Pablo Neira Ayuso says:
====================
Netfilter updates for net-next
The following large patchset contains Netfilter updates for your
net-next tree. My initial intention was to send you this in two goes but
when I looked back twice I already had this burden on top of me.
Several updates for IPVS from Marco Angaroni:
1) Allow SIP connections originating from real-servers to be load
balanced by the SIP persistence engine as is already implemented
in the other direction.
2) Release connections immediately for One-packet-scheduling (OPS)
in IPVS, instead of making it via timer and rcu callback.
3) Skip deleting conntracks for each one packet in OPS, and don't call
nf_conntrack_alter_reply() since no reply is expected.
4) Enable drop on exhaustion for OPS + SIP persistence.
Miscelaneous conntrack updates from Florian Westphal, including fix for
hash resize:
5) Move conntrack generation counter out of conntrack pernet structure
since this is only used by the init_ns to allow hash resizing.
6) Use get_random_once() from packet path to collect hash random seed
instead of our compound.
7) Don't disable BH from ____nf_conntrack_find() for statistics,
use NF_CT_STAT_INC_ATOMIC() instead.
8) Fix lookup race during conntrack hash resizing.
9) Introduce clash resolution on conntrack insertion for connectionless
protocol.
Then, Florian's netns rework to get rid of per-netns conntrack table,
thus we use one single table for them all. There was consensus on this
change during the NFWS 2015 and, on top of that, it has recently been
pointed as a source of multiple problems from unpriviledged netns:
11) Use a single conntrack hashtable for all namespaces. Include netns
in object comparisons and make it part of the hash calculation.
Adapt early_drop() to consider netns.
12) Use single expectation and NAT hashtable for all namespaces.
13) Use a single slab cache for all namespaces for conntrack objects.
14) Skip full table scanning from nf_ct_iterate_cleanup() if the pernet
conntrack counter tells us the table is empty (ie. equals zero).
Fixes for nf_tables interval set element handling, support to set
conntrack connlabels and allow set names up to 32 bytes.
15) Parse element flags from element deletion path and pass it up to the
backend set implementation.
16) Allow adjacent intervals in the rbtree set type for dynamic interval
updates.
17) Add support to set connlabel from nf_tables, from Florian Westphal.
18) Allow set names up to 32 bytes in nf_tables.
Several x_tables fixes and updates:
19) Fix incorrect use of IS_ERR_VALUE() in x_tables, original patch
from Andrzej Hajda.
And finally, miscelaneous netfilter updates such as:
20) Disable automatic helper assignment by default. Note this proc knob
was introduced by a900689264 ("netfilter: nf_ct_helper: allow to
disable automatic helper assignment") 4 years ago to start moving
towards explicit conntrack helper configuration via iptables CT
target.
21) Get rid of obsolete and inconsistent debugging instrumentation
in x_tables.
22) Remove unnecessary check for null after ip6_route_output().
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Vivien Didelot says:
====================
net: dsa: mv88e6xxx: turn into monolithic driver
This patchset merges all mv88e6* drivers supported by the shared
mv88e6xxx code into a single mv88e6xxx DSA switch driver.
Some flags are added to describe the capabilities of a switch model,
such as the presence of a PPU, EEPROM, some old or new registers, etc.
First these flags are used to conditionally support the same set of
functions in every driver, then specific driver files are removed in
favor of the common mv88e6xxx driver.
Only the merge of driver specific setup code assumes a few differences.
If these differences such as frames priorities are really needed for
some models, they can easily be brought back in a future patch.
Some inconsistencies might show up, such as the need for
MV88E6XXX_FLAG_PPU and MV88E6XXX_FLAG_PPU_ACTIVE flags. But this
patchset does not aim to fix them yet. A future patch can do that if
they end up being unwanted.
The patchset has been tested on interconnected 88E6352 and 88E6185.
Changes v1 -> v2:
- add missing MV88E6XXX_FLAG_EEPROM flag checks
- remove a few remaining _ prefixes
- remove MV88E6XXX_FLAG_CORE_TAG_TYPE which is a specific default
Changes RFC -> v1:
- introduce flags in a separate patch
- do not refactor anything yet
- do not add new functions prefixed with _
- drop packet discarding and mentioned tested platforms
- factorize family flags
- update text for NET_DSA_MV88E6XXX Kconfig entry
====================
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Tested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now that all drivers support the same set of functions and the same
setup code, drop every model-specific DSA switch driver and replace them
with a common mv88e6xxx driver.
This merges the info tables into one, removes the function exports, the
model-specific files, and update the defconfigs.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6131 is the only driver to set the tag protocol to DSA_TAG_PROTO_DSA.
Since it works fine with DSA_TAG_PROTO_EDSA, change its value, like all
other mv88e6xxx drivers.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Provide a shared mv88e6xxx_setup function to the drivers.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6131 is the only driver which setups the priority of IGMP/MLD snoop
frames and ARP frames to the highest setting. Drop such change until we
figure out a common configuration for all switch models.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
All switch models setup the GLOBAL_CONTROL_2 register with slightly
differences.
Since the cascade mode is valid even in a single chip setup, factorize
such configuration.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
All switch drivers configure the GLOBAL_MONITOR_CONTROL register with
slightly changes.
Assume the setup of the upstream port, and configure it as the port to
which ingress and egress and ARP monitor frames are to be sent.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The 6131 switch models have a Core Tag Type register. Their setup code
is setting it to 0x8100, which is the reset default.
Drop this specific part which is correctly configured on reset anyway.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
All switch models configure the GLOBAL_CONTROL register with slightly
differences.
Discarding packets with excessive collisions
(GLOBAL_CONTROL_DISCARD_EXCESS) is specific to 6352 and similar
switches, and setting a maximum frame size
(GLOBAL_CONTROL_MAX_FRAME_1632) is specific to 6185 and similar
switches.
As we are centralizing the chips setup, skip these settings and don't
discard any frames yet, until we found out that such discarding by the
hardware is necessary.
Assume a common setup to enable the PHY Polling Unit if present, don't
discard any packets, and mask all interrupt sources.
Tested on 88E6352 and 88E6185.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Every driver is calling mv88e6xxx_setup_global after
mv88e6xxx_setup_common. Call the former in the latter.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a MV88E6XXX_FLAG_PPU_ACTIVE flag to describe how to reset the
switch, and merge the reset call to the common setup code.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a MV88E6XXX_FLAG_ATU flag to identify switch models with an Address
Translation Unit.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a MV88E6XXX_FLAG_VTU flag to indentify switch models with a VLAN
Table Unit.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add MV88E6XXX_FLAG_PORTSTATE and MV88E6XXX_FLAG_VLANTABLE flags to
identify switch models with required 802.1D operations.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Only 6131 was not supporting the port registers access yet. Assume such
support and use the unlock access routines in the meantime.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a MV88E6XXX_FLAG_EEE flag to describe switch models featuring Energy
Efficient Ethernet. Use it to conditionally support such access in the
common code.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some switch models have a dedicated register for Switch MAC/WoF/WoL.
This register, when present, is used to indirectly set the switch MAC
address, instead of a direct write to 3 global registers.
Identify this feature and share a common mv88e6xxx_set_addr function.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add MV88E6XXX_FLAG_TEMP and MV88E6XXX_FLAG_TEMP_LIMIT flags to describe
switch models featuring a temperature access. Use them to centralize the
access to the temperature feature.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a MV88E6XXX_FLAG_EEPROM flag to describe switch models featuring an
EEPROM and distribute the EEPROM access routines to all models.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some switch has dedicated SMI PHY Command and Data registers, used to
indirectly access the PHYs, instead of direct access.
Identify these switch models and make mv88e6xxx_phy_{read,write} generic
enough to support every models.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a MV88E6XXX_FLAG_PPU flag to describe switch models with a PHY
Polling Unit. This allows to merge PPU specific PHY access code in the
share code.
Make the mv88e6xxx_ppu_disable and mv88e6xxx_phy_{read,write}_ppu
functions use unlocked register accesses in order to call them in
mv88e6xxx_phy_{read,write} in a locked context.
Since the PPU code is shared, also remove NET_DSA_MV88E6XXX_NEED_PPU.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a flags bitmap to the info structure in order to identify features
supported or not by the different switch models.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
An earlier patch changed lookup side to also net_eq() namespaces after
obtaining a reference on the conntrack, so a single kmemcache can be used.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
We already include netns address in the hash, so we only need to use
net_eq in find_appropriate_src and can then put all entries into
same table.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Will be needed soon when we place all in the same hash table.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Sergei Shtylyov says:
====================
sh_eth: couple of software reset bit cleanups
Here's a set of 2 patches against DaveM's 'net-next.git' repo. We can save
on the repetitive chip reset code...
[1/2] sh_eth: call sh_eth_tsu_write() from sh_eth_chip_reset_giga()
[2/2] sh_eth: reuse sh_eth_chip_reset()
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
All the chip_reset() methods repeat the code writing to the ARSTR register
and delaying for 1 ms, so that we can reuse sh_eth_chip_reset() twice.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
sh_eth_chip_reset_giga() doesn't really need to use direct iowrite32() when
writing to the ARSTR register, it can use sh_eth_tsu_write() as all other
chip_reset() methods.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now that mdiobus_scan() doesn't return NULL on failure anymore, this driver
no longer needs to check for it...
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Johan Hedberg says:
====================
pull request: bluetooth-next 2016-05-07
Here are a few more Bluetooth patches for the 4.7 kernel:
- NULL pointer fix in hci_intel driver
- New Intel Bluetooth controller id in btusb driver
- Added device tree binding documentation for Marvel's bt-sd8xxx
- Platform specific wakeup interrupt support for btmrvl driver
Please let me know if there are any issues pulling. Thanks.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Refactor tcp_skb_cb to create two overlaping areas to store
state for incoming or outgoing skbs based on comments by
Neal Cardwell to tcp_nv patch:
AFAICT this patch would not require an increase in the size of
sk_buff cb[] if it were to take advantage of the fact that the
tcp_skb_cb header.h4 and header.h6 fields are only used in the packet
reception code path, and this in_flight field is only used on the
transmit side.
Signed-off-by: Lawrence Brakmo <brakmo@fb.com>
Acked-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When using ifb+netem on ingress on SIT/IPIP/GRE traffic,
GRO packets are not properly processed.
Segmentation should not be forced, since ifb is already adding
quite a performance hit.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
TC_ACT_STOLEN is used when ingress traffic is mirred/redirected
to say ifb.
Packet is not dropped, but consumed.
Only TC_ACT_SHOT is a clear indication something went wrong.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of custom approach re-use generic helpers to convert byte to hex
format.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On small embedded routers, one wants to control maximal amount of
memory used by fq_codel, instead of controlling number of packets or
bytes, since GRO/TSO make these not practical.
Assuming skb->truesize is accurate, we have to keep track of
skb->truesize sum for skbs in queue.
This patch adds a new TCA_FQ_CODEL_MEMORY_LIMIT attribute.
I chose a default value of 32 MBytes, which looks reasonable even
for heavy duty usages. (Prior fq_codel users should not be hurt
when they upgrade their kernels)
Two fields are added to tc_fq_codel_qd_stats to report :
- Current memory usage
- Number of drops caused by memory limits
# tc qd replace dev eth1 root est 1sec 4sec fq_codel memory_limit 4M
..
# tc -s -d qd sh dev eth1
qdisc fq_codel 8008: root refcnt 257 limit 10240p flows 1024
quantum 1514 target 5.0ms interval 100.0ms memory_limit 4Mb ecn
Sent 2083566791363 bytes 1376214889 pkt (dropped 4994406, overlimits 0
requeues 21705223)
rate 9841Mbit 812549pps backlog 3906120b 376p requeues 21705223
maxpacket 68130 drop_overlimit 4994406 new_flow_count 28855414
ecn_mark 0 memory_used 4190048 drop_overmemory 4994406
new_flows_len 1 old_flows_len 177
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Dave Täht <dave.taht@gmail.com>
Cc: Sebastian Möller <moeller0@gmx.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add an implementation of Qualcomm's IPC router protocol, used to
communicate with service providing remote processors.
Signed-off-by: Courtney Cavin <courtney.cavin@sonymobile.com>
Signed-off-by: Bjorn Andersson <bjorn.andersson@sonymobile.com>
[bjorn: Cope with 0 being a valid node id and implement RTM_NEWADDR]
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce compile stubs for the SMD API, allowing consumers to be
compile tested.
Acked-by: Andy Gross <andy.gross@linaro.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
When cxgb4 is enabled with CONFIG_CHELSIO_T4_DCB set, VI enable command
gets called with DCB enabled. But when we have a back to back setup with
DCB enabled on one side and non-DCB on the Peer side. Firmware doesn't
send any DCB_L2_CFG, and DCB priority is never set for Tx queue.
But driver resets the queue priority and state machine whenever there
is a link down, this patch fixes it by adding a check to reset only if
cxgb4_dcb_enabled() returns true.
Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When creating macvtaps that are expected to have the same ifindex
in different network namespaces, only the first one will succeed.
The others will fail with a sysfs_warn_dup warning due to them trying
to create the following sysfs link (with 'NN' the ifindex of macvtapX):
/sys/class/macvtap/tapNN -> /sys/devices/virtual/net/macvtapX/tapNN
This is reproducible by running the following commands:
ip netns add ns1
ip netns add ns2
ip link add veth0 type veth peer name veth1
ip link set veth0 netns ns1
ip link set veth1 netns ns2
ip netns exec ns1 ip l add link veth0 macvtap0 type macvtap
ip netns exec ns2 ip l add link veth1 macvtap1 type macvtap
The last command will fail with "RTNETLINK answers: File exists" (along
with the kernel warning) but retrying it will work because the ifindex
was incremented.
The 'net' device class is isolated between network namespaces so each
one has its own hierarchy of net devices.
This isn't the case for the 'macvtap' device class.
The problem occurs half-way through the netdev registration, when
`macvtap_device_event` is called-back to create the 'tapNN' macvtap
class device under the 'macvtapX' net class device.
This patch adds namespace support to the 'macvtap' device class so
that /sys/class/macvtap is no longer shared between net namespaces.
However, making the macvtap sysfs class namespace-aware has the side
effect of changing /sys/devices/virtual/net/macvtapX/tapNN into
/sys/devices/virtual/net/macvtapX/macvtap/tapNN.
This is due to Commit 24b1442 ("Driver-core: Always create class
directories for classses that support namespaces") and the fact that
class devices supporting namespaces are really not supposed to be placed
directly under other class devices.
To avoid breaking userland, a tapNN symlink pointing to macvtap/tapNN is
created inside the macvtapX directory.
Signed-off-by: Marc Angel <marc@arista.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
I forgot that ip_send_unicast_reply() is not BH safe (yet).
Disabling preemption before calling it was not a good move.
Fixes: c10d9310ed ("tcp: do not assume TCP code is non preemptible")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Andres Lagar-Cavilla <andreslc@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Alexei Starovoitov says:
====================
bpf: introduce direct packet access
This set of patches introduce 'direct packet access' from
cls_bpf and act_bpf programs (which are root only).
Current bpf programs use LD_ABS, LD_INS instructions which have
to do 'if (off < skb_headlen)' for every packet access.
It's ok for socket filters, but too slow for XDP, since single
LD_ABS insn consumes 3% of cpu. Therefore we have to amortize the cost
of length check over multiple packet accesses via direct access
to skb->data, data_end pointers.
The existing packet parser typically look like:
if (load_half(skb, offsetof(struct ethhdr, h_proto)) != ETH_P_IP)
return 0;
if (load_byte(skb, ETH_HLEN + offsetof(struct iphdr, protocol)) != IPPROTO_UDP ||
load_byte(skb, ETH_HLEN) != 0x45)
return 0;
...
with 'direct packet access' the bpf program becomes:
void *data = (void *)(long)skb->data;
void *data_end = (void *)(long)skb->data_end;
struct eth_hdr *eth = data;
struct iphdr *iph = data + sizeof(*eth);
if (data + sizeof(*eth) + sizeof(*iph) + sizeof(*udp) > data_end)
return 0;
if (eth->h_proto != htons(ETH_P_IP))
return 0;
if (iph->protocol != IPPROTO_UDP || iph->ihl != 5)
return 0;
...
which is more natural to write and significantly faster.
See patch 6 for performance tests:
21Mpps(old) vs 24Mpps(new) with just 5 loads.
For more complex parsers the performance gain is higher.
The other approach implemented in [1] was adding two new instructions
to interpreter and JITs and was too hard to use from llvm side.
The approach presented here doesn't need any instruction changes,
but the verifier has to work harder to check safety of the packet access.
Patch 1 prepares the code and Patch 2 adds new checks for direct
packet access and all of them are gated with 'env->allow_ptr_leaks'
which is true for root only.
Patch 3 improves search pruning for large programs.
Patch 4 wires in verifier's changes with net/core/filter side.
Patch 5 updates docs
Patches 6 and 7 add tests.
[1] https://git.kernel.org/cgit/linux/kernel/git/ast/bpf.git/?h=ld_abs_dw
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
add few tests for "pointer to packet" logic of the verifier
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
parse_simple.c - packet parser exapmle with single length check that
filters out udp packets for port 9
parse_varlen.c - variable length parser that understand multiple vlan headers,
ipip, ipip6 and ip options to filter out udp or tcp packets on port 9.
The packet is parsed layer by layer with multitple length checks.
parse_ldabs.c - classic style of packet parsing using LD_ABS instruction.
Same functionality as parse_simple.
simple = 24.1Mpps per core
varlen = 22.7Mpps
ldabs = 21.4Mpps
Parser with LD_ABS instructions is slower than full direct access parser
which does more packet accesses and checks.
These examples demonstrate the choice bpf program authors can make between
flexibility of the parser vs speed.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
explain how verifier checks safety of packet access
and update email addresses.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
allow cls_bpf and act_bpf programs access skb->data and skb->data_end pointers.
The bpf helpers that change skb->data need to update data_end pointer as well.
The verifier checks that programs always reload data, data_end pointers
after calls to such bpf helpers.
We cannot add 'data_end' pointer to struct qdisc_skb_cb directly,
since it's embedded as-is by infiniband ipoib, so wrapper struct is needed.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
since UNKNOWN_VALUE type is weaker than CONST_IMM we can un-teach
verifier its recognition of constants in conditional branches
without affecting safety.
Ex:
if (reg == 123) {
.. here verifier was marking reg->type as CONST_IMM
instead keep reg as UNKNOWN_VALUE
}
Two verifier states with UNKNOWN_VALUE are equivalent, whereas
CONST_IMM_X != CONST_IMM_Y, since CONST_IMM is used for stack range
verification and other cases.
So help search pruning by marking registers as UNKNOWN_VALUE
where possible instead of CONST_IMM.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Extended BPF carried over two instructions from classic to access
packet data: LD_ABS and LD_IND. They're highly optimized in JITs,
but due to their design they have to do length check for every access.
When BPF is processing 20M packets per second single LD_ABS after JIT
is consuming 3% cpu. Hence the need to optimize it further by amortizing
the cost of 'off < skb_headlen' over multiple packet accesses.
One option is to introduce two new eBPF instructions LD_ABS_DW and LD_IND_DW
with similar usage as skb_header_pointer().
The kernel part for interpreter and x64 JIT was implemented in [1], but such
new insns behave like old ld_abs and abort the program with 'return 0' if
access is beyond linear data. Such hidden control flow is hard to workaround
plus changing JITs and rolling out new llvm is incovenient.
Therefore allow cls_bpf/act_bpf program access skb->data directly:
int bpf_prog(struct __sk_buff *skb)
{
struct iphdr *ip;
if (skb->data + sizeof(struct iphdr) + ETH_HLEN > skb->data_end)
/* packet too small */
return 0;
ip = skb->data + ETH_HLEN;
/* access IP header fields with direct loads */
if (ip->version != 4 || ip->saddr == 0x7f000001)
return 1;
[...]
}
This solution avoids introduction of new instructions. llvm stays
the same and all JITs stay the same, but verifier has to work extra hard
to prove safety of the above program.
For XDP the direct store instructions can be allowed as well.
The skb->data is NET_IP_ALIGNED, so for common cases the verifier can check
the alignment. The complex packet parsers where packet pointer is adjusted
incrementally cannot be tracked for alignment, so allow byte access in such cases
and misaligned access on architectures that define efficient_unaligned_access
[1] https://git.kernel.org/cgit/linux/kernel/git/ast/bpf.git/?h=ld_abs_dw
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
cleanup verifier code and prepare it for addition of "pointer to packet" logic
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
