Paul Durrant says:
====================
make skb_checksum_setup generally available
Both xen-netfront and xen-netback need to be able to set up the partial
checksum offset of an skb and may also need to recalculate the pseudo-
header checksum in the process. This functionality is currently private
and duplicated between the two drivers.
Patch #1 of this series moves the implementation into the core network code
as there is nothing xen-specific about it and it is potentially useful to
any network driver.
Patch #2 removes the private implementation from netback.
Patch #3 removes the private implementation from netfront.
v2:
- Put skb_checksum_setup in skbuff.c rather than dev.c
- remove inline
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Use skb_checksum_setup to set up partial checksum offsets rather
then a private implementation.
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: David Vrabel <david.vrabel@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use skb_checksum_setup to set up partial checksum offsets rather
then a private implementation.
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Cc: Ian Campbell <ian.campbell@citrix.com>
Cc: Wei Liu <wei.liu2@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds a function to set up the partial checksum offset for IP
packets (and optionally re-calculate the pseudo-header checksum) into the
core network code.
The implementation was previously private and duplicated between xen-netback
and xen-netfront, however it is not xen-specific and is potentially useful
to any network driver.
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Cc: David Miller <davem@davemloft.net>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Veaceslav Falico <vfalico@redhat.com>
Cc: Alexander Duyck <alexander.h.duyck@intel.com>
Cc: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
And it can become static.
Cc: Stephen Hemminger <stephen@networkplumber.org>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 60e453a940 ("USBNET: fix handling padding packet")
added an extra SG entry in case padding is necessary, but
failed to update the initialisation of the list. This can
cause list traversal to fall off the end of the list,
resulting in an oops.
Fixes: 60e453a940 ("USBNET: fix handling padding packet")
Reported-by: Thomas Kear <thomas@kear.co.nz>
Cc: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Bjørn Mork <bjorn@mork.no>
Tested-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The commit c466a9b2b3
(net: 3com: slight optimization of addr compare)
cause a warning: "passing argument 1 of 'ether_addr_equal'
from incompatible pointer type", so fix it.
I think julia will convert ether_addr_equal to ether_addr_equal_64bits later.
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Ding Tianhong <dingtianhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The commit 6878f79a8b
(net: qlcnic: slight optimization of addr compare)
cause a warning "sparse: incorrect type in argument 2
(different type sizes)", so fix it.
I think julia will convert ether_addr_equal to ether_addr_equal_64bits later.
Cc: Himanshu Madhani <himanshu.madhani@qlogic.com>
Cc: Rajesh Borundia <rajesh.borundia@qlogic.com>
Signed-off-by: Ding Tianhong <dingtianhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
sh_eth_error() in case of a TX error tries to print a message using 2 dev_err()
calls with the first string not finished by '\n', so that the resulting message
would inevitably come out garbled, with something like "3net eth0: " inserted
in the middle. Avoid that by merging 2 calls into one.
While at it, insert an empty line after the nearby declaration.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Reviewed-by: Simon Horman <horms+renesas@verge.net.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
net/xfrm/xfrm_policy.c
Steffen Klassert says:
====================
This pull request has a merge conflict between commits be7928d20b
("net: xfrm: xfrm_policy: fix inline not at beginning of declaration") and
da7c224b1b ("net: xfrm: xfrm_policy: silence compiler warning") from
the net-next tree and commit 2f3ea9a95c ("xfrm: checkpatch erros with
inline keyword position") from the ipsec-next tree.
The version from net-next can be used, like it is done in linux-next.
1) Checkpatch cleanups, from Weilong Chen.
2) Fix lockdep complaints when pktgen is used with IPsec,
from Fan Du.
3) Update pktgen to allow any combination of IPsec transport/tunnel mode
and AH/ESP/IPcomp type, from Fan Du.
4) Make pktgen_dst_metrics static, Fengguang Wu.
5) Compile fix for pktgen when CONFIG_XFRM is not set,
from Fan Du.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix inet_diag_dump_icsk() to reflect the fact that both TCP_TIME_WAIT
and TCP_FIN_WAIT2 connections are represented by inet_timewait_sock
(not just TIME_WAIT), and for such sockets the tw_substate field holds
the real state, which can be either TCP_TIME_WAIT or TCP_FIN_WAIT2.
This brings the inet_diag state-matching code in line with the field
it uses to populate idiag_state. This is also analogous to the info
exported in /proc/net/tcp, where get_tcp4_sock() exports sk->sk_state
and get_timewait4_sock() exports tw->tw_substate.
Before fixing this, (a) neither "ss -nemoi" nor "ss -nemoi state
fin-wait-2" would return a socket in TCP_FIN_WAIT2; and (b) "ss -nemoi
state time-wait" would also return sockets in state TCP_FIN_WAIT2.
This is an old bug that predates 05dbc7b ("tcp/dccp: remove twchain").
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Cc: Eric Dumazet <edumazet@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Veaceslav Falico says:
====================
bonding: fix bond_3ad RCU usage
While digging through bond_3ad.c I've found that the RCU usage there is
just wrong - it's used as a kind of mutex/spinlock instead of RCU.
v3->v4: remove useless goto and wrap __get_first_agg() in proper RCU.
v2->v3: make bond_3ad_set_carrier() use RCU read lock for the whole
function, so that all other functions will be protected by RCU as well.
This way we can use _rcu variants everywhere.
v1->v2: use generic primitives instead of _rcu ones cause we can hold RTNL
lock without RCU one, which is still safe.
This patchset is on top of bond_3ad.c cleanup:
http://www.spinics.net/lists/netdev/msg265447.html
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, the implementation is meaningless - once again, we take the
slave structure and use it after we've exited RCU critical section.
Fix this by removing the rcu_read_lock() from __get_active_agg(), and
ensuring that all its callers are holding RCU.
Fixes: be79bd048 ("bonding: add RCU for bond_3ad_state_machine_handler()")
CC: dingtianhong@huawei.com
CC: Jay Vosburgh <fubar@us.ibm.com>
CC: Andy Gospodarek <andy@greyhouse.net>
Signed-off-by: Veaceslav Falico <vfalico@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, the RCU read lock usage is just wrong - it gets the slave struct
under RCU and continues to use it when RCU lock is released.
However, it's still safe to do this cause we didn't need the
rcu_read_lock() initially - all of the __get_first_agg() callers are either
holding RCU read lock or the RTNL lock, so that we can't sync while in it.
Fixes: be79bd048 ("bonding: add RCU for bond_3ad_state_machine_handler()")
CC: dingtianhong@huawei.com
CC: Jay Vosburgh <fubar@us.ibm.com>
CC: Andy Gospodarek <andy@greyhouse.net>
Signed-off-by: Veaceslav Falico <vfalico@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, its usage is just plainly wrong. It first gets a slave under
RCU, and, after releasing the RCU lock, continues to use it - whilst it can
be freed.
Fix this by ensuring that bond_3ad_set_carrier() holds RCU till it uses its
slave (or its agg).
Fixes: be79bd048a ("bonding: add RCU for bond_3ad_state_machine_handler()")
CC: dingtianhong@huawei.com
CC: Jay Vosburgh <fubar@us.ibm.com>
CC: Andy Gospodarek <andy@greyhouse.net>
Signed-off-by: Veaceslav Falico <vfalico@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- drop dependency against CRC16
- move to new release version
- add size check at compile time for packet structs
- update copyright years in every file
- implement new bonding/interface alternation feature
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)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=BKos
-----END PGP SIGNATURE-----
Merge tag 'batman-adv-for-davem' of git://git.open-mesh.org/linux-merge
Included changes:
- drop dependency against CRC16
- move to new release version
- add size check at compile time for packet structs
- update copyright years in every file
- implement new bonding/interface alternation feature
Signed-off-by: David S. Miller <davem@davemloft.net>
Shahed Shaikh says:
====================
This series includes following changes:
o SRIOV and VLAN filtering related enhancements which includes
- Do MAC learning for PF
- Restrict VF from configuring any VLAN mode
- Enable flooding on PF
- Turn on promiscuous mode for PF
o Bug fix in qlcnic_sriov_cleanup() introduced by commit
154d0c81("qlcnic: VLAN enhancement for 84XX adapters")
o Beaconing support for 83xx and 84xx series adapters
o Allow 82xx adapter to perform IPv6 LRO even if destination IP address is not
programmed.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
o Enabling BIT_9 while configuring hardware LRO allows adapter to
perform LRO even if destination IP address is not programmed in adapter.
Signed-off-by: Shahed Shaikh <shahed.shaikh@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
o Add __QLCNIC_SRIOV_ENABLE bit check before doing SRIOV cleanup
Signed-off-by: Manish Chopra <manish.chopra@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
o Refactored code to handle beaconing test for all adapters.
o Use GET_LED_CONFIG mailbox command for 83xx/84xx series adapter
to detect current beaconing state of the adapter.
Signed-off-by: Himanshu Madhani <himanshu.madhani@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
o MAC learning will be done for SRIOV PF to help program VLAN filters
onto adapter. This will help VNIC traffic to flow through without
flooding traffic.
Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
o By default, SRIOV PF will have promiscous mode on.
Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
o On enabling VF flood bit, PF driver will be able to receive traffic
from all its VFs.
Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
o Adapter should allow vlan traffic only for vlans configured on a VF.
On configuring any vlan mode from VF, adapter will allow any vlan
traffic to pass for that VF. Do not allow VF to configure this mode.
Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, after changing the MTU for a device, dev_set_mtu() calls
NETDEV_CHANGEMTU notification, however doesn't verify it's return code -
which can be NOTIFY_BAD - i.e. some of the net notifier blocks refused this
change, and continues nevertheless.
To fix this, verify the return code, and if it's an error - then revert the
MTU to the original one, notify again and pass the error code.
CC: Jiri Pirko <jiri@resnulli.us>
CC: "David S. Miller" <davem@davemloft.net>
CC: Eric Dumazet <edumazet@google.com>
CC: Alexander Duyck <alexander.h.duyck@intel.com>
CC: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: Veaceslav Falico <vfalico@redhat.com>
Reviewed-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
Document how to use one AF_PACKET mmap socket for RX and TX.
Signed-off-by: Norbert van Bolhuis <nvbolhuis@aimvalley.nl>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10G PHYs don't currently support running the state machine, which
is implicitly setup via of_phy_connect(). Therefore, it is necessary
to implement an OF version of phy_attach(), which does everything
except start the state machine.
Signed-off-by: Andy Fleming <afleming@gmail.com>
Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
phy_attach_direct() may now attach to a generic 10G driver. It can
also be used exactly as phy_connect_direct(), which will be useful
when using of_mdio, as phy_connect (and therefore of_phy_connect)
start the PHY state machine, which is currently irrelevant for 10G
PHYs.
Signed-off-by: Andy Fleming <afleming@gmail.com>
Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Very incomplete, but will allow for binding an ethernet controller
to it.
Signed-off-by: Andy Fleming <afleming@gmail.com>
Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Then other generic phy driver such as generic 10g phy driver can join it.
Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Andy Fleming <afleming@gmail.com>
Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Need an extra parameter to read or write Clause 45 PHYs, so
need a different API with the extra parameter.
Signed-off-by: Andy Fleming <afleming@gmail.com>
Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Avoid needless export of local functions
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: James Chapman <jchapman@katalix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix a bunch of whole lot of namespace issues with the Broadcom bnx2x driver
found by running 'make namespacecheck'
* global variables must be prefixed with bnx2x_
naming a variable int_mode, or num_queue is invitation to disaster
* make local functions static
* move some inline's used in one file out of header
(this driver has a bad case of inline-itis)
* remove resulting dead code fallout
bnx2x_pfc_statistic,
bnx2x_emac_get_pfc_stat
bnx2x_init_vlan_mac_obj,
Looks like vlan mac support in this driver was a botch from day one
either never worked, or not implemented or missing support functions
Compile tested only.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
If used 64 bit compiler GCC warns that:
drivers/net/ethernet/smsc/smc91x.c:1897:7:
warning: cast from pointer to integer of different
size [-Wpointer-to-int-cast]
This patch fixes this by changing typecast from "unsigned int" to "unsigned long"
CC: "David S. Miller" <davem@davemloft.net>
CC: Jingoo Han <jg1.han@samsung.com>
CC: netdev@vger.kernel.org
Signed-off-by: Pankaj Dubey <pankaj.dubey@samsung.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Simplify the GRE header length calculation in gre_gso_segment().
Switch to an approach that is simpler, faster, and more general. The
new approach will continue to be correct even if we add support for
the optional variable-length routing info that may be present in a GRE
header.
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: H.K. Jerry Chu <hkchu@google.com>
Cc: Pravin B Shelar <pshelar@nicira.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It is not necessary at all.
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
tp->root is a void* pointer, no need to cast it.
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
tcf_match_indev() is called in fast path, it is not wise to
search for a netdev by ifindex and then compare by its name,
just compare the ifindex.
Also, dev->name could be changed by user-space, therefore
the match would be always fail, but dev->ifindex could
be consistent.
BTW, this will also save some bytes from the core struct of u32.
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It will be needed by the next patch.
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Refactor tcf_add_notify() and factor out tcf_del_notify().
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is no need to store the index separatedly
since tcf_hashinfo is allocated statically too.
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
GRO layer has a limit of 8 flows being held in GRO list,
for performance reason.
When a packet comes for a flow not yet in the list,
and list is full, we immediately give it to upper
stacks, lowering aggregation performance.
With TSO auto sizing and FQ packet scheduler, this situation
happens more often.
This patch changes strategy to simply evict the oldest flow of
the list. This works better because of the nature of packet
trains for which GRO is efficient. This also has the effect
of lowering the GRO latency if many flows are competing.
Tested :
Used a 40Gbps NIC, with 4 RX queues, and 200 concurrent TCP_STREAM
netperf.
Before patch, aggregate rate is 11Gbps (while a single flow can reach
30Gbps)
After patch, line rate is reached.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Jerry Chu <hkchu@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In order to use the native GRO handling of encapsulated protocols on
mlx4, we need to call napi_gro_receive() instead of netif_receive_skb()
unless busy polling is in action.
While we are at it, rename mlx4_en_cq_ll_polling() to
mlx4_en_cq_busy_polling()
Tested with GRE tunnel : GRO aggregation is now performed on the
ethernet device instead of being done later on gre device.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Amir Vadai <amirv@mellanox.com>
Cc: Jerry Chu <hkchu@google.com>
Cc: Or Gerlitz <ogerlitz@mellanox.com>
Acked-By: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fixes the following sparse warning:
net/ipv4/gre_offload.c:253:5: warning:
symbol 'gre_gro_complete' was not declared. Should it be static?
Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
Hannes Frederic Sowa says:
====================
path mtu hardening patches
After a lot of back and forth I want to propose these changes regarding
path mtu hardening and give an outline why I think this is the best way
how to proceed:
This set contains the following patches:
* ipv4: introduce ip_dst_mtu_maybe_forward and protect forwarding path against pmtu spoofing
* ipv6: introduce ip6_dst_mtu_forward and protect forwarding path with it
* ipv4: introduce hardened ip_no_pmtu_disc mode
The first one switches the forwarding path of IPv4 to use the interface
mtu by default and ignore a possible discovered path mtu. It provides
a sysctl to switch back to the original behavior (see discussion below).
The second patch does the same thing unconditionally for IPv6. I don't
provide a knob for IPv6 to switch to original behavior (please see
below).
The third patch introduces a hardened pmtu mode, where only pmtu
information are accepted where the protocol is able to do more stringent
checks on the icmp piggyback payload (please see the patch commit msg
for further details).
Why is this change necessary?
First of all, RFC 1191 4. Router specification says:
"When a router is unable to forward a datagram because it exceeds the
MTU of the next-hop network and its Don't Fragment bit is set, the
router is required to return an ICMP Destination Unreachable message
to the source of the datagram, with the Code indicating
"fragmentation needed and DF set". ..."
For some time now fragmentation has been considered problematic, e.g.:
* http://www.hpl.hp.com/techreports/Compaq-DEC/WRL-87-3.pdf
* http://tools.ietf.org/search/rfc4963
Most of them seem to agree that fragmentation should be avoided because
of efficiency, data corruption or security concerns.
Recently it was shown possible that correctly guessing IP ids could lead
to data injection on DNS packets:
<https://sites.google.com/site/hayashulman/files/fragmentation-poisoning.pdf>
While we can try to completly stop fragmentation on the end host
(this is e.g. implemented via IP_PMTUDISC_INTERFACE), we cannot stop
fragmentation completly on the forwarding path. On the end host the
application has to deal with MTUs and has to choose fallback methods
if fragmentation could be an attack vector. This is already the case for
most DNS software, where a maximum UDP packet size can be configured. But
until recently they had no control over local fragmentation and could
thus emit fragmented packets.
On the forwarding path we can just try to delay the fragmentation to
the last hop where this is really necessary. Current kernel already does
that but only because routers don't receive feedback of path mtus, these are
only send back to the end host system. But it is possible to maliciously
insert path mtu inforamtion via ICMP packets which have an icmp echo_reply
payload, because we cannot validate those notifications against local
sockets. DHCP clients which establish an any-bound RAW-socket could also
start processing unwanted fragmentation-needed packets.
Why does IPv4 has a knob to revert to old behavior while IPv6 doesn't?
IPv4 does fragmentation on the path while IPv6 does always respond with
packet-too-big errors. The interface MTU will always be greater than
the path MTU information. So we would discard packets we could actually
forward because of malicious information. After this change we would
let the hop, which really could not forward the packet, notify the host
of this problem.
IPv4 allowes fragmentation mid-path. In case someone does use a software
which tries to discover such paths and assumes that the kernel is handling
the discovered pmtu information automatically. This should be an extremly
rare case, but because I could not exclude the possibility this knob is
provided. Also this software could insert non-locked mtu information
into the kernel. We cannot distinguish that from path mtu information
currently. Premature fragmentation could solve some problems in wrongly
configured networks, thus this switch is provided.
One frag-needed packet could reduce the path mtu down to 522 bytes
(route/min_pmtu).
Misc:
IPv6 neighbor discovery could advertise mtu information for an
interface. These information update the ipv6-specific interface mtu and
thus get used by the forwarding path.
Tunnel and xfrm output path will still honour path mtu and also respond
with Packet-too-Big or fragmentation-needed errors if needed.
Changelog for all patches:
v2)
* enabled ip_forward_use_pmtu by default
* reworded
v3)
* disabled ip_forward_use_pmtu by default
* reworded
v4)
* renamed ip_dst_mtu_secure to ip_dst_mtu_maybe_forward
* updated changelog accordingly
* removed unneeded !!(... & ...) double negations
v2)
* by default we honour pmtu information
3)
* only honor interface mtu
* rewritten and simplified
* no knob to fall back to old mode any more
v2)
* reworded Documentation
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
