1 ===================== 1 =====================
2 BPF Type Format (BTF) 2 BPF Type Format (BTF)
3 ===================== 3 =====================
4 4
5 1. Introduction 5 1. Introduction
6 =============== !! 6 ***************
7 7
8 BTF (BPF Type Format) is the metadata format w 8 BTF (BPF Type Format) is the metadata format which encodes the debug info
9 related to BPF program/map. The name BTF was u 9 related to BPF program/map. The name BTF was used initially to describe data
10 types. The BTF was later extended to include f 10 types. The BTF was later extended to include function info for defined
11 subroutines, and line info for source/line inf 11 subroutines, and line info for source/line information.
12 12
13 The debug info is used for map pretty print, f 13 The debug info is used for map pretty print, function signature, etc. The
14 function signature enables better bpf program/ 14 function signature enables better bpf program/function kernel symbol. The line
15 info helps generate source annotated translate 15 info helps generate source annotated translated byte code, jited code and
16 verifier log. 16 verifier log.
17 17
18 The BTF specification contains two parts, 18 The BTF specification contains two parts,
19 * BTF kernel API 19 * BTF kernel API
20 * BTF ELF file format 20 * BTF ELF file format
21 21
22 The kernel API is the contract between user sp 22 The kernel API is the contract between user space and kernel. The kernel
23 verifies the BTF info before using it. The ELF 23 verifies the BTF info before using it. The ELF file format is a user space
24 contract between ELF file and libbpf loader. 24 contract between ELF file and libbpf loader.
25 25
26 The type and string sections are part of the B 26 The type and string sections are part of the BTF kernel API, describing the
27 debug info (mostly types related) referenced b 27 debug info (mostly types related) referenced by the bpf program. These two
28 sections are discussed in details in :ref:`BTF 28 sections are discussed in details in :ref:`BTF_Type_String`.
29 29
30 .. _BTF_Type_String: 30 .. _BTF_Type_String:
31 31
32 2. BTF Type and String Encoding 32 2. BTF Type and String Encoding
33 =============================== !! 33 *******************************
34 34
35 The file ``include/uapi/linux/btf.h`` provides 35 The file ``include/uapi/linux/btf.h`` provides high-level definition of how
36 types/strings are encoded. 36 types/strings are encoded.
37 37
38 The beginning of data blob must be:: 38 The beginning of data blob must be::
39 39
40 struct btf_header { 40 struct btf_header {
41 __u16 magic; 41 __u16 magic;
42 __u8 version; 42 __u8 version;
43 __u8 flags; 43 __u8 flags;
44 __u32 hdr_len; 44 __u32 hdr_len;
45 45
46 /* All offsets are in bytes relative t 46 /* All offsets are in bytes relative to the end of this header */
47 __u32 type_off; /* offset of t 47 __u32 type_off; /* offset of type section */
48 __u32 type_len; /* length of t 48 __u32 type_len; /* length of type section */
49 __u32 str_off; /* offset of s 49 __u32 str_off; /* offset of string section */
50 __u32 str_len; /* length of s 50 __u32 str_len; /* length of string section */
51 }; 51 };
52 52
53 The magic is ``0xeB9F``, which has different e 53 The magic is ``0xeB9F``, which has different encoding for big and little
54 endian systems, and can be used to test whethe 54 endian systems, and can be used to test whether BTF is generated for big- or
55 little-endian target. The ``btf_header`` is de 55 little-endian target. The ``btf_header`` is designed to be extensible with
56 ``hdr_len`` equal to ``sizeof(struct btf_heade 56 ``hdr_len`` equal to ``sizeof(struct btf_header)`` when a data blob is
57 generated. 57 generated.
58 58
59 2.1 String Encoding 59 2.1 String Encoding
60 ------------------- !! 60 ===================
61 61
62 The first string in the string section must be 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- 63 string table is a concatenation of other null-terminated strings.
64 64
65 2.2 Type Encoding 65 2.2 Type Encoding
66 ----------------- !! 66 =================
67 67
68 The type id ``0`` is reserved for ``void`` typ 68 The type id ``0`` is reserved for ``void`` type. The type section is parsed
69 sequentially and type id is assigned to each r 69 sequentially and type id is assigned to each recognized type starting from id
70 ``1``. Currently, the following types are supp 70 ``1``. Currently, the following types are supported::
71 71
72 #define BTF_KIND_INT 1 /* 72 #define BTF_KIND_INT 1 /* Integer */
73 #define BTF_KIND_PTR 2 /* 73 #define BTF_KIND_PTR 2 /* Pointer */
74 #define BTF_KIND_ARRAY 3 /* 74 #define BTF_KIND_ARRAY 3 /* Array */
75 #define BTF_KIND_STRUCT 4 /* 75 #define BTF_KIND_STRUCT 4 /* Struct */
76 #define BTF_KIND_UNION 5 /* 76 #define BTF_KIND_UNION 5 /* Union */
77 #define BTF_KIND_ENUM 6 /* !! 77 #define BTF_KIND_ENUM 6 /* Enumeration */
78 #define BTF_KIND_FWD 7 /* 78 #define BTF_KIND_FWD 7 /* Forward */
79 #define BTF_KIND_TYPEDEF 8 /* 79 #define BTF_KIND_TYPEDEF 8 /* Typedef */
80 #define BTF_KIND_VOLATILE 9 /* 80 #define BTF_KIND_VOLATILE 9 /* Volatile */
81 #define BTF_KIND_CONST 10 /* 81 #define BTF_KIND_CONST 10 /* Const */
82 #define BTF_KIND_RESTRICT 11 /* 82 #define BTF_KIND_RESTRICT 11 /* Restrict */
83 #define BTF_KIND_FUNC 12 /* 83 #define BTF_KIND_FUNC 12 /* Function */
84 #define BTF_KIND_FUNC_PROTO 13 /* 84 #define BTF_KIND_FUNC_PROTO 13 /* Function Proto */
85 #define BTF_KIND_VAR 14 /* 85 #define BTF_KIND_VAR 14 /* Variable */
86 #define BTF_KIND_DATASEC 15 /* 86 #define BTF_KIND_DATASEC 15 /* Section */
87 #define BTF_KIND_FLOAT 16 /* 87 #define BTF_KIND_FLOAT 16 /* Floating point */
88 #define BTF_KIND_DECL_TAG 17 /* <<
89 #define BTF_KIND_TYPE_TAG 18 /* <<
90 #define BTF_KIND_ENUM64 19 /* <<
91 88
92 Note that the type section encodes debug info, 89 Note that the type section encodes debug info, not just pure types.
93 ``BTF_KIND_FUNC`` is not a type, and it repres 90 ``BTF_KIND_FUNC`` is not a type, and it represents a defined subprogram.
94 91
95 Each type contains the following common data:: 92 Each type contains the following common data::
96 93
97 struct btf_type { 94 struct btf_type {
98 __u32 name_off; 95 __u32 name_off;
99 /* "info" bits arrangement 96 /* "info" bits arrangement
100 * bits 0-15: vlen (e.g. # of struct' 97 * bits 0-15: vlen (e.g. # of struct's members)
101 * bits 16-23: unused 98 * bits 16-23: unused
102 * bits 24-28: kind (e.g. int, ptr, ar 99 * bits 24-28: kind (e.g. int, ptr, array...etc)
103 * bits 29-30: unused 100 * bits 29-30: unused
104 * bit 31: kind_flag, currently us 101 * bit 31: kind_flag, currently used by
105 * struct, union, fwd, enu !! 102 * struct, union and fwd
106 */ 103 */
107 __u32 info; 104 __u32 info;
108 /* "size" is used by INT, ENUM, STRUCT !! 105 /* "size" is used by INT, ENUM, STRUCT and UNION.
109 * "size" tells the size of the type i 106 * "size" tells the size of the type it is describing.
110 * 107 *
111 * "type" is used by PTR, TYPEDEF, VOL 108 * "type" is used by PTR, TYPEDEF, VOLATILE, CONST, RESTRICT,
112 * FUNC, FUNC_PROTO, DECL_TAG and TYPE !! 109 * FUNC and FUNC_PROTO.
113 * "type" is a type_id referring to an 110 * "type" is a type_id referring to another type.
114 */ 111 */
115 union { 112 union {
116 __u32 size; 113 __u32 size;
117 __u32 type; 114 __u32 type;
118 }; 115 };
119 }; 116 };
120 117
121 For certain kinds, the common data are followe 118 For certain kinds, the common data are followed by kind-specific data. The
122 ``name_off`` in ``struct btf_type`` specifies 119 ``name_off`` in ``struct btf_type`` specifies the offset in the string table.
123 The following sections detail encoding of each 120 The following sections detail encoding of each kind.
124 121
125 2.2.1 BTF_KIND_INT 122 2.2.1 BTF_KIND_INT
126 ~~~~~~~~~~~~~~~~~~ 123 ~~~~~~~~~~~~~~~~~~
127 124
128 ``struct btf_type`` encoding requirement: 125 ``struct btf_type`` encoding requirement:
129 * ``name_off``: any valid offset 126 * ``name_off``: any valid offset
130 * ``info.kind_flag``: 0 127 * ``info.kind_flag``: 0
131 * ``info.kind``: BTF_KIND_INT 128 * ``info.kind``: BTF_KIND_INT
132 * ``info.vlen``: 0 129 * ``info.vlen``: 0
133 * ``size``: the size of the int type in bytes 130 * ``size``: the size of the int type in bytes.
134 131
135 ``btf_type`` is followed by a ``u32`` with the 132 ``btf_type`` is followed by a ``u32`` with the following bits arrangement::
136 133
137 #define BTF_INT_ENCODING(VAL) (((VAL) & 0x 134 #define BTF_INT_ENCODING(VAL) (((VAL) & 0x0f000000) >> 24)
138 #define BTF_INT_OFFSET(VAL) (((VAL) & 0x 135 #define BTF_INT_OFFSET(VAL) (((VAL) & 0x00ff0000) >> 16)
139 #define BTF_INT_BITS(VAL) ((VAL) & 0x 136 #define BTF_INT_BITS(VAL) ((VAL) & 0x000000ff)
140 137
141 The ``BTF_INT_ENCODING`` has the following att 138 The ``BTF_INT_ENCODING`` has the following attributes::
142 139
143 #define BTF_INT_SIGNED (1 << 0) 140 #define BTF_INT_SIGNED (1 << 0)
144 #define BTF_INT_CHAR (1 << 1) 141 #define BTF_INT_CHAR (1 << 1)
145 #define BTF_INT_BOOL (1 << 2) 142 #define BTF_INT_BOOL (1 << 2)
146 143
147 The ``BTF_INT_ENCODING()`` provides extra info 144 The ``BTF_INT_ENCODING()`` provides extra information: signedness, char, or
148 bool, for the int type. The char and bool enco 145 bool, for the int type. The char and bool encoding are mostly useful for
149 pretty print. At most one encoding can be spec 146 pretty print. At most one encoding can be specified for the int type.
150 147
151 The ``BTF_INT_BITS()`` specifies the number of 148 The ``BTF_INT_BITS()`` specifies the number of actual bits held by this int
152 type. For example, a 4-bit bitfield encodes `` 149 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 150 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_B 151 for the type. The maximum value of ``BTF_INT_BITS()`` is 128.
155 152
156 The ``BTF_INT_OFFSET()`` specifies the startin 153 The ``BTF_INT_OFFSET()`` specifies the starting bit offset to calculate values
157 for this int. For example, a bitfield struct m 154 for this int. For example, a bitfield struct member has:
158 155
159 * btf member bit offset 100 from the start of 156 * btf member bit offset 100 from the start of the structure,
160 * btf member pointing to an int type, 157 * btf member pointing to an int type,
161 * the int type has ``BTF_INT_OFFSET() = 2`` a 158 * the int type has ``BTF_INT_OFFSET() = 2`` and ``BTF_INT_BITS() = 4``
162 159
163 Then in the struct memory layout, this member 160 Then in the struct memory layout, this member will occupy ``4`` bits starting
164 from bits ``100 + 2 = 102``. 161 from bits ``100 + 2 = 102``.
165 162
166 Alternatively, the bitfield struct member can 163 Alternatively, the bitfield struct member can be the following to access the
167 same bits as the above: 164 same bits as the above:
168 165
169 * btf member bit offset 102, 166 * btf member bit offset 102,
170 * btf member pointing to an int type, 167 * btf member pointing to an int type,
171 * the int type has ``BTF_INT_OFFSET() = 0`` a 168 * the int type has ``BTF_INT_OFFSET() = 0`` and ``BTF_INT_BITS() = 4``
172 169
173 The original intention of ``BTF_INT_OFFSET()`` 170 The original intention of ``BTF_INT_OFFSET()`` is to provide flexibility of
174 bitfield encoding. Currently, both llvm and pa 171 bitfield encoding. Currently, both llvm and pahole generate
175 ``BTF_INT_OFFSET() = 0`` for all int types. 172 ``BTF_INT_OFFSET() = 0`` for all int types.
176 173
177 2.2.2 BTF_KIND_PTR 174 2.2.2 BTF_KIND_PTR
178 ~~~~~~~~~~~~~~~~~~ 175 ~~~~~~~~~~~~~~~~~~
179 176
180 ``struct btf_type`` encoding requirement: 177 ``struct btf_type`` encoding requirement:
181 * ``name_off``: 0 178 * ``name_off``: 0
182 * ``info.kind_flag``: 0 179 * ``info.kind_flag``: 0
183 * ``info.kind``: BTF_KIND_PTR 180 * ``info.kind``: BTF_KIND_PTR
184 * ``info.vlen``: 0 181 * ``info.vlen``: 0
185 * ``type``: the pointee type of the pointer 182 * ``type``: the pointee type of the pointer
186 183
187 No additional type data follow ``btf_type``. 184 No additional type data follow ``btf_type``.
188 185
189 2.2.3 BTF_KIND_ARRAY 186 2.2.3 BTF_KIND_ARRAY
190 ~~~~~~~~~~~~~~~~~~~~ 187 ~~~~~~~~~~~~~~~~~~~~
191 188
192 ``struct btf_type`` encoding requirement: 189 ``struct btf_type`` encoding requirement:
193 * ``name_off``: 0 190 * ``name_off``: 0
194 * ``info.kind_flag``: 0 191 * ``info.kind_flag``: 0
195 * ``info.kind``: BTF_KIND_ARRAY 192 * ``info.kind``: BTF_KIND_ARRAY
196 * ``info.vlen``: 0 193 * ``info.vlen``: 0
197 * ``size/type``: 0, not used 194 * ``size/type``: 0, not used
198 195
199 ``btf_type`` is followed by one ``struct btf_a 196 ``btf_type`` is followed by one ``struct btf_array``::
200 197
201 struct btf_array { 198 struct btf_array {
202 __u32 type; 199 __u32 type;
203 __u32 index_type; 200 __u32 index_type;
204 __u32 nelems; 201 __u32 nelems;
205 }; 202 };
206 203
207 The ``struct btf_array`` encoding: 204 The ``struct btf_array`` encoding:
208 * ``type``: the element type 205 * ``type``: the element type
209 * ``index_type``: the index type 206 * ``index_type``: the index type
210 * ``nelems``: the number of elements for thi 207 * ``nelems``: the number of elements for this array (``0`` is also allowed).
211 208
212 The ``index_type`` can be any regular int type 209 The ``index_type`` can be any regular int type (``u8``, ``u16``, ``u32``,
213 ``u64``, ``unsigned __int128``). The original 210 ``u64``, ``unsigned __int128``). The original design of including
214 ``index_type`` follows DWARF, which has an ``i 211 ``index_type`` follows DWARF, which has an ``index_type`` for its array type.
215 Currently in BTF, beyond type verification, th 212 Currently in BTF, beyond type verification, the ``index_type`` is not used.
216 213
217 The ``struct btf_array`` allows chaining throu 214 The ``struct btf_array`` allows chaining through element type to represent
218 multidimensional arrays. For example, for ``in 215 multidimensional arrays. For example, for ``int a[5][6]``, the following type
219 information illustrates the chaining: 216 information illustrates the chaining:
220 217
221 * [1]: int 218 * [1]: int
222 * [2]: array, ``btf_array.type = [1]``, ``bt 219 * [2]: array, ``btf_array.type = [1]``, ``btf_array.nelems = 6``
223 * [3]: array, ``btf_array.type = [2]``, ``bt 220 * [3]: array, ``btf_array.type = [2]``, ``btf_array.nelems = 5``
224 221
225 Currently, both pahole and llvm collapse multi 222 Currently, both pahole and llvm collapse multidimensional array into
226 one-dimensional array, e.g., for ``a[5][6]``, 223 one-dimensional array, e.g., for ``a[5][6]``, the ``btf_array.nelems`` is
227 equal to ``30``. This is because the original 224 equal to ``30``. This is because the original use case is map pretty print
228 where the whole array is dumped out so one-dim 225 where the whole array is dumped out so one-dimensional array is enough. As
229 more BTF usage is explored, pahole and llvm ca 226 more BTF usage is explored, pahole and llvm can be changed to generate proper
230 chained representation for multidimensional ar 227 chained representation for multidimensional arrays.
231 228
232 2.2.4 BTF_KIND_STRUCT 229 2.2.4 BTF_KIND_STRUCT
233 ~~~~~~~~~~~~~~~~~~~~~ 230 ~~~~~~~~~~~~~~~~~~~~~
234 2.2.5 BTF_KIND_UNION 231 2.2.5 BTF_KIND_UNION
235 ~~~~~~~~~~~~~~~~~~~~ 232 ~~~~~~~~~~~~~~~~~~~~
236 233
237 ``struct btf_type`` encoding requirement: 234 ``struct btf_type`` encoding requirement:
238 * ``name_off``: 0 or offset to a valid C ide 235 * ``name_off``: 0 or offset to a valid C identifier
239 * ``info.kind_flag``: 0 or 1 236 * ``info.kind_flag``: 0 or 1
240 * ``info.kind``: BTF_KIND_STRUCT or BTF_KIND 237 * ``info.kind``: BTF_KIND_STRUCT or BTF_KIND_UNION
241 * ``info.vlen``: the number of struct/union 238 * ``info.vlen``: the number of struct/union members
242 * ``info.size``: the size of the struct/unio 239 * ``info.size``: the size of the struct/union in bytes
243 240
244 ``btf_type`` is followed by ``info.vlen`` numb 241 ``btf_type`` is followed by ``info.vlen`` number of ``struct btf_member``.::
245 242
246 struct btf_member { 243 struct btf_member {
247 __u32 name_off; 244 __u32 name_off;
248 __u32 type; 245 __u32 type;
249 __u32 offset; 246 __u32 offset;
250 }; 247 };
251 248
252 ``struct btf_member`` encoding: 249 ``struct btf_member`` encoding:
253 * ``name_off``: offset to a valid C identifi 250 * ``name_off``: offset to a valid C identifier
254 * ``type``: the member type 251 * ``type``: the member type
255 * ``offset``: <see below> 252 * ``offset``: <see below>
256 253
257 If the type info ``kind_flag`` is not set, the 254 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 255 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 typ 256 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 257 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. 258 ``BTF_INT_BITS()`` encodes the bitfield size.
262 259
263 If the ``kind_flag`` is set, the ``btf_member. 260 If the ``kind_flag`` is set, the ``btf_member.offset`` contains both member
264 bitfield size and bit offset. The bitfield siz 261 bitfield size and bit offset. The bitfield size and bit offset are calculated
265 as below.:: 262 as below.::
266 263
267 #define BTF_MEMBER_BITFIELD_SIZE(val) ((va 264 #define BTF_MEMBER_BITFIELD_SIZE(val) ((val) >> 24)
268 #define BTF_MEMBER_BIT_OFFSET(val) ((va 265 #define BTF_MEMBER_BIT_OFFSET(val) ((val) & 0xffffff)
269 266
270 In this case, if the base type is an int type, 267 In this case, if the base type is an int type, it must be a regular int type:
271 268
272 * ``BTF_INT_OFFSET()`` must be 0. 269 * ``BTF_INT_OFFSET()`` must be 0.
273 * ``BTF_INT_BITS()`` must be equal to ``{1,2 270 * ``BTF_INT_BITS()`` must be equal to ``{1,2,4,8,16} * 8``.
274 271
275 Commit 9d5f9f701b18 introduced ``kind_flag`` a !! 272 The following kernel patch introduced ``kind_flag`` and explained why both
276 exist. !! 273 modes exist:
>> 274
>> 275 https://github.com/torvalds/linux/commit/9d5f9f701b1891466fb3dbb1806ad97716f95cc3#diff-fa650a64fdd3968396883d2fe8215ff3
277 276
278 2.2.6 BTF_KIND_ENUM 277 2.2.6 BTF_KIND_ENUM
279 ~~~~~~~~~~~~~~~~~~~ 278 ~~~~~~~~~~~~~~~~~~~
280 279
281 ``struct btf_type`` encoding requirement: 280 ``struct btf_type`` encoding requirement:
282 * ``name_off``: 0 or offset to a valid C ide 281 * ``name_off``: 0 or offset to a valid C identifier
283 * ``info.kind_flag``: 0 for unsigned, 1 for !! 282 * ``info.kind_flag``: 0
284 * ``info.kind``: BTF_KIND_ENUM 283 * ``info.kind``: BTF_KIND_ENUM
285 * ``info.vlen``: number of enum values 284 * ``info.vlen``: number of enum values
286 * ``size``: 1/2/4/8 !! 285 * ``size``: 4
287 286
288 ``btf_type`` is followed by ``info.vlen`` numb 287 ``btf_type`` is followed by ``info.vlen`` number of ``struct btf_enum``.::
289 288
290 struct btf_enum { 289 struct btf_enum {
291 __u32 name_off; 290 __u32 name_off;
292 __s32 val; 291 __s32 val;
293 }; 292 };
294 293
295 The ``btf_enum`` encoding: 294 The ``btf_enum`` encoding:
296 * ``name_off``: offset to a valid C identifi 295 * ``name_off``: offset to a valid C identifier
297 * ``val``: any value 296 * ``val``: any value
298 297
299 If the original enum value is signed and the s <<
300 that value will be sign extended into 4 bytes. <<
301 the value will be truncated into 4 bytes. <<
302 <<
303 2.2.7 BTF_KIND_FWD 298 2.2.7 BTF_KIND_FWD
304 ~~~~~~~~~~~~~~~~~~ 299 ~~~~~~~~~~~~~~~~~~
305 300
306 ``struct btf_type`` encoding requirement: 301 ``struct btf_type`` encoding requirement:
307 * ``name_off``: offset to a valid C identifi 302 * ``name_off``: offset to a valid C identifier
308 * ``info.kind_flag``: 0 for struct, 1 for un 303 * ``info.kind_flag``: 0 for struct, 1 for union
309 * ``info.kind``: BTF_KIND_FWD 304 * ``info.kind``: BTF_KIND_FWD
310 * ``info.vlen``: 0 305 * ``info.vlen``: 0
311 * ``type``: 0 306 * ``type``: 0
312 307
313 No additional type data follow ``btf_type``. 308 No additional type data follow ``btf_type``.
314 309
315 2.2.8 BTF_KIND_TYPEDEF 310 2.2.8 BTF_KIND_TYPEDEF
316 ~~~~~~~~~~~~~~~~~~~~~~ 311 ~~~~~~~~~~~~~~~~~~~~~~
317 312
318 ``struct btf_type`` encoding requirement: 313 ``struct btf_type`` encoding requirement:
319 * ``name_off``: offset to a valid C identifi 314 * ``name_off``: offset to a valid C identifier
320 * ``info.kind_flag``: 0 315 * ``info.kind_flag``: 0
321 * ``info.kind``: BTF_KIND_TYPEDEF 316 * ``info.kind``: BTF_KIND_TYPEDEF
322 * ``info.vlen``: 0 317 * ``info.vlen``: 0
323 * ``type``: the type which can be referred b 318 * ``type``: the type which can be referred by name at ``name_off``
324 319
325 No additional type data follow ``btf_type``. 320 No additional type data follow ``btf_type``.
326 321
327 2.2.9 BTF_KIND_VOLATILE 322 2.2.9 BTF_KIND_VOLATILE
328 ~~~~~~~~~~~~~~~~~~~~~~~ 323 ~~~~~~~~~~~~~~~~~~~~~~~
329 324
330 ``struct btf_type`` encoding requirement: 325 ``struct btf_type`` encoding requirement:
331 * ``name_off``: 0 326 * ``name_off``: 0
332 * ``info.kind_flag``: 0 327 * ``info.kind_flag``: 0
333 * ``info.kind``: BTF_KIND_VOLATILE 328 * ``info.kind``: BTF_KIND_VOLATILE
334 * ``info.vlen``: 0 329 * ``info.vlen``: 0
335 * ``type``: the type with ``volatile`` quali 330 * ``type``: the type with ``volatile`` qualifier
336 331
337 No additional type data follow ``btf_type``. 332 No additional type data follow ``btf_type``.
338 333
339 2.2.10 BTF_KIND_CONST 334 2.2.10 BTF_KIND_CONST
340 ~~~~~~~~~~~~~~~~~~~~~ 335 ~~~~~~~~~~~~~~~~~~~~~
341 336
342 ``struct btf_type`` encoding requirement: 337 ``struct btf_type`` encoding requirement:
343 * ``name_off``: 0 338 * ``name_off``: 0
344 * ``info.kind_flag``: 0 339 * ``info.kind_flag``: 0
345 * ``info.kind``: BTF_KIND_CONST 340 * ``info.kind``: BTF_KIND_CONST
346 * ``info.vlen``: 0 341 * ``info.vlen``: 0
347 * ``type``: the type with ``const`` qualifie 342 * ``type``: the type with ``const`` qualifier
348 343
349 No additional type data follow ``btf_type``. 344 No additional type data follow ``btf_type``.
350 345
351 2.2.11 BTF_KIND_RESTRICT 346 2.2.11 BTF_KIND_RESTRICT
352 ~~~~~~~~~~~~~~~~~~~~~~~~ 347 ~~~~~~~~~~~~~~~~~~~~~~~~
353 348
354 ``struct btf_type`` encoding requirement: 349 ``struct btf_type`` encoding requirement:
355 * ``name_off``: 0 350 * ``name_off``: 0
356 * ``info.kind_flag``: 0 351 * ``info.kind_flag``: 0
357 * ``info.kind``: BTF_KIND_RESTRICT 352 * ``info.kind``: BTF_KIND_RESTRICT
358 * ``info.vlen``: 0 353 * ``info.vlen``: 0
359 * ``type``: the type with ``restrict`` quali 354 * ``type``: the type with ``restrict`` qualifier
360 355
361 No additional type data follow ``btf_type``. 356 No additional type data follow ``btf_type``.
362 357
363 2.2.12 BTF_KIND_FUNC 358 2.2.12 BTF_KIND_FUNC
364 ~~~~~~~~~~~~~~~~~~~~ 359 ~~~~~~~~~~~~~~~~~~~~
365 360
366 ``struct btf_type`` encoding requirement: 361 ``struct btf_type`` encoding requirement:
367 * ``name_off``: offset to a valid C identifi 362 * ``name_off``: offset to a valid C identifier
368 * ``info.kind_flag``: 0 363 * ``info.kind_flag``: 0
369 * ``info.kind``: BTF_KIND_FUNC 364 * ``info.kind``: BTF_KIND_FUNC
370 * ``info.vlen``: linkage information (BTF_FU !! 365 * ``info.vlen``: 0
371 or BTF_FUNC_EXTERN - see :r <<
372 * ``type``: a BTF_KIND_FUNC_PROTO type 366 * ``type``: a BTF_KIND_FUNC_PROTO type
373 367
374 No additional type data follow ``btf_type``. 368 No additional type data follow ``btf_type``.
375 369
376 A BTF_KIND_FUNC defines not a type, but a subp 370 A BTF_KIND_FUNC defines not a type, but a subprogram (function) whose
377 signature is defined by ``type``. The subprogr 371 signature is defined by ``type``. The subprogram is thus an instance of that
378 type. The BTF_KIND_FUNC may in turn be referen 372 type. The BTF_KIND_FUNC may in turn be referenced by a func_info in the
379 :ref:`BTF_Ext_Section` (ELF) or in the argumen 373 :ref:`BTF_Ext_Section` (ELF) or in the arguments to :ref:`BPF_Prog_Load`
380 (ABI). 374 (ABI).
381 375
382 Currently, only linkage values of BTF_FUNC_STA <<
383 supported in the kernel. <<
384 <<
385 2.2.13 BTF_KIND_FUNC_PROTO 376 2.2.13 BTF_KIND_FUNC_PROTO
386 ~~~~~~~~~~~~~~~~~~~~~~~~~~ 377 ~~~~~~~~~~~~~~~~~~~~~~~~~~
387 378
388 ``struct btf_type`` encoding requirement: 379 ``struct btf_type`` encoding requirement:
389 * ``name_off``: 0 380 * ``name_off``: 0
390 * ``info.kind_flag``: 0 381 * ``info.kind_flag``: 0
391 * ``info.kind``: BTF_KIND_FUNC_PROTO 382 * ``info.kind``: BTF_KIND_FUNC_PROTO
392 * ``info.vlen``: # of parameters 383 * ``info.vlen``: # of parameters
393 * ``type``: the return type 384 * ``type``: the return type
394 385
395 ``btf_type`` is followed by ``info.vlen`` numb 386 ``btf_type`` is followed by ``info.vlen`` number of ``struct btf_param``.::
396 387
397 struct btf_param { 388 struct btf_param {
398 __u32 name_off; 389 __u32 name_off;
399 __u32 type; 390 __u32 type;
400 }; 391 };
401 392
402 If a BTF_KIND_FUNC_PROTO type is referred by a 393 If a BTF_KIND_FUNC_PROTO type is referred by a BTF_KIND_FUNC type, then
403 ``btf_param.name_off`` must point to a valid C 394 ``btf_param.name_off`` must point to a valid C identifier except for the
404 possible last argument representing the variab 395 possible last argument representing the variable argument. The btf_param.type
405 refers to parameter type. 396 refers to parameter type.
406 397
407 If the function has variable arguments, the la 398 If the function has variable arguments, the last parameter is encoded with
408 ``name_off = 0`` and ``type = 0``. 399 ``name_off = 0`` and ``type = 0``.
409 400
410 2.2.14 BTF_KIND_VAR 401 2.2.14 BTF_KIND_VAR
411 ~~~~~~~~~~~~~~~~~~~ 402 ~~~~~~~~~~~~~~~~~~~
412 403
413 ``struct btf_type`` encoding requirement: 404 ``struct btf_type`` encoding requirement:
414 * ``name_off``: offset to a valid C identifi 405 * ``name_off``: offset to a valid C identifier
415 * ``info.kind_flag``: 0 406 * ``info.kind_flag``: 0
416 * ``info.kind``: BTF_KIND_VAR 407 * ``info.kind``: BTF_KIND_VAR
417 * ``info.vlen``: 0 408 * ``info.vlen``: 0
418 * ``type``: the type of the variable 409 * ``type``: the type of the variable
419 410
420 ``btf_type`` is followed by a single ``struct 411 ``btf_type`` is followed by a single ``struct btf_variable`` with the
421 following data:: 412 following data::
422 413
423 struct btf_var { 414 struct btf_var {
424 __u32 linkage; 415 __u32 linkage;
425 }; 416 };
426 417
427 ``btf_var.linkage`` may take the values: BTF_V !! 418 ``struct btf_var`` encoding:
428 see :ref:`BTF_Var_Linkage_Constants`. !! 419 * ``linkage``: currently only static variable 0, or globally allocated
>> 420 variable in ELF sections 1
429 421
430 Not all type of global variables are supported 422 Not all type of global variables are supported by LLVM at this point.
431 The following is currently available: 423 The following is currently available:
432 424
433 * static variables with or without section a 425 * static variables with or without section attributes
434 * global variables with section attributes 426 * global variables with section attributes
435 427
436 The latter is for future extraction of map key 428 The latter is for future extraction of map key/value type id's from a
437 map definition. 429 map definition.
438 430
439 2.2.15 BTF_KIND_DATASEC 431 2.2.15 BTF_KIND_DATASEC
440 ~~~~~~~~~~~~~~~~~~~~~~~ 432 ~~~~~~~~~~~~~~~~~~~~~~~
441 433
442 ``struct btf_type`` encoding requirement: 434 ``struct btf_type`` encoding requirement:
443 * ``name_off``: offset to a valid name assoc 435 * ``name_off``: offset to a valid name associated with a variable or
444 one of .data/.bss/.rodata 436 one of .data/.bss/.rodata
445 * ``info.kind_flag``: 0 437 * ``info.kind_flag``: 0
446 * ``info.kind``: BTF_KIND_DATASEC 438 * ``info.kind``: BTF_KIND_DATASEC
447 * ``info.vlen``: # of variables 439 * ``info.vlen``: # of variables
448 * ``size``: total section size in bytes (0 a 440 * ``size``: total section size in bytes (0 at compilation time, patched
449 to actual size by BPF loaders su 441 to actual size by BPF loaders such as libbpf)
450 442
451 ``btf_type`` is followed by ``info.vlen`` numb 443 ``btf_type`` is followed by ``info.vlen`` number of ``struct btf_var_secinfo``.::
452 444
453 struct btf_var_secinfo { 445 struct btf_var_secinfo {
454 __u32 type; 446 __u32 type;
455 __u32 offset; 447 __u32 offset;
456 __u32 size; 448 __u32 size;
457 }; 449 };
458 450
459 ``struct btf_var_secinfo`` encoding: 451 ``struct btf_var_secinfo`` encoding:
460 * ``type``: the type of the BTF_KIND_VAR var 452 * ``type``: the type of the BTF_KIND_VAR variable
461 * ``offset``: the in-section offset of the v 453 * ``offset``: the in-section offset of the variable
462 * ``size``: the size of the variable in byte 454 * ``size``: the size of the variable in bytes
463 455
464 2.2.16 BTF_KIND_FLOAT 456 2.2.16 BTF_KIND_FLOAT
465 ~~~~~~~~~~~~~~~~~~~~~ 457 ~~~~~~~~~~~~~~~~~~~~~
466 458
467 ``struct btf_type`` encoding requirement: 459 ``struct btf_type`` encoding requirement:
468 * ``name_off``: any valid offset 460 * ``name_off``: any valid offset
469 * ``info.kind_flag``: 0 461 * ``info.kind_flag``: 0
470 * ``info.kind``: BTF_KIND_FLOAT 462 * ``info.kind``: BTF_KIND_FLOAT
471 * ``info.vlen``: 0 463 * ``info.vlen``: 0
472 * ``size``: the size of the float type in byt 464 * ``size``: the size of the float type in bytes: 2, 4, 8, 12 or 16.
473 465
474 No additional type data follow ``btf_type``. 466 No additional type data follow ``btf_type``.
475 467
476 2.2.17 BTF_KIND_DECL_TAG <<
477 ~~~~~~~~~~~~~~~~~~~~~~~~ <<
478 <<
479 ``struct btf_type`` encoding requirement: <<
480 * ``name_off``: offset to a non-empty string <<
481 * ``info.kind_flag``: 0 <<
482 * ``info.kind``: BTF_KIND_DECL_TAG <<
483 * ``info.vlen``: 0 <<
484 * ``type``: ``struct``, ``union``, ``func``, <<
485 <<
486 ``btf_type`` is followed by ``struct btf_decl_ <<
487 <<
488 struct btf_decl_tag { <<
489 __u32 component_idx; <<
490 }; <<
491 <<
492 The ``name_off`` encodes btf_decl_tag attribut <<
493 The ``type`` should be ``struct``, ``union``, <<
494 For ``var`` or ``typedef`` type, ``btf_decl_ta <<
495 For the other three types, if the btf_decl_tag <<
496 applied to the ``struct``, ``union`` or ``func <<
497 ``btf_decl_tag.component_idx`` must be ``-1``. <<
498 the attribute is applied to a ``struct``/``uni <<
499 a ``func`` argument, and ``btf_decl_tag.compon <<
500 valid index (starting from 0) pointing to a me <<
501 <<
502 2.2.18 BTF_KIND_TYPE_TAG <<
503 ~~~~~~~~~~~~~~~~~~~~~~~~ <<
504 <<
505 ``struct btf_type`` encoding requirement: <<
506 * ``name_off``: offset to a non-empty string <<
507 * ``info.kind_flag``: 0 <<
508 * ``info.kind``: BTF_KIND_TYPE_TAG <<
509 * ``info.vlen``: 0 <<
510 * ``type``: the type with ``btf_type_tag`` at <<
511 <<
512 Currently, ``BTF_KIND_TYPE_TAG`` is only emitt <<
513 It has the following btf type chain: <<
514 :: <<
515 <<
516 ptr -> [type_tag]* <<
517 -> [const | volatile | restrict | typede <<
518 -> base_type <<
519 <<
520 Basically, a pointer type points to zero or mo <<
521 type_tag, then zero or more const/volatile/res <<
522 and finally the base type. The base type is on <<
523 int, ptr, array, struct, union, enum, func_pro <<
524 <<
525 2.2.19 BTF_KIND_ENUM64 <<
526 ~~~~~~~~~~~~~~~~~~~~~~ <<
527 <<
528 ``struct btf_type`` encoding requirement: <<
529 * ``name_off``: 0 or offset to a valid C ide <<
530 * ``info.kind_flag``: 0 for unsigned, 1 for <<
531 * ``info.kind``: BTF_KIND_ENUM64 <<
532 * ``info.vlen``: number of enum values <<
533 * ``size``: 1/2/4/8 <<
534 <<
535 ``btf_type`` is followed by ``info.vlen`` numb <<
536 <<
537 struct btf_enum64 { <<
538 __u32 name_off; <<
539 __u32 val_lo32; <<
540 __u32 val_hi32; <<
541 }; <<
542 <<
543 The ``btf_enum64`` encoding: <<
544 * ``name_off``: offset to a valid C identifi <<
545 * ``val_lo32``: lower 32-bit value for a 64- <<
546 * ``val_hi32``: high 32-bit value for a 64-b <<
547 <<
548 If the original enum value is signed and the s <<
549 that value will be sign extended into 8 bytes. <<
550 <<
551 2.3 Constant Values <<
552 ------------------- <<
553 <<
554 .. _BTF_Function_Linkage_Constants: <<
555 <<
556 2.3.1 Function Linkage Constant Values <<
557 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ <<
558 .. table:: Function Linkage Values and Meaning <<
559 <<
560 =================== ===== =========== <<
561 kind value description <<
562 =================== ===== =========== <<
563 ``BTF_FUNC_STATIC`` 0x0 definition of su <<
564 ``BTF_FUNC_GLOBAL`` 0x1 definition of su <<
565 ``BTF_FUNC_EXTERN`` 0x2 declaration of a <<
566 =================== ===== =========== <<
567 <<
568 <<
569 .. _BTF_Var_Linkage_Constants: <<
570 <<
571 2.3.2 Variable Linkage Constant Values <<
572 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ <<
573 .. table:: Variable Linkage Values and Meaning <<
574 <<
575 ============================ ===== ======= <<
576 kind value descrip <<
577 ============================ ===== ======= <<
578 ``BTF_VAR_STATIC`` 0x0 definit <<
579 ``BTF_VAR_GLOBAL_ALLOCATED`` 0x1 definit <<
580 ``BTF_VAR_GLOBAL_EXTERN`` 0x2 declara <<
581 ============================ ===== ======= <<
582 <<
583 3. BTF Kernel API 468 3. BTF Kernel API
584 ================= !! 469 *****************
585 470
586 The following bpf syscall command involves BTF 471 The following bpf syscall command involves BTF:
587 * BPF_BTF_LOAD: load a blob of BTF data int 472 * BPF_BTF_LOAD: load a blob of BTF data into kernel
588 * BPF_MAP_CREATE: map creation with btf key 473 * BPF_MAP_CREATE: map creation with btf key and value type info.
589 * BPF_PROG_LOAD: prog load with btf functio 474 * BPF_PROG_LOAD: prog load with btf function and line info.
590 * BPF_BTF_GET_FD_BY_ID: get a btf fd 475 * BPF_BTF_GET_FD_BY_ID: get a btf fd
591 * BPF_OBJ_GET_INFO_BY_FD: btf, func_info, l 476 * BPF_OBJ_GET_INFO_BY_FD: btf, func_info, line_info
592 and other btf related info are returned. 477 and other btf related info are returned.
593 478
594 The workflow typically looks like: 479 The workflow typically looks like:
595 :: 480 ::
596 481
597 Application: 482 Application:
598 BPF_BTF_LOAD 483 BPF_BTF_LOAD
599 | 484 |
600 v 485 v
601 BPF_MAP_CREATE and BPF_PROG_LOAD 486 BPF_MAP_CREATE and BPF_PROG_LOAD
602 | 487 |
603 V 488 V
604 ...... 489 ......
605 490
606 Introspection tool: 491 Introspection tool:
607 ...... 492 ......
608 BPF_{PROG,MAP}_GET_NEXT_ID (get prog/map 493 BPF_{PROG,MAP}_GET_NEXT_ID (get prog/map id's)
609 | 494 |
610 V 495 V
611 BPF_{PROG,MAP}_GET_FD_BY_ID (get a prog/ 496 BPF_{PROG,MAP}_GET_FD_BY_ID (get a prog/map fd)
612 | 497 |
613 V 498 V
614 BPF_OBJ_GET_INFO_BY_FD (get bpf_prog_inf 499 BPF_OBJ_GET_INFO_BY_FD (get bpf_prog_info/bpf_map_info with btf_id)
615 | 500 | |
616 V 501 V |
617 BPF_BTF_GET_FD_BY_ID (get btf_fd) 502 BPF_BTF_GET_FD_BY_ID (get btf_fd) |
618 | 503 | |
619 V 504 V |
620 BPF_OBJ_GET_INFO_BY_FD (get btf) 505 BPF_OBJ_GET_INFO_BY_FD (get btf) |
621 | 506 | |
622 V 507 V V
623 pretty print types, dump func signatures 508 pretty print types, dump func signatures and line info, etc.
624 509
625 510
626 3.1 BPF_BTF_LOAD 511 3.1 BPF_BTF_LOAD
627 ---------------- !! 512 ================
628 513
629 Load a blob of BTF data into kernel. A blob of 514 Load a blob of BTF data into kernel. A blob of data, described in
630 :ref:`BTF_Type_String`, can be directly loaded 515 :ref:`BTF_Type_String`, can be directly loaded into the kernel. A ``btf_fd``
631 is returned to a userspace. 516 is returned to a userspace.
632 517
633 3.2 BPF_MAP_CREATE 518 3.2 BPF_MAP_CREATE
634 ------------------ !! 519 ==================
635 520
636 A map can be created with ``btf_fd`` and speci 521 A map can be created with ``btf_fd`` and specified key/value type id.::
637 522
638 __u32 btf_fd; /* fd pointing to 523 __u32 btf_fd; /* fd pointing to a BTF type data */
639 __u32 btf_key_type_id; /* BTF typ 524 __u32 btf_key_type_id; /* BTF type_id of the key */
640 __u32 btf_value_type_id; /* BTF typ 525 __u32 btf_value_type_id; /* BTF type_id of the value */
641 526
642 In libbpf, the map can be defined with extra a 527 In libbpf, the map can be defined with extra annotation like below:
643 :: 528 ::
644 529
645 struct { !! 530 struct bpf_map_def SEC("maps") btf_map = {
646 __uint(type, BPF_MAP_TYPE_ARRAY); !! 531 .type = BPF_MAP_TYPE_ARRAY,
647 __type(key, int); !! 532 .key_size = sizeof(int),
648 __type(value, struct ipv_counts); !! 533 .value_size = sizeof(struct ipv_counts),
649 __uint(max_entries, 4); !! 534 .max_entries = 4,
650 } btf_map SEC(".maps"); !! 535 };
651 !! 536 BPF_ANNOTATE_KV_PAIR(btf_map, int, struct ipv_counts);
652 During ELF parsing, libbpf is able to extract !! 537
653 them to BPF_MAP_CREATE attributes automaticall !! 538 Here, the parameters for macro BPF_ANNOTATE_KV_PAIR are map name, key and
>> 539 value types for the map. During ELF parsing, libbpf is able to extract
>> 540 key/value type_id's and assign them to BPF_MAP_CREATE attributes
>> 541 automatically.
654 542
655 .. _BPF_Prog_Load: 543 .. _BPF_Prog_Load:
656 544
657 3.3 BPF_PROG_LOAD 545 3.3 BPF_PROG_LOAD
658 ----------------- !! 546 =================
659 547
660 During prog_load, func_info and line_info can 548 During prog_load, func_info and line_info can be passed to kernel with proper
661 values for the following attributes: 549 values for the following attributes:
662 :: 550 ::
663 551
664 __u32 insn_cnt; 552 __u32 insn_cnt;
665 __aligned_u64 insns; 553 __aligned_u64 insns;
666 ...... 554 ......
667 __u32 prog_btf_fd; /* fd poin 555 __u32 prog_btf_fd; /* fd pointing to BTF type data */
668 __u32 func_info_rec_size; /* 556 __u32 func_info_rec_size; /* userspace bpf_func_info size */
669 __aligned_u64 func_info; /* func in 557 __aligned_u64 func_info; /* func info */
670 __u32 func_info_cnt; /* number 558 __u32 func_info_cnt; /* number of bpf_func_info records */
671 __u32 line_info_rec_size; /* 559 __u32 line_info_rec_size; /* userspace bpf_line_info size */
672 __aligned_u64 line_info; /* line in 560 __aligned_u64 line_info; /* line info */
673 __u32 line_info_cnt; /* number 561 __u32 line_info_cnt; /* number of bpf_line_info records */
674 562
675 The func_info and line_info are an array of be 563 The func_info and line_info are an array of below, respectively.::
676 564
677 struct bpf_func_info { 565 struct bpf_func_info {
678 __u32 insn_off; /* [0, insn_cnt - 1] 566 __u32 insn_off; /* [0, insn_cnt - 1] */
679 __u32 type_id; /* pointing to a BTF 567 __u32 type_id; /* pointing to a BTF_KIND_FUNC type */
680 }; 568 };
681 struct bpf_line_info { 569 struct bpf_line_info {
682 __u32 insn_off; /* [0, insn_cnt - 1] 570 __u32 insn_off; /* [0, insn_cnt - 1] */
683 __u32 file_name_off; /* offset to st 571 __u32 file_name_off; /* offset to string table for the filename */
684 __u32 line_off; /* offset to string 572 __u32 line_off; /* offset to string table for the source line */
685 __u32 line_col; /* line number and c 573 __u32 line_col; /* line number and column number */
686 }; 574 };
687 575
688 func_info_rec_size is the size of each func_in 576 func_info_rec_size is the size of each func_info record, and
689 line_info_rec_size is the size of each line_in 577 line_info_rec_size is the size of each line_info record. Passing the record
690 size to kernel make it possible to extend the 578 size to kernel make it possible to extend the record itself in the future.
691 579
692 Below are requirements for func_info: 580 Below are requirements for func_info:
693 * func_info[0].insn_off must be 0. 581 * func_info[0].insn_off must be 0.
694 * the func_info insn_off is in strictly incr 582 * the func_info insn_off is in strictly increasing order and matches
695 bpf func boundaries. 583 bpf func boundaries.
696 584
697 Below are requirements for line_info: 585 Below are requirements for line_info:
698 * the first insn in each func must have a li 586 * the first insn in each func must have a line_info record pointing to it.
699 * the line_info insn_off is in strictly incr 587 * the line_info insn_off is in strictly increasing order.
700 588
701 For line_info, the line number and column numb 589 For line_info, the line number and column number are defined as below:
702 :: 590 ::
703 591
704 #define BPF_LINE_INFO_LINE_NUM(line_col) 592 #define BPF_LINE_INFO_LINE_NUM(line_col) ((line_col) >> 10)
705 #define BPF_LINE_INFO_LINE_COL(line_col) 593 #define BPF_LINE_INFO_LINE_COL(line_col) ((line_col) & 0x3ff)
706 594
707 3.4 BPF_{PROG,MAP}_GET_NEXT_ID 595 3.4 BPF_{PROG,MAP}_GET_NEXT_ID
708 ------------------------------ !! 596 ==============================
709 597
710 In kernel, every loaded program, map or btf ha 598 In kernel, every loaded program, map or btf has a unique id. The id won't
711 change during the lifetime of a program, map, 599 change during the lifetime of a program, map, or btf.
712 600
713 The bpf syscall command BPF_{PROG,MAP}_GET_NEX 601 The bpf syscall command BPF_{PROG,MAP}_GET_NEXT_ID returns all id's, one for
714 each command, to user space, for bpf program o 602 each command, to user space, for bpf program or maps, respectively, so an
715 inspection tool can inspect all programs and m 603 inspection tool can inspect all programs and maps.
716 604
717 3.5 BPF_{PROG,MAP}_GET_FD_BY_ID 605 3.5 BPF_{PROG,MAP}_GET_FD_BY_ID
718 ------------------------------- !! 606 ===============================
719 607
720 An introspection tool cannot use id to get det 608 An introspection tool cannot use id to get details about program or maps.
721 A file descriptor needs to be obtained first f 609 A file descriptor needs to be obtained first for reference-counting purpose.
722 610
723 3.6 BPF_OBJ_GET_INFO_BY_FD 611 3.6 BPF_OBJ_GET_INFO_BY_FD
724 -------------------------- !! 612 ==========================
725 613
726 Once a program/map fd is acquired, an introspe 614 Once a program/map fd is acquired, an introspection tool can get the detailed
727 information from kernel about this fd, some of 615 information from kernel about this fd, some of which are BTF-related. For
728 example, ``bpf_map_info`` returns ``btf_id`` a 616 example, ``bpf_map_info`` returns ``btf_id`` and key/value type ids.
729 ``bpf_prog_info`` returns ``btf_id``, func_inf 617 ``bpf_prog_info`` returns ``btf_id``, func_info, and line info for translated
730 bpf byte codes, and jited_line_info. 618 bpf byte codes, and jited_line_info.
731 619
732 3.7 BPF_BTF_GET_FD_BY_ID 620 3.7 BPF_BTF_GET_FD_BY_ID
733 ------------------------ !! 621 ========================
734 622
735 With ``btf_id`` obtained in ``bpf_map_info`` a 623 With ``btf_id`` obtained in ``bpf_map_info`` and ``bpf_prog_info``, bpf
736 syscall command BPF_BTF_GET_FD_BY_ID can retri 624 syscall command BPF_BTF_GET_FD_BY_ID can retrieve a btf fd. Then, with
737 command BPF_OBJ_GET_INFO_BY_FD, the btf blob, 625 command BPF_OBJ_GET_INFO_BY_FD, the btf blob, originally loaded into the
738 kernel with BPF_BTF_LOAD, can be retrieved. 626 kernel with BPF_BTF_LOAD, can be retrieved.
739 627
740 With the btf blob, ``bpf_map_info``, and ``bpf 628 With the btf blob, ``bpf_map_info``, and ``bpf_prog_info``, an introspection
741 tool has full btf knowledge and is able to pre 629 tool has full btf knowledge and is able to pretty print map key/values, dump
742 func signatures and line info, along with byte 630 func signatures and line info, along with byte/jit codes.
743 631
744 4. ELF File Format Interface 632 4. ELF File Format Interface
745 ============================ !! 633 ****************************
746 634
747 4.1 .BTF section 635 4.1 .BTF section
748 ---------------- !! 636 ================
749 637
750 The .BTF section contains type and string data 638 The .BTF section contains type and string data. The format of this section is
751 same as the one describe in :ref:`BTF_Type_Str 639 same as the one describe in :ref:`BTF_Type_String`.
752 640
753 .. _BTF_Ext_Section: 641 .. _BTF_Ext_Section:
754 642
755 4.2 .BTF.ext section 643 4.2 .BTF.ext section
756 -------------------- !! 644 ====================
757 645
758 The .BTF.ext section encodes func_info, line_i !! 646 The .BTF.ext section encodes func_info and line_info which needs loader
759 which needs loader manipulation before loading !! 647 manipulation before loading into the kernel.
760 648
761 The specification for .BTF.ext section is defi 649 The specification for .BTF.ext section is defined at ``tools/lib/bpf/btf.h``
762 and ``tools/lib/bpf/btf.c``. 650 and ``tools/lib/bpf/btf.c``.
763 651
764 The current header of .BTF.ext section:: 652 The current header of .BTF.ext section::
765 653
766 struct btf_ext_header { 654 struct btf_ext_header {
767 __u16 magic; 655 __u16 magic;
768 __u8 version; 656 __u8 version;
769 __u8 flags; 657 __u8 flags;
770 __u32 hdr_len; 658 __u32 hdr_len;
771 659
772 /* All offsets are in bytes relative t 660 /* All offsets are in bytes relative to the end of this header */
773 __u32 func_info_off; 661 __u32 func_info_off;
774 __u32 func_info_len; 662 __u32 func_info_len;
775 __u32 line_info_off; 663 __u32 line_info_off;
776 __u32 line_info_len; 664 __u32 line_info_len;
777 <<
778 /* optional part of .BTF.ext header */ <<
779 __u32 core_relo_off; <<
780 __u32 core_relo_len; <<
781 }; 665 };
782 666
783 It is very similar to .BTF section. Instead of 667 It is very similar to .BTF section. Instead of type/string section, it
784 contains func_info, line_info and core_relo su !! 668 contains func_info and line_info section. See :ref:`BPF_Prog_Load` for details
785 See :ref:`BPF_Prog_Load` for details about fun !! 669 about func_info and line_info record format.
786 record format. <<
787 670
788 The func_info is organized as below.:: 671 The func_info is organized as below.::
789 672
790 func_info_rec_size /* __u32 !! 673 func_info_rec_size
791 btf_ext_info_sec for section #1 /* func_i 674 btf_ext_info_sec for section #1 /* func_info for section #1 */
792 btf_ext_info_sec for section #2 /* func_i 675 btf_ext_info_sec for section #2 /* func_info for section #2 */
793 ... 676 ...
794 677
795 ``func_info_rec_size`` specifies the size of ` 678 ``func_info_rec_size`` specifies the size of ``bpf_func_info`` structure when
796 .BTF.ext is generated. ``btf_ext_info_sec``, d 679 .BTF.ext is generated. ``btf_ext_info_sec``, defined below, is a collection of
797 func_info for each specific ELF section.:: 680 func_info for each specific ELF section.::
798 681
799 struct btf_ext_info_sec { 682 struct btf_ext_info_sec {
800 __u32 sec_name_off; /* offset to sec 683 __u32 sec_name_off; /* offset to section name */
801 __u32 num_info; 684 __u32 num_info;
802 /* Followed by num_info * record_size 685 /* Followed by num_info * record_size number of bytes */
803 __u8 data[0]; 686 __u8 data[0];
804 }; 687 };
805 688
806 Here, num_info must be greater than 0. 689 Here, num_info must be greater than 0.
807 690
808 The line_info is organized as below.:: 691 The line_info is organized as below.::
809 692
810 line_info_rec_size /* __u32 !! 693 line_info_rec_size
811 btf_ext_info_sec for section #1 /* line_i 694 btf_ext_info_sec for section #1 /* line_info for section #1 */
812 btf_ext_info_sec for section #2 /* line_i 695 btf_ext_info_sec for section #2 /* line_info for section #2 */
813 ... 696 ...
814 697
815 ``line_info_rec_size`` specifies the size of ` 698 ``line_info_rec_size`` specifies the size of ``bpf_line_info`` structure when
816 .BTF.ext is generated. 699 .BTF.ext is generated.
817 700
818 The interpretation of ``bpf_func_info->insn_of 701 The interpretation of ``bpf_func_info->insn_off`` and
819 ``bpf_line_info->insn_off`` is different betwe 702 ``bpf_line_info->insn_off`` is different between kernel API and ELF API. For
820 kernel API, the ``insn_off`` is the instructio 703 kernel API, the ``insn_off`` is the instruction offset in the unit of ``struct
821 bpf_insn``. For ELF API, the ``insn_off`` is t 704 bpf_insn``. For ELF API, the ``insn_off`` is the byte offset from the
822 beginning of section (``btf_ext_info_sec->sec_ 705 beginning of section (``btf_ext_info_sec->sec_name_off``).
823 706
824 The core_relo is organized as below.:: <<
825 <<
826 core_relo_rec_size /* __u32 <<
827 btf_ext_info_sec for section #1 /* core_r <<
828 btf_ext_info_sec for section #2 /* core_r <<
829 <<
830 ``core_relo_rec_size`` specifies the size of ` <<
831 structure when .BTF.ext is generated. All ``bp <<
832 within a single ``btf_ext_info_sec`` describe <<
833 section named by ``btf_ext_info_sec->sec_name_ <<
834 <<
835 See :ref:`Documentation/bpf/llvm_reloc.rst <bt <<
836 for more information on CO-RE relocations. <<
837 <<
838 4.2 .BTF_ids section 707 4.2 .BTF_ids section
839 -------------------- !! 708 ====================
840 709
841 The .BTF_ids section encodes BTF ID values tha 710 The .BTF_ids section encodes BTF ID values that are used within the kernel.
842 711
843 This section is created during the kernel comp 712 This section is created during the kernel compilation with the help of
844 macros defined in ``include/linux/btf_ids.h`` 713 macros defined in ``include/linux/btf_ids.h`` header file. Kernel code can
845 use them to create lists and sets (sorted list 714 use them to create lists and sets (sorted lists) of BTF ID values.
846 715
847 The ``BTF_ID_LIST`` and ``BTF_ID`` macros defi 716 The ``BTF_ID_LIST`` and ``BTF_ID`` macros define unsorted list of BTF ID values,
848 with following syntax:: 717 with following syntax::
849 718
850 BTF_ID_LIST(list) 719 BTF_ID_LIST(list)
851 BTF_ID(type1, name1) 720 BTF_ID(type1, name1)
852 BTF_ID(type2, name2) 721 BTF_ID(type2, name2)
853 722
854 resulting in following layout in .BTF_ids sect 723 resulting in following layout in .BTF_ids section::
855 724
856 __BTF_ID__type1__name1__1: 725 __BTF_ID__type1__name1__1:
857 .zero 4 726 .zero 4
858 __BTF_ID__type2__name2__2: 727 __BTF_ID__type2__name2__2:
859 .zero 4 728 .zero 4
860 729
861 The ``u32 list[];`` variable is defined to acc 730 The ``u32 list[];`` variable is defined to access the list.
862 731
863 The ``BTF_ID_UNUSED`` macro defines 4 zero byt 732 The ``BTF_ID_UNUSED`` macro defines 4 zero bytes. It's used when we
864 want to define unused entry in BTF_ID_LIST, li 733 want to define unused entry in BTF_ID_LIST, like::
865 734
866 BTF_ID_LIST(bpf_skb_output_btf_ids) 735 BTF_ID_LIST(bpf_skb_output_btf_ids)
867 BTF_ID(struct, sk_buff) 736 BTF_ID(struct, sk_buff)
868 BTF_ID_UNUSED 737 BTF_ID_UNUSED
869 BTF_ID(struct, task_struct) 738 BTF_ID(struct, task_struct)
870 739
871 The ``BTF_SET_START/END`` macros pair defines 740 The ``BTF_SET_START/END`` macros pair defines sorted list of BTF ID values
872 and their count, with following syntax:: 741 and their count, with following syntax::
873 742
874 BTF_SET_START(set) 743 BTF_SET_START(set)
875 BTF_ID(type1, name1) 744 BTF_ID(type1, name1)
876 BTF_ID(type2, name2) 745 BTF_ID(type2, name2)
877 BTF_SET_END(set) 746 BTF_SET_END(set)
878 747
879 resulting in following layout in .BTF_ids sect 748 resulting in following layout in .BTF_ids section::
880 749
881 __BTF_ID__set__set: 750 __BTF_ID__set__set:
882 .zero 4 751 .zero 4
883 __BTF_ID__type1__name1__3: 752 __BTF_ID__type1__name1__3:
884 .zero 4 753 .zero 4
885 __BTF_ID__type2__name2__4: 754 __BTF_ID__type2__name2__4:
886 .zero 4 755 .zero 4
887 756
888 The ``struct btf_id_set set;`` variable is def 757 The ``struct btf_id_set set;`` variable is defined to access the list.
889 758
890 The ``typeX`` name can be one of following:: 759 The ``typeX`` name can be one of following::
891 760
892 struct, union, typedef, func 761 struct, union, typedef, func
893 762
894 and is used as a filter when resolving the BTF 763 and is used as a filter when resolving the BTF ID value.
895 764
896 All the BTF ID lists and sets are compiled in 765 All the BTF ID lists and sets are compiled in the .BTF_ids section and
897 resolved during the linking phase of kernel bu 766 resolved during the linking phase of kernel build by ``resolve_btfids`` tool.
898 767
899 5. Using BTF 768 5. Using BTF
900 ============ !! 769 ************
901 770
902 5.1 bpftool map pretty print 771 5.1 bpftool map pretty print
903 ---------------------------- !! 772 ============================
904 773
905 With BTF, the map key/value can be printed bas 774 With BTF, the map key/value can be printed based on fields rather than simply
906 raw bytes. This is especially valuable for lar 775 raw bytes. This is especially valuable for large structure or if your data
907 structure has bitfields. For example, for the 776 structure has bitfields. For example, for the following map,::
908 777
909 enum A { A1, A2, A3, A4, A5 }; 778 enum A { A1, A2, A3, A4, A5 };
910 typedef enum A ___A; 779 typedef enum A ___A;
911 struct tmp_t { 780 struct tmp_t {
912 char a1:4; 781 char a1:4;
913 int a2:4; 782 int a2:4;
914 int :4; 783 int :4;
915 __u32 a3:4; 784 __u32 a3:4;
916 int b; 785 int b;
917 ___A b1:4; 786 ___A b1:4;
918 enum A b2:4; 787 enum A b2:4;
919 }; 788 };
920 struct { !! 789 struct bpf_map_def SEC("maps") tmpmap = {
921 __uint(type, BPF_MAP_TYPE_ARRAY); !! 790 .type = BPF_MAP_TYPE_ARRAY,
922 __type(key, int); !! 791 .key_size = sizeof(__u32),
923 __type(value, struct tmp_t); !! 792 .value_size = sizeof(struct tmp_t),
924 __uint(max_entries, 1); !! 793 .max_entries = 1,
925 } tmpmap SEC(".maps"); !! 794 };
>> 795 BPF_ANNOTATE_KV_PAIR(tmpmap, int, struct tmp_t);
926 796
927 bpftool is able to pretty print like below: 797 bpftool is able to pretty print like below:
928 :: 798 ::
929 799
930 [{ 800 [{
931 "key": 0, 801 "key": 0,
932 "value": { 802 "value": {
933 "a1": 0x2, 803 "a1": 0x2,
934 "a2": 0x4, 804 "a2": 0x4,
935 "a3": 0x6, 805 "a3": 0x6,
936 "b": 7, 806 "b": 7,
937 "b1": 0x8, 807 "b1": 0x8,
938 "b2": 0xa 808 "b2": 0xa
939 } 809 }
940 } 810 }
941 ] 811 ]
942 812
943 5.2 bpftool prog dump 813 5.2 bpftool prog dump
944 --------------------- !! 814 =====================
945 815
946 The following is an example showing how func_i 816 The following is an example showing how func_info and line_info can help prog
947 dump with better kernel symbol names, function 817 dump with better kernel symbol names, function prototypes and line
948 information.:: 818 information.::
949 819
950 $ bpftool prog dump jited pinned /sys/fs/b 820 $ bpftool prog dump jited pinned /sys/fs/bpf/test_btf_haskv
951 [...] 821 [...]
952 int test_long_fname_2(struct dummy_tracepo 822 int test_long_fname_2(struct dummy_tracepoint_args * arg):
953 bpf_prog_44a040bf25481309_test_long_fname_ 823 bpf_prog_44a040bf25481309_test_long_fname_2:
954 ; static int test_long_fname_2(struct dumm 824 ; static int test_long_fname_2(struct dummy_tracepoint_args *arg)
955 0: push %rbp 825 0: push %rbp
956 1: mov %rsp,%rbp 826 1: mov %rsp,%rbp
957 4: sub $0x30,%rsp 827 4: sub $0x30,%rsp
958 b: sub $0x28,%rbp 828 b: sub $0x28,%rbp
959 f: mov %rbx,0x0(%rbp) 829 f: mov %rbx,0x0(%rbp)
960 13: mov %r13,0x8(%rbp) 830 13: mov %r13,0x8(%rbp)
961 17: mov %r14,0x10(%rbp) 831 17: mov %r14,0x10(%rbp)
962 1b: mov %r15,0x18(%rbp) 832 1b: mov %r15,0x18(%rbp)
963 1f: xor %eax,%eax 833 1f: xor %eax,%eax
964 21: mov %rax,0x20(%rbp) 834 21: mov %rax,0x20(%rbp)
965 25: xor %esi,%esi 835 25: xor %esi,%esi
966 ; int key = 0; 836 ; int key = 0;
967 27: mov %esi,-0x4(%rbp) 837 27: mov %esi,-0x4(%rbp)
968 ; if (!arg->sock) 838 ; if (!arg->sock)
969 2a: mov 0x8(%rdi),%rdi 839 2a: mov 0x8(%rdi),%rdi
970 ; if (!arg->sock) 840 ; if (!arg->sock)
971 2e: cmp $0x0,%rdi 841 2e: cmp $0x0,%rdi
972 32: je 0x0000000000000070 842 32: je 0x0000000000000070
973 34: mov %rbp,%rsi 843 34: mov %rbp,%rsi
974 ; counts = bpf_map_lookup_elem(&btf_map, & 844 ; counts = bpf_map_lookup_elem(&btf_map, &key);
975 [...] 845 [...]
976 846
977 5.3 Verifier Log 847 5.3 Verifier Log
978 ---------------- !! 848 ================
979 849
980 The following is an example of how line_info c 850 The following is an example of how line_info can help debugging verification
981 failure.:: 851 failure.::
982 852
983 /* The code at tools/testing/selftests/ 853 /* The code at tools/testing/selftests/bpf/test_xdp_noinline.c
984 * is modified as below. 854 * is modified as below.
985 */ 855 */
986 data = (void *)(long)xdp->data; 856 data = (void *)(long)xdp->data;
987 data_end = (void *)(long)xdp->data_end; 857 data_end = (void *)(long)xdp->data_end;
988 /* 858 /*
989 if (data + 4 > data_end) 859 if (data + 4 > data_end)
990 return XDP_DROP; 860 return XDP_DROP;
991 */ 861 */
992 *(u32 *)data = dst->dst; 862 *(u32 *)data = dst->dst;
993 863
994 $ bpftool prog load ./test_xdp_noinline.o 864 $ bpftool prog load ./test_xdp_noinline.o /sys/fs/bpf/test_xdp_noinline type xdp
995 ; data = (void *)(long)xdp->data; 865 ; data = (void *)(long)xdp->data;
996 224: (79) r2 = *(u64 *)(r10 -112) 866 224: (79) r2 = *(u64 *)(r10 -112)
997 225: (61) r2 = *(u32 *)(r2 +0) 867 225: (61) r2 = *(u32 *)(r2 +0)
998 ; *(u32 *)data = dst->dst; 868 ; *(u32 *)data = dst->dst;
999 226: (63) *(u32 *)(r2 +0) = r1 869 226: (63) *(u32 *)(r2 +0) = r1
1000 invalid access to packet, off=0 size= 870 invalid access to packet, off=0 size=4, R2(id=0,off=0,r=0)
1001 R2 offset is outside of the packet 871 R2 offset is outside of the packet
1002 872
1003 6. BTF Generation 873 6. BTF Generation
1004 ================= !! 874 *****************
1005 875
1006 You need latest pahole 876 You need latest pahole
1007 877
1008 https://git.kernel.org/pub/scm/devel/pahole 878 https://git.kernel.org/pub/scm/devel/pahole/pahole.git/
1009 879
1010 or llvm (8.0 or later). The pahole acts as a 880 or llvm (8.0 or later). The pahole acts as a dwarf2btf converter. It doesn't
1011 support .BTF.ext and btf BTF_KIND_FUNC type y 881 support .BTF.ext and btf BTF_KIND_FUNC type yet. For example,::
1012 882
1013 -bash-4.4$ cat t.c 883 -bash-4.4$ cat t.c
1014 struct t { 884 struct t {
1015 int a:2; 885 int a:2;
1016 int b:3; 886 int b:3;
1017 int c:2; 887 int c:2;
1018 } g; 888 } g;
1019 -bash-4.4$ gcc -c -O2 -g t.c 889 -bash-4.4$ gcc -c -O2 -g t.c
1020 -bash-4.4$ pahole -JV t.o 890 -bash-4.4$ pahole -JV t.o
1021 File t.o: 891 File t.o:
1022 [1] STRUCT t kind_flag=1 size=4 vlen=3 892 [1] STRUCT t kind_flag=1 size=4 vlen=3
1023 a type_id=2 bitfield_size=2 bit 893 a type_id=2 bitfield_size=2 bits_offset=0
1024 b type_id=2 bitfield_size=3 bit 894 b type_id=2 bitfield_size=3 bits_offset=2
1025 c type_id=2 bitfield_size=2 bit 895 c type_id=2 bitfield_size=2 bits_offset=5
1026 [2] INT int size=4 bit_offset=0 nr_bits 896 [2] INT int size=4 bit_offset=0 nr_bits=32 encoding=SIGNED
1027 897
1028 The llvm is able to generate .BTF and .BTF.ex 898 The llvm is able to generate .BTF and .BTF.ext directly with -g for bpf target
1029 only. The assembly code (-S) is able to show 899 only. The assembly code (-S) is able to show the BTF encoding in assembly
1030 format.:: 900 format.::
1031 901
1032 -bash-4.4$ cat t2.c 902 -bash-4.4$ cat t2.c
1033 typedef int __int32; 903 typedef int __int32;
1034 struct t2 { 904 struct t2 {
1035 int a2; 905 int a2;
1036 int (*f2)(char q1, __int32 q2, ...); 906 int (*f2)(char q1, __int32 q2, ...);
1037 int (*f3)(); 907 int (*f3)();
1038 } g2; 908 } g2;
1039 int main() { return 0; } 909 int main() { return 0; }
1040 int test() { return 0; } 910 int test() { return 0; }
1041 -bash-4.4$ clang -c -g -O2 --target=bpf t !! 911 -bash-4.4$ clang -c -g -O2 -target bpf t2.c
1042 -bash-4.4$ readelf -S t2.o 912 -bash-4.4$ readelf -S t2.o
1043 ...... 913 ......
1044 [ 8] .BTF PROGBITS 914 [ 8] .BTF PROGBITS 0000000000000000 00000247
1045 000000000000016e 0000000000000000 915 000000000000016e 0000000000000000 0 0 1
1046 [ 9] .BTF.ext PROGBITS 916 [ 9] .BTF.ext PROGBITS 0000000000000000 000003b5
1047 0000000000000060 0000000000000000 917 0000000000000060 0000000000000000 0 0 1
1048 [10] .rel.BTF.ext REL 918 [10] .rel.BTF.ext REL 0000000000000000 000007e0
1049 0000000000000040 0000000000000010 919 0000000000000040 0000000000000010 16 9 8
1050 ...... 920 ......
1051 -bash-4.4$ clang -S -g -O2 --target=bpf t !! 921 -bash-4.4$ clang -S -g -O2 -target bpf t2.c
1052 -bash-4.4$ cat t2.s 922 -bash-4.4$ cat t2.s
1053 ...... 923 ......
1054 .section .BTF,"",@progbits 924 .section .BTF,"",@progbits
1055 .short 60319 # 925 .short 60319 # 0xeb9f
1056 .byte 1 926 .byte 1
1057 .byte 0 927 .byte 0
1058 .long 24 928 .long 24
1059 .long 0 929 .long 0
1060 .long 220 930 .long 220
1061 .long 220 931 .long 220
1062 .long 122 932 .long 122
1063 .long 0 # 933 .long 0 # BTF_KIND_FUNC_PROTO(id = 1)
1064 .long 218103808 # 934 .long 218103808 # 0xd000000
1065 .long 2 935 .long 2
1066 .long 83 # 936 .long 83 # BTF_KIND_INT(id = 2)
1067 .long 16777216 # 937 .long 16777216 # 0x1000000
1068 .long 4 938 .long 4
1069 .long 16777248 # 939 .long 16777248 # 0x1000020
1070 ...... 940 ......
1071 .byte 0 # 941 .byte 0 # string offset=0
1072 .ascii ".text" # 942 .ascii ".text" # string offset=1
1073 .byte 0 943 .byte 0
1074 .ascii "/home/yhs/tmp-pahole/t2. 944 .ascii "/home/yhs/tmp-pahole/t2.c" # string offset=7
1075 .byte 0 945 .byte 0
1076 .ascii "int main() { return 0; } 946 .ascii "int main() { return 0; }" # string offset=33
1077 .byte 0 947 .byte 0
1078 .ascii "int test() { return 0; } 948 .ascii "int test() { return 0; }" # string offset=58
1079 .byte 0 949 .byte 0
1080 .ascii "int" # 950 .ascii "int" # string offset=83
1081 ...... 951 ......
1082 .section .BTF.ext,"",@prog 952 .section .BTF.ext,"",@progbits
1083 .short 60319 # 953 .short 60319 # 0xeb9f
1084 .byte 1 954 .byte 1
1085 .byte 0 955 .byte 0
1086 .long 24 956 .long 24
1087 .long 0 957 .long 0
1088 .long 28 958 .long 28
1089 .long 28 959 .long 28
1090 .long 44 960 .long 44
1091 .long 8 # 961 .long 8 # FuncInfo
1092 .long 1 # 962 .long 1 # FuncInfo section string offset=1
1093 .long 2 963 .long 2
1094 .long .Lfunc_begin0 964 .long .Lfunc_begin0
1095 .long 3 965 .long 3
1096 .long .Lfunc_begin1 966 .long .Lfunc_begin1
1097 .long 5 967 .long 5
1098 .long 16 # 968 .long 16 # LineInfo
1099 .long 1 # 969 .long 1 # LineInfo section string offset=1
1100 .long 2 970 .long 2
1101 .long .Ltmp0 971 .long .Ltmp0
1102 .long 7 972 .long 7
1103 .long 33 973 .long 33
1104 .long 7182 # 974 .long 7182 # Line 7 Col 14
1105 .long .Ltmp3 975 .long .Ltmp3
1106 .long 7 976 .long 7
1107 .long 58 977 .long 58
1108 .long 8206 # 978 .long 8206 # Line 8 Col 14
1109 979
1110 7. Testing 980 7. Testing
1111 ========== !! 981 **********
1112 <<
1113 The kernel BPF selftest `tools/testing/selfte <<
1114 provides an extensive set of BTF-related test <<
1115 982
1116 .. Links !! 983 Kernel bpf selftest `test_btf.c` provides extensive set of BTF-related tests.
1117 .. _tools/testing/selftests/bpf/prog_tests/bt <<
1118 https://git.kernel.org/pub/scm/linux/kerne <<
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.