John Fastabend [Sat, 13 Oct 2018 00:46:01 +0000 (02:46 +0200)]
tls: add bpf support to sk_msg handling
This work adds BPF sk_msg verdict program support to kTLS
allowing BPF and kTLS to be combined together. Previously kTLS
and sk_msg verdict programs were mutually exclusive in the
ULP layer which created challenges for the orchestrator when
trying to apply TCP based policy, for example. To resolve this,
leveraging the work from previous patches that consolidates
the use of sk_msg, we can finally enable BPF sk_msg verdict
programs so they continue to run after the kTLS socket is
created. No change in behavior when kTLS is not used in
combination with BPF, the kselftest suite for kTLS also runs
successfully.
Joint work with Daniel.
Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
John Fastabend [Sat, 13 Oct 2018 00:46:00 +0000 (02:46 +0200)]
tls: replace poll implementation with read hook
Instead of re-implementing poll routine use the poll callback to
trigger read from kTLS, we reuse the stream_memory_read callback
which is simpler and achieves the same. This helps to align sockmap
and kTLS so we can more easily embed BPF in kTLS.
Joint work with Daniel.
Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Daniel Borkmann [Sat, 13 Oct 2018 00:45:59 +0000 (02:45 +0200)]
tls: convert to generic sk_msg interface
Convert kTLS over to make use of sk_msg interface for plaintext and
encrypted scattergather data, so it reuses all the sk_msg helpers
and data structure which later on in a second step enables to glue
this to BPF.
This also allows to remove quite a bit of open coded helpers which
are covered by the sk_msg API. Recent changes in kTLs 51edc7d81b3e
("tls: Remove redundant vars from tls record structure") and 8faeac9f15a7 ("tls: Add support for inplace records encryption")
changed the data path handling a bit; while we've kept the latter
optimization intact, we had to undo the former change to better
fit the sk_msg model, hence the sg_aead_in and sg_aead_out have
been brought back and are linked into the sk_msg sgs. Now the kTLS
record contains a msg_plaintext and msg_encrypted sk_msg each.
In the original code, the zerocopy_from_iter() has been used out
of TX but also RX path. For the strparser skb-based RX path,
we've left the zerocopy_from_iter() in decrypt_internal() mostly
untouched, meaning it has been moved into tls_setup_from_iter()
with charging logic removed (as not used from RX). Given RX path
is not based on sk_msg objects, we haven't pursued setting up a
dummy sk_msg to call into sk_msg_zerocopy_from_iter(), but it
could be an option to prusue in a later step.
Joint work with John.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Daniel Borkmann [Sat, 13 Oct 2018 00:45:58 +0000 (02:45 +0200)]
bpf, sockmap: convert to generic sk_msg interface
Add a generic sk_msg layer, and convert current sockmap and later
kTLS over to make use of it. While sk_buff handles network packet
representation from netdevice up to socket, sk_msg handles data
representation from application to socket layer.
This means that sk_msg framework spans across ULP users in the
kernel, and enables features such as introspection or filtering
of data with the help of BPF programs that operate on this data
structure.
Latter becomes in particular useful for kTLS where data encryption
is deferred into the kernel, and as such enabling the kernel to
perform L7 introspection and policy based on BPF for TLS connections
where the record is being encrypted after BPF has run and came to
a verdict. In order to get there, first step is to transform open
coding of scatter-gather list handling into a common core framework
that subsystems can use.
The code itself has been split and refactored into three bigger
pieces: i) the generic sk_msg API which deals with managing the
scatter gather ring, providing helpers for walking and mangling,
transferring application data from user space into it, and preparing
it for BPF pre/post-processing, ii) the plain sock map itself
where sockets can be attached to or detached from; these bits
are independent of i) which can now be used also without sock
map, and iii) the integration with plain TCP as one protocol
to be used for processing L7 application data (later this could
e.g. also be extended to other protocols like UDP). The semantics
are the same with the old sock map code and therefore no change
of user facing behavior or APIs. While pursuing this work it
also helped finding a number of bugs in the old sockmap code
that we've fixed already in earlier commits. The test_sockmap
kselftest suite passes through fine as well.
Joint work with John.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Daniel Borkmann [Sat, 13 Oct 2018 00:45:57 +0000 (02:45 +0200)]
tcp, ulp: remove ulp bits from sockmap
In order to prepare sockmap logic to be used in combination with kTLS
we need to detangle it from ULP, and further split it in later commits
into a generic API.
Joint work with John.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Daniel Borkmann [Sat, 13 Oct 2018 00:45:56 +0000 (02:45 +0200)]
tcp, ulp: enforce sock_owned_by_me upon ulp init and cleanup
Whenever the ULP data on the socket is mangled, enforce that the
caller has the socket lock held as otherwise things may race with
initialization and cleanup callbacks from ulp ops as both would
mangle internal socket state.
Joint work with John.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
bpf: wait for running BPF programs when updating map-in-map
The map-in-map frequently serves as a mechanism for atomic
snapshotting of state that a BPF program might record. The current
implementation is dangerous to use in this way, however, since
userspace has no way of knowing when all programs that might have
retrieved the "old" value of the map may have completed.
This change ensures that map update operations on map-in-map map types
always wait for all references to the old map to drop before returning
to userspace.
Signed-off-by: Daniel Colascione <dancol@google.com> Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Anders Roxell [Wed, 10 Oct 2018 14:27:04 +0000 (16:27 +0200)]
selftests: bpf: install script with_addr.sh
When test_flow_dissector.sh runs it complains that it can't find script
with_addr.sh:
./test_flow_dissector.sh: line 81: ./with_addr.sh: No such file or
directory
Rework so that with_addr.sh gets installed, add it to
TEST_PROGS_EXTENDED variable.
Fixes: 8f6ecdaf3d4a ("selftests/bpf: test bpf flow dissection") Signed-off-by: Anders Roxell <anders.roxell@linaro.org> Acked-by: Song Liu <songliubraving@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Anders Roxell [Wed, 10 Oct 2018 09:21:23 +0000 (11:21 +0200)]
selftests: bpf: add config fragment LWTUNNEL
When test_lwt_seg6local.sh was added commit b70e5433008a
("selftests/bpf: test for seg6local End.BPF action") config fragment
wasn't added, and without CONFIG_LWTUNNEL enabled we see this:
Error: CONFIG_LWTUNNEL is not enabled in this kernel.
selftests: test_lwt_seg6local [FAILED]
Signed-off-by: Anders Roxell <anders.roxell@linaro.org> Acked-by: Song Liu <songliubraving@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
====================
While implementing PoC building blocks for eBPF code XDP+TC that can
manipulate VLANs headers, I discovered a bug in generic-XDP.
The fix should be backported to stable kernels. Even-though
generic-XDP was introduced in v4.12, I think the bug is not exposed
until v4.14 in the mentined fixes commit.
====================
Acked-by: Song Liu <songliubraving@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
selftests/bpf: add XDP selftests for modifying and popping VLAN headers
This XDP selftest also contain a small TC-bpf component. It provoke
the generic-XDP bug fixed in previous commit.
The selftest itself shows how to do VLAN manipulation from XDP and TC.
The test demonstrate how XDP ingress can remove a VLAN tag, and how TC
egress can add back a VLAN tag.
This use-case originates from a production need by ISP (kviknet.dk),
who gets DSL-lines terminated as VLAN Q-in-Q tagged packets, and want
to avoid having an net_device for every end-customer on the box doing
the L2 to L3 termination.
The test-setup is done via a veth-pair and creating two network
namespaces (ns1 and ns2). The 'ns1' simulate the ISP network that are
loading the BPF-progs stripping and adding VLAN IDs. The 'ns2'
simulate the DSL-customer that are using VLAN tagged packets.
Running the script with --interactive, will simply not call the
cleanup function. This gives the effect of creating a testlab, that
the users can inspect and play with. The --verbose option will simply
request that the shell will print input lines as they are read, this
include comments, which in effect make the comments visible docs.
net: fix generic XDP to handle if eth header was mangled
XDP can modify (and resize) the Ethernet header in the packet.
There is a bug in generic-XDP, because skb->protocol and skb->pkt_type
are setup before reaching (netif_receive_)generic_xdp.
This bug was hit when XDP were popping VLAN headers (changing
eth->h_proto), as skb->protocol still contains VLAN-indication
(ETH_P_8021Q) causing invocation of skb_vlan_untag(skb), which corrupt
the packet (basically popping the VLAN again).
This patch catch if XDP changed eth header in such a way, that SKB
fields needs to be updated.
V2: on request from Song Liu, use ETH_HLEN instead of mac_len,
in __skb_push() as eth_type_trans() use ETH_HLEN in paired skb_pull_inline().
Fixes: 1cc8f7127f25 ("net: xdp: support xdp generic on virtual devices") Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
====================
Currently when map a lookup fails, user space API can not make any
distinction whether given key was not found or lookup is not supported
by particular map.
In this series we modify return value of maps which do not support
lookup. Lookup on such map implementation will return -EOPNOTSUPP.
bpf() syscall with BPF_MAP_LOOKUP_ELEM command will set EOPNOTSUPP
errno. We also handle this error in bpftool to print appropriate
message.
Patch 1: adds handling of BPF_MAP_LOOKUP ELEM command of bpf syscall
such that errno will set to EOPNOTSUPP when map doesn't support lookup
Patch 2: Modifies the return value of map_lookup_elem() to EOPNOTSUPP
for maps which do not support lookup
Patch 3: Splits do_dump() in bpftool/map.c. Element printing code is
moved out into new function dump_map_elem(). This was done in order to
reduce deep indentation and accomodate further changes.
Patch 4: Changes in bpftool to print strerror() message when lookup
error is occured. This will result in appropriate message like
"Operation not supported" when map doesn't support lookup.
Patch 5: test_verifier: change fixup map naming convention as
suggested by Alexei
Patch 6: Added verifier tests to check whether verifier rejects call
to bpf_map_lookup_elem from bpf program. For all map types those
do not support map lookup.
====================
Prashant Bhole [Tue, 9 Oct 2018 01:04:53 +0000 (10:04 +0900)]
selftests/bpf: test_verifier, change names of fixup maps
Currently fixup map are named like fixup_map1, fixup_map2, and so on.
As suggested by Alexei let's change change map names such that we can
identify map type by looking at the name.
This patch is basically a find and replace change:
fixup_map1 -> fixup_map_hash_8b
fixup_map2 -> fixup_map_hash_48b
fixup_map3 -> fixup_map_hash_16b
fixup_map4 -> fixup_map_array_48b
Suggested-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Prashant Bhole <bhole_prashant_q7@lab.ntt.co.jp> Acked-by: Alexei Starovoitov <ast@kernel.org> Acked-by: Song Liu <songliubraving@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Prashant Bhole [Tue, 9 Oct 2018 01:04:51 +0000 (10:04 +0900)]
tools/bpf: bpftool, split the function do_dump()
do_dump() function in bpftool/map.c has deep indentations. In order
to reduce deep indent, let's move element printing code out of
do_dump() into dump_map_elem() function.
Signed-off-by: Prashant Bhole <bhole_prashant_q7@lab.ntt.co.jp> Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Acked-by: Song Liu <songliubraving@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Prashant Bhole [Tue, 9 Oct 2018 01:04:49 +0000 (10:04 +0900)]
bpf: error handling when map_lookup_elem isn't supported
The error value returned by map_lookup_elem doesn't differentiate
whether lookup was failed because of invalid key or lookup is not
supported.
Lets add handling for -EOPNOTSUPP return value of map_lookup_elem()
method of map, with expectation from map's implementation that it
should return -EOPNOTSUPP if lookup is not supported.
The errno for bpf syscall for BPF_MAP_LOOKUP_ELEM command will be set
to EOPNOTSUPP if map lookup is not supported.
Signed-off-by: Prashant Bhole <bhole_prashant_q7@lab.ntt.co.jp> Acked-by: Alexei Starovoitov <ast@kernel.org> Acked-by: Song Liu <songliubraving@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Wenwen Wang [Sun, 7 Oct 2018 20:23:15 +0000 (15:23 -0500)]
bpf: btf: Fix a missing check bug
In btf_parse_hdr(), the length of the btf data header is firstly copied
from the user space to 'hdr_len' and checked to see whether it is larger
than 'btf_data_size'. If yes, an error code EINVAL is returned. Otherwise,
the whole header is copied again from the user space to 'btf->hdr'.
However, after the second copy, there is no check between
'btf->hdr->hdr_len' and 'hdr_len' to confirm that the two copies get the
same value. Given that the btf data is in the user space, a malicious user
can race to change the data between the two copies. By doing so, the user
can provide malicious data to the kernel and cause undefined behavior.
This patch adds a necessary check after the second copy, to make sure
'btf->hdr->hdr_len' has the same value as 'hdr_len'. Otherwise, an error
code EINVAL will be returned.
Signed-off-by: Wenwen Wang <wang6495@umn.edu> Acked-by: Song Liu <songliubraving@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The following pull-request contains BPF updates for your *net-next* tree.
The main changes are:
1) sk_lookup_[tcp|udp] and sk_release helpers from Joe Stringer which allow
BPF programs to perform lookups for sockets in a network namespace. This would
allow programs to determine early on in processing whether the stack is
expecting to receive the packet, and perform some action (eg drop,
forward somewhere) based on this information.
2) per-cpu cgroup local storage from Roman Gushchin.
Per-cpu cgroup local storage is very similar to simple cgroup storage
except all the data is per-cpu. The main goal of per-cpu variant is to
implement super fast counters (e.g. packet counters), which don't require
neither lookups, neither atomic operations in a fast path.
The example of these hybrid counters is in selftests/bpf/netcnt_prog.c
3) allow HW offload of programs with BPF-to-BPF function calls from Quentin Monnet
4) support more than 64-byte key/value in HW offloaded BPF maps from Jakub Kicinski
5) rename of libbpf interfaces from Andrey Ignatov.
libbpf is maturing as a library and should follow good practices in
library design and implementation to play well with other libraries.
This patch set brings consistent naming convention to global symbols.
6) relicense libbpf as LGPL-2.1 OR BSD-2-Clause from Alexei Starovoitov
to let Apache2 projects use libbpf
7) various AF_XDP fixes from Björn and Magnus
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
====================
Netfilter updates for net-next
The following patchset contains Netfilter updates for your net-next tree:
1) Support for matching on ipsec policy already set in the route, from
Florian Westphal.
2) Split set destruction into deactivate and destroy phase to make it
fit better into the transaction infrastructure, also from Florian.
This includes a patch to warn on imbalance when setting the new
activate and deactivate interfaces.
3) Release transaction list from the workqueue to remove expensive
synchronize_rcu() from configuration plane path. This speeds up
configuration plane quite a bit. From Florian Westphal.
4) Add new xfrm/ipsec extension, this new extension allows you to match
for ipsec tunnel keys such as source and destination address, spi and
reqid. From Máté Eckl and Florian Westphal.
5) Add secmark support, this includes connsecmark too, patches
from Christian Gottsche.
6) Allow to specify remaining bytes in xt_quota, from Chenbo Feng.
One follow up patch to calm a clang warning for this one, from
Nathan Chancellor.
7) Flush conntrack entries based on layer 3 family, from Kristian Evensen.
8) New revision for cgroups2 to shrink the path field.
9) Get rid of obsolete need_conntrack(), as a result from recent
demodularization works.
10) Use WARN_ON instead of BUG_ON, from Florian Westphal.
11) Unused exported symbol in nf_nat_ipv4_fn(), from Florian.
12) Remove superfluous check for timeout netlink parser and dump
functions in layer 4 conntrack helpers.
13) Unnecessary redundant rcu read side locks in NAT redirect,
from Taehee Yoo.
14) Pass nf_hook_state structure to error handlers, patch from
Florian Westphal.
15) Remove ->new() interface from layer 4 protocol trackers. Place
them in the ->packet() interface. From Florian.
16) Place conntrack ->error() handling in the ->packet() interface.
Patches from Florian Westphal.
17) Remove unused parameter in the pernet initialization path,
also from Florian.
18) Remove additional parameter to specify layer 3 protocol when
looking up for protocol tracker. From Florian.
19) Shrink array of layer 4 protocol trackers, from Florian.
20) Check for linear skb only once from the ALG NAT mangling
codebase, from Taehee Yoo.
21) Use rhashtable_walk_enter() instead of deprecated
rhashtable_walk_init(), also from Taehee.
22) No need to flush all conntracks when only one single address
is gone, from Tan Hu.
23) Remove redundant check for NAT flags in flowtable code, from
Taehee Yoo.
24) Use rhashtable_lookup() instead of rhashtable_lookup_fast()
from netfilter codebase, since rcu read lock side is already
assumed in this path.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Arnd Bergmann [Mon, 8 Oct 2018 18:30:01 +0000 (11:30 -0700)]
bpf: fix building without CONFIG_INET
The newly added TCP and UDP handling fails to link when CONFIG_INET
is disabled:
net/core/filter.o: In function `sk_lookup':
filter.c:(.text+0x7ff8): undefined reference to `tcp_hashinfo'
filter.c:(.text+0x7ffc): undefined reference to `tcp_hashinfo'
filter.c:(.text+0x8020): undefined reference to `__inet_lookup_established'
filter.c:(.text+0x8058): undefined reference to `__inet_lookup_listener'
filter.c:(.text+0x8068): undefined reference to `udp_table'
filter.c:(.text+0x8070): undefined reference to `udp_table'
filter.c:(.text+0x808c): undefined reference to `__udp4_lib_lookup'
net/core/filter.o: In function `bpf_sk_release':
filter.c:(.text+0x82e8): undefined reference to `sock_gen_put'
Wrap the related sections of code in #ifdefs for the config option.
Furthermore, sk_lookup() should always have been marked 'static', this
also avoids a warning about a missing prototype when building with
'make W=1'.
Fixes: 2a1cf81b2c34 ("bpf: Add helper to retrieve socket in BPF") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Joe Stringer <joe@wand.net.nz> Acked-by: Song Liu <songliubraving@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
dpaa2-eth: Don't account Tx confirmation frames on NAPI poll
Until now, both Rx and Tx confirmation frames handled during
NAPI poll were counted toward the NAPI budget. However, Tx
confirmations are lighter to process than Rx frames, which can
skew the amount of work actually done inside one NAPI cycle.
Update the code to only count Rx frames toward the NAPI budget
and set a separate threshold on how many Tx conf frames can be
processed in one poll cycle.
The NAPI poll routine stops when either the budget is consumed
by Rx frames or when Tx confirmation frames reach this threshold.
Signed-off-by: Ioana Radulescu <ruxandra.radulescu@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
YueHaibing [Mon, 8 Oct 2018 14:07:50 +0000 (14:07 +0000)]
net: mscc: ocelot: remove set but not used variable 'phy_mode'
Fixes gcc '-Wunused-but-set-variable' warning:
drivers/net/ethernet/mscc/ocelot_board.c: In function 'mscc_ocelot_probe':
drivers/net/ethernet/mscc/ocelot_board.c:262:17: warning:
variable 'phy_mode' set but not used [-Wunused-but-set-variable]
enum phy_mode phy_mode;
It never used since introduction in
commit 6cb28b8a0de3 ("net: mscc: ocelot: make use of SerDes PHYs for handling their configuration")
Signed-off-by: YueHaibing <yuehaibing@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 8 Oct 2018 18:00:24 +0000 (11:00 -0700)]
Merge branch 'more-pmtu-selftests'
Sabrina Dubroca says:
====================
selftests: add more PMTU tests
The current selftests for PMTU cover VTI tunnels, but there's nothing
about the generation and handling of PMTU exceptions by intermediate
routers. This series adds and improves existing helpers, then adds
IPv4 and IPv6 selftests with a setup involving an intermediate router.
Joint work with Stefano Brivio.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Sabrina Dubroca [Mon, 8 Oct 2018 12:37:05 +0000 (14:37 +0200)]
selftests: pmtu: add basic IPv4 and IPv6 PMTU tests
Commit 55c6e7bb3b58 ("selftests: net: Introduce first PMTU test") and
follow-ups introduced some PMTU tests, but they all rely on tunneling,
and, particularly, on VTI.
These new tests use simple routing to exercise the generation and
update of PMTU exceptions in IPv4 and IPv6.
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net> Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Reviewed-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Sabrina Dubroca [Mon, 8 Oct 2018 12:37:04 +0000 (14:37 +0200)]
selftests: pmtu: extend MTU parsing helper to locked MTU
The mtu_parse helper introduced in commit 0cb165250f7c ("selftests:
pmtu: Factor out MTU parsing helper") can only handle "mtu 1234", but
not "mtu lock 1234". Extend it, so that we can do IPv4 tests with PMTU
smaller than net.ipv4.route.min_pmtu
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net> Reviewed-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio [Mon, 8 Oct 2018 12:37:03 +0000 (14:37 +0200)]
selftests: pmtu: Introduce check_pmtu_value()
Introduce and use a function that checks PMTU values against
expected values and logs error messages, to remove some clutter.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Signed-off-by: Sabrina Dubroca <sd@queasysnail.net> Reviewed-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
isdn/gigaset/isocdata: mark expected switch fall-through
Notice that in this particular case, I replaced the
"--v-- fall through --v--" comment with a proper
"fall through", which is what GCC is expecting to
find.
This fix is part of the ongoing efforts to enabling
-Wimplicit-fallthrough
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
rtnetlink: Add support for rigid checking of data in dump request
There are many use cases where a user wants to influence what is
returned in a dump for some rtnetlink command: one is wanting data
for a different namespace than the one the request is received and
another is limiting the amount of data returned in the dump to a
specific set of interest to userspace, reducing the cpu overhead of
both kernel and userspace. Unfortunately, the kernel has historically
not been strict with checking for the proper header or checking the
values passed in the header. This lenient implementation has allowed
iproute2 and other packages to pass any struct or data in the dump
request as long as the family is the first byte. For example, ifinfomsg
struct is used by iproute2 for all generic dump requests - links,
addresses, routes and rules when it is really only valid for link
requests.
There is 1 is example where the kernel deals with the wrong struct: link
dumps after VF support was added. Older iproute2 was sending rtgenmsg as
the header instead of ifinfomsg so a patch was added to try and detect
old userspace vs new: 0b8bb2e51b53 ("rtnetlink: fix userspace API breakage for iproute2 < v3.9.0")
The latest example is Christian's patch set wanting to return addresses for
a target namespace. It guesses the header struct is an ifaddrmsg and if it
guesses wrong a netlink warning is generated in the kernel log on every
address dump which is unacceptable.
Another example where the kernel is a bit lenient is route dumps: iproute2
can send either a request with either ifinfomsg or a rtmsg as the header
struct, yet the kernel always treats the header as an rtmsg (see
inet_dump_fib and rtm_flags check). The header inconsistency impacts the
ability to add kernel side filters for route dumps - a necessary feature
for scale setups with 100k+ routes.
How to resolve the problem of not breaking old userspace yet be able to
move forward with new features such as kernel side filtering which are
crucial for efficient operation at high scale?
This patch set addresses the problem by adding a new socket flag,
NETLINK_DUMP_STRICT_CHK, that userspace can use with setsockopt to
request strict checking of headers and attributes on dump requests and
hence unlock the ability to use kernel side filters as they are added.
Kernel side, the dump handlers are updated to verify the message contains
at least the expected header struct:
RTM_GETLINK: ifinfomsg
RTM_GETADDR: ifaddrmsg
RTM_GETMULTICAST: ifaddrmsg
RTM_GETANYCAST: ifaddrmsg
RTM_GETADDRLABEL: ifaddrlblmsg
RTM_GETROUTE: rtmsg
RTM_GETSTATS: if_stats_msg
RTM_GETNEIGH: ndmsg
RTM_GETNEIGHTBL: ndtmsg
RTM_GETNSID: rtgenmsg
RTM_GETRULE: fib_rule_hdr
RTM_GETNETCONF: netconfmsg
RTM_GETMDB: br_port_msg
And then every field in the header struct should be 0 with the exception
of the family. There are a few exceptions to this rule where the kernel
already influences the data returned by values in the struct. Next the
message should not contain attributes unless the kernel implements
filtering for it. Any unexpected data causes the dump to fail with EINVAL.
If the new flag is honored by the kernel and the dump contents adjusted
by any data passed in the request, the dump handler can set the
NLM_F_DUMP_FILTERED flag in the netlink message header.
For old userspace on new kernel there is no impact as all checks are
wrapped in a check on the new strict flag. For new userspace on old
kernel, the data in the headers and any appended attributes are
silently ignored though the setsockopt failing is the clue to userspace
the feature is not supported. New userspace on new kernel gets the
requested data dump.
iproute2 patches can be found here:
https://github.com/dsahern/iproute2 dump-enhancements
Major changes since v1
- inner header is supposed to be 4-bytes aligned. So for dumps that
should not have attributes appended changed the check to use:
if (nlmsg_attrlen(nlh, sizeof(hdr)))
Only impacts patches with headers that are not multiples of 4-bytes
(rtgenmsg, netconfmsg), but applied the change to all patches not
calling nlmsg_parse for consistency.
- Added nlmsg_parse_strict and nla_parse_strict for tighter control on
attribute parsing. There should be no unknown attribute types or extra
bytes.
- Moved validation to a helper in most cases
Changes since rfc-v2
- dropped the NLM_F_DUMP_FILTERED flag from target nsid dumps per
Jiri's objections
- changed the opt-in uapi from a netlink message flag to a socket
flag. setsockopt provides an api for userspace to definitively
know if the kernel supports strict checking on dumps.
- re-ordered patches to peel off the extack on dumps if needed to
keep this set size within limits
- misc cleanups in patches based on testing
====================
Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:44 +0000 (20:16 -0700)]
rtnetlink: Update rtnl_fdb_dump for strict data checking
Update rtnl_fdb_dump for strict data checking. If the flag is set,
the dump request is expected to have an ndmsg struct as the header
potentially followed by one or more attributes. Any data passed in the
header or as an attribute is taken as a request to influence the data
returned. Only values supported by the dump handler are allowed to be
non-0 or set in the request. At the moment only the NDA_IFINDEX and
NDA_MASTER attributes are supported.
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:43 +0000 (20:16 -0700)]
rtnetlink: Move input checking for rtnl_fdb_dump to helper
Move the existing input checking for rtnl_fdb_dump into a helper,
valid_fdb_dump_legacy. This function will retain the current
logic that works around the 2 headers that userspace has been
allowed to send up to this point.
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:42 +0000 (20:16 -0700)]
net/bridge: Update br_mdb_dump for strict data checking
Update br_mdb_dump for strict data checking. If the flag is set,
the dump request is expected to have a br_port_msg struct as the
header. All elements of the struct are expected to be 0 and no
attributes can be appended.
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:41 +0000 (20:16 -0700)]
net: Update netconf dump handlers for strict data checking
Update inet_netconf_dump_devconf, inet6_netconf_dump_devconf, and
mpls_netconf_dump_devconf for strict data checking. If the flag is set,
the dump request is expected to have an netconfmsg struct as the header.
The struct only has the family member and no attributes can be appended.
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:40 +0000 (20:16 -0700)]
net/ipv6: Update ip6addrlbl_dump for strict data checking
Update ip6addrlbl_dump for strict data checking. If the flag is set,
the dump request is expected to have an ifaddrlblmsg struct as the
header. All elements of the struct are expected to be 0 and no
attributes can be appended.
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:39 +0000 (20:16 -0700)]
net/fib_rules: Update fib_nl_dumprule for strict data checking
Update fib_nl_dumprule for strict data checking. If the flag is set,
the dump request is expected to have fib_rule_hdr struct as the header.
All elements of the struct are expected to be 0 and no attributes can
be appended.
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:38 +0000 (20:16 -0700)]
net/namespace: Update rtnl_net_dumpid for strict data checking
Update rtnl_net_dumpid for strict data checking. If the flag is set,
the dump request is expected to have an rtgenmsg struct as the header
which has the family as the only element. No data may be appended.
Signed-off-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:37 +0000 (20:16 -0700)]
net/neighbor: Update neightbl_dump_info for strict data checking
Update neightbl_dump_info for strict data checking. If the flag is set,
the dump request is expected to have an ndtmsg struct as the header.
All elements of the struct are expected to be 0 and no attributes can
be appended.
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:36 +0000 (20:16 -0700)]
net/neighbor: Update neigh_dump_info for strict data checking
Update neigh_dump_info for strict data checking. If the flag is set,
the dump request is expected to have an ndmsg struct as the header
potentially followed by one or more attributes. Any data passed in the
header or as an attribute is taken as a request to influence the data
returned. Only values supported by the dump handler are allowed to be
non-0 or set in the request. At the moment only the NDA_IFINDEX and
NDA_MASTER attributes are supported.
Existing code does not fail the dump if nlmsg_parse fails. That behavior
is kept for non-strict checking.
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:35 +0000 (20:16 -0700)]
rtnetlink: Update fib dumps for strict data checking
Add helper to check netlink message for route dumps. If the strict flag
is set the dump request is expected to have an rtmsg struct as the header.
All elements of the struct are expected to be 0 with the exception of
rtm_flags (which is used by both ipv4 and ipv6 dumps) and no attributes
can be appended. rtm_flags can only have RTM_F_CLONED and RTM_F_PREFIX
set.
Update inet_dump_fib, inet6_dump_fib, mpls_dump_routes, ipmr_rtm_dumproute,
and ip6mr_rtm_dumproute to call this helper if strict data checking is
enabled.
Signed-off-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:34 +0000 (20:16 -0700)]
rtnetlink: Update ipmr_rtm_dumplink for strict data checking
Update ipmr_rtm_dumplink for strict data checking. If the flag is set,
the dump request is expected to have an ifinfomsg struct as the header.
All elements of the struct are expected to be 0 and no attributes can
be appended.
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:33 +0000 (20:16 -0700)]
rtnetlink: Update inet6_dump_ifinfo for strict data checking
Update inet6_dump_ifinfo for strict data checking. If the flag is
set, the dump request is expected to have an ifinfomsg struct as
the header. All elements of the struct are expected to be 0 and no
attributes can be appended.
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:32 +0000 (20:16 -0700)]
rtnetlink: Update rtnl_stats_dump for strict data checking
Update rtnl_stats_dump for strict data checking. If the flag is set,
the dump request is expected to have an if_stats_msg struct as the header.
All elements of the struct are expected to be 0 except filter_mask which
must be non-0 (legacy behavior). No attributes are supported.
Signed-off-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:31 +0000 (20:16 -0700)]
rtnetlink: Update rtnl_bridge_getlink for strict data checking
Update rtnl_bridge_getlink for strict data checking. If the flag is set,
the dump request is expected to have an ifinfomsg struct as the header
potentially followed by one or more attributes. Any data passed in the
header or as an attribute is taken as a request to influence the data
returned. Only values supported by the dump handler are allowed to be
non-0 or set in the request. At the moment only the IFLA_EXT_MASK
attribute is supported.
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:30 +0000 (20:16 -0700)]
rtnetlink: Update rtnl_dump_ifinfo for strict data checking
Update rtnl_dump_ifinfo for strict data checking. If the flag is set,
the dump request is expected to have an ifinfomsg struct as the header
potentially followed by one or more attributes. Any data passed in the
header or as an attribute is taken as a request to influence the data
returned. Only values supported by the dump handler are allowed to be
non-0 or set in the request. At the moment only the IFA_TARGET_NETNSID,
IFLA_EXT_MASK, IFLA_MASTER, and IFLA_LINKINFO attributes are supported.
Existing code does not fail the dump if nlmsg_parse fails. That behavior
is kept for non-strict checking.
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:29 +0000 (20:16 -0700)]
net/ipv6: Update inet6_dump_addr for strict data checking
Update inet6_dump_addr for strict data checking. If the flag is set, the
dump request is expected to have an ifaddrmsg struct as the header
potentially followed by one or more attributes. Any data passed in the
header or as an attribute is taken as a request to influence the data
returned. Only values suppored by the dump handler are allowed to be
non-0 or set in the request. At the moment only the IFA_TARGET_NETNSID
attribute is supported. Follow on patches can add support for other fields
(e.g., honor ifa_index and only return data for the given device index).
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:28 +0000 (20:16 -0700)]
net/ipv4: Update inet_dump_ifaddr for strict data checking
Update inet_dump_ifaddr for strict data checking. If the flag is set,
the dump request is expected to have an ifaddrmsg struct as the header
potentially followed by one or more attributes. Any data passed in the
header or as an attribute is taken as a request to influence the data
returned. Only values supported by the dump handler are allowed to be
non-0 or set in the request. At the moment only the IFA_TARGET_NETNSID
attribute is supported. Follow on patches can support for other fields
(e.g., honor ifa_index and only return data for the given device index).
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:27 +0000 (20:16 -0700)]
netlink: Add new socket option to enable strict checking on dumps
Add a new socket option, NETLINK_DUMP_STRICT_CHK, that userspace
can use via setsockopt to request strict checking of headers and
attributes on dump requests.
To get dump features such as kernel side filtering based on data in
the header or attributes appended to the dump request, userspace
must call setsockopt() for NETLINK_DUMP_STRICT_CHK and a non-zero
value. Since the netlink sock and its flags are private to the
af_netlink code, the strict checking flag is passed to dump handlers
via a flag in the netlink_callback struct.
For old userspace on new kernel there is no impact as all of the data
checks in later patches are wrapped in a check on the new strict flag.
For new userspace on old kernel, the setsockopt will fail and even if
new userspace sets data in the headers and appended attributes the
kernel will silently ignore it. Moving forward when the setsockopt
succeeds, the new userspace on old kernel means the dump request can
pass an attribute the kernel does not understand. The dump will then
fail as the older kernel does not understand it.
New userspace on new kernel setting the socket option gets the benefit
of the improved data dump.
Kernel side the NETLINK_DUMP_STRICT_CHK uapi is converted to a generic
NETLINK_F_STRICT_CHK flag which can potentially be leveraged for tighter
checking on the NEW, DEL, and SET commands.
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:25 +0000 (20:16 -0700)]
netlink: Add strict version of nlmsg_parse and nla_parse
nla_parse is currently lenient on message parsing, allowing type to be 0
or greater than max expected and only logging a message
"netlink: %d bytes leftover after parsing attributes in process `%s'."
if the netlink message has unknown data at the end after parsing. What this
could mean is that the header at the front of the attributes is actually
wrong and the parsing is shifted from what is expected.
Add a new strict version that actually fails with EINVAL if there are any
bytes remaining after the parsing loop completes, if the atttrbitue type
is 0 or greater than max expected.
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Mon, 8 Oct 2018 03:16:22 +0000 (20:16 -0700)]
netlink: Pass extack to dump handlers
Declare extack in netlink_dump and pass to dump handlers via
netlink_callback. Add any extack message after the dump_done_errno
allowing error messages to be returned. This will be useful when
strict checking is done on dump requests, returning why the dump
fails EINVAL.
Signed-off-by: David Ahern <dsahern@gmail.com> Acked-by: Christian Brauner <christian@brauner.io> Signed-off-by: David S. Miller <davem@davemloft.net>
Al Viro [Mon, 8 Oct 2018 10:22:44 +0000 (06:22 -0400)]
net: sched: cls_u32: simplify the hell out u32_delete() emptiness check
Now that we have the knode count, we can instantly check if
any hnodes are non-empty. And that kills the check for extra
references to root hnode - those could happen only if there was
a knode to carry such a link.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Al Viro [Mon, 8 Oct 2018 10:22:43 +0000 (06:22 -0400)]
net: sched: cls_u32: keep track of knodes count in tc_u_common
allows to simplify u32_delete() considerably
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Al Viro [Mon, 8 Oct 2018 10:22:42 +0000 (06:22 -0400)]
net: sched: cls_u32: get rid of tp_c
Both hnode ->tp_c and tp_c argument of u32_set_parms()
the latter is redundant, the former - never read...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Al Viro [Mon, 8 Oct 2018 10:22:41 +0000 (06:22 -0400)]
net: sched: cls_u32: the tp_c argument of u32_set_parms() is always tp->data
It must be tc_u_common associated with that tp (i.e. tp->data).
Proof:
* both ->ht_up and ->tp_c are assign-once
* ->tp_c of anything inserted into tp_c->hlist is tp_c
* hnodes never get reinserted into the lists or moved
between those, so anything found by u32_lookup_ht(tp->data, ...)
will have ->tp_c equal to tp->data.
* tp->root->tp_c == tp->data.
* ->ht_up of anything inserted into hnode->ht[...] is
equal to hnode.
* knodes never get reinserted into hash chains or moved
between those, so anything returned by u32_lookup_key(ht, ...)
will have ->ht_up equal to ht.
* any knode returned by u32_get(tp, ...) will have ->ht_up->tp_c
point to tp->data
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Al Viro [Mon, 8 Oct 2018 10:22:40 +0000 (06:22 -0400)]
net: sched: cls_u32: pass tc_u_common to u32_set_parms() instead of tc_u_hnode
the only thing we used ht for was ht->tp_c and callers can get that
without going through ->tp_c at all; start with lifting that into
the callers, next commits will massage those, eventually removing
->tp_c altogether.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Al Viro [Mon, 8 Oct 2018 10:22:39 +0000 (06:22 -0400)]
net: sched: cls_u32: clean tc_u_common hashtable
* calculate key *once*, not for each hash chain element
* let tc_u_hash() return the pointer to chain head rather than index -
callers are cleaner that way.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Al Viro [Mon, 8 Oct 2018 10:22:38 +0000 (06:22 -0400)]
net: sched: cls_u32: get rid of tc_u_common ->rcu
unused
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Al Viro [Mon, 8 Oct 2018 10:22:37 +0000 (06:22 -0400)]
net: sched: cls_u32: get rid of tc_u_knode ->tp
not used anymore
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Al Viro [Mon, 8 Oct 2018 10:22:36 +0000 (06:22 -0400)]
net: sched: cls_u32: get rid of unused argument of u32_destroy_key()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Al Viro [Mon, 8 Oct 2018 10:22:35 +0000 (06:22 -0400)]
net: sched: cls_u32: make sure that divisor is a power of 2
Tested by modifying iproute2 to allow sending a divisor > 255
Tested-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Al Viro [Mon, 8 Oct 2018 10:22:34 +0000 (06:22 -0400)]
net: sched: cls_u32: disallow linking to root hnode
Operation makes no sense. Nothing will actually break if we do so
(depth limit in u32_classify() will prevent infinite loops), but
according to maintainers it's best prohibited outright.
NOTE: doing so guarantees that u32_destroy() will trigger the call
of u32_destroy_hnode(); we might want to make that unconditional.
Test:
tc qdisc add dev eth0 ingress
tc filter add dev eth0 parent ffff: protocol ip prio 100 u32 \
link 800: offset at 0 mask 0f00 shift 6 plus 0 eat match ip protocol 6 ff
should fail with
Error: cls_u32: Not linking to root node
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Al Viro [Mon, 8 Oct 2018 10:22:33 +0000 (06:22 -0400)]
net: sched: cls_u32: mark root hnode explicitly
... and produce consistent error on attempt to delete such.
Existing check in u32_delete() is inconsistent - after
tc qdisc add dev eth0 ingress
tc filter add dev eth0 parent ffff: protocol ip prio 100 handle 1: u32 \
divisor 1
tc filter add dev eth0 parent ffff: protocol ip prio 200 handle 2: u32 \
divisor 1
both
tc filter delete dev eth0 parent ffff: protocol ip prio 100 handle 801: u32
and
tc filter delete dev eth0 parent ffff: protocol ip prio 100 handle 800: u32
will fail (at least with refcounting fixes), but the former will complain
about an attempt to remove a busy table, while the latter will recognize
it as root and yield "Not allowed to delete root node" instead.
The problem with the existing check is that several tcf_proto instances
might share the same tp->data and handle-to-hnode lookup will be the same
for all of them. So comparing an hnode to be deleted with tp->root won't
catch the case when one tp is used to try deleting the root of another.
Solution is trivial - mark the root hnodes explicitly upon allocation and
check for that.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
net: phy: mscc: add support for VSC8584 and VSC8574 Microsemi quad-port PHYs
RESEND: rebased on top of latest net-next and on top of latest version of
"net: phy: mscc: various improvements to Microsemi PHY driver" patch
series.
Both PHYs are 4-port PHY that are 10/100/1000BASE-T, 100BASE-FX, 1000BASE-X
and triple-speed copper SFP capable, can communicate with the MAC via
SGMII, QSGMII or 1000BASE-X, supports downshifting and can set the blinking
pattern of each of its 4 LEDs, supports SyncE as well as HP Auto-MDIX
detection.
VSC8574 supports WOL and VSC8584 supports hardware offloading of MACsec.
This patch series add support for 10/100/1000BASE-T, SGMII/QSGMII link with
the MAC, downshifting, HP Auto-MDIX detection and blinking pattern for
their 4 LEDs.
They have also an internal Intel 8051 microcontroller whose firmware needs
to be patched when the PHY is reset. If the 8051's firmware has the
expected CRC, its patching can be skipped. The microcontroller can be
accessed from any port of the PHY, though the CRC function can only be done
through the PHY that is the base PHY of the package (internal address 0)
due to a limitation of the firmware.
The GPIO register bank is a set of registers that are common to all PHYs in
the package. So any modification in any register of this bank affects all
PHYs of the package.
If the PHYs haven't been reset before booting the Linux kernel and were
configured to use interrupts for e.g. link status updates, it is required
to clear the interrupts mask register of all PHYs before being able to use
interrupts with any PHY. The first PHY of the package that will be init
will take care of clearing all PHYs interrupts mask registers. Thus, we
need to keep track of the init sequence in the package, if it's already
been done or if it's to be done.
Most of the init sequence of a PHY of the package is common to all PHYs in
the package, thus we use the SMI broadcast feature which enables us to
propagate a write in one register of one PHY to all PHYs in the same
package.
We also introduce a new development board called PCB120 which exists in
variants for VSC8584 and VSC8574 (and that's the only difference to the
best of my knowledge).
I suggest patches 1 to 3 go through net tree and patches 4 and 5 go
through MIPS tree. Patches going through net tree and those going through
MIPS tree do not depend on one another.
This patch series depends on this patch series:
(https://lore.kernel.org/lkml/20181008100728.24959-1-quentin.schulz@bootlin.com/)
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Quentin Schulz [Mon, 8 Oct 2018 10:14:43 +0000 (12:14 +0200)]
net: phy: mscc: add support for VSC8574 PHY
The VSC8574 PHY is a 4-ports PHY that is 10/100/1000BASE-T, 100BASE-FX,
1000BASE-X and triple-speed copper SFP capable, can communicate with
the MAC via SGMII, QSGMII or 1000BASE-X, supports WOL, downshifting and
can set the blinking pattern of each of its 4 LEDs, supports SyncE as
well as HP Auto-MDIX detection.
This adds support for 10/100/1000BASE-T, SGMII/QSGMII link with the MAC,
WOL, downshifting, HP Auto-MDIX detection and blinking pattern for its 4
LEDs.
The VSC8574 has also an internal Intel 8051 microcontroller whose
firmware needs to be patched when the PHY is reset. If the 8051's
firmware has the expected CRC, its patching can be skipped. The
microcontroller can be accessed from any port of the PHY, though the CRC
function can only be done through the PHY that is the base PHY of the
package (internal address 0) due to a limitation of the firmware.
The GPIO register bank is a set of registers that are common to all PHYs
in the package. So any modification in any register of this bank affects
all PHYs of the package.
If the PHYs haven't been reset before booting the Linux kernel and were
configured to use interrupts for e.g. link status updates, it is
required to clear the interrupts mask register of all PHYs before being
able to use interrupts with any PHY. The first PHY of the package that
will be init will take care of clearing all PHYs interrupts mask
registers. Thus, we need to keep track of the init sequence in the
package, if it's already been done or if it's to be done.
Most of the init sequence of a PHY of the package is common to all PHYs
in the package, thus we use the SMI broadcast feature which enables us
to propagate a write in one register of one PHY to all PHYs in the same
package.
Signed-off-by: Quentin Schulz <quentin.schulz@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Quentin Schulz [Mon, 8 Oct 2018 10:14:42 +0000 (12:14 +0200)]
net: phy: mscc: add support for VSC8584 PHY
The VSC8584 PHY is a 4-ports PHY that is 10/100/1000BASE-T, 100BASE-FX,
1000BASE-X and triple-speed copper SFP capable, can communicate with the
MAC via SGMII, QSGMII or 1000BASE-X, supports downshifting and can set
the blinking pattern of each of its 4 LEDs, supports hardware offloading
of MACsec and supports SyncE as well as HP Auto-MDIX detection.
This adds support for 10/100/1000BASE-T, SGMII/QSGMII link with the MAC,
downshifting, HP Auto-MDIX detection and blinking pattern for its 4
LEDs.
The VSC8584 has also an internal Intel 8051 microcontroller whose
firmware needs to be patched when the PHY is reset. If the 8051's
firmware has the expected CRC, its patching can be skipped. The
microcontroller can be accessed from any port of the PHY, though the CRC
function can only be done through the PHY that is the base PHY of the
package (internal address 0) due to a limitation of the firmware.
The GPIO register bank is a set of registers that are common to all PHYs
in the package. So any modification in any register of this bank affects
all PHYs of the package.
If the PHYs haven't been reset before booting the Linux kernel and were
configured to use interrupts for e.g. link status updates, it is
required to clear the interrupts mask register of all PHYs before being
able to use interrupts with any PHY. The first PHY of the package that
will be init will take care of clearing all PHYs interrupts mask
registers. Thus, we need to keep track of the init sequence in the
package, if it's already been done or if it's to be done.
Most of the init sequence of a PHY of the package is common to all PHYs
in the package, thus we use the SMI broadcast feature which enables us
to propagate a write in one register of one PHY to all PHYs in the same
package.
The revA of the VSC8584 PHY (which is not and will not be publicly
released) should NOT patch the firmware of the microcontroller or it'll
make things worse, the easiest way is just to not support it.
Signed-off-by: Quentin Schulz <quentin.schulz@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Quentin Schulz [Mon, 8 Oct 2018 10:14:41 +0000 (12:14 +0200)]
dt-bindings: net: vsc8531: add two additional LED modes for VSC8584
The VSC8584 (and most likely other PHYs in the same generation) has two
additional LED modes that can be picked, so let's add them.
Reviewed-by: Rob Herring <robh@kernel.org> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Quentin Schulz <quentin.schulz@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
net: phy: mscc: various improvements to Microsemi PHY driver
The Microsemi PHYs have multiple banks of registers (called pages).
Registers can only be accessed from one page, if we need a register from
another page, we need to switch the page and the registers of all other
pages are not accessible anymore.
Basically, to read register 5 from page 0, 1, 2, etc., you do the same
phy_read(phydev, 5); but you need to set the desired page beforehand.
In order to guarantee that two concurrent functions do not change the
page, we need to do some locking per page. This can be achieved with the
use of phy_select_page and phy_restore_page functions but phy_write/read
calls in-between those two functions shall be replaced by their
lock-free alternative __phy_write/read.
The Microsemi PHYs have several counters so let's make them available as PHY
statistics.
The VSC 8530/31/40/41 also need to update their EEE init sequence in order to
avoid packet losses and improve performance.
This patch series also makes some minor cosmetic changes to the driver.
v3:
- add reviewed-by,
- use phy_read/write/modify_paged whenever possible instead of the
combo phy_select_page, __phy_read/write/modify, phy_restore_page when
only one __phy_read/write/modify was executed,
v2:
- add patch to migrate MSCC driver to use phy_restore/select_page,
- migrate all patches from v1 to use those two functions,
- put the multiple lines of constants writes in an array and iterate over
it to write the values,
- add reviewed-bys,
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Here, the rc variable is either used only for the condition right after
the assignment or right before being used as the return value of the
function it's being used in.
So let's remove this unneeded temporary variable whenever possible.
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Quentin Schulz <quentin.schulz@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Quentin Schulz [Mon, 8 Oct 2018 10:07:27 +0000 (12:07 +0200)]
net: phy: mscc: shorten `x != 0` condition to `x`
`if (x != 0)` is basically a more verbose version of `if (x)` so let's
use the latter so it's consistent throughout the whole driver.
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Quentin Schulz <quentin.schulz@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Quentin Schulz [Mon, 8 Oct 2018 10:07:26 +0000 (12:07 +0200)]
net: phy: mscc: remove unneeded parenthesis
The == operator precedes the || operator, so we can remove the
parenthesis around (a == b) || (c == d).
The condition is rather explicit and short so removing the parenthesis
definitely does not make it harder to read.
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Quentin Schulz <quentin.schulz@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Raju Lakkaraju [Mon, 8 Oct 2018 10:07:25 +0000 (12:07 +0200)]
net: phy: mscc: Add EEE init sequence
Microsemi PHYs (VSC 8530/31/40/41) need to update the Energy Efficient
Ethernet initialization sequence.
In order to avoid certain link state errors that could result in link
drops and packet loss, the physical coding sublayer (PCS) must be
updated with settings related to EEE in order to improve performance.
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju@microchip.com> Signed-off-by: Quentin Schulz <quentin.schulz@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Raju Lakkaraju [Mon, 8 Oct 2018 10:07:24 +0000 (12:07 +0200)]
net: phy: mscc: add ethtool statistics counters
There are a few counters available in the PHY: receive errors, false
carriers, link disconnects, media CRC errors and valids counters.
So let's expose those in the PHY driver.
Use the priv structure as the next PHY to be supported has a few
additional counters.
Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju@microsemi.com> Signed-off-by: Quentin Schulz <quentin.schulz@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Quentin Schulz [Mon, 8 Oct 2018 10:07:23 +0000 (12:07 +0200)]
net: phy: mscc: migrate to phy_select/restore_page functions
The Microsemi PHYs have multiple banks of registers (called pages).
Registers can only be accessed from one page, if we need a register from
another page, we need to switch the page and the registers of all other
pages are not accessible anymore.
Basically, to read register 5 from page 0, 1, 2, etc., you do the same
phy_read(phydev, 5); but you need to set the desired page beforehand.
In order to guarantee that two concurrent functions do not change the
page, we need to do some locking per page. This can be achieved with the
use of phy_select_page and phy_restore_page functions but phy_write/read
calls in-between those two functions shall be replaced by their
lock-free alternative __phy_write/read.
Let's migrate this driver to those functions.
Suggested-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Quentin Schulz <quentin.schulz@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Yangbo Lu [Mon, 8 Oct 2018 07:44:28 +0000 (15:44 +0800)]
net: dpaa2: rename rtc as ptp in dpaa2-ptp driver
In dpaa2-ptp driver, it's odd to use rtc in names of
some functions and structures except these dprtc APIs.
This patch is to use ptp instead of rtc in names.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Yangbo Lu [Mon, 8 Oct 2018 07:44:27 +0000 (15:44 +0800)]
net: dpaa2: fix dependency of config FSL_DPAA2_ETH
The NETDEVICES dependency and ETHERNET dependency hadn't
been required since dpaa2-eth was moved out of staging.
Also allowed COMPILE_TEST for dpaa2-eth.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann [Mon, 8 Oct 2018 08:24:14 +0000 (10:24 +0200)]
Merge branch 'bpf-to-bpf-calls-nfp'
Quentin Monnet says:
====================
This patch series adds support for hardware offload of programs containing
BPF-to-BPF function calls. First, a new callback is added to the kernel
verifier, to collect information after the main part of the verification
has been performed. Then support for BPF-to-BPF calls is incrementally
added to the nfp driver, before offloading programs containing such calls
is eventually allowed by lifting the restriction in the kernel verifier, in
the last patch. Please refer to individual patches for details.
Many thanks to Jiong and Jakub for their precious help and contribution on
the main patches for the JIT-compiler, and everything related to stack
accesses.
====================
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Quentin Monnet [Sun, 7 Oct 2018 11:56:58 +0000 (12:56 +0100)]
bpf: allow offload of programs with BPF-to-BPF function calls
Now that there is at least one driver supporting BPF-to-BPF function
calls, lift the restriction, in the verifier, on hardware offload of
eBPF programs containing such calls. But prevent jit_subprogs(), still
in the verifier, from being run for offloaded programs.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jiong Wang <jiong.wang@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Quentin Monnet [Sun, 7 Oct 2018 11:56:57 +0000 (12:56 +0100)]
nfp: bpf: support pointers to other stack frames for BPF-to-BPF calls
Mark instructions that use pointers to areas in the stack outside of the
current stack frame, and process them accordingly in mem_op_stack().
This way, we also support BPF-to-BPF calls where the caller passes a
pointer to data in its own stack frame to the callee (typically, when
the caller passes an address to one of its local variables located in
the stack, as an argument).
Thanks to Jakub and Jiong for figuring out how to deal with this case,
I just had to turn their email discussion into this patch.
Suggested-by: Jiong Wang <jiong.wang@netronome.com> Suggested-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jiong Wang <jiong.wang@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Quentin Monnet [Sun, 7 Oct 2018 11:56:56 +0000 (12:56 +0100)]
nfp: bpf: optimise save/restore for R6~R9 based on register usage
When pre-processing the instructions, it is trivial to detect what
subprograms are using R6, R7, R8 or R9 as destination registers. If a
subprogram uses none of those, then we do not need to jump to the
subroutines dedicated to saving and restoring callee-saved registers in
its prologue and epilogue.
This patch introduces detection of callee-saved registers in subprograms
and prevents the JIT from adding calls to those subroutines whenever we
can: we save some instructions in the translated program, and some time
at runtime on BPF-to-BPF calls and returns.
If no subprogram needs to save those registers, we can avoid appending
the subroutines at the end of the program.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Quentin Monnet [Sun, 7 Oct 2018 11:56:55 +0000 (12:56 +0100)]
nfp: bpf: fix return address from register-saving subroutine to callee
On performing a BPF-to-BPF call, we first jump to a subroutine that
pushes callee-saved registers (R6~R9) to the stack, and from there we
goes to the start of the callee next. In order to do so, the caller must
pass to the subroutine the address of the NFP instruction to jump to at
the end of that subroutine. This cannot be reliably implemented when
translated the caller, as we do not always know the start offset of the
callee yet.
This patch implement the required fixup step for passing the start
offset in the callee via the register used by the subroutine to hold its
return address.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Quentin Monnet [Sun, 7 Oct 2018 11:56:54 +0000 (12:56 +0100)]
nfp: bpf: update fixup function for BPF-to-BPF calls support
Relocation for targets of BPF-to-BPF calls are required at the end of
translation. Update the nfp_fixup_branches() function in that regard.
When checking that the last instruction of each bloc is a branch, we
must account for the length of the instructions required to pop the
return address from the stack.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com> Signed-off-by: Jiong Wang <jiong.wang@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Quentin Monnet [Sun, 7 Oct 2018 11:56:53 +0000 (12:56 +0100)]
nfp: bpf: account for additional stack usage when checking stack limit
Offloaded programs using BPF-to-BPF calls use the stack to store the
return address when calling into a subprogram. Callees also need some
space to save eBPF registers R6 to R9. And contrarily to kernel
verifier, we align stack frames on 64 bytes (and not 32). Account for
all this when checking the stack size limit before JIT-ing the program.
This means we have to recompute maximum stack usage for the program, we
cannot get the value from the kernel.
In addition to adapting the checks on stack usage, move them to the
finalize() callback, now that we have it and because such checks are
part of the verification step rather than translation.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Quentin Monnet [Sun, 7 Oct 2018 11:56:52 +0000 (12:56 +0100)]
nfp: bpf: add main logics for BPF-to-BPF calls support in nfp driver
This is the main patch for the logics of BPF-to-BPF calls in the nfp
driver.
The functions called on BPF_JUMP | BPF_CALL and BPF_JUMP | BPF_EXIT were
used to call helpers and exit from the program, respectively; make them
usable for calling into, or returning from, a BPF subprogram as well.
For all calls, push the return address as well as the callee-saved
registers (R6 to R9) to the stack, and pop them upon returning from the
calls. In order to limit the overhead in terms of instruction number,
this is done through dedicated subroutines. Jumping to the callee
actually consists in jumping to the subroutine, that "returns" to the
callee: this will require some fixup for passing the address in a later
patch. Similarly, returning consists in jumping to the subroutine, which
pops registers and then return directly to the caller (but no fixup is
needed here).
Return to the caller is performed with the RTN instruction newly added
to the JIT.
For the few steps where we need to know what subprogram an instruction
belongs to, the struct nfp_insn_meta is extended with a new subprog_idx
field.
Note that checks on the available stack size, to take into account the
additional requirements associated to BPF-to-BPF calls (storing R6-R9
and return addresses), are added in a later patch.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com> Signed-off-by: Jiong Wang <jiong.wang@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Quentin Monnet [Sun, 7 Oct 2018 11:56:51 +0000 (12:56 +0100)]
nfp: bpf: account for BPF-to-BPF calls when preparing nfp JIT
Similarly to "exit" or "helper call" instructions, BPF-to-BPF calls will
require additional processing before translation starts, in order to
record and mark jump destinations.
We also mark the instructions where each subprogram begins. This will be
used in a following commit to determine where to add prologues for
subprograms.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jiong Wang <jiong.wang@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
projects / kernel.git / log