Danit Goldberg [Mon, 18 Nov 2019 07:49:12 +0000 (09:49 +0200)]
ip link: Add support to get SR-IOV VF node GUID and port GUID
Extend iplink to show VF GUIDs (IFLA_VF_IB_NODE_GUID, IFLA_VF_IB_PORT_GUID),
giving the ability for user-space application to print GUID values.
This ability is added to the one of setting new node GUID and port GUID values.
Suitable ip link command:
- ip link show <device>
For example:
- ip link set ib4 vf 0 node_guid 22:44:33:00:33:11:00:33
- ip link set ib4 vf 0 port_guid 10:21:33:12:00:11:22:10
- ip link show ib4
ib4: <BROADCAST,MULTICAST> mtu 4092 qdisc noop state DOWN mode DEFAULT group default qlen 256
link/infiniband 00:00:0a:2d:fe:80:00:00:00:00:00:00:ec:0d:9a:03:00:44:36:8d brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:ff:ff:ff
vf 0 link/infiniband 00:00:0a:2d:fe:80:00:00:00:00:00:00:ec:0d:9a:03:00:44:36:8d brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:ff:ff:ff,
spoof checking off, NODE_GUID 22:44:33:00:33:11:00:33, PORT_GUID 10:21:33:12:00:11:22:10, link-state disable, trust off, query_rss off
Signed-off-by: Danit Goldberg <danitg@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: David Ahern <dsahern@gmail.com>
David Ahern [Sat, 9 Nov 2019 01:34:24 +0000 (01:34 +0000)]
Merge branch 'nsid-cleanup' into next
Guillaume Nault says:
====================
It's currently hard to review ipnetns. The netns ids are inconsistently
treated as signed or unsigned and most helper functions aren't prepared
to use negative ids.
Netns id attributes can be negative: NETNSA_NSID_NOT_ASSIGNED =3D=3D -1.
So let's consistently treat nsids as signed and also reject negative
values in functions that are supposed to only handle assigned netns
ids.
While there, let's drop the extra blank line generated by some command
line parsing errors (patch 5/5).
Guillaume Nault [Fri, 8 Nov 2019 17:00:12 +0000 (18:00 +0100)]
ipnetns: treat NETNSA_NSID and NETNSA_CURRENT_NSID as signed
These attributes are signed (with -1 meaning NETNSA_NSID_NOT_ASSIGNED).
So let's use rta_getattr_s32() and print_int() instead of their
unsigned counterpart to avoid confusion.
Signed-off-by: Guillaume Nault <gnault@redhat.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Jakub Kicinski [Tue, 5 Nov 2019 21:17:05 +0000 (13:17 -0800)]
devlink: fix referencing namespace by PID
netns parameter for devlink reload is supposed to take PID
as well as string name. However, the PID parsing has two
bugs:
- the opts->netns member is unsigned so the < 0
condition is always false;
- the parameter list is not rewinded after parsing as
a name, so parsing as a pid uses the wrong argument.
Fixes: 08e8e1ca3e05 ("devlink: extend reload command to add support for network namespace change") Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Jakub Kicinski [Tue, 5 Nov 2019 21:13:36 +0000 (13:13 -0800)]
devlink: require resource parameters
If devlink resource set parameters are not provided it crashes:
$ devlink resource set netdevsim/netdevsim0
Segmentation fault (core dumped)
This is because even though DL_OPT_RESOURCE_PATH and
DL_OPT_RESOURCE_SIZE are passed as o_required, the validation
table doesn't contain a relevant string.
Fixes: 8cd644095842 ("devlink: Add support for devlink resource abstraction") Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Vlad Buslov [Wed, 30 Oct 2019 14:20:40 +0000 (16:20 +0200)]
tc: implement support for action flags
Implement setting and printing of action flags with single available flag
value "no_percpu" that translates to kernel UAPI TCA_ACT_FLAGS value
TCA_ACT_FLAGS_NO_PERCPU_STATS. Update man page with information regarding
usage of action flags.
Example usage:
# tc actions add action gact drop no_percpu
# sudo tc actions list action gact
total acts 1
action order 0: gact action drop
random type none pass val 0
index 1 ref 1 bind 0
no_percpu
Signed-off-by: Vlad Buslov <vladbu@mellanox.com> Signed-off-by: David Ahern <dsahern@gmail.com>
fread(3) returns size_t data type which is unsigned, thus check
`if (fread(...) < 0)' is always false. To check if fread(3) has
failed, user should check error indicator with ferror(3).
This commit also changes read logic a little bit by being less
forgiving for errors. Previous logic was checking if fread(3)
read *at least* required ammount of data, now code checks if
fread(3) read *exactly* expected ammount of data. This makes
sense because code parses very specific binary file, and reading
even 1 less/more byte than expected, will later corrupt data anyway.
Signed-off-by: Michał Łyszczek <michal.lyszczek@bofc.pl> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Vlad Buslov [Tue, 29 Oct 2019 17:53:46 +0000 (19:53 +0200)]
tc: remove duplicated NEXT_ARG_FWD() in parse_ct()
Function parse_ct() manually calls NEXT_ARG_FWD() after
parse_action_control_dflt(). This is redundant because
parse_action_control_dflt() modifies argc and argv itself. Moreover, such
implementation parses out any following actions option. For example, adding
action ct with cookie errors:
$ sudo tc actions add action ct cookie 111111111111
Bad action type 111111111111
Usage: ... gact <ACTION> [RAND] [INDEX]
Where: ACTION := reclassify | drop | continue | pass | pipe |
goto chain <CHAIN_INDEX> | jump <JUMP_COUNT>
RAND := random <RANDTYPE> <ACTION> <VAL>
RANDTYPE := netrand | determ
VAL : = value not exceeding 10000
JUMP_COUNT := Absolute jump from start of action list
INDEX := index value used
With fix:
$ sudo tc actions add action ct cookie 111111111111
$ sudo tc actions list action ct
total acts 1
action order 0: ct zone 0 pipe
index 1 ref 1 bind 0
cookie 111111111111
Michał Łyszczek [Thu, 24 Oct 2019 21:20:43 +0000 (23:20 +0200)]
rdma/sys.c: fix possible out-of-bound array access
netns_modes_str[] array has 2 elements, when netns_mode is 2,
condition (2 <= 2) will be true and `mode_str = netns_modes_str[2]'
will be executed, which will result in out-of-bound read.
Signed-off-by: Michał Łyszczek <michal.lyszczek@bofc.pl> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Jiri Pirko [Thu, 24 Oct 2019 10:20:52 +0000 (12:20 +0200)]
ip: allow to use alternative names as handle
Extend ll_name_to_index() to get the index of a netdevice using
alternative interface name. Allow alternative long names to pass checks
in couple of ip link/addr commands.
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Jiri Pirko [Thu, 24 Oct 2019 10:20:50 +0000 (12:20 +0200)]
lib/ll_map: cache alternative names
Alternative names are related to the "parent name". That means,
whenever ll_remember_index() is called to add/delete/update and it founds
the "parent name" im object by ifindex, processes related
alternative name im objects too. Put them in a list which holds the
relationship with the parent.
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David Ahern <dsahern@gmail.com>
David Ahern [Sun, 27 Oct 2019 17:28:49 +0000 (10:28 -0700)]
Merge branch 'rdma-mr-stats' into next
Leon Romanovsky says:
====================
This is supplementary part of "ODP information and statistics"
kernel series.
https://lore.kernel.org/linux-rdma/20191016062308.11886-1-leon@kernel.org
Michał Łyszczek [Tue, 22 Oct 2019 20:09:23 +0000 (22:09 +0200)]
ipnetns: do not check netns NAME when -all is specified
When `-all' argument is specified netns runs cmd on all namespaces
and NAME is not used, but netns nevertheless checks if argv[1] is a
valid namespace name ignoring the fact that argv[1] contains cmd
and not NAME. This results in bug where user cannot specify
absolute path to command.
# ip -all netns exec /usr/bin/whoami
Invalid netns name "/usr/bin/whoami"
This forces user to have his command in PATH.
Solution is simply to not validate argv[1] when `-all' argument is
specified.
Signed-off-by: Michał Łyszczek <michal.lyszczek@bofc.pl> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Davide Caratti [Mon, 7 Oct 2019 10:16:44 +0000 (12:16 +0200)]
ss: allow dumping kTLS info
now that INET_DIAG_INFO requests can dump TCP ULP information, extend 'ss'
to allow diagnosing kTLS when it is attached to a TCP socket. While at it,
import kTLS uAPI definitions from the latest net-next tree.
CC: Andrea Claudi <aclaudi@redhat.com> Co-developed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Davide Caratti <dcaratti@redhat.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Nicolas Dichtel [Mon, 7 Oct 2019 13:44:47 +0000 (15:44 +0200)]
ipnetns: enable to dump nsid conversion table
This patch enables to dump/get nsid from a netns into another netns.
Example:
$ ./test.sh
+ ip netns add foo
+ ip netns add bar
+ touch /var/run/netns/init_net
+ mount --bind /proc/1/ns/net /var/run/netns/init_net
+ ip netns set init_net 11
+ ip netns set foo 12
+ ip netns set bar 13
+ ip netns
init_net (id: 11)
bar (id: 13)
foo (id: 12)
+ ip -n foo netns set init_net 21
+ ip -n foo netns set foo 22
+ ip -n foo netns set bar 23
+ ip -n foo netns
init_net (id: 21)
bar (id: 23)
foo (id: 22)
+ ip -n bar netns set init_net 31
+ ip -n bar netns set foo 32
+ ip -n bar netns set bar 33
+ ip -n bar netns
init_net (id: 31)
bar (id: 33)
foo (id: 32)
+ ip netns list-id target-nsid 12
nsid 21 current-nsid 11 (iproute2 netns name: init_net)
nsid 22 current-nsid 12 (iproute2 netns name: foo)
nsid 23 current-nsid 13 (iproute2 netns name: bar)
+ ip -n foo netns list-id target-nsid 21
nsid 11 current-nsid 21 (iproute2 netns name: init_net)
nsid 12 current-nsid 22 (iproute2 netns name: foo)
nsid 13 current-nsid 23 (iproute2 netns name: bar)
+ ip -n bar netns list-id target-nsid 33 nsid 32
nsid 32 current-nsid 32 (iproute2 netns name: foo)
+ ip -n bar netns list-id target-nsid 31 nsid 32
nsid 12 current-nsid 32 (iproute2 netns name: foo)
+ ip netns list-id nsid 13
nsid 13 (iproute2 netns name: bar)
CC: Petr Oros <poros@redhat.com> Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Tested-by: Petr Oros <poros@redhat.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Aya Levin [Wed, 2 Oct 2019 14:35:16 +0000 (17:35 +0300)]
devlink: Fix inconsistency between command input and output
In devlink health show command the reporter's name parameter is called
reporter, but in the output the reporter's name is referred to as name
Before this patch:
$ devlink health show pci/0000:04:00.0 reporter tx
pci/0000:04:00.0:
name tx
state healthy error 0 recover 0 grace_period 500 auto_recover true
After this patch:
$ devlink health show pci/0000:04:00.0 reporter tx
pci/0000:04:00.0:
reporter tx
state healthy error 0 recover 0 grace_period 500 auto_recover true
Reported-by: Jiri Pirko <jiri@mellanox.com> Fixes: 2f1242efe9d0 ("devlink: Add devlink health show command") Signed-off-by: Aya Levin <ayal@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Tariq Toukan <tariqt@mellanox.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Aya Levin [Wed, 2 Oct 2019 14:35:15 +0000 (17:35 +0300)]
devlink: Left justification on FMSG output
FMSG output is dynamic, space separator must be on the left hand side of
the value. Otherwise output has redundant left indentation regardless
the hierarchy.
Fixes: 844a61764c6f ("devlink: Add helper functions for name and value separately") Signed-off-by: Aya Levin <ayal@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Tariq Toukan <tariqt@mellanox.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Andrea Claudi [Tue, 1 Oct 2019 10:32:17 +0000 (12:32 +0200)]
tc: fix segmentation fault on gact action
tc segfaults if gact action is used without action or index:
$ ip link add type dummy
$ tc actions add action pipe index 1
$ tc filter add dev dummy0 parent ffff: protocol ip \
pref 10 u32 match ip src 127.0.0.2 flowid 1:10 action gact
Segmentation fault
We expect tc to fail gracefully with an error message.
This happens if gact is the last argument of the incomplete
command. In this case the "gact" action is parsed, the macro
NEXT_ARG_FWD() is executed and the next matches() crashes
because of null argv pointer.
To avoid this, simply use NEXT_ARG() instead.
With this change in place:
$ ip link add type dummy
$ tc actions add action pipe index 1
$ tc filter add dev dummy0 parent ffff: protocol ip \
pref 10 u32 match ip src 127.0.0.2 flowid 1:10 action gact
Command line is not complete. Try option "help"
Fixes: fa4958897314 ("tc: Fix binding of gact action by index.") Reported-by: Davide Caratti <dcaratti@redhat.com> Signed-off-by: Andrea Claudi <aclaudi@redhat.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Damien Robert [Mon, 30 Sep 2019 21:11:37 +0000 (23:11 +0200)]
man: add reference to `ip route add encap ... src`
The ability to specify the source adresse for 'encap ip' / 'encap ip6'
was added in commit 94a8722f2f78f04c47678cf864ac234a38366709 but the man
page was not updated.
Also fixes a missing page in ip-route.8.in.
Signed-off-by: Damien Robert <damien.olivier.robert+git@gmail.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Jiri Pirko [Thu, 3 Oct 2019 09:51:15 +0000 (11:51 +0200)]
devlink: extend reload command to add support for network namespace change
Extend existing devlink reload command by adding option "netns" by which
user can instruct kernel to reload the devlink instance into specified
network namespace.
Example:
$ ip netns add testns1
$ devlink dev reload netdevsim/netdevsim10 netns testns1
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David Ahern <dsahern@kernel.org>
Leon Romanovsky [Wed, 2 Oct 2019 13:49:34 +0000 (16:49 +0300)]
rdma: Relax requirement to have PID for HW objects
RDMA has weak connection between PIDs and HW objects, because
the latter tied to file descriptors for their lifetime management.
The outcome of such connection is that for the following scenario,
the returned PID will be 0 (not-valid):
1. Create FD and context
2. Share it with ephemeral child
3. Create any object and exit that child
This flow was revealed in testing environment and of course real users
are not running such scenario, because it makes no sense at all in RDMA
world.
Let's do two changes in the code to support such workflow anyway:
1. Remove need to provide PID/kernel name. Code already supports it,
just need to remove extra validation.
2. Ball-out in case PID is 0.
Thomas Haller [Wed, 25 Sep 2019 10:24:03 +0000 (12:24 +0200)]
man: add note to ip-macsec manual about necessary key management
The man page of ip-macsec and the existance of the tool makes it seem like
the user could just configure static keys once, and be done with it. That is
not the case. Some form or key management must be done in user space.
Add a note about that.
Signed-off-by: Thomas Haller <thaller@redhat.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Joe Stringer [Fri, 20 Sep 2019 02:04:47 +0000 (19:04 -0700)]
bpf: Fix race condition with map pinning
If two processes attempt to invoke bpf_map_attach() at the same time,
then they will both create maps, then the first will successfully pin
the map to the filesystem and the second will not pin the map, but will
continue operating with a reference to its own copy of the map. As a
result, the sharing of the same map will be broken from the two programs
that were concurrently loaded via loaders using this library.
Fix this by adding a retry in the case where the pinning fails because
the map already exists on the filesystem. In that case, re-attempt
opening a fd to the map on the filesystem as it shows that another
program already created and pinned a map at that location.
Signed-off-by: Joe Stringer <joe@wand.net.nz> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Andrea Claudi [Mon, 16 Sep 2019 13:00:55 +0000 (15:00 +0200)]
bpf: replace snprintf with asprintf when dealing with long buffers
This reduces stack usage, as asprintf allocates memory on the heap.
This indirectly fixes a snprintf truncation warning (from gcc v9.2.1):
bpf.c: In function ‘bpf_get_work_dir’:
bpf.c:784:49: warning: ‘snprintf’ output may be truncated before the last format character [-Wformat-truncation=]
784 | snprintf(bpf_wrk_dir, sizeof(bpf_wrk_dir), "%s/", mnt);
| ^
bpf.c:784:2: note: ‘snprintf’ output between 2 and 4097 bytes into a destination of size 4096
784 | snprintf(bpf_wrk_dir, sizeof(bpf_wrk_dir), "%s/", mnt);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Fixes: e42256699cac ("bpf: make tc's bpf loader generic and move into lib") Signed-off-by: Andrea Claudi <aclaudi@redhat.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Nicolas Dichtel [Mon, 16 Sep 2019 15:36:27 +0000 (17:36 +0200)]
link_xfrm: don't force to set phydev
Since linux commit 22d6552f827e ("xfrm interface: fix management of
phydev"), phydev is not mandatory anymore.
Note that it also could be useful before the above commit to not force the
user to put a phydev (the kernel was checking it anyway).
For example, it was useful to not set it in case of x-netns, because the
phydev is not available in the current netns:
Before the patch:
$ ip netns add foo
$ ip link add xfrm1 type xfrm dev eth1 if_id 1
$ ip link set xfrm1 netns foo
$ ip -n foo link set xfrm1 type xfrm dev eth1 if_id 2
Cannot find device "eth1"
$ ip -n foo link set xfrm1 type xfrm if_id 2
must specify physical device
Fixes: 286446c1e8c7 ("ip: support for xfrm interfaces") Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Acked-by: Matt Ellison <matt@arroyo.io> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
The 'fw_load_policy' devlink parameter now supports an unknown value.
Suggested-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Add support for the new devlink parameter along with string to uint
conversion.
Signed-off-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David Ahern <dsahern@gmail.com>
David Dai [Wed, 4 Sep 2019 15:06:51 +0000 (10:06 -0500)]
iproute2-next: police: support 64bit rate and peakrate in tc utility
For high speed adapter like Mellanox CX-5 card, it can reach upto
100 Gbits per second bandwidth. Currently htb already supports 64bit rate
in tc utility. However police action rate and peakrate are still limited
to 32bit value (upto 32 Gbits per second). Taking advantage of the 2 new
attributes TCA_POLICE_RATE64 and TCA_POLICE_PEAKRATE64 from kernel,
tc can use them to break the 32bit limit, and still keep the backward
binary compatibility.
Tested-by: David Dai <zdai@linux.vnet.ibm.com> Signed-off-by: David Dai <zdai@linux.vnet.ibm.com> Signed-off-by: David Ahern <dsahern@gmail.com>
David Ahern [Wed, 4 Sep 2019 15:09:52 +0000 (08:09 -0700)]
nexthop: Add space after blackhole
Add a space after 'blackhole' is missing to properly separate the
protocol when it is given.
Fixes: 63df8e8543b0 ("Add support for nexthop objects") Signed-off-by: David Ahern <dsahern@gmail.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Andrea Claudi [Wed, 4 Sep 2019 17:26:14 +0000 (19:26 +0200)]
devlink: fix segfault on health command
devlink segfaults when using grace_period without reporter
$ devlink health set pci/0000:00:09.0 grace_period 3500
Segmentation fault
devlink is instead supposed to gracefully fail printing a warning
message
$ devlink health set pci/0000:00:09.0 grace_period 3500
Reporter's name is expected.
This happens because DL_OPT_HEALTH_REPORTER_NAME and
DL_OPT_HEALTH_REPORTER_GRACEFUL_PERIOD are both defined as BIT(27).
When dl_opts_put() parse options and grace_period is set, it erroneously
tries to set reporter name to null.
This is fixed simply shifting by 1 bit enumeration starting with
DL_OPT_HEALTH_REPORTER_GRACEFUL_PERIOD.
Fixes: b18d89195b16 ("devlink: Add devlink health set command") Signed-off-by: Andrea Claudi <aclaudi@redhat.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Donald Sharp [Sat, 10 Aug 2019 00:18:43 +0000 (20:18 -0400)]
ip nexthop: Allow flush|list operations to specify a specific protocol
In the case where we have a large number of nexthops from a specific
protocol, allow the flush and list operations to take a protocol
to limit the commands scopes.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com> Signed-off-by: David Ahern <dsahern@gmail.com>
Ido Schimmel [Tue, 13 Aug 2019 08:31:41 +0000 (11:31 +0300)]
devlink: Add devlink trap set and show commands
The trap set command allows the user to set the action of an individual
trap. Example:
# devlink trap set netdevsim/netdevsim10 trap blackhole_route action trap
The trap show command allows the user to get the current status of an
individual trap or a dump of all traps in case one is not specified.
When '-s' is specified the trap's statistics are shown. When '-v' is
specified the metadata types the trap can provide are shown. Example:
Donald Sharp [Sat, 10 Aug 2019 00:18:42 +0000 (20:18 -0400)]
ip nexthop: Add space to display properly when showing a group
When displaying a nexthop group made up of other nexthops, the display
line shows this when you have additional data at the end:
id 42 group 43/44/45/46/47/48/49/50/51/52/53/54/55/56/57/58/59/60/61/62/63/64/65/66/67/68/69/70/71/72/73/74proto zebra
Modify code so that it shows:
id 42 group 43/44/45/46/47/48/49/50/51/52/53/54/55/56/57/58/59/60/61/62/63/64/65/66/67/68/69/70/71/72/73/74 proto zebra
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com> Reviewed-by: David Ahern <dsahern@gmail.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Ido Schimmel [Mon, 12 Aug 2019 10:17:06 +0000 (13:17 +0300)]
tc: Fix block-handle support for filter operations
The revert of batchsize accidently reverted more than it should
and broke shared block functionality. Fix this by restoring the
original functionality.
To reproduce:
dst_ip 192.0.2.0/24 action drop
Unknown filter "block", hence option "10" is unparsable
Fixes: e991c04d64c0 ("Revert "tc: Add batchsize feature for filter and actions"") Signed-off-by: Ido Schimmel <idosch@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Jiri Pirko [Mon, 5 Aug 2019 09:56:56 +0000 (11:56 +0200)]
devlink: finish queue.h to list.h transition
Loose the "q" from the names and name the structure fields in the same
way rest of the code does. Also, fix list_add arg order which leads
to segfault.
Fixes: 33267017faf1 ("iproute2: devlink: port from sys/queue.h to list.h") Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
projects / mirror_iproute2.git / log