We store the rule blob per (possible) cpu. Unfortunately this means we can
waste lot of memory on big smp machines. ipt_entry structure ('rule head')
is 112 byte, so e.g. with maxcpu=64 one single rule eats
close to 8k RAM.
Since previous patch made counters percpu it appears there is nothing
left in the rule blob that needs to be percpu.
On my test system (144 possible cpus, 400k dummy rules) this
change saves close to 9 Gigabyte of RAM.
Florian Westphal [Wed, 10 Jun 2015 23:34:54 +0000 (01:34 +0200)]
netfilter: xtables: use percpu rule counters
The binary arp/ip/ip6tables ruleset is stored per cpu.
The only reason left as to why we need percpu duplication are the rule
counters embedded into ipt_entry et al -- since each cpu has its own copy
of the rules, all counters can be lockless.
The downside is that the more cpus are supported, the more memory is
required. Rules are not just duplicated per online cpu but for each
possible cpu, i.e. if maxcpu is 144, then rule is duplicated 144 times,
not for the e.g. 64 cores present.
To save some memory and also improve utilization of shared caches it
would be preferable to only store the rule blob once.
So we first need to separate counters and the rule blob.
Instead of using entry->counters, allocate this percpu and store the
percpu address in entry->counters.pcnt on CONFIG_SMP.
This change makes no sense as-is; it is merely an intermediate step to
remove the percpu duplication of the rule set in a followup patch.
Suggested-by: Eric Dumazet <edumazet@google.com> Acked-by: Jesper Dangaard Brouer <brouer@redhat.com> Reported-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: Florian Westphal <fw@strlen.de> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
netfilter: bridge: restore vlan tag when refragmenting
If bridge netfilter is used with both
bridge-nf-call-iptables and bridge-nf-filter-vlan-tagged enabled
then ip fragments in VLAN frames are sent without the vlan header.
This has never worked reliably. Turns out this relied on pre-3.5
behaviour where skb frag_list was used to store ip fragments;
ip_fragment() then re-used these skbs.
But since commit 3cc4949269e01f39443d0fcfffb5bc6b47878d45
("ipv4: use skb coalescing in defragmentation") this is no longer
the case. ip_do_fragment now needs to allocate new skbs, but these
don't contain the vlan tag information anymore.
Fix it by storing vlan information of the ressembled skb in the
br netfilter percpu frag area, and restore them for each of the
fragments.
Fixes: 3cc4949269e01f3 ("ipv4: use skb coalescing in defragmentation") Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
net: ip_fragment: remove BRIDGE_NETFILTER mtu special handling
since commit d6b915e29f4adea9
("ip_fragment: don't forward defragmented DF packet") the largest
fragment size is available in the IPCB.
Therefore we no longer need to care about 'encapsulation'
overhead of stripped PPPOE/VLAN headers since ip_do_fragment
doesn't use device mtu in such cases.
Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
IPv6 fragmented packets are not forwarded on an ethernet bridge
with netfilter ip6_tables loaded. e.g. steps to reproduce
1) create a simple bridge like this
modprobe br_netfilter
brctl addbr br0
brctl addif br0 eth0
brctl addif br0 eth2
ifconfig eth0 up
ifconfig eth2 up
ifconfig br0 up
2) place a host with an IPv6 address on each side of the bridge
set IPv6 address on host A:
ip -6 addr add fd01:2345:6789:1::1/64 dev eth0
set IPv6 address on host B:
ip -6 addr add fd01:2345:6789:1::2/64 dev eth0
3) run a simple ping command on host A with packets > MTU
ping6 -s 4000 fd01:2345:6789:1::2
4) wait some time and run e.g. "ip6tables -t nat -nvL" on the bridge
IPv6 fragmented packets traverse the bridge cleanly until somebody runs.
"ip6tables -t nat -nvL". As soon as it is run (and netfilter modules are
loaded) IPv6 fragmented packets do not traverse the bridge any more (you
see no more responses in ping's output).
After applying this patch IPv6 fragmented packets traverse the bridge
cleanly in above scenario.
Signed-off-by: Bernhard Thaler <bernhard.thaler@wvnet.at>
[pablo@netfilter.org: small changes to br_nf_dev_queue_xmit] Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Bernhard Thaler [Sat, 30 May 2015 13:28:28 +0000 (15:28 +0200)]
netfilter: bridge: refactor frag_max_size
Currently frag_max_size is member of br_input_skb_cb and copied back and
forth using IPCB(skb) and BR_INPUT_SKB_CB(skb) each time it is changed or
used.
Attach frag_max_size to nf_bridge_info and set value in pre_routing and
forward functions. Use its value in forward and xmit functions.
Signed-off-by: Bernhard Thaler <bernhard.thaler@wvnet.at> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This does not work with ip6tables on a bridge in NAT66 scenario
because the REDIRECT/DNAT/SNAT is not correctly detected.
The bridge pre-routing (finish) netfilter hook has to check for a possible
redirect and then fix the destination mac address. This allows to use the
ip6tables rules for local REDIRECT/DNAT/SNAT REDIRECT similar to the IPv4
iptables version.
This patch makes it possible to use IPv6 NAT66 on a bridge. It was tested
on a bridge with two interfaces using SNAT/DNAT NAT66 rules.
Reported-by: Artie Hamilton <artiemhamilton@yahoo.com> Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
[bernhard.thaler@wvnet.at: rebased, add indirect call to ip6_route_input()]
[bernhard.thaler@wvnet.at: rebased, split into separate patches] Signed-off-by: Bernhard Thaler <bernhard.thaler@wvnet.at> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
netfilter: conntrack: warn the user if there is a better helper to use
After db29a9508a92 ("netfilter: conntrack: disable generic tracking for
known protocols"), if the specific helper is built but not loaded
(a standard for most distributions) systems with a restrictive firewall
but weak configuration regarding netfilter modules to load, will
silently stop working.
This patch then puts a warning message so the sysadmin knows where to
start looking into. It's a pr_warn_once regardless of protocol itself
but it should be enough to give a hint on where to look.
Cc: Florian Westphal <fw@strlen.de> Cc: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
We put shinfo->gso_segs in TCP_SKB_CB(skb) a while back for performance
reasons.
This was in commit cd7d8498c9a5 ("tcp: change tcp_skb_pcount() location")
This patch series complete the job for gso_size and gso_type, so that
we do not bring 2 extra cache lines in tcp write xmit fast path,
and making tcp_init_tso_segs() simpler and faster.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Thu, 11 Jun 2015 16:15:19 +0000 (09:15 -0700)]
tcp: remove obsolete check in tcp_set_skb_tso_segs()
We had various issues in the past when TCP stack was modifying
gso_size/gso_segs while clones were in flight.
Commit c52e2421f73 ("tcp: must unclone packets before mangling them")
fixed these bugs and added a WARN_ON_ONCE(skb_cloned(skb)); in
tcp_set_skb_tso_segs()
These bugs are now fixed, and because TCP stack now only sets
shinfo->gso_size|segs on the clone itself, the check can be removed.
As a result of this change, compiler inlines tcp_set_skb_tso_segs() in
tcp_init_tso_segs()
Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Thu, 11 Jun 2015 16:15:18 +0000 (09:15 -0700)]
tcp: fill shinfo->gso_size at last moment
In commit cd7d8498c9a5 ("tcp: change tcp_skb_pcount() location") we stored
gso_segs in a temporary cache hot location.
This patch does the same for gso_size.
This allows to save 2 cache line misses in tcp xmit path for
the last packet that is considered but not sent because of
various conditions (cwnd, tso defer, receiver window, TSQ...)
Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Thu, 11 Jun 2015 16:15:16 +0000 (09:15 -0700)]
tcp: fill shinfo->gso_type at last moment
Our goal is to touch skb_shinfo(skb) only when absolutely needed,
to avoid two cache line misses in TCP output path for last skb
that is considered but not sent because of various conditions
(cwnd, tso defer, receiver window, TSQ...)
A packet is GSO only when skb_shinfo(skb)->gso_size is not zero.
We can set skb_shinfo(skb)->gso_type to sk->sk_gso_type even for
non GSO packets.
Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Thu, 11 Jun 2015 16:15:15 +0000 (09:15 -0700)]
tcp: reserve tcp_skb_mss() to tcp stack
tcp_gso_segment() and tcp_gro_receive() are not strictly
part of TCP stack. They should not assume tcp_skb_mss(skb)
is in fact skb_shinfo(skb)->gso_size.
This will allow us to change tcp_skb_mss() in following patches.
Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Scott Feldman [Thu, 11 Jun 2015 15:19:01 +0000 (08:19 -0700)]
switchdev: fix BUG when port driver doesn't support set attr op
Fix a BUG_ON() where CONFIG_NET_SWITCHDEV is set but the driver for a
bridged port does not support switchdev_port_attr_set op. Don't BUG_ON()
if -EOPNOTSUPP is returned.
Also change BUG_ON() to netdev_err since this is a normal error path and
does not warrant the use of BUG_ON(), which is reserved for unrecoverable
errs.
Signed-off-by: Scott Feldman <sfeldma@gmail.com> Reported-by: Brenden Blanco <bblanco@plumgrid.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ivan Vecera [Thu, 11 Jun 2015 13:52:31 +0000 (15:52 +0200)]
bna: use netdev_* and dev_* instead of printk and pr_*
...and remove some of them. It is not necessary to log when .probe() and
.remove() are called or when TxQ is started or stopped. Also log level
of some of them was changed to more appropriate one (link up/down,
firmware loading failure.
Signed-off-by: Ivan Vecera <ivecera@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ivan Vecera [Thu, 11 Jun 2015 13:52:27 +0000 (15:52 +0200)]
bna: remove useless pointer assignment
Pointer cmpl used to iterate through completion entries is updated at
the beginning of while loop as well as at the end. The update at the end
of the loop is useless.
Signed-off-by: Ivan Vecera <ivecera@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ivan Vecera [Thu, 11 Jun 2015 13:52:26 +0000 (15:52 +0200)]
bna: use memdup_user to copy userspace buffers
Patch converts kzalloc->copy_from_user sequence to memdup_user. There
is also one useless assignment of NULL to bnad->regdata as it is followed
by assignment of kzalloc output.
Signed-off-by: Ivan Vecera <ivecera@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ivan Vecera [Thu, 11 Jun 2015 13:52:24 +0000 (15:52 +0200)]
bna: remove TX_E_PRIO_CHANGE event and BNA_TX_F_PRIO_CHANGED flag
TX_E_PRIO_CHANGE event is never sent for bna_tx so it doesn't need to be
handled. After this change bna_tx->flags cannot contain
BNA_TX_F_PRIO_CHANGED flag and it can be also eliminated.
Signed-off-by: Ivan Vecera <ivecera@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ivan Vecera [Thu, 11 Jun 2015 13:52:23 +0000 (15:52 +0200)]
bna: remove paused from bna_rx_config and flags from bna_rxf
The bna_rx_config struct member paused can be removed as it is never
written and as it cannot have non-zero value the bna_rxf struct member
flags also cannot have BNA_RXF_F_PAUSED value and is always zero.
So the flags member can be removed as well as bna_rxf_flags enum and
the code-paths that needs to have non-zero bna_rxf->flags.
This clean-up makes bna_rxf_sm_paused state unsed and can be also removed.
Signed-off-by: David S. Miller <davem@davemloft.net>
Ivan Vecera [Thu, 11 Jun 2015 13:52:22 +0000 (15:52 +0200)]
bna: remove RXF_E_PAUSE and RXF_E_RESUME events
RXF_E_PAUSE & RXF_E_RESUME events are never sent for bna_rxf object so
they needn't to be handled. The bna_rxf's state bna_rxf_sm_fltr_clr_wait
and function bna_rxf_fltr_clear are unused after this so remove them also.
Signed-off-by: Ivan Vecera <ivecera@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ivan Vecera [Thu, 11 Jun 2015 13:52:13 +0000 (15:52 +0200)]
bna: use ether_addr_copy instead of memcpy
Parameters of all ether_addr_copy instances were checked for proper
alignment. Alignment of bnad_bcast_addr is forced to 2 as the implicit
alignment is 1.
I have also renamed address parameter of bnad_set_mac_address() to addr.
The name mac_addr was a little bit confusing as the real parameter is
struct sockaddr *.
v2: added __aligned directive to bnad_bcast_addr, renamed parameter of
bnad_set_mac_address() (thx joe@perches.com)
Signed-off-by: Ivan Vecera <ivecera@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Achiad Shochat [Thu, 11 Jun 2015 11:47:33 +0000 (14:47 +0300)]
net/mlx5e: Add transport domain to the ethernet TIRs/TISs
Allocate and use transport domain by the Ethernet driver code.
Signed-off-by: Achiad Shochat <achiad@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Achiad Shochat [Thu, 11 Jun 2015 11:47:32 +0000 (14:47 +0300)]
net/mlx5_core: Add transport domain alloc/dealloc support
Each transport object, namely TIR and TIS, must have a transport domain
number (TDN) identifier.
The driver wrongly assumed that it is OK to use TDN=0 without explicit
TDN allocation from the device.
The TDN will also be used for isolating different processes once user
mode Ethernet will be supported.
Signed-off-by: Achiad Shochat <achiad@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Saeed Mahameed [Thu, 11 Jun 2015 11:47:31 +0000 (14:47 +0300)]
net/mlx5e: Support NETIF_F_SG
When NETIF_F_SG is set, each send WQE may have a different size since
each skb can have different number of fragments as of LSO header etc.
This implies that a given WQE may wrap around the send queue, i.e begin
at its end and continue at its start. While it is legal by the device spec,
we preferred a solution that avoids it - when building of current WQE is
done, if the next WQE may wrap around the send queue, fill the send queue
with NOPs WQEs till its end, so that the next WQE will begin at send queue
start.
NOP WQE for itself cannot wrap around the send queue since it is of
minimal size - 64 bytes, and all send WQEs are a multiple of that size.
Signed-off-by: Achiad Shochat <achiad@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Gal Pressman [Thu, 11 Jun 2015 11:47:30 +0000 (14:47 +0300)]
net/mlx5e: Enforce max flow-tables level >= 3
The Ethernet driver requires at least 3 flow table levels to
operate, enforce that.
Signed-off-by: Gal Pressman <galp@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Saeed Mahameed [Thu, 11 Jun 2015 11:47:29 +0000 (14:47 +0300)]
net/mlx5e: Disable client vlan TX acceleration
We need to resolve a HW configuration issue for enabling HW CVLAN
insertion. Meanwhile, no need to implement the VLAN insertion in
the driver, rather use the generic kernel VLAN insertion method.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Saeed Mahameed [Thu, 11 Jun 2015 11:47:28 +0000 (14:47 +0300)]
net/mlx5e: Add HW cacheline start padding
Enable HW cacheline start padding and align RX WQE size to cacheline
while considering HW start padding. Also, fix dma_unmap call to use
the correct SKB data buffer size.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Saeed Mahameed [Thu, 11 Jun 2015 11:47:27 +0000 (14:47 +0300)]
net/mlx5e: Fix HW MTU settings
Previously we configured HW MTU to be netdev->mtu, actually we
need to configure netdev->mtu + (ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN).
Also, query MTU can not fail, hence make the relevant helper a
void functionm, add mlx5e_set_dev_port_mtu, helper function to
handle MTU setting.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Dan Carpenter [Thu, 11 Jun 2015 08:50:01 +0000 (11:50 +0300)]
net/mlx5_core: fix an error code
We return success if mlx5e_alloc_sq_db() fails but we should return an
error code.
Fixes: f62b8bb8f2d3 ('net/mlx5: Extend mlx5_core to support ConnectX-4 Ethernet functionality') Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Acked-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Geoff Levand [Wed, 10 Jun 2015 17:19:48 +0000 (10:19 -0700)]
net/ps3_gelic: Fix build error with DEBUG
When the DEBUG preprocessor macro is defined the ps3_gelic_net driver build
fails due to an undeclared routine gelic_descr_get_status(). This problem
was introduced during the code cleanup of commit 6b0c21cede22be1f68f0a632c0ca38008ce1abe7 (net: Fix p3_gelic_net sparse warnings),
which re-arranged the ordering of some of the gelic routines.
This change just moves the gelic_descr_get_status() routine up in the
ps3_gelic_net.c source file. There is no functional change.
Fixes build errors like these:
drivers/net/ethernet/toshiba/ps3_gelic_net.c: error: implicit declaration of function gelic_descr_get_status
Signed-off-by: Geoff Levand <geoff@infradead.org> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 11 Jun 2015 07:32:21 +0000 (00:32 -0700)]
Merge branch 'broadcom-MDIO-turn-around'
Florian Fainelli says:
====================
net: broadcom MDIO support for broken turn-around
These two patches update the GENET and UniMAC MDIO controllers to deal with
PHYs that are known to have a broken turn-around bug (e.g: BCM53125 and others)
This utilizes the infrastructure that code recently added to do that in 'net-next'.
Note that the changes look nearly identical and I will try to address the MDIO
code duplication between GENET and UniMAC in a future patch series.
Changes in v2:
- remove brcmphy.h include in mdio-bcm-unimac.c
- use the same comment as with GENET's MDIO read function
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Fainelli [Wed, 10 Jun 2015 19:24:11 +0000 (12:24 -0700)]
net: phy: mdio-bcm-unimac: handle broken turn-around for specific PHYs
Some Ethernet PHYs/switches such as Broadcom's BCM53125 have a hardware bug
which makes them not release the MDIO line during turn-around time. This gets
flagged by the UniMAC MDIO controller as a read failure, and we fail the read
transaction.
Check the MDIO bus phy_ignore_ta_mask bitmask for the PHY we are reading
from and if it is listed in this bitmask, ignore the read failure and
proceed with returning the data we read out of the controller.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Fainelli [Wed, 10 Jun 2015 19:24:10 +0000 (12:24 -0700)]
net: bcmgenet: handle broken turn-around for specific PHYs
Some Ethernet PHYs/switches such as Broadcom's BCM53125 have a hardware
bug which makes them not release the MDIO line during turn-around time.
This gets flagged by the GENET MDIO controller as a read failure, and we
fail the read transaction.
Check the MDIO bus phy_ignore_ta_mask bitmask for the PHY we are reading
from and if it is listed in this bitmask, ignore the read failure and
proceed with returning the data we read out of the controller.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Sergei Shtylyov [Wed, 10 Jun 2015 22:02:30 +0000 (01:02 +0300)]
Renesas Ethernet AVB PTP clock driver
Ethernet AVB device includes the gPTP timer, so we can implement a PTP clock
driver. We're doing that in a separate file, with the main Ethernet driver
calling the PTP driver's [de]initialization and interrupt handler functions.
Unfortunately, the clock seems tightly coupled with the AVB-DMAC, so when that
one leaves the operation mode, we have to unregister the PTP clock... :-(
Based on the original patches by Masaru Nagai.
Signed-off-by: Masaru Nagai <masaru.nagai.vx@renesas.com> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Sergei Shtylyov [Wed, 10 Jun 2015 22:01:43 +0000 (01:01 +0300)]
Renesas Ethernet AVB driver proper
Ethernet AVB includes an Gigabit Ethernet controller (E-MAC) that is basically
compatible with SuperH Gigabit Ethernet E-MAC. Ethernet AVB has a dedicated
direct memory access controller (AVB-DMAC) that is a new design compared to the
SuperH E-DMAC. The AVB-DMAC is compliant with 3 standards formulated for IEEE
802.1BA: IEEE 802.1AS timing and synchronization protocol, IEEE 802.1Qav real-
time transfer, and the IEEE 802.1Qat stream reservation protocol.
The driver only supports device tree probing, so the binding document is
included in this patch.
Based on the original patches by Mitsuhiro Kimura.
Signed-off-by: Mitsuhiro Kimura <mitsuhiro.kimura.kc@renesas.com> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> Signed-off-by: David S. Miller <davem@davemloft.net>
CAIA Delay-Gradient (CDG) is a TCP congestion control that modifies
the TCP sender in order to [1]:
o Use the delay gradient as a congestion signal.
o Back off with an average probability that is independent of the RTT.
o Coexist with flows that use loss-based congestion control, i.e.,
flows that are unresponsive to the delay signal.
o Tolerate packet loss unrelated to congestion. (Disabled by default.)
Its FreeBSD implementation was presented for the ICCRG in July 2012;
slides are available at http://www.ietf.org/proceedings/84/iccrg.html
Running the experiment scenarios in [1] suggests that our implementation
achieves more goodput compared with FreeBSD 10.0 senders, although it also
causes more queueing delay for a given backoff factor.
The loss tolerance heuristic is disabled by default due to safety concerns
for its use in the Internet [2, p. 45-46].
We use a variant of the Hybrid Slow start algorithm in tcp_cubic to reduce
the probability of slow start overshoot.
[1] D.A. Hayes and G. Armitage. "Revisiting TCP congestion control using
delay gradients." In Networking 2011, pages 328-341. Springer, 2011.
[2] K.K. Jonassen. "Implementing CAIA Delay-Gradient in Linux."
MSc thesis. Department of Informatics, University of Oslo, 2015.
Cc: Eric Dumazet <edumazet@google.com> Cc: Yuchung Cheng <ycheng@google.com> Cc: Stephen Hemminger <stephen@networkplumber.org> Cc: Neal Cardwell <ncardwell@google.com> Cc: David Hayes <davihay@ifi.uio.no> Cc: Andreas Petlund <apetlund@simula.no> Cc: Dave Taht <dave.taht@bufferbloat.net> Cc: Nicolas Kuhn <nicolas.kuhn@telecom-bretagne.eu> Signed-off-by: Kenneth Klette Jonassen <kennetkl@ifi.uio.no> Acked-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Upcoming tcp_cdg uses tcp_enter_cwr() to initiate PRR. Export this
function so that CDG can be compiled as a module.
Cc: Eric Dumazet <edumazet@google.com> Cc: Yuchung Cheng <ycheng@google.com> Cc: Stephen Hemminger <stephen@networkplumber.org> Cc: Neal Cardwell <ncardwell@google.com> Cc: David Hayes <davihay@ifi.uio.no> Cc: Andreas Petlund <apetlund@simula.no> Cc: Dave Taht <dave.taht@bufferbloat.net> Cc: Nicolas Kuhn <nicolas.kuhn@telecom-bretagne.eu> Signed-off-by: Kenneth Klette Jonassen <kennetkl@ifi.uio.no> Acked-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Scott Feldman [Thu, 11 Jun 2015 00:04:49 +0000 (17:04 -0700)]
switchdev: fix handling for drivers not supporting IPv4 fib add/del ops
If CONFIG_NET_SWITCHDEV is enabled, but port driver does not implement
support for IPv4 FIB add/del ops, don't fail route add/del offload
operations. Route adds will not be marked as OFFLOAD. Routes will be
installed in the kernel FIB, as usual.
This was report/fixed by Florian when testing DSA driver with net-next on
devices with L2 offload support but no L3 offload support. What he reported
was an initial route installed from DHCP client would fail (route not
installed to kernel FIB). This was triggering the setting of
ipv4.fib_offload_disabled, which would disable route offloading after the
first failure. So subsequent attempts to install the route would succeed.
There is follow-on work/discussion to address the handling of route install
failures, but for now, let's differentiate between no support and failed
support.
Reported-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: Scott Feldman <sfeldma@gmail.com> Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
This patch series converts existing in-tree users of the Broadcom pseudo-PHY
address (30) used to configure MDIO-connected switches to share a constant in a
shared header files.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Fainelli [Thu, 11 Jun 2015 01:08:01 +0000 (18:08 -0700)]
net: dsa: bcm_sf2: Utilize BRCM_PSEUDO_PHY_ADDR
Utilize the newly introduced BRCM_PSEUDO_PHY_ADDR constant from
brcmphy.h instead of open-coding the Broadcom Ethernet switches
pseudo-PHY address (30).
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Fainelli [Thu, 11 Jun 2015 01:08:00 +0000 (18:08 -0700)]
bgmac: Utilize BRCM_PSEUDO_PHY_ADDR
What BGMAC defines as BGMAC_PHY_NOREGS is in fact the Broadcom Ethernet
switches' pseudo-PHY address (30), utilize the newly introduced constant
from brcmphy.h
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Fainelli [Thu, 11 Jun 2015 01:07:59 +0000 (18:07 -0700)]
b44: Utilize BRCM_PSEUDO_PHY_ADDR
What B44 has been locally using as B44_PHY_ADDR_NO_LOCAL_PHY is in fact
the Broadcom Ethernet switches pseudo-PHY address (30). Update the
header to use the newly introduced constant and update comments so they
are within 80 columns and consistent.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Fainelli [Thu, 11 Jun 2015 01:07:57 +0000 (18:07 -0700)]
net: phy: broadcom: include phy.h for brcmphy.h
We utilize inline functions from the PHY library, make sure that we do
include phy.h in brcmphy.h in order for the code including brcmphy.h not
to have to resolve this inclusion dependency.
Fixes: 705314797b8b ("net: phy: broadcom: move shadow 0x1C register accessors to brcmphy.h") Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Thu, 11 Jun 2015 05:11:17 +0000 (22:11 -0700)]
net: tcp: dctcp_update_alpha() fixes.
dctcp_alpha can be read by from dctcp_get_info() without
synchro, so use WRITE_ONCE() to prevent compiler from using
dctcp_alpha as a temporary variable.
Also, playing with small dctcp_shift_g (like 1), can expose
an overflow with 32bit values shifted 9 times before divide.
Use an u64 field to avoid this problem, and perform the divide
only if acked_bytes_ecn is not zero.
Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 11 Jun 2015 05:49:49 +0000 (22:49 -0700)]
Merge tag 'mac80211-next-for-davem-2015-06-10' of git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211-next
Johannes Berg says:
====================
For this round we mostly have fixes:
* mesh fixes from Alexis Green and Chun-Yeow Yeoh,
* a documentation fix from Jakub Kicinski,
* a missing channel release (from Michal Kazior),
* a fix for a signal strength reporting bug (from Sara Sharon),
* handle deauth while associating (myself),
* don't report mangled TX SKB back to userspace for status (myself),
* handle aggregation session timeouts properly in fast-xmit (myself)
However, there are also a few cleanups and one big change that
affects all drivers (and that required me to pull in your tree)
to change the mac80211 HW flags to use an unsigned long bitmap
so that we can extend them more easily - we're running out of
flags even with a cleanup to remove the two unused ones.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Stephen Smalley [Wed, 10 Jun 2015 12:44:59 +0000 (08:44 -0400)]
net/unix: support SCM_SECURITY for stream sockets
SCM_SECURITY was originally only implemented for datagram sockets,
not for stream sockets. However, SCM_CREDENTIALS is supported on
Unix stream sockets. For consistency, implement Unix stream support
for SCM_SECURITY as well. Also clean up the existing code and get
rid of the superfluous UNIXSID macro.
Motivated by https://bugzilla.redhat.com/show_bug.cgi?id=1224211,
where systemd was using SCM_CREDENTIALS and assumed wrongly that
SCM_SECURITY was also supported on Unix stream sockets.
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov> Acked-by: Paul Moore <paul@paul-moore.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Raghu Vatsavayi [Wed, 10 Jun 2015 01:15:23 +0000 (18:15 -0700)]
Add support of Cavium Liquidio ethernet adapters
Following patch V8 adds support for Cavium Liquidio pci express
based 10Gig ethernet adapters.
1) Consolidated all debug macros to either call dev_* or
netdev_* macros directly, feedback from previous patch.
2) Changed soft commands to avoid crash when running
in interrupt context.
3) Fixed link status not reflecting correct status when NetworkManager
is running. Added MODULE_FIRMWARE declarations.
Following were the previous patches.
Patch V7:
1) Minor comments from v6 release regarding debug statements.
2) Fix for large multicast lists.
3) Fixed lockup issue if port initialization fails.
4) Enabled MSI by default.
https://patchwork.ozlabs.org/patch/464441/
Patch V6:
1) Addressed the uint64 vs u64 issue, feedback from previous patch.
2) Consolidated some receive processing routines.
3) Removed link status polling method.
https://patchwork.ozlabs.org/patch/459514/
Patch V5:
Based on the feedback from earlier patches with regards to
consolidation of common functions like device init, register
programming for cn66xx and cn68xx devices.
https://patchwork.ozlabs.org/patch/438979/
Patch V4:
Following were the changes based on the feedback from earlier patch:
1) Added mmiowb while synchronizing queue updates and other hw
interactions.
2) Statistics will now be incremented non-atomically per each ring.
liquidio_get_stats will add stats of each ring while reporting the
total statistics counts.
3) Modified liquidio_ioctl to return proper return codes.
4) Modified device naming to use standard Ethernet naming.
5) Global function names in the driver will have lio_/liquidio_/octeon_
prefix.
6) Ethtool related changes for:
Removed redundant stats and jiffies.
Use default ethtool handler of link status.
Speed setting will make use of ethtool_cmd_speed_set.
7) Added checks for pci_map_* return codes.
8) Check for signals while waiting in interruptible mode
https://patchwork.ozlabs.org/patch/435073/
Patch v3:
Implemented feedback from previous patch like:
Removed NAPI Config and DEBUG config options, added BQL and xmit_more
support.
https://patchwork.ozlabs.org/patch/422749/
Patch V2:
Implemented feedback from previous patch.
https://patchwork.ozlabs.org/patch/413539/
First Patch:
https://patchwork.ozlabs.org/patch/412946/
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com> Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com> Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com> Signed-off-by: Robert Richter <Robert.Richter@caviumnetworks.com> Signed-off-by: Aleksey Makarov <Aleksey.Makarov@caviumnetworks.com> Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 11 Jun 2015 05:14:14 +0000 (22:14 -0700)]
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue
Jeff Kirsher says:
====================
Intel Wired LAN Driver Updates 2015-06-09
This series contains updates to ixgbe only.
The series adds additional support for x550 support, such as WoL and
auto-negotiation of flow control. Adds new PHY support (external PHY)
for x550, as well as the new methods/functions needed to support the new
PHY's. Fixed a bug found in code inspection, where a check was missed
when clearing counters for x550. Also fixed the init code flow for copper
x550 devices.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 11 Jun 2015 05:00:27 +0000 (22:00 -0700)]
Merge branch 'cxgb4-next'
Hariprasad Shenai says:
====================
Add cim_ma_la, cim_pif_la and tx_rate debugfs entry
This patch series adds the following.
Add debugfs entry to dump CIM MA logic analyzer and CIM PIF logic analyzer logs.
Also adds support to dump tx rate of the channels in the adapter
This patch series has been created against net-next tree and includes
patches on cxgb4 driver.
We have included all the maintainers of respective drivers. Kindly review
the change and let us know in case of any review comments.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
bridge: make br_fdb_delete also check if the port matches
Before this patch the user-specified bridge port was ignored when
deleting an fdb entry and thus one could delete an entry that belonged
to any port.
Example (eth0 and eth1 are br0 ports):
bridge fdb add 00:11:22:33:44:55 dev eth0 master
bridge fdb del 00:11:22:33:44:55 dev eth1 master
(succeeds)
after the patch:
bridge fdb add 00:11:22:33:44:55 dev eth0 master
bridge fdb del 00:11:22:33:44:55 dev eth1 master
RTNETLINK answers: No such file or directory
Based on a patch by Wilson Kok.
Reported-by: Wilson Kok <wkok@cumulusnetworks.com> Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
this is a pull request of a two patches for net-next.
The first patch is by Tomas Krcka, he fixes the (currently unused)
register address for acceptance filters. Oliver Hartkopp contributes a
patch for the cangw, where an optional UID is added to reference
routing jobs.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Johannes Berg [Tue, 2 Jun 2015 19:39:54 +0000 (21:39 +0200)]
mac80211: convert HW flags to unsigned long bitmap
As we're running out of hardware capability flags pretty quickly,
convert them to use the regular test_bit() style unsigned long
bitmaps.
This introduces a number of helper functions/macros to set and to
test the bits, along with new debugfs code.
The occurrences of an explicit __clear_bit() are intentional, the
drivers were never supposed to change their supported bits on the
fly. We should investigate changing this to be a per-frame flag.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Alexis Green [Tue, 9 Jun 2015 23:20:24 +0000 (16:20 -0700)]
mac80211: Fix a case of incorrect metric used when forwarding a PREQ
This patch fixes a bug in hwmp_preq_frame_process where the wrong metric
can be used when forwarding a PREQ. This happens because the code uses
the same metric variable to record the value of the metric to the source
of the PREQ and the value of the metric to the target of the PREQ.
This comes into play when both reply and forward are set which happens
when IEEE80211_PREQ_PROACTIVE_PREP_FLAG is set and when MP_F_DO | MP_F_RF
is set. The original code had a special case to handle the first case
but not the second.
The patch uses distinct variables for the two metrics which makes the
code flow much clearer and removes the need to restore the original
value of metric when forwarding.
Signed-off-by: Alexis Green <agreen@cococorp.com> CC: Jesse Jones <jjones@cococorp.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
David S. Miller [Wed, 10 Jun 2015 03:23:52 +0000 (20:23 -0700)]
Merge tag 'batman-adv-for-davem' of git://git.open-mesh.org/linux-merge
Antonio Quartulli says:
====================
Included changes:
- use common Jenkins hash instead of private implementation
- extend internal routing API
- properly re-arrange header files inclusion
- clarify precedence between '&' and '?'
- remove unused ethhdr variable in batadv_gw_dhcp_recipient_get()
- ensure per-VLAN structs are updated upon MAC change
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Don Skidmore [Wed, 10 Jun 2015 00:48:54 +0000 (17:48 -0700)]
ixgbe: Set lan_id before using I2C
We need to call the set_lan_id before accessing I2C and this wasn't
being done so this patch corrects that. Likewise we do the same for
QSFP just to be consistent.
In the X550 case this is even more important as with out it the mux
is not controlled properly.
Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com> Tested-by: Krishneil Singh <krishneil.k.singh@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Don Skidmore [Wed, 10 Jun 2015 00:44:58 +0000 (17:44 -0700)]
ixgbe: add link check for X550 copper
This patch adds ixgbe_check_link_t_X550em for checking copper PHY
link. We check that both the MAC and external PHY have link. This
is to avoid a false link up between the internal and external PHY
when the external PHY doesn't have link.
Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com> Tested-by: Krishneil Singh <krishneil.k.singh@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Don Skidmore [Wed, 10 Jun 2015 00:36:53 +0000 (17:36 -0700)]
ixgbe: fix X550 PHY function pointers
A subset of the X550 function pointers didn't have default methods. This
didn't cause any issue with previous X550 devices as they were all
redefined. However future devices will need these default values.
Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com> Tested-by: Krishneil Singh <krishneil.k.singh@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Don Skidmore [Wed, 10 Jun 2015 00:02:35 +0000 (17:02 -0700)]
ixgbe: fix X550 devices init flow
Reviewing the X550 copper init flow with the Si team resulted in a
new simplified flow. We no longer wait for the PHY FW initialization
complete bit to be set as this bit is only set once by the PHY at power
on and then cleared on the first read. So only the first instance of
running SW (or possibly MAC FW) needs to initialize the PHY.
The PHY initialization has been simplified and now only requires that
the PHY FW be un-stalled
low-power mode or enabled the transceiver
Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com> Tested-by: Krishneil Singh <krishneil.k.singh@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Don Skidmore [Tue, 9 Jun 2015 23:59:04 +0000 (16:59 -0700)]
ixgbe: fix issue with sfp events with new X550 devices
Add checks for systems that don't have SFP's to avoid incorrectly
acting on interrupts that are falsely interpreted as SFP events.
We do this by updating the ixgbe_is_sfp to be aware of the
new X550 devices. This also includes a modified check generating
the EICR mask to be more forward-looking.
Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com> Tested-by: Krishneil Singh <krishneil.k.singh@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Don Skidmore [Tue, 9 Jun 2015 23:52:02 +0000 (16:52 -0700)]
ixgbe: add support for interrupts from X550 external PHY
This patch adds support for receiving interrupts from a external copper
PHY for the X550 part. This includes enabling, detection as well as
re-enablement.
Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com> Tested-by: Krishneil Singh <krishneil.k.singh@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
projects / linux-beck.git / log