commit 50b045a8c0ccf44f76640ac3eea8d80ca53979a3 upstream.

One of the biggest issues we face right now with picking LRU map over
regular hash table is that a map walk out of user space, for example,
to just dump the existing entries or to remove certain ones, will
completely mess up LRU eviction heuristics and wrong entries such
as just created ones will get evicted instead. The reason for this
is that we mark an entry as "in use" via bpf_lru_node_set_ref() from
system call lookup side as well. Thus upon walk, all entries are
being marked, so information of actual least recently used ones
are "lost".

In case of Cilium where it can be used (besides others) as a BPF
based connection tracker, this current behavior causes disruption
upon control plane changes that need to walk the map from user space
to evict certain entries. Discussion result from bpfconf [0] was that
we should simply just remove marking from system call side as no
good use case could be found where it's actually needed there.
Therefore this patch removes marking for regular LRU and per-CPU
flavor. If there ever should be a need in future, the behavior could
be selected via map creation flag, but due to mentioned reason we
avoid this here.

  [0] http://vger.kernel.org/bpfconf.html

Fixes: 29ba732a ("bpf: Add BPF_MAP_TYPE_LRU_HASH")
Fixes: 8f844938 ("bpf: Add BPF_MAP_TYPE_LRU_PERCPU_HASH")
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NMartin KaFai Lau <kafai@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

mainline inclusion
from mainline-5.0
commit a89fac57b5d0
category: bugfix
bugzilla: 9352
CVE: NA

-------------------------------------------------

Lockdep warns about false positive:
[   12.492084] 00000000e6b28347 (&head->lock){+...}, at: pcpu_freelist_push+0x2a/0x40
[   12.492696] but this lock was taken by another, HARDIRQ-safe lock in the past:
[   12.493275]  (&rq->lock){-.-.}
[   12.493276]
[   12.493276]
[   12.493276] and interrupts could create inverse lock ordering between them.
[   12.493276]
[   12.494435]
[   12.494435] other info that might help us debug this:
[   12.494979]  Possible interrupt unsafe locking scenario:
[   12.494979]
[   12.495518]        CPU0                    CPU1
[   12.495879]        ----                    ----
[   12.496243]   lock(&head->lock);
[   12.496502]                                local_irq_disable();
[   12.496969]                                lock(&rq->lock);
[   12.497431]                                lock(&head->lock);
[   12.497890]   <Interrupt>
[   12.498104]     lock(&rq->lock);
[   12.498368]
[   12.498368]  *** DEADLOCK ***
[   12.498368]
[   12.498837] 1 lock held by dd/276:
[   12.499110]  #0: 00000000c58cb2ee (rcu_read_lock){....}, at: trace_call_bpf+0x5e/0x240
[   12.499747]
[   12.499747] the shortest dependencies between 2nd lock and 1st lock:
[   12.500389]  -> (&rq->lock){-.-.} {
[   12.500669]     IN-HARDIRQ-W at:
[   12.500934]                       _raw_spin_lock+0x2f/0x40
[   12.501373]                       scheduler_tick+0x4c/0xf0
[   12.501812]                       update_process_times+0x40/0x50
[   12.502294]                       tick_periodic+0x27/0xb0
[   12.502723]                       tick_handle_periodic+0x1f/0x60
[   12.503203]                       timer_interrupt+0x11/0x20
[   12.503651]                       __handle_irq_event_percpu+0x43/0x2c0
[   12.504167]                       handle_irq_event_percpu+0x20/0x50
[   12.504674]                       handle_irq_event+0x37/0x60
[   12.505139]                       handle_level_irq+0xa7/0x120
[   12.505601]                       handle_irq+0xa1/0x150
[   12.506018]                       do_IRQ+0x77/0x140
[   12.506411]                       ret_from_intr+0x0/0x1d
[   12.506834]                       _raw_spin_unlock_irqrestore+0x53/0x60
[   12.507362]                       __setup_irq+0x481/0x730
[   12.507789]                       setup_irq+0x49/0x80
[   12.508195]                       hpet_time_init+0x21/0x32
[   12.508644]                       x86_late_time_init+0xb/0x16
[   12.509106]                       start_kernel+0x390/0x42a
[   12.509554]                       secondary_startup_64+0xa4/0xb0
[   12.510034]     IN-SOFTIRQ-W at:
[   12.510305]                       _raw_spin_lock+0x2f/0x40
[   12.510772]                       try_to_wake_up+0x1c7/0x4e0
[   12.511220]                       swake_up_locked+0x20/0x40
[   12.511657]                       swake_up_one+0x1a/0x30
[   12.512070]                       rcu_process_callbacks+0xc5/0x650
[   12.512553]                       __do_softirq+0xe6/0x47b
[   12.512978]                       irq_exit+0xc3/0xd0
[   12.513372]                       smp_apic_timer_interrupt+0xa9/0x250
[   12.513876]                       apic_timer_interrupt+0xf/0x20
[   12.514343]                       default_idle+0x1c/0x170
[   12.514765]                       do_idle+0x199/0x240
[   12.515159]                       cpu_startup_entry+0x19/0x20
[   12.515614]                       start_kernel+0x422/0x42a
[   12.516045]                       secondary_startup_64+0xa4/0xb0
[   12.516521]     INITIAL USE at:
[   12.516774]                      _raw_spin_lock_irqsave+0x38/0x50
[   12.517258]                      rq_attach_root+0x16/0xd0
[   12.517685]                      sched_init+0x2f2/0x3eb
[   12.518096]                      start_kernel+0x1fb/0x42a
[   12.518525]                      secondary_startup_64+0xa4/0xb0
[   12.518986]   }
[   12.519132]   ... key      at: [<ffffffff82b7bc28>] __key.71384+0x0/0x8
[   12.519649]   ... acquired at:
[   12.519892]    pcpu_freelist_pop+0x7b/0xd0
[   12.520221]    bpf_get_stackid+0x1d2/0x4d0
[   12.520563]    ___bpf_prog_run+0x8b4/0x11a0
[   12.520887]
[   12.521008] -> (&head->lock){+...} {
[   12.521292]    HARDIRQ-ON-W at:
[   12.521539]                     _raw_spin_lock+0x2f/0x40
[   12.521950]                     pcpu_freelist_push+0x2a/0x40
[   12.522396]                     bpf_get_stackid+0x494/0x4d0
[   12.522828]                     ___bpf_prog_run+0x8b4/0x11a0
[   12.523296]    INITIAL USE at:
[   12.523537]                    _raw_spin_lock+0x2f/0x40
[   12.523944]                    pcpu_freelist_populate+0xc0/0x120
[   12.524417]                    htab_map_alloc+0x405/0x500
[   12.524835]                    __do_sys_bpf+0x1a3/0x1a90
[   12.525253]                    do_syscall_64+0x4a/0x180
[   12.525659]                    entry_SYSCALL_64_after_hwframe+0x49/0xbe
[   12.526167]  }
[   12.526311]  ... key      at: [<ffffffff838f7668>] __key.13130+0x0/0x8
[   12.526812]  ... acquired at:
[   12.527047]    __lock_acquire+0x521/0x1350
[   12.527371]    lock_acquire+0x98/0x190
[   12.527680]    _raw_spin_lock+0x2f/0x40
[   12.527994]    pcpu_freelist_push+0x2a/0x40
[   12.528325]    bpf_get_stackid+0x494/0x4d0
[   12.528645]    ___bpf_prog_run+0x8b4/0x11a0
[   12.528970]
[   12.529092]
[   12.529092] stack backtrace:
[   12.529444] CPU: 0 PID: 276 Comm: dd Not tainted 5.0.0-rc3-00018-g2fa53f892422 #475
[   12.530043] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.el7 04/01/2014
[   12.530750] Call Trace:
[   12.530948]  dump_stack+0x5f/0x8b
[   12.531248]  check_usage_backwards+0x10c/0x120
[   12.531598]  ? ___bpf_prog_run+0x8b4/0x11a0
[   12.531935]  ? mark_lock+0x382/0x560
[   12.532229]  mark_lock+0x382/0x560
[   12.532496]  ? print_shortest_lock_dependencies+0x180/0x180
[   12.532928]  __lock_acquire+0x521/0x1350
[   12.533271]  ? find_get_entry+0x17f/0x2e0
[   12.533586]  ? find_get_entry+0x19c/0x2e0
[   12.533902]  ? lock_acquire+0x98/0x190
[   12.534196]  lock_acquire+0x98/0x190
[   12.534482]  ? pcpu_freelist_push+0x2a/0x40
[   12.534810]  _raw_spin_lock+0x2f/0x40
[   12.535099]  ? pcpu_freelist_push+0x2a/0x40
[   12.535432]  pcpu_freelist_push+0x2a/0x40
[   12.535750]  bpf_get_stackid+0x494/0x4d0
[   12.536062]  ___bpf_prog_run+0x8b4/0x11a0

It has been explained that is a false positive here:
https://lkml.org/lkml/2018/7/25/756
Recap:
- stackmap uses pcpu_freelist
- The lock in pcpu_freelist is a percpu lock
- stackmap is only used by tracing bpf_prog
- A tracing bpf_prog cannot be run if another bpf_prog
  has already been running (ensured by the percpu bpf_prog_active counter).

Eric pointed out that this lockdep splats stops other
legit lockdep splats in selftests/bpf/test_progs.c.

Fix this by calling local_irq_save/restore for stackmap.

Another false positive had also been worked around by calling
local_irq_save in commit 89ad2fa3 ("bpf: fix lockdep splat").
That commit added unnecessary irq_save/restore to fast path of
bpf hash map. irqs are already disabled at that point, since htab
is holding per bucket spin_lock with irqsave.

Let's reduce overhead for htab by introducing __pcpu_freelist_push/pop
function w/o irqsave and convert pcpu_freelist_push/pop to irqsave
to be used elsewhere (right now only in stackmap).
It stops lockdep false positive in stackmap with a bit of acceptable overhead.

Fixes: 557c0c6e ("bpf: convert stackmap to pre-allocation")
Reported-by: NNaresh Kamboju <naresh.kamboju@linaro.org>
Reported-by: NEric Dumazet <eric.dumazet@gmail.com>
Acked-by: NMartin KaFai Lau <kafai@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NCheng Jian <cj.chengjian@huawei.com>
Reviewed-by: NHanjun Guo <guohanjun@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

All BPF hash and LRU maps currently have a known and global seed
we feed into jhash() which is 0. This is suboptimal, thus fix it
by generating a random seed upon hashtab setup time which we can
later on feed into jhash() on lookup, update and deletions.

Fixes: 0f8e4bd8 ("bpf: add hashtable type of eBPF maps")
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NSong Liu <songliubraving@fb.com>
Reviewed-by: NEduardo Valentin <eduval@amazon.com>

Commit a26ca7c9 ("bpf: btf: Add pretty print support to
the basic arraymap") and 699c86d6 ("bpf: btf: add pretty
print for hash/lru_hash maps") enabled support for BTF and
dumping via BPF fs for array and hash/lru map. However, both
can be decoupled from each other such that regular BPF maps
can be supported for attaching BTF key/value information,
while not all maps necessarily need to dump via map_seq_show_elem()
callback.

The basic sanity check which is a prerequisite for all maps
is that key/value size has to match in any case, and some maps
can have extra checks via map_check_btf() callback, e.g.
probing certain types or indicating no support in general. With
that we can also enable retrieving BTF info for per-cpu map
types and lpm.
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NYonghong Song <yhs@fb.com>

Commit a26ca7c9 ("bpf: btf: Add pretty print support to
the basic arraymap") added pretty print support to array map.
This patch adds pretty print for hash and lru_hash maps.
The following example shows the pretty-print result of
a pinned hashmap:

    struct map_value {
            int count_a;
            int count_b;
    };

    cat /sys/fs/bpf/pinned_hash_map:

    87907: {87907,87908}
    57354: {37354,57355}
    76625: {76625,76626}
    ...
Signed-off-by: NYonghong Song <yhs@fb.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

Decrement the number of elements in the map in case the allocation
of a new node fails.

Fixes: 6c905981 ("bpf: pre-allocate hash map elements")
Signed-off-by: NMauricio Vasquez B <mauricio.vasquez@polito.it>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

While some of the BPF map lookup helpers provide a ->map_gen_lookup()
callback for inlining the map lookup altogether it is not available
for every map, so the remaining ones have to call bpf_map_lookup_elem()
helper which does a dispatch to map->ops->map_lookup_elem(). In
times of retpolines, this will control and trap speculative execution
rather than letting it do its work for the indirect call and will
therefore cause a slowdown. Likewise, bpf_map_update_elem() and
bpf_map_delete_elem() do not have an inlined version and need to call
into their map->ops->map_update_elem() resp. map->ops->map_delete_elem()
handlers.

Before:

  # bpftool prog dump xlated id 1
    0: (bf) r2 = r10
    1: (07) r2 += -8
    2: (7a) *(u64 *)(r2 +0) = 0
    3: (18) r1 = map[id:1]
    5: (85) call __htab_map_lookup_elem#232656
    6: (15) if r0 == 0x0 goto pc+4
    7: (71) r1 = *(u8 *)(r0 +35)
    8: (55) if r1 != 0x0 goto pc+1
    9: (72) *(u8 *)(r0 +35) = 1
   10: (07) r0 += 56
   11: (15) if r0 == 0x0 goto pc+4
   12: (bf) r2 = r0
   13: (18) r1 = map[id:1]
   15: (85) call bpf_map_delete_elem#215008  <-- indirect call via
   16: (95) exit                                 helper

After:

  # bpftool prog dump xlated id 1
    0: (bf) r2 = r10
    1: (07) r2 += -8
    2: (7a) *(u64 *)(r2 +0) = 0
    3: (18) r1 = map[id:1]
    5: (85) call __htab_map_lookup_elem#233328
    6: (15) if r0 == 0x0 goto pc+4
    7: (71) r1 = *(u8 *)(r0 +35)
    8: (55) if r1 != 0x0 goto pc+1
    9: (72) *(u8 *)(r0 +35) = 1
   10: (07) r0 += 56
   11: (15) if r0 == 0x0 goto pc+4
   12: (bf) r2 = r0
   13: (18) r1 = map[id:1]
   15: (85) call htab_lru_map_delete_elem#238240  <-- direct call
   16: (95) exit

In all three lookup/update/delete cases however we can use the actual
address of the map callback directly if we find that there's only a
single path with a map pointer leading to the helper call, meaning
when the map pointer has not been poisoned from verifier side.
Example code can be seen above for the delete case.
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NSong Liu <songliubraving@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>

All map types reimplement the field-by-field copy of union bpf_attr
members into struct bpf_map.  Add a helper to perform this operation.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NQuentin Monnet <quentin.monnet@netronome.com>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

Use the new callback to perform allocation checks for hash maps.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NQuentin Monnet <quentin.monnet@netronome.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

Number of attribute checks are currently performed after hashtab
is already allocated.  Move them to be able to split them out to
the check function later on.  Checks have to now be performed on
the attr union directly instead of the members of bpf_map, since
bpf_map will be allocated later.  No functional changes.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NQuentin Monnet <quentin.monnet@netronome.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

While using large percpu maps, htab_map_alloc() can hold
cpu for hundreds of ms.

This patch adds cond_resched() calls to percpu alloc/free
call sites, all running in process context.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>

Introduce the map read/write flags to the eBPF syscalls that returns the
map fd. The flags is used to set up the file mode when construct a new
file descriptor for bpf maps. To not break the backward capability, the
f_flags is set to O_RDWR if the flag passed by syscall is 0. Otherwise
it should be O_RDONLY or O_WRONLY. When the userspace want to modify or
read the map content, it will check the file mode to see if it is
allowed to make the change.
Signed-off-by: NChenbo Feng <fengc@google.com>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

PCPU_MIN_UNIT_SIZE is an implementation detail of the percpu
allocator. Given we support __GFP_NOWARN now, lets just let
the allocation request fail naturally instead. The two call
sites from BPF mistakenly assumed __GFP_NOWARN would work, so
no changes needed to their actual __alloc_percpu_gfp() calls
which use the flag already.
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NJohn Fastabend <john.fastabend@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch writes 'node->ref = 1' only if node->ref is 0.
The number of lookups/s for a ~1M entries LRU map increased by
~30% (260097 to 343313).

Other writes on 'node->ref = 0' is not changed.  In those cases, the
same cache line has to be changed anyway.

First column: Size of the LRU hash
Second column: Number of lookups/s

Before:
> echo "$((2**20+1)): $(./map_perf_test 1024 1 $((2**20+1)) 10000000 | awk '{print $3}')"
1048577: 260097

After:
> echo "$((2**20+1)): $(./map_perf_test 1024 1 $((2**20+1)) 10000000 | awk '{print $3}')"
1048577: 343313
Signed-off-by: NMartin KaFai Lau <kafai@fb.com>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Inline the lru map lookup to save the cost in making calls to
bpf_map_lookup_elem() and htab_lru_map_lookup_elem().

Different LRU hash size is tested.  The benefit diminishes when
the cache miss starts to dominate in the bigger LRU hash.
Considering the change is simple, it is still worth to optimize.

First column: Size of the LRU hash
Second column: Number of lookups/s

Before:
> for i in $(seq 9 20); do echo "$((2**i+1)): $(./map_perf_test 1024 1 $((2**i+1)) 10000000 | awk '{print $3}')"; done
513: 1132020
1025: 1056826
2049: 1007024
4097: 853298
8193: 742723
16385: 712600
32769: 688142
65537: 677028
131073: 619437
262145: 498770
524289: 316695
1048577: 260038

After:
> for i in $(seq 9 20); do echo "$((2**i+1)): $(./map_perf_test 1024 1 $((2**i+1)) 10000000 | awk '{print $3}')"; done
513: 1221851
1025: 1144695
2049: 1049902
4097: 884460
8193: 773731
16385: 729673
32769: 721989
65537: 715530
131073: 671665
262145: 516987
524289: 321125
1048577: 260048
Signed-off-by: NMartin KaFai Lau <kafai@fb.com>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, iproute2's BPF ELF loader works fine with array of maps
when retrieving the fd from a pinned node and doing a selfcheck
against the provided map attributes from the object file, but we
fail to do the same for hash of maps and thus refuse to get the
map from pinned node.

Reason is that when allocating hash of maps, fd_htab_map_alloc() will
set the value size to sizeof(void *), and any user space map creation
requests are forced to set 4 bytes as value size. Thus, selfcheck
will complain about exposed 8 bytes on 64 bit archs vs. 4 bytes from
object file as value size. Contract is that fdinfo or BPF_MAP_GET_FD_BY_ID
returns the value size used to create the map.

Fix it by handling it the same way as we do for array of maps, which
means that we leave value size at 4 bytes and in the allocation phase
round up value size to 8 bytes. alloc_htab_elem() needs an adjustment
in order to copy rounded up 8 bytes due to bpf_fd_htab_map_update_elem()
calling into htab_map_update_elem() with the pointer of the map
pointer as value. Unlike array of maps where we just xchg(), we're
using the generic htab_map_update_elem() callback also used from helper
calls, which published the key/value already on return, so we need
to ensure to memcpy() the right size.

Fixes: bcc6b1b7 ("bpf: Add hash of maps support")
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NMartin KaFai Lau <kafai@fb.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, iproute2's BPF ELF loader works fine with array of maps
when retrieving the fd from a pinned node and doing a selfcheck
against the provided map attributes from the object file, but we
fail to do the same for hash of maps and thus refuse to get the
map from pinned node.

Reason is that when allocating hash of maps, fd_htab_map_alloc() will
set the value size to sizeof(void *), and any user space map creation
requests are forced to set 4 bytes as value size. Thus, selfcheck
will complain about exposed 8 bytes on 64 bit archs vs. 4 bytes from
object file as value size. Contract is that fdinfo or BPF_MAP_GET_FD_BY_ID
returns the value size used to create the map.

Fix it by handling it the same way as we do for array of maps, which
means that we leave value size at 4 bytes and in the allocation phase
round up value size to 8 bytes. alloc_htab_elem() needs an adjustment
in order to copy rounded up 8 bytes due to bpf_fd_htab_map_update_elem()
calling into htab_map_update_elem() with the pointer of the map
pointer as value. Unlike array of maps where we just xchg(), we're
using the generic htab_map_update_elem() callback also used from helper
calls, which published the key/value already on return, so we need
to ensure to memcpy() the right size.

Fixes: bcc6b1b7 ("bpf: Add hash of maps support")
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NMartin KaFai Lau <kafai@fb.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Avoid two successive functions calls for the map in map lookup, first
is the bpf_map_lookup_elem() helper call, and second the callback via
map->ops->map_lookup_elem() to get to the map in map implementation.
Implementation inlines array and htab flavor for map in map lookups.
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The current map creation API does not allow to provide the numa-node
preference.  The memory usually comes from where the map-creation-process
is running.  The performance is not ideal if the bpf_prog is known to
always run in a numa node different from the map-creation-process.

One of the use case is sharding on CPU to different LRU maps (i.e.
an array of LRU maps).  Here is the test result of map_perf_test on
the INNER_LRU_HASH_PREALLOC test if we force the lru map used by
CPU0 to be allocated from a remote numa node:

[ The machine has 20 cores. CPU0-9 at node 0. CPU10-19 at node 1 ]

># taskset -c 10 ./map_perf_test 512 8 1260000 8000000
5:inner_lru_hash_map_perf pre-alloc 1628380 events per sec
4:inner_lru_hash_map_perf pre-alloc 1626396 events per sec
3:inner_lru_hash_map_perf pre-alloc 1626144 events per sec
6:inner_lru_hash_map_perf pre-alloc 1621657 events per sec
2:inner_lru_hash_map_perf pre-alloc 1621534 events per sec
1:inner_lru_hash_map_perf pre-alloc 1620292 events per sec
7:inner_lru_hash_map_perf pre-alloc 1613305 events per sec
0:inner_lru_hash_map_perf pre-alloc 1239150 events per sec  #<<<

After specifying numa node:
># taskset -c 10 ./map_perf_test 512 8 1260000 8000000
5:inner_lru_hash_map_perf pre-alloc 1629627 events per sec
3:inner_lru_hash_map_perf pre-alloc 1628057 events per sec
1:inner_lru_hash_map_perf pre-alloc 1623054 events per sec
6:inner_lru_hash_map_perf pre-alloc 1616033 events per sec
2:inner_lru_hash_map_perf pre-alloc 1614630 events per sec
4:inner_lru_hash_map_perf pre-alloc 1612651 events per sec
7:inner_lru_hash_map_perf pre-alloc 1609337 events per sec
0:inner_lru_hash_map_perf pre-alloc 1619340 events per sec #<<<

This patch adds one field, numa_node, to the bpf_attr.  Since numa node 0
is a valid node, a new flag BPF_F_NUMA_NODE is also added.  The numa_node
field is honored if and only if the BPF_F_NUMA_NODE flag is set.

Numa node selection is not supported for percpu map.

This patch does not change all the kmalloc.  F.e.
'htab = kzalloc()' is not changed since the object
is small enough to stay in the cache.
Signed-off-by: NMartin KaFai Lau <kafai@fb.com>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@fb.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch allows userspace to do BPF_MAP_LOOKUP_ELEM on
BPF_MAP_TYPE_PROG_ARRAY,
BPF_MAP_TYPE_ARRAY_OF_MAPS and
BPF_MAP_TYPE_HASH_OF_MAPS.

The lookup returns a prog-id or map-id to the userspace.
The userspace can then use the BPF_PROG_GET_FD_BY_ID
or BPF_MAP_GET_FD_BY_ID to get a fd.
Signed-off-by: NMartin KaFai Lau <kafai@fb.com>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When iterating through a map, we need to find a key that does not exist
in the map so map_get_next_key will give us the first key of the map.
This often requires a lot of guessing in production systems.

This patch makes map_get_next_key return the first key when the key
pointer in the parameter is NULL.
Signed-off-by: NTeng Qin <qinteng@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There's no need to have struct bpf_map_type_list since
it just contains a list_head, the type, and the ops
pointer. Since the types are densely packed and not
actually dynamically registered, it's much easier and
smaller to have an array of type->ops pointer. Also
initialize this array statically to remove code needed
to initialize it.

In order to save duplicating the list, move it to the
types header file added by the previous patch and
include it in the same fashion.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds hash of maps support (hashmap->bpf_map).
BPF_MAP_TYPE_HASH_OF_MAPS is added.

A map-in-map contains a pointer to another map and lets call
this pointer 'inner_map_ptr'.

Notes on deleting inner_map_ptr from a hash map:

1. For BPF_F_NO_PREALLOC map-in-map, when deleting
   an inner_map_ptr, the htab_elem itself will go through
   a rcu grace period and the inner_map_ptr resides
   in the htab_elem.

2. For pre-allocated htab_elem (!BPF_F_NO_PREALLOC),
   when deleting an inner_map_ptr, the htab_elem may
   get reused immediately.  This situation is similar
   to the existing prealloc-ated use cases.

   However, the bpf_map_fd_put_ptr() calls bpf_map_put() which calls
   inner_map->ops->map_free(inner_map) which will go
   through a rcu grace period (i.e. all bpf_map's map_free
   currently goes through a rcu grace period).  Hence,
   the inner_map_ptr is still safe for the rcu reader side.

This patch also includes BPF_MAP_TYPE_HASH_OF_MAPS to the
check_map_prealloc() in the verifier.  preallocation is a
must for BPF_PROG_TYPE_PERF_EVENT.  Hence, even we don't expect
heavy updates to map-in-map, enforcing BPF_F_NO_PREALLOC for map-in-map
is impossible without disallowing BPF_PROG_TYPE_PERF_EVENT from using
map-in-map first.
Signed-off-by: NMartin KaFai Lau <kafai@fb.com>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In both kmalloc and prealloc mode the bpf_map_update_elem() is using
per-cpu extra_elems to do atomic update when the map is full.
There are two issues with it. The logic can be misused, since it allows
max_entries+num_cpus elements to be present in the map. And alloc_extra_elems()
at map creation time can fail percpu alloc for large map values with a warn:
WARNING: CPU: 3 PID: 2752 at ../mm/percpu.c:892 pcpu_alloc+0x119/0xa60
illegal size (32824) or align (8) for percpu allocation

The fixes for both of these issues are different for kmalloc and prealloc modes.
For prealloc mode allocate extra num_possible_cpus elements and store
their pointers into extra_elems array instead of actual elements.
Hence we can use these hidden(spare) elements not only when the map is full
but during bpf_map_update_elem() that replaces existing element too.
That also improves performance, since pcpu_freelist_pop/push is avoided.
Unfortunately this approach cannot be used for kmalloc mode which needs
to kfree elements after rcu grace period. Therefore switch it back to normal
kmalloc even when full and old element exists like it was prior to
commit 6c905981 ("bpf: pre-allocate hash map elements").

Add tests to check for over max_entries and large map values.
Reported-by: NDave Jones <davej@codemonkey.org.uk>
Fixes: 6c905981 ("bpf: pre-allocate hash map elements")
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NMartin KaFai Lau <kafai@fb.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Optimize:
bpf_call
  bpf_map_lookup_elem
    map->ops->map_lookup_elem
      htab_map_lookup_elem
        __htab_map_lookup_elem
into:
bpf_call
  __htab_map_lookup_elem

to improve performance of JITed programs.
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

when all map elements are pre-allocated one cpu can delete and reuse htab_elem
while another cpu is still walking the hlist. In such case the lookup may
miss the element. Convert hlist to hlist_nulls to avoid such scenario.
When bucket lock is taken there is no need to take such precautions,
so only convert map_lookup and map_get_next to nulls.
The race window is extremely small and only reproducible with explicit
udelay() inside lookup_nulls_elem_raw()

Similar to hlist add hlist_nulls_for_each_entry_safe() and
hlist_nulls_entry_safe() helpers.

Fixes: 6c905981 ("bpf: pre-allocate hash map elements")
Reported-by: NJonathan Perry <jonperry@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

when htab_elem is removed from the bucket list the htab_elem.hash_node.next
field should not be overridden too early otherwise we have a tiny race window
between lookup and delete.
The bug was discovered by manual code analysis and reproducible
only with explicit udelay() in lookup_elem_raw().

Fixes: 6c905981 ("bpf: pre-allocate hash map elements")
Reported-by: NJonathan Perry <jonperry@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

All map types and prog types are registered to the BPF core through
bpf_register_map_type() and bpf_register_prog_type() during init and
remain unchanged thereafter. As by design we don't (and never will)
have any pluggable code that can register to that at any later point
in time, lets mark all the existing bpf_{map,prog}_type_list objects
in the tree as __ro_after_init, so they can be moved to read-only
section from then onwards.
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds two helpers, bpf_map_area_alloc() and bpf_map_area_free(),
that are to be used for map allocations. Using kmalloc() for very large
allocations can cause excessive work within the page allocator, so i) fall
back earlier to vmalloc() when the attempt is considered costly anyway,
and even more importantly ii) don't trigger OOM killer with any of the
allocators.

Since this is based on a user space request, for example, when creating
maps with element pre-allocation, we really want such requests to fail
instead of killing other user space processes.

Also, don't spam the kernel log with warnings should any of the allocations
fail under pressure. Given that, we can make backend selection in
bpf_map_area_alloc() generic, and convert all maps over to use this API
for spots with potentially large allocation requests.

Note, replacing the one kmalloc_array() is fine as overflow checks happen
earlier in htab_map_alloc(), since it must also protect the multiplication
for vmalloc() should kmalloc_array() fail.
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit 01b3f521 ("bpf: fix allocation warnings in bpf maps and
integer overflow") has added checks for the maximum allocateable size.
It (ab)used KMALLOC_SHIFT_MAX for that purpose.

While this is not incorrect it is not very clean because we already have
KMALLOC_MAX_SIZE for this very reason so let's change both checks to use
KMALLOC_MAX_SIZE instead.

The original motivation for using KMALLOC_SHIFT_MAX was to work around
an incorrect KMALLOC_MAX_SIZE which could lead to allocation warnings
but it is no longer needed since "slab: make sure that KMALLOC_MAX_SIZE
will fit into MAX_ORDER".

Link: http://lkml.kernel.org/r/20161220130659.16461-3-mhocko@kernel.orgSigned-off-by: NMichal Hocko <mhocko@suse.com>
Acked-by: NChristoph Lameter <cl@linux.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Andrey Konovalov <andreyknvl@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Provide a LRU version of the existing BPF_MAP_TYPE_PERCPU_HASH
Signed-off-by: NMartin KaFai Lau <kafai@fb.com>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Provide a LRU version of the existing BPF_MAP_TYPE_HASH.
Signed-off-by: NMartin KaFai Lau <kafai@fb.com>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Refactor the codes that populate the value
of a htab_elem in a BPF_MAP_TYPE_PERCPU_HASH
typed bpf_map.
Signed-off-by: NMartin KaFai Lau <kafai@fb.com>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit a6ed3ea6 ("bpf: restore behavior of bpf_map_update_elem")
added an extra per-cpu reserve to the hash table map to restore old
behaviour from pre prealloc times. When non-prealloc is in use for a
map, then problem is that once a hash table extra element has been
linked into the hash-table, and the hash table is destroyed due to
refcount dropping to zero, then htab_map_free() -> delete_all_elements()
will walk the whole hash table and drop all elements via htab_elem_free().
The problem is that the element from the extra reserve is first fed
to the wrong backend allocator and eventually freed twice.

Fixes: a6ed3ea6 ("bpf: restore behavior of bpf_map_update_elem")
Reported-by: NDmitry Vyukov <dvyukov@google.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The introduction of pre-allocated hash elements inadvertently broke
the behavior of bpf hash maps where users expected to call
bpf_map_update_elem() without considering that the map can be full.
Some programs do:
old_value = bpf_map_lookup_elem(map, key);
if (old_value) {
  ... prepare new_value on stack ...
  bpf_map_update_elem(map, key, new_value);
}
Before pre-alloc the update() for existing element would work even
in 'map full' condition. Restore this behavior.

The above program could have updated old_value in place instead of
update() which would be faster and most programs use that approach,
but sometimes the values are large and the programs use update()
helper to do atomic replacement of the element.
Note we cannot simply update element's value in-place like percpu
hash map does and have to allocate extra num_possible_cpu elements
and use this extra reserve when the map is full.

Fixes: 6c905981 ("bpf: pre-allocate hash map elements")
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If kprobe is placed on spin_unlock then calling kmalloc/kfree from
bpf programs is not safe, since the following dead lock is possible:
kfree->spin_lock(kmem_cache_node->lock)...spin_unlock->kprobe->
bpf_prog->map_update->kmalloc->spin_lock(of the same kmem_cache_node->lock)
and deadlocks.

The following solutions were considered and some implemented, but
eventually discarded
- kmem_cache_create for every map
- add recursion check to slow-path of slub
- use reserved memory in bpf_map_update for in_irq or in preempt_disabled
- kmalloc via irq_work

At the end pre-allocation of all map elements turned out to be the simplest
solution and since the user is charged upfront for all the memory, such
pre-allocation doesn't affect the user space visible behavior.

Since it's impossible to tell whether kprobe is triggered in a safe
location from kmalloc point of view, use pre-allocation by default
and introduce new BPF_F_NO_PREALLOC flag.

While testing of per-cpu hash maps it was discovered
that alloc_percpu(GFP_ATOMIC) has odd corner cases and often
fails to allocate memory even when 90% of it is free.
The pre-allocation of per-cpu hash elements solves this problem as well.

Turned out that bpf_map_update() quickly followed by
bpf_map_lookup()+bpf_map_delete() is very common pattern used
in many of iovisor/bcc/tools, so there is additional benefit of
pre-allocation, since such use cases are must faster.

Since all hash map elements are now pre-allocated we can remove
atomic increment of htab->count and save few more cycles.

Also add bpf_map_precharge_memlock() to check rlimit_memlock early to avoid
large malloc/free done by users who don't have sufficient limits.

Pre-allocation is done with vmalloc and alloc/free is done
via percpu_freelist. Here are performance numbers for different
pre-allocation algorithms that were implemented, but discarded
in favor of percpu_freelist:

1 cpu:
pcpu_ida	2.1M
pcpu_ida nolock	2.3M
bt		2.4M
kmalloc		1.8M
hlist+spinlock	2.3M
pcpu_freelist	2.6M

4 cpu:
pcpu_ida	1.5M
pcpu_ida nolock	1.8M
bt w/smp_align	1.7M
bt no/smp_align	1.1M
kmalloc		0.7M
hlist+spinlock	0.2M
pcpu_freelist	2.0M

8 cpu:
pcpu_ida	0.7M
bt w/smp_align	0.8M
kmalloc		0.4M
pcpu_freelist	1.5M

32 cpu:
kmalloc		0.13M
pcpu_freelist	0.49M

pcpu_ida nolock is a modified percpu_ida algorithm without
percpu_ida_cpu locks and without cross-cpu tag stealing.
It's faster than existing percpu_ida, but not as fast as pcpu_freelist.

bt is a variant of block/blk-mq-tag.c simlified and customized
for bpf use case. bt w/smp_align is using cache line for every 'long'
(similar to blk-mq-tag). bt no/smp_align allocates 'long'
bitmasks continuously to save memory. It's comparable to percpu_ida
and in some cases faster, but slower than percpu_freelist

hlist+spinlock is the simplest free list with single spinlock.
As expeceted it has very bad scaling in SMP.

kmalloc is existing implementation which is still available via
BPF_F_NO_PREALLOC flag. It's significantly slower in single cpu and
in 8 cpu setup it's 3 times slower than pre-allocation with pcpu_freelist,
but saves memory, so in cases where map->max_entries can be large
and number of map update/delete per second is low, it may make
sense to use it.
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

bpf_percpu_hash_update() expects rcu lock to be held and warns if it's not,
which pointed out a missing rcu read lock.

Fixes: 15a07b33 ("bpf: add lookup/update support for per-cpu hash and array maps")
Signed-off-by: NSasha Levin <sasha.levin@oracle.com>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The functions bpf_map_lookup_elem(map, key, value) and
bpf_map_update_elem(map, key, value, flags) need to get/set
values from all-cpus for per-cpu hash and array maps,
so that user space can aggregate/update them as necessary.

Example of single counter aggregation in user space:
  unsigned int nr_cpus = sysconf(_SC_NPROCESSORS_CONF);
  long values[nr_cpus];
  long value = 0;

  bpf_lookup_elem(fd, key, values);
  for (i = 0; i < nr_cpus; i++)
    value += values[i];

The user space must provide round_up(value_size, 8) * nr_cpus
array to get/set values, since kernel will use 'long' copy
of per-cpu values to try to copy good counters atomically.
It's a best-effort, since bpf programs and user space are racing
to access the same memory.
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Introduce BPF_MAP_TYPE_PERCPU_HASH map type which is used to do
accurate counters without need to use BPF_XADD instruction which turned
out to be too costly for high-performance network monitoring.
In the typical use case the 'key' is the flow tuple or other long
living object that sees a lot of events per second.

bpf_map_lookup_elem() returns per-cpu area.
Example:
struct {
  u32 packets;
  u32 bytes;
} * ptr = bpf_map_lookup_elem(&map, &key);
/* ptr points to this_cpu area of the value, so the following
 * increments will not collide with other cpus
 */
ptr->packets ++;
ptr->bytes += skb->len;

bpf_update_elem() atomically creates a new element where all per-cpu
values are zero initialized and this_cpu value is populated with
given 'value'.
Note that non-per-cpu hash map always allocates new element
and then deletes old after rcu grace period to maintain atomicity
of update. Per-cpu hash map updates element values in-place.
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Both htab_map_update_elem() and htab_map_delete_elem() can be
called from eBPF program, and they may be in kernel hot path,
so it isn't efficient to use a per-hashtable lock in this two
helpers.

The per-hashtable spinlock is used for protecting bucket's
hlist, and per-bucket lock is just enough. This patch converts
the per-hashtable lock into per-bucket spinlock, so that
contention can be decreased a lot.
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>