scripts/bpf_helpers_doc.py

   1 #!/usr/bin/env python3
   2 # SPDX-License-Identifier: GPL-2.0-only
   3 #
   4 # Copyright (C) 2018-2019 Netronome Systems, Inc.
   5
   6 # In case user attempts to run with Python 2.
   7 from __future__ import print_function
   8
   9 import argparse
  10 import re
  11 import sys, os
  12
  13 class NoHelperFound(BaseException):
  14     pass
  15
  16 class ParsingError(BaseException):
  17     def __init__(self, line='<line not provided>', reader=None):
  18         if reader:
  19             BaseException.__init__(self,
  20                                    'Error at file offset %d, parsing line: %s' %
  21                                    (reader.tell(), line))
  22         else:
  23             BaseException.__init__(self, 'Error parsing line: %s' % line)
  24
  25 class Helper(object):
  26     """
  27     An object representing the description of an eBPF helper function.
  28     @proto: function prototype of the helper function
  29     @desc: textual description of the helper function
  30     @ret: description of the return value of the helper function
  31     """
  32     def __init__(self, proto='', desc='', ret=''):
  33         self.proto = proto
  34         self.desc = desc
  35         self.ret = ret
  36
  37     def proto_break_down(self):
  38         """
  39         Break down helper function protocol into smaller chunks: return type,
  40         name, distincts arguments.
  41         """
  42         arg_re = re.compile('((\w+ )*?(\w+|...))( (\**)(\w+))?$')
  43         res = {}
  44         proto_re = re.compile('(.+) (\**)(\w+)\(((([^,]+)(, )?){1,5})\)$')
  45
  46         capture = proto_re.match(self.proto)
  47         res['ret_type'] = capture.group(1)
  48         res['ret_star'] = capture.group(2)
  49         res['name']     = capture.group(3)
  50         res['args'] = []
  51
  52         args    = capture.group(4).split(', ')
  53         for a in args:
  54             capture = arg_re.match(a)
  55             res['args'].append({
  56                 'type' : capture.group(1),
  57                 'star' : capture.group(5),
  58                 'name' : capture.group(6)
  59             })
  60
  61         return res
  62
  63 class HeaderParser(object):
  64     """
  65     An object used to parse a file in order to extract the documentation of a
  66     list of eBPF helper functions. All the helpers that can be retrieved are
  67     stored as Helper object, in the self.helpers() array.
  68     @filename: name of file to parse, usually include/uapi/linux/bpf.h in the
  69                kernel tree
  70     """
  71     def __init__(self, filename):
  72         self.reader = open(filename, 'r')
  73         self.line = ''
  74         self.helpers = []
  75
  76     def parse_helper(self):
  77         proto    = self.parse_proto()
  78         desc     = self.parse_desc()
  79         ret      = self.parse_ret()
  80         return Helper(proto=proto, desc=desc, ret=ret)
  81
  82     def parse_proto(self):
  83         # Argument can be of shape:
  84         #   - "void"
  85         #   - "type  name"
  86         #   - "type *name"
  87         #   - Same as above, with "const" and/or "struct" in front of type
  88         #   - "..." (undefined number of arguments, for bpf_trace_printk())
  89         # There is at least one term ("void"), and at most five arguments.
  90         p = re.compile(' \* ?((.+) \**\w+\((((const )?(struct )?(\w+|\.\.\.)( \**\w+)?)(, )?){1,5}\))$')
  91         capture = p.match(self.line)
  92         if not capture:
  93             raise NoHelperFound
  94         self.line = self.reader.readline()
  95         return capture.group(1)
  96
  97     def parse_desc(self):
  98         p = re.compile(' \* ?(?:\t| {5,8})Description$')
  99         capture = p.match(self.line)
 100         if not capture:
 101             # Helper can have empty description and we might be parsing another
 102             # attribute: return but do not consume.
 103             return ''
 104         # Description can be several lines, some of them possibly empty, and it
 105         # stops when another subsection title is met.
 106         desc = ''
 107         while True:
 108             self.line = self.reader.readline()
 109             if self.line == ' *\n':
 110                 desc += '\n'
 111             else:
 112                 p = re.compile(' \* ?(?:\t| {5,8})(?:\t| {8})(.*)')
 113                 capture = p.match(self.line)
 114                 if capture:
 115                     desc += capture.group(1) + '\n'
 116                 else:
 117                     break
 118         return desc
 119
 120     def parse_ret(self):
 121         p = re.compile(' \* ?(?:\t| {5,8})Return$')
 122         capture = p.match(self.line)
 123         if not capture:
 124             # Helper can have empty retval and we might be parsing another
 125             # attribute: return but do not consume.
 126             return ''
 127         # Return value description can be several lines, some of them possibly
 128         # empty, and it stops when another subsection title is met.
 129         ret = ''
 130         while True:
 131             self.line = self.reader.readline()
 132             if self.line == ' *\n':
 133                 ret += '\n'
 134             else:
 135                 p = re.compile(' \* ?(?:\t| {5,8})(?:\t| {8})(.*)')
 136                 capture = p.match(self.line)
 137                 if capture:
 138                     ret += capture.group(1) + '\n'
 139                 else:
 140                     break
 141         return ret
 142
 143     def run(self):
 144         # Advance to start of helper function descriptions.
 145         offset = self.reader.read().find('* Start of BPF helper function descriptions:')
 146         if offset == -1:
 147             raise Exception('Could not find start of eBPF helper descriptions list')
 148         self.reader.seek(offset)
 149         self.reader.readline()
 150         self.reader.readline()
 151         self.line = self.reader.readline()
 152
 153         while True:
 154             try:
 155                 helper = self.parse_helper()
 156                 self.helpers.append(helper)
 157             except NoHelperFound:
 158                 break
 159
 160         self.reader.close()
 161
 162 ###############################################################################
 163
 164 class Printer(object):
 165     """
 166     A generic class for printers. Printers should be created with an array of
 167     Helper objects, and implement a way to print them in the desired fashion.
 168     @helpers: array of Helper objects to print to standard output
 169     """
 170     def __init__(self, helpers):
 171         self.helpers = helpers
 172
 173     def print_header(self):
 174         pass
 175
 176     def print_footer(self):
 177         pass
 178
 179     def print_one(self, helper):
 180         pass
 181
 182     def print_all(self):
 183         self.print_header()
 184         for helper in self.helpers:
 185             self.print_one(helper)
 186         self.print_footer()
 187
 188 class PrinterRST(Printer):
 189     """
 190     A printer for dumping collected information about helpers as a ReStructured
 191     Text page compatible with the rst2man program, which can be used to
 192     generate a manual page for the helpers.
 193     @helpers: array of Helper objects to print to standard output
 194     """
 195     def print_header(self):
 196         header = '''\
 197 .. Copyright (C) All BPF authors and contributors from 2014 to present.
 198 .. See git log include/uapi/linux/bpf.h in kernel tree for details.
 199 ..
 200 .. %%%LICENSE_START(VERBATIM)
 201 .. Permission is granted to make and distribute verbatim copies of this
 202 .. manual provided the copyright notice and this permission notice are
 203 .. preserved on all copies.
 204 ..
 205 .. Permission is granted to copy and distribute modified versions of this
 206 .. manual under the conditions for verbatim copying, provided that the
 207 .. entire resulting derived work is distributed under the terms of a
 208 .. permission notice identical to this one.
 209 ..
 210 .. Since the Linux kernel and libraries are constantly changing, this
 211 .. manual page may be incorrect or out-of-date.  The author(s) assume no
 212 .. responsibility for errors or omissions, or for damages resulting from
 213 .. the use of the information contained herein.  The author(s) may not
 214 .. have taken the same level of care in the production of this manual,
 215 .. which is licensed free of charge, as they might when working
 216 .. professionally.
 217 ..
 218 .. Formatted or processed versions of this manual, if unaccompanied by
 219 .. the source, must acknowledge the copyright and authors of this work.
 220 .. %%%LICENSE_END
 221 ..
 222 .. Please do not edit this file. It was generated from the documentation
 223 .. located in file include/uapi/linux/bpf.h of the Linux kernel sources
 224 .. (helpers description), and from scripts/bpf_helpers_doc.py in the same
 225 .. repository (header and footer).
 226
 227 ===========
 228 BPF-HELPERS
 229 ===========
 230 -------------------------------------------------------------------------------
 231 list of eBPF helper functions
 232 -------------------------------------------------------------------------------
 233
 234 :Manual section: 7
 235
 236 DESCRIPTION
 237 ===========
 238
 239 The extended Berkeley Packet Filter (eBPF) subsystem consists in programs
 240 written in a pseudo-assembly language, then attached to one of the several
 241 kernel hooks and run in reaction of specific events. This framework differs
 242 from the older, "classic" BPF (or "cBPF") in several aspects, one of them being
 243 the ability to call special functions (or "helpers") from within a program.
 244 These functions are restricted to a white-list of helpers defined in the
 245 kernel.
 246
 247 These helpers are used by eBPF programs to interact with the system, or with
 248 the context in which they work. For instance, they can be used to print
 249 debugging messages, to get the time since the system was booted, to interact
 250 with eBPF maps, or to manipulate network packets. Since there are several eBPF
 251 program types, and that they do not run in the same context, each program type
 252 can only call a subset of those helpers.
 253
 254 Due to eBPF conventions, a helper can not have more than five arguments.
 255
 256 Internally, eBPF programs call directly into the compiled helper functions
 257 without requiring any foreign-function interface. As a result, calling helpers
 258 introduces no overhead, thus offering excellent performance.
 259
 260 This document is an attempt to list and document the helpers available to eBPF
 261 developers. They are sorted by chronological order (the oldest helpers in the
 262 kernel at the top).
 263
 264 HELPERS
 265 =======
 266 '''
 267         print(header)
 268
 269     def print_footer(self):
 270         footer = '''
 271 EXAMPLES
 272 ========
 273
 274 Example usage for most of the eBPF helpers listed in this manual page are
 275 available within the Linux kernel sources, at the following locations:
 276
 277 * *samples/bpf/*
 278 * *tools/testing/selftests/bpf/*
 279
 280 LICENSE
 281 =======
 282
 283 eBPF programs can have an associated license, passed along with the bytecode
 284 instructions to the kernel when the programs are loaded. The format for that
 285 string is identical to the one in use for kernel modules (Dual licenses, such
 286 as "Dual BSD/GPL", may be used). Some helper functions are only accessible to
 287 programs that are compatible with the GNU Privacy License (GPL).
 288
 289 In order to use such helpers, the eBPF program must be loaded with the correct
 290 license string passed (via **attr**) to the **bpf**\ () system call, and this
 291 generally translates into the C source code of the program containing a line
 292 similar to the following:
 293
 294 ::
 295
 296         char ____license[] __attribute__((section("license"), used)) = "GPL";
 297
 298 IMPLEMENTATION
 299 ==============
 300
 301 This manual page is an effort to document the existing eBPF helper functions.
 302 But as of this writing, the BPF sub-system is under heavy development. New eBPF
 303 program or map types are added, along with new helper functions. Some helpers
 304 are occasionally made available for additional program types. So in spite of
 305 the efforts of the community, this page might not be up-to-date. If you want to
 306 check by yourself what helper functions exist in your kernel, or what types of
 307 programs they can support, here are some files among the kernel tree that you
 308 may be interested in:
 309
 310 * *include/uapi/linux/bpf.h* is the main BPF header. It contains the full list
 311   of all helper functions, as well as many other BPF definitions including most
 312   of the flags, structs or constants used by the helpers.
 313 * *net/core/filter.c* contains the definition of most network-related helper
 314   functions, and the list of program types from which they can be used.
 315 * *kernel/trace/bpf_trace.c* is the equivalent for most tracing program-related
 316   helpers.
 317 * *kernel/bpf/verifier.c* contains the functions used to check that valid types
 318   of eBPF maps are used with a given helper function.
 319 * *kernel/bpf/* directory contains other files in which additional helpers are
 320   defined (for cgroups, sockmaps, etc.).
 321 * The bpftool utility can be used to probe the availability of helper functions
 322   on the system (as well as supported program and map types, and a number of
 323   other parameters). To do so, run **bpftool feature probe** (see
 324   **bpftool-feature**\ (8) for details). Add the **unprivileged** keyword to
 325   list features available to unprivileged users.
 326
 327 Compatibility between helper functions and program types can generally be found
 328 in the files where helper functions are defined. Look for the **struct
 329 bpf_func_proto** objects and for functions returning them: these functions
 330 contain a list of helpers that a given program type can call. Note that the
 331 **default:** label of the **switch ... case** used to filter helpers can call
 332 other functions, themselves allowing access to additional helpers. The
 333 requirement for GPL license is also in those **struct bpf_func_proto**.
 334
 335 Compatibility between helper functions and map types can be found in the
 336 **check_map_func_compatibility**\ () function in file *kernel/bpf/verifier.c*.
 337
 338 Helper functions that invalidate the checks on **data** and **data_end**
 339 pointers for network processing are listed in function
 340 **bpf_helper_changes_pkt_data**\ () in file *net/core/filter.c*.
 341
 342 SEE ALSO
 343 ========
 344
 345 **bpf**\ (2),
 346 **bpftool**\ (8),
 347 **cgroups**\ (7),
 348 **ip**\ (8),
 349 **perf_event_open**\ (2),
 350 **sendmsg**\ (2),
 351 **socket**\ (7),
 352 **tc-bpf**\ (8)'''
 353         print(footer)
 354
 355     def print_proto(self, helper):
 356         """
 357         Format function protocol with bold and italics markers. This makes RST
 358         file less readable, but gives nice results in the manual page.
 359         """
 360         proto = helper.proto_break_down()
 361
 362         print('**%s %s%s(' % (proto['ret_type'],
 363                               proto['ret_star'].replace('*', '\\*'),
 364                               proto['name']),
 365               end='')
 366
 367         comma = ''
 368         for a in proto['args']:
 369             one_arg = '{}{}'.format(comma, a['type'])
 370             if a['name']:
 371                 if a['star']:
 372                     one_arg += ' {}**\ '.format(a['star'].replace('*', '\\*'))
 373                 else:
 374                     one_arg += '** '
 375                 one_arg += '*{}*\\ **'.format(a['name'])
 376             comma = ', '
 377             print(one_arg, end='')
 378
 379         print(')**')
 380
 381     def print_one(self, helper):
 382         self.print_proto(helper)
 383
 384         if (helper.desc):
 385             print('\tDescription')
 386             # Do not strip all newline characters: formatted code at the end of
 387             # a section must be followed by a blank line.
 388             for line in re.sub('\n$', '', helper.desc, count=1).split('\n'):
 389                 print('{}{}'.format('\t\t' if line else '', line))
 390
 391         if (helper.ret):
 392             print('\tReturn')
 393             for line in helper.ret.rstrip().split('\n'):
 394                 print('{}{}'.format('\t\t' if line else '', line))
 395
 396         print('')
 397
 398 class PrinterHelpers(Printer):
 399     """
 400     A printer for dumping collected information about helpers as C header to
 401     be included from BPF program.
 402     @helpers: array of Helper objects to print to standard output
 403     """
 404
 405     type_fwds = [
 406             'struct bpf_fib_lookup',
 407             'struct bpf_sk_lookup',
 408             'struct bpf_perf_event_data',
 409             'struct bpf_perf_event_value',
 410             'struct bpf_pidns_info',
 411             'struct bpf_redir_neigh',
 412             'struct bpf_sock',
 413             'struct bpf_sock_addr',
 414             'struct bpf_sock_ops',
 415             'struct bpf_sock_tuple',
 416             'struct bpf_spin_lock',
 417             'struct bpf_sysctl',
 418             'struct bpf_tcp_sock',
 419             'struct bpf_tunnel_key',
 420             'struct bpf_xfrm_state',
 421             'struct linux_binprm',
 422             'struct pt_regs',
 423             'struct sk_reuseport_md',
 424             'struct sockaddr',
 425             'struct tcphdr',
 426             'struct seq_file',
 427             'struct tcp6_sock',
 428             'struct tcp_sock',
 429             'struct tcp_timewait_sock',
 430             'struct tcp_request_sock',
 431             'struct udp6_sock',
 432             'struct task_struct',
 433
 434             'struct __sk_buff',
 435             'struct sk_msg_md',
 436             'struct xdp_md',
 437             'struct path',
 438             'struct btf_ptr',
 439             'struct inode',
 440             'struct socket',
 441             'struct file',
 442     ]
 443     known_types = {
 444             '...',
 445             'void',
 446             'const void',
 447             'char',
 448             'const char',
 449             'int',
 450             'long',
 451             'unsigned long',
 452
 453             '__be16',
 454             '__be32',
 455             '__wsum',
 456
 457             'struct bpf_fib_lookup',
 458             'struct bpf_perf_event_data',
 459             'struct bpf_perf_event_value',
 460             'struct bpf_pidns_info',
 461             'struct bpf_redir_neigh',
 462             'struct bpf_sk_lookup',
 463             'struct bpf_sock',
 464             'struct bpf_sock_addr',
 465             'struct bpf_sock_ops',
 466             'struct bpf_sock_tuple',
 467             'struct bpf_spin_lock',
 468             'struct bpf_sysctl',
 469             'struct bpf_tcp_sock',
 470             'struct bpf_tunnel_key',
 471             'struct bpf_xfrm_state',
 472             'struct linux_binprm',
 473             'struct pt_regs',
 474             'struct sk_reuseport_md',
 475             'struct sockaddr',
 476             'struct tcphdr',
 477             'struct seq_file',
 478             'struct tcp6_sock',
 479             'struct tcp_sock',
 480             'struct tcp_timewait_sock',
 481             'struct tcp_request_sock',
 482             'struct udp6_sock',
 483             'struct task_struct',
 484             'struct path',
 485             'struct btf_ptr',
 486             'struct inode',
 487             'struct socket',
 488             'struct file',
 489     }
 490     mapped_types = {
 491             'u8': '__u8',
 492             'u16': '__u16',
 493             'u32': '__u32',
 494             'u64': '__u64',
 495             's8': '__s8',
 496             's16': '__s16',
 497             's32': '__s32',
 498             's64': '__s64',
 499             'size_t': 'unsigned long',
 500             'struct bpf_map': 'void',
 501             'struct sk_buff': 'struct __sk_buff',
 502             'const struct sk_buff': 'const struct __sk_buff',
 503             'struct sk_msg_buff': 'struct sk_msg_md',
 504             'struct xdp_buff': 'struct xdp_md',
 505     }
 506     # Helpers overloaded for different context types.
 507     overloaded_helpers = [
 508         'bpf_get_socket_cookie',
 509         'bpf_sk_assign',
 510     ]
 511
 512     def print_header(self):
 513         header = '''\
 514 /* This is auto-generated file. See bpf_helpers_doc.py for details. */
 515
 516 /* Forward declarations of BPF structs */'''
 517
 518         print(header)
 519         for fwd in self.type_fwds:
 520             print('%s;' % fwd)
 521         print('')
 522
 523     def print_footer(self):
 524         footer = ''
 525         print(footer)
 526
 527     def map_type(self, t):
 528         if t in self.known_types:
 529             return t
 530         if t in self.mapped_types:
 531             return self.mapped_types[t]
 532         print("Unrecognized type '%s', please add it to known types!" % t,
 533               file=sys.stderr)
 534         sys.exit(1)
 535
 536     seen_helpers = set()
 537
 538     def print_one(self, helper):
 539         proto = helper.proto_break_down()
 540
 541         if proto['name'] in self.seen_helpers:
 542             return
 543         self.seen_helpers.add(proto['name'])
 544
 545         print('/*')
 546         print(" * %s" % proto['name'])
 547         print(" *")
 548         if (helper.desc):
 549             # Do not strip all newline characters: formatted code at the end of
 550             # a section must be followed by a blank line.
 551             for line in re.sub('\n$', '', helper.desc, count=1).split('\n'):
 552                 print(' *{}{}'.format(' \t' if line else '', line))
 553
 554         if (helper.ret):
 555             print(' *')
 556             print(' * Returns')
 557             for line in helper.ret.rstrip().split('\n'):
 558                 print(' *{}{}'.format(' \t' if line else '', line))
 559
 560         print(' */')
 561         print('static %s %s(*%s)(' % (self.map_type(proto['ret_type']),
 562                                       proto['ret_star'], proto['name']), end='')
 563         comma = ''
 564         for i, a in enumerate(proto['args']):
 565             t = a['type']
 566             n = a['name']
 567             if proto['name'] in self.overloaded_helpers and i == 0:
 568                     t = 'void'
 569                     n = 'ctx'
 570             one_arg = '{}{}'.format(comma, self.map_type(t))
 571             if n:
 572                 if a['star']:
 573                     one_arg += ' {}'.format(a['star'])
 574                 else:
 575                     one_arg += ' '
 576                 one_arg += '{}'.format(n)
 577             comma = ', '
 578             print(one_arg, end='')
 579
 580         print(') = (void *) %d;' % len(self.seen_helpers))
 581         print('')
 582
 583 ###############################################################################
 584
 585 # If script is launched from scripts/ from kernel tree and can access
 586 # ../include/uapi/linux/bpf.h, use it as a default name for the file to parse,
 587 # otherwise the --filename argument will be required from the command line.
 588 script = os.path.abspath(sys.argv[0])
 589 linuxRoot = os.path.dirname(os.path.dirname(script))
 590 bpfh = os.path.join(linuxRoot, 'include/uapi/linux/bpf.h')
 591
 592 argParser = argparse.ArgumentParser(description="""
 593 Parse eBPF header file and generate documentation for eBPF helper functions.
 594 The RST-formatted output produced can be turned into a manual page with the
 595 rst2man utility.
 596 """)
 597 argParser.add_argument('--header', action='store_true',
 598                        help='generate C header file')
 599 if (os.path.isfile(bpfh)):
 600     argParser.add_argument('--filename', help='path to include/uapi/linux/bpf.h',
 601                            default=bpfh)
 602 else:
 603     argParser.add_argument('--filename', help='path to include/uapi/linux/bpf.h')
 604 args = argParser.parse_args()
 605
 606 # Parse file.
 607 headerParser = HeaderParser(args.filename)
 608 headerParser.run()
 609
 610 # Print formatted output to standard output.
 611 if args.header:
 612     printer = PrinterHelpers(headerParser.helpers)
 613 else:
 614     printer = PrinterRST(headerParser.helpers)
 615 printer.print_all()