linux-stable.git/tools/lib/bpf/libbpf.map, branch linux-5.2.y

bpf, libbpf: add support for BTF Var and DataSec

2019-04-10T00:05:47+00:00

This adds libbpf support for BTF Var and DataSec kinds. Main point
here is that libbpf needs to do some preparatory work before the
whole BTF object can be loaded into the kernel, that is, fixing up
of DataSec size taken from the ELF section size and non-static
variable offset which needs to be taken from the ELF's string section.

Upstream LLVM doesn't fix these up since at time of BTF emission
it is too early in the compilation process thus this information
isn't available yet, hence loader needs to take care of it.

Note, deduplication handling has not been in the scope of this work
and needs to be addressed in a future commit.

Signed-off-by: Daniel Borkmann 
Link: https://reviews.llvm.org/D59441
Acked-by: Martin KaFai Lau 
Signed-off-by: Alexei Starovoitov

bpf, libbpf: support global data/bss/rodata sections

2019-04-10T00:05:47+00:00

This work adds BPF loader support for global data sections
to libbpf. This allows to write BPF programs in more natural
C-like way by being able to define global variables and const
data.

Back at LPC 2018 [0] we presented a first prototype which
implemented support for global data sections by extending BPF
syscall where union bpf_attr would get additional memory/size
pair for each section passed during prog load in order to later
add this base address into the ldimm64 instruction along with
the user provided offset when accessing a variable. Consensus
from LPC was that for proper upstream support, it would be
more desirable to use maps instead of bpf_attr extension as
this would allow for introspection of these sections as well
as potential live updates of their content. This work follows
this path by taking the following steps from loader side:

 1) In bpf_object__elf_collect() step we pick up ".data",
    ".rodata", and ".bss" section information.

 2) If present, in bpf_object__init_internal_map() we add
    maps to the obj's map array that corresponds to each
    of the present sections. Given section size and access
    properties can differ, a single entry array map is
    created with value size that is corresponding to the
    ELF section size of .data, .bss or .rodata. These
    internal maps are integrated into the normal map
    handling of libbpf such that when user traverses all
    obj maps, they can be differentiated from user-created
    ones via bpf_map__is_internal(). In later steps when
    we actually create these maps in the kernel via
    bpf_object__create_maps(), then for .data and .rodata
    sections their content is copied into the map through
    bpf_map_update_elem(). For .bss this is not necessary
    since array map is already zero-initialized by default.
    Additionally, for .rodata the map is frozen as read-only
    after setup, such that neither from program nor syscall
    side writes would be possible.

 3) In bpf_program__collect_reloc() step, we record the
    corresponding map, insn index, and relocation type for
    the global data.

 4) And last but not least in the actual relocation step in
    bpf_program__relocate(), we mark the ldimm64 instruction
    with src_reg = BPF_PSEUDO_MAP_VALUE where in the first
    imm field the map's file descriptor is stored as similarly
    done as in BPF_PSEUDO_MAP_FD, and in the second imm field
    (as ldimm64 is 2-insn wide) we store the access offset
    into the section. Given these maps have only single element
    ldimm64's off remains zero in both parts.

 5) On kernel side, this special marked BPF_PSEUDO_MAP_VALUE
    load will then store the actual target address in order
    to have a 'map-lookup'-free access. That is, the actual
    map value base address + offset. The destination register
    in the verifier will then be marked as PTR_TO_MAP_VALUE,
    containing the fixed offset as reg->off and backing BPF
    map as reg->map_ptr. Meaning, it's treated as any other
    normal map value from verification side, only with
    efficient, direct value access instead of actual call to
    map lookup helper as in the typical case.

Currently, only support for static global variables has been
added, and libbpf rejects non-static global variables from
loading. This can be lifted until we have proper semantics
for how BPF will treat multi-object BPF loads. From BTF side,
libbpf will set the value type id of the types corresponding
to the ".bss", ".data" and ".rodata" names which LLVM will
emit without the object name prefix. The key type will be
left as zero, thus making use of the key-less BTF option in
array maps.

Simple example dump of program using globals vars in each
section:

  # bpftool prog
  [...]
  6784: sched_cls  name load_static_dat  tag a7e1291567277844  gpl
        loaded_at 2019-03-11T15:39:34+0000  uid 0
        xlated 1776B  jited 993B  memlock 4096B  map_ids 2238,2237,2235,2236,2239,2240

  # bpftool map show id 2237
  2237: array  name test_glo.bss  flags 0x0
        key 4B  value 64B  max_entries 1  memlock 4096B
  # bpftool map show id 2235
  2235: array  name test_glo.data  flags 0x0
        key 4B  value 64B  max_entries 1  memlock 4096B
  # bpftool map show id 2236
  2236: array  name test_glo.rodata  flags 0x80
        key 4B  value 96B  max_entries 1  memlock 4096B

  # bpftool prog dump xlated id 6784
  int load_static_data(struct __sk_buff * skb):
  ; int load_static_data(struct __sk_buff *skb)
     0: (b7) r6 = 0
  ; test_reloc(number, 0, &num0);
     1: (63) *(u32 *)(r10 -4) = r6
     2: (bf) r2 = r10
  ; int load_static_data(struct __sk_buff *skb)
     3: (07) r2 += -4
  ; test_reloc(number, 0, &num0);
     4: (18) r1 = map[id:2238]
     6: (18) r3 = map[id:2237][0]+0    <-- direct addr in .bss area
     8: (b7) r4 = 0
     9: (85) call array_map_update_elem#100464
    10: (b7) r1 = 1
  ; test_reloc(number, 1, &num1);
  [...]
  ; test_reloc(string, 2, str2);
   120: (18) r8 = map[id:2237][0]+16   <-- same here at offset +16
   122: (18) r1 = map[id:2239]
   124: (18) r3 = map[id:2237][0]+16
   126: (b7) r4 = 0
   127: (85) call array_map_update_elem#100464
   128: (b7) r1 = 120
  ; str1[5] = 'x';
   129: (73) *(u8 *)(r9 +5) = r1
  ; test_reloc(string, 3, str1);
   130: (b7) r1 = 3
   131: (63) *(u32 *)(r10 -4) = r1
   132: (b7) r9 = 3
   133: (bf) r2 = r10
  ; int load_static_data(struct __sk_buff *skb)
   134: (07) r2 += -4
  ; test_reloc(string, 3, str1);
   135: (18) r1 = map[id:2239]
   137: (18) r3 = map[id:2235][0]+16   <-- direct addr in .data area
   139: (b7) r4 = 0
   140: (85) call array_map_update_elem#100464
   141: (b7) r1 = 111
  ; __builtin_memcpy(&str2[2], "hello", sizeof("hello"));
   142: (73) *(u8 *)(r8 +6) = r1       <-- further access based on .bss data
   143: (b7) r1 = 108
   144: (73) *(u8 *)(r8 +5) = r1
  [...]

For Cilium use-case in particular, this enables migrating configuration
constants from Cilium daemon's generated header defines into global
data sections such that expensive runtime recompilations with LLVM can
be avoided altogether. Instead, the ELF file becomes effectively a
"template", meaning, it is compiled only once (!) and the Cilium daemon
will then rewrite relevant configuration data from the ELF's .data or
.rodata sections directly instead of recompiling the program. The
updated ELF is then loaded into the kernel and atomically replaces
the existing program in the networking datapath. More info in [0].

Based upon recent fix in LLVM, commit c0db6b6bd444 ("[BPF] Don't fail
for static variables").

  [0] LPC 2018, BPF track, "ELF relocation for static data in BPF",
      http://vger.kernel.org/lpc-bpf2018.html#session-3

Signed-off-by: Daniel Borkmann 
Acked-by: Andrii Nakryiko 
Acked-by: Martin KaFai Lau 
Signed-off-by: Alexei Starovoitov

tools lib bpf: Introduce bpf_program__get_prog_info_linear()

2019-03-19T19:52:06+00:00

Currently, bpf_prog_info includes 9 arrays. The user has the option to
fetch any combination of these arrays. However, this requires a lot of
handling.

This work becomes more tricky when we need to store bpf_prog_info to a
file, because these arrays are allocated independently.

This patch introduces 'struct bpf_prog_info_linear', which stores arrays
of bpf_prog_info in continuous memory.

Helper functions are introduced to unify the work to get different sets
of bpf_prog_info.  Specifically, bpf_program__get_prog_info_linear()
allows the user to select which arrays to fetch, and handles details for
the user.

Please see the comments right before 'enum bpf_prog_info_array' for more
details and examples.

Signed-off-by: Song Liu 
Reviewed-by: Jiri Olsa 
Acked-by: Daniel Borkmann 
Link: https://lkml.kernel.org/r/ce92c091-e80d-a0c1-4aa0-987706c42b20@iogearbox.net
Tested-by: Arnaldo Carvalho de Melo 
Cc: Alexei Starovoitov 
Cc: kernel-team@fb.com
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Stanislav Fomichev 
Link: http://lkml.kernel.org/r/20190312053051.2690567-3-songliubraving@fb.com
Signed-off-by: Arnaldo Carvalho de Melo

libbpf: add support for using AF_XDP sockets

2019-02-25T22:21:42+00:00

This commit adds AF_XDP support to libbpf. The main reason for this is
to facilitate writing applications that use AF_XDP by offering
higher-level APIs that hide many of the details of the AF_XDP
uapi. This is in the same vein as libbpf facilitates XDP adoption by
offering easy-to-use higher level interfaces of XDP
functionality. Hopefully this will facilitate adoption of AF_XDP, make
applications using it simpler and smaller, and finally also make it
possible for applications to benefit from optimizations in the AF_XDP
user space access code. Previously, people just copied and pasted the
code from the sample application into their application, which is not
desirable.

The interface is composed of two parts:

* Low-level access interface to the four rings and the packet
* High-level control plane interface for creating and setting
  up umems and af_xdp sockets as well as a simple XDP program.

Tested-by: Björn Töpel 
Signed-off-by: Magnus Karlsson 
Signed-off-by: Daniel Borkmann

libbpf: Introduce bpf_object__btf

2019-02-15T14:20:54+00:00

Add new accessor for bpf_object to get opaque struct btf * from it.

struct btf * is needed for all operations with BTF and it's present in
bpf_object. The only thing missing is a way to get it.

Example use-case is to get BTF key_type_id and value_type_id for a map in
bpf_object. It can be done with btf__get_map_kv_tids() but that function
requires struct btf *.

Similar API can be added for struct btf_ext but no use-case for it yet.

Signed-off-by: Andrey Ignatov 
Acked-by: Yonghong Song 
Signed-off-by: Daniel Borkmann

libbpf: Introduce bpf_map__resize

2019-02-15T14:20:42+00:00

Add bpf_map__resize() to change max_entries for a map.

Quite often necessary map size is unknown at compile time and can be
calculated only at run time.

Currently the following approach is used to do so:
* bpf_object__open_buffer() to open Elf file from a buffer;
* bpf_object__find_map_by_name() to find relevant map;
* bpf_map__def() to get map attributes and create struct
  bpf_create_map_attr from them;
* update max_entries in bpf_create_map_attr;
* bpf_create_map_xattr() to create new map with updated max_entries;
* bpf_map__reuse_fd() to replace the map in bpf_object with newly
  created one.

And after all this bpf_object can finally be loaded. The map will have
new size.

It 1) is quite a lot of steps; 2) doesn't take BTF into account.

For "2)" even more steps should be made and some of them require changes
to libbpf (e.g. to get struct btf * from bpf_object).

Instead the whole problem can be solved by introducing simple
bpf_map__resize() API that checks the map and sets new max_entries if
the map is not loaded yet.

So the new steps are:
* bpf_object__open_buffer() to open Elf file from a buffer;
* bpf_object__find_map_by_name() to find relevant map;
* bpf_map__resize() to update max_entries.

That's much simpler and works with BTF.

Signed-off-by: Andrey Ignatov 
Acked-by: Yonghong Song 
Signed-off-by: Daniel Borkmann

tools/bpf: remove btf__get_strings() superseded by raw data API

2019-02-08T20:04:13+00:00

Now that we have btf__get_raw_data() it's trivial for tests to iterate
over all strings for testing purposes, which eliminates the need for
btf__get_strings() API.

Signed-off-by: Andrii Nakryiko 
Acked-by: Yonghong Song 
Signed-off-by: Alexei Starovoitov

btf: expose API to work with raw btf_ext data

2019-02-08T20:04:13+00:00

This patch changes struct btf_ext to retain original data in sequential
block of memory, which makes it possible to expose
btf_ext__get_raw_data() interface similar to btf__get_raw_data(), allowing
users of libbpf to get access to raw representation of .BTF.ext section.

Signed-off-by: Andrii Nakryiko 
Acked-by: Yonghong Song 
Signed-off-by: Alexei Starovoitov

btf: expose API to work with raw btf data

2019-02-08T20:04:13+00:00

This patch exposes new API btf__get_raw_data() that allows to get a copy
of raw BTF data out of struct btf. This is useful for external programs
that need to manipulate raw data, e.g., pahole using btf__dedup() to
deduplicate BTF type info and then writing it back to file.

Signed-off-by: Andrii Nakryiko 
Acked-by: Song Liu 
Signed-off-by: Alexei Starovoitov

btf: separate btf creation and loading

2019-02-08T20:04:13+00:00

This change splits out previous btf__new functionality of constructing
struct btf and loading it into kernel into two:
- btf__new() just creates and initializes struct btf
- btf__load() attempts to load existing struct btf into kernel

btf__free will still close BTF fd, if it was ever loaded successfully
into kernel.

This change allows users of libbpf to manipulate BTF using its API,
without the need to unnecessarily load it into kernel.

One of the intended use cases is pahole, which will do DWARF to BTF
conversion and then use libbpf to do type deduplication, while then
handling ELF sections overwriting and other concerns on its own.

Fixes: 2d3feca8c44f ("bpf: btf: print map dump and lookup with btf info")
Signed-off-by: Andrii Nakryiko 
Acked-by: Song Liu 
Signed-off-by: Alexei Starovoitov