The helper functions like bpf_map_lookup_elem(map, key) were only
allowing 'key' to point to the initialized stack area.
That is causing performance degradation when programs need to process
millions of packets per second and need to copy contents of the packet
into the stack just to pass the stack pointer into the lookup() function.
Allow such helpers read from the packet directly.
All helpers that expect ARG_PTR_TO_MAP_KEY, ARG_PTR_TO_MAP_VALUE,
ARG_PTR_TO_STACK assume byte aligned pointer, so no alignment concerns,
only need to check that helper will not be accessing beyond
the packet range verified by the prior 'if (ptr < data_end)' condition.
For now allow this feature for XDP programs only. Later it can be
relaxed for the clsact programs as well.
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fixes the following sparse warning:

net/ipv6/sit.c:1129:6: warning:
 symbol 'ipip6_valid_ip_proto' was not declared. Should it be static?
Signed-off-by: NWei Yongjun <weiyj.lk@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Sargun Dhillon says:

====================
Add test_current_task_under_cgroup bpf helper and test

This patchset includes a helper and an example to determine whether the probe is
currently executing in the context of a specific cgroup based on a cgroup bpf
map / array. The helper checks the cgroupsv2 hierarchy based on the handle in
the map and if the current cgroup is equal to it, or a descendant of it. The
helper was tested with the example program, and it was verified that the correct
behaviour occurs in the interrupt context.

In an earlier version of this patchset I had added an "opensnoop"-like tool, and
I realized I was basically reimplementing a lot of the code that already exists
in the bcc repo. So, instead I decided to write a test that creates a new mount
namespace, mounts up the cgroupv2 hierarchy, and does some basic tests.  I used
the sync syscall as a canary for these tests because it's a simple, 0-arg
syscall. Once this patch is accepted, adding support to opensnoop will be easy.

I also added a task_under_cgroup_hierarchy function in cgroups.h, as this
pattern is used in a couple places. Converting those can be done in a later
patchset.

Thanks to Alexei, Tejun, and Daniel for providing review.

v1->v2: Clean up
v2->v3: Move around ifdefs out of *.c files, add an "integration" test
v3->v4: De-genercize arraymap fetching function;
	rename helper from in_cgroup to under_cgroup (makes much more sense)
	Split adding cgroups task_under_cgroup_hierarchy function
v4->v5: Fix formatting
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This test has a BPF program which writes the last known pid to call the
sync syscall within a given cgroup to a map.

The user mode program creates its own mount namespace, and mounts the
cgroupsv2  hierarchy in there, as on all current test systems
(Ubuntu 16.04, Debian), the cgroupsv2 vfs is unmounted by default.
Once it does this, it proceeds to test.

The test checks for positive and negative condition. It ensures that
when it's part of a given cgroup, its pid is captured in the map,
and that when it leaves the cgroup, this doesn't happen.

It populate a cgroups arraymap prior to execution in userspace. This means
that the program must be run in the same cgroups namespace as the programs
that are being traced.
Signed-off-by: NSargun Dhillon <sargun@sargun.me>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Tejun Heo <tj@kernel.org>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This adds a bpf helper that's similar to the skb_in_cgroup helper to check
whether the probe is currently executing in the context of a specific
subset of the cgroupsv2 hierarchy. It does this based on membership test
for a cgroup arraymap. It is invalid to call this in an interrupt, and
it'll return an error. The helper is primarily to be used in debugging
activities for containers, where you may have multiple programs running in
a given top-level "container".
Signed-off-by: NSargun Dhillon <sargun@sargun.me>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Tejun Heo <tj@kernel.org>
Acked-by: NTejun Heo <tj@kernel.org>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This commit adds an inline function to cgroup.h to check whether a given
task is under a given cgroup hierarchy. This is to avoid having to put
ifdefs in .c files to gate access to cgroups. When cgroups are disabled
this always returns true.
Signed-off-by: NSargun Dhillon <sargun@sargun.me>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Tejun Heo <tj@kernel.org>
Acked-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Simon Wunderlich says:

====================
This feature patchset includes the following changes (mostly
chronological order):

 - bump version strings, by Simon Wunderlich

 - kerneldoc clean up, by Sven Eckelmann

 - enable RTNL automatic loading and according documentation
   changes, by Sven Eckelmann (2 patches)

 - fix/improve interface removal and associated locking, by
   Sven Eckelmann (3 patches)

 - clean up unused variables, by Linus Luessing

 - implement Gateway selection code for B.A.T.M.A.N. V by
   Antonio Quartulli (4 patches)

 - rewrite TQ comparison by Markus Pargmann

 - fix Cocinelle warnings on bool vs integers (by Fenguang Wu/Intels
   kbuild test robot) and bitwise arithmetic operations (by Linus
   Luessing)

 - rewrite packet creation for forwarding for readability and to avoid
   reference count mistakes, by Linus Luessing

 - use kmem_cache for translation table, which results in more efficient
   storing of translation table entries, by Sven Eckelmann

 - rewrite/clarify reference handling for send_skb_unicast, by Sven
   Eckelmann

 - fix debug messages when updating routes, by Sven Eckelmann
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Bert Kenward says:

====================
sfc: SFN8000 support improvements

This series improves support for the recently released SFN8000 series
of adapters. Specifically, it retrieves interrupt moderation timer
settings directly from the adapter and uses those settings. It also
uses a new event queue initialisation interface, allowing specification
of a performance objective rather than enabling individual flags.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

On SFN8000 series adapters the MC provides a method to get the timer
quantum and the maximum timer setting. We revert to the old values if the
new call is unavailable.
Signed-off-by: NBert Kenward <bkenward@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

SFN8000-series NICs require a new method of setting interrupt moderation,
via MCDI. This is indicated by a workaround flag. This new MCDI command
takes an explicit time value rather than a number of ticks. It therefore
makes sense to also store the moderation values in terms of time, since
that is what the ethtool interface is interested in.
Signed-off-by: NBert Kenward <bkenward@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Rather than explicitly specifying flags we can now specify a desired
performance target to the firmware, ie higher throughput or lower latency.
For now we use the default "auto" configuration.
Signed-off-by: NBert Kenward <bkenward@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NBert Kenward <bkenward@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NBert Kenward <bkenward@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NBert Kenward <bkenward@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since these critical sections protected by page_lock are all entered
from the user context or bottom half context, they can be replaced
with the spin_lock() or spin_lock_bh instead of spin_lock_irqsave().
Signed-off-by: NSean Wang <sean.wang@mediatek.com>
Acked-by: NJohn Crispin <john@phrozen.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This is a driver for the ENA family of networking devices.
Signed-off-by: NNetanel Belgazal <netanel@annapurnalabs.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Kedareswara rao Appana says:

====================
net: phy: Add xilinx gmiitorgmii converter support

The Gigabit Media Independent Interface (GMII) to Reduced Gigabit Media
Independent Interface (RGMII) core provides the RGMII between RGMII-compliant
Ethernet physical media devices (PHY) and the Gigabit Ethernet controller.
This core can be used in all three modes of operation(10/100/1000 Mb/s).
The Management Data Input/Output (MDIO) interface is used to configure the
Speed of operation. This core can switch dynamically between the three
Different speed modes by configuring the conveter register through mdio write.

The conveter sits b/w the MAC and external phy like below

MACB <==> GMII2RGMII <==> RGMII_PHY

        MDIO    <========> GMII2RGMII
MCAB <=======>
                <========> RGMII

Using MAC MDIO bus we can access both the converter and the external PHY.
We need to program the line speed of the converter during run time based
On the external phy negotiated speed.

This patch series does the below
---> Add mask for Control register 10Mbps speed.
---> Add support for xilinx gmiitorgmii converter.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds support for gmiitorgmii converter.

The GMII to RGMII IP core provides the Reduced Gigabit Media
Independent Interface (RGMII) between Ethernet physical media
Devices and the Gigabit Ethernet controller. This core can
Switch dynamically between the three different speed modes of
Operation by configuring the converter register through mdio write.

MDIO interface is used to set operating speed of Ethernet MAC.

This converter sits between the MAC and the external phy
MAC <==> GMII2RGMII <==> RGMII_PHY
Signed-off-by: NKedareswara rao Appana <appanad@xilinx.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Device-tree binding documentation for xilinx gmiitorgmii converter.
Signed-off-by: NKedareswara rao Appana <appanad@xilinx.com>
Acked-by: NRob Herring <robh@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds mask for the Control register
10Mbps speed.
Reviewed-by: NFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NKedareswara rao Appana <appanad@xilinx.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The ethtool api {get|set}_settings is deprecated.
We move this driver to new api {get|set}_link_ksettings.
Signed-off-by: NPhilippe Reynes <tremyfr@gmail.com>
Tested-by: NSimon Horman <horms+renesas@verge.net.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The private structure contain a pointer to phydev, but the structure
net_device already contain such pointer. So we can remove the pointer
phy_dev in the private structure, and update the driver to use the
one contained in struct net_device.
Signed-off-by: NPhilippe Reynes <tremyfr@gmail.com>
Tested-by: NSimon Horman <horms+renesas@verge.net.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The commit 555c8a86 ("bpf: avoid stack copy and use skb ctx for event output")
started using 20 of initially reserved upper 32-bits of 'flags' argument
in bpf_perf_event_output(). Adjust corresponding prototype in samples/bpf/bpf_helpers.h
Signed-off-by: NAdam Barth <arb@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds support for 64 bit addressing and BDs.
-> Enable 64 bit addressing in DMACFG register.
-> Set DMA mask when design config register shows support for 64 bit addr.
-> Add new BD words for higher address when 64 bit DMA support is present.
-> Add and update TBQPH and RBQPH for MSB of BD pointers.
-> Change extraction and updation of buffer addresses to use
64 bit address.
-> In gem_rx extract address in one place insted of two and use a
separate flag for RXUSED.
Signed-off-by: NHarini Katakam <harinik@xilinx.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds the driver implementation for ethtool link_ksettings
callbacks. qed driver now defines/uses the qed specific masks for
representing link capability values. qede driver maps these values to
to new link modes defined by the kernel implementation of link_ksettings.

Please consider applying this to 'net-next' branch.
Signed-off-by: NSudarsana Reddy Kalluru <sudarsana.kalluru@qlogic.com>
Signed-off-by: NYuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Ivan Khoronzhuk says:

====================
net: ethernet: ti: cpsw: split driver data and per ndev data

In dual_emac mode the driver can handle 2 network devices. Each of them can use
its own private data and common data/resources. This patchset splits common driver
data/resources and private per net device data.
It leads to:
- reduce memory usage
- increase code readability
- allows add a bunch of simplification
- create prerequisites to add multi-channel support,
  when channels are shared between net devices

Doesn't have bad impact on performance.
v2: https://lkml.org/lkml/2016/8/6/108

Since v2:
- removed patch:
  net: ethernet: ti: cpsw: fix int dbg message
- replaced patch:
  "net: ethernet: ti: cpsw: remove redundant check in napi poll"
  on "net: ethernet: ti: cpsw: remove intr dbg msg from poll handlers"
- removed macro "cpsw_get_slave_ndev"
- corrected some commits

Since v1:
- added several patch improvements
- avoided variable reordering in structures
- removed static variable for common function
- split big patch on several patches:
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The ale, cpts, version, rx_packet_max, bus_freq, interrupt pacing
parameters are common per net device that uses the same h/w. So,
move them to common driver structure.
Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Reviewed-by: NMugunthan V N <mugunthanvnm@ti.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The napi structs are common for both net devices in dual_emac
mode, In order to not hold duplicate links to them, move to
cpsw_common.
Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Reviewed-by: NMugunthan V N <mugunthanvnm@ti.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

These data are common for net devs in dual_emac mode. No need to hold
it for every priv instance, so move them under cpsw_common.
Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Reviewed-by: NMugunthan V N <mugunthanvnm@ti.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The irq data are common for net devs in dual_emac mode. So no need to
hold these data in every priv struct, move them under cpsw_common.
Also delete irq_num var, as after optimization it's not needed.
Correct number of irqs to 2, as anyway, driver is using only 2,
at least for now.
Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Reviewed-by: NMugunthan V N <mugunthanvnm@ti.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Every net device private struct holds links to shared cpdma resources.
No need to save and every time synchronize these resources per net dev.
So, move it to common driver struct.
Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Reviewed-by: NMugunthan V N <mugunthanvnm@ti.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The pointers on h/w registers are common for every cpsw_private
instance, so no need to hold them for every ndev.
Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Reviewed-by: NMugunthan V N <mugunthanvnm@ti.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

No need to hold pdev link when only dev is needed.
This allows to simplify a bunch of cpsw->pdev->dev now and farther.
Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Reviewed-by: NMugunthan V N <mugunthanvnm@ti.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch simply create holder for common data and as a start moves
pdev var to it.
Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Reviewed-by: NMugunthan V N <mugunthanvnm@ti.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

No need to check const slave num in runtime for every packet,
and ndev for slaves w/o ndev is anyway NULL. So remove redundant
check and macro.
Reviewed-by: NMugunthan V N <mugunthanvnm@ti.com>
Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There is no need to hold link to clk, it's used only once
while probe.
Reviewed-by: NMugunthan V N <mugunthanvnm@ti.com>
Reviewed-by: NGrygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There is no need in priv here.
Reviewed-by: NMugunthan V N <mugunthanvnm@ti.com>
Reviewed-by: NGrygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

At poll handler no possibility to figure out which network device is
handling packets, as cpdma channels are common for both network
devices in dual_emac mode. Currently, the messages are printed only
for one device, in fact, there is two. This print msg is incorrect
and seems is not very useful, so drop it from poll handler.
Reviewed-by: NMugunthan V N <mugunthanvnm@ti.com>
Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As second net dev is created only in case of dual_emac mode, port
number can be figured out in simpler way. Also no need to pass
redundant ndev struct.
Reviewed-by: NMugunthan V N <mugunthanvnm@ti.com>
Reviewed-by: NGrygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The PPTP is encapsulated by GRE header with that GRE_VERSION bits
must contain one. But current GRE RPS needs the GRE_VERSION must be
zero. So RPS does not work for PPTP traffic.

In my test environment, there are four MIPS cores, and all traffic
are passed through by PPTP. As a result, only one core is 100% busy
while other three cores are very idle. After this patch, the usage
of four cores are balanced well.
Signed-off-by: NGao Feng <fgao@ikuai8.com>
Reviewed-by: NPhilip Prindeville <philipp@redfish-solutions.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>