Add helpers to get the policy's signed/unsigned range
validation data.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use a validation type instead, so we can later expose
the NLA_* values to userspace for policy descriptions.
Some transformations were done with this spatch:
@@
identifier p;
expression X, L, A;
@@
struct nla_policy p[X] = {
[A] =
-{ .type = NLA_EXACT_LEN_WARN, .len = L },
+NLA_POLICY_EXACT_LEN_WARN(L),
...
};
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since NLA_MSECS is really equivalent to NLA_U64, allow
it to have range validation as well.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Using a pointer to a struct indicating the min/max values,
extend the ability to do range validation for arbitrary
values. Small values in the s16 range can be kept in the
policy directly.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the netlink policy, we currently have a void *validation_data
that's pointing to different things:
* a u32 value for bitfield32,
* the netlink policy for nested/nested array
* the string for NLA_REJECT
Remove the pointer and place appropriate type-safe items in the
union instead.
While at it, completely dissolve the pointer for the bitfield32
case and just put the value there directly.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a sysctl to control hrtimer slack, default of 100 usec.
This gives the opportunity to reduce system overhead,
and help very short RTT flows.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In commit 86de5921a3 ("tcp: defer SACK compression after DupThresh")
I added a TCP_FASTRETRANS_THRESH bias to tp->compressed_ack in order
to enable sack compression only after 3 dupacks.
Since we plan to relax this rule for flows that involve
stacks not requiring this old rule, this patch adds
a distinct tp->dup_ack_counter.
This means the TCP_FASTRETRANS_THRESH value is now used
in a single location that a future patch can adjust:
if (tp->dup_ack_counter < TCP_FASTRETRANS_THRESH) {
tp->dup_ack_counter++;
goto send_now;
}
This patch also introduces tcp_sack_compress_send_ack()
helper to ease following patch comprehension.
This patch refines LINUX_MIB_TCPACKCOMPRESSED to not
count the acks that we had to send if the timer expires
or tcp_sack_compress_send_ack() is sending an ack.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Saeed Mahameed says:
====================
mlx5-updates-2020-04-30
1) Add release all pages support, From Eran.
to release all FW pages at once on driver unload, when supported by FW.
2) From Maxim and Tariq, Trivial Data path cleanup and code improvements
in preparation for their next features, TLS offload and TX performance
improvements
3) Multiple cleanups.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
- add SPDX header;
- add a document title;
- adjust titles and chapters, adding proper markups;
- mark code blocks and literals as such;
- adjust identation, whitespaces and blank lines where needed;
- add to networking/index.rst.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
- add SPDX header;
- adjust title markup;
- mark code blocks and literals as such;
- adjust identation, whitespaces and blank lines where needed;
- add to networking/index.rst.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Not much to be done here:
- add SPDX header;
- adjust titles and chapters, adding proper markups;
- add to networking/index.rst.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds ability to filter sockets based on cgroup v2 ID.
Such filter is helpful in ss utility for filtering sockets by
cgroup pathname.
Signed-off-by: Dmitry Yakunin <zeil@yandex-team.ru>
Reviewed-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds cgroup v2 ID to common inet diag message attributes.
Cgroup v2 ID is kernfs ID (ino or ino+gen). This attribute allows filter
inet diag output by cgroup ID obtained by name_to_handle_at() syscall.
When net_cls or net_prio cgroup is activated this ID is equal to 1 (root
cgroup ID) for newly created sockets.
Some notes about this ID:
1) gets initialized in socket() syscall
2) incoming socket gets ID from listening socket
(not during accept() syscall)
3) not changed when process get moved to another cgroup
4) can point to deleted cgroup (refcounting)
v2:
- use CONFIG_SOCK_CGROUP_DATA instead if CONFIG_CGROUPS
v3:
- fix attr size by using nla_total_size_64bit() (Eric Dumazet)
- more detailed commit message (Konstantin Khlebnikov)
Signed-off-by: Dmitry Yakunin <zeil@yandex-team.ru>
Reviewed-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Acked-By: Tejun Heo <tj@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
mlx5 updates for both net-next and rdma-next:
1) HW bits and definitions for TLS and IPsec offlaods
2) Release all pages capability bits
3) New command interface helpers and some code cleanup as a result
4) Move qp.c out of mlx5 core driver into mlx5_ib rdma driver
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Pablo Neira Ayuso says:
====================
Netfilter updates for net-next
The following patchset contains Netfilter updates for nf-next:
1) Add IPS_HW_OFFLOAD status bit, from Bodong Wang.
2) Remove 128-bit limit on the set element data area, rise it
to 64 bytes.
3) Report EOPNOTSUPP for unsupported NAT types and flags.
4) Set up nft_nat flags from the control plane path.
5) Add helper functions to set up the nf_nat_range2 structure.
6) Add netmap support for nft_nat.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Current route nexthop API maintains user space compatibility
with old route API by default. Dumps and netlink notifications
support both new and old API format. In systems which have
moved to the new API, this compatibility mode cancels some
of the performance benefits provided by the new nexthop API.
This patch adds new sysctl nexthop_compat_mode which is on
by default but provides the ability to turn off compatibility
mode allowing systems to run entirely with the new routing
API. Old route API behaviour and support is not modified by this
sysctl.
Uses a single sysctl to cover both ipv4 and ipv6 following
other sysctls. Covers dumps and delete notifications as
suggested by David Ahern.
Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add new TX WQE field for Connect-X6DX trailer insertion support,
when set, the HW adds a trailer to the packet, the WQE trailer
association flags are used to set to HW the header which the
trailer belongs.
Signed-off-by: Raed Salem <raeds@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Add a bit in HCA capabilities layout to indicate if release all pages is
supported.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Reviewed-by: Moshe Shemesh <moshe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Add needed structure layouts and defines for pci sync for fw update
event. The downstream patches will include event handlers for this event
type.
Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Add needed structure layouts and defines for MFRL (Management Firmware
Reset Level) register. This structure will be used for the firmware
upgrade and reset flow in the downstream patches.
Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
The imm_inval_pkey field can hold four different types of data,
depends on the usage, the data could be one of the below:
- Immediate field of the received message
- Invalidate rkey
- Pkey of the packet
- Flow table metadata
Current implementation doesn't reflect the intended usage of the
field at usage time.
Reflect the different types by replace this field with a union,
modify code where this field is used to reflect its intended
usage.
Signed-off-by: Raed Salem <raeds@mellanox.com>
Reviewed-by: Huy Nguyen <huyn@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
The alignment value is part of the input structure, so use it and spare
extra memory allocation when is not needed.
Now, using the new ability when allocating icm for Direct-Rule
insertion.
Signed-off-by: Ariel Levkovich <lariel@mellanox.com>
Signed-off-by: Erez Shitrit <erezsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Add COPY type to modify_header action. IPsec feature is the first
feature that needs COPY steering action.
Signed-off-by: Huy Nguyen <huyn@mellanox.com>
Signed-off-by: Raed Salem <raeds@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Acked-by: Leon Romanovsky <leonro@mellanox.com>
This patch allows you to NAT the network address prefix onto another
network address prefix, a.k.a. netmapping.
Userspace must specify the NF_NAT_RANGE_NETMAP flag and the prefix
address through the NFTA_NAT_REG_ADDR_MIN and NFTA_NAT_REG_ADDR_MAX
netlink attributes.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Extend switchdev API to add support for MRP. The HW is notified in
following cases:
SWITCHDEV_OBJ_ID_MRP: This is used when a MRP instance is added/removed
from the MRP ring.
SWITCHDEV_OBJ_ID_RING_ROLE_MRP: This is used when the role of the node
changes. The current supported roles are MRM and MRC.
SWITCHDEV_OBJ_ID_RING_TEST_MRP: This is used when to start/stop sending
MRP_Test frames on the mrp ring ports. This is called only on nodes that have
the role MRM. In case this fails then the SW will generate the frames.
SWITCHDEV_OBJ_ID_RING_STATE_STATE: This is used when the ring changes it states
to open or closed. This is required to notify HW because the MRP_Test frame
contains the field MRP_InState which contains this information.
SWITCHDEV_ATTR_ID_MRP_PORT_STATE: This is used when the port's state is
changed. It can be in blocking/forwarding mode.
SWITCHDEV_ATTR_ID_MRP_PORT_ROLE: This is used when port's role changes. The
roles of the port can be primary/secondary. This is required to notify HW
because the MRP_Test frame contains the field MRP_PortRole that contains this
information.
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds a new port attribute, IFLA_BRPORT_MRP_RING_OPEN, which allows
to notify the userspace when the port lost the continuite of MRP frames.
This attribute is set by kernel whenever the SW or HW detects that the ring is
being open or closed.
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
To integrate MRP into the bridge, first the bridge needs to be aware of ports
that are part of an MRP ring and which rings are on the bridge.
Therefore extend bridge interface with the following:
- add new flag(BR_MPP_AWARE) to the net bridge ports, this bit will be
set when the port is added to an MRP instance. In this way it knows if
the frame was received on MRP ring port
- add new flag(BR_MRP_LOST_CONT) to the net bridge ports, this bit will be set
when the port lost the continuity of MRP Test frames.
- add a list of MRP instances
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add new nested netlink attribute to configure the MRP. These attributes are used
by the userspace to add/delete/configure MRP instances and by the kernel to
notify the userspace when the MRP ring gets open/closed. MRP nested attribute
has the following attributes:
IFLA_BRIDGE_MRP_INSTANCE - the parameter type is br_mrp_instance which contains
the instance id, and the ifindex of the two ports. The ports can't be part of
multiple instances. This is used to create/delete MRP instances.
IFLA_BRIDGE_MRP_PORT_STATE - the parameter type is u32. Which can be forwarding,
blocking or disabled.
IFLA_BRIDGE_MRP_PORT_ROLE - the parameter type is br_mrp_port_role which
contains the instance id and the role. The role can be primary or secondary.
IFLA_BRIDGE_MRP_RING_STATE - the parameter type is br_mrp_ring_state which
contains the instance id and the state. The state can be open or closed.
IFLA_BRIDGE_MRP_RING_ROLE - the parameter type is br_mrp_ring_role which
contains the instance id and the ring role. The role can be MRM or MRC.
IFLA_BRIDGE_MRP_START_TEST - the parameter type is br_mrp_start_test which
contains the instance id, the interval at which to send the MRP_Test frames,
how many test frames can be missed before declaring the ring open and the
period which represent for how long to send the test frames.
Also add the file include/uapi/linux/mrp_bridge.h which defines all the types
used by MRP that are also needed by the userpace.
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This bit indicates that the conntrack entry is offloaded to hardware
flow table. nf_conntrack entry will be tagged with [HW_OFFLOAD] if
it's offload to hardware.
cat /proc/net/nf_conntrack
ipv4 2 tcp 6 \
src=1.1.1.17 dst=1.1.1.16 sport=56394 dport=5001 \
src=1.1.1.16 dst=1.1.1.17 sport=5001 dport=56394 [HW_OFFLOAD] \
mark=0 zone=0 use=3
Note that HW_OFFLOAD/OFFLOAD/ASSURED are mutually exclusive.
Changelog:
* V1->V2:
- Remove check of lastused from stats. It was meant for cases such
as removing driver module while traffic still running. Better to
handle such cases from garbage collector.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Paul Blakey <paulb@mellanox.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
In MPTCP, the receive window is shared across all subflows, because it
refers to the mptcp-level sequence space.
MPTCP receivers already place incoming packets on the mptcp socket
receive queue and will charge it to the mptcp socket rcvbuf until
userspace consumes the data.
Update __tcp_select_window to use the occupancy of the parent/mptcp
socket instead of the subflow socket in case the tcp socket is part
of a logical mptcp connection.
This commit doesn't change choice of initial window for passive or active
connections.
While it would be possible to change those as well, this adds complexity
(especially when handling MP_JOIN requests). Furthermore, the MPTCP RFC
specifically says that a MPTCP sender 'MUST NOT use the RCV.WND field
of a TCP segment at the connection level if it does not also carry a DSS
option with a Data ACK field.'
SYN/SYNACK packets do not carry a DSS option with a Data ACK field.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull timer fixlet from Ingo Molnar:
"A single fix for a comment that may show up in DocBook output"
* tag 'timers-urgent-2020-04-25' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
vdso/datapage: Use correct clock mode name in comment
Pull networking fixes from David Miller:
1) Fix memory leak in netfilter flowtable, from Roi Dayan.
2) Ref-count leaks in netrom and tipc, from Xiyu Yang.
3) Fix warning when mptcp socket is never accepted before close, from
Florian Westphal.
4) Missed locking in ovs_ct_exit(), from Tonghao Zhang.
5) Fix large delays during PTP synchornization in cxgb4, from Rahul
Lakkireddy.
6) team_mode_get() can hang, from Taehee Yoo.
7) Need to use kvzalloc() when allocating fw tracer in mlx5 driver,
from Niklas Schnelle.
8) Fix handling of bpf XADD on BTF memory, from Jann Horn.
9) Fix BPF_STX/BPF_B encoding in x86 bpf jit, from Luke Nelson.
10) Missing queue memory release in iwlwifi pcie code, from Johannes
Berg.
11) Fix NULL deref in macvlan device event, from Taehee Yoo.
12) Initialize lan87xx phy correctly, from Yuiko Oshino.
13) Fix looping between VRF and XFRM lookups, from David Ahern.
14) etf packet scheduler assumes all sockets are full sockets, which is
not necessarily true. From Eric Dumazet.
15) Fix mptcp data_fin handling in RX path, from Paolo Abeni.
16) fib_select_default() needs to handle nexthop objects, from David
Ahern.
17) Use GFP_ATOMIC under spinlock in mac80211_hwsim, from Wei Yongjun.
18) vxlan and geneve use wrong nlattr array, from Sabrina Dubroca.
19) Correct rx/tx stats in bcmgenet driver, from Doug Berger.
20) BPF_LDX zero-extension is encoded improperly in x86_32 bpf jit, fix
from Luke Nelson.
* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (100 commits)
selftests/bpf: Fix a couple of broken test_btf cases
tools/runqslower: Ensure own vmlinux.h is picked up first
bpf: Make bpf_link_fops static
bpftool: Respect the -d option in struct_ops cmd
selftests/bpf: Add test for freplace program with expected_attach_type
bpf: Propagate expected_attach_type when verifying freplace programs
bpf: Fix leak in LINK_UPDATE and enforce empty old_prog_fd
bpf, x86_32: Fix logic error in BPF_LDX zero-extension
bpf, x86_32: Fix clobbering of dst for BPF_JSET
bpf, x86_32: Fix incorrect encoding in BPF_LDX zero-extension
bpf: Fix reStructuredText markup
net: systemport: suppress warnings on failed Rx SKB allocations
net: bcmgenet: suppress warnings on failed Rx SKB allocations
macsec: avoid to set wrong mtu
mac80211: sta_info: Add lockdep condition for RCU list usage
mac80211: populate debugfs only after cfg80211 init
net: bcmgenet: correct per TX/RX ring statistics
net: meth: remove spurious copyright text
net: phy: bcm84881: clear settings on link down
chcr: Fix CPU hard lockup
...
Alexei Starovoitov says:
====================
pull-request: bpf 2020-04-24
The following pull-request contains BPF updates for your *net* tree.
We've added 17 non-merge commits during the last 5 day(s) which contain
a total of 19 files changed, 203 insertions(+), 85 deletions(-).
The main changes are:
1) link_update fix, from Andrii.
2) libbpf get_xdp_id fix, from David.
3) xadd verifier fix, from Jann.
4) x86-32 JIT fixes, from Luke and Wang.
5) test_btf fix, from Stanislav.
6) freplace verifier fix, from Toke.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Since 6e2d85ec05 ("net: phy: Stop with excessive soft reset")
we don't need genphy_no_soft_reset() any longer. Not setting
callback soft_reset results in a no-op now.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull PNP cleanup from Rafael Wysocki:
"Make the PNP code use list_for_each_entry() in a few places instead of
open-coding it (Jason Gunthorpe)"
* tag 'pnp-5.7-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
pnp: Use list_for_each_entry() instead of open coding
Pull block fixes from Jens Axboe:
"A few fixes/changes that should go into this release:
- null_blk zoned fixes (Damien)
- blkdev_close() sync improvement (Douglas)
- Fix regression in blk-iocost that impacted (at least) systemtap
(Waiman)
- Comment fix, header removal (Zhiqiang, Jianpeng)"
* tag 'block-5.7-2020-04-24' of git://git.kernel.dk/linux-block:
null_blk: Cleanup zoned device initialization
null_blk: Fix zoned command handling
block: remove unused header
blk-iocost: Fix error on iocost_ioc_vrate_adj
bdev: Reduce time holding bd_mutex in sync in blkdev_close()
buffer: remove useless comment and WB_REASON_FREE_MORE_MEM, reason.
Pull tracing fixes from Steven Rostedt:
"A few tracing fixes:
- Two fixes for memory leaks detected by kmemleak
- Removal of some dead code
- A few local functions turned static"
* tag 'trace-v5.7-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
tracing: Convert local functions in tracing_map.c to static
tracing: Remove DECLARE_TRACE_NOARGS
ftrace: Fix memory leak caused by not freeing entry in unregister_ftrace_direct()
tracing: Fix memory leaks in trace_events_hist.c
Pull Kbuild fixes from Masahiro Yamada:
- fix scripts/config to properly handle ':' in string type CONFIG
options
- fix unneeded rebuilds of DT schema check rule
- git rid of ordering dependency between <linux/vermagic.h> and
<linux/module.h> to fix build errors in some network drivers
- clean up generated headers of host arch with 'make ARCH=um mrproper'
* tag 'kbuild-fixes-v5.7' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
h8300: ignore vmlinux.lds
Documentation: kbuild: fix the section title format
um: ensure `make ARCH=um mrproper` removes arch/$(SUBARCH)/include/generated/
arch: split MODULE_ARCH_VERMAGIC definitions out to <asm/vermagic.h>
kbuild: fix DT binding schema rule again to avoid needless rebuilds
scripts/config: allow colons in option strings for sed
Pull sound fixes from Takashi Iwai:
"This became a slightly big pull request, as the accumulated ASoC fixes
are included here. Some highlights:
- Revert of ASoC DAI startup changes that caused regression on some
x86 platforms
- Regression fix in HD-audio power management and driver blacklist
- A collection of ASoC DAPM and topology fixes
- Continued USB-audio fixes and quirks
- Lots of small device-specific fixes
- Rockchip S/PDIF DT stuff update for validation issues"
* tag 'sound-5.7-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (51 commits)
ALSA: hda: Always use jackpoll helper for jack update after resume
ALSA: hda/realtek - Add new codec supported for ALC245
ALSA: usb-audio: Fix usb audio refcnt leak when getting spdif
ALSA: usb-audio: Add connector notifier delegation
ALSA: usb-audio: Apply async workaround for Scarlett 2i4 2nd gen
ASoC: wm8960: Fix wrong clock after suspend & resume
ALSA: usx2y: Fix potential NULL dereference
ALSA: usb-audio: Add quirk for Focusrite Scarlett 2i2
ASoC: wm89xx: Add missing dependency
ASoC: dapm: fixup dapm kcontrol widget
ASoC: rsnd: Fix "status check failed" spam for multi-SSI
ASoC: rsnd: Don't treat master SSI in multi SSI setup as parent
ASoC: meson: gx-card: fix codec-to-codec link setup
ASoC: meson: axg-card: fix codec-to-codec link setup
ALSA: usb-audio: Add static mapping table for ALC1220-VB-based mobos
ALSA: hda: Remove ASUS ROG Zenith from the blacklist
ALSA: hda/realtek - Fix unexpected init_amp override
ALSA: usb-audio: Filter out unsupported sample rates on Focusrite devices
ASoC: SOF: Intel: add min/max channels for SSP on Baytrail/Broadwell
ASoC: stm32: sai: fix sai probe
...
Back in commit 3b47d30396 ("net: gro: add a per device gro flush timer")
we added the ability to arm one high resolution timer, that we used
to keep not-complete packets in GRO engine a bit longer, hoping that further
frames might be added to them.
Since then, we added the napi_complete_done() interface, and commit
364b605573 ("net: busy-poll: return busypolling status to drivers")
allowed drivers to avoid re-arming NIC interrupts if we made a promise
that their NAPI poll() handler would be called in the near future.
This infrastructure can be leveraged, thanks to a new device parameter,
which allows to arm the napi hrtimer, instead of re-arming the device
hard IRQ.
We have noticed that on some servers with 32 RX queues or more, the chit-chat
between the NIC and the host caused by IRQ delivery and re-arming could hurt
throughput by ~20% on 100Gbit NIC.
In contrast, hrtimers are using local (percpu) resources and might have lower
cost.
The new tunable, named napi_defer_hard_irqs, is placed in the same hierarchy
than gro_flush_timeout (/sys/class/net/ethX/)
By default, both gro_flush_timeout and napi_defer_hard_irqs are zero.
This patch does not change the prior behavior of gro_flush_timeout
if used alone : NIC hard irqs should be rearmed as before.
One concrete usage can be :
echo 20000 >/sys/class/net/eth1/gro_flush_timeout
echo 10 >/sys/class/net/eth1/napi_defer_hard_irqs
If at least one packet is retired, then we will reset napi counter
to 10 (napi_defer_hard_irqs), ensuring at least 10 periodic scans
of the queue.
On busy queues, this should avoid NIC hard IRQ, while before this patch IRQ
avoidance was only possible if napi->poll() was exhausting its budget
and not call napi_complete_done().
This feature also can be used to work around some non-optimal NIC irq
coalescing strategies.
Having the ability to insert XX usec delays between each napi->poll()
can increase cache efficiency, since we increase batch sizes.
It also keeps serving cpus not idle too long, reducing tail latencies.
Co-developed-by: Luigi Rizzo <lrizzo@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
