Sameeh Jubran [Mon, 3 Jun 2019 14:43:26 +0000 (17:43 +0300)]
net: ena: add support for changing max_header_size in LLQ mode
Up until now the driver always used a single setting for the sizes
of the different parts of the llq entry - 128 for entry size, 2 for
descriptors before header and 96 for maximum header size.
The current code makes sure that the parts of the llq entry are
compatible with each other and with the initial llq entry size given
by the device.
This commit changes this code to support any llq entry size
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: Sameeh Jubran <sameehj@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Sameeh Jubran [Mon, 3 Jun 2019 14:43:25 +0000 (17:43 +0300)]
net: ena: allow automatic fallback to polling mode
Enable fallback to polling mode for Admin queue
when identified a command response arrival
without an accompanying MSI-X interrupt
Signed-off-by: Igor Chauskin <igorch@amazon.com>
Signed-off-by: Sameeh Jubran <sameehj@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Sameeh Jubran [Mon, 3 Jun 2019 14:43:24 +0000 (17:43 +0300)]
net: ena: documentation: update ena.txt
Small cosmetic changes to ena.txt
Signed-off-by: Sameeh Jubran <sameehj@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Sameeh Jubran [Mon, 3 Jun 2019 14:43:23 +0000 (17:43 +0300)]
net: ena: add newline at the end of pr_err prints
Some pr_err prints lacked '\n' in the end. Added where missing.
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: Sameeh Jubran <sameehj@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Sameeh Jubran [Mon, 3 Jun 2019 14:43:22 +0000 (17:43 +0300)]
net: ena: arrange ena_probe() function variables in reverse christmas tree
Reverse christmas tree arrangement is when strings are written from longer
to shorter with each line. Most of our functions are abiding this
arrangement but this function does not.
In this commit we arrange the variables of ena_probe() in reverse christmas
tree.
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: Sameeh Jubran <sameehj@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Sameeh Jubran [Mon, 3 Jun 2019 14:43:21 +0000 (17:43 +0300)]
net: ena: replace free_tx/rx_ids union with single free_ids field in ena_ring
struct ena_ring holds a union of free_rx_ids and free_tx_ids.
Both of the above fields mean the exact same thing and are used
exactly the same way.
Furthermore, these fields are always used with a prefix of the
type of ring. So for tx it will be tx_ring->free_tx_ids, and for
rx it will be rx_ring->free_rx_ids, which shows how redundant the
"_tx" and "_rx" parts are.
Furthermore still, this may lead to confusing code like where
tx_ring->free_rx_ids which works correctly but looks like a mess.
This commit removes the aforementioned redundancy by replacing the
free_rx/tx_ids union with a single free_ids field.
It also changes a single goto label name from err_free_tx_ids: to
err_tx_free_ids: for consistency with the above new notation.
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: Sameeh Jubran <sameehj@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Arthur Kiyanovski [Mon, 3 Jun 2019 14:43:20 +0000 (17:43 +0300)]
net: ena: ethtool: add extra properties retrieval via get_priv_flags
This commit adds a mechanism for exposing different device
properties via ethtool's priv_flags. The strings are provided
by the device and copied to user space through the driver.
In this commit we:
Add commands, structs and defines necessary for handling
extra properties
Add functions for:
Allocation/destruction of a buffer for extra properties strings.
Retreival of extra properties strings and flags from the network device.
Handle the allocation of a buffer for extra properties strings.
* Initialize buffer with extra properties strings from the
network device at driver startup.
Use ethtool's get_priv_flags to expose extra properties of
the ENA device
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: Sameeh Jubran <sameehj@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Sameeh Jubran [Mon, 3 Jun 2019 14:43:19 +0000 (17:43 +0300)]
net: ena: add handling of llq max tx burst size
There is a maximum TX burst size that the ENA device can handle.
It is exposed by the device to the driver and the driver
needs to comply with it to avoid bugs.
In this commit we:
1. Add ena_com_is_doorbell_needed(), which calculates the number of
llq entries that will be used to hold a packet, and will return
true if they exceed the number of allowed entries in a burst.
If the function returns true, a doorbell needs to be invoked
to send this packet in the next burst.
2. Follow the available entries in the current burst:
- Every doorbell a new burst begins
- With each write of an llq entry, the available entries in the
current burst are decreased by 1.
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: Sameeh Jubran <sameehj@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rasmus Villemoes [Mon, 3 Jun 2019 08:04:09 +0000 (08:04 +0000)]
net: dsa: mv88e6xxx: make mv88e6xxx_g1_stats_wait static
mv88e6xxx_g1_stats_wait has no users outside global1.c, so make it
static.
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Reviewed-by: Vivien Didelot <vivien.didelot@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rasmus Villemoes [Mon, 3 Jun 2019 07:52:46 +0000 (07:52 +0000)]
net: dsa: mv88e6xxx: fix comments and macro names in mv88e6390_g1_mgmt_rsvd2cpu
The macros have an extraneous '800' (after 0180C2 there should be just
six nibbles, with X representing one), while the comments have
interchanged c2 and 80 and an extra :00.
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Fri, 31 May 2019 18:44:09 +0000 (12:44 -0600)]
nexthop: Add entry to MAINTAINERS
Add entry to MAINTAINERS file for new nexthop code.
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 3 Jun 2019 01:15:38 +0000 (18:15 -0700)]
Merge branch 'r8169-replace-several-function-pointers-with-direct-calls'
Heiner Kallweit says:
====================
r8169: replace several function pointers with direct calls
This series removes most function pointers from struct rtl8169_private
and uses direct calls instead. This simplifies the code and avoids
the penalty of indirect calls in times of retpoline.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Heiner Kallweit [Fri, 31 May 2019 17:55:11 +0000 (19:55 +0200)]
r8169: avoid tso csum function indirection
Replace indirect call to tso_csum with direct calls. To do this we have
to move rtl_chip_supports_csum_v2().
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Heiner Kallweit [Fri, 31 May 2019 17:54:03 +0000 (19:54 +0200)]
r8169: remove struct jumbo_ops
The jumbo_ops are used in just one place, so we can simplify the code
and avoid the penalty of indirect calls in times of retpoline.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Heiner Kallweit [Fri, 31 May 2019 17:53:28 +0000 (19:53 +0200)]
r8169: remove struct mdio_ops
The mdio_ops are used in just one place, so we can simplify the code
and avoid the penalty of indirect calls in times of retpoline.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Heiner Kallweit [Fri, 31 May 2019 17:17:15 +0000 (19:17 +0200)]
r8169: improve r8169_csum_workaround
Use helper skb_is_gso() and simplify access to tx_dropped.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Heiner Kallweit [Fri, 31 May 2019 17:14:44 +0000 (19:14 +0200)]
net: ethernet: improve eth_platform_get_mac_address
pci_device_to_OF_node(to_pci_dev(dev)) is the same as dev->of_node,
so we can simplify the code. In addition add an empty line before
the return statement.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 3 Jun 2019 01:08:47 +0000 (18:08 -0700)]
Merge branch 'ifa_list-RCU'
Florian Westphal says:
====================
net: add rcu annotations for ifa_list
v3: fix typo in patch1 commit message
All other patches are unchanged.
v2: remove ifa_list iteration in afs instead of conversion
Eric Dumazet reported following problem:
It looks that unless RTNL is held, accessing ifa_list needs proper RCU
protection. indev->ifa_list can be changed under us by another cpu
(which owns RTNL) [..]
A proper rcu_dereference() with an happy sparse support would require
adding __rcu attribute.
This patch series does that: add __rcu to the ifa_list pointers.
That makes sparse complain, so the series also adds the required
rcu_assign_pointer/dereference helpers where needed.
All patches except the last one are preparation work.
Two new macros are introduced for in_ifaddr walks.
Last patch adds the __rcu annotations and the assign_pointer/dereference
helper calls.
This patch is a bit large, but I found no better way -- other
approaches (annotate-first or add helpers-first) all result in
mid-series sparse warnings.
This series is submitted vs. net-next rather than net for several
reasons:
1. Its (mostly) compile-tested only
2. 3rd patch changes behaviour wrt. secondary addresses
(see changelog)
3. The problem exists for a very long time (2004), so it doesn't
seem to be urgent to fix this -- rcu use to free ifa_list
predates the git era.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Westphal [Fri, 31 May 2019 16:27:09 +0000 (18:27 +0200)]
net: ipv4: provide __rcu annotation for ifa_list
ifa_list is protected by rcu, yet code doesn't reflect this.
Add the __rcu annotations and fix up all places that are now reported by
sparse.
I've done this in the same commit to not add intermediate patches that
result in new warnings.
Reported-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Westphal [Fri, 31 May 2019 16:27:08 +0000 (18:27 +0200)]
drivers: use in_dev_for_each_ifa_rtnl/rcu
Like previous patches, use the new iterator macros to avoid sparse
warnings once proper __rcu annotations are added.
Compile tested only.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Westphal [Fri, 31 May 2019 16:27:07 +0000 (18:27 +0200)]
net: use new in_dev_ifa iterators
Use in_dev_for_each_ifa_rcu/rtnl instead.
This prevents sparse warnings once proper __rcu annotations are added.
Signed-off-by: Florian Westphal <fw@strlen.de>
t di# Last commands done (6 commands done):
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Westphal [Fri, 31 May 2019 16:27:06 +0000 (18:27 +0200)]
netfilter: use in_dev_for_each_ifa_rcu
Netfilter hooks are always running under rcu read lock, use
the new iterator macro so sparse won't complain once we add
proper __rcu annotations.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Westphal [Fri, 31 May 2019 16:27:05 +0000 (18:27 +0200)]
devinet: use in_dev_for_each_ifa_rcu in more places
This also replaces spots that used for_primary_ifa().
for_primary_ifa() aborts the loop on the first secondary address seen.
Replace it with either the rcu or rtnl variant of in_dev_for_each_ifa(),
but two places will now also consider secondary addresses too:
inet_addr_onlink() and inet_ifa_byprefix().
I do not understand why they should ignore secondary addresses.
Why would a secondary address not be considered 'on link'?
When matching a prefix, why ignore a matching secondary address?
Other places get converted as well, but gain "->flags & SECONDARY" check.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Westphal [Fri, 31 May 2019 16:27:04 +0000 (18:27 +0200)]
net: inetdevice: provide replacement iterators for in_ifaddr walk
The ifa_list is protected either by rcu or rtnl lock, but the
current iterators do not account for this.
This adds two iterators as replacement, a later patch in
the series will update them with the needed rcu/rtnl_dereference calls.
Its not done in this patch yet to avoid sparse warnings -- the fields
lack the proper __rcu annotation.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Westphal [Fri, 31 May 2019 16:27:03 +0000 (18:27 +0200)]
afs: do not send list of client addresses
David Howells says:
I'm told that there's not really any point populating the list.
Current OpenAFS ignores it, as does AuriStor - and IBM AFS 3.6 will
do the right thing.
The list is actually useless as it's the client's view of the world,
not the servers, so if there's any NAT in the way its contents are
invalid. Further, it doesn't support IPv6 addresses.
On that basis, feel free to make it an empty list and remove all the
interface enumeration.
V1 of this patch reworked the function to use a new helper for the
ifa_list iteration to avoid sparse warnings once the proper __rcu
annotations get added in struct in_device later.
But, in light of the above, just remove afs_get_ipv4_interfaces.
Compile tested only.
Cc: David Howells <dhowells@redhat.com>
Cc: linux-afs@lists.infradead.org
Signed-off-by: Florian Westphal <fw@strlen.de>
Tested-by: David Howells <dhowells@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Colin Ian King [Fri, 31 May 2019 13:27:38 +0000 (14:27 +0100)]
qed: remove redundant assignment to rc
The variable rc is assigned with a value that is never read and
it is re-assigned a new value later on. The assignment is redundant
and can be removed.
Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 3 Jun 2019 00:48:58 +0000 (17:48 -0700)]
Merge tag 'isdn-removal' of https://git./linux/kernel/git/arnd/playground
Arnd Bergmann says:
====================
isdn: deprecate non-mISDN drivers
When isdn4linux came up in the context of another patch series, I
remembered that we had discussed removing it a while ago.
It turns out that the suggestion from Karsten Keil wa to remove I4L
in 2018 after the last public ISDN networks are shut down. This has
happened now (with a very small number of exceptions), so I guess it's
time to try again.
We currently have three ISDN stacks in the kernel: the original
isdn4linux (with the hisax driver), the newer CAPI (with four drivers),
and finally the mISDN stack (supporting roughly the same hardware as
hisax).
As far as I can tell, anyone using ISDN with mainline kernel drivers in
the past few years uses mISDN, and this is typically used for voice-only
PBX installations that don't require a public network.
The older stacks support additional features for data networks, but those
typically make no sense any more if there is no network to connect to.
My proposal for this time is to kill off isdn4linux entirely, as it seems
to have been unusable for quite a while. This code has been abandoned
for many years and it does cause problems for treewide maintenance as
it tends to do everything that we try to stop doing.
Birger Harzenetter mentioned that is is still using i4l in order to
make use of the 'divert' feature that is not part of mISDN, but has
otherwise moved on to mISDN for normal operation, like apparently
everyone else.
CAPI in turn is not quite as obsolete, but two of the drivers (avm
and hysdn) don't seem to be used at all, while another one (gigaset)
will stop being maintained as Paul Bolle is no longer able to
test it after the network gets shut down in September.
All three are now moved into drivers/staging to let others speak
up in case there are remaining users.
This leaves Bluetooth CMTP as the only remaining user of CAPI, but
Marcel Holtmann wishes to keep maintaining it.
For the discussion on version 1, see [2]
Unfortunately, Karsten Keil as the maintainer has not participated in
the discussion.
Arnd
[1] https://patchwork.kernel.org/patch/
8484861/#
17900371
[2] https://listserv.isdn4linux.de/pipermail/isdn4linux/2019-April/thread.html
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Sun, 2 Jun 2019 20:49:49 +0000 (13:49 -0700)]
Merge branch 'mscc-ocelot-tc-flower'
Horatiu Vultur says:
====================
Add hw offload of TC flower on MSCC Ocelot
This patch series enables hardware offload for flower filter used in
traffic controller on MSCC Ocelot board.
v2->v3 changes:
- remove the check for shared blocks
v1->v2 changes:
- when declaring variables use reverse christmas tree
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Horatiu Vultur [Fri, 31 May 2019 07:16:57 +0000 (09:16 +0200)]
net: mscc: ocelot: Hardware ofload for tc flower filter
Hardware offload of port filtering are now supported via tc command using
flower filter. ACL rules are used to enable the hardware offload.
The following keys are supported:
vlan_id
vlan_prio
dst_mac/src_mac for non IP frames
dst_ip/src_ip
dst_port/src_port
The following actions are supported:
trap
drop
These filters are supported only on the ingress schedulare.
Add:
tc qdisc add dev eth3 ingress
tc filter ad dev eth3 parent ffff: ip_proto ip flower \
ip_proto tcp dst_port 80 action drop
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Horatiu Vultur [Fri, 31 May 2019 07:16:56 +0000 (09:16 +0200)]
net: mscc: ocelot: Add support for tcam
Add ACL support using the TCAM. Using ACL it is possible to create rules
in hardware to filter/redirect frames.
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Thu, 30 May 2019 19:06:36 +0000 (12:06 -0700)]
selftests: Add test cases for nexthop objects
Add functional test cases for nexthop objects.
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Sat, 1 Jun 2019 23:21:19 +0000 (16:21 -0700)]
Merge git://git./linux/kernel/git/pablo/nf-next
Pablo Neira Ayuso says:
====================
Netfilter/IPVS updates for net-next
The following patchset container Netfilter/IPVS update for net-next:
1) Add UDP tunnel support for ICMP errors in IPVS.
Julian Anastasov says:
This patchset is a followup to the commit that adds UDP/GUE tunnel:
"ipvs: allow tunneling with gue encapsulation".
What we do is to put tunnel real servers in hash table (patch 1),
add function to lookup tunnels (patch 2) and use it to strip the
embedded tunnel headers from ICMP errors (patch 3).
2) Extend xt_owner to match for supplementary groups, from
Lukasz Pawelczyk.
3) Remove unused oif field in flow_offload_tuple object, from
Taehee Yoo.
4) Release basechain counters from workqueue to skip synchronize_rcu()
call. From Florian Westphal.
5) Replace skb_make_writable() by skb_ensure_writable(). Patchset
from Florian Westphal.
6) Checksum support for gue encapsulation in IPVS, from Jacky Hu.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Sat, 1 Jun 2019 04:21:18 +0000 (21:21 -0700)]
Merge git://git./linux/kernel/git/bpf/bpf-next
Alexei Starovoitov says:
====================
pull-request: bpf-next 2019-05-31
The following pull-request contains BPF updates for your *net-next* tree.
Lots of exciting new features in the first PR of this developement cycle!
The main changes are:
1) misc verifier improvements, from Alexei.
2) bpftool can now convert btf to valid C, from Andrii.
3) verifier can insert explicit ZEXT insn when requested by 32-bit JITs.
This feature greatly improves BPF speed on 32-bit architectures. From Jiong.
4) cgroups will now auto-detach bpf programs. This fixes issue of thousands
bpf programs got stuck in dying cgroups. From Roman.
5) new bpf_send_signal() helper, from Yonghong.
6) cgroup inet skb programs can signal CN to the stack, from Lawrence.
7) miscellaneous cleanups, from many developers.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Alan Maguire [Fri, 31 May 2019 17:47:14 +0000 (18:47 +0100)]
selftests/bpf: measure RTT from xdp using xdping
xdping allows us to get latency estimates from XDP. Output looks
like this:
./xdping -I eth4 192.168.55.8
Setting up XDP for eth4, please wait...
XDP setup disrupts network connectivity, hit Ctrl+C to quit
Normal ping RTT data
[Ignore final RTT; it is distorted by XDP using the reply]
PING 192.168.55.8 (192.168.55.8) from 192.168.55.7 eth4: 56(84) bytes of data.
64 bytes from 192.168.55.8: icmp_seq=1 ttl=64 time=0.302 ms
64 bytes from 192.168.55.8: icmp_seq=2 ttl=64 time=0.208 ms
64 bytes from 192.168.55.8: icmp_seq=3 ttl=64 time=0.163 ms
64 bytes from 192.168.55.8: icmp_seq=8 ttl=64 time=0.275 ms
4 packets transmitted, 4 received, 0% packet loss, time 3079ms
rtt min/avg/max/mdev = 0.163/0.237/0.302/0.054 ms
XDP RTT data:
64 bytes from 192.168.55.8: icmp_seq=5 ttl=64 time=0.02808 ms
64 bytes from 192.168.55.8: icmp_seq=6 ttl=64 time=0.02804 ms
64 bytes from 192.168.55.8: icmp_seq=7 ttl=64 time=0.02815 ms
64 bytes from 192.168.55.8: icmp_seq=8 ttl=64 time=0.02805 ms
The xdping program loads the associated xdping_kern.o BPF program
and attaches it to the specified interface. If run in client
mode (the default), it will add a map entry keyed by the
target IP address; this map will store RTT measurements, current
sequence number etc. Finally in client mode the ping command
is executed, and the xdping BPF program will use the last ICMP
reply, reformulate it as an ICMP request with the next sequence
number and XDP_TX it. After the reply to that request is received
we can measure RTT and repeat until the desired number of
measurements is made. This is why the sequence numbers in the
normal ping are 1, 2, 3 and 8. We XDP_TX a modified version
of ICMP reply 4 and keep doing this until we get the 4 replies
we need; hence the networking stack only sees reply 8, where
we have XDP_PASSed it upstream since we are done.
In server mode (-s), xdping simply takes ICMP requests and replies
to them in XDP rather than passing the request up to the networking
stack. No map entry is required.
xdping can be run in native XDP mode (the default, or specified
via -N) or in skb mode (-S).
A test program test_xdping.sh exercises some of these options.
Note that native XDP does not seem to XDP_TX for veths, hence -N
is not tested. Looking at the code, it looks like XDP_TX is
supported so I'm not sure if that's expected. Running xdping in
native mode for ixgbe as both client and server works fine.
Changes since v4
- close fds on cleanup (Song Liu)
Changes since v3
- fixed seq to be __be16 (Song Liu)
- fixed fd checks in xdping.c (Song Liu)
Changes since v2
- updated commit message to explain why seq number of last
ICMP reply is 8 not 4 (Song Liu)
- updated types of seq number, raddr and eliminated csum variable
in xdpclient/xdpserver functions as it was not needed (Song Liu)
- added XDPING_DEFAULT_COUNT definition and usage specification of
default/max counts (Song Liu)
Changes since v1
- moved from RFC to PATCH
- removed unused variable in ipv4_csum() (Song Liu)
- refactored ICMP checks into icmp_check() function called by client
and server programs and reworked client and server programs due
to lack of shared code (Song Liu)
- added checks to ensure that SKB and native mode are not requested
together (Song Liu)
Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
David S. Miller [Sat, 1 Jun 2019 00:13:19 +0000 (17:13 -0700)]
Merge branch '40GbE' of git://git./linux/kernel/git/jkirsher/next-queue
Jeff Kirsher says:
====================
Intel Wired LAN Driver Updates 2019-05-31
This series contains updates to the iavf driver.
Nathan Chancellor converts the use of gnu_printf to printf.
Aleksandr modifies the driver to limit the number of RSS queues to the
number of online CPUs in order to avoid creating misconfigured RSS
queues.
Gustavo A. R. Silva converts a couple of instances where sizeof() can be
replaced with struct_size().
Alice makes the remaining changes to the iavf driver to cleanup all the
old "i40evf" references in the driver to iavf, including the file names
that still contained the old driver reference. There was no functional
changes made, just cosmetic to reduce any confusion going forward now
that the iavf driver is the virtual function driver for both i40e and
ice drivers.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiong Wang [Thu, 30 May 2019 20:23:18 +0000 (21:23 +0100)]
bpf: doc: update answer for 32-bit subregister question
There has been quite a few progress around the two steps mentioned in the
answer to the following question:
Q: BPF 32-bit subregister requirements
This patch updates the answer to reflect what has been done.
v2:
- Add missing full stop. (Song Liu)
- Minor tweak on one sentence. (Song Liu)
v1:
- Integrated rephrase from Quentin and Jakub
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Jiong Wang <jiong.wang@netronome.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Alexei Starovoitov [Fri, 31 May 2019 23:52:56 +0000 (16:52 -0700)]
Merge branch 'map-charge-cleanup'
Roman Gushchin says:
====================
During my work on memcg-based memory accounting for bpf maps
I've done some cleanups and refactorings of the existing
memlock rlimit-based code. It makes it more robust, unifies
size to pages conversion, size checks and corresponding error
codes. Also it adds coverage for cgroup local storage and
socket local storage maps.
It looks like some preliminary work on the mm side might be
required to start working on the memcg-based accounting,
so I'm sending these patches as a separate patchset.
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Roman Gushchin [Thu, 30 May 2019 01:03:59 +0000 (18:03 -0700)]
bpf: move memory size checks to bpf_map_charge_init()
Most bpf map types doing similar checks and bytes to pages
conversion during memory allocation and charging.
Let's unify these checks by moving them into bpf_map_charge_init().
Signed-off-by: Roman Gushchin <guro@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Roman Gushchin [Thu, 30 May 2019 01:03:58 +0000 (18:03 -0700)]
bpf: rework memlock-based memory accounting for maps
In order to unify the existing memlock charging code with the
memcg-based memory accounting, which will be added later, let's
rework the current scheme.
Currently the following design is used:
1) .alloc() callback optionally checks if the allocation will likely
succeed using bpf_map_precharge_memlock()
2) .alloc() performs actual allocations
3) .alloc() callback calculates map cost and sets map.memory.pages
4) map_create() calls bpf_map_init_memlock() which sets map.memory.user
and performs actual charging; in case of failure the map is
destroyed
<map is in use>
1) bpf_map_free_deferred() calls bpf_map_release_memlock(), which
performs uncharge and releases the user
2) .map_free() callback releases the memory
The scheme can be simplified and made more robust:
1) .alloc() calculates map cost and calls bpf_map_charge_init()
2) bpf_map_charge_init() sets map.memory.user and performs actual
charge
3) .alloc() performs actual allocations
<map is in use>
1) .map_free() callback releases the memory
2) bpf_map_charge_finish() performs uncharge and releases the user
The new scheme also allows to reuse bpf_map_charge_init()/finish()
functions for memcg-based accounting. Because charges are performed
before actual allocations and uncharges after freeing the memory,
no bogus memory pressure can be created.
In cases when the map structure is not available (e.g. it's not
created yet, or is already destroyed), on-stack bpf_map_memory
structure is used. The charge can be transferred with the
bpf_map_charge_move() function.
Signed-off-by: Roman Gushchin <guro@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Roman Gushchin [Thu, 30 May 2019 01:03:57 +0000 (18:03 -0700)]
bpf: group memory related fields in struct bpf_map_memory
Group "user" and "pages" fields of bpf_map into the bpf_map_memory
structure. Later it can be extended with "memcg" and other related
information.
The main reason for a such change (beside cosmetics) is to pass
bpf_map_memory structure to charging functions before the actual
allocation of bpf_map.
Signed-off-by: Roman Gushchin <guro@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Roman Gushchin [Thu, 30 May 2019 01:03:56 +0000 (18:03 -0700)]
bpf: add memlock precharge for socket local storage
Socket local storage maps lack the memlock precharge check,
which is performed before the memory allocation for
most other bpf map types.
Let's add it in order to unify all map types.
Signed-off-by: Roman Gushchin <guro@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Roman Gushchin [Thu, 30 May 2019 01:03:55 +0000 (18:03 -0700)]
bpf: add memlock precharge check for cgroup_local_storage
Cgroup local storage maps lack the memlock precharge check,
which is performed before the memory allocation for
most other bpf map types.
Let's add it in order to unify all map types.
Signed-off-by: Roman Gushchin <guro@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Alexei Starovoitov [Fri, 31 May 2019 23:41:30 +0000 (16:41 -0700)]
Merge branch 'propagate-cn-to-tcp'
Lawrence Brakmo says:
====================
This patchset adds support for propagating congestion notifications (cn)
to TCP from cgroup inet skb egress BPF programs.
Current cgroup skb BPF programs cannot trigger TCP congestion window
reductions, even when they drop a packet. This patch-set adds support
for cgroup skb BPF programs to send congestion notifications in the
return value when the packets are TCP packets. Rather than the
current 1 for keeping the packet and 0 for dropping it, they can
now return:
NET_XMIT_SUCCESS (0) - continue with packet output
NET_XMIT_DROP (1) - drop packet and do cn
NET_XMIT_CN (2) - continue with packet output and do cn
-EPERM - drop packet
Finally, HBM programs are modified to collect and return more
statistics.
There has been some discussion regarding the best place to manage
bandwidths. Some believe this should be done in the qdisc where it can
also be managed with a BPF program. We believe there are advantages
for doing it with a BPF program in the cgroup/skb callback. For example,
it reduces overheads in the cases where there is on primary workload and
one or more secondary workloads, where each workload is running on its
own cgroupv2. In this scenario, we only need to throttle the secondary
workloads and there is no overhead for the primary workload since there
will be no BPF program attached to its cgroup.
Regardless, we agree that this mechanism should not penalize those that
are not using it. We tested this by doing 1 byte req/reply RPCs over
loopback. Each test consists of 30 sec of back-to-back 1 byte RPCs.
Each test was repeated 50 times with a 1 minute delay between each set
of 10. We then calculated the average RPCs/sec over the 50 tests. We
compare upstream with upstream + patchset and no BPF program as well
as upstream + patchset and a BPF program that just returns ALLOW_PKT.
Here are the results:
upstream 80937 RPCs/sec
upstream + patches, no BPF program 80894 RPCs/sec
upstream + patches, BPF program 80634 RPCs/sec
These numbers indicate that there is no penalty for these patches
The use of congestion notifications improves the performance of HBM when
using Cubic. Without congestion notifications, Cubic will not decrease its
cwnd and HBM will need to drop a large percentage of the packets.
The following results are obtained for rate limits of 1Gbps,
between two servers using netperf, and only one flow. We also show how
reducing the max delayed ACK timer can improve the performance when
using Cubic.
Command used was:
./do_hbm_test.sh -l -D --stats -N -r=<rate> [--no_cn] [dctcp] \
-s=<server running netserver>
where:
<rate> is 1000
--no_cn specifies no cwr notifications
dctcp uses dctcp
Cubic DCTCP
Lim, DA Mbps cwnd cred drops Mbps cwnd cred drops
-------- ---- ---- ---- ----- ---- ---- ---- -----
1G, 40 35 462 -320 67% 995 1 -212 0.05%
1G, 40,cn 736 9 -78 0.07 995 1 -212 0.05
1G, 5,cn 941 2 -189 0.13 995 1 -212 0.05
Notes:
--no_cn has no effect with DCTCP
Lim = rate limit
DA = maximum delay ack timer
cred = credit in packets
drops = % packets dropped
v1->v2: Insures that only BPF_CGROUP_INET_EGRESS can return values 2 and 3
New egress values apply to all protocols, not just TCP
Cleaned up patch 4, Update BPF_CGROUP_RUN_PROG_INET_EGRESS callers
Removed changes to __tcp_transmit_skb (patch 5), no longer needed
Removed sample use of EDT
v2->v3: Removed the probe timer related changes
v3->v4: Replaced preempt_enable_no_resched() by preempt_enable()
in BPF_PROG_CGROUP_INET_EGRESS_RUN_ARRAY() macro
====================
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
brakmo [Tue, 28 May 2019 23:59:40 +0000 (16:59 -0700)]
bpf: Add more stats to HBM
Adds more stats to HBM, including average cwnd and rtt of all TCP
flows, percents of packets that are ecn ce marked and distribution
of return values.
Signed-off-by: Lawrence Brakmo <brakmo@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
brakmo [Tue, 28 May 2019 23:59:39 +0000 (16:59 -0700)]
bpf: Add cn support to hbm_out_kern.c
Update hbm_out_kern.c to support returning cn notifications.
Also updates relevant files to allow disabling cn notifications.
Signed-off-by: Lawrence Brakmo <brakmo@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
brakmo [Tue, 28 May 2019 23:59:38 +0000 (16:59 -0700)]
bpf: Update BPF_CGROUP_RUN_PROG_INET_EGRESS calls
Update BPF_CGROUP_RUN_PROG_INET_EGRESS() callers to support returning
congestion notifications from the BPF programs.
Signed-off-by: Lawrence Brakmo <brakmo@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
brakmo [Tue, 28 May 2019 23:59:37 +0000 (16:59 -0700)]
bpf: Update __cgroup_bpf_run_filter_skb with cn
For egress packets, __cgroup_bpf_fun_filter_skb() will now call
BPF_PROG_CGROUP_INET_EGRESS_RUN_ARRAY() instead of PROG_CGROUP_RUN_ARRAY()
in order to propagate congestion notifications (cn) requests to TCP
callers.
For egress packets, this function can return:
NET_XMIT_SUCCESS (0) - continue with packet output
NET_XMIT_DROP (1) - drop packet and notify TCP to call cwr
NET_XMIT_CN (2) - continue with packet output and notify TCP
to call cwr
-EPERM - drop packet
For ingress packets, this function will return -EPERM if any attached
program was found and if it returned != 1 during execution. Otherwise 0
is returned.
Signed-off-by: Lawrence Brakmo <brakmo@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
brakmo [Tue, 28 May 2019 23:59:36 +0000 (16:59 -0700)]
bpf: cgroup inet skb programs can return 0 to 3
Allows cgroup inet skb programs to return values in the range [0, 3].
The second bit is used to deterine if congestion occurred and higher
level protocol should decrease rate. E.g. TCP would call tcp_enter_cwr()
The bpf_prog must set expected_attach_type to BPF_CGROUP_INET_EGRESS
at load time if it uses the new return values (i.e. 2 or 3).
The expected_attach_type is currently not enforced for
BPF_PROG_TYPE_CGROUP_SKB. e.g Meaning the current bpf_prog with
expected_attach_type setting to BPF_CGROUP_INET_EGRESS can attach to
BPF_CGROUP_INET_INGRESS. Blindly enforcing expected_attach_type will
break backward compatibility.
This patch adds a enforce_expected_attach_type bit to only
enforce the expected_attach_type when it uses the new
return value.
Signed-off-by: Lawrence Brakmo <brakmo@fb.com>
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
brakmo [Tue, 28 May 2019 23:59:35 +0000 (16:59 -0700)]
bpf: Create BPF_PROG_CGROUP_INET_EGRESS_RUN_ARRAY
Create new macro BPF_PROG_CGROUP_INET_EGRESS_RUN_ARRAY() to be used by
__cgroup_bpf_run_filter_skb for EGRESS BPF progs so BPF programs can
request cwr for TCP packets.
Current cgroup skb programs can only return 0 or 1 (0 to drop the
packet. This macro changes the behavior so the low order bit
indicates whether the packet should be dropped (0) or not (1)
and the next bit is used for congestion notification (cn).
Hence, new allowed return values of CGROUP EGRESS BPF programs are:
0: drop packet
1: keep packet
2: drop packet and call cwr
3: keep packet and call cwr
This macro then converts it to one of NET_XMIT values or -EPERM
that has the effect of dropping the packet with no cn.
0: NET_XMIT_SUCCESS skb should be transmitted (no cn)
1: NET_XMIT_DROP skb should be dropped and cwr called
2: NET_XMIT_CN skb should be transmitted and cwr called
3: -EPERM skb should be dropped (no cn)
Note that when more than one BPF program is called, the packet is
dropped if at least one of programs requests it be dropped, and
there is cn if at least one program returns cn.
Signed-off-by: Lawrence Brakmo <brakmo@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Colin Ian King [Thu, 30 May 2019 19:04:38 +0000 (20:04 +0100)]
xen-netback: remove redundant assignment to err
The variable err is assigned with the value -ENOMEM that is never
read and it is re-assigned a new value later on. The assignment is
redundant and can be removed.
Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Colin Ian King [Thu, 30 May 2019 15:57:54 +0000 (16:57 +0100)]
nexthop: remove redundant assignment to err
The variable err is initialized with a value that is never read
and err is reassigned a few statements later. This initialization
is redundant and can be removed.
Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 31 May 2019 19:37:46 +0000 (12:37 -0700)]
Merge branch 'phylink-sfp-updates'
Russell King says:
====================
phylink/sfp updates
This is a series of updates to phylink and sfp:
- Remove an unused net device argument from the phylink MII ioctl
emulation code.
- add support for using interrupts when using a GPIO for link status
tracking, rather than polling it at one second intervals. This
reduces the need to wakeup the CPU every second.
- add support to the MII ioctl API to read and write Clause 45 PHY
registers. I don't know how desirable this is for mainline, but I
have used this facility extensively to investigate the Marvell
88x3310 PHY. A recent illustration of use for this was debugging
the PHY-without-firmware problem recently reported.
- add mandatory attach/detach methods for the upstream side of sfp
bus code, which will allow us to remove the "netdev" structure from
the SFP layers.
- remove the "netdev" structure from the SFP upstream registration
calls, which simplifies PHY to SFP links.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Russell King [Tue, 28 May 2019 09:57:39 +0000 (10:57 +0100)]
net: sfp: remove sfp-bus use of netdevs
The sfp-bus code now no longer has any use for the network device
structure, so remove its use.
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Russell King [Tue, 28 May 2019 09:57:34 +0000 (10:57 +0100)]
net: sfp: add mandatory attach/detach methods for sfp buses
Add attach and detach methods for SFP buses, which will allow us to get
rid of the netdev storage in sfp-bus.
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Russell King [Tue, 28 May 2019 09:57:29 +0000 (10:57 +0100)]
net: phy: allow Clause 45 access via mii ioctl
Allow userspace to generate Clause 45 MII access cycles via phylib.
This is useful for tools such as mii-diag to be able to inspect Clause
45 PHYs.
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Russell King [Tue, 28 May 2019 09:57:23 +0000 (10:57 +0100)]
net: phylink: support for link gpio interrupt
Add support for using GPIO interrupts with a fixed-link GPIO rather than
polling the GPIO every second and invoking the phylink resolution. This
avoids unnecessary calls to mac_config().
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Russell King [Tue, 28 May 2019 09:57:18 +0000 (10:57 +0100)]
net: phylink: remove netdev from phylink mii ioctl emulation
The netdev used in the phylink ioctl emulation is never used, so let's
remove it.
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 31 May 2019 17:49:43 +0000 (10:49 -0700)]
Merge git://git./linux/kernel/git/davem/net
The phylink conflict was between a bug fix by Russell King
to make sure we have a consistent PHY interface mode, and
a change in net-next to pull some code in phylink_resolve()
into the helper functions phylink_mac_link_{up,down}()
On the dp83867 side it's mostly overlapping changes, with
the 'net' side removing a condition that was supposed to
trigger for RGMII but because of how it was coded never
actually could trigger.
Signed-off-by: David S. Miller <davem@davemloft.net>
Pablo Neira Ayuso [Fri, 31 May 2019 09:15:26 +0000 (11:15 +0200)]
netfilter: nf_conntrack_bridge: fix CONFIG_IPV6=y
This patch fixes a few problems with CONFIG_IPV6=y and
CONFIG_NF_CONNTRACK_BRIDGE=m:
In file included from net/netfilter/utils.c:5:
include/linux/netfilter_ipv6.h: In function 'nf_ipv6_br_defrag':
include/linux/netfilter_ipv6.h:110:9: error: implicit declaration of function 'nf_ct_frag6_gather'; did you mean 'nf_ct_attach'? [-Werror=implicit-function-declaration]
And these too:
net/ipv6/netfilter.c:242:2: error: unknown field 'br_defrag' specified in initializer
net/ipv6/netfilter.c:243:2: error: unknown field 'br_fragment' specified in initializer
This patch includes an original chunk from wenxu.
Fixes: 764dd163ac92 ("netfilter: nf_conntrack_bridge: add support for IPv6")
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Reported-by: Yuehaibing <yuehaibing@huawei.com>
Reported-by: kbuild test robot <lkp@intel.com>
Reported-by: wenxu <wenxu@ucloud.cn>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: wenxu <wenxu@ucloud.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jacky Hu [Thu, 30 May 2019 00:16:40 +0000 (08:16 +0800)]
ipvs: add checksum support for gue encapsulation
Add checksum support for gue encapsulation with the tun_flags parameter,
which could be one of the values below:
IP_VS_TUNNEL_ENCAP_FLAG_NOCSUM
IP_VS_TUNNEL_ENCAP_FLAG_CSUM
IP_VS_TUNNEL_ENCAP_FLAG_REMCSUM
Signed-off-by: Jacky Hu <hengqing.hu@gmail.com>
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Florian Westphal [Thu, 23 May 2019 13:44:12 +0000 (15:44 +0200)]
netfilter: replace skb_make_writable with skb_ensure_writable
This converts all remaining users and then removes skb_make_writable.
Suggested-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Florian Westphal [Thu, 23 May 2019 13:44:11 +0000 (15:44 +0200)]
netfilter: tcpmss, optstrip: prefer skb_ensure_writable
This also changes optstrip to only make the tcp header writeable
rather than the entire packet.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Florian Westphal [Thu, 23 May 2019 13:44:10 +0000 (15:44 +0200)]
netfilter: xt_HL: prefer skb_ensure_writable
Also, make the argument to be only the needed size of the header
we're altering, no need to pull in the full packet into linear area.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Florian Westphal [Thu, 23 May 2019 13:44:09 +0000 (15:44 +0200)]
netfilter: nf_tables: prefer skb_ensure_writable
.. so skb_make_writable can be removed.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Florian Westphal [Thu, 23 May 2019 13:44:08 +0000 (15:44 +0200)]
netfilter: ipv4: prefer skb_ensure_writable
.. so skb_make_writable can be removed soon.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Florian Westphal [Thu, 23 May 2019 13:44:07 +0000 (15:44 +0200)]
netfilter: conntrack, nat: prefer skb_ensure_writable
like previous patches -- convert conntrack to use the core helper.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Florian Westphal [Thu, 23 May 2019 13:44:06 +0000 (15:44 +0200)]
netfilter: ipvs: prefer skb_ensure_writable
It does the same thing, use it instead so we can remove skb_make_writable.
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Florian Westphal [Thu, 23 May 2019 13:44:05 +0000 (15:44 +0200)]
netfilter: bridge: convert skb_make_writable to skb_ensure_writable
Back in the day, skb_ensure_writable did not exist. By now, both functions
have the same precondition:
I. skb_make_writable will test in this order:
1. wlen > skb->len -> error
2. if not cloned and wlen <= headlen -> OK
3. If cloned and wlen bytes of clone writeable -> OK
After those checks, skb is either not cloned but needs to pull from
nonlinear area, or writing to head would also alter data of another clone.
In both cases skb_make_writable will then call __pskb_pull_tail, which will
kmalloc a new memory area to use for skb->head.
IOW, after successful skb_make_writable call, the requested length is in
linear area and can be modified, even if skb was cloned.
II. skb_ensure_writable will do this instead:
1. call pskb_may_pull. This handles case 1 above.
After this, wlen is in linear area, but skb might be cloned.
2. return if skb is not cloned
3. return if wlen byte of clone are writeable.
4. fully copy the skb.
So post-conditions are the same:
*len bytes are writeable in linear area without altering any payload data
of a clone, all header pointers might have been changed.
Only differences are that skb_ensure_writable is in the core, whereas
skb_make_writable lives in netfilter core and the inverted return value.
skb_make_writable returns 0 on error, whereas skb_ensure_writable returns
negative value.
For the normal cases performance is similar:
A. skb is not cloned and in linear area:
pskb_may_pull is inline helper, so neither function copies.
B. skb is cloned, write is in linear area and clone is writeable:
both funcions return with step 3.
This series removes skb_make_writable from the kernel.
While at it, pass the needed value instead, its less confusing that way:
There is no special-handling of "0-length" argument in either
skb_make_writable or skb_ensure_writable.
bridge already makes sure ethernet header is in linear area, only purpose
of the make_writable() is is to copy skb->head in case of cloned skbs.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Florian Westphal [Wed, 22 May 2019 21:35:11 +0000 (23:35 +0200)]
netfilter: nf_tables: free base chain counters from worker
No need to use synchronize_rcu() here, just swap the two pointers
and have the release occur from work queue after commit has completed.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Taehee Yoo [Wed, 15 May 2019 19:02:31 +0000 (04:02 +0900)]
netfilter: nf_flow_table: remove unnecessary variable in flow_offload_tuple
The oifidx in the struct flow_offload_tuple is not used anymore.
Signed-off-by: Taehee Yoo <ap420073@gmail.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Lukasz Pawelczyk [Fri, 10 May 2019 11:46:22 +0000 (13:46 +0200)]
netfilter: xt_owner: Add supplementary groups option
The XT_OWNER_SUPPL_GROUPS flag causes GIDs specified with XT_OWNER_GID
to be also checked in the supplementary groups of a process.
f_cred->group_info cannot be modified during its lifetime and f_cred
holds a reference to it so it's safe to use.
Signed-off-by: Lukasz Pawelczyk <l.pawelczyk@samsung.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Julian Anastasov [Sun, 5 May 2019 12:14:40 +0000 (15:14 +0300)]
ipvs: strip udp tunnel headers from icmp errors
Recognize UDP tunnels in received ICMP errors and
properly strip the tunnel headers. GUE is what we
have for now.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Julian Anastasov [Sun, 5 May 2019 12:14:39 +0000 (15:14 +0300)]
ipvs: add function to find tunnels
Add ip_vs_find_tunnel() to match tunnel headers
by family, address and optional port. Use it to
properly find the tunnel real server used in
received ICMP errors.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Julian Anastasov [Sun, 5 May 2019 12:14:38 +0000 (15:14 +0300)]
ipvs: allow rs_table to contain different real server types
Before now rs_table was used only for NAT real servers.
Change it to allow TUN real severs from different types,
possibly hashed with different port key.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Arnd Bergmann [Sat, 20 Apr 2019 20:28:45 +0000 (22:28 +0200)]
isdn: move capi drivers to staging
I tried to find any indication of whether the capi drivers are still in
use, and have not found anything from a long time ago.
With public ISDN networks almost completely shut down over the past 12
months, there is very little you can actually do with this hardware. The
main remaining use case would be to connect ISDN voice phones to an
in-house installation with Asterisk or LCR, but anyone trying this in
turn seems to be using either the mISDN driver stack, or out-of-tree
drivers from the hardware vendors.
I may of course have missed something, so I would suggest moving these
three drivers (avm, hysdn, gigaset) into drivers/staging/ just in case
someone still uses them.
If nobody complains, we can remove them entirely in six months, or
otherwise move the core code and any drivers that are still needed back
into drivers/isdn.
As Paul Bolle notes, he is still testing the gigaset driver as long as
he can, but the Dutch ISDN network will be shut down in September 2019,
which puts an end to that.
Marcel Holtmann still maintains the Bluetooth CMTP profile and wants to
keep that alive, so the actual CAPI subsystem code remains in place for
now, after all other drivers are gone, CMTP and CAPI can be merged into
a single driver directory.
Cc: Marcel Holtmann <marcel@holtmann.org>
Cc: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Arnd Bergmann [Thu, 18 Apr 2019 20:57:08 +0000 (22:57 +0200)]
isdn: hdlc: move into mISDN
The last remnant of the isdn4linux interface is now the isdnhdlc
support, used by the netjet driver. Move it next to that driver.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Arnd Bergmann [Thu, 18 Apr 2019 20:43:36 +0000 (22:43 +0200)]
isdn: remove isdn4linux
With all isdn4linux hardware drivers gone, this is only a wrapper around
CAPI to support old user space. However, from looking at the mailing
list, it seems that the last time anyone asked about it was in 2014,
when the upgrade from a linux-2.4 installation failed, and mISDN was
suggested as a replacement.
The largest public ISDN network (Deutsche Telekom) was supposed to be
shut down 2018, which must have drastically reduced the number of legacy
installations.
When we last discussed removing i4l in 2016, Karsten Keil suggested
revisiting this in 2018. I guess this is overdue.
Link: http://listserv.isdn4linux.de/pipermail/isdn4linux/2014-October/006165.html
Link: https://patchwork.kernel.org/patch/8484861/#17900371
Link: https://listserv.isdn4linux.de/pipermail/isdn4linux/2019-April/thread.html
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Arnd Bergmann [Thu, 18 Apr 2019 20:47:35 +0000 (22:47 +0200)]
isdn: remove hisax driver
With the decline of ISDN, this seems to have become almost completely
obsolete, and even in the past years before that, almost all remaining
users appear to have used mISDN instead.
Birger Harzenetter noted that he is still using i4l/hisax to take
advantage of the 'divert' driver for call diversion, but otherwise uses
mISDN on the same hardware. This is a rare edge case as far as I
can tell, but we are still breaking an actively used work flow
(see https://xkcd.com/1172/).
We debated moving i4l/hisax to staging as an intermediate step, but as
he is not likely to change the setup, and that would just delay breaking
this use case. The alternatives here are to stay on stable kernels
< 5.2, to create an external driver repository for isdn4linux, or to
add divert functionality to mISDN.
Cc: Birger Harzenetter <WIMPy@yeti.dk>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Arnd Bergmann [Thu, 18 Apr 2019 20:45:09 +0000 (22:45 +0200)]
isdn: gigaset: remove i4l support
isdn4linux is getting removed, and the gigaset driver can still
use the CAPI support, so this can all go away.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Alice Michael [Wed, 17 Apr 2019 22:17:37 +0000 (15:17 -0700)]
iavf: update comments and file checks to match iavf
Some small things were missed with recent name changes
from i40e to iavf. Having a separate patch allows to
correct the small misses in one place.
Signed-off-by: Alice Michael <alice.michael@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Alice Michael [Wed, 17 Apr 2019 22:17:36 +0000 (15:17 -0700)]
iavf: rename i40e_device to iavf_device
Renaming remaining defines from i40e to iavf
Signed-off-by: Alice Michael <alice.michael@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Alice Michael [Wed, 17 Apr 2019 22:17:35 +0000 (15:17 -0700)]
iavf: change remaining i40e defines to be iavf
There were a couple of erroneously missed i40e names to
update to iavf left after the larger chunks. Updated them
separately so now they should all be aligned as iavf.
Signed-off-by: Alice Michael <alice.michael@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Alice Michael [Wed, 17 Apr 2019 22:17:34 +0000 (15:17 -0700)]
iavf: rename iavf_client.h defines to match driver name
The defines in iavf_client.h were still vastly i40e, and they
should be iavf.
Signed-off-by: Alice Michael <alice.michael@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Alice Michael [Wed, 17 Apr 2019 22:17:33 +0000 (15:17 -0700)]
iavf: rename iavf_status structure flags
rename the flags inside of iavf_status from I40E_*
to IAVF_*
Signed-off-by: Alice Michael <alice.michael@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Alice Michael [Wed, 17 Apr 2019 22:17:32 +0000 (15:17 -0700)]
iavf: replace i40e variables with iavf
Update the old variables and flags marked as i40e to match
the iavf name of the driver.
Signed-off-by: Alice Michael <alice.michael@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Alice Michael [Wed, 17 Apr 2019 22:17:31 +0000 (15:17 -0700)]
iavf: rename i40e functions to be iavf
Update the old i40e function names to be iavf
Signed-off-by: Alice Michael <alice.michael@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Sergey Nemov [Wed, 17 Apr 2019 22:17:30 +0000 (15:17 -0700)]
iavf: change iavf_status_code to iavf_status
Instead of typedefing the enum iavf_status_code with iavf_status,
just shorten the enum itself and get rid of typedef.
Signed-off-by: Sergey Nemov <sergey.nemov@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Alice Michael [Wed, 17 Apr 2019 22:17:29 +0000 (15:17 -0700)]
iavf: Rename i40e_adminq* files to iavf_adminq*
With the rename of the iavf driver, there were some
files that were missed in renaming. Update these to
be iavf as well.
Signed-off-by: Alice Michael <alice.michael@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Gustavo A. R. Silva [Fri, 29 Mar 2019 23:38:50 +0000 (16:38 -0700)]
iavf: iavf_client: use struct_size() helper
One of the more common cases of allocation size calculations is finding
the size of a structure that has a zero-sized array at the end, along
with memory for some number of elements for that array. For example:
struct foo {
int stuff;
struct boo entry[];
};
size = sizeof(struct foo) + count * sizeof(struct boo);
Instead of leaving these open-coded and prone to type mistakes, we can
now use the new struct_size() helper:
size = struct_size(instance, entry, count);
This code was detected with the help of Coccinelle.
Signed-off-by: "Gustavo A. R. Silva" <gustavo@embeddedor.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Gustavo A. R. Silva [Fri, 29 Mar 2019 23:38:48 +0000 (16:38 -0700)]
iavf: use struct_size() in kzalloc()
One of the more common cases of allocation size calculations is finding
the size of a structure that has a zero-sized array at the end, along
with memory for some number of elements for that array. For example:
struct foo {
int stuff;
struct boo entry[];
};
size = sizeof(struct foo) + count * sizeof(struct boo);
instance = kzalloc(size, GFP_KERNEL)
Instead of leaving these open-coded and prone to type mistakes, we can
now use the new struct_size() helper:
instance = kzalloc(struct_size(instance, entry, count), GFP_KERNEL)
Notice that, in this case, variable bufsz is not necessary, hence it
is removed.
This code was detected with the help of Coccinelle.
Signed-off-by: "Gustavo A. R. Silva" <gustavo@embeddedor.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Aleksandr Loktionov [Fri, 29 Mar 2019 19:48:50 +0000 (12:48 -0700)]
iavf: Limiting RSS queues to CPUs
Limiting RSS queues number to online CPUs number in order to
avoid issues with creating misconfigured RSS queues.
Signed-off-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Nathan Chancellor [Thu, 10 Jan 2019 04:21:57 +0000 (21:21 -0700)]
iavf: Use printf instead of gnu_printf for iavf_debug_d
Clang warns:
In file included from drivers/net/ethernet/intel/iavf/iavf_main.c:4:
In file included from drivers/net/ethernet/intel/iavf/iavf.h:37:
In file included from drivers/net/ethernet/intel/iavf/iavf_type.h:8:
drivers/net/ethernet/intel/iavf/iavf_osdep.h:49:18: warning: 'format' attribute argument not supported: gnu_printf [-Wignored-attributes]
__attribute__ ((format(gnu_printf, 3, 4)));
^
1 warning generated.
We can convert from gnu_printf to printf without any side effects for
two reasons:
1. All iavf_debug instances use standard printf formats, as pointed out
by Miguel Ojeda at the below link, meaning gnu_printf is not strictly
required.
2. However, GCC has aliased printf to gnu_printf on Linux since at least
2010 based on git history.
From gcc/c-family/c-format.c:
/* Attributes such as "printf" are equivalent to those such as
"gnu_printf" unless this is overridden by a target. */
static const target_ovr_attr gnu_target_overrides_format_attributes[] =
{
{ "gnu_printf", "printf" },
{ "gnu_scanf", "scanf" },
{ "gnu_strftime", "strftime" },
{ "gnu_strfmon", "strfmon" },
{ NULL, NULL }
};
The mentioned override only happens on Windows (mingw32). Changing from
gnu_printf to printf is a no-op for GCC and stops Clang from warning.
Link: https://github.com/ClangBuiltLinux/linux/issues/111
Suggested-by: Miguel Ojeda <miguel.ojeda.sandonis@gmail.com>
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Linus Torvalds [Fri, 31 May 2019 04:11:22 +0000 (21:11 -0700)]
Merge git://git./linux/kernel/git/davem/net
Pull networking fixes from David Miller:
1) Fix OOPS during nf_tables rule dump, from Florian Westphal.
2) Use after free in ip_vs_in, from Yue Haibing.
3) Fix various kTLS bugs (NULL deref during device removal resync,
netdev notification ignoring, etc.) From Jakub Kicinski.
4) Fix ipv6 redirects with VRF, from David Ahern.
5) Memory leak fix in igmpv3_del_delrec(), from Eric Dumazet.
6) Missing memory allocation failure check in ip6_ra_control(), from
Gen Zhang. And likewise fix ip_ra_control().
7) TX clean budget logic error in aquantia, from Igor Russkikh.
8) SKB leak in llc_build_and_send_ui_pkt(), from Eric Dumazet.
9) Double frees in mlx5, from Parav Pandit.
10) Fix lost MAC address in r8169 during PCI D3, from Heiner Kallweit.
11) Fix botched register access in mvpp2, from Antoine Tenart.
12) Use after free in napi_gro_frags(), from Eric Dumazet.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (89 commits)
net: correct zerocopy refcnt with udp MSG_MORE
ethtool: Check for vlan etype or vlan tci when parsing flow_rule
net: don't clear sock->sk early to avoid trouble in strparser
net-gro: fix use-after-free read in napi_gro_frags()
net: dsa: tag_8021q: Create a stable binary format
net: dsa: tag_8021q: Change order of rx_vid setup
net: mvpp2: fix bad MVPP2_TXQ_SCHED_TOKEN_CNTR_REG queue value
ipv4: tcp_input: fix stack out of bounds when parsing TCP options.
mlxsw: spectrum: Prevent force of 56G
mlxsw: spectrum_acl: Avoid warning after identical rules insertion
net: dsa: mv88e6xxx: fix handling of upper half of STATS_TYPE_PORT
r8169: fix MAC address being lost in PCI D3
net: core: support XDP generic on stacked devices.
netvsc: unshare skb in VF rx handler
udp: Avoid post-GRO UDP checksum recalculation
net: phy: dp83867: Set up RGMII TX delay
net: phy: dp83867: do not call config_init twice
net: phy: dp83867: increase SGMII autoneg timer duration
net: phy: dp83867: fix speed 10 in sgmii mode
net: phy: marvell10g: report if the PHY fails to boot firmware
...
Linus Torvalds [Fri, 31 May 2019 04:05:23 +0000 (21:05 -0700)]
Merge tag 'arm64-fixes' of git://git./linux/kernel/git/arm64/linux
Pull arm64 fixes from Will Deacon:
"The fixes are still trickling in for arm64, but the only really
significant one here is actually fixing a regression in the botched
module relocation range checking merged for -rc2.
Hopefully we've nailed it this time.
- Fix implementation of our set_personality() system call, which
wasn't being wrapped properly
- Fix system call function types to keep CFI happy
- Fix siginfo layout when delivering SIGKILL after a kernel fault
- Really fix module relocation range checking"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: use the correct function type for __arm64_sys_ni_syscall
arm64: use the correct function type in SYSCALL_DEFINE0
arm64: fix syscall_fn_t type
signal/arm64: Use force_sig not force_sig_fault for SIGKILL
arm64/module: revert to unsigned interpretation of ABS16/32 relocations
arm64: Fix the arm64_personality() syscall wrapper redirection
Linus Torvalds [Fri, 31 May 2019 03:52:40 +0000 (20:52 -0700)]
Merge tag 'for-5.2-rc2-tag' of git://git./linux/kernel/git/kdave/linux
Pull btrfs fixes from David Sterba:
"A few more fixes for bugs reported by users, fuzzing tools and
regressions:
- fix crashes in relocation:
+ resuming interrupted balance operation does not properly clean
up orphan trees
+ with enabled qgroups, resuming needs to be more careful about
block groups due to limited context when updating qgroups
- fsync and logging fixes found by fuzzing
- incremental send fixes for no-holes and clone
- fix spin lock type used in timer function for zstd"
* tag 'for-5.2-rc2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
Btrfs: fix race updating log root item during fsync
Btrfs: fix wrong ctime and mtime of a directory after log replay
Btrfs: fix fsync not persisting changed attributes of a directory
btrfs: qgroup: Check bg while resuming relocation to avoid NULL pointer dereference
btrfs: reloc: Also queue orphan reloc tree for cleanup to avoid BUG_ON()
Btrfs: incremental send, fix emission of invalid clone operations
Btrfs: incremental send, fix file corruption when no-holes feature is enabled
btrfs: correct zstd workspace manager lock to use spin_lock_bh()
btrfs: Ensure replaced device doesn't have pending chunk allocation
Linus Torvalds [Fri, 31 May 2019 03:35:48 +0000 (20:35 -0700)]
Merge tag 'configfs-for-5.2-2' of git://git.infradead.org/users/hch/configfs
Pull configs fix from Christoph Hellwig:
- fix a use after free in configfs_d_iput (Sahitya Tummala)
* tag 'configfs-for-5.2-2' of git://git.infradead.org/users/hch/configfs:
configfs: Fix use-after-free when accessing sd->s_dentry
Linus Torvalds [Fri, 31 May 2019 02:58:59 +0000 (19:58 -0700)]
Merge tag 'sound-5.2-rc3' of git://git./linux/kernel/git/tiwai/sound
Pull sound fixes from Takashi Iwai:
"No big surprises here, just a few device-specific fixes.
HD-audio received several fixes for Acer, Dell, Huawei and other
laptops as well as the workaround for the new Intel chipset. One
significant one-liner fix is the disablement of the node-power saving
on Realtek codecs, which may potentially cover annoying bugs like the
background noises or click noises on many devices.
Other than that, a fix for FireWire bit definitions, and another fix
for LINE6 USB audio bug that was discovered by syzkaller"
* tag 'sound-5.2-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: fireface: Use ULL suffixes for 64-bit constants
ALSA: hda/realtek - Improve the headset mic for Acer Aspire laptops
ALSA: line6: Assure canceling delayed work at disconnection
ALSA: hda - Force polling mode on CNL for fixing codec communication
ALSA: hda/realtek - Enable micmute LED for Huawei laptops
ALSA: hda/realtek - Set default power save node to 0
ALSA: hda/realtek - Check headset type by unplug and resume
Linus Torvalds [Thu, 30 May 2019 23:33:37 +0000 (16:33 -0700)]
Merge tag 'clk-fixes-for-linus' of git://git./linux/kernel/git/clk/linux
Pull clk driver fixes from Stephen Boyd:
- Don't expose the SiFive clk driver on non-RISCV architectures
- Fix some bits describing clks in the imx8mm driver
- Always call clk domain code in the TI driver so non-legacy platforms
work
* tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
clk: ti: clkctrl: Fix clkdm_clk handling
clk: imx: imx8mm: fix int pll clk gate
clk: sifive: restrict Kconfig scope for the FU540 PRCI driver
Willem de Bruijn [Thu, 30 May 2019 22:01:21 +0000 (18:01 -0400)]
net: correct zerocopy refcnt with udp MSG_MORE
TCP zerocopy takes a uarg reference for every skb, plus one for the
tcp_sendmsg_locked datapath temporarily, to avoid reaching refcnt zero
as it builds, sends and frees skbs inside its inner loop.
UDP and RAW zerocopy do not send inside the inner loop so do not need
the extra sock_zerocopy_get + sock_zerocopy_put pair. Commit
52900d22288ed ("udp: elide zerocopy operation in hot path") introduced
extra_uref to pass the initial reference taken in sock_zerocopy_alloc
to the first generated skb.
But, sock_zerocopy_realloc takes this extra reference at the start of
every call. With MSG_MORE, no new skb may be generated to attach the
extra_uref to, so refcnt is incorrectly 2 with only one skb.
Do not take the extra ref if uarg && !tcp, which implies MSG_MORE.
Update extra_uref accordingly.
This conditional assignment triggers a false positive may be used
uninitialized warning, so have to initialize extra_uref at define.
Changes v1->v2: fix typo in Fixes SHA1
Fixes: 52900d22288e7 ("udp: elide zerocopy operation in hot path")
Reported-by: syzbot <syzkaller@googlegroups.com>
Diagnosed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 30 May 2019 22:17:05 +0000 (15:17 -0700)]
Merge branch '100GbE' of git://git./linux/kernel/git/jkirsher/next-queue
Jeff Kirsher says:
====================
100GbE Intel Wired LAN Driver Updates 2019-05-30
This series contains updates to ice driver only.
Brett continues his work with interrupt handling by fixing an issue
where were writing to the incorrect register to disable all VF
interrupts.
Tony consolidates the unicast and multicast MAC filters into a single
new function.
Anirudh adds support for virtual channel vector mapping to receive and
transmit queues. This uses a bitmap to associate indicated queues with
the specified vector. Makes several cosmetic code cleanups, as well as
update the driver to align with the current specification for managing
MAC operation codes (opcodes).
Paul adds support for Forward Error Correction (FEC) and also adds the
ethtool get and set handlers to modify FEC parameters.
Bruce cleans up the driver code to fix a number of issues, such as,
reducing the scope of some local variables, reduce the number of
de-references by changing a local variable and reorder the code to
remove unnecessary "goto's".
Dave adds switch rules to be able to handle LLDP packets and in the
process, fix a couple of issues found, like stop treating DCBx state of
"not started" as an error and stop hard coding the filter information
flag to transmit.
Jacob updates the driver to allow for more granular debugging by
developers by using a distinct separate bit for dumping firmware logs.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>