scripts/bpf_helpers_doc.py

   1 #!/usr/bin/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         print('Parsed description of %d helper function(s)' % len(self.helpers),
 162               file=sys.stderr)
 163
 164 ###############################################################################
 165
 166 class Printer(object):
 167     """
 168     A generic class for printers. Printers should be created with an array of
 169     Helper objects, and implement a way to print them in the desired fashion.
 170     @helpers: array of Helper objects to print to standard output
 171     """
 172     def __init__(self, helpers):
 173         self.helpers = helpers
 174
 175     def print_header(self):
 176         pass
 177
 178     def print_footer(self):
 179         pass
 180
 181     def print_one(self, helper):
 182         pass
 183
 184     def print_all(self):
 185         self.print_header()
 186         for helper in self.helpers:
 187             self.print_one(helper)
 188         self.print_footer()
 189
 190 class PrinterRST(Printer):
 191     """
 192     A printer for dumping collected information about helpers as a ReStructured
 193     Text page compatible with the rst2man program, which can be used to
 194     generate a manual page for the helpers.
 195     @helpers: array of Helper objects to print to standard output
 196     """
 197     def print_header(self):
 198         header = '''\
 199 .. Copyright (C) All BPF authors and contributors from 2014 to present.
 200 .. See git log include/uapi/linux/bpf.h in kernel tree for details.
 201 ..
 202 .. %%%LICENSE_START(VERBATIM)
 203 .. Permission is granted to make and distribute verbatim copies of this
 204 .. manual provided the copyright notice and this permission notice are
 205 .. preserved on all copies.
 206 ..
 207 .. Permission is granted to copy and distribute modified versions of this
 208 .. manual under the conditions for verbatim copying, provided that the
 209 .. entire resulting derived work is distributed under the terms of a
 210 .. permission notice identical to this one.
 211 ..
 212 .. Since the Linux kernel and libraries are constantly changing, this
 213 .. manual page may be incorrect or out-of-date.  The author(s) assume no
 214 .. responsibility for errors or omissions, or for damages resulting from
 215 .. the use of the information contained herein.  The author(s) may not
 216 .. have taken the same level of care in the production of this manual,
 217 .. which is licensed free of charge, as they might when working
 218 .. professionally.
 219 ..
 220 .. Formatted or processed versions of this manual, if unaccompanied by
 221 .. the source, must acknowledge the copyright and authors of this work.
 222 .. %%%LICENSE_END
 223 ..
 224 .. Please do not edit this file. It was generated from the documentation
 225 .. located in file include/uapi/linux/bpf.h of the Linux kernel sources
 226 .. (helpers description), and from scripts/bpf_helpers_doc.py in the same
 227 .. repository (header and footer).
 228
 229 ===========
 230 BPF-HELPERS
 231 ===========
 232 -------------------------------------------------------------------------------
 233 list of eBPF helper functions
 234 -------------------------------------------------------------------------------
 235
 236 :Manual section: 7
 237
 238 DESCRIPTION
 239 ===========
 240
 241 The extended Berkeley Packet Filter (eBPF) subsystem consists in programs
 242 written in a pseudo-assembly language, then attached to one of the several
 243 kernel hooks and run in reaction of specific events. This framework differs
 244 from the older, "classic" BPF (or "cBPF") in several aspects, one of them being
 245 the ability to call special functions (or "helpers") from within a program.
 246 These functions are restricted to a white-list of helpers defined in the
 247 kernel.
 248
 249 These helpers are used by eBPF programs to interact with the system, or with
 250 the context in which they work. For instance, they can be used to print
 251 debugging messages, to get the time since the system was booted, to interact
 252 with eBPF maps, or to manipulate network packets. Since there are several eBPF
 253 program types, and that they do not run in the same context, each program type
 254 can only call a subset of those helpers.
 255
 256 Due to eBPF conventions, a helper can not have more than five arguments.
 257
 258 Internally, eBPF programs call directly into the compiled helper functions
 259 without requiring any foreign-function interface. As a result, calling helpers
 260 introduces no overhead, thus offering excellent performance.
 261
 262 This document is an attempt to list and document the helpers available to eBPF
 263 developers. They are sorted by chronological order (the oldest helpers in the
 264 kernel at the top).
 265
 266 HELPERS
 267 =======
 268 '''
 269         print(header)
 270
 271     def print_footer(self):
 272         footer = '''
 273 EXAMPLES
 274 ========
 275
 276 Example usage for most of the eBPF helpers listed in this manual page are
 277 available within the Linux kernel sources, at the following locations:
 278
 279 * *samples/bpf/*
 280 * *tools/testing/selftests/bpf/*
 281
 282 LICENSE
 283 =======
 284
 285 eBPF programs can have an associated license, passed along with the bytecode
 286 instructions to the kernel when the programs are loaded. The format for that
 287 string is identical to the one in use for kernel modules (Dual licenses, such
 288 as "Dual BSD/GPL", may be used). Some helper functions are only accessible to
 289 programs that are compatible with the GNU Privacy License (GPL).
 290
 291 In order to use such helpers, the eBPF program must be loaded with the correct
 292 license string passed (via **attr**) to the **bpf**\ () system call, and this
 293 generally translates into the C source code of the program containing a line
 294 similar to the following:
 295
 296 ::
 297
 298         char ____license[] __attribute__((section("license"), used)) = "GPL";
 299
 300 IMPLEMENTATION
 301 ==============
 302
 303 This manual page is an effort to document the existing eBPF helper functions.
 304 But as of this writing, the BPF sub-system is under heavy development. New eBPF
 305 program or map types are added, along with new helper functions. Some helpers
 306 are occasionally made available for additional program types. So in spite of
 307 the efforts of the community, this page might not be up-to-date. If you want to
 308 check by yourself what helper functions exist in your kernel, or what types of
 309 programs they can support, here are some files among the kernel tree that you
 310 may be interested in:
 311
 312 * *include/uapi/linux/bpf.h* is the main BPF header. It contains the full list
 313   of all helper functions, as well as many other BPF definitions including most
 314   of the flags, structs or constants used by the helpers.
 315 * *net/core/filter.c* contains the definition of most network-related helper
 316   functions, and the list of program types from which they can be used.
 317 * *kernel/trace/bpf_trace.c* is the equivalent for most tracing program-related
 318   helpers.
 319 * *kernel/bpf/verifier.c* contains the functions used to check that valid types
 320   of eBPF maps are used with a given helper function.
 321 * *kernel/bpf/* directory contains other files in which additional helpers are
 322   defined (for cgroups, sockmaps, etc.).
 323
 324 Compatibility between helper functions and program types can generally be found
 325 in the files where helper functions are defined. Look for the **struct
 326 bpf_func_proto** objects and for functions returning them: these functions
 327 contain a list of helpers that a given program type can call. Note that the
 328 **default:** label of the **switch ... case** used to filter helpers can call
 329 other functions, themselves allowing access to additional helpers. The
 330 requirement for GPL license is also in those **struct bpf_func_proto**.
 331
 332 Compatibility between helper functions and map types can be found in the
 333 **check_map_func_compatibility**\ () function in file *kernel/bpf/verifier.c*.
 334
 335 Helper functions that invalidate the checks on **data** and **data_end**
 336 pointers for network processing are listed in function
 337 **bpf_helper_changes_pkt_data**\ () in file *net/core/filter.c*.
 338
 339 SEE ALSO
 340 ========
 341
 342 **bpf**\ (2),
 343 **cgroups**\ (7),
 344 **ip**\ (8),
 345 **perf_event_open**\ (2),
 346 **sendmsg**\ (2),
 347 **socket**\ (7),
 348 **tc-bpf**\ (8)'''
 349         print(footer)
 350
 351     def print_proto(self, helper):
 352         """
 353         Format function protocol with bold and italics markers. This makes RST
 354         file less readable, but gives nice results in the manual page.
 355         """
 356         proto = helper.proto_break_down()
 357
 358         print('**%s %s%s(' % (proto['ret_type'],
 359                               proto['ret_star'].replace('*', '\\*'),
 360                               proto['name']),
 361               end='')
 362
 363         comma = ''
 364         for a in proto['args']:
 365             one_arg = '{}{}'.format(comma, a['type'])
 366             if a['name']:
 367                 if a['star']:
 368                     one_arg += ' {}**\ '.format(a['star'].replace('*', '\\*'))
 369                 else:
 370                     one_arg += '** '
 371                 one_arg += '*{}*\\ **'.format(a['name'])
 372             comma = ', '
 373             print(one_arg, end='')
 374
 375         print(')**')
 376
 377     def print_one(self, helper):
 378         self.print_proto(helper)
 379
 380         if (helper.desc):
 381             print('\tDescription')
 382             # Do not strip all newline characters: formatted code at the end of
 383             # a section must be followed by a blank line.
 384             for line in re.sub('\n$', '', helper.desc, count=1).split('\n'):
 385                 print('{}{}'.format('\t\t' if line else '', line))
 386
 387         if (helper.ret):
 388             print('\tReturn')
 389             for line in helper.ret.rstrip().split('\n'):
 390                 print('{}{}'.format('\t\t' if line else '', line))
 391
 392         print('')
 393
 394 class PrinterHelpers(Printer):
 395     """
 396     A printer for dumping collected information about helpers as C header to
 397     be included from BPF program.
 398     @helpers: array of Helper objects to print to standard output
 399     """
 400
 401     type_fwds = [
 402             'struct bpf_fib_lookup',
 403             'struct bpf_perf_event_data',
 404             'struct bpf_perf_event_value',
 405             'struct bpf_sock',
 406             'struct bpf_sock_addr',
 407             'struct bpf_sock_ops',
 408             'struct bpf_sock_tuple',
 409             'struct bpf_spin_lock',
 410             'struct bpf_sysctl',
 411             'struct bpf_tcp_sock',
 412             'struct bpf_tunnel_key',
 413             'struct bpf_xfrm_state',
 414             'struct pt_regs',
 415             'struct sk_reuseport_md',
 416             'struct sockaddr',
 417             'struct tcphdr',
 418
 419             'struct __sk_buff',
 420             'struct sk_msg_md',
 421             'struct xdp_md',
 422     ]
 423     known_types = {
 424             '...',
 425             'void',
 426             'const void',
 427             'char',
 428             'const char',
 429             'int',
 430             'long',
 431             'unsigned long',
 432
 433             '__be16',
 434             '__be32',
 435             '__wsum',
 436
 437             'struct bpf_fib_lookup',
 438             'struct bpf_perf_event_data',
 439             'struct bpf_perf_event_value',
 440             'struct bpf_sock',
 441             'struct bpf_sock_addr',
 442             'struct bpf_sock_ops',
 443             'struct bpf_sock_tuple',
 444             'struct bpf_spin_lock',
 445             'struct bpf_sysctl',
 446             'struct bpf_tcp_sock',
 447             'struct bpf_tunnel_key',
 448             'struct bpf_xfrm_state',
 449             'struct pt_regs',
 450             'struct sk_reuseport_md',
 451             'struct sockaddr',
 452             'struct tcphdr',
 453     }
 454     mapped_types = {
 455             'u8': '__u8',
 456             'u16': '__u16',
 457             'u32': '__u32',
 458             'u64': '__u64',
 459             's8': '__s8',
 460             's16': '__s16',
 461             's32': '__s32',
 462             's64': '__s64',
 463             'size_t': 'unsigned long',
 464             'struct bpf_map': 'void',
 465             'struct sk_buff': 'struct __sk_buff',
 466             'const struct sk_buff': 'const struct __sk_buff',
 467             'struct sk_msg_buff': 'struct sk_msg_md',
 468             'struct xdp_buff': 'struct xdp_md',
 469     }
 470
 471     def print_header(self):
 472         header = '''\
 473 /* This is auto-generated file. See bpf_helpers_doc.py for details. */
 474
 475 /* Forward declarations of BPF structs */'''
 476
 477         print(header)
 478         for fwd in self.type_fwds:
 479             print('%s;' % fwd)
 480         print('')
 481
 482     def print_footer(self):
 483         footer = ''
 484         print(footer)
 485
 486     def map_type(self, t):
 487         if t in self.known_types:
 488             return t
 489         if t in self.mapped_types:
 490             return self.mapped_types[t]
 491         print("Unrecognized type '%s', please add it to known types!" % t,
 492               file=sys.stderr)
 493         sys.exit(1)
 494
 495     seen_helpers = set()
 496
 497     def print_one(self, helper):
 498         proto = helper.proto_break_down()
 499
 500         if proto['name'] in self.seen_helpers:
 501             return
 502         self.seen_helpers.add(proto['name'])
 503
 504         print('/*')
 505         print(" * %s" % proto['name'])
 506         print(" *")
 507         if (helper.desc):
 508             # Do not strip all newline characters: formatted code at the end of
 509             # a section must be followed by a blank line.
 510             for line in re.sub('\n$', '', helper.desc, count=1).split('\n'):
 511                 print(' *{}{}'.format(' \t' if line else '', line))
 512
 513         if (helper.ret):
 514             print(' *')
 515             print(' * Returns')
 516             for line in helper.ret.rstrip().split('\n'):
 517                 print(' *{}{}'.format(' \t' if line else '', line))
 518
 519         print(' */')
 520         print('static %s %s(*%s)(' % (self.map_type(proto['ret_type']),
 521                                       proto['ret_star'], proto['name']), end='')
 522         comma = ''
 523         for i, a in enumerate(proto['args']):
 524             t = a['type']
 525             n = a['name']
 526             if proto['name'] == 'bpf_get_socket_cookie' and i == 0:
 527                     t = 'void'
 528                     n = 'ctx'
 529             one_arg = '{}{}'.format(comma, self.map_type(t))
 530             if n:
 531                 if a['star']:
 532                     one_arg += ' {}'.format(a['star'])
 533                 else:
 534                     one_arg += ' '
 535                 one_arg += '{}'.format(n)
 536             comma = ', '
 537             print(one_arg, end='')
 538
 539         print(') = (void *) %d;' % len(self.seen_helpers))
 540         print('')
 541
 542 ###############################################################################
 543
 544 # If script is launched from scripts/ from kernel tree and can access
 545 # ../include/uapi/linux/bpf.h, use it as a default name for the file to parse,
 546 # otherwise the --filename argument will be required from the command line.
 547 script = os.path.abspath(sys.argv[0])
 548 linuxRoot = os.path.dirname(os.path.dirname(script))
 549 bpfh = os.path.join(linuxRoot, 'include/uapi/linux/bpf.h')
 550
 551 argParser = argparse.ArgumentParser(description="""
 552 Parse eBPF header file and generate documentation for eBPF helper functions.
 553 The RST-formatted output produced can be turned into a manual page with the
 554 rst2man utility.
 555 """)
 556 argParser.add_argument('--header', action='store_true',
 557                        help='generate C header file')
 558 if (os.path.isfile(bpfh)):
 559     argParser.add_argument('--filename', help='path to include/uapi/linux/bpf.h',
 560                            default=bpfh)
 561 else:
 562     argParser.add_argument('--filename', help='path to include/uapi/linux/bpf.h')
 563 args = argParser.parse_args()
 564
 565 # Parse file.
 566 headerParser = HeaderParser(args.filename)
 567 headerParser.run()
 568
 569 # Print formatted output to standard output.
 570 if args.header:
 571     printer = PrinterHelpers(headerParser.helpers)
 572 else:
 573     printer = PrinterRST(headerParser.helpers)
 574 printer.print_all()