Daniel Borkmann [Tue, 5 Sep 2017 00:24:32 +0000 (02:24 +0200)]
bpf: consolidate dumps to use bpf_dump_prog_info
Consolidate dump of prog info to use bpf_dump_prog_info() when possible.
Moving forward, we want to have a consistent output for BPF progs when
being dumped. E.g. in cls/act case we used to dump tag as a separate
netlink attribute before we had BPF_OBJ_GET_INFO_BY_FD bpf(2) command.
Move dumping tag into bpf_dump_prog_info() as well, and only dump the
netlink attribute for older kernels. Also, reuse bpf_dump_prog_info()
for XDP case, so we can dump tag and whether program was jited, which
we currently don't show.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Simon Horman [Tue, 5 Sep 2017 11:06:24 +0000 (13:06 +0200)]
tc actions: store and dump correct length of user cookies
Correct two errors which cancel each other out:
* Do not send twice the length of the actual provided by the user to the kernel
* Do not dump half the length of the cookie provided by the kernel
As the cookie is now stored in the kernel at its correct length rather
than double the that length cookies of up to the maximum size of 16 bytes
may now be stored rather than a maximum of half that length.
Output of dump is the same before and after this change,
but the data stored in the kernel is now exactly the cookie
rather than the cookie + as many trailing zeros.
Before:
# tc filter add dev eth0 protocol ip parent ffff: \
flower ip_proto udp action drop \
cookie 0123456789abcdef0123456789abcdef
RTNETLINK answers: Invalid argument
After:
# tc filter add dev eth0 protocol ip parent ffff: \
flower ip_proto udp action drop \
cookie 0123456789abcdef0123456789abcdef
# tc filter show dev eth0 ingress
eth_type ipv4
ip_proto udp
not_in_hw
action order 1: gact action drop
random type none pass val 0
index 1 ref 1 bind 1 installed 1 sec used 1 sec
Action statistics:
Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
backlog 0b 0p requeues 0
cookie len 16 0123456789abcdef0123456789abcdef
Fixes: fd8b3d2c1b9b ("actions: Add support for user cookies") Cc: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Simon Horman <simon.horman@netronome.com>
Phil Sutter [Tue, 29 Aug 2017 15:09:45 +0000 (17:09 +0200)]
lib/bpf: Fix bytecode-file parsing
The signedness of char type is implementation dependent, and there are
architectures on which it is unsigned by default. In that case, the
check whether fgetc() returned EOF failed because the return value was
assigned an (unsigned) char variable prior to comparison with EOF (which
is defined to -1). Fix this by using int as type for 'c' variable, which
also matches the declaration of fgetc().
While being at it, fix the parser logic to correctly handle multiple
empty lines and consecutive whitespace and tab characters to further
improve the parser's robustness. Note that this will still detect double
separator characters, so doesn't soften up the parser too much.
Fixes: 3da3ebfca85b8 ("bpf: Make bytecode-file reading a little more robust") Cc: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Phil Sutter <phil@nwl.cc> Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Michal Kubecek [Fri, 1 Sep 2017 16:39:16 +0000 (18:39 +0200)]
iplink: double the buffer size also in iplink_get()
Commit 72b365e8e0fd ("libnetlink: Double the dump buffer size") increased
the buffer size for "ip link show" command to 32 KB to handle NICs with
large number of VFs. With "dev" filter, a different code path is taken and
iplink_get() still uses only 16 KB buffer.
The size of 32768 is not very future-proof as NICs supporting 120-128 VFs
are already in use so that single RTM_NEWLINK message in the dump can
exceed 30000 bytes. But it's what rtnl_talk() and rtnl_dump_filter_l() use
so let's be consistent. Once this proves insufficient, all three sizes
should be increased.
Michal Kubecek [Fri, 1 Sep 2017 16:39:11 +0000 (18:39 +0200)]
iplink: check for message truncation in iplink_get()
If message length exceeds maxlen argument of rtnl_talk(), it is truncated
to maxlen but unlike in the case of truncation to the length of local
buffer in rtnl_talk(), the caller doesn't get any indication of a problem.
In particular, iplink_get() passes the truncated message on and parsing it
results in various warnings and sometimes even a segfault (observed with
"ip link show dev ..." for a NIC with 125 VFs).
Handle message truncation in iplink_get() the same way as truncation in
rtnl_talk() would be handled: return an error.
Phil Sutter [Fri, 1 Sep 2017 14:08:08 +0000 (16:08 +0200)]
link_gre6: Fix for changing tclass/flowlabel
When trying to change tclass or flowlabel of a GREv6 tunnel which has
the respective value set already, the code accidentally bitwise OR'ed
the old and the new value, leading to unexpected results. Fix this by
clearing the relevant bits of flowinfo variable prior to assigning the
new value.
Fixes: af89576d7a8c4 ("iproute2: GRE over IPv6 tunnel support.") Signed-off-by: Phil Sutter <phil@nwl.cc>
Phil Sutter [Mon, 28 Aug 2017 17:31:22 +0000 (19:31 +0200)]
ss: Fix for added diag support check
Commit 9f66764e308e9 ("libnetlink: Add test for error code returned from
netlink reply") changed rtnl_dump_filter_l() to return an error in case
NLMSG_DONE would contain one, even if it was ENOENT.
This in turn breaks ss when it tries to dump DCCP sockets on a system
without support for it: The function tcp_show(), which is shared between
TCP and DCCP, will start parsing /proc since inet_show_netlink() returns
an error - yet it parses /proc/net/tcp which doesn't make sense for DCCP
sockets at all.
On my system, a call to 'ss' without further arguments prints the list
of connected TCP sockets twice.
Fix this by introducing a dedicated function dccp_show() which does not
have a fallback to /proc, just like sctp_show(). And since tcp_show()
is no longer "multi-purpose", drop it's socktype parameter.
Fixes: 9f66764e308e9 ("libnetlink: Add test for error code returned from netlink reply") Signed-off-by: Phil Sutter <phil@nwl.cc>
Phil Sutter [Thu, 24 Aug 2017 09:41:30 +0000 (11:41 +0200)]
lib/fs: Fix and simplify make_path()
Calling stat() before mkdir() is racey: The entry might change in
between. Also, the call to stat() seems to exist only to check if the
directory exists already. So simply call mkdir() unconditionally and
catch only errors other than EEXIST.
Phil Sutter [Thu, 24 Aug 2017 09:41:26 +0000 (11:41 +0200)]
ss: Make struct tcpstat fields 'timer' and 'timeout' unsigned
Both 'timer' and 'timeout' variables of struct tcpstat are either
scanned as unsigned values from /proc/net/tcp{,6} or copied from
'idiag_timer' and 'idiag_expries' fields of struct inet_diag_msg, which
itself are unsigned. Therefore they may be unsigned as well, which
eliminates the need to check for negative values.
Phil Sutter [Thu, 24 Aug 2017 09:51:50 +0000 (11:51 +0200)]
lib/ll_map: Choose size of new cache items at run-time
Instead of having a fixed buffer of 16 bytes for the interface name,
tailor size of new ll_cache entry using the interface name's actual
length. This also makes sure the following call to strcpy() is safe.
Phil Sutter [Thu, 24 Aug 2017 09:51:49 +0000 (11:51 +0200)]
tc/m_xt: Fix for potential string buffer overflows
- Use strncpy() when writing to target->t->u.user.name and make sure the
final byte remains untouched (xtables_calloc() set it to zero).
- 'tname' length sanitization was completely wrong: If it's length
exceeded the 16 bytes available in 'k', passing a length value of 16
to strncpy() would overwrite the previously NULL'ed 'k[15]'. Also, the
sanitization has to happen if 'tname' is exactly 16 bytes long as
well.
Phil Sutter [Thu, 24 Aug 2017 09:51:48 +0000 (11:51 +0200)]
lnstat_util: Simplify alloc_and_open() a bit
Relying upon callers and using unsafe strcpy() is probably not the best
idea. Aside from that, using snprintf() allows to format the string for
lf->path in one go.
Phil Sutter [Thu, 24 Aug 2017 09:51:47 +0000 (11:51 +0200)]
lib/inet_proto: Review inet_proto_{a2n,n2a}()
The original intent was to make sure strings written by those functions
are NUL-terminated at all times, though it was suggested to get rid of
the 15 char protocol name limit as well which this patch accomplishes.
In addition to that, simplify inet_proto_a2n() a bit: Use the error
checking in get_u8() to find out whether passed 'buf' contains a valid
decimal number instead of checking the first character's value manually.
Phil Sutter [Thu, 24 Aug 2017 09:51:46 +0000 (11:51 +0200)]
lib/fs: Fix format string in find_fs_mount()
A field width of 4096 allows fscanf() to store that amount of characters
into the given buffer, though that doesn't include the terminating NULL
byte. Decrease the value by one to leave space for it.
Phil Sutter [Thu, 24 Aug 2017 09:51:45 +0000 (11:51 +0200)]
ipntable: Avoid memory allocation for filter.name
The original issue was that filter.name might end up unterminated if
user provided string was too long. But in fact it is not necessary to
copy the commandline parameter at all: just make filter.name point to it
instead.
ss: fix help/man TCP-STATE description for listening
There's some misleading information in --help and ss(8) manpage about
TCP-STATE named 'listen'.
ss doesn't know such a state, but it knows 'listening' state.
$ ss -tua state listen
ss: wrong state name: listen
$ ss -tua state listening
[...]
Addresses: https://bugs.debian.org/872990 Reported-by: Pavel Lyulchenko <p.lyulchenko@gmail.com> Signed-off-by: Andreas Henriksson <andreas@fatal.se>
Phil Sutter [Mon, 21 Aug 2017 16:36:52 +0000 (18:36 +0200)]
devlink: Check return code of strslashrsplit()
This function shouldn't fail because all callers of
__dl_argv_handle_port() make sure the passed string contains enough
slashes already, but better make sure if this changes in future the
function won't access uninitialized data.
Phil Sutter [Mon, 21 Aug 2017 09:27:00 +0000 (11:27 +0200)]
iplink_can: Prevent overstepping array bounds
can_state_names array contains at most CAN_STATE_MAX fields, so allowing
an index to it to be equal to that number is wrong. While here, also
make sure the array is indeed that big so nothing bad happens if
CAN_STATE_MAX ever increases.
Phil Sutter [Thu, 17 Aug 2017 17:09:31 +0000 (19:09 +0200)]
tc/m_gact: Drop dead code
The use of 'ok' variable in parse_gact() is ineffective: The second
conditional increments it either if *argv is 'gact' or if
parse_action_control() doesn't fail (in which case exit() is called).
So this is effectively an unconditional increment and since no decrement
happens anywhere, all remaining checks for 'ok != 0' can be dropped.
Phil Sutter [Thu, 17 Aug 2017 17:09:27 +0000 (19:09 +0200)]
iproute: Fix for missing 'Oifs:' display
Covscan complained about dead code but after reading it, I assume the
author's intention was to prefix the interface list with 'Oifs: '.
Initializing first to 1 and setting it to 0 after above prefix was
printed should fix it.
Leon Romanovsky [Sun, 20 Aug 2017 09:58:22 +0000 (12:58 +0300)]
rdma: Add basic infrastructure for RDMA tool
RDMA devices are cross-functional devices from one side,
but very tailored for the specific markets from another.
Such diversity caused to spread of RDMA related configuration
across various tools, e.g. devlink, ip, ethtool, ib specific and
vendor specific solutions.
This patch adds ability to fill device and port information
by reading RDMA netlink.
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Phil Sutter [Thu, 17 Aug 2017 17:09:30 +0000 (19:09 +0200)]
iproute_lwtunnel: csum_mode value checking was ineffective
ila_csum_name2mode() returning -1 on error but being declared as
returning __u8 doesn't make much sense. Change the code to correctly
detect this issue. Checking for __u8 overruns shouldn't be necessary
though since ila_csum_name2mode() return values are well-defined.
Phil Sutter [Thu, 17 Aug 2017 17:09:31 +0000 (19:09 +0200)]
examples: Some shell fixes to cbq.init
This addresses the following issues:
- $@ is an array, so don't use it in quoted strings - use $* instead.
- Add missing quotes to components of [ ] expressions. These are not
strictly necessary since the output of 'wc -l' should be a single word
only, but in case of errors, bash prints "integer expression expected"
instead of "too many arguments".
- Use -print0/-0 when piping from find to xargs to allow for filenames
which contain whitespace.
- Quote arguments to 'eval' to prevent word-splitting.
Daniel Borkmann [Wed, 9 Aug 2017 22:15:41 +0000 (00:15 +0200)]
bpf: unbreak libelf linkage for bpf obj loader
Commit 69fed534a533 ("change how Config is used in Makefile's") moved
HAVE_MNL specific CFLAGS/LDLIBS for building with libmnl out of the
top level Makefile into sub-Makefiles. However, it also removed the
HAVE_ELF specific CFLAGS/LDLIBS entirely, which breaks the BPF object
loader for tc and ip with "No ELF library support compiled in." despite
having libelf detected in configure script. Fix it similarly as in 69fed534a533 for HAVE_ELF.
Fixes: 69fed534a533 ("change how Config is used in Makefile's") Reported-by: Jeffrey Panneman <jeffrey.panneman@tno.nl> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
David Ahern [Wed, 9 Aug 2017 15:43:27 +0000 (08:43 -0700)]
lib: Dump ext-ack string by default
In time, errfn can be implemented for link, route, etc commands to
give a much more detailed response (e.g., point to the attribute
that failed). Doing so is much more complicated to process the
message and convert attribute ids to names.
In any case the error string returned by the kernel should be dumped
to the user, so make that happen now.
The ikey and okey value are normal u32 values. The input accepts
them in dotted, hex or decimal form. For output, hex seems like
the best form since they are not really addresses.
Suggested-by: Christian Langrock <christian.langrock@secunet.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
The recent LIBMNL changes was made more difficult to debug because
of how Config is handle in clean make. The Config file is generated
by top level make, but since it is not recursive, the values generated
would not be visible on a clean make.
The change is to not include Config in top level make, and move
all the conditionals down into sub makefiles. Not ideal, but beter
than going full autoconf route. Or forcing separate configure
step.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
According to the IPv4 behavior of 'ip' it should be possible
to omit the arguments for local and remote address.
Without this patch omitting these parameters would lead to
uninitialized memory being interpreted as IPv6 addresses.
Reported-by: Christian Langrock <christian.langrock@secunet.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
When ip netns {add|delete} is first run, it bind-mounts /var/run/netns
on top of itself, then marks it as shared. However, if there are already
bind-mounts in the directory from other tools, these would not be
propagated. Fix this by recursively bind-mounting.
Signed-off-by: Casey Callendrello <casey.callendrello@coreos.com> Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
iproute: Add support for extended ack to rtnl_talk
Add support for extended ack error reporting via libmnl.
Add a new function rtnl_talk_extack that takes a callback as an input
arg. If a netlink response contains extack attributes, the callback is
is invoked with the the err string, offset in the message and a pointer
to the message returned by the kernel.
If iproute2 is built without libmnl, it will still work but
extended error reports from kernel will not be available.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Phil Sutter [Tue, 1 Aug 2017 16:36:11 +0000 (18:36 +0200)]
Really fix get_addr() and get_prefix() error messages
Both functions take the desired address family as a parameter. So using
that to notify the user what address family was expected is correct,
unlike using dst->family which will tell the user only what address
family was specified.
The situation which commit 334af76143368 tried to fix was when 'ip'
would accept addresses from multiple families. In that case, the family
parameter is set to AF_UNSPEC so that get_addr_1() may accept any valid
address.
This patch introduces a wrapper around family_name() which returns the
string "any valid" for AF_UNSPEC instead of the three question marks
unsuitable for use in error messages.
Tests for AF_UNSPEC:
| # ip a a 256.10.166.1/24 dev d0
| Error: any valid prefix is expected rather than "256.10.166.1/24".
| # ip neighbor add proxy 2001:db8::g dev d0
| Error: any valid address is expected rather than "2001:db8::g".
Tests for explicit address family:
| # ip -6 addrlabel add prefix 1.1.1.1/24 label 123
| Error: inet6 prefix is expected rather than "1.1.1.1/24".
| # ip -4 addrlabel add prefix dead:beef::1/24 label 123
| Error: inet prefix is expected rather than "dead:beef::1/24".
Reported-by: Jaroslav Aster <jaster@redhat.com> Fixes: 334af76143368 ("fix get_addr() and get_prefix() error messages") Signed-off-by: Phil Sutter <phil@nwl.cc>
Phil Sutter [Wed, 2 Aug 2017 12:57:56 +0000 (14:57 +0200)]
bpf: Make bytecode-file reading a little more robust
bpf_parse_string() will now correctly handle:
- Extraneous whitespace,
- OPs on multiple lines and
- overlong file names.
The added feature of allowing to have OPs on multiple lines (like e.g.
tcpdump prints them) is rather a side effect of fixing detection of
malformed bytecode files having random content on a second line, like
e.g.:
Hangbin Liu [Thu, 27 Jul 2017 09:44:15 +0000 (17:44 +0800)]
utils: return default family when rtm_family is not RTNL_FAMILY_IPMR/IP6MR
When we get a multicast route, the rtm_type is RTN_MULTICAST, but the
rtm_family may be AF_INET. If we only check the type with RTNL_FAMILY_IPMR,
we will get malformed address. e.g.
+ ip -4 route add multicast 172.111.1.1 dev em1 table main
Before fix:
+ ip route list type multicast table main
multicast ac6f:101:800:400:400:0:3c00:0 dev em1 scope link
After fix:
+ ip route list type multicast table main
multicast 172.111.1.1 dev em1 scope link
Fixes: 56e3eb4c3400 ("ip: route: fix multicast route dumps") Signed-off-by: Hangbin Liu <liuhangbin@gmail.com> Acked-by: Phil Sutter <phil@nwl.cc>
ip netns accepts invalid input as namespace name like an empty string or a
string longer than the maximum file name length.
Check that the netns name is not empty and less than or equal to NAME_MAX.
Ability to change geneve device attributes was added to kernel through
commit 5b861f6baa3a ("geneve: add rtnl changelink support"), however one
cannot do the same through ip-link(8) command. Changing the allowed
geneve device attributes using 'ip link set <geneve_name> type geneve id
<geneve_id> <allowed_attributes>' currently fails with 'operation not
supported' error. This patch adds support for it.
Daniel Borkmann [Sat, 22 Jul 2017 23:22:18 +0000 (01:22 +0200)]
bpf: improve error reporting around tail calls
Currently, it's still quite hard to figure out if a prog passed the
verifier, but later gets rejected due to different tail call ownership.
Figure out whether that is the case and provide appropriate error
messages to the user.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>