Pablo Neira Ayuso says:
====================
Netfilter/IPVS updates for net-next
The following patchset contains Netfilter updates for net-next:
1) Remove the broute pseudo hook, implement this from the bridge
prerouting hook instead. Now broute becomes real table in ebtables,
from Florian Westphal. This also includes a size reduction patch for the
bridge control buffer area via squashing boolean into bitfields and
a selftest.
2) Add OS passive fingerprint version matching, from Fernando Fernandez.
3) Support for gue encapsulation for IPVS, from Jacky Hu.
4) Add support for NAT to the inet family, from Florian Westphal.
This includes support for masquerade, redirect and nat extensions.
5) Skip interface lookup in flowtable, use device in the dst object.
6) Add jiffies64_to_msecs() and use it, from Li RongQing.
7) Remove unused parameter in nf_tables_set_desc_parse(), from Colin Ian King.
8) Statify several functions, patches from YueHaibing and Florian Westphal.
9) Add an optimized version of nf_inet_addr_cmp(), from Li RongQing.
10) Merge route extension to core, also from Florian.
11) Use IS_ENABLED(CONFIG_NF_NAT) instead of NF_NAT_NEEDED, from Florian.
12) Merge ip/ip6 masquerade extensions, from Florian. This includes
netdevice notifier unification.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
As reported by Guenter Roeck, the new bit-locking using
BIT(1) doesn't work on the m68k architecture. m68k only requires
2-byte alignment for words and longwords, so there is only one
unused bit in pointers to structs - We current use two, one for the
NULLS marker at the end of the linked list, and one for the bit-lock
in the head of the list.
The two uses don't need to conflict as we never need the head of the
list to be a NULLS marker - the marker is only needed to check if an
object has moved to a different table, and the bucket head cannot
move. The NULLS marker is only needed in a ->next pointer.
As we already have different types for the bucket head pointer (struct
rhash_lock_head) and the ->next pointers (struct rhash_head), it is
fairly easy to treat the lsb differently in each.
So: Initialize buckets heads to NULL, and use the lsb for locking.
When loading the pointer from the bucket head, if it is NULL (ignoring
the lock big), report as being the expected NULLS marker.
When storing a value into a bucket head, if it is a NULLS marker,
store NULL instead.
And convert all places that used bit 1 for locking, to use bit 0.
Fixes: 8f0db01800 ("rhashtable: use bit_spin_locks to protect hash bucket.")
Reported-by: Guenter Roeck <linux@roeck-us.net>
Tested-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The only times rht_ptr_locked() is used, it is to store a new
value in a bucket-head. This is the only time it makes sense
to use it too. So replace it by a function which does the
whole task: Sets the lock bit and assigns to a bucket head.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rather than dereferencing a pointer to a bucket and then passing the
result to rht_ptr(), we now pass in the pointer and do the dereference
in rht_ptr().
This requires that we pass in the tbl and hash as well to support RCU
checks, and means that the various rht_for_each functions can expect a
pointer that can be dereferenced without further care.
There are two places where we dereference a bucket pointer
where there is no testable protection - in each case we know
that we much have exclusive access without having taken a lock.
The previous code used rht_dereference() to pretend that holding
the mutex provided protects, but holding the mutex never provides
protection for accessing buckets.
So instead introduce rht_ptr_exclusive() that can be used when
there is known to be exclusive access without holding any locks.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch only moves some code around, it doesn't
change the code at all.
A subsequent patch will benefit from this as it needs
to add calls to functions which are now defined before the
call-site, but weren't before.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
With these annotations, the rhashtable now gets no
warnings when compiled with "C=1" for sparse checking.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann says:
====================
pull-request: bpf-next 2019-04-12
The following pull-request contains BPF updates for your *net-next* tree.
The main changes are:
1) Improve BPF verifier scalability for large programs through two
optimizations: i) remove verifier states that are not useful in pruning,
ii) stop walking parentage chain once first LIVE_READ is seen. Combined
gives approx 20x speedup. Increase limits for accepting large programs
under root, and add various stress tests, from Alexei.
2) Implement global data support in BPF. This enables static global variables
for .data, .rodata and .bss sections to be properly handled which allows
for more natural program development. This also opens up the possibility
to optimize program workflow by compiling ELFs only once and later only
rewriting section data before reload, from Daniel and with test cases and
libbpf refactoring from Joe.
3) Add config option to generate BTF type info for vmlinux as part of the
kernel build process. DWARF debug info is converted via pahole to BTF.
Latter relies on libbpf and makes use of BTF deduplication algorithm which
results in 100x savings compared to DWARF data. Resulting .BTF section is
typically about 2MB in size, from Andrii.
4) Add BPF verifier support for stack access with variable offset from
helpers and add various test cases along with it, from Andrey.
5) Extend bpf_skb_adjust_room() growth BPF helper to mark inner MAC header
so that L2 encapsulation can be used for tc tunnels, from Alan.
6) Add support for input __sk_buff context in BPF_PROG_TEST_RUN so that
users can define a subset of allowed __sk_buff fields that get fed into
the test program, from Stanislav.
7) Add bpf fs multi-dimensional array tests for BTF test suite and fix up
various UBSAN warnings in bpftool, from Yonghong.
8) Generate a pkg-config file for libbpf, from Luca.
9) Dump program's BTF id in bpftool, from Prashant.
10) libbpf fix to use smaller BPF log buffer size for AF_XDP's XDP
program, from Magnus.
11) kallsyms related fixes for the case when symbols are not present in
BPF selftests and samples, from Daniel
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
This makes broute a normal ebtables table, hooking at PREROUTING.
The broute hook is removed.
It uses skb->cb to signal to bridge rx handler that the skb should be
routed instead of being bridged.
This change is backwards compatible with ebtables as no userspace visible
parts are changed.
This means we can also remove the !ops test in ebt_register_table,
it was only there for broute table sake.
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: David S. Miller <davem@davemloft.net>
Acked-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Commit b0b9395d86 ("bpf: support input __sk_buff context in
BPF_PROG_TEST_RUN") started using bpf_check_uarg_tail_zero in
BPF_PROG_TEST_RUN. However, bpf_check_uarg_tail_zero is not defined
for !CONFIG_BPF_SYSCALL:
net/bpf/test_run.c: In function ‘bpf_ctx_init’:
net/bpf/test_run.c:142:9: error: implicit declaration of function ‘bpf_check_uarg_tail_zero’ [-Werror=implicit-function-declaration]
err = bpf_check_uarg_tail_zero(data_in, max_size, size);
^~~~~~~~~~~~~~~~~~~~~~~~
Let's not build net/bpf/test_run.c when CONFIG_BPF_SYSCALL is not set.
Reported-by: kbuild test robot <lkp@intel.com>
Fixes: b0b9395d86 ("bpf: support input __sk_buff context in BPF_PROG_TEST_RUN")
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Given we'll be reusing BPF array maps for global data/bss/rodata
sections, we need a way to associate BTF DataSec type as its map
value type. In usual cases we have this ugly BPF_ANNOTATE_KV_PAIR()
macro hack e.g. via 38d5d3b3d5 ("bpf: Introduce BPF_ANNOTATE_KV_PAIR")
to get initial map to type association going. While more use cases
for it are discouraged, this also won't work for global data since
the use of array map is a BPF loader detail and therefore unknown
at compilation time. For array maps with just a single entry we make
an exception in terms of BTF in that key type is declared optional
if value type is of DataSec type. The latter LLVM is guaranteed to
emit and it also aligns with how we regard global data maps as just
a plain buffer area reusing existing map facilities for allowing
things like introspection with existing tools.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This patch adds a new BPF_MAP_FREEZE command which allows to
"freeze" the map globally as read-only / immutable from syscall
side.
Map permission handling has been refactored into map_get_sys_perms()
and drops FMODE_CAN_WRITE in case of locked map. Main use case is
to allow for setting up .rodata sections from the BPF ELF which
are loaded into the kernel, meaning BPF loader first allocates
map, sets up map value by copying .rodata section into it and once
complete, it calls BPF_MAP_FREEZE on the map fd to prevent further
modifications.
Right now BPF_MAP_FREEZE only takes map fd as argument while remaining
bpf_attr members are required to be zero. I didn't add write-only
locking here as counterpart since I don't have a concrete use-case
for it on my side, and I think it makes probably more sense to wait
once there is actually one. In that case bpf_attr can be extended
as usual with a flag field and/or others where flag 0 means that
we lock the map read-only hence this doesn't prevent to add further
extensions to BPF_MAP_FREEZE upon need.
A map creation flag like BPF_F_WRONCE was not considered for couple
of reasons: i) in case of a generic implementation, a map can consist
of more than just one element, thus there could be multiple map
updates needed to set the map into a state where it can then be
made immutable, ii) WRONCE indicates exact one-time write before
it is then set immutable. A generic implementation would set a bit
atomically on map update entry (if unset), indicating that every
subsequent update from then onwards will need to bail out there.
However, map updates can fail, so upon failure that flag would need
to be unset again and the update attempt would need to be repeated
for it to be eventually made immutable. While this can be made
race-free, this approach feels less clean and in combination with
reason i), it's not generic enough. A dedicated BPF_MAP_FREEZE
command directly sets the flag and caller has the guarantee that
map is immutable from syscall side upon successful return for any
future syscall invocations that would alter the map state, which
is also more intuitive from an API point of view. A command name
such as BPF_MAP_LOCK has been avoided as it's too close with BPF
map spin locks (which already has BPF_F_LOCK flag). BPF_MAP_FREEZE
is so far only enabled for privileged users.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This work adds two new map creation flags BPF_F_RDONLY_PROG
and BPF_F_WRONLY_PROG in order to allow for read-only or
write-only BPF maps from a BPF program side.
Today we have BPF_F_RDONLY and BPF_F_WRONLY, but this only
applies to system call side, meaning the BPF program has full
read/write access to the map as usual while bpf(2) calls with
map fd can either only read or write into the map depending
on the flags. BPF_F_RDONLY_PROG and BPF_F_WRONLY_PROG allows
for the exact opposite such that verifier is going to reject
program loads if write into a read-only map or a read into a
write-only map is detected. For read-only map case also some
helpers are forbidden for programs that would alter the map
state such as map deletion, update, etc. As opposed to the two
BPF_F_RDONLY / BPF_F_WRONLY flags, BPF_F_RDONLY_PROG as well
as BPF_F_WRONLY_PROG really do correspond to the map lifetime.
We've enabled this generic map extension to various non-special
maps holding normal user data: array, hash, lru, lpm, local
storage, queue and stack. Further generic map types could be
followed up in future depending on use-case. Main use case
here is to forbid writes into .rodata map values from verifier
side.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This generic extension to BPF maps allows for directly loading
an address residing inside a BPF map value as a single BPF
ldimm64 instruction!
The idea is similar to what BPF_PSEUDO_MAP_FD does today, which
is a special src_reg flag for ldimm64 instruction that indicates
that inside the first part of the double insns's imm field is a
file descriptor which the verifier then replaces as a full 64bit
address of the map into both imm parts. For the newly added
BPF_PSEUDO_MAP_VALUE src_reg flag, the idea is the following:
the first part of the double insns's imm field is again a file
descriptor corresponding to the map, and the second part of the
imm field is an offset into the value. The verifier will then
replace both imm parts with an address that points into the BPF
map value at the given value offset for maps that support this
operation. Currently supported is array map with single entry.
It is possible to support more than just single map element by
reusing both 16bit off fields of the insns as a map index, so
full array map lookup could be expressed that way. It hasn't
been implemented here due to lack of concrete use case, but
could easily be done so in future in a compatible way, since
both off fields right now have to be 0 and would correctly
denote a map index 0.
The BPF_PSEUDO_MAP_VALUE is a distinct flag as otherwise with
BPF_PSEUDO_MAP_FD we could not differ offset 0 between load of
map pointer versus load of map's value at offset 0, and changing
BPF_PSEUDO_MAP_FD's encoding into off by one to differ between
regular map pointer and map value pointer would add unnecessary
complexity and increases barrier for debugability thus less
suitable. Using the second part of the imm field as an offset
into the value does /not/ come with limitations since maximum
possible value size is in u32 universe anyway.
This optimization allows for efficiently retrieving an address
to a map value memory area without having to issue a helper call
which needs to prepare registers according to calling convention,
etc, without needing the extra NULL test, and without having to
add the offset in an additional instruction to the value base
pointer. The verifier then treats the destination register as
PTR_TO_MAP_VALUE with constant reg->off from the user passed
offset from the second imm field, and guarantees that this is
within bounds of the map value. Any subsequent operations are
normally treated as typical map value handling without anything
extra needed from verification side.
The two map operations for direct value access have been added to
array map for now. In future other types could be supported as
well depending on the use case. The main use case for this commit
is to allow for BPF loader support for global variables that
reside in .data/.rodata/.bss sections such that we can directly
load the address of them with minimal additional infrastructure
required. Loader support has been added in subsequent commits for
libbpf library.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
genphy_read_status() so far checks phydev->supported, not the actual
PHY capabilities. This can make a difference if the supported speeds
have been limited by of_set_phy_supported() or phy_set_max_speed().
It seems that this issue only affects the link partner advertisements
as displayed by ethtool. Also this patch wouldn't apply to older
kernels because linkmode bitmaps have been introduced recently.
Therefore net-next.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Saeed Mamameed says:
====================
mlx5-updates-2019-04-02
This series provides misc updates to mlx5 driver
1) Aya Levin (1): Handle event of power detection in the PCIE slot
2) Eli Britstein (6):
Some TC VLAN related updates and fixes to the previous VLAN modify action
support patchset.
Offload TC e-switch rules with egress/ingress VLAN devices
3) Max Gurtovoy (1): Fix double mutex initialization in esiwtch.c
4) Tariq Toukan (3): Misc small updates
A write memory barrier is sufficient in EQ ci update
Obsolete param field holding a constant value
Unify logic of MTU boundaries
5) Tonghao Zhang (4): Misc updates to en_tc.c
Make the log friendly when decapsulation offload not supported
Remove 'parse_attr' argument in parse_tc_fdb_actions()
Deletes unnecessary setting of esw_attr->parse_attr
Return -EOPNOTSUPP when attempting to offload an unsupported action
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Add version option support to the nftables "osf" expression.
Signed-off-by: Fernando Fernandez Mancera <ffmancera@riseup.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
NF_NAT_NEEDED is true whenever nat support for either ipv4 or ipv6 is
enabled. Now that the af-specific nat configuration switches have been
removed, IS_ENABLED(CONFIG_NF_NAT) has the same effect.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
very little code, so it really doesn't make sense to have extra
modules or even a kconfig knob for this.
Merge them and make functionality available unconditionally.
The merge makes inet family route support trivial, so add it
as well here.
Before:
text data bss dec hex filename
835 832 0 1667 683 nft_chain_route_ipv4.ko
870 832 0 1702 6a6 nft_chain_route_ipv6.ko
111568 2556 529 114653 1bfdd nf_tables.ko
After:
text data bss dec hex filename
113133 2556 529 116218 1c5fa nf_tables.ko
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
there is a similar helper in net/netfilter/nf_tables_api.c,
this maybe become a common request someday, so move it to
time.c
Signed-off-by: Zhang Yu <zhangyu31@baidu.com>
Signed-off-by: Li RongQing <lirongqing@baidu.com>
Acked-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
After commit a297569fe0 ("net/udp: do not touch skb->peeked unless
really needed") the 'peeked' argument of __skb_try_recv_datagram()
and friends is always equal to !!'flags & MSG_PEEK'.
Since such argument is really a boolean info, and the callers have
already 'flags & MSG_PEEK' handy, we can remove it and clean-up the
code a bit.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Native bit_spin_locks are not tracked by lockdep.
The bit_spin_locks used for rhashtable buckets are local
to the rhashtable implementation, so there is little opportunity
for the sort of misuse that lockdep might detect.
However locks are held while a hash function or compare
function is called, and if one of these took a lock,
a misbehaviour is possible.
As it is quite easy to add lockdep support this unlikely
possibility seems to be enough justification.
So create a lockdep class for bucket bit_spin_lock and attach
through a lockdep_map in each bucket_table.
Without the 'nested' annotation in rhashtable_rehash_one(), lockdep
correctly reports a possible problem as this lock is taken
while another bucket lock (in another table) is held. This
confirms that the added support works.
With the correct nested annotation in place, lockdep reports
no problems.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch changes rhashtables to use a bit_spin_lock on BIT(1) of the
bucket pointer to lock the hash chain for that bucket.
The benefits of a bit spin_lock are:
- no need to allocate a separate array of locks.
- no need to have a configuration option to guide the
choice of the size of this array
- locking cost is often a single test-and-set in a cache line
that will have to be loaded anyway. When inserting at, or removing
from, the head of the chain, the unlock is free - writing the new
address in the bucket head implicitly clears the lock bit.
For __rhashtable_insert_fast() we ensure this always happens
when adding a new key.
- even when lockings costs 2 updates (lock and unlock), they are
in a cacheline that needs to be read anyway.
The cost of using a bit spin_lock is a little bit of code complexity,
which I think is quite manageable.
Bit spin_locks are sometimes inappropriate because they are not fair -
if multiple CPUs repeatedly contend of the same lock, one CPU can
easily be starved. This is not a credible situation with rhashtable.
Multiple CPUs may want to repeatedly add or remove objects, but they
will typically do so at different buckets, so they will attempt to
acquire different locks.
As we have more bit-locks than we previously had spinlocks (by at
least a factor of two) we can expect slightly less contention to
go with the slightly better cache behavior and reduced memory
consumption.
To enhance type checking, a new struct is introduced to represent the
pointer plus lock-bit
that is stored in the bucket-table. This is "struct rhash_lock_head"
and is empty. A pointer to this needs to be cast to either an
unsigned lock, or a "struct rhash_head *" to be useful.
Variables of this type are most often called "bkt".
Previously "pprev" would sometimes point to a bucket, and sometimes a
->next pointer in an rhash_head. As these are now different types,
pprev is NULL when it would have pointed to the bucket. In that case,
'blk' is used, together with correct locking protocol.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rather than returning a pointer to a static nulls, rht_bucket_var()
now returns NULL if the bucket doesn't exist.
This will make the next patch, which stores a bitlock in the
bucket pointer, somewhat cleaner.
This change involves introducing __rht_bucket_nested() which is
like rht_bucket_nested(), but doesn't provide the static nulls,
and changing rht_bucket_nested() to call this and possible
provide a static nulls - as is still needed for the non-var case.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 9c225f2655 ("vfs: atomic f_pos accesses as per POSIX") added
locking for file.f_pos access and in particular made concurrent read and
write not possible - now both those functions take f_pos lock for the
whole run, and so if e.g. a read is blocked waiting for data, write will
deadlock waiting for that read to complete.
This caused regression for stream-like files where previously read and
write could run simultaneously, but after that patch could not do so
anymore. See e.g. commit 581d21a2d0 ("xenbus: fix deadlock on writes
to /proc/xen/xenbus") which fixes such regression for particular case of
/proc/xen/xenbus.
The patch that added f_pos lock in 2014 did so to guarantee POSIX thread
safety for read/write/lseek and added the locking to file descriptors of
all regular files. In 2014 that thread-safety problem was not new as it
was already discussed earlier in 2006.
However even though 2006'th version of Linus's patch was adding f_pos
locking "only for files that are marked seekable with FMODE_LSEEK (thus
avoiding the stream-like objects like pipes and sockets)", the 2014
version - the one that actually made it into the tree as 9c225f2655 -
is doing so irregardless of whether a file is seekable or not.
See
https://lore.kernel.org/lkml/53022DB1.4070805@gmail.com/https://lwn.net/Articles/180387https://lwn.net/Articles/180396
for historic context.
The reason that it did so is, probably, that there are many files that
are marked non-seekable, but e.g. their read implementation actually
depends on knowing current position to correctly handle the read. Some
examples:
kernel/power/user.c snapshot_read
fs/debugfs/file.c u32_array_read
fs/fuse/control.c fuse_conn_waiting_read + ...
drivers/hwmon/asus_atk0110.c atk_debugfs_ggrp_read
arch/s390/hypfs/inode.c hypfs_read_iter
...
Despite that, many nonseekable_open users implement read and write with
pure stream semantics - they don't depend on passed ppos at all. And for
those cases where read could wait for something inside, it creates a
situation similar to xenbus - the write could be never made to go until
read is done, and read is waiting for some, potentially external, event,
for potentially unbounded time -> deadlock.
Besides xenbus, there are 14 such places in the kernel that I've found
with semantic patch (see below):
drivers/xen/evtchn.c:667:8-24: ERROR: evtchn_fops: .read() can deadlock .write()
drivers/isdn/capi/capi.c:963:8-24: ERROR: capi_fops: .read() can deadlock .write()
drivers/input/evdev.c:527:1-17: ERROR: evdev_fops: .read() can deadlock .write()
drivers/char/pcmcia/cm4000_cs.c:1685:7-23: ERROR: cm4000_fops: .read() can deadlock .write()
net/rfkill/core.c:1146:8-24: ERROR: rfkill_fops: .read() can deadlock .write()
drivers/s390/char/fs3270.c:488:1-17: ERROR: fs3270_fops: .read() can deadlock .write()
drivers/usb/misc/ldusb.c:310:1-17: ERROR: ld_usb_fops: .read() can deadlock .write()
drivers/hid/uhid.c:635:1-17: ERROR: uhid_fops: .read() can deadlock .write()
net/batman-adv/icmp_socket.c:80:1-17: ERROR: batadv_fops: .read() can deadlock .write()
drivers/media/rc/lirc_dev.c:198:1-17: ERROR: lirc_fops: .read() can deadlock .write()
drivers/leds/uleds.c:77:1-17: ERROR: uleds_fops: .read() can deadlock .write()
drivers/input/misc/uinput.c:400:1-17: ERROR: uinput_fops: .read() can deadlock .write()
drivers/infiniband/core/user_mad.c:985:7-23: ERROR: umad_fops: .read() can deadlock .write()
drivers/gnss/core.c:45:1-17: ERROR: gnss_fops: .read() can deadlock .write()
In addition to the cases above another regression caused by f_pos
locking is that now FUSE filesystems that implement open with
FOPEN_NONSEEKABLE flag, can no longer implement bidirectional
stream-like files - for the same reason as above e.g. read can deadlock
write locking on file.f_pos in the kernel.
FUSE's FOPEN_NONSEEKABLE was added in 2008 in a7c1b990f7 ("fuse:
implement nonseekable open") to support OSSPD. OSSPD implements /dev/dsp
in userspace with FOPEN_NONSEEKABLE flag, with corresponding read and
write routines not depending on current position at all, and with both
read and write being potentially blocking operations:
See
https://github.com/libfuse/osspdhttps://lwn.net/Articles/308445https://github.com/libfuse/osspd/blob/14a9cff0/osspd.c#L1406https://github.com/libfuse/osspd/blob/14a9cff0/osspd.c#L1438-L1477https://github.com/libfuse/osspd/blob/14a9cff0/osspd.c#L1479-L1510
Corresponding libfuse example/test also describes FOPEN_NONSEEKABLE as
"somewhat pipe-like files ..." with read handler not using offset.
However that test implements only read without write and cannot exercise
the deadlock scenario:
https://github.com/libfuse/libfuse/blob/fuse-3.4.2-3-ga1bff7d/example/poll.c#L124-L131https://github.com/libfuse/libfuse/blob/fuse-3.4.2-3-ga1bff7d/example/poll.c#L146-L163https://github.com/libfuse/libfuse/blob/fuse-3.4.2-3-ga1bff7d/example/poll.c#L209-L216
I've actually hit the read vs write deadlock for real while implementing
my FUSE filesystem where there is /head/watch file, for which open
creates separate bidirectional socket-like stream in between filesystem
and its user with both read and write being later performed
simultaneously. And there it is semantically not easy to split the
stream into two separate read-only and write-only channels:
https://lab.nexedi.com/kirr/wendelin.core/blob/f13aa600/wcfs/wcfs.go#L88-169
Let's fix this regression. The plan is:
1. We can't change nonseekable_open to include &~FMODE_ATOMIC_POS -
doing so would break many in-kernel nonseekable_open users which
actually use ppos in read/write handlers.
2. Add stream_open() to kernel to open stream-like non-seekable file
descriptors. Read and write on such file descriptors would never use
nor change ppos. And with that property on stream-like files read and
write will be running without taking f_pos lock - i.e. read and write
could be running simultaneously.
3. With semantic patch search and convert to stream_open all in-kernel
nonseekable_open users for which read and write actually do not
depend on ppos and where there is no other methods in file_operations
which assume @offset access.
4. Add FOPEN_STREAM to fs/fuse/ and open in-kernel file-descriptors via
steam_open if that bit is present in filesystem open reply.
It was tempting to change fs/fuse/ open handler to use stream_open
instead of nonseekable_open on just FOPEN_NONSEEKABLE flags, but
grepping through Debian codesearch shows users of FOPEN_NONSEEKABLE,
and in particular GVFS which actually uses offset in its read and
write handlers
https://codesearch.debian.net/search?q=-%3Enonseekable+%3Dhttps://gitlab.gnome.org/GNOME/gvfs/blob/1.40.0-6-gcbc54396/client/gvfsfusedaemon.c#L1080https://gitlab.gnome.org/GNOME/gvfs/blob/1.40.0-6-gcbc54396/client/gvfsfusedaemon.c#L1247-1346https://gitlab.gnome.org/GNOME/gvfs/blob/1.40.0-6-gcbc54396/client/gvfsfusedaemon.c#L1399-1481
so if we would do such a change it will break a real user.
5. Add stream_open and FOPEN_STREAM handling to stable kernels starting
from v3.14+ (the kernel where 9c225f2655 first appeared).
This will allow to patch OSSPD and other FUSE filesystems that
provide stream-like files to return FOPEN_STREAM | FOPEN_NONSEEKABLE
in their open handler and this way avoid the deadlock on all kernel
versions. This should work because fs/fuse/ ignores unknown open
flags returned from a filesystem and so passing FOPEN_STREAM to a
kernel that is not aware of this flag cannot hurt. In turn the kernel
that is not aware of FOPEN_STREAM will be < v3.14 where just
FOPEN_NONSEEKABLE is sufficient to implement streams without read vs
write deadlock.
This patch adds stream_open, converts /proc/xen/xenbus to it and adds
semantic patch to automatically locate in-kernel places that are either
required to be converted due to read vs write deadlock, or that are just
safe to be converted because read and write do not use ppos and there
are no other funky methods in file_operations.
Regarding semantic patch I've verified each generated change manually -
that it is correct to convert - and each other nonseekable_open instance
left - that it is either not correct to convert there, or that it is not
converted due to current stream_open.cocci limitations.
The script also does not convert files that should be valid to convert,
but that currently have .llseek = noop_llseek or generic_file_llseek for
unknown reason despite file being opened with nonseekable_open (e.g.
drivers/input/mousedev.c)
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Cc: Yongzhi Pan <panyongzhi@gmail.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: David Vrabel <david.vrabel@citrix.com>
Cc: Juergen Gross <jgross@suse.com>
Cc: Miklos Szeredi <miklos@szeredi.hu>
Cc: Tejun Heo <tj@kernel.org>
Cc: Kirill Tkhai <ktkhai@virtuozzo.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Julia Lawall <Julia.Lawall@lip6.fr>
Cc: Nikolaus Rath <Nikolaus@rath.org>
Cc: Han-Wen Nienhuys <hanwen@google.com>
Signed-off-by: Kirill Smelkov <kirr@nexedi.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Merge misc fixes from Andrew Morton:
"14 fixes"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
kernel/sysctl.c: fix out-of-bounds access when setting file-max
mm/util.c: fix strndup_user() comment
sh: fix multiple function definition build errors
MAINTAINERS: add maintainer and replacing reviewer ARM/NUVOTON NPCM
MAINTAINERS: fix bad pattern in ARM/NUVOTON NPCM
mm: writeback: use exact memcg dirty counts
psi: clarify the units used in pressure files
mm/huge_memory.c: fix modifying of page protection by insert_pfn_pmd()
hugetlbfs: fix memory leak for resv_map
mm: fix vm_fault_t cast in VM_FAULT_GET_HINDEX()
lib/lzo: fix bugs for very short or empty input
include/linux/bitrev.h: fix constant bitrev
kmemleak: powerpc: skip scanning holes in the .bss section
lib/string.c: implement a basic bcmp
Since commit a983b5ebee ("mm: memcontrol: fix excessive complexity in
memory.stat reporting") memcg dirty and writeback counters are managed
as:
1) per-memcg per-cpu values in range of [-32..32]
2) per-memcg atomic counter
When a per-cpu counter cannot fit in [-32..32] it's flushed to the
atomic. Stat readers only check the atomic. Thus readers such as
balance_dirty_pages() may see a nontrivial error margin: 32 pages per
cpu.
Assuming 100 cpus:
4k x86 page_size: 13 MiB error per memcg
64k ppc page_size: 200 MiB error per memcg
Considering that dirty+writeback are used together for some decisions the
errors double.
This inaccuracy can lead to undeserved oom kills. One nasty case is
when all per-cpu counters hold positive values offsetting an atomic
negative value (i.e. per_cpu[*]=32, atomic=n_cpu*-32).
balance_dirty_pages() only consults the atomic and does not consider
throttling the next n_cpu*32 dirty pages. If the file_lru is in the
13..200 MiB range then there's absolutely no dirty throttling, which
burdens vmscan with only dirty+writeback pages thus resorting to oom
kill.
It could be argued that tiny containers are not supported, but it's more
subtle. It's the amount the space available for file lru that matters.
If a container has memory.max-200MiB of non reclaimable memory, then it
will also suffer such oom kills on a 100 cpu machine.
The following test reliably ooms without this patch. This patch avoids
oom kills.
$ cat test
mount -t cgroup2 none /dev/cgroup
cd /dev/cgroup
echo +io +memory > cgroup.subtree_control
mkdir test
cd test
echo 10M > memory.max
(echo $BASHPID > cgroup.procs && exec /memcg-writeback-stress /foo)
(echo $BASHPID > cgroup.procs && exec dd if=/dev/zero of=/foo bs=2M count=100)
$ cat memcg-writeback-stress.c
/*
* Dirty pages from all but one cpu.
* Clean pages from the non dirtying cpu.
* This is to stress per cpu counter imbalance.
* On a 100 cpu machine:
* - per memcg per cpu dirty count is 32 pages for each of 99 cpus
* - per memcg atomic is -99*32 pages
* - thus the complete dirty limit: sum of all counters 0
* - balance_dirty_pages() only sees atomic count -99*32 pages, which
* it max()s to 0.
* - So a workload can dirty -99*32 pages before balance_dirty_pages()
* cares.
*/
#define _GNU_SOURCE
#include <err.h>
#include <fcntl.h>
#include <sched.h>
#include <stdlib.h>
#include <stdio.h>
#include <sys/stat.h>
#include <sys/sysinfo.h>
#include <sys/types.h>
#include <unistd.h>
static char *buf;
static int bufSize;
static void set_affinity(int cpu)
{
cpu_set_t affinity;
CPU_ZERO(&affinity);
CPU_SET(cpu, &affinity);
if (sched_setaffinity(0, sizeof(affinity), &affinity))
err(1, "sched_setaffinity");
}
static void dirty_on(int output_fd, int cpu)
{
int i, wrote;
set_affinity(cpu);
for (i = 0; i < 32; i++) {
for (wrote = 0; wrote < bufSize; ) {
int ret = write(output_fd, buf+wrote, bufSize-wrote);
if (ret == -1)
err(1, "write");
wrote += ret;
}
}
}
int main(int argc, char **argv)
{
int cpu, flush_cpu = 1, output_fd;
const char *output;
if (argc != 2)
errx(1, "usage: output_file");
output = argv[1];
bufSize = getpagesize();
buf = malloc(getpagesize());
if (buf == NULL)
errx(1, "malloc failed");
output_fd = open(output, O_CREAT|O_RDWR);
if (output_fd == -1)
err(1, "open(%s)", output);
for (cpu = 0; cpu < get_nprocs(); cpu++) {
if (cpu != flush_cpu)
dirty_on(output_fd, cpu);
}
set_affinity(flush_cpu);
if (fsync(output_fd))
err(1, "fsync(%s)", output);
if (close(output_fd))
err(1, "close(%s)", output);
free(buf);
}
Make balance_dirty_pages() and wb_over_bg_thresh() work harder to
collect exact per memcg counters. This avoids the aforementioned oom
kills.
This does not affect the overhead of memory.stat, which still reads the
single atomic counter.
Why not use percpu_counter? memcg already handles cpus going offline, so
no need for that overhead from percpu_counter. And the percpu_counter
spinlocks are more heavyweight than is required.
It probably also makes sense to use exact dirty and writeback counters
in memcg oom reports. But that is saved for later.
Link: http://lkml.kernel.org/r/20190329174609.164344-1-gthelen@google.com
Signed-off-by: Greg Thelen <gthelen@google.com>
Reviewed-by: Roman Gushchin <guro@fb.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: <stable@vger.kernel.org> [4.16+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
clang points out with hundreds of warnings that the bitrev macros have a
problem with constant input:
drivers/hwmon/sht15.c:187:11: error: variable '__x' is uninitialized when used within its own initialization
[-Werror,-Wuninitialized]
u8 crc = bitrev8(data->val_status & 0x0F);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
include/linux/bitrev.h:102:21: note: expanded from macro 'bitrev8'
__constant_bitrev8(__x) : \
~~~~~~~~~~~~~~~~~~~^~~~
include/linux/bitrev.h:67:11: note: expanded from macro '__constant_bitrev8'
u8 __x = x; \
~~~ ^
Both the bitrev and the __constant_bitrev macros use an internal
variable named __x, which goes horribly wrong when passing one to the
other.
The obvious fix is to rename one of the variables, so this adds an extra
'_'.
It seems we got away with this because
- there are only a few drivers using bitrev macros
- usually there are no constant arguments to those
- when they are constant, they tend to be either 0 or (unsigned)-1
(drivers/isdn/i4l/isdnhdlc.o, drivers/iio/amplifiers/ad8366.c) and
give the correct result by pure chance.
In fact, the only driver that I could find that gets different results
with this is drivers/net/wan/slic_ds26522.c, which in turn is a driver
for fairly rare hardware (adding the maintainer to Cc for testing).
Link: http://lkml.kernel.org/r/20190322140503.123580-1-arnd@arndb.de
Fixes: 556d2f055b ("ARM: 8187/1: add CONFIG_HAVE_ARCH_BITREVERSE to support rbit instruction")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Cc: Zhao Qiang <qiang.zhao@nxp.com>
Cc: Yalin Wang <yalin.wang@sonymobile.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull syscall-get-arguments cleanup and fixes from Steven Rostedt:
"Andy Lutomirski approached me to tell me that the
syscall_get_arguments() implementation in x86 was horrible and gcc
certainly gets it wrong.
He said that since the tracepoints only pass in 0 and 6 for i and n
repectively, it should be optimized for that case. Inspecting the
kernel, I discovered that all users pass in 0 for i and only one file
passing in something other than 6 for the number of arguments. That
code happens to be my own code used for the special syscall tracing.
That can easily be converted to just using 0 and 6 as well, and only
copying what is needed. Which is probably the faster path anyway for
that case.
Along the way, a couple of real fixes came from this as the
syscall_get_arguments() function was incorrect for csky and riscv.
x86 has been optimized to for the new interface that removes the
variable number of arguments, but the other architectures could still
use some loving and take more advantage of the simpler interface"
* tag 'trace-5.1-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
syscalls: Remove start and number from syscall_set_arguments() args
syscalls: Remove start and number from syscall_get_arguments() args
csky: Fix syscall_get_arguments() and syscall_set_arguments()
riscv: Fix syscall_get_arguments() and syscall_set_arguments()
tracing/syscalls: Pass in hardcoded 6 into syscall_get_arguments()
ptrace: Remove maxargs from task_current_syscall()
Minor comment merge conflict in mlx5.
Staging driver has a fixup due to the skb->xmit_more changes
in 'net-next', but was removed in 'net'.
Signed-off-by: David S. Miller <davem@davemloft.net>
Handle event of power state change in the PCIE slot. When the event
occurs, check if query power state and PCI power fields is supported. If
so, read these fields from MPEIN (management PCIE info) register and
issue a corresponding message.
Signed-off-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
This merge commit includes some misc shared code updates from mlx5-next branch needed
for net-next.
1) From Maxim, Remove un-used macros and spinlock from mlx5 code.
2) From Aya, Expose Management PCIE info register layout and add rate limit
print macros.
3) From Tariq, Compilation warning fix in fs_core.c
4) From Vu, Huy and Saeed, Improve mlx5 initialization flow:
The goal is to provide a better logical separation of mlx5 core
device initialization flow and will help to seamlessly support
creating different mlx5 device types such as PF, VF and SF
mlx5 sub-function virtual devices.
Mlx5_core driver needs to separate HCA resources from pci resources.
Its initialize/load/unload will be broken into stages:
1. Initialize common data structures
2. Setup function which initializes pci resources (for PF/VF)
or some other specific resources for virtual device
3. Initialize software objects according to hardware capabilities
4. Load all mlx5_core components
It is also necessary to detach mlx5_core mdev name/message from pci
device mdev->pdev name/message for a clearer report/debug of
different mlx5 device types.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Pull networking fixes from David Miller:
1) Several hash table refcount fixes in batman-adv, from Sven
Eckelmann.
2) Use after free in bpf_evict_inode(), from Daniel Borkmann.
3) Fix mdio bus registration in ixgbe, from Ivan Vecera.
4) Unbounded loop in __skb_try_recv_datagram(), from Paolo Abeni.
5) ila rhashtable corruption fix from Herbert Xu.
6) Don't allow upper-devices to be added to vrf devices, from Sabrina
Dubroca.
7) Add qmi_wwan device ID for Olicard 600, from Bjørn Mork.
8) Don't leave skb->next poisoned in __netif_receive_skb_list_ptype,
from Alexander Lobakin.
9) Missing IDR checks in mlx5 driver, from Aditya Pakki.
10) Fix false connection termination in ktls, from Jakub Kicinski.
11) Work around some ASPM issues with r8169 by disabling rx interrupt
coalescing on certain chips. From Heiner Kallweit.
12) Properly use per-cpu qstat values on NOLOCK qdiscs, from Paolo
Abeni.
13) Fully initialize sockaddr_in structures in SCTP, from Xin Long.
14) Various BPF flow dissector fixes from Stanislav Fomichev.
15) Divide by zero in act_sample, from Davide Caratti.
16) Fix bridging multicast regression introduced by rhashtable
conversion, from Nikolay Aleksandrov.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (106 commits)
ibmvnic: Fix completion structure initialization
ipv6: sit: reset ip header pointer in ipip6_rcv
net: bridge: always clear mcast matching struct on reports and leaves
libcxgb: fix incorrect ppmax calculation
vlan: conditional inclusion of FCoE hooks to match netdevice.h and bnx2x
sch_cake: Make sure we can write the IP header before changing DSCP bits
sch_cake: Use tc_skb_protocol() helper for getting packet protocol
tcp: Ensure DCTCP reacts to losses
net/sched: act_sample: fix divide by zero in the traffic path
net: thunderx: fix NULL pointer dereference in nicvf_open/nicvf_stop
net: hns: Fix sparse: some warnings in HNS drivers
net: hns: Fix WARNING when remove HNS driver with SMMU enabled
net: hns: fix ICMP6 neighbor solicitation messages discard problem
net: hns: Fix probabilistic memory overwrite when HNS driver initialized
net: hns: Use NAPI_POLL_WEIGHT for hns driver
net: hns: fix KASAN: use-after-free in hns_nic_net_xmit_hw()
flow_dissector: rst'ify documentation
ipv6: Fix dangling pointer when ipv6 fragment
net-gro: Fix GRO flush when receiving a GSO packet.
flow_dissector: document BPF flow dissector environment
...
syzbot reports:
BUG: using __this_cpu_read() in preemptible code:
caller is dev_recursion_level include/linux/netdevice.h:3052 [inline]
__this_cpu_preempt_check+0x246/0x270 lib/smp_processor_id.c:47
dev_recursion_level include/linux/netdevice.h:3052 [inline]
ip6_skb_dst_mtu include/net/ip6_route.h:245 [inline]
I erronously downgraded a this_cpu_read to __this_cpu_read when
moving dev_recursion_level() around.
Reported-by: syzbot+51471b4aae195285a4a3@syzkaller.appspotmail.com
Fixes: 97cdcf37b5 ("net: place xmit recursion in softnet data")
Signed-off-by: Florian Westphal <fw@strlen.de>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
task_current_syscall() has a single user that passes in 6 for maxargs, which
is the maximum arguments that can be used to get system calls from
syscall_get_arguments(). Instead of passing in a number of arguments to
grab, just get 6 arguments. The args argument even specifies that it's an
array of 6 items.
This will also allow changing syscall_get_arguments() to not get a variable
number of arguments, but always grab 6.
Linus also suggested not passing in a bunch of arguments to
task_current_syscall() but to instead pass in a pointer to a structure, and
just fill the structure. struct seccomp_data has almost all the parameters
that is needed except for the stack pointer (sp). As seccomp_data is part of
uapi, and I'm afraid to change it, a new structure was created
"syscall_info", which includes seccomp_data and adds the "sp" field.
Link: http://lkml.kernel.org/r/20161107213233.466776454@goodmis.org
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: linux-fsdevel@vger.kernel.org
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
The original patch didn't consider the case that autoneg process
finishes successfully but both link partners have no mode in common.
In this case there's no link, nevertheless we may be interested in
what the link partner advertised.
Like phydev->link we set phydev->autoneg_complete in
genphy_update_link() and use the stored value in genphy_read_status().
This way we don't have to read register BMSR again.
Fixes: b6163f194c ("net: phy: improve genphy_read_status")
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Large verifier speed improvements allow to increase
verifier complexity limit.
Now regardless of the program composition and its size it takes
little time for the verifier to hit insn_processed limit.
On typical x86 machine non-debug kernel processes 1M instructions
in 1/10 of a second.
(before these speed improvements specially crafted programs
could be hitting multi-second verification times)
Full kasan kernel with debug takes ~1 second for the same 1M insns.
Hence bump the BPF_COMPLEXITY_LIMIT_INSNS limit to 1M.
Also increase the number of instructions per program
from 4k to internal BPF_COMPLEXITY_LIMIT_INSNS limit.
4k limit was confusing to users, since small programs with hundreds
of insns could be hitting BPF_COMPLEXITY_LIMIT_INSNS limit.
Sometimes adding more insns and bpf_trace_printk debug statements
would make the verifier accept the program while removing
code would make the verifier reject it.
Some user space application started to add #define MAX_FOO to
their programs and do:
MAX_FOO=100;
again:
compile with MAX_FOO;
try to load;
if (fails_to_load) { reduce MAX_FOO; goto again; }
to be able to fit maximum amount of processing into single program.
Other users artificially split their single program into a set of programs
and use all 32 iterations of tail_calls to increase compute limits.
And the most advanced folks used unlimited tc-bpf filter list
to execute many bpf programs.
Essentially the users managed to workaround 4k insn limit.
This patch removes the limit for root programs from uapi.
BPF_COMPLEXITY_LIMIT_INSNS is the kernel internal limit
and success to load the program no longer depends on program size,
but on 'smartness' of the verifier only.
The verifier will continue to get smarter with every kernel release.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Branch instructions, branch targets and calls in a bpf program are
the places where the verifier remembers states that led to successful
verification of the program.
These states are used to prune brute force program analysis.
For unprivileged programs there is a limit of 64 states per such
'branching' instructions (maximum length is tracked by max_states_per_insn
counter introduced in the previous patch).
Simply reducing this threshold to 32 or lower increases insn_processed
metric to the point that small valid programs get rejected.
For root programs there is no limit and cilium programs can have
max_states_per_insn to be 100 or higher.
Walking 100+ states multiplied by number of 'branching' insns during
verification consumes significant amount of cpu time.
Turned out simple LRU-like mechanism can be used to remove states
that unlikely will be helpful in future search pruning.
This patch introduces hit_cnt and miss_cnt counters:
hit_cnt - this many times this state successfully pruned the search
miss_cnt - this many times this state was not equivalent to other states
(and that other states were added to state list)
The heuristic introduced in this patch is:
if (sl->miss_cnt > sl->hit_cnt * 3 + 3)
/* drop this state from future considerations */
Higher numbers increase max_states_per_insn (allow more states to be
considered for pruning) and slow verification speed, but do not meaningfully
reduce insn_processed metric.
Lower numbers drop too many states and insn_processed increases too much.
Many different formulas were considered.
This one is simple and works well enough in practice.
(the analysis was done on selftests/progs/* and on cilium programs)
The end result is this heuristic improves verification speed by 10 times.
Large synthetic programs that used to take a second more now take
1/10 of a second.
In cases where max_states_per_insn used to be 100 or more, now it's ~10.
There is a slight increase in insn_processed for cilium progs:
before after
bpf_lb-DLB_L3.o 1831 1838
bpf_lb-DLB_L4.o 3029 3218
bpf_lb-DUNKNOWN.o 1064 1064
bpf_lxc-DDROP_ALL.o 26309 26935
bpf_lxc-DUNKNOWN.o 33517 34439
bpf_netdev.o 9713 9721
bpf_overlay.o 6184 6184
bpf_lcx_jit.o 37335 39389
And 2-3 times improvement in the verification speed.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
In order to understand the verifier bottlenecks add various stats
and extend log_level:
log_level 1 and 2 are kept as-is:
bit 0 - level=1 - print every insn and verifier state at branch points
bit 1 - level=2 - print every insn and verifier state at every insn
bit 2 - level=4 - print verifier error and stats at the end of verification
When verifier rejects the program the libbpf is trying to load the program twice.
Once with log_level=0 (no messages, only error code is reported to user space)
and second time with log_level=1 to tell the user why the verifier rejected it.
With introduction of bit 2 - level=4 the libbpf can choose to always use that
level and load programs once, since the verification speed is not affected and
in case of error the verbose message will be available.
Note that the verifier stats are not part of uapi just like all other
verbose messages. They're expected to change in the future.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Similar to genphy_c45_pma_read_abilities() add a function to dynamically
detect the abilities of a Clause 22 PHY. This is mainly copied from
genphy_config_init().
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Expose PRM layout for handling MPEIN (Management PCIE Info). It will be
used in the downstream patch for querying MPEIN via the driver.
Signed-off-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Add bar_addr field to store bar-0 address to avoid calling
pci_resource_start with hard-coded bar-0 as parameter.
Also note that different mlx5 device types will have bar_addr
on different bars.
This patch does not change any functionality.
Signed-off-by: Huy Nguyen <huyn@mellanox.com>
Signed-off-by: Vu Pham <vuhuong@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Software structure initialization should be in mdev_init stage.
This provides a better logical separation of mlx5 core device
initialization flow and will help to seamlessly support creating different
mlx5 device types such as PF, VF and SF mlx5 sub-function virtual device.
This patch does not change any functionality.
Signed-off-by: Vu Pham <vuhuong@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
As there is no user of mlx5_write64 that passes a spinlock to
mlx5_write64, remove this functionality and simplify the function.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
MLX5_*_DOORBELL_LOCK macros provided a way to avoid locking for
mlx5_write64 on 64-bit platforms where it's not necessary. Currently all
calls to mlx5_write64 don't use a spinlock, so the macros became unused.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
This is the last driver using always-0 skb->xmit_more.
Switch it to netdev_xmit_more and remove the now unused xmit_more flag
from sk_buff.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
