Recent work in f4d0525921 ("bpf: Add map_meta_equal map ops") and 134fede4ee
("bpf: Relax max_entries check for most of the inner map types") added support
for dynamic inner max elements for most map-in-map types. Exceptions were maps
like array or prog array where the map_gen_lookup() callback uses the maps'
max_entries field as a constant when emitting instructions.
We recently implemented Maglev consistent hashing into Cilium's load balancer
which uses map-in-map with an outer map being hash and inner being array holding
the Maglev backend table for each service. This has been designed this way in
order to reduce overall memory consumption given the outer hash map allows to
avoid preallocating a large, flat memory area for all services. Also, the
number of service mappings is not always known a-priori.
The use case for dynamic inner array map entries is to further reduce memory
overhead, for example, some services might just have a small number of back
ends while others could have a large number. Right now the Maglev backend table
for small and large number of backends would need to have the same inner array
map entries which adds a lot of unneeded overhead.
Dynamic inner array map entries can be realized by avoiding the inlined code
generation for their lookup. The lookup will still be efficient since it will
be calling into array_map_lookup_elem() directly and thus avoiding retpoline.
The patch adds a BPF_F_INNER_MAP flag to map creation which therefore skips
inline code generation and relaxes array_map_meta_equal() check to ignore both
maps' max_entries. This also still allows to have faster lookups for map-in-map
when BPF_F_INNER_MAP is not specified and hence dynamic max_entries not needed.
Example code generation where inner map is dynamic sized array:
# bpftool p d x i 125
int handle__sys_enter(void * ctx):
; int handle__sys_enter(void *ctx)
0: (b4) w1 = 0
; int key = 0;
1: (63) *(u32 *)(r10 -4) = r1
2: (bf) r2 = r10
;
3: (07) r2 += -4
; inner_map = bpf_map_lookup_elem(&outer_arr_dyn, &key);
4: (18) r1 = map[id:468]
6: (07) r1 += 272
7: (61) r0 = *(u32 *)(r2 +0)
8: (35) if r0 >= 0x3 goto pc+5
9: (67) r0 <<= 3
10: (0f) r0 += r1
11: (79) r0 = *(u64 *)(r0 +0)
12: (15) if r0 == 0x0 goto pc+1
13: (05) goto pc+1
14: (b7) r0 = 0
15: (b4) w6 = -1
; if (!inner_map)
16: (15) if r0 == 0x0 goto pc+6
17: (bf) r2 = r10
;
18: (07) r2 += -4
; val = bpf_map_lookup_elem(inner_map, &key);
19: (bf) r1 = r0 | No inlining but instead
20: (85) call array_map_lookup_elem#149280 | call to array_map_lookup_elem()
; return val ? *val : -1; | for inner array lookup.
21: (15) if r0 == 0x0 goto pc+1
; return val ? *val : -1;
22: (61) r6 = *(u32 *)(r0 +0)
; }
23: (bc) w0 = w6
24: (95) exit
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20201010234006.7075-4-daniel@iogearbox.net
Add an efficient ingress to ingress netns switch that can be used out of tc BPF
programs in order to redirect traffic from host ns ingress into a container
veth device ingress without having to go via CPU backlog queue [0]. For local
containers this can also be utilized and path via CPU backlog queue only needs
to be taken once, not twice. On a high level this borrows from ipvlan which does
similar switch in __netif_receive_skb_core() and then iterates via another_round.
This helps to reduce latency for mentioned use cases.
Pod to remote pod with redirect(), TCP_RR [1]:
# percpu_netperf 10.217.1.33
RT_LATENCY: 122.450 (per CPU: 122.666 122.401 122.333 122.401 )
MEAN_LATENCY: 121.210 (per CPU: 121.100 121.260 121.320 121.160 )
STDDEV_LATENCY: 120.040 (per CPU: 119.420 119.910 125.460 115.370 )
MIN_LATENCY: 46.500 (per CPU: 47.000 47.000 47.000 45.000 )
P50_LATENCY: 118.500 (per CPU: 118.000 119.000 118.000 119.000 )
P90_LATENCY: 127.500 (per CPU: 127.000 128.000 127.000 128.000 )
P99_LATENCY: 130.750 (per CPU: 131.000 131.000 129.000 132.000 )
TRANSACTION_RATE: 32666.400 (per CPU: 8152.200 8169.842 8174.439 8169.897 )
Pod to remote pod with redirect_peer(), TCP_RR:
# percpu_netperf 10.217.1.33
RT_LATENCY: 44.449 (per CPU: 43.767 43.127 45.279 45.622 )
MEAN_LATENCY: 45.065 (per CPU: 44.030 45.530 45.190 45.510 )
STDDEV_LATENCY: 84.823 (per CPU: 66.770 97.290 84.380 90.850 )
MIN_LATENCY: 33.500 (per CPU: 33.000 33.000 34.000 34.000 )
P50_LATENCY: 43.250 (per CPU: 43.000 43.000 43.000 44.000 )
P90_LATENCY: 46.750 (per CPU: 46.000 47.000 47.000 47.000 )
P99_LATENCY: 52.750 (per CPU: 51.000 54.000 53.000 53.000 )
TRANSACTION_RATE: 90039.500 (per CPU: 22848.186 23187.089 22085.077 21919.130 )
[0] https://linuxplumbersconf.org/event/7/contributions/674/attachments/568/1002/plumbers_2020_cilium_load_balancer.pdf
[1] https://github.com/borkmann/netperf_scripts/blob/master/percpu_netperf
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20201010234006.7075-3-daniel@iogearbox.net
Follow-up to address David's feedback that we should better describe internals
of the bpf_redirect_neigh() helper.
Suggested-by: David Ahern <dsahern@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Reviewed-by: David Ahern <dsahern@gmail.com>
Link: https://lore.kernel.org/bpf/20201010234006.7075-2-daniel@iogearbox.net
The main purpose of the profiler test to check different llvm generation
patterns to make sure the verifier can load these large programs.
Note that profiler.inc.h test doesn't follow strict kernel coding style.
The code was formatted in the kernel style, but variable declarations are
kept as-is to preserve original llvm IR pattern.
profiler1.c should pass with older and newer llvm
profiler[23].c may fail on older llvm that don't have:
https://reviews.llvm.org/D85570
because llvm may do speculative code motion optimization that
will generate code like this:
// r9 is a pointer to map_value
// r7 is a scalar
17: bf 96 00 00 00 00 00 00 r6 = r9
18: 0f 76 00 00 00 00 00 00 r6 += r7
19: a5 07 01 00 01 01 00 00 if r7 < 257 goto +1
20: bf 96 00 00 00 00 00 00 r6 = r9
// r6 is used here
The verifier will reject such code with the error:
"math between map_value pointer and register with unbounded min value is not allowed"
At insn 18 the r7 is indeed unbounded. The later insn 19 checks the bounds and
the insn 20 undoes map_value addition. It is currently impossible for the
verifier to understand such speculative pointer arithmetic. Hence llvm D85570
addresses it on the compiler side.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20201009011240.48506-4-alexei.starovoitov@gmail.com
The llvm register allocator may use two different registers representing the
same virtual register. In such case the following pattern can be observed:
1047: (bf) r9 = r6
1048: (a5) if r6 < 0x1000 goto pc+1
1050: ...
1051: (a5) if r9 < 0x2 goto pc+66
1052: ...
1053: (bf) r2 = r9 /* r2 needs to have upper and lower bounds */
This is normal behavior of greedy register allocator.
The slides 137+ explain why regalloc introduces such register copy:
http://llvm.org/devmtg/2018-04/slides/Yatsina-LLVM%20Greedy%20Register%20Allocator.pdf
There is no way to tell llvm 'not to do this'.
Hence the verifier has to recognize such patterns.
In order to track this information without backtracking allocate ID
for scalars in a similar way as it's done for find_good_pkt_pointers().
When the verifier encounters r9 = r6 assignment it will assign the same ID
to both registers. Later if either register range is narrowed via conditional
jump propagate the register state into the other register.
Clear register ID in adjust_reg_min_max_vals() for any alu instruction. The
register ID is ignored for scalars in regsafe() and doesn't affect state
pruning. mark_reg_unknown() clears the ID. It's used to process call, endian
and other instructions. Hence ID is explicitly cleared only in
adjust_reg_min_max_vals() and in 32-bit mov.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20201009011240.48506-2-alexei.starovoitov@gmail.com
Add selftests validating libbpf's auto-resizing of load/store instructions
when used with CO-RE relocations. An explicit and manual approach with using
bpf_core_read() is also demonstrated and tested. Separate BPF program is
supposed to fail due to using signed integers of sizes that differ from
kernel's sizes.
To reliably simulate 32-bit BTF (i.e., the one with sizeof(long) ==
sizeof(void *) == 4), selftest generates its own custom BTF and passes it as
a replacement for real kernel BTF. This allows to test 32/64-bitness mix on
all architectures.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20201008001025.292064-5-andrii@kernel.org
Use generalized BTF parsing logic, making it possible to parse BTF both from
ELF file, as well as a raw BTF dump. This makes it easier to write custom
tests with manually generated BTFs.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20201008001025.292064-4-andrii@kernel.org
Add support for patching instructions of the following form:
- rX = *(T *)(rY + <off>);
- *(T *)(rX + <off>) = rY;
- *(T *)(rX + <off>) = <imm>, where T is one of {u8, u16, u32, u64}.
For such instructions, if the actual kernel field recorded in CO-RE relocation
has a different size than the one recorded locally (e.g., from vmlinux.h),
then libbpf will adjust T to an appropriate 1-, 2-, 4-, or 8-byte loads.
In general, such transformation is not always correct and could lead to
invalid final value being loaded or stored. But two classes of cases are
always safe:
- if both local and target (kernel) types are unsigned integers, but of
different sizes, then it's OK to adjust load/store instruction according to
the necessary memory size. Zero-extending nature of such instructions and
unsignedness make sure that the final value is always correct;
- pointer size mismatch between BPF target architecture (which is always
64-bit) and 32-bit host kernel architecture can be similarly resolved
automatically, because pointer is essentially an unsigned integer. Loading
32-bit pointer into 64-bit BPF register with zero extension will leave
correct pointer in the register.
Both cases are necessary to support CO-RE on 32-bit kernels, as `unsigned
long` in vmlinux.h generated from 32-bit kernel is 32-bit, but when compiled
with BPF program for BPF target it will be treated by compiler as 64-bit
integer. Similarly, pointers in vmlinux.h are 32-bit for kernel, but treated
as 64-bit values by compiler for BPF target. Both problems are now resolved by
libbpf for direct memory reads.
But similar transformations are useful in general when kernel fields are
"resized" from, e.g., unsigned int to unsigned long (or vice versa).
Now, similar transformations for signed integers are not safe to perform as
they will result in incorrect sign extension of the value. If such situation
is detected, libbpf will emit helpful message and will poison the instruction.
Not failing immediately means that it's possible to guard the instruction
based on kernel version (or other conditions) and make sure it's not
reachable.
If there is a need to read signed integers that change sizes between different
kernels, it's possible to use BPF_CORE_READ_BITFIELD() macro, which works both
with bitfields and non-bitfield integers of any signedness and handles
sign-extension properly. Also, bpf_core_read() with proper size and/or use of
bpf_core_field_size() relocation could allow to deal with such complicated
situations explicitly, if not so conventiently as direct memory reads.
Selftests added in a separate patch in progs/test_core_autosize.c demonstrate
both direct memory and probed use cases.
BPF_CORE_READ() is not changed and it won't deal with such situations as
automatically as direct memory reads due to the signedness integer
limitations, which are much harder to detect and control with compiler macro
magic. So it's encouraged to utilize direct memory reads as much as possible.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20201008001025.292064-3-andrii@kernel.org
Bypass CO-RE relocations step for BPF programs that are not going to be
loaded. This allows to have BPF programs compiled in and disabled dynamically
if kernel is not supposed to provide enough relocation information. In such
case, there won't be unnecessary warnings about failed relocations.
Fixes: d929758101 ("libbpf: Support disabling auto-loading BPF programs")
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20201008001025.292064-2-andrii@kernel.org
Fix a compatibility problem when the old XDP_SHARED_UMEM mode is used
together with the xsk_socket__create() call. In the old XDP_SHARED_UMEM
mode, only sharing of the same device and queue id was allowed, and
in this mode, the fill ring and completion ring were shared between
the AF_XDP sockets.
Therefore, it was perfectly fine to call the xsk_socket__create() API
for each socket and not use the new xsk_socket__create_shared() API.
This behavior was ruined by the commit introducing XDP_SHARED_UMEM
support between different devices and/or queue ids. This patch restores
the ability to use xsk_socket__create in these circumstances so that
backward compatibility is not broken.
Fixes: 2f6324a393 ("libbpf: Support shared umems between queues and devices")
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/1602070946-11154-1-git-send-email-magnus.karlsson@gmail.com
Commit 4976b718c3 ("bpf: Introduce pseudo_btf_id") switched
the order of check_subprogs() and resolve_pseudo_ldimm() in
the verifier. Now an empty prog expects to see the error "last
insn is not an the prog of a single invalid ldimm exit or jmp"
instead, because the check for subprogs comes first. It's now
pointless to validate that half of ldimm64 won't be the last
instruction.
Tested:
# ./test_verifier
Summary: 1129 PASSED, 537 SKIPPED, 0 FAILED
and the full set of bpf selftests.
Fixes: 4976b718c3 ("bpf: Introduce pseudo_btf_id")
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20201007022857.2791884-1-haoluo@google.com
bpf_program__set_attach_target(prog, fd, ...) will always fail when
fd = 0 (attach to a kernel symbol) because obj->btf_vmlinux is NULL
and there is no way to set it (at the moment btf_vmlinux is meant
to be temporary storage for use in bpf_object__load_xattr()).
Fix this by using libbpf_find_vmlinux_btf_id().
At some point we may want to opportunistically cache btf_vmlinux
so it can be reused with multiple programs.
Signed-off-by: Luigi Rizzo <lrizzo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Petar Penkov <ppenkov@google.com>
Link: https://lore.kernel.org/bpf/20201005224528.389097-1-lrizzo@google.com
This add a test to make sure that we can still pin maps with
reused map fd.
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20201006021345.3817033-4-liuhangbin@gmail.com
Say a user reuse map fd after creating a map manually and set the
pin_path, then load the object via libbpf.
In libbpf bpf_object__create_maps(), bpf_object__reuse_map() will
return 0 if there is no pinned map in map->pin_path. Then after
checking if map fd exist, we should also check if pin_path was set
and do bpf_map__pin() instead of continue the loop.
Fix it by creating map if fd not exist and continue checking pin_path
after that.
Suggested-by: Andrii Nakryiko <andrii.nakryiko@gmail.com>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20201006021345.3817033-3-liuhangbin@gmail.com
Previously we forgot to close the map fd if bpf_map_update_elem()
failed during map slot init, which will leak map fd.
Let's move map slot initialization to new function init_map_slots() to
simplify the code. And close the map fd if init slot failed.
Reported-by: Andrii Nakryiko <andrii.nakryiko@gmail.com>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20201006021345.3817033-2-liuhangbin@gmail.com
Test bpf_per_cpu_ptr() and bpf_this_cpu_ptr(). Test two paths in the
kernel. If the base pointer points to a struct, the returned reg is
of type PTR_TO_BTF_ID. Direct pointer dereference can be applied on
the returned variable. If the base pointer isn't a struct, the
returned reg is of type PTR_TO_MEM, which also supports direct pointer
dereference.
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20200929235049.2533242-7-haoluo@google.com
Add bpf_this_cpu_ptr() to help access percpu var on this cpu. This
helper always returns a valid pointer, therefore no need to check
returned value for NULL. Also note that all programs run with
preemption disabled, which means that the returned pointer is stable
during all the execution of the program.
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20200929235049.2533242-6-haoluo@google.com
Add bpf_per_cpu_ptr() to help bpf programs access percpu vars.
bpf_per_cpu_ptr() has the same semantic as per_cpu_ptr() in the kernel
except that it may return NULL. This happens when the cpu parameter is
out of range. So the caller must check the returned value.
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20200929235049.2533242-5-haoluo@google.com
Selftests for typed ksyms. Tests two types of ksyms: one is a struct,
the other is a plain int. This tests two paths in the kernel. Struct
ksyms will be converted into PTR_TO_BTF_ID by the verifier while int
typed ksyms will be converted into PTR_TO_MEM.
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20200929235049.2533242-4-haoluo@google.com
If a ksym is defined with a type, libbpf will try to find the ksym's btf
information from kernel btf. If a valid btf entry for the ksym is found,
libbpf can pass in the found btf id to the verifier, which validates the
ksym's type and value.
Typeless ksyms (i.e. those defined as 'void') will not have such btf_id,
but it has the symbol's address (read from kallsyms) and its value is
treated as a raw pointer.
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20200929235049.2533242-3-haoluo@google.com
Pseudo_btf_id is a type of ld_imm insn that associates a btf_id to a
ksym so that further dereferences on the ksym can use the BTF info
to validate accesses. Internally, when seeing a pseudo_btf_id ld insn,
the verifier reads the btf_id stored in the insn[0]'s imm field and
marks the dst_reg as PTR_TO_BTF_ID. The btf_id points to a VAR_KIND,
which is encoded in btf_vminux by pahole. If the VAR is not of a struct
type, the dst reg will be marked as PTR_TO_MEM instead of PTR_TO_BTF_ID
and the mem_size is resolved to the size of the VAR's type.
>From the VAR btf_id, the verifier can also read the address of the
ksym's corresponding kernel var from kallsyms and use that to fill
dst_reg.
Therefore, the proper functionality of pseudo_btf_id depends on (1)
kallsyms and (2) the encoding of kernel global VARs in pahole, which
should be available since pahole v1.18.
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20200929235049.2533242-2-haoluo@google.com
This patch adds a test to ensure the child sk inherited everything
from the bpf_sock_ops_cb_flags of the listen sk:
1. Sets one more cb_flags (BPF_SOCK_OPS_STATE_CB_FLAG) to the listen sk
in test_tcp_hdr_options.c
2. Saves the skops->bpf_sock_ops_cb_flags when handling the newly
established passive connection
3. CHECK() it is the same as the listen sk
This also covers the fastopen case as the existing test_tcp_hdr_options.c
does.
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20201002013454.2542367-1-kafai@fb.com
When using -Werror=missing-braces, compiler complains about missing braces.
Let's use use ={} initialization which should do the job:
tools/testing/selftests/bpf/prog_tests/sockmap_basic.c: In function 'test_sockmap_iter':
tools/testing/selftests/bpf/prog_tests/sockmap_basic.c:181:8: error: missing braces around initializer [-Werror=missing-braces]
union bpf_iter_link_info linfo = {0};
^
tools/testing/selftests/bpf/prog_tests/sockmap_basic.c:181:8: error: (near initialization for 'linfo.map') [-Werror=missing-braces]
tools/testing/selftests/bpf/prog_tests/sockmap_basic.c: At top level:
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20201002000451.1794044-1-sdf@google.com
Fixes clang error:
tools/testing/selftests/bpf/prog_tests/xdp_noinline.c:35:6: error: variable 'duration' is uninitialized when used here [-Werror,-Wuninitialized]
if (CHECK(!skel, "skel_open_and_load", "failed\n"))
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20201001225440.1373233-1-sdf@google.com
Daniel Borkmann says:
====================
pull-request: bpf-next 2020-10-01
The following pull-request contains BPF updates for your *net-next* tree.
We've added 90 non-merge commits during the last 8 day(s) which contain
a total of 103 files changed, 7662 insertions(+), 1894 deletions(-).
Note that once bpf(/net) tree gets merged into net-next, there will be a small
merge conflict in tools/lib/bpf/btf.c between commit 1245008122 ("libbpf: Fix
native endian assumption when parsing BTF") from the bpf tree and the commit
3289959b97 ("libbpf: Support BTF loading and raw data output in both endianness")
from the bpf-next tree. Correct resolution would be to stick with bpf-next, it
should look like:
[...]
/* check BTF magic */
if (fread(&magic, 1, sizeof(magic), f) < sizeof(magic)) {
err = -EIO;
goto err_out;
}
if (magic != BTF_MAGIC && magic != bswap_16(BTF_MAGIC)) {
/* definitely not a raw BTF */
err = -EPROTO;
goto err_out;
}
/* get file size */
[...]
The main changes are:
1) Add bpf_snprintf_btf() and bpf_seq_printf_btf() helpers to support displaying
BTF-based kernel data structures out of BPF programs, from Alan Maguire.
2) Speed up RCU tasks trace grace periods by a factor of 50 & fix a few race
conditions exposed by it. It was discussed to take these via BPF and
networking tree to get better testing exposure, from Paul E. McKenney.
3) Support multi-attach for freplace programs, needed for incremental attachment
of multiple XDP progs using libxdp dispatcher model, from Toke Høiland-Jørgensen.
4) libbpf support for appending new BTF types at the end of BTF object, allowing
intrusive changes of prog's BTF (useful for future linking), from Andrii Nakryiko.
5) Several BPF helper improvements e.g. avoid atomic op in cookie generator and add
a redirect helper into neighboring subsys, from Daniel Borkmann.
6) Allow map updates on sockmaps from bpf_iter context in order to migrate sockmaps
from one to another, from Lorenz Bauer.
7) Fix 32 bit to 64 bit assignment from latest alu32 bounds tracking which caused
a verifier issue due to type downgrade to scalar, from John Fastabend.
8) Follow-up on tail-call support in BPF subprogs which optimizes x64 JIT prologue
and epilogue sections, from Maciej Fijalkowski.
9) Add an option to perf RB map to improve sharing of event entries by avoiding remove-
on-close behavior. Also, add BPF_PROG_TEST_RUN for raw_tracepoint, from Song Liu.
10) Fix a crash in AF_XDP's socket_release when memory allocation for UMEMs fails,
from Magnus Karlsson.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Add tests for perf event array with and without BPF_F_PRESERVE_ELEMS.
Add a perf event to array via fd mfd. Without BPF_F_PRESERVE_ELEMS, the
perf event is removed when mfd is closed. With BPF_F_PRESERVE_ELEMS, the
perf event is removed when the map is freed.
Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200930224927.1936644-3-songliubraving@fb.com
Currently, perf event in perf event array is removed from the array when
the map fd used to add the event is closed. This behavior makes it
difficult to the share perf events with perf event array.
Introduce perf event map that keeps the perf event open with a new flag
BPF_F_PRESERVE_ELEMS. With this flag set, perf events in the array are not
removed when the original map fd is closed. Instead, the perf event will
stay in the map until 1) it is explicitly removed from the array; or 2)
the array is freed.
Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200930224927.1936644-2-songliubraving@fb.com
Fix a build failure on arm64, due to missing alignment information for
the .BTF_ids section:
resolve_btfids.test.o: in function `test_resolve_btfids':
tools/testing/selftests/bpf/prog_tests/resolve_btfids.c:140:(.text+0x29c): relocation truncated to fit: R_AARCH64_LDST32_ABS_LO12_NC against `.BTF_ids'
ld: tools/testing/selftests/bpf/prog_tests/resolve_btfids.c:140: warning: one possible cause of this error is that the symbol is being referenced in the indicated code as if it had a larger alignment than was declared where it was defined
In vmlinux, the .BTF_ids section is aligned to 4 bytes by vmlinux.lds.h.
In test_progs however, .BTF_ids doesn't have alignment constraints. The
arm64 linker expects the btf_id_set.cnt symbol, a u32, to be naturally
aligned but finds it misaligned and cannot apply the relocation. Enforce
alignment of .BTF_ids to 4 bytes.
Fixes: cd04b04de1 ("selftests/bpf: Add set test to resolve_btfids")
Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Link: https://lore.kernel.org/bpf/20200930093559.2120126-1-jean-philippe@linaro.org
Test that drop monitor correctly captures both software and hardware
originated packet drops.
# ./drop_monitor_tests.sh
Software drops test
TEST: Capturing active software drops [ OK ]
TEST: Capturing inactive software drops [ OK ]
Hardware drops test
TEST: Capturing active hardware drops [ OK ]
TEST: Capturing inactive hardware drops [ OK ]
Tests passed: 4
Tests failed: 0
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a test for PFC. Runs 10MB of traffic through a bottleneck and checks
that none of it gets lost.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a test for headroom configuration. This covers projection of ETS
configuration to ingress, PFC, adjustments for MTU, the qdisc / TC
mode and the effect of egress SPAN session on buffer configuration.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
mlnx_qos is a script for configuration of DCB. Despite the name it is not
actually Mellanox-specific in any way. It is currently the only ad-hoc tool
available (in contrast to a daemon that manages an interface on an ongoing
basis). However, it is very verbose and parsing out error messages is not
really possible. Add a wrapper that makes it easier to use the tool.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some selftests may not need any actual ports. Technically those are not
forwarding selftests, but devlink_lib can still be handy. Fall back on
NETIF_NO_CABLE in those cases.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a helper that answers the cell size of the devlink device.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Changing pool type from static to dynamic causes reinterpretation of
threshold values. They therefore need to be saved before pool type is
changed, then the pool type can be changed, and then the new values need
to be set up.
For that reason, set cannot subsume save, because it would be saving the
wrong thing, with possibly a nonsensical value, and restore would then fail
to restore the nonsensical value.
Thus extract a _save() from each of the relevant _set()'s. This way it is
possible to save everything up front, then to tweak it, and then restore in
the required order.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add test validating that btf_dump works fine with BTFs that are modified and
incrementally generated.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200929232843.1249318-5-andriin@fb.com
Ensure that btf_dump can accommodate new BTF types being appended to BTF
instance after struct btf_dump was created. This came up during attemp to
use btf_dump for raw type dumping in selftests, but given changes are not
excessive, it's good to not have any gotchas in API usage, so I decided to
support such use case in general.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200929232843.1249318-2-andriin@fb.com
Port of tail_call_static() helper function from Cilium's BPF code base [0]
to libbpf, so others can easily consume it as well. We've been using this
in production code for some time now. The main idea is that we guarantee
that the kernel's BPF infrastructure and JIT (here: x86_64) can patch the
JITed BPF insns with direct jumps instead of having to fall back to using
expensive retpolines. By using inline asm, we guarantee that the compiler
won't merge the call from different paths with potentially different
content of r2/r3.
We're also using Cilium's __throw_build_bug() macro (here as: __bpf_unreachable())
in different places as a neat trick to trigger compilation errors when
compiler does not remove code at compilation time. This works for the BPF
back end as it does not implement the __builtin_trap().
[0] f5537c2602
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/1656a082e077552eb46642d513b4a6bde9a7dd01.1601477936.git.daniel@iogearbox.net
Add a redirect_neigh() helper as redirect() drop-in replacement
for the xmit side. Main idea for the helper is to be very similar
in semantics to the latter just that the skb gets injected into
the neighboring subsystem in order to let the stack do the work
it knows best anyway to populate the L2 addresses of the packet
and then hand over to dev_queue_xmit() as redirect() does.
This solves two bigger items: i) skbs don't need to go up to the
stack on the host facing veth ingress side for traffic egressing
the container to achieve the same for populating L2 which also
has the huge advantage that ii) the skb->sk won't get orphaned in
ip_rcv_core() when entering the IP routing layer on the host stack.
Given that skb->sk neither gets orphaned when crossing the netns
as per 9c4c325252 ("skbuff: preserve sock reference when scrubbing
the skb.") the helper can then push the skbs directly to the phys
device where FQ scheduler can do its work and TCP stack gets proper
backpressure given we hold on to skb->sk as long as skb is still
residing in queues.
With the helper used in BPF data path to then push the skb to the
phys device, I observed a stable/consistent TCP_STREAM improvement
on veth devices for traffic going container -> host -> host ->
container from ~10Gbps to ~15Gbps for a single stream in my test
environment.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Reviewed-by: David Ahern <dsahern@gmail.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Cc: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/bpf/f207de81629e1724899b73b8112e0013be782d35.1601477936.git.daniel@iogearbox.net
Similarly to 5a52ae4e32 ("bpf: Allow to retrieve cgroup v1 classid
from v2 hooks"), add a helper to retrieve cgroup v1 classid solely
based on the skb->sk, so it can be used as key as part of BPF map
lookups out of tc from host ns, in particular given the skb->sk is
retained these days when crossing net ns thanks to 9c4c325252
("skbuff: preserve sock reference when scrubbing the skb."). This
is similar to bpf_skb_cgroup_id() which implements the same for v2.
Kubernetes ecosystem is still operating on v1 however, hence net_cls
needs to be used there until this can be dropped in with the v2
helper of bpf_skb_cgroup_id().
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/ed633cf27a1c620e901c5aa99ebdefb028dce600.1601477936.git.daniel@iogearbox.net
Libbpf compiles .o's for static and shared library modes separately, so no
need to specify -fPIC for both. Keep it only for shared library mode.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20200929220604.833631-3-andriin@fb.com
For some reason compiler doesn't complain about uninitialized variable, fixed
in previous patch, if libbpf is compiled without -O2 optimization level. So do
compile it with -O2 and never let similar issue slip by again. -Wall is added
unconditionally, so no need to specify it again.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20200929220604.833631-2-andriin@fb.com
Fix obvious unitialized variable use that wasn't reported by compiler. libbpf
Makefile changes to catch such errors are added separately.
Fixes: 3289959b97 ("libbpf: Support BTF loading and raw data output in both endianness")
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20200929220604.833631-1-andriin@fb.com
This test makes a lot of narrow load checks while assuming little
endian architecture, and therefore fails on s390.
Fix by introducing LSB and LSW macros and using them to perform narrow
loads.
Fixes: 0ab5539f85 ("selftests/bpf: Tests for BPF_SK_LOOKUP attach point")
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200929201814.44360-1-iii@linux.ibm.com
