On Tue, Jul 17, 2018 at 12:46 AM, Mark Bloch wrote:
>> From: Or Gerlitz [mailto:gerlitz...@gmail.com]
>> > If NIC RX flow tables support decap opertion, enable it on creation.
>> opertion --> operation
saw it?
>> > If NIC TX flow tables support reformat opertion, enable it on creation.
ds->enabled_port_mask only contains a bitmask of user-facing enabled
ports, we also need to allow programming CFP rules that target CPU ports
(e.g: ports 5 and 8).
Signed-off-by: Florian Fainelli
---
drivers/net/dsa/bcm_sf2_cfp.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff
When we are in Wake-on-LAN, we operate with the host sofware not running
a network stack, so we want to the switch to flood packets in order to
cause a system wake-up when matching specific filters (unicast or
multicast). This was not necessary before since we supported Magic
Packet which are
On Tue, Jul 17, 2018 at 08:36:37AM -0700, Florian Fainelli wrote:
> Hi all,
>
> This patch series adds support for allowing Wake-on-LAN to wake-up the
> system using configurable filters. This is particular useful in the context
> of Android where wake on MDNS is a requirement.
>
> We support
Fix the warning below by calling rhashtable_lookup under
RCU read lock.
[ 342.450870] WARNING: suspicious RCU usage
[ 342.455856] 4.18.0-rc2+ #17 Tainted: G O
[ 342.462210] -
[ 342.467202] ./include/linux/rhashtable.h:481 suspicious
Fix the warning below by calling the ptr_ring_consume_bh,
which uses spin_[un]lock_bh.
[ 179.064300]
[ 179.069073] WARNING: inconsistent lock state
[ 179.073846] 4.18.0-rc2+ #18 Not tainted
[ 179.078133]
[ 179.082907]
The driver builds fine even with CONFIG_OF=n since we now have stubs that are
provided.
Signed-off-by: Florian Fainelli
---
drivers/net/phy/Kconfig | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index
Hi David,
This patch series drops the CONFIG_OF dependency that some Broadcom drivers
had, this is no longer necessary and goes against allowing build testing
on more platforms.
Let me know if kbuild or your own builds somehow fail.
Florian Fainelli (3):
net: phy: Drop OF dependency for
NET_DSA_BCM_SF2 does not need to depend on CONFIG_OF anymore since we have
stubs when that option is disabled.
Signed-off-by: Florian Fainelli
---
drivers/net/dsa/Kconfig | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/dsa/Kconfig b/drivers/net/dsa/Kconfig
index
Both BCMGENET and SYSTEMPORT build just fine with CONFIG_OF=n, we do have a
dependency on HAS_IOMEM that was not being reflected for SYSTEMPORT so add
that.
Signed-off-by: Florian Fainelli
---
drivers/net/ethernet/broadcom/Kconfig | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff
> On 7/16/2018 8:08 PM, Steve Wise wrote:
> > Hey Max:
> >
> >
>
> Hey,
>
> > On 7/16/2018 11:46 AM, Max Gurtovoy wrote:
> >>
> >>
> >> On 7/16/2018 5:59 PM, Sagi Grimberg wrote:
> >>>
> Hi,
> I've tested this patch and seems problematic at this moment.
> >>>
> >>> Problematic how?
On Sun, Jul 15, 2018 at 12:06 PM, Kamal Heib wrote:
> Move the definition of mlx5e_priv_flags into en_ethtool.c because it's
> only used there.
>
> Fixes: 4e59e2888139 ("net/mlx5e: Introduce net device priv flags
> infrastructure")
> Signed-off-by: Kamal Heib
> ---
LGTM, but why to net ?
this
On Tue, Jul 17, 2018 at 08:36:40AM -0700, Florian Fainelli wrote:
> When we are in Wake-on-LAN, we operate with the host sofware not running
> a network stack, so we want to the switch to flood packets
Hi Florian
Just to be sure...
By flood, you mean from a user port to the CPU. You don't mean
On 7/16/18 3:09 PM, David Miller wrote:
>
> Is this a -stable candidate?
>
I think so. The API is not doing what the user requested, even though
the route add does not fail.
Create a helper function to turn on/off MPD, this will be used to avoid
duplicating code as we are going to add additional types of wake-up
types.
Signed-off-by: Florian Fainelli
---
drivers/net/ethernet/broadcom/bcmsysport.c | 20 +---
1 file changed, 13 insertions(+), 7
Add the ability to specify that a filter, programmed through
ethtool::rxnfc will be used as a wake-up source. sopass which is a
48-bit wide storage is used to indicate which filters (as bits) can be
used for wake-up.
Signed-off-by: Florian Fainelli
---
include/uapi/linux/ethtool.h | 3 ++-
1
Allow re-purposing the wol->sopass storage area to specify a bitmask of filters
(programmed previously via ethtool::rxnfc) to be used as wake-up patterns.
Signed-off-by: Florian Fainelli
---
ethtool-copy.h | 1 +
ethtool.c | 35 ++-
2 files changed, 35
Hi all,
This patch series adds support for allowing Wake-on-LAN to wake-up the
system using configurable filters. This is particular useful in the context
of Android where wake on MDNS is a requirement.
We support this by using the bcm_sf2 Compact Field Processor (CFP) which
supports matching
We already properly resume from Wake-on-LAN whether such a condition
occured or not, no need to process the WoL interrupt for functional
changes since that could race with other settings.
Signed-off-by: Florian Fainelli
---
drivers/net/ethernet/broadcom/bcmsysport.c | 4 +---
1 file changed, 1
The SYSTEMPORT MAC allows up to 8 filters to be programmed to wake-up
from LAN. Verify that we have up to 8 filters and program them to the
appropriate RXCHK entries to be matched (along with their masks).
We need to update the entry and exit to Wake-on-LAN mode to keep the
RXCHK engine running
Propagate wol->sopass when WAKE_FILTER is set since that contains the
bitmap of filters to be enabled for wake-up.
Signed-off-by: Florian Fainelli
---
drivers/net/dsa/bcm_sf2.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/dsa/bcm_sf2.c
On 07/17/2018 08:54 AM, Andrew Lunn wrote:
> On Tue, Jul 17, 2018 at 08:36:40AM -0700, Florian Fainelli wrote:
>> When we are in Wake-on-LAN, we operate with the host sofware not running
>> a network stack, so we want to the switch to flood packets
>
> Hi Florian
>
> Just to be sure...
>
>
Allow other drivers and parts of kernel to use the same define for
CRC32 polynomial, instead of duplicating it in many places. This code
does not bring any functional changes, except moving existing code.
Signed-off-by: Krzysztof Kozlowski
---
include/linux/crc32poly.h | 20
Create initial unit tests for the tc fw filter.
Signed-off-by: Keara Leibovitz
---
.../selftests/tc-testing/tc-tests/filters/fw.json | 1049
1 file changed, 1049 insertions(+)
create mode 100644 tools/testing/selftests/tc-testing/tc-tests/filters/fw.json
diff --git
> >> ethtool -s gphy wol f filters 0x2
> >
> > What does this 0x2 represent?
>
> 0x2 = bit 1 is set, which corresponds to the filter ID that was returned
> from the previous ethtool::rxnfc command invocation. If ethtool
> --config-nfc returned 3, then we would have used filters 0x8, etc.
It
Hi,
Stephen Hemminger writes:
> On Fri, 13 Jul 2018 17:06:11 -0700
> Vinicius Costa Gomes wrote:
>
>> +while (fscanf(f, "%ms %x %" PRIu32 "\n", _str, ,
>> ) != EOF) {
>> +struct rtattr *entry;
>> +
>> +err = str_to_entry_cmd(cmd_str);
>> +
Allow programs and maps to be re-used across different netdevs,
as long as they belong to the same struct bpf_offload_dev.
Update the bpf_offload_prog_map_match() helper for the verifier
and export a new helper for the drivers to use when checking
programs at attachment time.
Signed-off-by: Jakub
Grouping netdevsim devices into "ASICs" will soon be supported.
Add switch_id attribute to all netdevsims. For now each netdevsim
will have its switch_id matching the device id.
Signed-off-by: Jakub Kicinski
Reviewed-by: Quentin Monnet
---
drivers/net/netdevsim/netdev.c| 24
On Tue, Jul 17, 2018 at 10:43 AM David Ahern wrote:
>
> On 7/17/18 11:40 AM, Cong Wang wrote:
> > On Tue, Jul 17, 2018 at 5:11 AM wrote:
> >>
> >> From: David Ahern
> >>
> >> Nikita Leshenko reported that neighbor entries in one namespace can
> >> evict neighbor entries in another. The problem
Create a higher-level entity to represent a device/ASIC to allow
programs and maps to be shared between device ports. The extra
work is required to make sure we don't destroy BPF objects as
soon as the netdev for which they were loaded gets destroyed,
as other ports may still be using them. When
Factor out sharable netdevsim sub-object and use IFLA_LINK to link
netdevsims together at creation time. Sharable object will have
its own DebugFS directory.
Signed-off-by: Jakub Kicinski
Reviewed-by: Quentin Monnet
---
drivers/net/netdevsim/netdev.c| 87 ---
A set of new API functions exported for the drivers will soon use
'bpf_offload_dev_' as a prefix. Rename the bpf_offload_dev_match()
which is internal to the core (used by the verifier) to avoid any
confusion.
Signed-off-by: Jakub Kicinski
Reviewed-by: Quentin Monnet
---
include/linux/bpf.h
Move bound program information from netdevsim to shared sub-object,
as programs will soon be shared between netdevs of the same ASIC.
Signed-off-by: Jakub Kicinski
Reviewed-by: Quentin Monnet
---
drivers/net/netdevsim/bpf.c | 30 -
BPF code should unregister the offload capabilities from .ndo_uninit(),
to make sure the operation is atomic with unlist_netdevice(). Plumb
the init/uninit NDOs for vNICs and representors.
Signed-off-by: Jakub Kicinski
Reviewed-by: Quentin Monnet
---
Currently we have two lists of offloaded objects - programs and maps.
Netdevice unregister notifier scans those lists to orphan objects
associated with device being unregistered. This puts unnecessary
(even if negligible) burden on all netdev unregister calls in BPF-
-enabled kernel. The lists
Hi!
This patchset adds support for sharing BPF objects within one ASIC.
This will allow us to reuse of the same program on multiple ports of
a device leading to better code store utilization. It also enables
sharing maps between programs attached to different ports of a device.
v2:
- rename
Allow program sharing between netdevs of the same NFP ASIC.
Signed-off-by: Jakub Kicinski
Reviewed-by: Quentin Monnet
---
drivers/net/ethernet/netronome/nfp/bpf/offload.c | 10 ++
1 file changed, 2 insertions(+), 8 deletions(-)
diff --git
Add tests for sharing programs and maps between different netdevs.
Use netdevsim's ability to pretend multiple netdevs belong to the
same "ASIC".
Signed-off-by: Jakub Kicinski
Reviewed-by: Quentin Monnet
---
tools/testing/selftests/bpf/test_offload.py | 146 +++-
1 file
Allow program sharing between devices which were linked together.
Signed-off-by: Jakub Kicinski
Reviewed-by: Quentin Monnet
---
drivers/net/netdevsim/bpf.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/netdevsim/bpf.c b/drivers/net/netdevsim/bpf.c
index
On 07/17/2018 06:47 PM, Alexei Starovoitov wrote:
> On Tue, Jul 17, 2018 at 06:10:38PM +0300, Tariq Toukan wrote:
>> Fix the warning below by calling rhashtable_lookup under
>> RCU read lock.
>>
>> [ 342.450870] WARNING: suspicious RCU usage
>> [ 342.455856] 4.18.0-rc2+ #17 Tainted: G
On 07/17/2018 08:47 AM, Andrew Lunn wrote:
> On Tue, Jul 17, 2018 at 08:36:37AM -0700, Florian Fainelli wrote:
>> Hi all,
>>
>> This patch series adds support for allowing Wake-on-LAN to wake-up the
>> system using configurable filters. This is particular useful in the context
>> of Android
Do not define again the polynomial but use header with existing define.
Signed-off-by: Krzysztof Kozlowski
---
lib/decompress_bunzip2.c | 3 ++-
lib/xz/xz_crc32.c| 3 ++-
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/lib/decompress_bunzip2.c b/lib/decompress_bunzip2.c
On Tue, Jul 17, 2018 at 2:16 AM Paolo Abeni wrote:
>
> Hi,
>
> On Mon, 2018-07-16 at 16:39 -0700, Cong Wang wrote:
> > On Fri, Jul 13, 2018 at 2:55 AM Paolo Abeni wrote:
> > >
> > > When mirred is invoked from the ingress path, and it wants to redirect
> > > the processed packet, it can now use
The removed code would be called in two situations:
1. interface is brought up never or >10s after driver load
2. after close()
Case 1 we can handle cleaner by ensuring chip is powered down when
leaving probe(). open() callback will power up the chip.
In case 2 we call rtl_pll_power_down() twice
On 07/17/2018 09:14 AM, Andrew Lunn wrote:
> On Tue, Jul 17, 2018 at 08:36:44AM -0700, Florian Fainelli wrote:
>> The SYSTEMPORT MAC allows up to 8 filters to be programmed to wake-up
>> from LAN. Verify that we have up to 8 filters and program them to the
>> appropriate RXCHK entries to be
On Tue, Jul 17, 2018 at 06:10:38PM +0300, Tariq Toukan wrote:
> Fix the warning below by calling rhashtable_lookup under
> RCU read lock.
>
> [ 342.450870] WARNING: suspicious RCU usage
> [ 342.455856] 4.18.0-rc2+ #17 Tainted: G O
> [ 342.462210] -
> [
On Tue, Jul 17, 2018 at 08:36:44AM -0700, Florian Fainelli wrote:
> The SYSTEMPORT MAC allows up to 8 filters to be programmed to wake-up
> from LAN. Verify that we have up to 8 filters and program them to the
> appropriate RXCHK entries to be matched (along with their masks).
>
> We need to
Allow user-space to provide tos and ttl to be set for the tunnel headers.
Signed-off-by: Or Gerlitz
Reviewed-by: Roi Dayan
Acked-by: Jiri Pirko
---
include/uapi/linux/tc_act/tc_tunnel_key.h | 2 ++
net/sched/act_tunnel_key.c| 20 ++--
2 files changed, 20
Add dissection of the tos and ttl from the ip tunnel headers
fields in case a match is needed on them.
Signed-off-by: Or Gerlitz
Reviewed-by: Roi Dayan
Acked-by: Jiri Pirko
---
include/net/flow_dissector.h | 2 +-
net/core/flow_dissector.c| 14 +-
2 files changed, 14
Hi Dave,
This series comes to address the case to set (encap) and match (decap)
also the tos and ttl fields of TC based IP tunnels.
Example encap (1st one) and decap (2nd) that use the new fields
tc filter add dev eth0_0 protocol ip parent : prio 10 flower \
src_mac
Allow users to set rules matching on ipv4 tos and ttl or
ipv6 traffic-class and hoplimit of tunnel headers.
Signed-off-by: Or Gerlitz
Reviewed-by: Roi Dayan
Acked-by: Jiri Pirko
---
include/uapi/linux/pkt_cls.h | 5 +
net/sched/cls_flower.c | 43
On 07/17/2018 09:21 AM, Andrew Lunn wrote:
ethtool -s gphy wol f filters 0x2
>>>
>>> What does this 0x2 represent?
>>
>> 0x2 = bit 1 is set, which corresponds to the filter ID that was returned
>> from the previous ethtool::rxnfc command invocation. If ethtool
>> --config-nfc returned 3,
> -Original Message-
> From: linux-rdma-ow...@vger.kernel.org [mailto:linux-rdma-
> ow...@vger.kernel.org] On Behalf Of Or Gerlitz
> Sent: Tuesday, July 17, 2018 5:47 AM
> To: Mark Bloch
> Cc: Doug Ledford ; Jason Gunthorpe
> ; Leon Romanovsky ; RDMA
> mailing list ; Saeed Mahameed
> ;
> +if (bitmap_weight(priv->filters, WAKE_FILTER_BITS) >
> + RXCHK_BRCM_TAG_MAX) {
> +bitmap_zero(priv->filters, WAKE_FILTER_BITS);
> +return -ENOSPC;
> +}
>
On Tue, Jul 17, 2018 at 09:28:52AM -0700, Florian Fainelli wrote:
>
>
> On 07/17/2018 09:21 AM, Andrew Lunn wrote:
> ethtool -s gphy wol f filters 0x2
> >>>
> >>> What does this 0x2 represent?
> >>
> >> 0x2 = bit 1 is set, which corresponds to the filter ID that was returned
> >> from the
On Tue, Jul 17, 2018 at 08:00:58AM -0700, Saeed Mahameed wrote:
> On Sun, Jul 15, 2018 at 12:06 PM, Kamal Heib wrote:
> > Move the definition of mlx5e_priv_flags into en_ethtool.c because it's
> > only used there.
> >
> > Fixes: 4e59e2888139 ("net/mlx5e: Introduce net device priv flags
> >
Add dependencies on PCI where necessary.
Fixes: 7e2bc7fb65 ("net: cavium: Drop dependency of NET_VENDOR_CAVIUM on PCI")
Signed-off-by: Alexander Sverdlin
---
drivers/net/ethernet/cavium/Kconfig | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git
> >> struct ethtool_wolinfo *wol)
> >> {
> >>struct bcm_sysport_priv *priv = netdev_priv(dev);
> >>struct device *kdev = >pdev->dev;
> >> - u32 supported = WAKE_MAGIC | WAKE_MAGICSECURE;
> >> + u32 supported = WAKE_MAGIC | WAKE_MAGICSECURE | WAKE_FILTER;
> >> +
On 07/17/2018 09:49 AM, Andrew Lunn wrote:
struct ethtool_wolinfo *wol)
{
struct bcm_sysport_priv *priv = netdev_priv(dev);
struct device *kdev = >pdev->dev;
- u32 supported = WAKE_MAGIC | WAKE_MAGICSECURE;
+ u32 supported =
On Mon, Jul 16, 2018 at 1:31 AM Vlad Buslov wrote:
>
>
> On Fri 13 Jul 2018 at 21:51, Cong Wang wrote:
> > On Fri, Jul 13, 2018 at 6:30 AM Vlad Buslov wrote:
> >>
> >>
> >> On Fri 13 Jul 2018 at 03:52, Cong Wang wrote:
> >> > On Thu, Jul 5, 2018 at 7:24 AM Vlad Buslov wrote:
> >> >>
> >> >>
GCC 8 complains:
net/core/pktgen.c: In function ‘pktgen_if_write’:
net/core/pktgen.c:1419:4: warning: ‘strncpy’ output may be truncated copying
between 0 and 31 bytes from a string of length 127 [-Wstringop-truncation]
strncpy(pkt_dev->src_max, buf, len);
On 2018-07-17 11:32, Michal Soltys wrote:
> On 07/17/2018 01:53 AM, Mahesh Bandewar (महेश बंडेवार) wrote:
>> On Mon, Jul 16, 2018 at 2:24 PM, Jay Vosburgh
>> wrote:
>>> Mahesh Bandewar wrote:
>>>
From: Mahesh Bandewar
Commit b89f04c61efe ("bonding: deliver link-local packets with
Implement loading of .BTF section from object file and build up
internal table for retrieving key/value id related to maps in
the BPF program. Latter is done by setting up struct btf_type
table.
One of the issues is that there's a disconnect between the data
types used in the map and struct
Implement missing bpf to bpf calls support. The loader will
recognize .text section and handle relocation entries that
are emitted by LLVM.
First step is processing of map related relocation entries
for .text section, and in a second step loader will copy .text
section into program section and
For slow processors using bit-banging MDIO, 20ms can be too short a
timeout when waiting for the transmit timestamp to become
available. Double it to 40ms.
Signed-off-by: Andrew Lunn
---
drivers/net/dsa/mv88e6xxx/hwtstamp.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
The 6165 family supports a more restricted version of hardware time
stamps. Only L2 PTP is supported. All ports have to use the same
EtherType, and transport spec configuration. PTP can only be
enabled/disabled globally, not per port.
Signed-off-by: Andrew Lunn
---
Do not bail out when AF_ALG is not supported by the kernel and
only do so when a map is requested in object ns where we're
calculating the hash. Otherwise, the loader can operate just
fine, therefore lets not fail early when it's not needed.
Signed-off-by: Daniel Borkmann
---
lib/bpf.c | 74
No need to spam the user with this if it can be fixed gracefully
anyway. Therefore, move it under verbose option.
Signed-off-by: Daniel Borkmann
---
lib/bpf.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/lib/bpf.c b/lib/bpf.c
index 4e26c0d..42093db 100644
---
Main part of this set is to: i) avoid strict af_alg kernel dependency,
ii) add loader support for bpf to bpf calls and iii) add btf loader
support with an option to annotate maps. For details please see the
individual patches. Thanks!
Daniel Borkmann (5):
bpf: import btf uapi kernel header
Needed later on for BTF loader.
Signed-off-by: Daniel Borkmann
---
include/uapi/linux/btf.h | 113 +++
1 file changed, 113 insertions(+)
create mode 100644 include/uapi/linux/btf.h
diff --git a/include/uapi/linux/btf.h b/include/uapi/linux/btf.h
new
Now that all the basic refactoring has been done we can add phylib
support. This patch series was successfully tested on:
RTL8168h
RTL8168evl
RTL8169sb
Changes in v2:
- return error in mdio ops if phyaddr > 0
- advertise pause modes
- added reviewed-by for several patches
Changes in v3:
- return
The 6165 family does not have per port PTP control registers. Also, it
places the timestamp data in different registers. Abstract the current
implementation of 6352 compatible PTP devices so that 6165 can be
added.
Signed-off-by: Andrew Lunn
---
drivers/net/dsa/mv88e6xxx/chip.h | 5 +++
The mv88e6165 family has its global clock in the PTP global
registers. It does not support any form of PTP events. Add a function
to read the clock, fill in an ops structure, and register it with the
two members of the family.
Signed-off-by: Andrew Lunn
---
drivers/net/dsa/mv88e6xxx/chip.c | 4
The MV88E6165 PTP registers are all in AVB bank F, unlike newer
generations which spread them over AVB bank E and F. Implement AVB ops
for the MV88E6165 which hides this difference.
Signed-off-by: Andrew Lunn
---
drivers/net/dsa/mv88e6xxx/chip.c| 2 ++
We can remove rtl8169_set_speed_xmii() now that phylib handles all this.
Signed-off-by: Heiner Kallweit
---
drivers/net/ethernet/realtek/r8169.c | 90
1 file changed, 90 deletions(-)
diff --git a/drivers/net/ethernet/realtek/r8169.c
The only remaining usage of the struct mii_if_info member is to store the
information whether the chip is GMII-capable. So we can replace it with
a simple flag.
Signed-off-by: Heiner Kallweit
Reviewed-by: Andrew Lunn
---
drivers/net/ethernet/realtek/r8169.c | 38 +---
1
Switch to using phy_ethtool_nway_reset().
Signed-off-by: Heiner Kallweit
Reviewed-by: Andrew Lunn
Reviewed-by: Florian Fainelli
---
drivers/net/ethernet/realtek/Kconfig | 1 -
drivers/net/ethernet/realtek/r8169.c | 9 +
2 files changed, 1 insertion(+), 9 deletions(-)
diff --git
Use genphy_soft_reset() instead of open-coding a PHY soft reset. We have
to do an explicit PHY soft reset because some chips use the genphy driver
which uses a no-op as soft_reset callback.
Signed-off-by: Heiner Kallweit
Reviewed-by: Andrew Lunn
Reviewed-by: Florian Fainelli
---
Instead of accessing the PHYstatus register we can use the information
phylib stores in the phy_device structure.
Signed-off-by: Heiner Kallweit
Reviewed-by: Florian Fainelli
---
drivers/net/ethernet/realtek/r8169.c | 9 +
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git
Add basic phylib support to r8169. All now unneeded old PHY handling code
will be removed in subsequent patches.
Signed-off-by: Heiner Kallweit
---
v2:
- return error in mdio ops if phyaddr > 0
- advertise pause modes
v3:
- return ENODEV in mdio ops if phyaddr > 0
- consider other minor review
Use phy_resume() / phy_suspend() instead of open coding this functionality.
The chip version specific differences are handled by the respective PHY
drivers.
The call to r8168_phy_power_down() in r8168_pll_power_down() can be
removed because phylib takes care now. The relevant scenarios are:
-
Use phy_ethtool_(g|s)et_link_ksettings() for the respective ethtool_ops
callbacks.
Signed-off-by: Heiner Kallweit
Reviewed-by: Florian Fainelli
---
drivers/net/ethernet/realtek/r8169.c | 35 +++-
1 file changed, 3 insertions(+), 32 deletions(-)
diff --git
Use new phylib functions phy_speed_down() and phy_speed_up().
Signed-off-by: Heiner Kallweit
---
drivers/net/ethernet/realtek/r8169.c | 29
1 file changed, 4 insertions(+), 25 deletions(-)
diff --git a/drivers/net/ethernet/realtek/r8169.c
Switch to using phy_mii_ioctl().
Signed-off-by: Heiner Kallweit
Reviewed-by: Andrew Lunn
Reviewed-by: Florian Fainelli
---
drivers/net/ethernet/realtek/r8169.c | 25 +++--
1 file changed, 3 insertions(+), 22 deletions(-)
diff --git a/drivers/net/ethernet/realtek/r8169.c
In order to remove performance impact of having the extra u32 in every
single flowi, this change removes the flowi_xfrm struct, prefering to
take the if_id as a method parameter where needed.
In the inbound direction, if_id is only needed during the
__xfrm_check_policy() function, and the if_id
The mv88e6165 family supports PTP, but its registers use a different
layout to the currently supported devices. Abstract accessing the PTP
registers into a set of ops, so making space for a second
implementation.
Signed-off-by: Andrew Lunn
---
drivers/net/dsa/mv88e6xxx/chip.c | 11 +++
The mv88e6165 family of switches supports PTP. It is however not fully
compatible with the current PTP support in the mv88e6xxx driver. This
patchset adds a level of abstraction to the PTP code, and then adds
the code needed to support the mv88e6165 family.
Andrew Lunn (8):
net: dsa: mv88e6xxx:
For the 6352 and newer switches, the PTP Ethertype defaults to
ETH_P_1588. Hence it was not explicitly set. The 6165 however defaults
to 0. So explicitly set the EtherType.
Signed-off-by: Andrew Lunn
---
drivers/net/dsa/mv88e6xxx/hwtstamp.c | 5 +
drivers/net/dsa/mv88e6xxx/ptp.h | 3
The 6165 only supports layer L2 PTP, where as the more modern devices
also support UDP and UDPv6, i.e. L4. Abstract the supported receive
filters.
Signed-off-by: Andrew Lunn
---
drivers/net/dsa/mv88e6xxx/chip.h | 1 +
drivers/net/dsa/mv88e6xxx/hwtstamp.c | 26 ++
On Wed, 18 Jul 2018 01:31:22 +0200, Daniel Borkmann wrote:
> # bpftool map dump id 386
>[{
> "key": 0,
> "value": {
> "": {
> "value": 0,
> "ifindex": 0,
> "mac": []
> }
> }
> },{
>
From: Heiner Kallweit
Date: Tue, 17 Jul 2018 22:42:40 +0200
> Now that all the basic refactoring has been done we can add phylib
> support. This patch series was successfully tested on:
> RTL8168h
> RTL8168evl
> RTL8169sb
>
> Changes in v2:
> - return error in mdio ops if phyaddr > 0
> -
By giving each register its own liveness chain, we elide the skip_callee()
logic. Instead, each register's parent is the state it inherits from;
both check_func_call() and prepare_func_exit() automatically connect
reg states to the correct chain since when they copy the reg state across
(r1-r5
From: Heiner Kallweit
Date: Tue, 17 Jul 2018 21:21:37 +0200
> The removed code would be called in two situations:
> 1. interface is brought up never or >10s after driver load
> 2. after close()
>
> Case 1 we can handle cleaner by ensuring chip is powered down when
> leaving probe(). open()
On 7/16/18 4:19 PM, Roopa Prabhu wrote:
> From: Roopa Prabhu
>
> NUD_NOARP entries are filtered out by default by iproute2.
> We dont want NUD_NOARP with NTF_EXT_LEARNED flag filtered out.
> This patch extends the default filter check for ip neigh show
> to include the NTF_EXT_LEARNED flag.
>
>
On 7/14/18 1:10 AM, Qiaobin Fu wrote:
> @@ -156,6 +162,9 @@ parse_skbedit(struct action_util *a, int *argc_p, char
> ***argv_p, int tca_id,
> if (flags & SKBEDIT_F_PTYPE)
> addattr_l(n, MAX_MSG, TCA_SKBEDIT_PTYPE,
> , sizeof(ptype));
> + if
From: Andrew Lunn
Date: Tue, 17 Jul 2018 23:48:53 +0200
> @@ -319,6 +337,8 @@ int mv88e6xxx_ptp_setup(struct mv88e6xxx_chip *chip)
> {
> int i;
>
> + const struct mv88e6xxx_ptp_ops *ptp_ops = chip->info->ops->ptp_ops;
> +
Please keep the local variables together here.
Otherwise,
On Tue, Jul 17, 2018 at 11:48:52PM +0200, Andrew Lunn wrote:
> The mv88e6165 family of switches supports PTP. It is however not fully
> compatible with the current PTP support in the mv88e6xxx driver. This
> patchset adds a level of abstraction to the PTP code, and then adds
> the code needed to
On Tue, Jul 17, 2018 at 10:53:25AM -0700, Jakub Kicinski wrote:
> Create a higher-level entity to represent a device/ASIC to allow
> programs and maps to be shared between device ports. The extra
> work is required to make sure we don't destroy BPF objects as
> soon as the netdev for which they
From: Florian Fainelli
Date: Tue, 17 Jul 2018 08:36:39 -0700
> @@ -755,7 +755,8 @@ static int bcm_sf2_cfp_rule_set(struct dsa_switch *ds,
> int port,
> port_num = fs->ring_cookie / SF2_NUM_EGRESS_QUEUES;
>
> if (fs->ring_cookie == RX_CLS_FLOW_DISC ||
> -
/net/ipv4/tcp.c:2011 tcp_recvmsg+0x694/0xba0
Suggested-by: 積丹尼 Dan Jacobson
Signed-off-by: Randy Dunlap
---
v2: drop __func__ because it duplicates part of the error message.
net/ipv4/tcp.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
--- linux-next-20180717.orig/net/ipv4/tcp.c
+++
1 - 100 of 122 matches
Mail list logo
