]>
git.ipfire.org Git - thirdparty/iproute2.git/log
Matthieu Baerts [Wed, 23 Aug 2023 07:24:08 +0000 (09:24 +0200)]
ss: mptcp: print missing info counters
These new counters have been added in different kernel versions:
- v5.12: local_addr_used, local_addr_max
- v5.13: csum_enabled
- v6.5: retransmits, bytes_retrans, bytes_sent, bytes_received,
bytes_acked
It is interesting to display them if they are available.
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/415
Acked-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Andrea Claudi <aclaudi@redhat.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Matthieu Baerts [Wed, 23 Aug 2023 07:24:07 +0000 (09:24 +0200)]
ss: mptcp: display seq related counters as decimal
This is aligned with what is printed for TCP sockets.
The main difference here is that these counters can be larger (u32 vs
u64) but WireShark and TCPDump are also printing these MPTCP counters as
decimal and they look fine.
So it sounds better to do the same here with ss for those who want to
easily count how many bytes have been exchanged between two runs without
having to think in hexa.
Acked-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Andrea Claudi <aclaudi@redhat.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Matthieu Baerts [Wed, 23 Aug 2023 07:24:06 +0000 (09:24 +0200)]
ss: mptcp: display info counters as unsigned
Some counters from mptcp_info structure were stored as an unsigned
number (u8) but displayed as a signed one.
Even if it is unlikely these u8 counters -- number of subflows and
ADD_ADDR -- have a value bigger than 2^7, it still sounds better to
display them as unsigned.
Fixes: 9c3be2c0 ("ss: mptcp: add msk diag interface support")
Acked-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Andrea Claudi <aclaudi@redhat.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Maximilian Bosch [Tue, 22 Aug 2023 12:33:07 +0000 (14:33 +0200)]
ip-vrf: recommend using CAP_BPF rather than CAP_SYS_ADMIN
The CAP_SYS_ADMIN capability allows far too much, to quote
`capabilities(7)`:
Note: this capability is overloaded; see Notes to kernel developers, below.
In the case of `ip-vrf(8)` this is needed to load a BPF program.
According to the same section of the same man-page, using `CAP_BPF` is
preferred if that's the reason for `CAP_SYS_ADMIN`;
perform the same BPF operations as are governed by CAP_BPF (but the latter, weaker capability is preferred for accessing
that functionality).
Local testing revealed that `ip vrf exec` for an unprivileged user is
sufficient if the `CAP_BPF` capability is given rather than
`CAP_SYS_ADMIN`.
In a previous version of the patch[1] it was mentioned that
CAP_SYS_ADMIN was still required for Linux <5.8, however it was
suggested to not make man-pages dependent on the kernel version. Also,
it was suggested to improve the wording and the formatting of the entire
paragraph mentioning capabilities which was also done.
Signed-off-by: Maximilian Bosch <maximilian@mbosch.me>
[1] https://lore.kernel.org/netdev/e6t4ucjdrcitzneh2imygsaxyb2aasxfn2q2a4zh5yqdx3vold@kutwh5kwixva/T/#m628a1900a7e5012bb87e6cb3c94af6c7281cf2bf
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Phil Sutter [Tue, 22 Aug 2023 12:19:16 +0000 (14:19 +0200)]
ss: Fix socket type check in packet_show_line()
The field is accessed before being assigned a meaningful value,
effectively disabling the checks.
Fixes: 4a0053b606a34 ("ss: Unify packet stats output from netlink and proc")
Signed-off-by: Phil Sutter <phil@nwl.cc>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Jiri Pirko [Mon, 14 Aug 2023 07:29:01 +0000 (09:29 +0200)]
devlink: spell out STATE in devlink port function help
Be in-sync with port help and port man page and spell out the possible
states instead of "STATE".
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Vladimir Oltean [Mon, 7 Aug 2023 22:09:36 +0000 (01:09 +0300)]
tc/taprio: fix JSON output when TCA_TAPRIO_ATTR_ADMIN_SCHED is present
When the kernel reports that a configuration change is pending
(and that the schedule is still in the administrative state and
not yet operational), we (tc -j -p qdisc show) produce the following
output:
[ {
"kind": "taprio",
"handle": "8001:",
"root": true,
"refcnt": 9,
"options": {
"tc": 8,
"map": [ 0,1,2,3,4,5,6,7,0,0,0,0,0,0,0,0 ],
"queues": [ {
"offset": 0,
"count": 1
},{
"offset": 1,
"count": 1
},{
"offset": 2,
"count": 1
},{
"offset": 3,
"count": 1
},{
"offset": 4,
"count": 1
},{
"offset": 5,
"count": 1
},{
"offset": 6,
"count": 1
},{
"offset": 7,
"count": 1
} ],
"clockid": "TAI",
"base_time": 0,
"cycle_time":
20000000 ,
"cycle_time_extension": 0,
"schedule": [ {
"index": 0,
"cmd": "S",
"gatemask": "0xff",
"interval":
20000000
} ],{
"base_time":
1691160103110424418 ,
"cycle_time":
20000000 ,
"cycle_time_extension": 0,
"schedule": [ {
"index": 0,
"cmd": "S",
"gatemask": "0xff",
"interval":
20000000
} ]
},
"max-sdu": [ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 ],
"fp": [ "E","E","E","E","E","E","E","E","E","E","E","E","E","E","E","E" ]
}
} ]
which is invalid json, because the second group of "base_time",
"cycle_time", etc etc is placed in an unlabeled sub-object. If we pipe
it into jq, it complains:
parse error: Objects must consist of key:value pairs at line 53, column 14
Since it represents the administrative schedule, give this unnamed JSON
object the "admin" name. We now print valid JSON which looks like this:
[ {
"kind": "taprio",
"handle": "8001:",
"root": true,
"refcnt": 9,
"options": {
"tc": 8,
"map": [ 0,1,2,3,4,5,6,7,0,0,0,0,0,0,0,0 ],
"queues": [ {
"offset": 0,
"count": 1
},{
"offset": 1,
"count": 1
},{
"offset": 2,
"count": 1
},{
"offset": 3,
"count": 1
},{
"offset": 4,
"count": 1
},{
"offset": 5,
"count": 1
},{
"offset": 6,
"count": 1
},{
"offset": 7,
"count": 1
} ],
"clockid": "TAI",
"base_time": 0,
"cycle_time":
20000000 ,
"cycle_time_extension": 0,
"schedule": [ {
"index": 0,
"cmd": "S",
"gatemask": "0xff",
"interval":
20000000
} ],
"admin": {
"base_time":
1691160511783528178 ,
"cycle_time":
20000000 ,
"cycle_time_extension": 0,
"schedule": [ {
"index": 0,
"cmd": "S",
"gatemask": "0xff",
"interval":
20000000
} ]
},
"max-sdu": [ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 ],
"fp": [ "E","E","E","E","E","E","E","E","E","E","E","E","E","E","E","E" ]
}
} ]
Fixes: 602fae856d80 ("taprio: Add support for changing schedules")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Vladimir Oltean [Mon, 7 Aug 2023 22:09:35 +0000 (01:09 +0300)]
tc/taprio: don't print netlink attributes which weren't reported by the kernel
When an admin schedule is pending and hasn't yet become operational, the
kernel will report only the parameters of the admin schedule in a nested
TCA_TAPRIO_ATTR_ADMIN_SCHED attribute.
However, we default to printing zeroes even for the parameters of the
operational base time, when that doesn't exist.
Fixes: 0dd16449356f ("tc: Add support for configuring the taprio scheduler")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Nicolas Escande [Fri, 4 Aug 2023 16:49:52 +0000 (18:49 +0200)]
man: bridge: update bridge link show
Add missing man page documentation for bridge link show features added in
commit
13a5d8fcb41b ("bridge: link: allow filtering on bridge name") and
commit
64108901b737 ("bridge: Add support for setting bridge port attributes")
Signed-off-by: Nicolas Escande <nico.escande@gmail.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Wed, 9 Aug 2023 20:21:20 +0000 (13:21 -0700)]
uapi: update headers
Based off ov 6.5-rc5
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Paolo Lungaroni [Mon, 31 Jul 2023 18:36:16 +0000 (20:36 +0200)]
seg6: man: ip-link.8: add description of NEXT-C-SID flavor for SRv6 End.X behavior
This patch extends the manpage by providing the description of NEXT-C-SID
support for the SRv6 End.X behavior as defined in RFC 8986 [1].
The code/logic required to handle the "flavors" framework has already been
merged into iproute2 by commit:
04a6b456bf74 ("seg6: add support for flavors in SRv6 End* behaviors").
Some examples:
ip -6 route add 2001:db8::1 encap seg6local action End.X nh6 fc00::1 flavors next-csid dev eth0
Standard Output:
ip -6 route show 2001:db8::1
2001:db8::1 encap seg6local action End.X nh6 fc00::1 flavors next-csid lblen 32 nflen 16 dev eth0 metric 1024 pref medium
JSON Output:
ip -6 -j -p route show 2001:db8::1
[ {
"dst": "2001:db8::1",
"encap": "seg6local",
"action": "End.X",
"nh6": "fc00::1",
"flavors": [ "next-csid" ],
"lblen": 32,
"nflen": 16,
"dev": "eth0",
"metric": 1024,
"flags": [ ],
"pref": "medium"
} ]
[1] - https://datatracker.ietf.org/doc/html/rfc8986
Signed-off-by: Paolo Lungaroni <paolo.lungaroni@uniroma2.it>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Jakub Kicinski [Mon, 31 Jul 2023 15:06:28 +0000 (08:06 -0700)]
ss: report when the RxNoPad optimization is set on TLS sockets
Similarly to RO ZC report when RxNoPad is set.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Jakub Kicinski [Mon, 31 Jul 2023 16:19:20 +0000 (09:19 -0700)]
ip: error out if iplink does not consume all options
dummy does not define .parse_opt, which make ip ignore all
trailing arguments, for example:
# ip link add type dummy a b c d e f name cheese
will work just fine (and won't call the device "cheese").
Error out in this case with a clear error message:
# ip link add type dummy a b c d e f name cheese
Garbage instead of arguments "a ...". Try "ip link help".
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Nicolas Escande [Wed, 26 Jul 2023 07:25:07 +0000 (09:25 +0200)]
bridge: link: allow filtering on bridge name
When using 'brige link show' we can either dump all links enslaved to any bridge
(called without arg ) or display a single link (called with dev arg).
However there is no way to dummp all links of a single bridge.
To do so, this adds new optional 'master XXX' arg to 'bridge link show' command.
usage: bridge link show master br0
Signed-off-by: Nicolas Escande <nico.escande@gmail.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Gioele Barabucci [Wed, 26 Jul 2023 06:14:09 +0000 (08:14 +0200)]
Read configuration files from /etc and /usr
Add support for the so called "stateless" configuration pattern (read
from /etc, fall back to /usr), giving system administrators a way to
define local configuration without changing any distro-provided files.
In practice this means that each configuration file FOO is loaded
from /usr/lib/iproute2/FOO unless /etc/iproute2/FOO exists.
Signed-off-by: Gioele Barabucci <gioele@svario.it>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Masatake YAMATO [Sun, 23 Jul 2023 16:42:57 +0000 (01:42 +0900)]
man: (ss) fix wrong margin
Signed-off-by: Masatake YAMATO <yamato@redhat.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Masatake YAMATO [Sun, 23 Jul 2023 16:42:56 +0000 (01:42 +0900)]
tc: fix a wrong file name in comment
Signed-off-by: Masatake YAMATO <yamato@redhat.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Trevor Gamblin [Thu, 20 Jul 2023 20:37:26 +0000 (16:37 -0400)]
bridge/mdb.c: include limits.h
While building iproute2 6.4.0 with musl using Yocto Project, errors such
as the following were encountered:
| mdb.c: In function 'mdb_parse_vni':
| mdb.c:666:47: error: 'ULONG_MAX' undeclared (first use in this function)
| 666 | if ((endptr && *endptr) || vni_num == ULONG_MAX)
| | ^~~~~~~~~
| mdb.c:666:47: note: 'ULONG_MAX' is defined in header '<limits.h>'; did you forget to '#include <limits.h>'?
Include limits.h in bridge/mdb.c to fix this issue. This change is based
on one in Alpine Linux, but the author there had no plans to submit:
https://git.alpinelinux.org/aports/commit/main/iproute2/include.patch?id=
bd46efb8a8da54948639cebcfa5b37bd608f1069
Signed-off-by: Trevor Gamblin <tgamblin@baylibre.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Chander Govindarajan [Tue, 25 Jul 2023 01:23:48 +0000 (18:23 -0700)]
misc/ifstat: fix incorrect output data in json mode
Due to this bug, in json mode (with the -j flag), the output was
always in absolute mode (as if passing in the -a flag) and not in
relative mode.
Signed-off-by: Chander Govindarajan <mail@chandergovind.org>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Sat, 22 Jul 2023 02:41:37 +0000 (19:41 -0700)]
Add missing SPDX headers
All headers and source in iproute2 should be using SPDX license info.
Add a couple that were missed, and take off boilerplate.
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Sat, 22 Jul 2023 02:38:52 +0000 (19:38 -0700)]
include: dual license the bpf helper includes
The files bpf_api.h and bpf_elf.h are useful for TC BPF programs
to use. And there is no requirement that those be GPL only;
we intend to allow BSD licensed BPF helpers as well.
This makes the file license same as libbpf.
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Chander Govindarajan [Mon, 17 Jul 2023 09:32:46 +0000 (15:02 +0530)]
misc/ifstat: fix incorrect output data in json mode
Due to this bug, in json mode (with the -j flag), the output was
always in absolute mode (as if passing in the -a flag) and not in
relative mode.
Signed-off-by: Chander Govindarajan <mail@chandergovind.org>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Ido Schimmel [Tue, 11 Jul 2023 06:59:03 +0000 (09:59 +0300)]
f_flower: Treat port 0 as valid
It is not currently possible to add a filter matching on port 0 despite
it being a valid port number. This is caused by cited commit which
treats a value of 0 as an indication that the port was not specified.
Instead of inferring that a port range was specified by checking that both
the minimum and the maximum ports are non-zero, simply add a boolean
argument to parse_range() and set it after parsing a port range.
Before:
# tc filter add dev swp1 ingress pref 1 proto ip flower ip_proto udp src_port 0 action pass
Illegal "src_port"
# tc filter add dev swp1 ingress pref 2 proto ip flower ip_proto udp dst_port 0 action pass
Illegal "dst_port"
# tc filter add dev swp1 ingress pref 3 proto ip flower ip_proto udp src_port 0-100 action pass
Illegal "src_port"
# tc filter add dev swp1 ingress pref 4 proto ip flower ip_proto udp dst_port 0-100 action pass
Illegal "dst_port"
After:
# tc filter add dev swp1 ingress pref 1 proto ip flower ip_proto udp src_port 0 action pass
# tc filter add dev swp1 ingress pref 2 proto ip flower ip_proto udp dst_port 0 action pass
# tc filter add dev swp1 ingress pref 3 proto ip flower ip_proto udp src_port 0-100 action pass
# tc filter add dev swp1 ingress pref 4 proto ip flower ip_proto udp dst_port 0-100 action pass
# tc filter show dev swp1 ingress | grep _port
src_port 0
dst_port 0
src_port 0-100
dst_port 0-100
Fixes: 767b6fd620dd ("tc: flower: fix port value truncation")
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Mon, 10 Jul 2023 17:00:07 +0000 (10:00 -0700)]
uapi: update headers to 6.5-rc1
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Vladimir Oltean [Wed, 5 Jul 2023 10:51:55 +0000 (13:51 +0300)]
tc/taprio: fix parsing of "fp" option when it doesn't appear last
When installing a Qdisc this way:
tc qdisc replace dev $ifname handle 8001: parent root stab overhead 24 taprio \
num_tc 8 \
map 0 1 2 3 4 5 6 7 \
queues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 \
base-time 0 \
sched-entry S 01 1216 \
sched-entry S fe 12368 \
fp P E E E E E E E \
flags 0x2
the parser will error out when it tries to parse the "fp" array and it
finds "flags" as one of the elements, expecting it to be one of "P" or
"E".
The way this is handled in the parsing of other array arguments of
variable size (max-sdu, map, queues etc) is to not fail, call PREV_ARG()
and attempt re-parsing the argument as something else. Do that for "fp"
as well.
Apparently mqprio handles this case correctly, so I must have forgotten
to apply the same treatment for taprio as well, during development.
Fixes: 5fbca3b469ec ("tc/taprio: add support for preemptible traffic classes")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Zahari Doychev [Thu, 29 Jun 2023 19:57:36 +0000 (21:57 +0200)]
f_flower: simplify cfm dump function
The standard print function can be used to print the cfm attributes in
both standard and json use cases. In this way no string buffer is needed
which simplifies the code.
Signed-off-by: Zahari Doychev <zdoychev@maxlinear.com>
Reviewed-by: Petr Machata <me@pmachata.org>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Wed, 28 Jun 2023 23:35:25 +0000 (16:35 -0700)]
ifstat: fix warning about conditional
Gcc with warnings enabled complains because the conditional.
if ((long)(a - b) < 0)
could be construed as never true. Change to simple comparison.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Wed, 28 Jun 2023 23:34:01 +0000 (16:34 -0700)]
ct: check for invalid proto
Previously since proto was __u8 an invalid proto would
be allowed. Gcc warns about never true conditional
since __u8 can never be negative.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Wed, 28 Jun 2023 23:33:10 +0000 (16:33 -0700)]
ss: fix warning about empty if()
With all warnings enabled gcc wants brackets around the
empty if() clause. "Yes I really want an empty clause"
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Wed, 28 Jun 2023 23:32:22 +0000 (16:32 -0700)]
fix fallthrough warnings
In lib/utils.c comment for fallthrough was in wrong place
and one was missing in xfrm_state.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Wed, 28 Jun 2023 23:31:33 +0000 (16:31 -0700)]
dcb: fully initialize flag table
And make the flag table const since only used for lookup.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Wed, 28 Jun 2023 15:20:57 +0000 (08:20 -0700)]
Merge git://git.kernel.org/pub/scm/network/iproute2/iproute2-next
Stephen Hemminger [Mon, 26 Jun 2023 16:12:23 +0000 (09:12 -0700)]
v6.4.0
Luca Boccassi [Thu, 15 Jun 2023 01:06:59 +0000 (02:06 +0100)]
man: fix typos found by Lintian
Signed-off-by: Luca Boccassi <bluca@debian.org>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Zahari Doychev [Wed, 21 Jun 2023 20:55:45 +0000 (22:55 +0200)]
f_flower: add cfm support
Add support for matching on CFM Maintenance Domain level and opcode.
# tc filter add dev ens6 ingress pref 1 proto cfm \
flower cfm op 1 mdl 5 action ok
# tc filter show dev ens6 ingress
filter protocol cfm pref 1 flower chain 0
filter protocol cfm pref 1 flower chain 0 handle 0x1
eth_type 8902
cfm mdl 5 op 1
not_in_hw
action order 1: gact action pass
random type none pass val 0
index 1 ref 1 bind 1
# tc -j -p filter show dev ens6 ingress
[ {
"protocol": "cfm",
"pref": 1,
"kind": "flower",
"chain": 0
},{
"protocol": "cfm",
"pref": 1,
"kind": "flower",
"chain": 0,
"options": {
"handle": 1,
"keys": {
"eth_type": "8902",
"cfm": {
"mdl": 5,
"op": 1
}
},
"not_in_hw": true,
"actions": [ {
"order": 1,
"kind": "gact",
"control_action": {
"type": "pass"
},
"prob": {
"random_type": "none",
"control_action": {
"type": "pass"
},
"val": 0
},
"index": 1,
"ref": 1,
"bind": 1
} ]
}
} ]
Signed-off-by: Zahari Doychev <zdoychev@maxlinear.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David Ahern <dsahern@kernel.org>
Edwin Peer [Sun, 11 Jun 2023 10:57:38 +0000 (13:57 +0300)]
iplink: filter stats using RTEXT_FILTER_SKIP_STATS
Don't request statistics we do not intend to render. This avoids the
possibility of a truncated IFLA_VFINFO_LIST when statistics are not
requested as well as the fetching of unnecessary data.
Signed-off-by: Edwin Peer <edwin.peer@broadcom.com>
Cc: Edwin Peer <espeer@gmail.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: David Ahern <dsahern@kernel.org>
David Ahern [Wed, 21 Jun 2023 16:13:05 +0000 (09:13 -0700)]
Update kernel headers
Update kernel headers to commit:
cc7554954848 ("net: micrel: Change to receive timestamp in the frame for lan8841")
Signed-off-by: David Ahern <dsahern@kernel.org>
Vladimir Oltean [Wed, 7 Jun 2023 15:45:04 +0000 (18:45 +0300)]
tc/taprio: print the offload xstats
When the kernel reports offload counters through TCA_STATS2 ->
TCA_STATS_APP for the taprio qdisc, decode and print them.
Usage:
# Global stats
$ tc -s qdisc show dev eth0 root
# Per-tc stats
$ tc -s class show dev eth0
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David Ahern <dsahern@kernel.org>
Ido Schimmel [Wed, 7 Jun 2023 15:35:50 +0000 (18:35 +0300)]
f_flower: Add l2_miss support
Add the ability to match on packets that encountered a layer 2 miss in
bridge driver's FDB / MDB. Example:
# tc filter add dev swp2 egress pref 1 proto all flower indev swp1 l2_miss 1 action drop
# tc filter add dev swp2 egress pref 1 proto all flower indev swp1 l2_miss 0 action drop
# tc filter show dev swp2 egress
filter protocol all pref 1 flower chain 0
filter protocol all pref 1 flower chain 0 handle 0x1
indev swp1
l2_miss 1
not_in_hw
action order 1: gact action drop
random type none pass val 0
index 1 ref 1 bind 1
filter protocol all pref 1 flower chain 0 handle 0x2
indev swp1
l2_miss 0
not_in_hw
action order 1: gact action drop
random type none pass val 0
index 2 ref 1 bind 1
# tc -j -p filter show dev swp2 egress
[ {
"protocol": "all",
"pref": 1,
"kind": "flower",
"chain": 0
},{
"protocol": "all",
"pref": 1,
"kind": "flower",
"chain": 0,
"options": {
"handle": 1,
"indev": "swp1",
"keys": {
"l2_miss": 1
},
"not_in_hw": true,
"actions": [ {
[...]
} ]
}
},{
"protocol": "all",
"pref": 1,
"kind": "flower",
"chain": 0,
"options": {
"handle": 2,
"indev": "swp1",
"keys": {
"l2_miss": 0
},
"not_in_hw": true,
"actions": [ {
[...]
} ]
}
} ]
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David Ahern <dsahern@kernel.org>
David Ahern [Fri, 9 Jun 2023 21:46:04 +0000 (15:46 -0600)]
Merge branch 'dcb-rewr-subcmd' into next
Daniel Machon says:
====================
========================================================================
Introduction:
========================================================================
This series introduces a new DCB subcommand: rewr, which is used to
configure the in-kernel DCB rewrite table [1].
Rewrite support is added as a separate DCB subcommand, rather than an
APP opt-in flag or similar. This goes in line with what we did to dcbnl,
where rewrite is a separate object. Obviously this requires a bit more
code to implement the new command, but much of the existing dcb-app code
(especially the bookkeeping code) can be reused. In some cases a little
adaptation is needed.
========================================================================
dcb-rewr parameters:
========================================================================
Initially, I have only made support for the prio-pcp and prio-dscp
parameters, as DSCP and PCP are the only selectors that currently have
a user [2] and to be honest, I am not even sure it makes sense to add
dgram, stream, ethtype rewrite support - At least the rewriter of Sparx5
does not support this. Any input here is much appreciated!
Examples:
Rewrite DSCP to 63 for packets with priority 1
$ dcb rewr add dev eth0 prio-dscp 1:63
Rewrite PCP 7 and DEI to 1 for packets with priority 1
$ dcb rewr add dev eth0 prio-pcp 1:7de
A new manpage has been added, to cover the new dcb-rewr subcommand, and
its parameters. Also I took the liberty to clean up a few things in the
dcb-app manpage.
========================================================================
Patch overview:
========================================================================
Patch #1 Adds a new field 'attr' to the dcb_app_table struct, which is
used to distinguish app and rewrite tables.
Patch #2 Replaces uses of %d with %u for unsigned int.
Patch #3 Moves colon out of callback functions.
Patch #4 Renames protocl print functions from _key to _pid
Patch #5 Modifies the _print_filtered() function for dcb-rewr reuse, by
introducing new callbacks.
Patch #6 Modifies existing dcb-app function dcb_app_table_remove_replaced
for reuse by dcb-rewr
Patch #7 Expose dcb-app functions required by dcb-rewr.
Patch #8 Adds the new dcb-rewr subcommand with initial support for
prio-pcp and prio-dscp rewrite.
Patch #9 Introduces symbol for max DSCP value and updates accordingly.
Patch #10 Adds the dcb-rewr.8 manpage
Patch #11 Adds references to dcb-apptrust and dcb-rewr in the dcb.8
manpage.
Patch #12 Cleans up the dcb-app.8 manpage.
[1] https://elixir.bootlin.com/linux/v6.4-rc1/source/net/dcb/dcbnl.c#L181
[2] https://elixir.bootlin.com/linux/v6.4-rc1/source/drivers/net/ethernet/microchip/sparx5/sparx5_dcb.c#L380
====================
Signed-off-by: David Ahern <dsahern@kernel.org>
Daniel Machon [Tue, 6 Jun 2023 07:19:47 +0000 (09:19 +0200)]
man: dcb-app: clean up a few mistakes
While referencing the dcb-app manpage, I spotted a few mistakes. Lets
fix them.
Reviewed-by: Petr Machata <me@pmachata.org>
Signed-off-by: Daniel Machon <daniel.machon@microchip.com>
Signed-off-by: David Ahern <dsahern@kernel.org>
Daniel Machon [Tue, 6 Jun 2023 07:19:46 +0000 (09:19 +0200)]
man: dcb: add additional references under 'SEE ALSO'
Add dcb-apptrust and dcb-rewr to the 'SEE ALSO' section of the dcb
manpage.
Reviewed-by: Petr Machata <me@pmachata.org>
Signed-off-by: Daniel Machon <daniel.machon@microchip.com>
Signed-off-by: David Ahern <dsahern@kernel.org>
Daniel Machon [Tue, 6 Jun 2023 07:19:45 +0000 (09:19 +0200)]
man: dcb-rewr: add new manpage for dcb-rewr
Add a new manpage for dcb-rewr. Most of the content is copied over from
dcb-app, as the same set of commands and parameters (in reverse) applies
to dcb-rewr.
Reviewed-by: Petr Machata <me@pmachata.org>
Signed-off-by: Daniel Machon <daniel.machon@microchip.com>
Signed-off-by: David Ahern <dsahern@kernel.org>
Daniel Machon [Tue, 6 Jun 2023 07:19:44 +0000 (09:19 +0200)]
dcb: rewr: add symbol for max DSCP value
A symbol, DCB_APP_PCP_MAX, for maximum PCP value, already exists. Lets
add a symbol DCB_APP_DSCP_MAX and update accordingly.
Signed-off-by: Daniel Machon <daniel.machon@microchip.com>
Reviewed-by: Petr Machata <me@pmachata.org>
Signed-off-by: David Ahern <dsahern@kernel.org>
Daniel Machon [Tue, 6 Jun 2023 07:19:43 +0000 (09:19 +0200)]
dcb: rewr: add new dcb-rewr subcommand
Add a new subcommand 'rewr' for configuring the in-kernel DCB rewrite
table. The rewrite table of the kernel is similar to the APP table,
therefore, much of the existing bookkeeping code from dcb-app, can be
reused in the dcb-rewr implementation.
Initially, only support for configuring PCP and DSCP-based rewrite has
been added.
Signed-off-by: Daniel Machon <daniel.machon@microchip.com>
Reviewed-by: Petr Machata <me@pmachata.org>
Signed-off-by: David Ahern <dsahern@kernel.org>
Daniel Machon [Tue, 6 Jun 2023 07:19:42 +0000 (09:19 +0200)]
dcb: app: expose functions required by dcb-rewr
In preparation for the dcb-rewr implementation, expose required
functions, and structs.
Signed-off-by: Daniel Machon <daniel.machon@microchip.com>
Reviewed-by: Petr Machata <me@pmachata.org>
Signed-off-by: David Ahern <dsahern@kernel.org>
Daniel Machon [Tue, 6 Jun 2023 07:19:41 +0000 (09:19 +0200)]
dcb: app: modify dcb_app_table_remove_replaced() for dcb-rewr reuse
When doing a replace command, entries are checked against selector and
protocol. Rewrite requires the check to be against selector and
priority.
Adapt the existing dcb_app_table_remove_replace function for this, by
using callback functions for selector, pid and prio checks.
Signed-off-by: Daniel Machon <daniel.machon@microchip.com>
Reviewed-by: Petr Machata <me@pmachata.org>
Signed-off-by: David Ahern <dsahern@kernel.org>
Daniel Machon [Tue, 6 Jun 2023 07:19:40 +0000 (09:19 +0200)]
dcb: app: modify dcb_app_print_filtered() for dcb-rewr reuse
Where dcb-app requires protocol to be the printed key, dcb-rewr requires
it to be the priority. Adapt existing dcb-app print functions for this.
dcb_app_print_filtered() has been modified, to take two callbacks; one
for printing the entire string (pid and prio), and one for the pid type
(dec, hex, dscp, pcp). This saves us for making one dedicated function
for each pid type for both app and rewr.
Also, printing the colon is now expected to be handled by the
print_pid_prio() callback.
Signed-off-by: Daniel Machon <daniel.machon@microchip.com>
Reviewed-by: Petr Machata <me@pmachata.org>
Signed-off-by: David Ahern <dsahern@kernel.org>
Daniel Machon [Tue, 6 Jun 2023 07:19:39 +0000 (09:19 +0200)]
dcb: app: rename dcb_app_print_key_*() functions
In preparation for changing the prototype of dcb_app_print_filtered(),
rename the _print_key_*() functions to _print_pid_*(), as the protocol
can both be key and value with the introduction of dcb-rewr.
Signed-off-by: Daniel Machon <daniel.machon@microchip.com>
Reviewed-by: Petr Machata <me@pmachata.org>
Signed-off-by: David Ahern <dsahern@kernel.org>
Daniel Machon [Tue, 6 Jun 2023 07:19:38 +0000 (09:19 +0200)]
dcb: app: move colon printing out of callbacks
In preparation for changing the prototype of dcb_app_print_filtered(),
move the colon printing out of the callbacks, and into
dcb_app_print_filtered().
Signed-off-by: Daniel Machon <daniel.machon@microchip.com>
Reviewed-by: Petr Machata <me@pmachata.org>
Signed-off-by: David Ahern <dsahern@kernel.org>
Daniel Machon [Tue, 6 Jun 2023 07:19:37 +0000 (09:19 +0200)]
dcb: app: replace occurrences of %d with %u for printing unsigned int
In preparation for changing the prototype of dcb_app_print_filtered(),
replace occurrences of %d for printing unsigned integer, with %u as it
ought to be.
Signed-off-by: Daniel Machon <daniel.machon@microchip.com>
Reviewed-by: Petr Machata <me@pmachata.org>
Signed-off-by: David Ahern <dsahern@kernel.org>
Daniel Machon [Tue, 6 Jun 2023 07:19:36 +0000 (09:19 +0200)]
dcb: app: add new dcbnl attribute field
Add a new field 'attr' to the dcb_app_table struct, in order to inject
different dcbnl get/set attributes for APP and rewrite.
Signed-off-by: Daniel Machon <daniel.machon@microchip.com>
Reviewed-by: Petr Machata <me@pmachata.org>
Signed-off-by: David Ahern <dsahern@kernel.org>
Stephen Hemminger [Fri, 9 Jun 2023 16:23:40 +0000 (09:23 -0700)]
uapi: update to bpf.h
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Thu, 8 Jun 2023 01:33:49 +0000 (18:33 -0700)]
rt_names: check for malloc() failure
Fixes issue reported by Gcc 13 analayzer.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Vladimir Nikishkin [Tue, 6 Jun 2023 02:32:02 +0000 (10:32 +0800)]
ip-link: add support for nolocalbypass in vxlan
Add userspace support for the [no]localbypass vxlan netlink
attribute. With localbypass on (default), the vxlan driver processes
the packets destined to the local machine by itself, bypassing the
userspace nework stack. With nolocalbypass the packets are always
forwarded to the userspace network stack, so userspace programs,
such as tcpdump have a chance to process them.
Signed-off-by: Vladimir Nikishkin <vladimir@nikishkin.pw>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Andrea Claudi <aclaudi@redhat.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David Ahern <dsahern@kernel.org>
David Ahern [Tue, 6 Jun 2023 14:30:22 +0000 (08:30 -0600)]
Update kernel headers
Update kernel headers to commit:
ae91f7e436f8 ("net/pppoe: fix a typo for the PPPOE_HASH_BITS_1 definition")
Signed-off-by: David Ahern <dsahern@kernel.org>
Stephen Hemminger [Tue, 6 Jun 2023 01:43:58 +0000 (18:43 -0700)]
whitespace cleanups
Remove trailing whitespace.
Make sure there is space after keywords.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Andrea Claudi [Wed, 31 May 2023 10:35:56 +0000 (12:35 +0200)]
treewide: fix indentation
Replace multiple whitespaces with tab where appropriate.
While at it, fix tc flower help message and remove some double
whitespaces.
Signed-off-by: Andrea Claudi <aclaudi@redhat.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Tested-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David Ahern <dsahern@kernel.org>
David Ahern [Mon, 5 Jun 2023 15:18:09 +0000 (09:18 -0600)]
Merge branch 'main' into next
Signed-off-by: David Ahern <dsahern@kernel.org>
Stephen Hemminger [Thu, 1 Jun 2023 17:14:11 +0000 (10:14 -0700)]
rdma: make rd_attr_check static
Function defined and used in only one file.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Thu, 1 Jun 2023 17:13:23 +0000 (10:13 -0700)]
xfrm: make xfrm_stat_print_nokeys static
This function is only used in one file.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Thu, 1 Jun 2023 17:12:32 +0000 (10:12 -0700)]
ip: make print_rta_gateway static
Function only used in one file.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Thu, 1 Jun 2023 17:10:31 +0000 (10:10 -0700)]
bridge: make print_vlan_info static
Function defined and used in only one file.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Thu, 1 Jun 2023 17:09:47 +0000 (10:09 -0700)]
libnetlink: drop unused rtnl_talk_iov
Function was defined but not used in current iproute2 code.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Thu, 1 Jun 2023 17:08:09 +0000 (10:08 -0700)]
utils: make local cmdline functions static
No need to expose these parts of command line parsing.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Fri, 2 Jun 2023 15:51:53 +0000 (08:51 -0700)]
ipaddress: accept symbolic names
The function rtnl_addproto_a2n() was defined but never used.
Use it to allow for symbolic names, and fix the function signatures
so protocol value is consistently __u8.
Fixes: bdb8d8549ed9 ("ip: Support IP address protocol")
Cc: petrm@nvidia.com
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Tue, 30 May 2023 21:54:10 +0000 (14:54 -0700)]
uapi: update headers to 6.4-rc4
Minor addition to in.h
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Tue, 23 May 2023 16:57:15 +0000 (09:57 -0700)]
vxlan: make option printing more consistent
Add new helper function print_bool_opt() which prints
with no prefix and use it for vxlan options.
If the option matches the expected default value,
it is not printed if in non JSON mode unless the details
setting is repeated.
Use a table for the vxlan options. This will change
the order of the printing of options.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Tue, 23 May 2023 16:56:05 +0000 (09:56 -0700)]
vxlan: use print_nll for gbp and gpe
The Gbp and Gpe are presence, not booleans so use print_null()
for them
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Andrea Claudi [Mon, 29 May 2023 21:42:16 +0000 (23:42 +0200)]
iproute_lwtunnel: fix array boundary check
seg6_mode_types is made up of 5 elements, so ARRAY_SIZE(seg6_mode_types)
evaluates to 5. Thus, when mode = 5, this function returns
seg6_mode_types[5], resulting in an out-of-bound access.
Fix this bailing out when mode is equal to or greater than 5.
Fixes: cf87da417bb4 ("iproute: add support for seg6 l2encap mode")
Signed-off-by: Andrea Claudi <aclaudi@redhat.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Andrea Claudi [Mon, 29 May 2023 21:41:41 +0000 (23:41 +0200)]
vdpa: propagate error from cmd_dev_vstats_show()
Error potentially returned from mnlu_gen_socket_sndrcv() are propagated
for each and every invocation in vdpa. Let's do the same here.
Fixes: 6f97e9c9337b ("vdpa: Add support for reading vdpa device statistics")
Signed-off-by: Andrea Claudi <aclaudi@redhat.com>
Reviewed-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Andrea Claudi [Mon, 29 May 2023 16:59:15 +0000 (18:59 +0200)]
ipstats: fix message reporting error
strerror() accepts any integer as arguments, but returns meaningful
error descriptions only for positive integers.
ipstats code uses strerror on a code path where either err is 0 or
-ENOMEM, thus resulting in a useless error message.
Fix this using errno and moving the error printing closer to the only
function populating it in this code path.
Fixes: df0b2c6d0098 ("ipstats: Add a shell of "show" command")
Signed-off-by: Andrea Claudi <aclaudi@redhat.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Andrea Claudi [Fri, 26 May 2023 17:27:20 +0000 (19:27 +0200)]
bridge: vni: remove useless checks on vni
After the (d == NULL || vni == NULL) check, vni cannot be NULL anymore.
This remove two useless conditional checks on vni value:
- the first check cannot be true, so remove the whole conditional block
- the second check is always true, so remove the check
Signed-off-by: Andrea Claudi <aclaudi@redhat.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Andrea Claudi [Fri, 26 May 2023 17:36:54 +0000 (19:36 +0200)]
ip: remove double space before 'allmulti' flag
Current output:
$ ip -d link show vxlan0
79: vxlan0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
link/ether b6:f6:12:c3:2d:52 brd ff:ff:ff:ff:ff:ff promiscuity 0 allmulti 0 minmtu 68 maxmtu 65535
Resulting output:
$ ip -d link show vxlan0
79: vxlan0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
link/ether b6:f6:12:c3:2d:52 brd ff:ff:ff:ff:ff:ff promiscuity 0 allmulti 0 minmtu 68 maxmtu 65535
Fixes: e98683accc28 ("link: display 'allmulti' counter")
Signed-off-by: Andrea Claudi <aclaudi@redhat.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Fri, 26 May 2023 18:55:58 +0000 (11:55 -0700)]
ll_type_n2a: use ARRAY_SIZE
Replace open coded version of array size computation.
And fix whitespace.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Andrea Claudi [Tue, 16 May 2023 09:48:04 +0000 (11:48 +0200)]
mptcp: add support for implicit flag
Kernel supports implicit flag since commit
d045b9eb95a9 ("mptcp:
introduce implicit endpoints"), included in v5.18.
Let's add support for displaying it to iproute2.
Before this change:
$ ip mptcp endpoint show
10.0.2.2 id 1 rawflags 10
After this change:
$ ip mptcp endpoint show
10.0.2.2 id 1 implicit
Signed-off-by: Andrea Claudi <aclaudi@redhat.com>
Signed-off-by: David Ahern <dsahern@kernel.org>
Stephen Hemminger [Fri, 12 May 2023 03:47:54 +0000 (20:47 -0700)]
CREDITS: add file
Record some of the historical contributors to iproute2.
Take Alexey out of README.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Tue, 9 May 2023 03:17:50 +0000 (20:17 -0700)]
tc/prio: handle possible truncated kernel response
Reported by -fanalyzer. If kernel did not send full qdisc
info, then uninitialized or null data could be referenced.
q_prio.c: In function ‘prio_print_opt’:
q_prio.c:105:57: warning: dereference of NULL ‘0’ [CWE-476] [-Wanalyzer-null-dereference]
105 | print_uint(PRINT_ANY, "bands", "bands %u ", qopt->bands);
| ~~~~^~~~~~~
‘prio_print_opt’: event 1
|
| 98 | if (opt == NULL)
| | ^
| | |
| | (1) following ‘false’ branch (when ‘opt’ is non-NULL)...
|
‘prio_print_opt’: event 2
|
|../include/uapi/linux/rtnetlink.h:228:38:
| 228 | #define RTA_PAYLOAD(rta) ((int)((rta)->rta_len) - RTA_LENGTH(0))
| | ~~~~~~^~~~~~~~~~
| | |
| | (2) ...to here
../include/libnetlink.h:236:19: note: in expansion of macro ‘RTA_PAYLOAD’
| 236 | ({ data = RTA_PAYLOAD(rta) >= len ? RTA_DATA(rta) : NULL; \
| | ^~~~~~~~~~~
q_prio.c:101:13: note: in expansion of macro ‘parse_rtattr_nested_compat’
| 101 | if (parse_rtattr_nested_compat(tb, TCA_PRIO_MAX, opt, qopt,
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~
|
‘prio_print_opt’: event 3
|
|../include/libnetlink.h:236:59:
| 236 | ({ data = RTA_PAYLOAD(rta) >= len ? RTA_DATA(rta) : NULL; \
q_prio.c:101:13: note: in expansion of macro ‘parse_rtattr_nested_compat’
| 101 | if (parse_rtattr_nested_compat(tb, TCA_PRIO_MAX, opt, qopt,
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~
|
‘prio_print_opt’: events 4-5
|
| 105 | print_uint(PRINT_ANY, "bands", "bands %u ", qopt->bands);
| | ~~~~^~~~~~~
| | |
| | (4) ...to here
| | (5) dereference of NULL ‘<unknown>’
|
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Tue, 9 May 2023 03:15:52 +0000 (20:15 -0700)]
rdma/utils: fix some analyzer warnings
Add error checks for cases where analyzer thinks it is possible
to us a possibly NULL value.
utils.c: In function ‘get_port_from_argv’:
utils.c:76:17: warning: use of NULL where non-null expected [CWE-476] [-Wanalyzer-null-argument]
76 | slash = strchr(rd_argv(rd), '/');
| ^~~~~~~~~~~~~~~~~~~~~~~~
‘get_port_from_argv’: events 1-2
|
| 68 | static int get_port_from_argv(struct rd *rd, uint32_t *port,
| | ^~~~~~~~~~~~~~~~~~
| | |
| | (1) entry to ‘get_port_from_argv’
|......
| 76 | slash = strchr(rd_argv(rd), '/');
| | ~
| | |
| | (2) inlined call to ‘rd_argv’ from ‘get_port_from_argv’
|
+--> ‘rd_argv’: event 3
|
| 18 | if (!rd_argc(rd))
| | ^
| | |
| | (3) following ‘true’ branch...
|
<------+
|
‘get_port_from_argv’: events 4-5
|
| 76 | slash = strchr(rd_argv(rd), '/');
| | ^~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (4) ...to here
| | (5) argument 1 (‘<unknown>’) NULL where non-null expected
|
In file included from rdma.h:10,
from utils.c:7:
/usr/include/string.h:246:14: note: argument 1 of ‘strchr’ must be non-null
246 | extern char *strchr (const char *__s, int __c)
| ^~~~~~
Fixes: 40df8263a0f0 ("rdma: Add dev object")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Tue, 9 May 2023 02:42:03 +0000 (19:42 -0700)]
nstat: fix potential NULL deref
Reported as:
CC nstat
nstat.c: In function ‘load_ugly_table’:
nstat.c:205:24: warning: dereference of NULL ‘p’ [CWE-476] [-Wanalyzer-null-dereference]
205 | while (*p) {
| ^~
‘main’: events 1-14
|
| 575 | int main(int argc, char *argv[])
| | ^~~~
| | |
| | (1) entry to ‘main’
|......
| 635 | if (scan_interval > 0) {
| | ~
| | |
| | (2) following ‘true’ branch...
| 636 | if (time_constant == 0)
| | ~~~~~~~~~~~~~~~~~~
| | |
| | (3) ...to here
|......
| 640 | if ((fd = socket(AF_UNIX, SOCK_STREAM, 0)) < 0) {
| | ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | | |
| | | (4) when ‘socket’ succeeds
| | (5) following ‘false’ branch (when ‘fd >= 0’)...
|......
| 644 | if (bind(fd, (struct sockaddr *)&sun, 2+1+strlen(sun.sun_path+1)) < 0) {
| | ~ ~~~~~~~~~~~~~~~~~~~~~~
| | | |
| | (7) following ‘false’ branch... (6) ...to here
|......
| 648 | if (listen(fd, 5) < 0) {
| | ~~~~~~~~~~~~~~
| | ||
| | |(8) ...to here
| | |(9) when ‘listen’ succeeds
| | (10) following ‘false’ branch...
|......
| 652 | if (daemon(0, 0)) {
| | ~~~~~~~~~~~~~
| | ||
| | |(11) ...to here
| | (12) following ‘false’ branch...
|......
| 656 | signal(SIGPIPE, SIG_IGN);
| | ~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (13) ...to here
| 657 | signal(SIGCHLD, sigchild);
| 658 | server_loop(fd);
| | ~~~~~~~~~~~~~~~
| | |
| | (14) calling ‘server_loop’ from ‘main’
|
+--> ‘server_loop’: events 15-16
|
| 472 | static void server_loop(int fd)
| | ^~~~~~~~~~~
| | |
| | (15) entry to ‘server_loop’
|......
| 483 | load_netstat();
| | ~~~~~~~~~~~~~~
| | |
| | (16) calling ‘load_netstat’ from ‘server_loop’
|
+--> ‘load_netstat’: events 17-20
|
| 302 | static void load_netstat(void)
| | ^~~~~~~~~~~~
| | |
| | (17) entry to ‘load_netstat’
|......
| 306 | if (fp) {
| | ~
| | |
| | (18) following ‘true’ branch (when ‘fp’ is non-NULL)...
| 307 | load_ugly_table(fp);
| | ~~~~~~~~~~~~~~~~~~~
| | |
| | (19) ...to here
| | (20) calling ‘load_ugly_table’ from ‘load_netstat’
|
+--> ‘load_ugly_table’: events 21-26
|
| 178 | static void load_ugly_table(FILE *fp)
| | ^~~~~~~~~~~~~~~
| | |
| | (21) entry to ‘load_ugly_table’
| 179 | {
| 180 | char *buf = NULL;
| | ~~~
| | |
| | (22) ‘buf’ is NULL
|......
| 186 | while ((nread = getline(&buf, &buflen, fp)) != -1) {
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (23) following ‘true’ branch...
|......
| 192 | p = strchr(buf, ':');
| | ~~~~~~~~~~~~~~~~
| | |
| | (24) ...to here
| | (25) when ‘strchr’ returns non-NULL
| 193 | if (!p) {
| | ~
| | |
| | (26) following ‘false’ branch (when ‘p’ is non-NULL)...
|
‘load_ugly_table’: event 27
|
|cc1:
| (27): ...to here
|
‘load_ugly_table’: events 28-40
|
| 205 | while (*p) {
| | ^~
| | |
| | (28) following ‘true’ branch...
| | (40) dereference of NULL ‘p’
|......
| 208 | if ((next = strchr(p, ' ')) != NULL)
| | ~ ~~~~~~~~~~~~~~
| | | |
| | | (29) ...to here
| | | (30) when ‘strchr’ returns NULL
| | (31) following ‘false’ branch (when ‘next’ is NULL)...
| 209 | *next++ = 0;
| 210 | else if ((next = strchr(p, '\n')) != NULL)
| | ~ ~~~~~~~~~~~~~~~
| | | |
| | | (32) ...to here
| | | (33) when ‘strchr’ returns NULL
| | (34) following ‘false’ branch (when ‘next’ is NULL)...
| 211 | *next++ = 0;
| 212 | if (off < sizeof(idbuf)) {
| | ~~~~~~~~~~~~~~~~~~~~
| | | |
| | | (35) ...to here
| | (36) following ‘false’ branch...
|......
| 216 | n = malloc(sizeof(*n));
| | ~~~~~~~~~~~~~~~~~~
| | |
| | (37) ...to here
| 217 | if (!n) {
| | ~
| | |
| | (38) following ‘false’ branch (when ‘n’ is non-NULL)...
|......
| 221 | n->id = strdup(idbuf);
| | ~~~~~~~~~~~~~
| | |
| | (39) ...to here
|
nstat.c:254:35: warning: dereference of NULL ‘n’ [CWE-476] [-Wanalyzer-null-dereference]
254 | n = n->next;
| ~~^~~~~~~~~
‘main’: events 1-14
|
| 575 | int main(int argc, char *argv[])
| | ^~~~
| | |
| | (1) entry to ‘main’
|......
| 635 | if (scan_interval > 0) {
| | ~
| | |
| | (2) following ‘true’ branch...
| 636 | if (time_constant == 0)
| | ~~~~~~~~~~~~~~~~~~
| | |
| | (3) ...to here
|......
| 640 | if ((fd = socket(AF_UNIX, SOCK_STREAM, 0)) < 0) {
| | ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | | |
| | | (4) when ‘socket’ succeeds
| | (5) following ‘false’ branch (when ‘fd >= 0’)...
|......
| 644 | if (bind(fd, (struct sockaddr *)&sun, 2+1+strlen(sun.sun_path+1)) < 0) {
| | ~ ~~~~~~~~~~~~~~~~~~~~~~
| | | |
| | (7) following ‘false’ branch... (6) ...to here
|......
| 648 | if (listen(fd, 5) < 0) {
| | ~~~~~~~~~~~~~~
| | ||
| | |(8) ...to here
| | |(9) when ‘listen’ succeeds
| | (10) following ‘false’ branch...
|......
| 652 | if (daemon(0, 0)) {
| | ~~~~~~~~~~~~~
| | ||
| | |(11) ...to here
| | (12) following ‘false’ branch...
|......
| 656 | signal(SIGPIPE, SIG_IGN);
| | ~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (13) ...to here
| 657 | signal(SIGCHLD, sigchild);
| 658 | server_loop(fd);
| | ~~~~~~~~~~~~~~~
| | |
| | (14) calling ‘server_loop’ from ‘main’
|
+--> ‘server_loop’: events 15-16
|
| 472 | static void server_loop(int fd)
| | ^~~~~~~~~~~
| | |
| | (15) entry to ‘server_loop’
|......
| 483 | load_netstat();
| | ~~~~~~~~~~~~~~
| | |
| | (16) calling ‘load_netstat’ from ‘server_loop’
|
+--> ‘load_netstat’: events 17-20
|
| 302 | static void load_netstat(void)
| | ^~~~~~~~~~~~
| | |
| | (17) entry to ‘load_netstat’
|......
| 306 | if (fp) {
| | ~
| | |
| | (18) following ‘true’ branch (when ‘fp’ is non-NULL)...
| 307 | load_ugly_table(fp);
| | ~~~~~~~~~~~~~~~~~~~
| | |
| | (19) ...to here
| | (20) calling ‘load_ugly_table’ from ‘load_netstat’
|
+--> ‘load_ugly_table’: events 21-25
|
| 178 | static void load_ugly_table(FILE *fp)
| | ^~~~~~~~~~~~~~~
| | |
| | (21) entry to ‘load_ugly_table’
|......
| 186 | while ((nread = getline(&buf, &buflen, fp)) != -1) {
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (22) following ‘true’ branch...
|......
| 192 | p = strchr(buf, ':');
| | ~~~~~~~~~~~~~~~~
| | |
| | (23) ...to here
| | (24) when ‘strchr’ returns non-NULL
| 193 | if (!p) {
| | ~
| | |
| | (25) following ‘false’ branch (when ‘p’ is non-NULL)...
|
‘load_ugly_table’: event 26
|
|cc1:
| (26): ...to here
|
‘load_ugly_table’: events 27-28
|
| 205 | while (*p) {
| | ^
| | |
| | (27) following ‘false’ branch...
|......
| 228 | nread = getline(&buf, &buflen, fp);
| | ~
| | |
| | (28) inlined call to ‘getline’ from ‘load_ugly_table’
|
+--> ‘getline’: event 29
|
|/usr/include/bits/stdio.h:120:10:
| 120 | return __getdelim (__lineptr, __n, '\n', __stream);
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (29) ...to here
|
<------+
|
‘load_ugly_table’: events 30-36
|
|nstat.c:229:20:
| 229 | if (nread == -1) {
| | ^
| | |
| | (30) following ‘false’ branch...
|......
| 234 | count2 = count_spaces(buf);
| | ~~~~~~~~~~~~~~~~~
| | |
| | (31) ...to here
|......
| 239 | if (!p) {
| | ~
| | |
| | (32) following ‘false’ branch (when ‘p’ is non-NULL)...
|......
| 244 | *p = 0;
| | ~~~~~~
| | |
| | (33) ...to here
| 245 | if (sscanf(p+1, "%llu", &n->val) != 1) {
| | ~
| | |
| | (34) following ‘false’ branch...
|......
| 251 | if (skip)
| | ~
| | |
| | (35) ...to here
|......
| 254 | n = n->next;
| | ~~~~~~~~~~~
| | |
| | (36) dereference of NULL ‘n’
|
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Tue, 9 May 2023 02:36:14 +0000 (19:36 -0700)]
netem: fix NULL deref on allocation failure
q_netem.c: In function ‘get_distribution’:
q_netem.c:159:35: warning: dereference of possibly-NULL ‘data’ [CWE-690] [-Wanalyzer-possible-null-dereference]
159 | data[n++] = x;
| ~~~~~~~~~~^~~
‘netem_parse_opt’: events 1-24
|
| 192 | static int netem_parse_opt(struct qdisc_util *qu, int argc, char **argv,
| | ^~~~~~~~~~~~~~~
| | |
| | (1) entry to ‘netem_parse_opt’
|......
| 212 | for ( ; argc > 0; --argc, ++argv) {
| | ~~~~~~~~
| | |
| | (2) following ‘true’ branch (when ‘argc > 0’)...
| 213 | if (matches(*argv, "limit") == 0) {
| | ~~~~~~~~~~~~~~~~~~~~~~~~
| | ||
| | |(3) ...to here
| | (4) following ‘true’ branch...
|......
| 219 | } else if (matches(*argv, "latency") == 0 ||
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | || |
| | |(5) ...to here (8) following ‘true’ branch...
| | (6) following ‘true’ branch...
| 220 | matches(*argv, "delay") == 0) {
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (7) ...to here
|......
| 243 | } else if (matches(*argv, "loss") == 0 ||
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | || |
| | |(9) ...to here (12) following ‘true’ branch...
| | (10) following ‘true’ branch...
| 244 | matches(*argv, "drop") == 0) {
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (11) ...to here
|......
| 366 | } else if (matches(*argv, "ecn") == 0) {
| | ~~~~~~~~~~~~~~~~~~~~~~
| | ||
| | |(13) ...to here
| | (14) following ‘true’ branch...
| 367 | present[TCA_NETEM_ECN] = 1;
| 368 | } else if (matches(*argv, "reorder") == 0) {
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~
| | ||
| | |(15) ...to here
| | (16) following ‘true’ branch...
|......
| 383 | } else if (matches(*argv, "corrupt") == 0) {
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~
| | ||
| | |(17) ...to here
| | (18) following ‘true’ branch...
|......
| 398 | } else if (matches(*argv, "gap") == 0) {
| | ~~~~~~~~~~~~~~~~~~~~~~
| | ||
| | |(19) ...to here
| | (20) following ‘true’ branch...
|......
| 404 | } else if (matches(*argv, "duplicate") == 0) {
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | ||
| | |(21) ...to here
| | (22) following ‘true’ branch...
|......
| 417 | } else if (matches(*argv, "distribution") == 0) {
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | ||
| | |(23) ...to here
| | (24) following ‘false’ branch...
|
‘netem_parse_opt’: event 25
|
|../include/utils.h:50:29:
| 50 | #define NEXT_ARG() do { argv++; if (--argc <= 0) incomplete_command(); } while(0)
| | ~~~~^~
| | |
| | (25) ...to here
q_netem.c:418:25: note: in expansion of macro ‘NEXT_ARG’
| 418 | NEXT_ARG();
| | ^~~~~~~~
|
‘netem_parse_opt’: event 26
|
|../include/utils.h:50:36:
| 50 | #define NEXT_ARG() do { argv++; if (--argc <= 0) incomplete_command(); } while(0)
| | ^
| | |
| | (26) following ‘false’ branch (when ‘argc != 0’)...
q_netem.c:418:25: note: in expansion of macro ‘NEXT_ARG’
| 418 | NEXT_ARG();
| | ^~~~~~~~
|
‘netem_parse_opt’: events 27-29
|
| 419 | dist_data = calloc(sizeof(dist_data[0]), MAX_DIST);
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (27) ...to here
| | (28) this call could return NULL
| 420 | dist_size = get_distribution(*argv, dist_data, MAX_DIST);
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (29) calling ‘get_distribution’ from ‘netem_parse_opt’
|
+--> ‘get_distribution’: events 30-31
|
| 124 | static int get_distribution(const char *type, __s16 *data, int maxdata)
| | ^~~~~~~~~~~~~~~~
| | |
| | (30) entry to ‘get_distribution’
|......
| 135 | if (f == NULL) {
| | ~
| | |
| | (31) following ‘false’ branch (when ‘f’ is non-NULL)...
|
‘get_distribution’: event 32
|
|cc1:
| (32): ...to here
|
‘get_distribution’: events 33-35
|
| 142 | while (getline(&line, &len, f) != -1) {
| | ~~~~~~~~~~~~~~~~~~~~~~~~^~~~~
| | |
| | (33) following ‘true’ branch...
|......
| 145 | if (*line == '\n' || *line == '#')
| | ~~~~~~
| | ||
| | |(34) ...to here
| | (35) following ‘false’ branch...
|
‘get_distribution’: event 36
|
|cc1:
| (36): ...to here
|
‘get_distribution’: events 37-41
|
| 150 | if (endp == p)
| | ^
| | |
| | (37) following ‘false’ branch...
|......
| 153 | if (n >= maxdata) {
| | ~
| | |
| | (38) ...to here
| | (39) following ‘false’ branch (when ‘n < maxdata’)...
|......
| 159 | data[n++] = x;
| | ~~~~~~~~~~~~~
| | | |
| | | (41) ‘data + (long unsigned int)n * 2’ could be NULL: unchecked value from (28)
| | (40) ...to here
|
Fixes: c1b81cb5fe92 ("netem potential dist table overflow")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Tue, 9 May 2023 02:25:33 +0000 (19:25 -0700)]
m_action: fix warning of overwrite of const string
The function get_action_kind() searches first for the given
action, then rescans on failure for "gact". In the process,
it would overwrite the argument. Avoid the warning
by using a const argument and not copying.
The problem dates back to pre-git history.
m_action.c: In function ‘get_action_kind’:
m_action.c:126:17: warning: write to string literal [-Wanalyzer-write-to-string-literal]
126 | strcpy(str, "gact");
| ^~~~~~~~~~~~~~~~~~~
‘do_action’: events 1-6
|
| 853 | int do_action(int argc, char **argv)
| | ^~~~~~~~~
| | |
| | (1) entry to ‘do_action’
|......
| 858 | while (argc > 0) {
| | ~~~~~~~~
| | |
| | (2) following ‘true’ branch...
| 859 |
| 860 | if (matches(*argv, "add") == 0) {
| | ~~~~~~~~~~~~~~~~~~~~~~
| | ||
| | |(3) ...to here
| | (4) following ‘false’ branch...
| 861 | ret = tc_action_modify(RTM_NEWACTION,
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (5) ...to here
| | (6) calling ‘tc_action_modify’ from ‘do_action’
| 862 | NLM_F_EXCL | NLM_F_CREATE,
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~
| 863 | &argc, &argv);
| | ~~~~~~~~~~~~~
|
+--> ‘tc_action_modify’: events 7-8
|
| 715 | static int tc_action_modify(int cmd, unsigned int flags,
| | ^~~~~~~~~~~~~~~~
| | |
| | (7) entry to ‘tc_action_modify’
|......
| 735 | if (parse_action(&argc, &argv, TCA_ACT_TAB, &req.n)) {
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (8) calling ‘parse_action’ from ‘tc_action_modify’
|
+--> ‘parse_action’: events 9-18
|
| 203 | int parse_action(int *argc_p, char ***argv_p, int tca_id, struct nlmsghdr *n)
| | ^~~~~~~~~~~~
| | |
| | (9) entry to ‘parse_action’
|......
| 217 | if (argc <= 0)
| | ~
| | |
| | (10) following ‘false’ branch...
|......
| 220 | tail2 = addattr_nest(n, MAX_MSG, tca_id);
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (11) ...to here
| 221 |
| 222 | while (argc > 0) {
| | ~~~~~~~~
| | |
| | (12) following ‘true’ branch...
| 223 |
| 224 | memset(k, 0, sizeof(k));
| | ~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (13) ...to here
| 225 |
| 226 | if (strcmp(*argv, "action") == 0) {
| | ~
| | |
| | (14) following ‘true’ branch (when the strings are equal)...
| 227 | argc--;
| | ~~~~~~
| | |
| | (15) ...to here
|......
| 231 | if (!gact_ld)
| | ~
| | |
| | (16) following ‘true’ branch...
| 232 | get_action_kind("gact");
| | ~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (17) ...to here
| | (18) calling ‘get_action_kind’ from ‘parse_action’
|
+--> ‘get_action_kind’: events 19-24
|
| 86 | static struct action_util *get_action_kind(char *str)
| | ^~~~~~~~~~~~~~~
| | |
| | (19) entry to ‘get_action_kind’
|......
| 114 | if (a == NULL)
| | ~
| | |
| | (20) following ‘true’ branch (when ‘a’ is NULL)...
| 115 | goto noexist;
| | ~~~~
| | |
| | (21) ...to here
|......
| 124 | if (!looked4gact) {
| | ~
| | |
| | (22) following ‘true’ branch (when ‘looked4gact == 0’)...
| 125 | looked4gact = 1;
| 126 | strcpy(str, "gact");
| | ~~~~~~~~~~~~~~~~~~~
| | |
| | (23) ...to here
| | (24) write to string literal here
|
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Tue, 9 May 2023 02:21:27 +0000 (19:21 -0700)]
tc_exec: don't dereference NULL on calloc failure
Reported as:
tc_exec.c: In function ‘do_exec’:
tc_exec.c:103:18: warning: dereference of NULL ‘eu’ [CWE-476] [-Wanalyzer-null-dereference]
103 | return eu->parse_eopt(eu, argc, argv);
| ~~^~~~~~~~~~~~
‘do_exec’: events 1-6
|
| 81 | int do_exec(int argc, char **argv)
| | ^~~~~~~
| | |
| | (1) entry to ‘do_exec’
|......
| 86 | if (argc < 1) {
| | ~
| | |
| | (2) following ‘false’ branch (when ‘argc > 0’)...
|......
| 91 | if (matches(*argv, "help") == 0) {
| | ~~~~~~~~~~~~~~~~~~~~~~~
| | ||
| | |(3) ...to here
| | (4) following ‘true’ branch...
|......
| 96 | strncpy(kind, *argv, sizeof(kind) - 1);
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (5) ...to here
| 97 |
| 98 | eu = get_exec_kind(kind);
| | ~~~~~~~~~~~~~~~~~~~
| | |
| | (6) calling ‘get_exec_kind’ from ‘do_exec’
|
+--> ‘get_exec_kind’: events 7-10
|
| 40 | static struct exec_util *get_exec_kind(const char *name)
| | ^~~~~~~~~~~~~
| | |
| | (7) entry to ‘get_exec_kind’
|......
| 63 | if (eu == NULL)
| | ~
| | |
| | (8) following ‘true’ branch (when ‘eu’ is NULL)...
| 64 | goto noexist;
| | ~~~~
| | |
| | (9) ...to here
|......
| 72 | if (eu) {
| | ~
| | |
| | (10) following ‘false’ branch (when ‘eu’ is NULL)...
|
‘get_exec_kind’: event 11
|
|cc1:
| (11): ...to here
|
<------+
|
‘do_exec’: events 12-13
|
| 98 | eu = get_exec_kind(kind);
| | ^~~~~~~~~~~~~~~~~~~
| | |
| | (12) return of NULL to ‘do_exec’ from ‘get_exec_kind’
|......
| 103 | return eu->parse_eopt(eu, argc, argv);
| | ~~~~~~~~~~~~~~
| | |
| | (13) dereference of NULL ‘eu’
|
Fixes: 4bd624467bc6 ("tc: built-in eBPF exec proxy")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Tue, 9 May 2023 02:15:43 +0000 (19:15 -0700)]
tc_util fix unitialized warning
tc_util.c: In function ‘parse_action_control_slash_spaces’:
tc_util.c:488:28: warning: use of uninitialized value ‘result2’ [CWE-457] [-Wanalyzer-use-of-uninitialized-value]
488 | *result2_p = result2;
| ~~~~~~~~~~~^~~~~~~~~
‘parse_action_control_slash_spaces’: events 1-5
|
| 455 | static int parse_action_control_slash_spaces(int *argc_p, char ***argv_p,
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (1) entry to ‘parse_action_control_slash_spaces’
|......
| 461 | int result1 = -1, result2;
| | ~~~~~~~
| | |
| | (2) region created on stack here
| | (3) capacity: 4 bytes
|......
| 467 | switch (ok) {
| | ~~~~~~
| | |
| | (4) following ‘case 0:’ branch...
|......
| 475 | ret = parse_action_control(&argc, &argv,
| | ~
| | |
| | (5) inlined call to ‘parse_action_control’ from ‘parse_action_control_slash_spaces’
|
+--> ‘parse_action_control’: events 6-7
|
| 432 | return __parse_action_control(argc_p, argv_p, result_p,
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (6) ...to here
| | (7) calling ‘__parse_action_control’ from ‘parse_action_control_slash_spaces’
| 433 | allow_num, false);
| | ~~~~~~~~~~~~~~~~~
|
‘__parse_action_control’: events 8-11
|
| 371 | static int __parse_action_control(int *argc_p, char ***argv_p, int *result_p,
| | ^~~~~~~~~~~~~~~~~~~~~~
| | |
| | (8) entry to ‘__parse_action_control’
|......
| 378 | if (!argc)
| | ~
| | |
| | (9) following ‘false’ branch (when ‘argc != 0’)...
| 379 | return -1;
| 380 | if (action_a2n(*argv, &result, allow_num) == -1) {
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (10) ...to here
| | (11) calling ‘action_a2n’ from ‘__parse_action_control’
|
+--> ‘action_a2n’: events 12-16
|
| 335 | int action_a2n(char *arg, int *result, bool allow_num)
| | ^~~~~~~~~~
| | |
| | (12) entry to ‘action_a2n’
|......
| 356 | for (iter = a2n; iter->a; iter++) {
| | ~~~~
| | |
| | (13) following ‘true’ branch...
| 357 | if (matches(arg, iter->a) != 0)
| | ~~~~~~~~~~~~~~~~~~~~~
| | |
| | (14) ...to here
|......
| 366 | if (result)
| | ~
| | |
| | (15) following ‘true’ branch (when ‘result’ is non-NULL)...
| 367 | *result = n;
| | ~~~~~~~~~~~
| | |
| | (16) ...to here
|
<------+
|
‘__parse_action_control’: event 17
|
| 380 | if (action_a2n(*argv, &result, allow_num) == -1) {
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (17) returning to ‘__parse_action_control’ from ‘action_a2n’
|
<------+
|
‘parse_action_control_slash_spaces’: event 18
|
| 475 | ret = parse_action_control(&argc, &argv,
| | ^
| | |
| | (18) inlined call to ‘parse_action_control’ from ‘parse_action_control_slash_spaces’
|
+--> ‘parse_action_control’: event 19
|
| 432 | return __parse_action_control(argc_p, argv_p, result_p,
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (19) returning to ‘parse_action_control_slash_spaces’ from ‘__parse_action_control’
| 433 | allow_num, false);
| | ~~~~~~~~~~~~~~~~~
|
<------+
|
‘parse_action_control_slash_spaces’: events 20-24
|
| 477 | if (ret)
| | ^
| | |
| | (20) following ‘false’ branch...
| 478 | return ret;
| 479 | ok++;
| | ~~~~
| | |
| | (21) ...to here
|......
| 487 | if (ok == 2)
| | ~
| | |
| | (22) following ‘true’ branch (when ‘ok == 2’)...
| 488 | *result2_p = result2;
| | ~~~~~~~~~~~~~~~~~~~~
| | |
| | (23) ...to here
| | (24) use of uninitialized value ‘result2’ here
|
tc_util.c:488:28: warning: use of uninitialized value ‘result2’ [CWE-457] [-Wanalyzer-use-of-uninitialized-value]
488 | *result2_p = result2;
| ~~~~~~~~~~~^~~~~~~~~
‘parse_action_control_slash’: events 1-5
|
| 505 | int parse_action_control_slash(int *argc_p, char ***argv_p,
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (1) entry to ‘parse_action_control_slash’
|......
| 510 | char *p = strchr(*argv, '/');
| | ~~~~~~~~~~~~~~~~~~
| | |
| | (2) when ‘strchr’ returns NULL
| 511 |
| 512 | if (!p)
| | ~
| | |
| | (3) following ‘true’ branch (when ‘p’ is NULL)...
| 513 | return parse_action_control_slash_spaces(argc_p, argv_p,
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (4) ...to here
| | (5) calling ‘parse_action_control_slash_spaces’ from ‘parse_action_control_slash’
| 514 | result1_p, result2_p,
| | ~~~~~~~~~~~~~~~~~~~~~
| 515 | allow_num);
| | ~~~~~~~~~~
|
+--> ‘parse_action_control_slash_spaces’: events 6-10
|
| 455 | static int parse_action_control_slash_spaces(int *argc_p, char ***argv_p,
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (6) entry to ‘parse_action_control_slash_spaces’
|......
| 461 | int result1 = -1, result2;
| | ~~~~~~~
| | |
| | (7) region created on stack here
| | (8) capacity: 4 bytes
|......
| 467 | switch (ok) {
| | ~~~~~~
| | |
| | (9) following ‘case 0:’ branch...
|......
| 475 | ret = parse_action_control(&argc, &argv,
| | ~
| | |
| | (10) inlined call to ‘parse_action_control’ from ‘parse_action_control_slash_spaces’
|
+--> ‘parse_action_control’: events 11-12
|
| 432 | return __parse_action_control(argc_p, argv_p, result_p,
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (11) ...to here
| | (12) calling ‘__parse_action_control’ from ‘parse_action_control_slash_spaces’
| 433 | allow_num, false);
| | ~~~~~~~~~~~~~~~~~
|
‘__parse_action_control’: events 13-16
|
| 371 | static int __parse_action_control(int *argc_p, char ***argv_p, int *result_p,
| | ^~~~~~~~~~~~~~~~~~~~~~
| | |
| | (13) entry to ‘__parse_action_control’
|......
| 378 | if (!argc)
| | ~
| | |
| | (14) following ‘false’ branch (when ‘argc != 0’)...
| 379 | return -1;
| 380 | if (action_a2n(*argv, &result, allow_num) == -1) {
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (15) ...to here
| | (16) calling ‘action_a2n’ from ‘__parse_action_control’
|
+--> ‘action_a2n’: events 17-21
|
| 335 | int action_a2n(char *arg, int *result, bool allow_num)
| | ^~~~~~~~~~
| | |
| | (17) entry to ‘action_a2n’
|......
| 356 | for (iter = a2n; iter->a; iter++) {
| | ~~~~
| | |
| | (18) following ‘true’ branch...
| 357 | if (matches(arg, iter->a) != 0)
| | ~~~~~~~~~~~~~~~~~~~~~
| | |
| | (19) ...to here
|......
| 366 | if (result)
| | ~
| | |
| | (20) following ‘true’ branch (when ‘result’ is non-NULL)...
| 367 | *result = n;
| | ~~~~~~~~~~~
| | |
| | (21) ...to here
|
<------+
|
‘__parse_action_control’: event 22
|
| 380 | if (action_a2n(*argv, &result, allow_num) == -1) {
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (22) returning to ‘__parse_action_control’ from ‘action_a2n’
|
<------+
|
‘parse_action_control_slash_spaces’: event 23
|
| 475 | ret = parse_action_control(&argc, &argv,
| | ^
| | |
| | (23) inlined call to ‘parse_action_control’ from ‘parse_action_control_slash_spaces’
|
+--> ‘parse_action_control’: event 24
|
| 432 | return __parse_action_control(argc_p, argv_p, result_p,
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (24) returning to ‘parse_action_control_slash_spaces’ from ‘__parse_action_control’
| 433 | allow_num, false);
| | ~~~~~~~~~~~~~~~~~
|
<------+
|
‘parse_action_control_slash_spaces’: events 25-29
|
| 477 | if (ret)
| | ^
| | |
| | (25) following ‘false’ branch...
| 478 | return ret;
| 479 | ok++;
| | ~~~~
| | |
| | (26) ...to here
|......
| 487 | if (ok == 2)
| | ~
| | |
| | (27) following ‘true’ branch (when ‘ok == 2’)...
| 488 | *result2_p = result2;
| | ~~~~~~~~~~~~~~~~~~~~
| | |
| | (28) ...to here
| | (29) use of uninitialized value ‘result2’ here
|
Fixes: e67aba559581 ("tc: actions: add helpers to parse and print control actions")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Tue, 9 May 2023 02:12:23 +0000 (19:12 -0700)]
tc_filter: fix unitialized warning
When run with -fanalyzer.
tc_filter.c: In function ‘tc_filter_list’:
tc_filter.c:718:17: warning: use of uninitialized value ‘chain_index’ [CWE-457] [-Wanalyzer-use-of-uninitialized-value]
718 | addattr32(&req.n, sizeof(req), TCA_CHAIN, chain_index);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
‘do_chain’: events 1-4
|
| 772 | int do_chain(int argc, char **argv)
| | ^~~~~~~~
| | |
| | (1) entry to ‘do_chain’
| 773 | {
| 774 | if (argc < 1)
| | ~
| | |
| | (2) following ‘true’ branch (when ‘argc <= 0’)...
| 775 | return tc_filter_list(RTM_GETCHAIN, 0, NULL);
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (3) ...to here
| | (4) calling ‘tc_filter_list’ from ‘do_chain’
|
+--> ‘tc_filter_list’: events 5-8
|
| 582 | static int tc_filter_list(int cmd, int argc, char **argv)
| | ^~~~~~~~~~~~~~
| | |
| | (5) entry to ‘tc_filter_list’
|......
| 597 | __u32 chain_index;
| | ~~~~~~~~~~~
| | |
| | (6) region created on stack here
| | (7) capacity: 4 bytes
|......
| 601 | while (argc > 0) {
| | ~~~~~~~~
| | |
| | (8) following ‘false’ branch (when ‘argc <= 0’)...
|
‘tc_filter_list’: event 9
|
|../include/uapi/linux/pkt_sched.h:72:35:
| 72 | #define TC_H_MAKE(maj,min) (((maj)&TC_H_MAJ_MASK)|((min)&TC_H_MIN_MASK))
| | ~~~~~~^~~~~~~~~~~~~~~
| | |
| | (9) ...to here
tc_filter.c:698:26: note: in expansion of macro ‘TC_H_MAKE’
| 698 | req.t.tcm_info = TC_H_MAKE(prio<<16, protocol);
| | ^~~~~~~~~
|
‘tc_filter_list’: events 10-16
|
| 702 | if (d[0]) {
| | ^
| | |
| | (10) following ‘false’ branch...
|......
| 707 | } else if (block_index) {
| | ~~~~~~~~~~~~
| | ||
| | |(11) ...to here
| | (12) following ‘false’ branch...
|......
| 717 | if (filter_chain_index_set)
| | ~~~~~~~~~~~~~~~~~~~~~~~
| | ||
| | |(13) ...to here
| | (14) following ‘true’ branch...
| 718 | addattr32(&req.n, sizeof(req), TCA_CHAIN, chain_index);
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (15) ...to here
| | (16) use of uninitialized value ‘chain_index’ here
|
tc_filter.c:718:17: warning: use of uninitialized value ‘chain_index’ [CWE-457] [-Wanalyzer-use-of-uninitialized-value]
718 | addattr32(&req.n, sizeof(req), TCA_CHAIN, chain_index);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
‘do_filter’: events 1-4
|
| 744 | int do_filter(int argc, char **argv)
| | ^~~~~~~~~
| | |
| | (1) entry to ‘do_filter’
| 745 | {
| 746 | if (argc < 1)
| | ~
| | |
| | (2) following ‘true’ branch (when ‘argc <= 0’)...
| 747 | return tc_filter_list(RTM_GETTFILTER, 0, NULL);
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (3) ...to here
| | (4) calling ‘tc_filter_list’ from ‘do_filter’
|
+--> ‘tc_filter_list’: events 5-8
|
| 582 | static int tc_filter_list(int cmd, int argc, char **argv)
| | ^~~~~~~~~~~~~~
| | |
| | (5) entry to ‘tc_filter_list’
|......
| 597 | __u32 chain_index;
| | ~~~~~~~~~~~
| | |
| | (6) region created on stack here
| | (7) capacity: 4 bytes
|......
| 601 | while (argc > 0) {
| | ~~~~~~~~
| | |
| | (8) following ‘false’ branch (when ‘argc <= 0’)...
|
‘tc_filter_list’: event 9
|
|../include/uapi/linux/pkt_sched.h:72:35:
| 72 | #define TC_H_MAKE(maj,min) (((maj)&TC_H_MAJ_MASK)|((min)&TC_H_MIN_MASK))
| | ~~~~~~^~~~~~~~~~~~~~~
| | |
| | (9) ...to here
tc_filter.c:698:26: note: in expansion of macro ‘TC_H_MAKE’
| 698 | req.t.tcm_info = TC_H_MAKE(prio<<16, protocol);
| | ^~~~~~~~~
|
‘tc_filter_list’: events 10-16
|
| 702 | if (d[0]) {
| | ^
| | |
| | (10) following ‘false’ branch...
|......
| 707 | } else if (block_index) {
| | ~~~~~~~~~~~~
| | ||
| | |(11) ...to here
| | (12) following ‘false’ branch...
|......
| 717 | if (filter_chain_index_set)
| | ~~~~~~~~~~~~~~~~~~~~~~~
| | ||
| | |(13) ...to here
| | (14) following ‘true’ branch...
| 718 | addattr32(&req.n, sizeof(req), TCA_CHAIN, chain_index);
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (15) ...to here
| | (16) use of uninitialized value ‘chain_index’ here
|
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Tue, 9 May 2023 02:05:38 +0000 (19:05 -0700)]
iproute_lwtunnel: fix possible use of NULL when malloc() fails
iproute_lwtunnel.c: In function ‘parse_srh’:
iproute_lwtunnel.c:903:9: warning: use of possibly-NULL ‘srh’ where non-null expected [CWE-690] [-Wanalyzer-possible-null-argument]
903 | memset(srh, 0, srhlen);
| ^~~~~~~~~~~~~~~~~~~~~~
‘parse_srh’: events 1-2
|
| 902 | srh = malloc(srhlen);
| | ^~~~~~~~~~~~~~
| | |
| | (1) this call could return NULL
| 903 | memset(srh, 0, srhlen);
| | ~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (2) argument 1 (‘srh’) from (1) could be NULL where non-null expected
|
In file included from iproute_lwtunnel.c:13:
/usr/include/string.h:61:14: note: argument 1 of ‘memset’ must be non-null
61 | extern void *memset (void *__s, int __c, size_t __n) __THROW __nonnull ((1));
| ^~~~~~
iproute_lwtunnel.c: In function ‘parse_encap_seg6’:
iproute_lwtunnel.c:980:9: warning: use of possibly-NULL ‘tuninfo’ where non-null expected [CWE-690] [-Wanalyzer-possible-null-argument]
980 | memset(tuninfo, 0, sizeof(*tuninfo) + srhlen);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
‘parse_encap_seg6’: events 1-2
|
| 934 | static int parse_encap_seg6(struct rtattr *rta, size_t len, int *argcp,
| | ^~~~~~~~~~~~~~~~
| | |
| | (1) entry to ‘parse_encap_seg6’
|......
| 976 | srh = parse_srh(segbuf, hmac, encap);
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (2) calling ‘parse_srh’ from ‘parse_encap_seg6’
|
+--> ‘parse_srh’: events 3-5
|
| 882 | static struct ipv6_sr_hdr *parse_srh(char *segbuf, int hmac, bool encap)
| | ^~~~~~~~~
| | |
| | (3) entry to ‘parse_srh’
|......
| 922 | if (hmac) {
| | ~
| | |
| | (4) following ‘false’ branch (when ‘hmac == 0’)...
|......
| 931 | return srh;
| | ~~~
| | |
| | (5) ...to here
|
<------+
|
‘parse_encap_seg6’: events 6-8
|
| 976 | srh = parse_srh(segbuf, hmac, encap);
| | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (6) returning to ‘parse_encap_seg6’ from ‘parse_srh’
|......
| 979 | tuninfo = malloc(sizeof(*tuninfo) + srhlen);
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (7) this call could return NULL
| 980 | memset(tuninfo, 0, sizeof(*tuninfo) + srhlen);
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (8) argument 1 (‘tuninfo’) from (7) could be NULL where non-null expected
|
/usr/include/string.h:61:14: note: argument 1 of ‘memset’ must be non-null
61 | extern void *memset (void *__s, int __c, size_t __n) __THROW __nonnull ((1));
| ^~~~~~
iproute_lwtunnel.c: In function ‘parse_rpl_srh’:
iproute_lwtunnel.c:1018:21: warning: dereference of possibly-NULL ‘srh’ [CWE-690] [-Wanalyzer-possible-null-dereference]
1018 | srh->hdrlen = (srhlen >> 3) - 1;
| ~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~
‘parse_rpl_srh’: events 1-2
|
| 1016 | srh = calloc(1, srhlen);
| | ^~~~~~~~~~~~~~~~~
| | |
| | (1) this call could return NULL
| 1017 |
| 1018 | srh->hdrlen = (srhlen >> 3) - 1;
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (2) ‘srh’ could be NULL: unchecked value from (1)
|
Fixes: 00e76d4da37f ("iproute: add helper functions for SRH processing")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Tue, 9 May 2023 02:02:20 +0000 (19:02 -0700)]
ipmaddr: fix dereference of NULL on malloc() failure
Found by -fanalyzer. This is a bug since beginning of initial
versions of ip multicast support (pre git).
ipmaddr.c: In function ‘read_dev_mcast’:
ipmaddr.c:105:25: warning: dereference of possibly-NULL ‘ma’ [CWE-690] [-Wanalyzer-possible-null-dereference]
105 | memcpy(ma, &m, sizeof(m));
| ^~~~~~~~~~~~~~~~~~~~~~~~~
‘do_multiaddr’: events 1-4
|
| 354 | int do_multiaddr(int argc, char **argv)
| | ^~~~~~~~~~~~
| | |
| | (1) entry to ‘do_multiaddr’
| 355 | {
| 356 | if (argc < 1)
| | ~
| | |
| | (2) following ‘true’ branch (when ‘argc <= 0’)...
| 357 | return multiaddr_list(0, NULL);
| | ~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (3) ...to here
| | (4) calling ‘multiaddr_list’ from ‘do_multiaddr’
|
+--> ‘multiaddr_list’: events 5-10
|
| 255 | static int multiaddr_list(int argc, char **argv)
| | ^~~~~~~~~~~~~~
| | |
| | (5) entry to ‘multiaddr_list’
|......
| 262 | while (argc > 0) {
| | ~~~~~~~~
| | |
| | (6) following ‘false’ branch (when ‘argc <= 0’)...
|......
| 275 | if (!filter.family || filter.family == AF_PACKET)
| | ~ ~~~~~~~~~~~~~
| | | |
| | | (7) ...to here
| | (8) following ‘true’ branch...
| 276 | read_dev_mcast(&list);
| | ~~~~~~~~~~~~~~~~~~~~~
| | |
| | (9) ...to here
| | (10) calling ‘read_dev_mcast’ from ‘multiaddr_list’
|
+--> ‘read_dev_mcast’: events 11-12
|
| 82 | static void read_dev_mcast(struct ma_info **result_p)
| | ^~~~~~~~~~~~~~
| | |
| | (11) entry to ‘read_dev_mcast’
|......
| 87 | if (!fp)
| | ~
| | |
| | (12) following ‘false’ branch (when ‘fp’ is non-NULL)...
|
‘read_dev_mcast’: event 13
|
|cc1:
| (13): ...to here
|
‘read_dev_mcast’: events 14-17
|
| 90 | while (fgets(buf, sizeof(buf), fp)) {
| | ^~~~~
| | |
| | (14) following ‘true’ branch...
| 91 | char hexa[256];
| 92 | struct ma_info m = { .addr.family = AF_PACKET };
| | ~
| | |
| | (15) ...to here
|......
| 103 | struct ma_info *ma = malloc(sizeof(m));
| | ~~~~~~~~~~~~~~~~~
| | |
| | (16) this call could return NULL
| 104 |
| 105 | memcpy(ma, &m, sizeof(m));
| | ~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (17) ‘ma’ could be NULL: unchecked value from (16)
|
ipmaddr.c: In function ‘read_igmp’:
ipmaddr.c:152:17: warning: dereference of possibly-NULL ‘ma’ [CWE-690] [-Wanalyzer-possible-null-dereference]
152 | memcpy(ma, &m, sizeof(m));
| ^~~~~~~~~~~~~~~~~~~~~~~~~
‘do_multiaddr’: events 1-4
|
| 354 | int do_multiaddr(int argc, char **argv)
| | ^~~~~~~~~~~~
| | |
| | (1) entry to ‘do_multiaddr’
| 355 | {
| 356 | if (argc < 1)
| | ~
| | |
| | (2) following ‘true’ branch (when ‘argc <= 0’)...
| 357 | return multiaddr_list(0, NULL);
| | ~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (3) ...to here
| | (4) calling ‘multiaddr_list’ from ‘do_multiaddr’
|
+--> ‘multiaddr_list’: events 5-10
|
| 255 | static int multiaddr_list(int argc, char **argv)
| | ^~~~~~~~~~~~~~
| | |
| | (5) entry to ‘multiaddr_list’
|......
| 262 | while (argc > 0) {
| | ~~~~~~~~
| | |
| | (6) following ‘false’ branch (when ‘argc <= 0’)...
|......
| 275 | if (!filter.family || filter.family == AF_PACKET)
| | ~~~~~~~~~~~~~
| | |
| | (7) ...to here
| 276 | read_dev_mcast(&list);
| 277 | if (!filter.family || filter.family == AF_INET)
| | ~
| | |
| | (8) following ‘true’ branch...
| 278 | read_igmp(&list);
| | ~~~~~~~~~~~~~~~~
| | |
| | (9) ...to here
| | (10) calling ‘read_igmp’ from ‘multiaddr_list’
|
+--> ‘read_igmp’: events 11-14
|
| 116 | static void read_igmp(struct ma_info **result_p)
| | ^~~~~~~~~
| | |
| | (11) entry to ‘read_igmp’
|......
| 126 | if (!fp)
| | ~
| | |
| | (12) following ‘false’ branch (when ‘fp’ is non-NULL)...
| 127 | return;
| 128 | if (!fgets(buf, sizeof(buf), fp)) {
| | ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | | |
| | | (13) ...to here
| | (14) following ‘false’ branch...
|
‘read_igmp’: event 15
|
|cc1:
| (15): ...to here
|
‘read_igmp’: events 16-19
|
| 133 | while (fgets(buf, sizeof(buf), fp)) {
| | ^~~~~
| | |
| | (16) following ‘true’ branch...
|......
| 136 | if (buf[0] != '\t') {
| | ~~~~~~
| | |
| | (17) ...to here
|......
| 151 | ma = malloc(sizeof(m));
| | ~~~~~~~~~~~~~~~~~
| | |
| | (18) this call could return NULL
| 152 | memcpy(ma, &m, sizeof(m));
| | ~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (19) ‘ma’ could be NULL: unchecked value from (18)
|
ipmaddr.c: In function ‘read_igmp6’:
ipmaddr.c:181:25: warning: dereference of possibly-NULL ‘ma’ [CWE-690] [-Wanalyzer-possible-null-dereference]
181 | memcpy(ma, &m, sizeof(m));
| ^~~~~~~~~~~~~~~~~~~~~~~~~
‘do_multiaddr’: events 1-4
|
| 354 | int do_multiaddr(int argc, char **argv)
| | ^~~~~~~~~~~~
| | |
| | (1) entry to ‘do_multiaddr’
| 355 | {
| 356 | if (argc < 1)
| | ~
| | |
| | (2) following ‘true’ branch (when ‘argc <= 0’)...
| 357 | return multiaddr_list(0, NULL);
| | ~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (3) ...to here
| | (4) calling ‘multiaddr_list’ from ‘do_multiaddr’
|
+--> ‘multiaddr_list’: events 5-10
|
| 255 | static int multiaddr_list(int argc, char **argv)
| | ^~~~~~~~~~~~~~
| | |
| | (5) entry to ‘multiaddr_list’
|......
| 262 | while (argc > 0) {
| | ~~~~~~~~
| | |
| | (6) following ‘false’ branch (when ‘argc <= 0’)...
|......
| 275 | if (!filter.family || filter.family == AF_PACKET)
| | ~~~~~~~~~~~~~
| | |
| | (7) ...to here
|......
| 279 | if (!filter.family || filter.family == AF_INET6)
| | ~
| | |
| | (8) following ‘true’ branch...
| 280 | read_igmp6(&list);
| | ~~~~~~~~~~~~~~~~~
| | |
| | (9) ...to here
| | (10) calling ‘read_igmp6’ from ‘multiaddr_list’
|
+--> ‘read_igmp6’: events 11-12
|
| 159 | static void read_igmp6(struct ma_info **result_p)
| | ^~~~~~~~~~
| | |
| | (11) entry to ‘read_igmp6’
|......
| 164 | if (!fp)
| | ~
| | |
| | (12) following ‘false’ branch (when ‘fp’ is non-NULL)...
|
‘read_igmp6’: event 13
|
|cc1:
| (13): ...to here
|
‘read_igmp6’: events 14-17
|
| 167 | while (fgets(buf, sizeof(buf), fp)) {
| | ^~~~~
| | |
| | (14) following ‘true’ branch...
| 168 | char hexa[256];
| 169 | struct ma_info m = { .addr.family = AF_INET6 };
| | ~
| | |
| | (15) ...to here
|......
| 179 | struct ma_info *ma = malloc(sizeof(m));
| | ~~~~~~~~~~~~~~~~~
| | |
| | (16) this call could return NULL
| 180 |
| 181 | memcpy(ma, &m, sizeof(m));
| | ~~~~~~~~~~~~~~~~~~~~~~~~~
| | |
| | (17) ‘ma’ could be NULL: unchecked value from (16)
|
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Tue, 9 May 2023 01:38:10 +0000 (18:38 -0700)]
lib/fs: fix file leak in task_get_name
Fixes the problem identified -fanalyzer.
Why did rdma choose to reimplement the same function as
exiting glibc pthread_getname().
fs.c: In function ‘get_task_name’:
fs.c:355:12: warning: leak of FILE ‘f’ [CWE-775] [-Wanalyzer-file-leak]
355 | if (!fgets(name, len, f))
| ^
‘get_task_name’: events 1-9
|
| 345 | if (!pid)
| | ^
| | |
| | (1) following ‘false’ branch (when ‘pid != 0’)...
|......
| 348 | if (snprintf(path, sizeof(path), "/proc/%d/comm", pid) >= sizeof(path))
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | ||
| | |(2) ...to here
| | (3) following ‘false’ branch...
|......
| 351 | f = fopen(path, "r");
| | ~~~~~~~~~~~~~~~~
| | |
| | (4) ...to here
| | (5) opened here
| 352 | if (!f)
| | ~
| | |
| | (6) assuming ‘f’ is non-NULL
| | (7) following ‘false’ branch (when ‘f’ is non-NULL)...
|......
| 355 | if (!fgets(name, len, f))
| | ~ ~~~~~~~~~~~~~~~~~~~
| | | |
| | | (8) ...to here
| | (9) following ‘true’ branch...
|
‘get_task_name’: event 10
|
|cc1:
| (10): ...to here
|
‘get_task_name’: event 11
|
| 355 | if (!fgets(name, len, f))
| | ^
| | |
| | (11) ‘f’ leaks here; was opened at (5)
|
fs.c:355:12: warning: leak of ‘f’ [CWE-401] [-Wanalyzer-malloc-leak]
‘get_task_name’: events 1-9
|
| 345 | if (!pid)
| | ^
| | |
| | (1) following ‘false’ branch (when ‘pid != 0’)...
|......
| 348 | if (snprintf(path, sizeof(path), "/proc/%d/comm", pid) >= sizeof(path))
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| | ||
| | |(2) ...to here
| | (3) following ‘false’ branch...
|......
| 351 | f = fopen(path, "r");
| | ~~~~~~~~~~~~~~~~
| | |
| | (4) ...to here
| | (5) allocated here
| 352 | if (!f)
| | ~
| | |
| | (6) assuming ‘f’ is non-NULL
| | (7) following ‘false’ branch (when ‘f’ is non-NULL)...
|......
| 355 | if (!fgets(name, len, f))
| | ~ ~~~~~~~~~~~~~~~~~~~
| | | |
| | | (8) ...to here
| | (9) following ‘true’ branch...
|
‘get_task_name’: event 10
|
|cc1:
| (10): ...to here
|
‘get_task_name’: event 11
|
| 355 | if (!fgets(name, len, f))
| | ^
| | |
| | (11) ‘f’ leaks here; was allocated at (5)
Fixes: 81bfd01a4c9e ("lib: move get_task_name() from rdma")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Wed, 10 May 2023 15:36:16 +0000 (08:36 -0700)]
Add MAINTAINERS file
Record the maintainers of subsections of iproute2.
The subtree maintainers are based off of most recent current
patches and maintainer of kernel portion of that subsystem.
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Acked-by: Petr Machata <me@pmachata.org> # For DCB
Acked-by: Parav Pandit <parav@nvidia.com>
Acked-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
zhaoshuang [Thu, 11 May 2023 00:37:26 +0000 (08:37 +0800)]
iproute2: optimize code and fix some mem-leak risk
Signed-off-by: zhaoshuang <izhaoshuang@163.com>
Reviewed-by: Pawel Chmielewski <pawel.chmielewski@intel.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Nicolas Dichtel [Thu, 11 May 2023 14:42:24 +0000 (16:42 +0200)]
ipnetns: fix fd leak with 'ip netns set'
There is no reason to open this netns file. set_netnsid_from_name() uses
netns_get_fd() for this purpose and uses the returned fd.
Reported-by: Stephen Hemminger <stephen@networkplumber.org>
Fixes: d182ee1307c7 ("ipnetns: allow to get and set netns ids")
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Thu, 11 May 2023 21:10:26 +0000 (14:10 -0700)]
ip-rule: more manual page grammer fixes
Add missing articles and replace use of passive voice.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Kamal Heib [Thu, 4 May 2023 12:09:18 +0000 (08:09 -0400)]
rdma: Report device protocol
Add support for reporting the device protocol.
11: mlx5_0: node_type ca protocol roce fw 12.28.2006
node_guid 248a:0703:004b:f094 sys_image_guid 248a:0703:004b:f094
12: mlx5_1: node_type ca protocol ib fw 12.28.2006
node_guid 248a:0703:0049:d4f0 sys_image_guid 248a:0703:0049:d4f0
13: mlx5_2: node_type ca protocol ib fw 12.28.2006
node_guid 248a:0703:0049:d4f1 sys_image_guid 248a:0703:0049:d4f0
19: siw0: node_type rnic protocol iw node_guid 0200:00ff:fe00:0000
sys_image_guid 0200:00ff:fe00:0000
Signed-off-by: Kamal Heib <kheib@redhat.com>
Acked-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: David Ahern <dsahern@kernel.org>
David Ahern [Wed, 10 May 2023 20:00:29 +0000 (14:00 -0600)]
Merge branch 'main' into next
Signed-off-by: David Ahern <dsahern@kernel.org>
Stephen Hemminger [Wed, 10 May 2023 15:16:40 +0000 (08:16 -0700)]
remove unnecessary checks for NULL before calling free()
The function free() handles the case wher argument is NULL
by doing nothing. So the extra checks are not needed.
Found by modified version of kernel coccinelle script.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Tue, 9 May 2023 21:22:20 +0000 (14:22 -0700)]
uapi: add capability.h
All kernel header files should come from local copy of sanitized
headers, rather than relying on what Linux distribution ships.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Mon, 8 May 2023 20:41:31 +0000 (13:41 -0700)]
uapi: update kernel headers 6.4-rc1
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Bilal Khan [Tue, 2 May 2023 09:44:32 +0000 (14:44 +0500)]
fixed the grammar in ip-rule(8) man page
a small grammatical error has been idenfied in the ip-rule(8) man page
Signed-off-by: Bilal Khan <bilalkhanrecovered@gmail.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Stephen Hemminger [Sat, 29 Apr 2023 19:10:38 +0000 (12:10 -0700)]
Merge git://git.kernel.org/pub/scm/network/iproute2/iproute2-next
Stephen Hemminger [Thu, 27 Apr 2023 15:55:24 +0000 (08:55 -0700)]
v6.3.0