8 BTF (BPF Type Format) is the metadata format which encodes the debug info
9 related to BPF program/map. The name BTF was used initially to describe data
10 types. The BTF was later extended to include function info for defined
11 subroutines, and line info for source/line information.
13 The debug info is used for map pretty print, function signature, etc. The
14 function signature enables better bpf program/function kernel symbol. The line
15 info helps generate source annotated translated byte code, jited code and
18 The BTF specification contains two parts,
22 The kernel API is the contract between user space and kernel. The kernel
23 verifies the BTF info before using it. The ELF file format is a user space
24 contract between ELF file and libbpf loader.
26 The type and string sections are part of the BTF kernel API, describing the
27 debug info (mostly types related) referenced by the bpf program. These two
28 sections are discussed in details in :ref:`BTF_Type_String`.
32 2. BTF Type and String Encoding
33 ===============================
35 The file ``include/uapi/linux/btf.h`` provides high-level definition of how
36 types/strings are encoded.
38 The beginning of data blob must be::
46 /* All offsets are in bytes relative to the end of this header */
47 __u32 type_off; /* offset of type section */
48 __u32 type_len; /* length of type section */
49 __u32 str_off; /* offset of string section */
50 __u32 str_len; /* length of string section */
53 The magic is ``0xeB9F``, which has different encoding for big and little
54 endian systems, and can be used to test whether BTF is generated for big- or
55 little-endian target. The ``btf_header`` is designed to be extensible with
56 ``hdr_len`` equal to ``sizeof(struct btf_header)`` when a data blob is
62 The first string in the string section must be a null string. The rest of
63 string table is a concatenation of other null-terminated strings.
68 The type id ``0`` is reserved for ``void`` type. The type section is parsed
69 sequentially and type id is assigned to each recognized type starting from id
70 ``1``. Currently, the following types are supported::
72 #define BTF_KIND_INT 1 /* Integer */
73 #define BTF_KIND_PTR 2 /* Pointer */
74 #define BTF_KIND_ARRAY 3 /* Array */
75 #define BTF_KIND_STRUCT 4 /* Struct */
76 #define BTF_KIND_UNION 5 /* Union */
77 #define BTF_KIND_ENUM 6 /* Enumeration up to 32-bit values */
78 #define BTF_KIND_FWD 7 /* Forward */
79 #define BTF_KIND_TYPEDEF 8 /* Typedef */
80 #define BTF_KIND_VOLATILE 9 /* Volatile */
81 #define BTF_KIND_CONST 10 /* Const */
82 #define BTF_KIND_RESTRICT 11 /* Restrict */
83 #define BTF_KIND_FUNC 12 /* Function */
84 #define BTF_KIND_FUNC_PROTO 13 /* Function Proto */
85 #define BTF_KIND_VAR 14 /* Variable */
86 #define BTF_KIND_DATASEC 15 /* Section */
87 #define BTF_KIND_FLOAT 16 /* Floating point */
88 #define BTF_KIND_DECL_TAG 17 /* Decl Tag */
89 #define BTF_KIND_TYPE_TAG 18 /* Type Tag */
90 #define BTF_KIND_ENUM64 19 /* Enumeration up to 64-bit values */
92 Note that the type section encodes debug info, not just pure types.
93 ``BTF_KIND_FUNC`` is not a type, and it represents a defined subprogram.
95 Each type contains the following common data::
99 /* "info" bits arrangement
100 * bits 0-15: vlen (e.g. # of struct's members)
102 * bits 24-28: kind (e.g. int, ptr, array...etc)
104 * bit 31: kind_flag, currently used by
105 * struct, union, fwd, enum and enum64.
108 /* "size" is used by INT, ENUM, STRUCT, UNION and ENUM64.
109 * "size" tells the size of the type it is describing.
111 * "type" is used by PTR, TYPEDEF, VOLATILE, CONST, RESTRICT,
112 * FUNC, FUNC_PROTO, DECL_TAG and TYPE_TAG.
113 * "type" is a type_id referring to another type.
121 For certain kinds, the common data are followed by kind-specific data. The
122 ``name_off`` in ``struct btf_type`` specifies the offset in the string table.
123 The following sections detail encoding of each kind.
128 ``struct btf_type`` encoding requirement:
129 * ``name_off``: any valid offset
130 * ``info.kind_flag``: 0
131 * ``info.kind``: BTF_KIND_INT
133 * ``size``: the size of the int type in bytes.
135 ``btf_type`` is followed by a ``u32`` with the following bits arrangement::
137 #define BTF_INT_ENCODING(VAL) (((VAL) & 0x0f000000) >> 24)
138 #define BTF_INT_OFFSET(VAL) (((VAL) & 0x00ff0000) >> 16)
139 #define BTF_INT_BITS(VAL) ((VAL) & 0x000000ff)
141 The ``BTF_INT_ENCODING`` has the following attributes::
143 #define BTF_INT_SIGNED (1 << 0)
144 #define BTF_INT_CHAR (1 << 1)
145 #define BTF_INT_BOOL (1 << 2)
147 The ``BTF_INT_ENCODING()`` provides extra information: signedness, char, or
148 bool, for the int type. The char and bool encoding are mostly useful for
149 pretty print. At most one encoding can be specified for the int type.
151 The ``BTF_INT_BITS()`` specifies the number of actual bits held by this int
152 type. For example, a 4-bit bitfield encodes ``BTF_INT_BITS()`` equals to 4.
153 The ``btf_type.size * 8`` must be equal to or greater than ``BTF_INT_BITS()``
154 for the type. The maximum value of ``BTF_INT_BITS()`` is 128.
156 The ``BTF_INT_OFFSET()`` specifies the starting bit offset to calculate values
157 for this int. For example, a bitfield struct member has:
159 * btf member bit offset 100 from the start of the structure,
160 * btf member pointing to an int type,
161 * the int type has ``BTF_INT_OFFSET() = 2`` and ``BTF_INT_BITS() = 4``
163 Then in the struct memory layout, this member will occupy ``4`` bits starting
164 from bits ``100 + 2 = 102``.
166 Alternatively, the bitfield struct member can be the following to access the
167 same bits as the above:
169 * btf member bit offset 102,
170 * btf member pointing to an int type,
171 * the int type has ``BTF_INT_OFFSET() = 0`` and ``BTF_INT_BITS() = 4``
173 The original intention of ``BTF_INT_OFFSET()`` is to provide flexibility of
174 bitfield encoding. Currently, both llvm and pahole generate
175 ``BTF_INT_OFFSET() = 0`` for all int types.
180 ``struct btf_type`` encoding requirement:
182 * ``info.kind_flag``: 0
183 * ``info.kind``: BTF_KIND_PTR
185 * ``type``: the pointee type of the pointer
187 No additional type data follow ``btf_type``.
192 ``struct btf_type`` encoding requirement:
194 * ``info.kind_flag``: 0
195 * ``info.kind``: BTF_KIND_ARRAY
197 * ``size/type``: 0, not used
199 ``btf_type`` is followed by one ``struct btf_array``::
207 The ``struct btf_array`` encoding:
208 * ``type``: the element type
209 * ``index_type``: the index type
210 * ``nelems``: the number of elements for this array (``0`` is also allowed).
212 The ``index_type`` can be any regular int type (``u8``, ``u16``, ``u32``,
213 ``u64``, ``unsigned __int128``). The original design of including
214 ``index_type`` follows DWARF, which has an ``index_type`` for its array type.
215 Currently in BTF, beyond type verification, the ``index_type`` is not used.
217 The ``struct btf_array`` allows chaining through element type to represent
218 multidimensional arrays. For example, for ``int a[5][6]``, the following type
219 information illustrates the chaining:
222 * [2]: array, ``btf_array.type = [1]``, ``btf_array.nelems = 6``
223 * [3]: array, ``btf_array.type = [2]``, ``btf_array.nelems = 5``
225 Currently, both pahole and llvm collapse multidimensional array into
226 one-dimensional array, e.g., for ``a[5][6]``, the ``btf_array.nelems`` is
227 equal to ``30``. This is because the original use case is map pretty print
228 where the whole array is dumped out so one-dimensional array is enough. As
229 more BTF usage is explored, pahole and llvm can be changed to generate proper
230 chained representation for multidimensional arrays.
232 2.2.4 BTF_KIND_STRUCT
233 ~~~~~~~~~~~~~~~~~~~~~
237 ``struct btf_type`` encoding requirement:
238 * ``name_off``: 0 or offset to a valid C identifier
239 * ``info.kind_flag``: 0 or 1
240 * ``info.kind``: BTF_KIND_STRUCT or BTF_KIND_UNION
241 * ``info.vlen``: the number of struct/union members
242 * ``info.size``: the size of the struct/union in bytes
244 ``btf_type`` is followed by ``info.vlen`` number of ``struct btf_member``.::
252 ``struct btf_member`` encoding:
253 * ``name_off``: offset to a valid C identifier
254 * ``type``: the member type
255 * ``offset``: <see below>
257 If the type info ``kind_flag`` is not set, the offset contains only bit offset
258 of the member. Note that the base type of the bitfield can only be int or enum
259 type. If the bitfield size is 32, the base type can be either int or enum
260 type. If the bitfield size is not 32, the base type must be int, and int type
261 ``BTF_INT_BITS()`` encodes the bitfield size.
263 If the ``kind_flag`` is set, the ``btf_member.offset`` contains both member
264 bitfield size and bit offset. The bitfield size and bit offset are calculated
267 #define BTF_MEMBER_BITFIELD_SIZE(val) ((val) >> 24)
268 #define BTF_MEMBER_BIT_OFFSET(val) ((val) & 0xffffff)
270 In this case, if the base type is an int type, it must be a regular int type:
272 * ``BTF_INT_OFFSET()`` must be 0.
273 * ``BTF_INT_BITS()`` must be equal to ``{1,2,4,8,16} * 8``.
275 The following kernel patch introduced ``kind_flag`` and explained why both
278 https://github.com/torvalds/linux/commit/9d5f9f701b1891466fb3dbb1806ad97716f95cc3#diff-fa650a64fdd3968396883d2fe8215ff3
283 ``struct btf_type`` encoding requirement:
284 * ``name_off``: 0 or offset to a valid C identifier
285 * ``info.kind_flag``: 0 for unsigned, 1 for signed
286 * ``info.kind``: BTF_KIND_ENUM
287 * ``info.vlen``: number of enum values
290 ``btf_type`` is followed by ``info.vlen`` number of ``struct btf_enum``.::
297 The ``btf_enum`` encoding:
298 * ``name_off``: offset to a valid C identifier
301 If the original enum value is signed and the size is less than 4,
302 that value will be sign extended into 4 bytes. If the size is 8,
303 the value will be truncated into 4 bytes.
308 ``struct btf_type`` encoding requirement:
309 * ``name_off``: offset to a valid C identifier
310 * ``info.kind_flag``: 0 for struct, 1 for union
311 * ``info.kind``: BTF_KIND_FWD
315 No additional type data follow ``btf_type``.
317 2.2.8 BTF_KIND_TYPEDEF
318 ~~~~~~~~~~~~~~~~~~~~~~
320 ``struct btf_type`` encoding requirement:
321 * ``name_off``: offset to a valid C identifier
322 * ``info.kind_flag``: 0
323 * ``info.kind``: BTF_KIND_TYPEDEF
325 * ``type``: the type which can be referred by name at ``name_off``
327 No additional type data follow ``btf_type``.
329 2.2.9 BTF_KIND_VOLATILE
330 ~~~~~~~~~~~~~~~~~~~~~~~
332 ``struct btf_type`` encoding requirement:
334 * ``info.kind_flag``: 0
335 * ``info.kind``: BTF_KIND_VOLATILE
337 * ``type``: the type with ``volatile`` qualifier
339 No additional type data follow ``btf_type``.
341 2.2.10 BTF_KIND_CONST
342 ~~~~~~~~~~~~~~~~~~~~~
344 ``struct btf_type`` encoding requirement:
346 * ``info.kind_flag``: 0
347 * ``info.kind``: BTF_KIND_CONST
349 * ``type``: the type with ``const`` qualifier
351 No additional type data follow ``btf_type``.
353 2.2.11 BTF_KIND_RESTRICT
354 ~~~~~~~~~~~~~~~~~~~~~~~~
356 ``struct btf_type`` encoding requirement:
358 * ``info.kind_flag``: 0
359 * ``info.kind``: BTF_KIND_RESTRICT
361 * ``type``: the type with ``restrict`` qualifier
363 No additional type data follow ``btf_type``.
368 ``struct btf_type`` encoding requirement:
369 * ``name_off``: offset to a valid C identifier
370 * ``info.kind_flag``: 0
371 * ``info.kind``: BTF_KIND_FUNC
373 * ``type``: a BTF_KIND_FUNC_PROTO type
375 No additional type data follow ``btf_type``.
377 A BTF_KIND_FUNC defines not a type, but a subprogram (function) whose
378 signature is defined by ``type``. The subprogram is thus an instance of that
379 type. The BTF_KIND_FUNC may in turn be referenced by a func_info in the
380 :ref:`BTF_Ext_Section` (ELF) or in the arguments to :ref:`BPF_Prog_Load`
383 2.2.13 BTF_KIND_FUNC_PROTO
384 ~~~~~~~~~~~~~~~~~~~~~~~~~~
386 ``struct btf_type`` encoding requirement:
388 * ``info.kind_flag``: 0
389 * ``info.kind``: BTF_KIND_FUNC_PROTO
390 * ``info.vlen``: # of parameters
391 * ``type``: the return type
393 ``btf_type`` is followed by ``info.vlen`` number of ``struct btf_param``.::
400 If a BTF_KIND_FUNC_PROTO type is referred by a BTF_KIND_FUNC type, then
401 ``btf_param.name_off`` must point to a valid C identifier except for the
402 possible last argument representing the variable argument. The btf_param.type
403 refers to parameter type.
405 If the function has variable arguments, the last parameter is encoded with
406 ``name_off = 0`` and ``type = 0``.
411 ``struct btf_type`` encoding requirement:
412 * ``name_off``: offset to a valid C identifier
413 * ``info.kind_flag``: 0
414 * ``info.kind``: BTF_KIND_VAR
416 * ``type``: the type of the variable
418 ``btf_type`` is followed by a single ``struct btf_variable`` with the
425 ``struct btf_var`` encoding:
426 * ``linkage``: currently only static variable 0, or globally allocated
427 variable in ELF sections 1
429 Not all type of global variables are supported by LLVM at this point.
430 The following is currently available:
432 * static variables with or without section attributes
433 * global variables with section attributes
435 The latter is for future extraction of map key/value type id's from a
438 2.2.15 BTF_KIND_DATASEC
439 ~~~~~~~~~~~~~~~~~~~~~~~
441 ``struct btf_type`` encoding requirement:
442 * ``name_off``: offset to a valid name associated with a variable or
443 one of .data/.bss/.rodata
444 * ``info.kind_flag``: 0
445 * ``info.kind``: BTF_KIND_DATASEC
446 * ``info.vlen``: # of variables
447 * ``size``: total section size in bytes (0 at compilation time, patched
448 to actual size by BPF loaders such as libbpf)
450 ``btf_type`` is followed by ``info.vlen`` number of ``struct btf_var_secinfo``.::
452 struct btf_var_secinfo {
458 ``struct btf_var_secinfo`` encoding:
459 * ``type``: the type of the BTF_KIND_VAR variable
460 * ``offset``: the in-section offset of the variable
461 * ``size``: the size of the variable in bytes
463 2.2.16 BTF_KIND_FLOAT
464 ~~~~~~~~~~~~~~~~~~~~~
466 ``struct btf_type`` encoding requirement:
467 * ``name_off``: any valid offset
468 * ``info.kind_flag``: 0
469 * ``info.kind``: BTF_KIND_FLOAT
471 * ``size``: the size of the float type in bytes: 2, 4, 8, 12 or 16.
473 No additional type data follow ``btf_type``.
475 2.2.17 BTF_KIND_DECL_TAG
476 ~~~~~~~~~~~~~~~~~~~~~~~~
478 ``struct btf_type`` encoding requirement:
479 * ``name_off``: offset to a non-empty string
480 * ``info.kind_flag``: 0
481 * ``info.kind``: BTF_KIND_DECL_TAG
483 * ``type``: ``struct``, ``union``, ``func``, ``var`` or ``typedef``
485 ``btf_type`` is followed by ``struct btf_decl_tag``.::
487 struct btf_decl_tag {
491 The ``name_off`` encodes btf_decl_tag attribute string.
492 The ``type`` should be ``struct``, ``union``, ``func``, ``var`` or ``typedef``.
493 For ``var`` or ``typedef`` type, ``btf_decl_tag.component_idx`` must be ``-1``.
494 For the other three types, if the btf_decl_tag attribute is
495 applied to the ``struct``, ``union`` or ``func`` itself,
496 ``btf_decl_tag.component_idx`` must be ``-1``. Otherwise,
497 the attribute is applied to a ``struct``/``union`` member or
498 a ``func`` argument, and ``btf_decl_tag.component_idx`` should be a
499 valid index (starting from 0) pointing to a member or an argument.
501 2.2.18 BTF_KIND_TYPE_TAG
502 ~~~~~~~~~~~~~~~~~~~~~~~~
504 ``struct btf_type`` encoding requirement:
505 * ``name_off``: offset to a non-empty string
506 * ``info.kind_flag``: 0
507 * ``info.kind``: BTF_KIND_TYPE_TAG
509 * ``type``: the type with ``btf_type_tag`` attribute
511 Currently, ``BTF_KIND_TYPE_TAG`` is only emitted for pointer types.
512 It has the following btf type chain:
516 -> [const | volatile | restrict | typedef]*
519 Basically, a pointer type points to zero or more
520 type_tag, then zero or more const/volatile/restrict/typedef
521 and finally the base type. The base type is one of
522 int, ptr, array, struct, union, enum, func_proto and float types.
524 2.2.19 BTF_KIND_ENUM64
525 ~~~~~~~~~~~~~~~~~~~~~~
527 ``struct btf_type`` encoding requirement:
528 * ``name_off``: 0 or offset to a valid C identifier
529 * ``info.kind_flag``: 0 for unsigned, 1 for signed
530 * ``info.kind``: BTF_KIND_ENUM64
531 * ``info.vlen``: number of enum values
534 ``btf_type`` is followed by ``info.vlen`` number of ``struct btf_enum64``.::
542 The ``btf_enum64`` encoding:
543 * ``name_off``: offset to a valid C identifier
544 * ``val_lo32``: lower 32-bit value for a 64-bit value
545 * ``val_hi32``: high 32-bit value for a 64-bit value
547 If the original enum value is signed and the size is less than 8,
548 that value will be sign extended into 8 bytes.
553 The following bpf syscall command involves BTF:
554 * BPF_BTF_LOAD: load a blob of BTF data into kernel
555 * BPF_MAP_CREATE: map creation with btf key and value type info.
556 * BPF_PROG_LOAD: prog load with btf function and line info.
557 * BPF_BTF_GET_FD_BY_ID: get a btf fd
558 * BPF_OBJ_GET_INFO_BY_FD: btf, func_info, line_info
559 and other btf related info are returned.
561 The workflow typically looks like:
568 BPF_MAP_CREATE and BPF_PROG_LOAD
575 BPF_{PROG,MAP}_GET_NEXT_ID (get prog/map id's)
578 BPF_{PROG,MAP}_GET_FD_BY_ID (get a prog/map fd)
581 BPF_OBJ_GET_INFO_BY_FD (get bpf_prog_info/bpf_map_info with btf_id)
584 BPF_BTF_GET_FD_BY_ID (get btf_fd) |
587 BPF_OBJ_GET_INFO_BY_FD (get btf) |
590 pretty print types, dump func signatures and line info, etc.
596 Load a blob of BTF data into kernel. A blob of data, described in
597 :ref:`BTF_Type_String`, can be directly loaded into the kernel. A ``btf_fd``
598 is returned to a userspace.
603 A map can be created with ``btf_fd`` and specified key/value type id.::
605 __u32 btf_fd; /* fd pointing to a BTF type data */
606 __u32 btf_key_type_id; /* BTF type_id of the key */
607 __u32 btf_value_type_id; /* BTF type_id of the value */
609 In libbpf, the map can be defined with extra annotation like below:
613 __uint(type, BPF_MAP_TYPE_ARRAY);
615 __type(value, struct ipv_counts);
616 __uint(max_entries, 4);
617 } btf_map SEC(".maps");
619 During ELF parsing, libbpf is able to extract key/value type_id's and assign
620 them to BPF_MAP_CREATE attributes automatically.
627 During prog_load, func_info and line_info can be passed to kernel with proper
628 values for the following attributes:
634 __u32 prog_btf_fd; /* fd pointing to BTF type data */
635 __u32 func_info_rec_size; /* userspace bpf_func_info size */
636 __aligned_u64 func_info; /* func info */
637 __u32 func_info_cnt; /* number of bpf_func_info records */
638 __u32 line_info_rec_size; /* userspace bpf_line_info size */
639 __aligned_u64 line_info; /* line info */
640 __u32 line_info_cnt; /* number of bpf_line_info records */
642 The func_info and line_info are an array of below, respectively.::
644 struct bpf_func_info {
645 __u32 insn_off; /* [0, insn_cnt - 1] */
646 __u32 type_id; /* pointing to a BTF_KIND_FUNC type */
648 struct bpf_line_info {
649 __u32 insn_off; /* [0, insn_cnt - 1] */
650 __u32 file_name_off; /* offset to string table for the filename */
651 __u32 line_off; /* offset to string table for the source line */
652 __u32 line_col; /* line number and column number */
655 func_info_rec_size is the size of each func_info record, and
656 line_info_rec_size is the size of each line_info record. Passing the record
657 size to kernel make it possible to extend the record itself in the future.
659 Below are requirements for func_info:
660 * func_info[0].insn_off must be 0.
661 * the func_info insn_off is in strictly increasing order and matches
664 Below are requirements for line_info:
665 * the first insn in each func must have a line_info record pointing to it.
666 * the line_info insn_off is in strictly increasing order.
668 For line_info, the line number and column number are defined as below:
671 #define BPF_LINE_INFO_LINE_NUM(line_col) ((line_col) >> 10)
672 #define BPF_LINE_INFO_LINE_COL(line_col) ((line_col) & 0x3ff)
674 3.4 BPF_{PROG,MAP}_GET_NEXT_ID
675 ------------------------------
677 In kernel, every loaded program, map or btf has a unique id. The id won't
678 change during the lifetime of a program, map, or btf.
680 The bpf syscall command BPF_{PROG,MAP}_GET_NEXT_ID returns all id's, one for
681 each command, to user space, for bpf program or maps, respectively, so an
682 inspection tool can inspect all programs and maps.
684 3.5 BPF_{PROG,MAP}_GET_FD_BY_ID
685 -------------------------------
687 An introspection tool cannot use id to get details about program or maps.
688 A file descriptor needs to be obtained first for reference-counting purpose.
690 3.6 BPF_OBJ_GET_INFO_BY_FD
691 --------------------------
693 Once a program/map fd is acquired, an introspection tool can get the detailed
694 information from kernel about this fd, some of which are BTF-related. For
695 example, ``bpf_map_info`` returns ``btf_id`` and key/value type ids.
696 ``bpf_prog_info`` returns ``btf_id``, func_info, and line info for translated
697 bpf byte codes, and jited_line_info.
699 3.7 BPF_BTF_GET_FD_BY_ID
700 ------------------------
702 With ``btf_id`` obtained in ``bpf_map_info`` and ``bpf_prog_info``, bpf
703 syscall command BPF_BTF_GET_FD_BY_ID can retrieve a btf fd. Then, with
704 command BPF_OBJ_GET_INFO_BY_FD, the btf blob, originally loaded into the
705 kernel with BPF_BTF_LOAD, can be retrieved.
707 With the btf blob, ``bpf_map_info``, and ``bpf_prog_info``, an introspection
708 tool has full btf knowledge and is able to pretty print map key/values, dump
709 func signatures and line info, along with byte/jit codes.
711 4. ELF File Format Interface
712 ============================
717 The .BTF section contains type and string data. The format of this section is
718 same as the one describe in :ref:`BTF_Type_String`.
725 The .BTF.ext section encodes func_info and line_info which needs loader
726 manipulation before loading into the kernel.
728 The specification for .BTF.ext section is defined at ``tools/lib/bpf/btf.h``
729 and ``tools/lib/bpf/btf.c``.
731 The current header of .BTF.ext section::
733 struct btf_ext_header {
739 /* All offsets are in bytes relative to the end of this header */
746 It is very similar to .BTF section. Instead of type/string section, it
747 contains func_info and line_info section. See :ref:`BPF_Prog_Load` for details
748 about func_info and line_info record format.
750 The func_info is organized as below.::
753 btf_ext_info_sec for section #1 /* func_info for section #1 */
754 btf_ext_info_sec for section #2 /* func_info for section #2 */
757 ``func_info_rec_size`` specifies the size of ``bpf_func_info`` structure when
758 .BTF.ext is generated. ``btf_ext_info_sec``, defined below, is a collection of
759 func_info for each specific ELF section.::
761 struct btf_ext_info_sec {
762 __u32 sec_name_off; /* offset to section name */
764 /* Followed by num_info * record_size number of bytes */
768 Here, num_info must be greater than 0.
770 The line_info is organized as below.::
773 btf_ext_info_sec for section #1 /* line_info for section #1 */
774 btf_ext_info_sec for section #2 /* line_info for section #2 */
777 ``line_info_rec_size`` specifies the size of ``bpf_line_info`` structure when
778 .BTF.ext is generated.
780 The interpretation of ``bpf_func_info->insn_off`` and
781 ``bpf_line_info->insn_off`` is different between kernel API and ELF API. For
782 kernel API, the ``insn_off`` is the instruction offset in the unit of ``struct
783 bpf_insn``. For ELF API, the ``insn_off`` is the byte offset from the
784 beginning of section (``btf_ext_info_sec->sec_name_off``).
789 The .BTF_ids section encodes BTF ID values that are used within the kernel.
791 This section is created during the kernel compilation with the help of
792 macros defined in ``include/linux/btf_ids.h`` header file. Kernel code can
793 use them to create lists and sets (sorted lists) of BTF ID values.
795 The ``BTF_ID_LIST`` and ``BTF_ID`` macros define unsorted list of BTF ID values,
796 with following syntax::
802 resulting in following layout in .BTF_ids section::
804 __BTF_ID__type1__name1__1:
806 __BTF_ID__type2__name2__2:
809 The ``u32 list[];`` variable is defined to access the list.
811 The ``BTF_ID_UNUSED`` macro defines 4 zero bytes. It's used when we
812 want to define unused entry in BTF_ID_LIST, like::
814 BTF_ID_LIST(bpf_skb_output_btf_ids)
815 BTF_ID(struct, sk_buff)
817 BTF_ID(struct, task_struct)
819 The ``BTF_SET_START/END`` macros pair defines sorted list of BTF ID values
820 and their count, with following syntax::
827 resulting in following layout in .BTF_ids section::
831 __BTF_ID__type1__name1__3:
833 __BTF_ID__type2__name2__4:
836 The ``struct btf_id_set set;`` variable is defined to access the list.
838 The ``typeX`` name can be one of following::
840 struct, union, typedef, func
842 and is used as a filter when resolving the BTF ID value.
844 All the BTF ID lists and sets are compiled in the .BTF_ids section and
845 resolved during the linking phase of kernel build by ``resolve_btfids`` tool.
850 5.1 bpftool map pretty print
851 ----------------------------
853 With BTF, the map key/value can be printed based on fields rather than simply
854 raw bytes. This is especially valuable for large structure or if your data
855 structure has bitfields. For example, for the following map,::
857 enum A { A1, A2, A3, A4, A5 };
869 __uint(type, BPF_MAP_TYPE_ARRAY);
871 __type(value, struct tmp_t);
872 __uint(max_entries, 1);
873 } tmpmap SEC(".maps");
875 bpftool is able to pretty print like below:
891 5.2 bpftool prog dump
892 ---------------------
894 The following is an example showing how func_info and line_info can help prog
895 dump with better kernel symbol names, function prototypes and line
898 $ bpftool prog dump jited pinned /sys/fs/bpf/test_btf_haskv
900 int test_long_fname_2(struct dummy_tracepoint_args * arg):
901 bpf_prog_44a040bf25481309_test_long_fname_2:
902 ; static int test_long_fname_2(struct dummy_tracepoint_args *arg)
907 f: mov %rbx,0x0(%rbp)
908 13: mov %r13,0x8(%rbp)
909 17: mov %r14,0x10(%rbp)
910 1b: mov %r15,0x18(%rbp)
912 21: mov %rax,0x20(%rbp)
915 27: mov %esi,-0x4(%rbp)
917 2a: mov 0x8(%rdi),%rdi
920 32: je 0x0000000000000070
922 ; counts = bpf_map_lookup_elem(&btf_map, &key);
928 The following is an example of how line_info can help debugging verification
931 /* The code at tools/testing/selftests/bpf/test_xdp_noinline.c
932 * is modified as below.
934 data = (void *)(long)xdp->data;
935 data_end = (void *)(long)xdp->data_end;
937 if (data + 4 > data_end)
940 *(u32 *)data = dst->dst;
942 $ bpftool prog load ./test_xdp_noinline.o /sys/fs/bpf/test_xdp_noinline type xdp
943 ; data = (void *)(long)xdp->data;
944 224: (79) r2 = *(u64 *)(r10 -112)
945 225: (61) r2 = *(u32 *)(r2 +0)
946 ; *(u32 *)data = dst->dst;
947 226: (63) *(u32 *)(r2 +0) = r1
948 invalid access to packet, off=0 size=4, R2(id=0,off=0,r=0)
949 R2 offset is outside of the packet
954 You need latest pahole
956 https://git.kernel.org/pub/scm/devel/pahole/pahole.git/
958 or llvm (8.0 or later). The pahole acts as a dwarf2btf converter. It doesn't
959 support .BTF.ext and btf BTF_KIND_FUNC type yet. For example,::
967 -bash-4.4$ gcc -c -O2 -g t.c
968 -bash-4.4$ pahole -JV t.o
970 [1] STRUCT t kind_flag=1 size=4 vlen=3
971 a type_id=2 bitfield_size=2 bits_offset=0
972 b type_id=2 bitfield_size=3 bits_offset=2
973 c type_id=2 bitfield_size=2 bits_offset=5
974 [2] INT int size=4 bit_offset=0 nr_bits=32 encoding=SIGNED
976 The llvm is able to generate .BTF and .BTF.ext directly with -g for bpf target
977 only. The assembly code (-S) is able to show the BTF encoding in assembly
984 int (*f2)(char q1, __int32 q2, ...);
987 int main() { return 0; }
988 int test() { return 0; }
989 -bash-4.4$ clang -c -g -O2 -target bpf t2.c
990 -bash-4.4$ readelf -S t2.o
992 [ 8] .BTF PROGBITS 0000000000000000 00000247
993 000000000000016e 0000000000000000 0 0 1
994 [ 9] .BTF.ext PROGBITS 0000000000000000 000003b5
995 0000000000000060 0000000000000000 0 0 1
996 [10] .rel.BTF.ext REL 0000000000000000 000007e0
997 0000000000000040 0000000000000010 16 9 8
999 -bash-4.4$ clang -S -g -O2 -target bpf t2.c
1002 .section .BTF,"",@progbits
1003 .short 60319 # 0xeb9f
1011 .long 0 # BTF_KIND_FUNC_PROTO(id = 1)
1012 .long 218103808 # 0xd000000
1014 .long 83 # BTF_KIND_INT(id = 2)
1015 .long 16777216 # 0x1000000
1017 .long 16777248 # 0x1000020
1019 .byte 0 # string offset=0
1020 .ascii ".text" # string offset=1
1022 .ascii "/home/yhs/tmp-pahole/t2.c" # string offset=7
1024 .ascii "int main() { return 0; }" # string offset=33
1026 .ascii "int test() { return 0; }" # string offset=58
1028 .ascii "int" # string offset=83
1030 .section .BTF.ext,"",@progbits
1031 .short 60319 # 0xeb9f
1040 .long 1 # FuncInfo section string offset=1
1047 .long 1 # LineInfo section string offset=1
1052 .long 7182 # Line 7 Col 14
1056 .long 8206 # Line 8 Col 14
1061 Kernel bpf selftest `test_btf.c` provides extensive set of BTF-related tests.