11 Dec, 2011
2 commits
-
Wrap the udp6 lookup into the proper ifdef-s.
Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
Eric Dumazet reported, that when inet_diag is built-in the udp_diag also goes
built-in and when ipv6 is a module the udp6 lookup symbol is not found.LD .tmp_vmlinux1
net/built-in.o: In function `udp_dump_one':
udp_diag.c:(.text+0xa2b40): undefined reference to `__udp6_lib_lookup'
make: *** [.tmp_vmlinux1] Erreur 1Fix this by making udp diag build mode depend on both -- inet diag and ipv6.
Reported-by: Eric Dumazet
Signed-off-by: Pavel Emelyanov
Acked-by: Eric Dumazet
Signed-off-by: David S. Miller
10 Dec, 2011
15 commits
-
Copy-s/tcp/udp/-paste from TCP bits.
Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
Do the same as TCP does -- iterate the given udp_table, filter
sockets with bytecode and dump sockets into reply message.The same filtering as for TCP applies, though only some of the
state bits really matter.Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
Do the same as TCP does -- lookup a socket in the given udp_table,
check cookie, fill the reply message with existing inet socket dumping
helper and send one back.Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
Introduce the transport level diag handler module for UDP (and UDP-lite)
sockets and register (empty for now) callbacks in the inet_diag module.Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
The UDP diag get_exact handler will require them to find a
socket by provided net, [sd]addr-s, [sd]ports and device.Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
Introduce two callbacks in inet_diag_handler -- one for dumping all
sockets (with filters) and the other one for dumping a single sk.Replace direct calls to icsk handlers with indirect calls to callbacks
provided by handlers.Make existing TCP and DCCP handlers use provided helpers for icsk-s.
The UDP diag module will provide its own.
Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
The existing inet_csk_diag_fill dumps the inet connection sock info
into the netlink inet_diag_message. Prepare this routine to be able
to dump only the inet_sock part of a socket if the icsk part is missing.This will be used by UDP diag module when dumping UDP sockets.
Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
The upcoming UDP module will require exactly this ability, so just
move the existing code to provide one.Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
Similar to previous patch: the 1st part locks the inet handler
and will get generalized and the 2nd one dumps icsk-s and will
be used by TCP and DCCP handlers.Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
The 1st part locks the inet handler and the 2nd one dump the
inet connection sock.In the next patches the 1st part will be generalized to call
the socket dumping routine indirectly (i.e. TCP/UDP/DCCP) and
the 2nd part will be used by TCP and DCCP handlers.Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
The netlink diag susbsys stores sk address bits in the nl message
as a "cookie" and uses one when dumps details about particular
socket.The same will be required for udp diag module, so introduce a heler
in inet_diag moduleSigned-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
There's an info_size value stored on inet_diag_handler, but for existing
code this value is effectively constant, so just use sizeof(struct tcp_info)
where required.Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
Now RED uses a Q0.32 number to store max_p (max probability), allow
RED/GRED/CHOKE to use/report full resolution at config/dump time.Old tc binaries are non aware of new attributes, and still set/get Plog.
New tc binary set/get both Plog and max_p for backward compatibility,
they display "probability value" if they get max_p from new kernels.# tc -d qdisc show dev ...
...
qdisc red 10: parent 1:1 limit 360Kb min 30Kb max 90Kb ecn ewma 5
probability 0.09 Scell_log 15Make sure we avoid potential divides by 0 in reciprocal_value(), if
(max_th - min_th) is big.Signed-off-by: Eric Dumazet
Signed-off-by: David S. Miller -
This reverts commit 865d9f9f748fdc1943679ea65d9ee1dc55e4a6ae.
This commit breaks the build with CONFIG_NETPRIO_CGROUP=y so
revert it. It does build as a module though. The SUBSYS macro
in the cgroup core code automatically defines a subsys structure
as extern. Long term we should fix the macro. And I need to
fully build test things.Tested with CONFIG_NETPRIO_CGROUP={y|m|n} with and without
CONFIG_CGROUPS defined.Signed-off-by: John Fastabend
CC: Neil Horman
Reported-By: Eric Dumazet
Signed-off-by: David S. Miller
09 Dec, 2011
11 commits
-
These tests are off by one because sock_diag_handlers[] only has AF_MAX
elements.Signed-off-by: Dan Carpenter
Acked-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
net_prio_subsys can be made static this removes the sparse
warning it was throwing.Signed-off-by: John Fastabend
Acked-by: Neil Horman
Signed-off-by: David S. Miller -
Add netpoll support to 802.1q vlan devices. Based on the netpoll support
in the bridging code. Tested on a forced_eth device with netconsole.Signed-off-by: Benjamin LaHaise
Signed-off-by: David S. Miller -
Adaptative RED AQM for linux, based on paper from Sally FLoyd,
Ramakrishna Gummadi, and Scott Shenker, August 2001 :http://icir.org/floyd/papers/adaptiveRed.pdf
Goal of Adaptative RED is to make max_p a dynamic value between 1% and
50% to reach the target average queue : (max_th - min_th) / 2Every 500 ms:
if (avg > target and max_p < target and max_p >= 0.01)
decrease max_p : max_p *= beta;target :[min_th + 0.4*(min_th - max_th),
min_th + 0.6*(min_th - max_th)].
alpha : min(0.01, max_p / 4)
beta : 0.9
max_P is a Q0.32 fixed point number (unsigned, with 32 bits mantissa)Changes against our RED implementation are :
max_p is no longer a negative power of two (1/(2^Plog)), but a Q0.32
fixed point number, to allow full range described in Adatative paper.To deliver a random number, we now use a reciprocal divide (thats really
a multiply), but this operation is done once per marked/droped packet
when in RED_BETWEEN_TRESH window, so added cost (compared to previous
AND operation) is near zero.dump operation gives current max_p value in a new TCA_RED_MAX_P
attribute.Example on a 10Mbit link :
tc qdisc add dev $DEV parent 1:1 handle 10: est 1sec 8sec red \
limit 400000 min 30000 max 90000 avpkt 1000 \
burst 55 ecn adaptative bandwidth 10Mbit# tc -s -d qdisc show dev eth3
...
qdisc red 10: parent 1:1 limit 400000b min 30000b max 90000b ecn
adaptative ewma 5 max_p=0.113335 Scell_log 15
Sent 50414282 bytes 34504 pkt (dropped 35, overlimits 1392 requeues 0)
rate 9749Kbit 831pps backlog 72056b 16p requeues 0
marked 1357 early 35 pdrop 0 other 0Signed-off-by: Eric Dumazet
Signed-off-by: David S. Miller -
Introduce functions handy to copy vlan ids from one driver's list to
another.Signed-off-by: Jiri Pirko
Signed-off-by: David S. Miller -
This allows to keep track of vids needed to be in rx vlan filters of
devices even if they are used in bond/team etc.vlan_info as well as vlan_group previously was, is allocated when first
vid is added and dealocated whan last vid is deleted.vlan_group definition is moved to private header.
Signed-off-by: Jiri Pirko
Signed-off-by: David S. Miller -
This patch adds wrapper for ndo_vlan_rx_add_vid/ndo_vlan_rx_kill_vid
functions. Check for NETIF_F_HW_VLAN_FILTER feature is done in this
wrapper.Signed-off-by: Jiri Pirko
Signed-off-by: David S. Miller -
As this structure is priv, name it approprietely. Also for pointer to it
use name "vlan".Signed-off-by: Jiri Pirko
Signed-off-by: David S. Miller -
If user has configured a MAC address that is not one of the existing
ports of the bridge, then we need to add a special entry in the forwarding
table. This forwarding table entry has no outgoing port so it has to be
treated a little differently. The special entry is reported by the netlink
interface with ifindex of bridge, but ignored by the old interface since there
is no usable way to put it in the ABI.Reported-by: Koki Sanagi
Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller -
Pass bridge to fdb_notify so it can determine correct namespace based
on namespace of bridge rather than namespace of destination port.
Also makes next patch easier.Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller -
Move fdb_notify outside of fdb_create. This fixes the problem
that notification of local entries are not flagged correctly.Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller
07 Dec, 2011
12 commits
-
BUG_ON is too strict in a number of circumstances,
use WARN_ON instead. Protocol errors should not halt the system.Signed-off-by: Sjur Brændeland
Signed-off-by: David S. Miller -
Fix bad assert on fragment size triggering false positive.
Signed-off-by: Sjur Brændeland
Signed-off-by: David S. Miller -
And return error pointers.
Signed-off-by: David S. Miller
-
Signed-off-by: David S. Miller
-
On a CONFIG_NET=y build
net/core/secure_seq.c:22: warning: 'seq_scale' defined but not
usedSigned-off-by: Stephen Boyd
Signed-off-by: David S. Miller -
This patch moves the sock_ code from inet_diag.c to generic sock_diag.c
file and provides necessary request_module-s calls and a pointer on
inet_diag_compat dumping routine.Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
Now all the code works with sock_diag_req-compatible structs, so it's
possible to stop using the inet_diag_type2proto in inet_csk_diag_fill.
Pass the inet_diag_req into it and use the sdiag_protocol field. At the
same time remove the explicit ext argument, since it's also on the req.However, this conversion is still required in _compat code, so just move
this routine, not remove.Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
The new API will specify family to work with. Teach the existing
socket walking code to bypass not interesting ones.To preserve compatibility with existing behavior the _compat code
sets interesting family to AF_UNSPEC to dump them all.Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
Make inet_diag_dumo work with given header instead of calculating
one from the nl message.The SOCK_DIAG_BY_FAMILY just passes skb's one through, the compat code
converts the old header to new one.Also fix the bytecode calculation to find one at proper offset.
Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
Make inet_diag_get_exact work with given header instead of calculating
one from the nl message.The SOCK_DIAG_BY_FAMILY just passes skb's one through, the compat code
converts the old header to new one.Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller -
This one coinsides with the sock_diag_req in the beginning and
contains only used fields from its previous analogue.The existing code is patched to use the _compat version of it
for now.Signed-off-by: Pavel Emelyanov
Signed-off-by: David S. Miller
