take this patch?
Thanks,
Fab
> From: Fabrizio Castro
> Sent: 16 January 2019 18:38
> Subject: [PATCH 11/11] dt-bindings: can: rcar_canfd: document r8a774c0 support
>
> Document the support for rcar_canfd on R8A774C0 SoC devices.
>
> S
Hello Marc,
I am sorry to bother you.
Do you think you can take this patch?
Thanks,
Fab
> From: Fabrizio Castro
> Sent: 17 January 2019 14:54
> Subject: [PATCH 3/3] dt-bindings: can: rcar_canfd: Document r8a774a1 support
>
> Document support for rcar_canfd on R8A774A1 SoC devices.
>
> Signed-
Good day,
According to this link:
http://linux-tc-notes.sourceforge.net/tc/doc/sch_noqueue.txt it should
not be possible to assign noqueue qdisc to physical devices or
classes. Yet, I can clearly do it even on the latest 5.0-rc series
kernels:
$ sudo tc qdisc add dev eth0 root noqueue
$ sudo tc q
On Wed, Feb 27, 2019 at 08:24:00PM -0800, Florian Fainelli wrote:
On 2/26/2019 10:45 AM, Ivan Khoronzhuk wrote:
Despite this is supposed to be used for Ethernet VLANs, not Ethernet
addresses with space for VID also can reuse this, so VID is considered
as virtual ID extension, not belonging stric
On Wed, Feb 27, 2019 at 08:09:44PM -0800, Florian Fainelli wrote:
On 2/26/2019 10:45 AM, Ivan Khoronzhuk wrote:
Update vlan mc and uc addresses with VID tag while propagating
addresses to lower devices, do this only if address is not synced.
It allows at end driver level to distinguish address
On Wed, Feb 27, 2019 at 08:13:34PM -0800, Florian Fainelli wrote:
On 2/26/2019 10:45 AM, Ivan Khoronzhuk wrote:
The vlan device address is held separately from uc/mc lists and
handled differently. The vlan dev address is bound with real device
address only if it's inherited from init, in all o
Add 36 pedit action tests to check pedit options described in tc-pedit(8)
man page. Test cases can be specified by categories: actions, pedit,
raw_op, layered_op. RAW_OP cases check offset option for u8, u16 and u32
offset size. LAYERED_OP cases check fields option for eth, ip, ip6,
tcp and udp hea
On Fri, Mar 01, 2019 at 09:13:58AM +0100, Jiri Pirko wrote:
> From: Jiri Pirko
>
> Follow-up patch is going to need a devlink port instance according to
> a netdev. Devlink port instance should be always available when devlink
> is used.
I noticed netdevsim has devlink interface but it does not
On 03/01/2019 12:39 PM, Arthur Fabre wrote:
> I can reproduce this on 4.19.0-3-amd64 both with, and without the JIT enabled.
>
> Dumping the "root" and "non-root" programs with bpftool,
> the subtraction instructions differ:
>
> "non-root":
>0: (85) call bpf_ktime_get_ns#74944
>1: (bf) r7
On 01. 03. 19 12:26, liweihang wrote:
Hi all,
We encountered a problem that if there are two devices in kernel v5.0 with
marvell phy 88E1510 connect with each other directly, one with autoneg on and
the other off. The one who has disabled auto-negotiation will failed on setting
speed and duplex
On Wed, Feb 27, 2019 at 08:29:20PM -0800, Florian Fainelli wrote:
On 2/26/2019 10:45 AM, Ivan Khoronzhuk wrote:
IVDF - individual virtual device filtering. Allows to set per vlan
l2 address filters on end real network device (for unicast and for
multicast) and drop redundant not expected packe
On Thu, Feb 28, 2019 at 05:30:56PM -0800, si-wei liu wrote:
>
>
> On 2/28/2019 6:26 AM, Michael S. Tsirkin wrote:
> > On Thu, Feb 28, 2019 at 01:32:12AM -0800, si-wei liu wrote:
> > > > > Will the
> > > > > change break userspace further?
> > > > >
> > > > > -Siwei
> > > > Didn't you show usersp
On Mon, Feb 25, 2019 at 07:27:43PM +0800, Xin Long wrote:
> sctp_csum_check() is called by sctp_s/dnat_handler() where it calls
> skb_make_writable() to ensure sctphdr to be linearized.
>
> So there's no need to get sctphdr by calling skb_header_pointer()
> in sctp_csum_check().
Applied, thanks.
On Mon, Feb 25, 2019 at 04:21:14PM -0800, David Ahern wrote:
> From: David Ahern
>
> Followup to a173f066c7cf ("netfilter: bridge: Don't sabotage nf_hook
> calls from an l3mdev"). Some packets (e.g., ndisc) do not have the skb
> device flipped to the l3mdev (e.g., VRF) device. Update ip_sabotage_
On Wed, Feb 27, 2019 at 10:19:10AM -0800, Sami Tolvanen wrote:
> Use struct device_attribute instead of struct idletimer_tg_attr, and
> the correct callback function type to avoid indirect call mismatches
> with Control Flow Integrity checking.
Applied, thanks.
When calculating the multipath hash for input routes the flow info is
not available and therefore should not be used.
Fixes: 24ba14406c5c ("route: Add multipath_hash in flowi_common to make
user-define hash")
Signed-off-by: Ido Schimmel
Cc: wenxu
---
net/ipv4/route.c | 2 +-
1 file changed, 1
On Fri, Mar 01, 2019 at 12:15:27AM -0800, Jeff Kirsher wrote:
> From: Nicholas Nunley
>
> Move option parsing code into find_option function.
>
> No behavior changes.
>
> Based on patch by Kan Liang
>
> Signed-off-by: Nicholas Nunley
Reviewed-by: Michal Kubecek
On Fri, Mar 01, 2019 at 11:54:22AM +0100, Jose Abreu wrote:
> For -net. Please review!
Hi Jose
Patches for net should have fixes: tags, making it clear what they
fix, and how far back the patches should be ported.
Thanks
Andrew
On Fri, Mar 01, 2019 at 12:15:28AM -0800, Jeff Kirsher wrote:
> From: Nicholas Nunley
>
> Move the definition of cmdline_coalesce out of do_scoalesce and into a
> macro so it can be resused across functions.
>
> No behavior change.
>
> Based on patch by Kan Liang
>
> Signed-off-by: Nicholas N
Fri, Mar 01, 2019 at 01:45:26PM CET, mkube...@suse.cz wrote:
>On Fri, Mar 01, 2019 at 09:13:58AM +0100, Jiri Pirko wrote:
>> From: Jiri Pirko
>>
>> Follow-up patch is going to need a devlink port instance according to
>> a netdev. Devlink port instance should be always available when devlink
>> i
On Fri, Mar 01, 2019 at 11:54:24AM +0100, Jose Abreu wrote:
> +static inline int phy_update_link(struct phy_device *phydev)
> +{
> + if (!phydev->drv)
> + return -EIO;
> +
> + if (phydev->drv->read_status)
> + return phydev->drv->read_status(phydev);
> + else if
On 03/01/2019 12:39 PM, Arthur Fabre wrote:
> I can reproduce this on 4.19.0-3-amd64 both with, and without the JIT enabled.
>
> Dumping the "root" and "non-root" programs with bpftool,
> the subtraction instructions differ:
>
> "non-root":
>0: (85) call bpf_ktime_get_ns#74944
>1: (bf) r7
Great, appreciated.
One more thing (since upgrading kernels takes time) do you think I can
amend eBPF on my side to avoid triggering this? Naive stuff like this
doesn't work sadly:
uint64_t delta = b + ~a + 1;
I tried couple more variants with uint32_t types, but to no avail. Ideas?
Marek
An XDP program can redirect packets between interfaces using either the
xdp_redirect() helper or the xdp_redirect_map() helper. Apart from the
flexibility of updating maps from userspace, the redirect_map() helper also
uses the map structure to batch packets, which results in a significant
(around
This series changes the xdp_redirect helper to use a hidden default map. The
redirect_map() helper also uses the map structure to batch packets, which
results in a significant (around 50%) performance boost for the _map variant.
However, the xdp_redirect() API is simpler if one just wants to redire
A common pattern when using xdp_redirect_map() is to create a device map
where the lookup key is simply ifindex. Because device maps are arrays,
this leaves holes in the map, and the map has to be sized to fit the
largest ifindex, regardless of how many devices actually are actually
needed in the m
The subsequent commits introducing default maps and a hash-based ifindex
devmap require a bit of refactoring of the devmap code. Perform this first
so the subsequent commits become easier to read.
Signed-off-by: Toke Høiland-Jørgensen
---
kernel/bpf/devmap.c | 177 ++
On Fri, Mar 01, 2019 at 12:00:46PM +0100, Antoine Tenart wrote:
> When the 88x2110 PHY support was added, the suspend and resume callbacks
> were forgotten. This patch adds them to the 88x2110 PHY callback
> definition.
>
> Signed-off-by: Antoine Tenart
Reviewed-by: Andrew Lunn
Andrew
On Fri, Mar 01, 2019 at 12:00:45PM +0100, Antoine Tenart wrote:
> This patch adds the suspend/resume callbacks for Marvell 10G PHYs. The
> three PCS (base-t, base-r and 1000base-x) are set in low power (the PCS
> are powered down) when the PHY isn't used.
>
> Signed-off-by: Antoine Tenart
Review
On Fri, Mar 01, 2019 at 12:15:29AM -0800, Jeff Kirsher wrote:
> From: Nicholas Nunley
>
> Introduce a new ioctl for applying per-queue parameters.
> Users can apply commands to specific queues by setting SUB_COMMAND and
> queue_mask with the following ethtool command:
>
> ethtool -Q|--per-queue
On Fri, Mar 01, 2019 at 12:00:47PM +0100, Antoine Tenart wrote:
> When the Marvell 10G PHYs are set out of reset, the LPOWER bit is set
> depending on an hardware configuration choice. We also do not know what
> is the PHY state at boot time. Hence, set the PHY in low power by
> default when this d
On 03/01/2019 03:10 PM, Marek Majkowski wrote:
> Great, appreciated.
>
> One more thing (since upgrading kernels takes time) do you think I can
> amend eBPF on my side to avoid triggering this? Naive stuff like this
> doesn't work sadly:
>
> uint64_t delta = b + ~a + 1;
>
> I tried couple mo
Hi Christoph,
I saw you sent this patch as part of another series back in August, but
it seems it was never applied and I can't find it even in -next. We made
some tests and this patch is helping a lot the PPv2 engine driver in
improving its performances.
Do you plan on re-sending it, or reworkin
On Fri, Mar 01, 2019 at 12:15:30AM -0800, Jeff Kirsher wrote:
> From: Nicholas Nunley
>
> Get all masked queues' coalesce settings from kernel and dump them one by
> one.
>
> Example:
>
> $ sudo ./ethtool --per-queue eth5 queue_mask 0x11 --show-coalesce
> Queue: 0
> Adaptive RX: off TX: off
On Fri, Mar 01, 2019 at 12:15:31AM -0800, Jeff Kirsher wrote:
> From: Nicholas Nunley
>
> This patch adds the ability to configure the coalesce settings from
> do_scoalesce on a per-queue basis.
>
> For each masked queue the current settings are read, modified, and written
> back to the kernel.
On 2019/3/1 下午9:38, Ido Schimmel wrote:
> When calculating the multipath hash for input routes the flow info is
> not available and therefore should not be used.
>
> Fixes: 24ba14406c5c ("route: Add multipath_hash in flowi_common to make
> user-define hash")
> Signed-off-by: Ido Schimmel
> Cc: we
1 - 100 of 363 matches
Mail list logo
