Dpipe - Each dpipe table can have one resource which is mapped to it.
The resource is presented via its full path. Furthermore, the number
of units consumed by single table entry is presented.
Resource - Each resource presents the dpipe tables that use it.
devlink: Add support for devlink resource abstraction
Add support for devlink resource abstraction. The resources are
represented by a tree based structure and are identified by a name and
a size. Some resources can present their real time occupancy.
First the resources exposed by the driver can be observed, for example:
$devlink resource show pci/0000:03:00.0
pci/0000:03:00.0:
name kvd size 245760 unit entry
resources:
name linear size 98304 occ 0 unit entry size_min 0 size_max 147456 size_gran 128
name hash_double size 60416 unit entry size_min 32768 size_max 180224 size_gran 128
name hash_single size 87040 unit entry size_min 65536 size_max 212992 size_gran 128
Some resource's size can be changed. Examples:
$devlink resource set pci/0000:03:00.0 path /kvd/hash_single size 73088
$devlink resource set pci/0000:03:00.0 path /kvd/hash_double size 74368
The changes do not apply immediately, this can be validate by the 'size_new'
attribute, which represents the pending changed size. For example
$devlink resource show pci/0000:03:00.0
pci/0000:03:00.0:
name kvd size 245760 unit entry size_valid false
resources:
name linear size 98304 size_new 147456 occ 0 unit entry size_min 0 size_max 147456 size_gran 128
name hash_double size 60416 unit entry size_min 32768 size_max 180224 size_gran 128
name hash_single size 87040 unit entry size_min 65536 size_max 212992 size_gran 128
In case of a pending change the nested resources present an indication
for a valid configuration of its children (sum of its children sizes
doesn't exceed the parent's size).
In order for the changes to take place hot reload is needed. The hot
reload through devlink will be introduced in the following patch.
Quentin Monnet [Thu, 22 Feb 2018 03:22:14 +0000 (19:22 -0800)]
README: re-add updated information link
The "Information" link was removed from README file in commit d7843207e6fd ("README: update location of git repositories, remove
broken info link"), because it redirected to a page that no longer
existed on the Linux Foundation wiki.
This page has just been restored, so we can add the link back again.
Since the previous link was a redirection, use the updated link instead.
Thanks to Luca Boccassi for investigating this issue, restoring and
updating the page.
Vincent Bernat [Tue, 20 Feb 2018 23:28:04 +0000 (00:28 +0100)]
color: disable color when json output is requested
Instead of declaring -color and -json exclusive, ignore -color when
-json is provided. The rationale is to allow to put -color in an alias
for ip while still being able to use -json. -color is merely a
presentation suggestion and we can assume there is nothing to color in
the JSON output.
Signed-off-by: Vincent Bernat <vincent@bernat.im> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Phil Sutter [Fri, 9 Feb 2018 17:49:38 +0000 (18:49 +0100)]
Remove leftovers from removed Latex documentation
Since there is no documentation in Latex format left, there is no need
to check for commands to build it. Also there is no need to ignore any
of the temporary files which were created by them.
Quentin Monnet [Fri, 9 Feb 2018 17:11:09 +0000 (09:11 -0800)]
README: update location of git repositories, remove broken info link
Reflect the recent change of location for the git repositories, and the
creation of the -next development repo, in README and README.devel.
Also remove the link to the Linux Foundation wiki that contained
information about iproute2. The link is now broken, I did not find any
alternative page to point to.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com> Acked-by: David Ahern <dsahern@gmail.com>
If the kernel receives a negative nsid it will automatically assign
the next available nsid. In this case alloc_netid() will set min and
max to 0 for ird_alloc(). And when max == 0 idr_alloc() will interpret
this as the maximum range, i.e. specific to nsids it will try to find
an id in the range [0,INT_MAX). This is intentionally supported in the
kernel for nsids.
Commit acbe9118ce80 ("ip netns: use strtol() instead of atoi()")
regressed ip netns in that respect although previously the use-case
was either accidentally supported or opaquely supported such that it
triggered the original commit. From what I can gather it went as
follows before: atoi() was called with a string indicating a negative
value which caused it to return -1 which was passed to the
kernel. Let's make it less opaque by introducing the keyword "auto":
ip netns set <netns-name> auto
will cause nsid to be set to -1 and the kernel will select an available
nsid.
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Leon Romanovsky [Wed, 31 Jan 2018 08:11:54 +0000 (10:11 +0200)]
rdma: Add QP resource tracking information
This patch adds ss-similar interface to view various resource
tracked objects. At this stage, only QP is presented.
1. Get all QPs for the specific device:
$ rdma res show qp link mlx5_4
link mlx5_4/- lqpn 8 type UD state RESET sq-psn 0 pid 0 comm [ib_ipoib]
link mlx5_4/1 lqpn 7 type UD state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 1 type GSI state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 0 type SMI state RTS sq-psn 0 pid 0 comm [ib_core]
$ rdma res show qp link mlx5_4/
link mlx5_4/- lqpn 8 type UD state RESET sq-psn 0 pid 0 comm [ib_ipoib]
link mlx5_4/1 lqpn 7 type UD state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 1 type GSI state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 0 type SMI state RTS sq-psn 0 pid 0 comm [ib_core]
2. Provide illegal port number (0 is illegal):
$ rdma res show qp link mlx5_4/0
Wrong device name
3. Get QPs of specific port:
$ rdma res show qp link mlx5_4/1
link mlx5_4/1 lqpn 7 type UD state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 1 type GSI state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 0 type SMI state RTS sq-psn 0 pid 0 comm [ib_core]
4. Get QPs which have not assigned port yet:
link mlx5_4/- lqpn 8 type UD state RESET sq-psn 0 pid 0 comm [ib_ipoib]
5. Limit to specific Local QPNs:
$ rdma res show qp link mlx5_4/1 lqpn 1-3,7
link mlx5_4/1 lqpn 7 type UD state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 1 type GSI state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 0 type SMI state RTS sq-psn 0 pid 0 comm [ib_core]
. Filter types (strings):
$ rdma res show qp link mlx5_4/1 type UD,gSi
link mlx5_4/1 lqpn 7 type UD state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 1 type GSI state RTS sq-psn 0 pid 0 comm [ib_core]
Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Leon Romanovsky [Wed, 31 Jan 2018 08:11:49 +0000 (10:11 +0200)]
rdma: Add filtering infrastructure
This patch adds general infrastructure to RDMAtool to handle various
filtering options needed for the downstream resource tracking patches.
The infrastructure is generic and stores filters in list of key<->value
entries. There are three types of filters:
1. Numeric - the values are intended to be digits combined with '-' to
mark range and ',' to mark multiple entries, e.g. pid 1-100,234,400-401
is perfectly legit filter to limit process ids.
2. String - the values are consist from strings and "," as a denominator.
3. Link - special case to allow '/' in string to provide link name, e.g.
link mlx4_1/2.
Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Leon Romanovsky [Wed, 31 Jan 2018 08:11:47 +0000 (10:11 +0200)]
rdma: Add option to provide "-" sign for the port number
According to the IBTA spec [1], the physical connected port is provided
for the QP in RTR-to-INIT stage performed by modify_qp(). It causes
to do not have port number for newly created QPs.
The following patch adds "-" sign to present absence of port, because
QPs are going to be associated with rdmatool link object, which needs
port number as an index.
Jakub Kicinski [Sat, 27 Jan 2018 09:19:04 +0000 (01:19 -0800)]
tc: fix second printing of requeues
Non-JSON tc qdisc output used to print the "requeues" statistic
twice. Commit 4fcec7f3665b ("tc: jsonify stats2") tried to preserve
this behaviour for both standard output and JSON, but used the wrong
statistic (q.qlen). Also duplicating keys in JSON is not allowed,
so the second occurrence should be completely skipped with JSON.
Fixes: 4fcec7f3665b ("tc: jsonify stats2") Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Jakub Kicinski [Fri, 26 Jan 2018 19:30:35 +0000 (11:30 -0800)]
ip: address: fix stats64 JSON object name
The JSON object name for statistics in ip link show is "stats644".
Looks like a typo, commit d0e720111aad ("ip: ipaddress.c: add support
for json output") contains an example with the expected "stats64" name.
The fact that no one has noticed until now is probably an indication
that no one is using this object. Hopefully it's not too late to fix
this, although IIUC this has already been in 4.13 and 4.14 releases :S
Fixes: d0e720111aad ("ip: ipaddress.c: add support for json output") Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Jakub Kicinski [Fri, 26 Jan 2018 19:27:56 +0000 (11:27 -0800)]
tc: red: JSON-ify RED output
Make JSON output work with RED Qdiscs. Float/double printing
helpers have to be added/uncommented to print the probability.
Since TC stats in general are not split out to a separate object
the xstats printed by this patch are not separated either.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David Ahern <dsahern@gmail.com>
David Ahern [Thu, 25 Jan 2018 17:32:27 +0000 (09:32 -0800)]
Merge branch 'get_addr_rta' into iproute2-next
Serhey Popovych says:
====================
Now we enhance get_addr() to return additional information about address
(e.g. if it unspecified or multicast) we want to have same functionality
for attributes in netlink message.
Introduce and use get_addr_rta() that parses given netlink attribute
into @inet_prefix data structure in the same way similar get_addr()
parses address from it's string representation.
Use attribute length to guess address family: force it by giving non
AF_UNSPEC @family to get_addr_rta() to ensure address is of expected
family.
Introduce and use inet_addr_match_rta() to further simplify and unify
code where get_addr_rta() intended to be used together with
inet_addr_match().
This is next step in ipv4 and ipv6 modules unification to prepare for
merge in the future.
Introduce and use tnl_print_endpoint() helper to print of tunnel
endpoint address.
Note that for AF_INET and AF_INET6 inet_ntop(3) is used that may return
NULL in case of failure and while unlikely format_host_rta() might
return NULL too. Handle this case when passing local/remote to
print_string().
Signed-off-by: Serhey Popovych <serhe.popovych@gmail.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Serhey Popovych [Wed, 24 Jan 2018 18:56:33 +0000 (20:56 +0200)]
utils: Introduce get_addr_rta() and inet_addr_match_rta()
First is used to get address from netlink attribute to
inet_prefix data structure. Use memcpy() with constant
value to let complier optimize by replacing a call by
inlining load/store instructions.
Second is used to match address in given netlink attribute
with one given as reference. It matches successfully if
no attribute is given (@rta is NULL), reference address
family is AF_UNSPEC or it's length isn't given; fails if
get_attr_rta() can't get attribute or it's family does
not match reference; calls inet_addr_match() to get final
verdict.
Signed-off-by: Serhey Popovych <serhe.popovych@gmail.com> Signed-off-by: David Ahern <dsahern@gmail.com>
David Ahern [Wed, 24 Jan 2018 18:02:27 +0000 (10:02 -0800)]
Merge branch 'unify_external' into iproute2-next
Serhey Popovych says:
====================
With this series I want to unify collect metadata
handling in tunnels:
1) Use "external" name for JSON and non-JSON output.
Do not *print* any options when tunnel in
collect metadata mode: gre6 already do
this, so just apply to others.
2) Do not *add* any attributes when configuring
gre tunnel in collect metadata mode.
Other tunnels (e.g. gre6, iptnl, ip6tnl)
alredy do that.
This is next step in ipv4 and ipv6 modules
unification to prepare for merge in the future.
Any comments, suggestions and criticism as always
welcome.
v2
For all tunnels implementing collect metadata
use "external" keyword for both JSON. Thanks
to Jiri Benc for detailed explanation.
====================
The lexer will go with the longest match, so previously
the starting double quotes of a string would be swallowed by
the [^ \t\r\n()]+ pattern leaving the user no way to
actually use strings with escape sequences.
Fix this by not allowing this case to start with double
quotes.
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
Serhey Popovych [Fri, 19 Jan 2018 16:44:03 +0000 (18:44 +0200)]
iplink: Use ll_name_to_index() instead of if_nametoindex()
While benefit from using ll_name_to_index() with populated
cache can potentially be exploited only in few places
(e.g. bridge fdb/mdb/vlan show routines) there is another
advantage of ll_name_to_index() over plain if_nametoindex():
in case of if_nametoindex() failure ll_name_to_index()
will attempt to get index from common name in form "if%d"
that may be returned from ll_index_to_name().
This makes output from ip(8) coherent with it's input.
Note that most of the code already switched from plain
if_nametoindex() to ll_name_to_index() to cached variant.
Signed-off-by: Serhey Popovych <serhe.popovych@gmail.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Serhey Popovych [Mon, 22 Jan 2018 14:50:08 +0000 (16:50 +0200)]
gre/gre6: Post merge fixes
Few minor changes after merge of 'master' into 'net-next' branch:
1) Follow 80 line length for printing erspan_index parameter
as we did in master with commit 2a8d0f6e9c3f ("gre/tunnel:
Print erspan_index using print_uint()").
2) Remove remnants of encapsulation option printing: now it
is done using tnl_print_encap() helper in commit bad76e6b1f44
("ip/tunnel: Abstract tunnel encapsulation options printing").
David Ahern [Sun, 21 Jan 2018 19:20:56 +0000 (11:20 -0800)]
Merge branch 'shared_block' into net-next
Jiri Pirko says:
====================
From: Jiri Pirko <jiri@mellanox.com>
Kernel allows to share all filters between qdiscs with use
of shared block.
Example:
block number 22. "22" is just an identification:
$ tc qdisc add dev ens7 ingress_block 22 ingress
^^^^^^^^^^^^^^^^
$ tc qdisc add dev ens8 ingress_block 22 ingress
^^^^^^^^^^^^^^^^
If we don't specify "block" command line option, no shared block would
be created:
$ tc qdisc add dev ens9 ingress
Now if we list the qdiscs, we will see the block index in the output:
$ tc qdisc
qdisc ingress ffff: dev ens7 parent ffff:fff1 ingress_block 22
qdisc ingress ffff: dev ens8 parent ffff:fff1 ingress_block 22
qdisc ingress ffff: dev ens9 parent ffff:fff1
To make is more visual, the situation looks like this:
Unlimited number of qdiscs may share the same block.
Block sharing is also supported for clsact qdisc:
$ tc qdisc add dev ens10 ingress_block 23 egress_block 24 clsact
$ tc qdisc show dev ens10
qdisc clsact ffff: dev ens10 parent ffff:fff1 ingress_block 23 egress_block 24
We can add filter using the block index:
$ tc filter add block 22 protocol ip pref 25 flower dst_ip 192.168.0.0/16 action drop
Note we cannot use the qdisc for filter manipulations of shared blocks:
$ tc filter add dev ens8 ingress protocol ip pref 1 flower dst_ip 192.168.100.2 action drop
Error: This filter block is shared. Please use the block index to manipulate the filters.
We will see the same output if we list filters for ingress qdisc of
ens7 and ens8, also for the block 22:
$ tc filter show block 22
filter protocol ip pref 25 flower chain 0
filter protocol ip pref 25 flower chain 0 handle 0x1
...
$ tc filter show dev ens7 ingress
filter block 22 protocol ip pref 25 flower chain 0
filter block 22 protocol ip pref 25 flower chain 0 handle 0x1
...
$ tc filter show dev ens8 ingress
filter block 22 protocol ip pref 25 flower chain 0
filter block 22 protocol ip pref 25 flower chain 0 handle 0x1
...
Jiri Pirko [Sat, 20 Jan 2018 10:00:29 +0000 (11:00 +0100)]
tc: implement ingress/egress block index attributes for qdiscs
During qdisc creation it is possible to specify shared block for bot
ingress and egress. Pass this values to kernel according to the command
line options.
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Jiri Pirko [Sat, 20 Jan 2018 10:00:28 +0000 (11:00 +0100)]
tc: introduce support for block-handle for filter operations
So far, qdisc was the only handle that could be used to manipulate
filters. Kernel added support for using block to manipulate it. So add
the support to use block index to manipulate filters. The magic
TCM_IFINDEX_MAGIC_BLOCK indicates the block index is in use.
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Jiri Pirko [Sat, 20 Jan 2018 10:00:27 +0000 (11:00 +0100)]
tc: introduce tc_qdisc_block_exists helper
This hepler used qdisc dump to list all qdisc and find if block index in
question is used by any of them. That means the block with specified
index exists.
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David Ahern <dsahern@gmail.com>
David Ahern [Sun, 21 Jan 2018 18:11:07 +0000 (10:11 -0800)]
Merge branch 'inet_get_addr' into net-next
Serhey Popovych says:
====================
It looks confusing to have multiple independent
routines to get internet address from it's string
representation: get_addr() and inet_get_addr().
Most complicated users of inet_get_addr() is
iplink_geneve.c and iplink_vxlan.c because they
required to handle both AF_INET and AF_INET6
for their local/remote endpoints.
On the other hand get_addr() does not provide
additional information like address type: need
to address this. to get rid of current and
possible future code duplications. Note that
this functionality is first step to make proto
independent handling of local/remote endpoints
in ip/tunnel code (there will be additional
series based on this one).
Also fix get_addr_1() and get_prefix() to make
sure it always provide correct ->family and
->bitlen.
As always comments, suggestions and criticism
are welcome.
Serhey Popovych [Thu, 18 Jan 2018 18:13:47 +0000 (20:13 +0200)]
ip: Get rid of inet_get_addr()
Both geneve and vxlan modules are converted to
use get_addr() we can replace inet_get_addr()
in less problematic places and finally get
rid of inet_get_addr().
Signed-off-by: Serhey Popovych <serhe.popovych@gmail.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Serhey Popovych [Thu, 18 Jan 2018 18:13:44 +0000 (20:13 +0200)]
utils: Fast inet address classification after get_addr()
It looks very useful to receive additional information
from get_addr_1() and get_addr() about address to simplify
caller and get rid of code duplications.
For now following information can be returned:
1) address is unspecified (zero)
2) address is multicast
3) address is internet: family is either AF_INET or
AF_INET6.
More information can be added in the future.
Introduce inline helpers to make code using this new
address classification interface more self explaining:
bool is_addrtype_inet(inet_prefix *addr)
true if @addr is inet address
bool is_addrtype_inet_unspec(inet_prefix *addr)
true if @addr is unspecified inet address
bool is_addrtype_inet_multi(inet_prefix *addr)
true if @addr is multicast inet address
bool is_addrtype_inet_not_unspec(inet_prefix *addr)
true if @addr is not unspecified inet address
false if @addr is not inet or unspecified inet
bool is_addrtype_inet_not_multi(inet_prefix *addr)
true if @addr is not multicast inet address
false if @addr is not inet or multicast inet
Last two are useful for case when we need inet address
that is not unspecified or multicast.
Signed-off-by: Serhey Popovych <serhe.popovych@gmail.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Serhey Popovych [Thu, 18 Jan 2018 18:13:43 +0000 (20:13 +0200)]
utils: Always specify family and ->bytelen in get_prefix_1()
Handle default/all/any special case in get_addr_1() to setup
->family and ->bytelen correctly.
Make get_addr_1() return ->bitlen == -2 instead of -1 to
distinguish default/all/any special case from the rest:
it is safe because all callers check ->bitlen < 0, not
explicit value -1.
Reduce intendation by one level and get rid of goto/label
to make code more readable.
Signed-off-by: Serhey Popovych <serhe.popovych@gmail.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Serhey Popovych [Thu, 18 Jan 2018 18:13:42 +0000 (20:13 +0200)]
utils: Always specify family for address in get_addr_1()
Set ->family correctly when string representing address
is "default", "all" or "any": get_addr_1() might be called
with AF_UNSPEC (e.g. get_addr() -> get_addr_1()).
Extend support for zero address to all address families,
not only AF_INET and AF_INET6 when one explicitly given
as @family: use af_byte_len() to correctly set address length.
Still assume AF_INET when @family is AF_UNSPEC.
Signed-off-by: Serhey Popovych <serhe.popovych@gmail.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Jakub Kicinski [Wed, 17 Jan 2018 07:50:54 +0000 (23:50 -0800)]
bpf: support map offload
When program is loaded with a specified ifindex, use that
ifindex also when creating maps.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: David Ahern <dsahern@gmail.com>
Jakub Kicinski [Tue, 16 Jan 2018 23:08:50 +0000 (15:08 -0800)]
tc: red: allow setting th_min and th_max to the same value
Setting th_min and th_max to the same value may be useful for DCTCP
deployments. The original DCTCP paper describes it as a simplest way
of achieving simple ECN threshold marking. Indeed, there doesn't seem
to be any simpler qdisc in Linux which would allow such a setup today.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Serhey Popovych [Thu, 18 Jan 2018 14:04:36 +0000 (16:04 +0200)]
tunnel: Return constant string without copying it
We return constant string from tnl_strproto(), no need
to copy it to temporary buffer and then return such
buffer as const: return constant string instead.
Signed-off-by: Serhey Popovych <serhe.popovych@gmail.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Get rid of code duplications and consolidate encapsulation
options printing in single function - tnl_print_encap().
Introduce and use tnl_encap_str() to format encapsulation
option string according to tempate and given values to avoid
code duplication and simplify it.
Use print_string() instead of fputs() and fprintf() to
print encapsulation for !is_json_context().
Print "unknown" parameter for "encap" type in PRINT_FP
context using "%s " format specifier and benefit from
complite time string merge.
Signed-off-by: Serhey Popovych <serhe.popovych@gmail.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Serhey Popovych [Thu, 18 Jan 2018 14:04:30 +0000 (16:04 +0200)]
ip/tunnel: Simplify and unify tos printing
For ip tunnels tos can be 0 when not configured, 1 when
inherited from encapsulated packet and rest specifying
diffserv (rfc2474) or tos (rfc1349) bits. It is stored
in packet tos/diffserv field and returned in tos
netlink attribute to userspace.
Simplify and unify tos printing by using print_0xhex()
and print_string() instead of fprintf() to output values.
Signed-off-by: Serhey Popovych <serhe.popovych@gmail.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Serhey Popovych [Thu, 18 Jan 2018 14:04:29 +0000 (16:04 +0200)]
ip/tunnel: Correct and unify ttl/hoplimit printing
Both ttl/hoplimit is from 1 to 255. Zero has special meaning:
use encapsulated packet value. In ip-link(8) -d output this
looks like "ttl/hoplimit inherit". In JSON we have "int" type
for ttl and therefore values from 0 (inherit) to 255.
To do the best in handling ttl/hoplimit we need to accept
both cases: missing attribute in netlink dump and zero value
for "inherit"ed case. Last one is broken since JSON output
introduction for gre/iptnl versions and was never true for
gre6/ip6tnl.
For all tunnels, except ip6tnl change JSON type from "int" to
"uint" to reflect true nature of the ttl.
Signed-off-by: Serhey Popovych <serhe.popovych@gmail.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Serhey Popovych [Thu, 18 Jan 2018 14:04:28 +0000 (16:04 +0200)]
iplink: Use ll_index_to_name() instead of if_indextoname()
There are two reasons for switching to cached variant:
1) ll_index_to_name() may return result from cache,
eliminating expensive ioctl() to the kernel.
Note that most of the code already switched from plain
if_indextoname() to ll_index_to_name() to cached variant
in print path because in most cases cache populated.
2) It always return name in the form "if%d", even if
entry is not in cache and ioctl() fails. This drops
"link_index" from JSON output.
Signed-off-by: Serhey Popovych <serhe.popovych@gmail.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Gal Pressman [Tue, 16 Jan 2018 13:41:59 +0000 (15:41 +0200)]
ipaddress: Make sure VF min/max rate API is supported before using it
When using the new minimum rate API and providing only one parameter
(minimum rate/maximum rate), we query the VF min and max rate regardless
of kernel support.
This resulted in segmentation fault in ipaddr_loop_each_vf, which tries
to access NULL pointer.
This patch identifies such cases by testing the VF table for NULL
pointer in IFLA_VF_RATE, and aborts the operation.
Aborting on the first VF is valid since if the kernel does not support
the new API for the first VF, it will not support it for the other VFs
as well.
Fixes: f89a2a05ffa9 ("Add support to configure SR-IOV VF minimum and maximum Tx rate through ip tool") Cc: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com> Signed-off-by: Gal Pressman <galp@mellanox.com> Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com> Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Gal Pressman [Tue, 16 Jan 2018 13:41:58 +0000 (15:41 +0200)]
iplink: Validate minimum tx rate is less than maximum tx rate
According to the documentation (man ip-link), the minimum TXRATE should
be always <= Maximum TXRATE, but commit f89a2a05ffa9 ("Add support to
configure SR-IOV VF minimum and maximum Tx rate through ip tool") didn't
enforce it.
Fixes: f89a2a05ffa9 ("Add support to configure SR-IOV VF minimum and maximum Tx rate through ip tool") Cc: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com> Signed-off-by: Gal Pressman <galp@mellanox.com> Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com> Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Phil Sutter [Fri, 12 Jan 2018 11:20:21 +0000 (12:20 +0100)]
tc: Optimize gact action lookup
When adding a filter with a gact action such as 'drop', tc first tries
to open a shared object with equivalent name (m_drop.so in this case)
before trying gact. Avoid this by matching the action name against those
handled by gact prior to calling get_action_kind().
Cc: Jiri Pirko <jiri@mellanox.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: David Ahern <dsahern@gmail.com> Signed-off-by: Phil Sutter <phil@nwl.cc>
David Ahern [Mon, 15 Jan 2018 16:25:30 +0000 (08:25 -0800)]
Merge branch 'tc-batch' into net-next
Chris Mi says:
====================
Currently in tc batch mode, only one command is read from the batch
file and sent to kernel to process. With this patchset, at most 128
commands can be accumulated before sending to kernel.
We introduced a new function in patch 1 to support for sending
multiple messages. In patch 2, we add this support for filter
add/delete/change/replace and actions add/change/replace commands.
But please note that kernel still processes the requests one by one.
To process the requests in parallel in kernel is another effort.
The time we're saving in this patchset is the user mode and kernel mode
context switch. So this patchset works on top of the current kernel.
Using the following script in kernel, we can generate 1,000,000 rules.
tools/testing/selftests/tc-testing/tdc_batch.py
Without this patchset, 'tc -b $file' exection time is:
real 0m15.555s
user 0m7.211s
sys 0m8.284s
With this patchset, 'tc -b $file' exection time is:
real 0m12.360s
user 0m6.082s
sys 0m6.213s
The insertion rate is improved more than 10%.
====================
Chris Mi [Fri, 12 Jan 2018 05:13:16 +0000 (14:13 +0900)]
tc: Add batchsize feature for filter and actions
Currently in tc batch mode, only one command is read from the batch
file and sent to kernel to process. With this support, at most 128
commands can be accumulated before sending to kernel.
Now it only works for the following successive commands:
1. filter add/delete/change/replace
2. actions add/change/replace
Signed-off-by: Chris Mi <chrism@mellanox.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Chris Mi [Fri, 12 Jan 2018 05:13:15 +0000 (14:13 +0900)]
lib/libnetlink: Add a new function rtnl_talk_iov
rtnl_talk can only send a single message to kernel. Add a new function
rtnl_talk_iov that can send multiple messages to kernel.
rtnl_talk_iov takes struct iovec * and iovlen as arguments.
Signed-off-by: Chris Mi <chrism@mellanox.com> Signed-off-by: David Ahern <dsahern@gmail.com>
The change to limit the read size from /dev/urandom is a tradeoff.
Reading too much can trigger an issue, but so it could if the
suggested 250 random chars would not contain enough [:alpha:] chars.
If they contain:
a) >=6 all is ok
b) [1-5] the devname would be shorter than expected (non fatal).
c) 0 things would break
In loops of hundreds of thousands it always was (a) for my, but since
if occuring in a test it might be hard to track what happened avoid
this issue by retrying on the length condition.
Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
In some test environments like e.g. Ubuntu & Debian autopkgtest it
can happen that while generating random device names the pipes
between tr and head are considered dead while processing.
That prints (non fatal) issues like:
Running ip/link/new_link.t [iproute2-this/4.13.0-17-generic]: tr:
write error: Broken pipe
tr: write error
PASS
This only happens if reading an infinite amount of chars with the
read from urandom, so reading a defined amount fixes the issue.
Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
This document described how kernel and tc used to handle
timing. In last two years, kernel has switched over to using
ktime. Nothing to see here, move along.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>