Phil Sutter [Wed, 29 Nov 2017 17:34:09 +0000 (18:34 +0100)]
man: tc-csum.8: Fix inconsistency in example description
Commit 6bbe5e6290db5 ("man: tc-csum.8: Fix example") changed both source
and destination IP addresses in example code but missed to update the
example's description accordingly.
Fixes: 6bbe5e6290db5 ("man: tc-csum.8: Fix example") Signed-off-by: Phil Sutter <phil@nwl.cc>
Robert Shearman [Tue, 28 Nov 2017 11:16:50 +0000 (11:16 +0000)]
vxlan: Make id optional when modifying a link
Specifying the IFLA_VXLAN_LINK attribute on a vxlan link modify is
optional in the kernel, so make the id argument optional for "ip link
set ..." to avoid a user needing to specify it when changing another
attribute.
Robert Shearman [Tue, 28 Nov 2017 11:16:21 +0000 (11:16 +0000)]
gre: Fix ttl inherit option
Specifying "... ttl inherit" currently does nothing on a GRE link
modify since the previous ttl value is retrieved up front. Fix this by
explicitly setting ttl to 0 when "inherit" is specified for the
option, since 0 represents the semantics of inherit.
Jiri Pirko [Sat, 25 Nov 2017 10:07:57 +0000 (11:07 +0100)]
tc: remove action cookie len from printout
Make the output same as input and avoid printout of unnecessary len.
Suggested-by: Stephen Hemminger <stephen@networkplumber.org> Fixes: fd8b3d2c1b9b ("actions: Add support for user cookies") Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Jakub Kicinski [Fri, 24 Nov 2017 02:12:07 +0000 (18:12 -0800)]
f_bpf: communicate ifindex for eBPF offload
Split parsing and loading of the eBPF program and if skip_sw is set
load the program for ifindex, to which the qdisc is attached.
Note that the ifindex will be ignored for programs which are already
loaded (e.g. when using pinned programs), but in that case we just
trust the user knows what he's doing. Hopefully we will get extack
soon in the driver to help debugging this case.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Jakub Kicinski [Fri, 24 Nov 2017 02:12:06 +0000 (18:12 -0800)]
tc_filter: resolve device name before parsing filter
Move resolving device name into an ifindex before calling filter
specific callbacks. This way if filters need the ifindex, they
can read it from the request.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Both BPF filter and action will allow users to specify run
multiple times, and only the last one will be considered by
the kernel. Explicitly refuse such command lines.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Jakub Kicinski [Fri, 24 Nov 2017 02:12:01 +0000 (18:12 -0800)]
bpf: split parse from program loading
Parsing command line is currently done together with potentially
loading a new eBPF program. This makes it more difficult to
provide additional parameters for loading (which may come after
the eBPF program info on the command line).
Split the two (only internally for now). Verbose parameter
has to be saved in struct bpf_cfg_in to be carried between
the stages.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Jakub Kicinski [Fri, 24 Nov 2017 02:12:00 +0000 (18:12 -0800)]
bpf: allocate opcode table in struct bpf_cfg_in
struct bpf_cfg_in already carries a pointer to sock_filter ops.
It's currently set to a local variable in bpf_parse_opt_tbl(),
shared between parsing and loading stages. Move the array
entirely to struct bpf_cfg_in, this will allow us to split
parsing and loading.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Jakub Kicinski [Fri, 24 Nov 2017 02:11:59 +0000 (18:11 -0800)]
bpf: keep parsed program mode in struct bpf_cfg_in
bpf_parse() will parse command line arguments to find out the
program mode. This mode will later be needed at loading time.
Instead of keeping it locally add it to struct bpf_cfg_in,
this will allow splitting parsing and loading stages.
enum bpf_mode has to be moved to the header file, because C
doesn't allow forward declaration of enums.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Jakub Kicinski [Fri, 24 Nov 2017 02:11:58 +0000 (18:11 -0800)]
bpf: pass program type in struct bpf_cfg_in
Program type is needed both for parsing and loading of
the program. Parsing may also induce the type based on
signatures from __bpf_prog_meta. Instead of passing
the type around keep it in struct bpf_cfg_in.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net>
This patch adapts the tc command line interface to allow bandwidth limits
to be specified as a percentage of the interface's capacity.
Adding this functionality requires passing the specified device string to
each class/qdisc which changes the prototype for a couple of functions: the
.parse_qopt and .parse_copt interfaces. The device string is a required
parameter for tc-qdisc and tc-class, and when not specified, the kernel
returns ENODEV. In this patch, if the user tries to specify a bandwidth
percentage without naming the device, we return an error from userspace.
Tom Herbert [Wed, 22 Nov 2017 20:05:35 +0000 (12:05 -0800)]
ila: support to configure checksum neutral-map-auto
Configuration support in both ip ila and ip LWT for checksum
neutral-map-auto. This is a mode of ILA where checksum
neutral mapping is assumed for packets (there is no C-bit
in the identifier to indicate checksum neutral).
Jakub Kicinski [Thu, 23 Nov 2017 01:00:53 +0000 (17:00 -0800)]
bpf: initialize the verifier log
If program loading fails before verifier prints its first
message, the verifier log will not be initialized. Always
set the first character of the log buffer to zero to make
sure we don't dump non-printable characters to the terminal.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Lorenzo Colitti [Mon, 20 Nov 2017 03:57:07 +0000 (12:57 +0900)]
iproute2: fixes to compile on some systems.
1. Put the declarations of strlcpy and strlcat inside
an #ifdef NEED_STRLCPY. Their declarations were already in a
similar #ifdef.
2. In bpf_scm.h, include sys/un.h for struct sockaddr_un.
3. In utils.h, include time.h for struct timeval.
Tested: builds on ubuntu 14.04 with "make clean distclean; ./configure && make -j64"
Tested: 4.14.1 builds on Android with Android-specific #ifndefs for missing library code Signed-off-by: Lorenzo Colitti <lorenzo@google.com>
Phil Sutter [Wed, 15 Nov 2017 14:01:31 +0000 (15:01 +0100)]
tc_util: Silence spurious compiler warning
GCC version 7.2.1 complains that 'result1' may be used uninitialized in
parse_action_control_slash_spaces(). This should not be possible in
practice, so the actual value 'result1' is initialized with does not
matter.
Phil Sutter [Wed, 15 Nov 2017 14:01:30 +0000 (15:01 +0100)]
tc_util: Drop needless pointer check
The function parse_action_control_slash() returns early if 'p' is NULL,
so after the first call to action_a2n(), 'p' is guaranteed not to be
NULL. Otherwise, the assignment '*p = 0' above would dereference the
NULL pointer already anyway, so just drop this check here.
Jon Maloy [Wed, 15 Nov 2017 16:25:44 +0000 (17:25 +0100)]
tipc: change family attribute from u32 to u16
commit 28033ae4e0f ("net: netlink: Update attr validation to require
exact length for some types") introduces a stricter control on attributes
of type NLA_U* and NLA_S*.
Since the tipc tool is sending a family attribute of u32 instead of as
expected u16 the tool is now effectively broken.
We fix this by changing the type of the said attribute.
Leon Romanovsky [Mon, 13 Nov 2017 10:21:19 +0000 (12:21 +0200)]
ip: Fix compilation break on old systems
As was reported [1], the iproute2 fails to compile on old systems,
in Cong's case, it was Fedora 19, in our case it was RedHat 7.2, which
failed with the following errors during compilation:
ipxfrm.c: In function ‘xfrm_selector_print’:
ipxfrm.c:479:7: error: ‘IPPROTO_MH’ undeclared (first use in this
function)
case IPPROTO_MH:
^
ipxfrm.c:479:7: note: each undeclared identifier is reported only once
for each function it appears in
ipxfrm.c: In function ‘xfrm_selector_upspec_parse’:
ipxfrm.c:1345:8: error: ‘IPPROTO_MH’ undeclared (first use in this
function)
case IPPROTO_MH:
^ make[1]: *** [ipxfrm.o] Error 1
The reason to it is the order of headers files. The IPPROTO_MH field is
set in kernel's UAPI header file (in6.h), but only in case
__UAPI_DEF_IPPROTO_V6 is set before. That define comes from other kernel's
header file (libc-compat.h) and is set in case there are no previous
libc relevant declarations.
In ip code, the include of <netdb.h> causes to indirect inclusion of
<netinet/in.h> and it sets __UAPI_DEF_IPPROTO_V6 to be zero and prevents from
IPPROTO_MH declaration.
This patch takes the simplest possible approach to fix the compilation
error by checking if IPPROTO_MH was defined before and in case it
wasn't, it defines it to be the same as in the kernel.
Ivan Vecera [Fri, 10 Nov 2017 06:20:13 +0000 (07:20 +0100)]
lib: make resolve_hosts variable common
Any iproute utility that uses any function from lib/utils.c needs
to declare its own resolve_hosts variable instance although it does
not need/use hostname resolving functionality (currently only 'ip'
and 'ss' commands uses this).
The patch declares single common instance of resolve_hosts directly
in utils.c so the existing ones can be removed (the same approach
that is used for timestamp_short).
In order to calculate the idleSlope parameter of CBS correctly, users
must take into account the entire packet size, including the overhead
from all layers.
Add some more details to the man page to clarify that, giving one
simple example and pointing users to the correct 802.1Q section for
further clarifications if needed.
Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
David Ahern [Thu, 9 Nov 2017 00:46:50 +0000 (09:46 +0900)]
libnetlink: Handle extack messages for non-error case
Kernel can now return non-fatal error messages in extack facility.
Update iproute2 to dump to use if present.
- rename nl_dump_ext_err to nl_dump_ext_ack
- rename errmsg to msg
- add call to nl_dump_ext_ack in rtnl_dump_done and __rtnl_talk for
non-error path
Signed-off-by: David Ahern <dsahern@gmail.com> Tested-by: Ido Schimmel <idosch@mellanox.com>
Thomas Egerer [Mon, 30 Oct 2017 18:11:46 +0000 (19:11 +0100)]
xfrm_{state, policy}: Allow to deleteall polices/states with marks
Using 'ip deleteall' with policies that have marks, fails unless you
eplicitely specify the mark values. This is very uncomfortable when
bulk-deleting policies and states. With this patch all relevant states
and policies are wiped by 'ip deleteall' regardless of their mark
values.
Signed-off-by: Thomas Egerer <thomas.egerer@secunet.com>
Thomas Egerer [Mon, 30 Oct 2017 18:11:45 +0000 (19:11 +0100)]
xfrm_policy: Do not attempt to deleteall a socket policy
Socket polices are added to a socket using setsockopt(2). They cannot be
deleted by iproute2. The attempt to delete them causes an error
(EINVAL).
To avoid this unnecessary error message all socket policies are skipped
in xfrm_policy_keep.
Signed-off-by: Thomas Egerer <thomas.egerer@secunet.com>
Thomas Egerer [Mon, 30 Oct 2017 18:11:44 +0000 (19:11 +0100)]
xfrm_policy: Add filter option for socket policies
Listing policies on systems with a lot of socket policies can be
confusing due to the number of returned polices. Even if socket polices
are not of interest, they cannot be filtered. This patch adds an option
to filter all socket policies from the output.
Signed-off-by: Thomas Egerer <thomas.egerer@secunet.com>
Amritha Nambiar [Fri, 3 Nov 2017 08:54:01 +0000 (01:54 -0700)]
flower: Represent HW traffic classes as classid values
This patch was previously submitted as RFC. Submitting this as
non-RFC now that the classid reservation scheme for hardware
traffic classes and offloads to route packets to a hardware
traffic class are accepted in net-next.
HW traffic classes 0 through 15 are represented using the
reserved classid values :ffe0 - :ffef.
Example:
Match Dst IPv4,Dst Port and route to TC1:
# tc filter add dev eth0 protocol ip parent ffff:\
prio 1 flower dst_ip 192.168.1.1/32\
ip_proto udp dst_port 12000 skip_sw\
hw_tc 1
The Credit Based Shaper (CBS) queueing discipline allows bandwidth
reservation with sub-milisecond precision. It is defined by the
802.1Q-2014 specification (section 8.6.8.2 and Annex L).
The syntax is:
tc qdisc add dev DEV parent NODE cbs locredit <LOCREDIT>
hicredit <HICREDIT> sendslope <SENDSLOPE>
idleslope <IDLESLOPE>
(The order is not important)
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Amritha Nambiar [Wed, 1 Nov 2017 07:45:42 +0000 (00:45 -0700)]
tc/mqprio: Offload mode and shaper options in mqprio
This patch was previously submitted as RFC. Submitting this as
non-RFC now that the tc/mqprio changes are accepted in net-next.
Adds new mqprio options for 'mode' and 'shaper'. The mode
option can take values for offload modes such as 'dcb' (default),
'channel' with the 'hw' option set to 1. The new 'channel' mode
supports offloading TCs and other queue configurations. The
'shaper' option is to support HW shapers ('dcb' default) and
takes the value 'bw_rlimit' for bandwidth rate limiting. The
parameters to the bw_rlimit shaper are minimum and maximum
bandwidth rates. New HW shapers in future can be supported
through the shaper attribute.
Mahesh Bandewar [Mon, 30 Oct 2017 20:57:51 +0000 (13:57 -0700)]
ip/ipvlan: enhance ability to add mode flags to existing modes
IPvlan supported bridge-only functionality prior to commits a190d04db937 ('ipvlan: introduce 'private' attribute for all
existing modes.') and fe89aa6b250c ('ipvlan: implement VEPA mode').
These two commits allow to configure the VEPA and private modes now.
This patch adds those options in ip command.
e.g.
bash:~# ip link add link eth0 name ipvl0 type ipvlan mode l2 private
-or-
bash:~# ip link add link eth0 type ipvl0 type ipvlan mode l2 vepa
Also the output will reflect the mode and the mode-flag accordingly.
e.g.
bash:~# ip -details link show ipvl0
4: ipvl0@eth0: <BROADCAST,MULTICAST> mtu 1500 qdisc ...
link/ether 00:1a:11:44:a5:3e brd ff:ff:ff:ff:ff:ff promiscuity 0
ipvlan mode l2 private addrgenmode eui64 numtxqueues 1 ...
Stefano Brivio [Tue, 31 Oct 2017 17:47:56 +0000 (18:47 +0100)]
ss: Fix width calculations when Netid or State columns are missing
If Netid or State columns are missing, we must not subtract one
for each of these two columns from the remaining screen width,
while distributing available space to columns. This one
character corresponding to one delimiting space has to be
subtracted only if the columns are actually printed.
Further, in the existing implementation, if the screen width is
an odd number, one additional character is added to the width of
one of the two columns.
But if both are not printed, this filling character needs to be
added somewhere else, in order to have the right spacing
allowing us to fill lines completely.
Address and port fields are printed in pairs (local and remote),
so we can't distribute the space to any of them, because it
would be doubled. Instead, print this additional space to the
right of the Send-Q column, to keep code changes to a minimum.
This is particularly visible with 'ss -f netlink -Z'. Before
this patch, with an 80 column terminal, we have:
$ ss -f netlink -Z|head -n3
Recv-Q Send-Q Local Address:Port Peer Address:Port
0 0 rtnl:evolution-calen/2049 * pro
c_ctx=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023
0 0 rtnl:clock-applet/1944 * pro
c_ctx=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023
After this patch, in both cases, the output is:
$ ss -f netlink -Z|head -n3
Recv-Q Send-Q Local Address:Port Peer Address:Port
0 0 rtnl:evolution-calen/2049 *
proc_ctx=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023
0 0 rtnl:clock-applet/1944 *
proc_ctx=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023
Stefano Brivio [Tue, 31 Oct 2017 17:47:54 +0000 (18:47 +0100)]
ss: Remove useless width specifier in process context print
Both local address and service, and remote address and service
fields are already printed out in netlink_show_one() before we
start printing process context, by calling sock_addr_print()
twice.
At this point, sock_addr_print() has already forced the remote
service field to be 'serv_width' wide -- that is, 'serv_width'
width has already been consumed, before we print process
context.
Hence, it makes no sense to force the display width of process
context to be 'serv_width' wide again: previous prints have
filled up the line already. Remove the width specifier and
prefix with a space instead, to keep this consistent with fields
which are displayed after the first output line.
Roman Mashak [Tue, 31 Oct 2017 18:24:19 +0000 (14:24 -0400)]
ip netns: use strtol() instead of atoi()
Use strtol-based API to parse and validate integer input; atoi() does
not detect errors and may yield undefined behaviour if result can't be
represented.
v2: use get_unsigned() since network namespace is really an unsigned value.
Roopa Prabhu [Sat, 28 Oct 2017 05:13:49 +0000 (22:13 -0700)]
iplink: bridge: support bridge port vlan_tunnel attribute
This config maps to IFLA_BRPORT_VLAN_TUNNEL bridge port netlink
flag attribute. This flag enables vlan to tunnel mapping on a bridge
port. It is off by default.
set vlan_tunnel attribute on bridge port vxlan0:
$ip link set dev vxlan0 type bridge_slave vlan_tunnel on
$ip link set dev vxlan0 type bridge_slave vlan_tunnel off
or via bridge command
$bridge link set dev vxlan0 vlan_tunnel on
$bridge link set dev vxlan0 vlan_tunnel off