Files
wuffs/fuzz/c/std/brotli_fuzzer.c
T
Claude b9be8feaee std/brotli: large-window mode, custom dictionaries and a fuzzer
Completes the decoder's feature set relative to libbrotlidec v1.2.1:

- QUIRK_LARGE_WINDOW mirrors BROTLI_DECODER_PARAM_LARGE_WINDOW: the
  large-window WBITS escape, window bits up to 30, the 544-symbol distance
  alphabet and 64-bit distance arithmetic, with workbuf_len growing to
  (1 << 30) + 542 when the quirk is set.
- decoder.add_dictionary! mirrors BrotliDecoderAttachDictionary for raw
  (LZ77 prefix) and serialized shared dictionaries: compound-dictionary
  distance mapping, the shared-dictionary parser (custom word lists,
  transform lists including the SHIFT transforms, context-dependent
  dictionary selection, prefix dictionaries) and all of its validation.
  Wuffs structs cannot hold slices, so the dictionary bytes are copied
  into the front of the caller's workbuf; the layout is documented in
  the README. Serialized dictionaries are accepted as in a
  BROTLI_EXPERIMENTAL build of libbrotli.
- fuzz/c/std/brotli_fuzzer.c, exercising the quirk, dictionaries and
  random I/O limits.

Tests: new golden vectors for large windows (including a 17 MiB
far-distance stream generated in the test), raw and serialized
dictionaries, nine more malformed inputs, and mimic comparisons that
attach the same dictionaries to libbrotlidec. Differential runs over
2.7 million mutated large-window and dictionary streams found no
divergence from libbrotlidec.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_011NoosQNe7DKWnqDCBeqw1p
2026-10-26 23:45:06 +00:00

214 lines
8.9 KiB
C

// Copyright 2026 The Wuffs Authors.
//
// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
// https://www.apache.org/licenses/LICENSE-2.0> or the MIT license
// <LICENSE-MIT or https://opensource.org/licenses/MIT>, at your
// option. This file may not be copied, modified, or distributed
// except according to those terms.
//
// SPDX-License-Identifier: Apache-2.0 OR MIT
// ----------------
// Silence the nested slash-star warning for the next comment's command line.
#pragma clang diagnostic push
#pragma clang diagnostic ignored "-Wcomment"
/*
This fuzzer (the fuzz function) is typically run indirectly, by a framework
such as https://github.com/google/oss-fuzz calling LLVMFuzzerTestOneInput.
When working on the fuzz implementation, or as a coherence check, defining
WUFFS_CONFIG__FUZZLIB_MAIN will let you manually run fuzz over a set of files:
gcc -DWUFFS_CONFIG__FUZZLIB_MAIN brotli_fuzzer.c
./a.out ../../../test/data/brotli/*
rm -f ./a.out
It should print "PASS", amongst other information, and exit(0).
*/
#pragma clang diagnostic pop
// Wuffs ships as a "single file C library" or "header file library" as per
// https://github.com/nothings/stb/blob/master/docs/stb_howto.txt
//
// To use that single file as a "foo.c"-like implementation, instead of a
// "foo.h"-like header, #define WUFFS_IMPLEMENTATION before #include'ing or
// compiling it.
#define WUFFS_IMPLEMENTATION
#if defined(WUFFS_CONFIG__FUZZLIB_MAIN)
// Defining the WUFFS_CONFIG__STATIC_FUNCTIONS macro is optional, but when
// combined with WUFFS_IMPLEMENTATION, it demonstrates making all of Wuffs'
// functions have static storage.
//
// This can help the compiler ignore or discard unused code, which can produce
// faster compiles and smaller binaries. Other motivations are discussed in the
// "ALLOW STATIC IMPLEMENTATION" section of
// https://raw.githubusercontent.com/nothings/stb/master/docs/stb_howto.txt
#define WUFFS_CONFIG__STATIC_FUNCTIONS
#endif // defined(WUFFS_CONFIG__FUZZLIB_MAIN)
// Defining the WUFFS_CONFIG__MODULE* macros are optional, but it lets users of
// release/c/etc.c choose which parts of Wuffs to build. That file contains the
// entire Wuffs standard library, implementing a variety of codecs and file
// formats. Without this macro definition, an optimizing compiler or linker may
// very well discard Wuffs code for unused codecs, but listing the Wuffs
// modules we use makes that process explicit. Preprocessing means that such
// code simply isn't compiled.
#define WUFFS_CONFIG__MODULES
#define WUFFS_CONFIG__MODULE__BASE
#define WUFFS_CONFIG__MODULE__BROTLI
// If building this program in an environment that doesn't easily accommodate
// relative includes, you can use the script/inline-c-relative-includes.go
// program to generate a stand-alone C file.
#include "../../../release/c/wuffs-unsupported-snapshot.c"
#include "../fuzzlib/fuzzlib.c"
// 64 KiB.
#define DST_BUFFER_ARRAY_SIZE 65536
// Wuffs allows either statically or dynamically allocated work buffers. This
// program exercises static allocation.
//
// We hard-code 64 MiB, as WUFFS_BROTLI__DECODER_WORKBUF_LEN_MAX_INCL_WORST_CASE
// is over 3 GiB and even the large-window workbuf_len is over 1 GiB. Using
// those can cause "relocation truncated to fit: R_X86_64_PC32 against symbol
// etc" compiler errors. Large-window streams with a window of 64 MiB or more
// are rejected with "#base: bad workbuf length", but smaller ones are fine.
#define WORK_BUFFER_ARRAY_SIZE 67108864
#if WORK_BUFFER_ARRAY_SIZE > 0
uint8_t g_work_buffer_array[WORK_BUFFER_ARRAY_SIZE];
#else
// Not all C/C++ compilers support 0-length arrays.
uint8_t g_work_buffer_array[1];
#endif
// g_raw_dictionary is a raw (LZ77 prefix) dictionary.
static const uint8_t g_raw_dictionary[] =
"The quick brown fox jumps over the lazy dog. "
"<html><head><title>Wuffs</title></head><body></body></html>";
// g_serialized_dictionary is a small serialized shared dictionary. It has an
// LZ77 prefix, a custom word list and a custom transform list (with
// BROTLI_TRANSFORM_SHIFT_FIRST and BROTLI_TRANSFORM_SHIFT_ALL transforms) and
// two context-selected word and transform list combinations.
static const uint8_t g_serialized_dictionary[] = {
0x91, 0x00, 0x11, 0x77, 0x75, 0x66, 0x66, 0x73, 0x20, 0x62, 0x72, 0x6F,
0x74, 0x6C, 0x69, 0x20, 0x66, 0x75, 0x7A, 0x7A, 0x01, 0x01, 0x01, 0x00,
0x01, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
0x00, 0x6F, 0x76, 0x65, 0x72, 0x6C, 0x61, 0x7A, 0x79, 0x71, 0x75, 0x69,
0x63, 0x6B, 0x62, 0x72, 0x6F, 0x77, 0x6E, 0x77, 0x69, 0x7A, 0x61, 0x72,
0x64, 0x73, 0x71, 0x75, 0x69, 0x65, 0x74, 0x6C, 0x79, 0x01, 0x06, 0x00,
0x01, 0x20, 0x02, 0xC3, 0xA9, 0x00, 0x06, 0x02, 0x00, 0x02, 0x02, 0x15,
0x00, 0x01, 0x16, 0x02, 0x02, 0x0B, 0x01, 0x00, 0x0E, 0x00, 0x02, 0x15,
0x02, 0x00, 0x00, 0x41, 0x00, 0xFF, 0xFF, 0x00, 0x00, 0x00, 0x00, 0x01,
0x80, 0x02, 0x00, 0x00, 0x01, 0x01, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00,
0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00,
0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00,
0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00,
0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00,
0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01,
};
const char* //
fuzz(wuffs_base__io_buffer* src, uint64_t hash) {
wuffs_brotli__decoder dec;
wuffs_base__status status = wuffs_brotli__decoder__initialize(
&dec, sizeof dec, WUFFS_VERSION,
(hash & 1) ? WUFFS_INITIALIZE__LEAVE_INTERNAL_BUFFERS_UNINITIALIZED : 0);
if (!wuffs_base__status__is_ok(&status)) {
return wuffs_base__status__message(&status);
}
// Enable large-window brotli for half of the inputs.
wuffs_brotli__decoder__set_quirk(&dec, WUFFS_BROTLI__QUIRK_LARGE_WINDOW,
(hash >> 1) & 1);
wuffs_base__slice_u8 workbuf =
wuffs_base__make_slice_u8(g_work_buffer_array, WORK_BUFFER_ARRAY_SIZE);
// Attach dictionaries for some of the inputs.
switch ((hash >> 2) & 7) {
case 1:
status = wuffs_brotli__decoder__add_dictionary(
&dec, WUFFS_BROTLI__DICTIONARY_TYPE_RAW,
wuffs_base__make_slice_u8((uint8_t*)g_raw_dictionary,
sizeof g_raw_dictionary - 1),
workbuf);
break;
case 2:
status = wuffs_brotli__decoder__add_dictionary(
&dec, WUFFS_BROTLI__DICTIONARY_TYPE_SERIALIZED,
wuffs_base__make_slice_u8((uint8_t*)g_serialized_dictionary,
sizeof g_serialized_dictionary),
workbuf);
break;
case 3:
status = wuffs_brotli__decoder__add_dictionary(
&dec, WUFFS_BROTLI__DICTIONARY_TYPE_SERIALIZED,
wuffs_base__make_slice_u8((uint8_t*)g_serialized_dictionary,
sizeof g_serialized_dictionary),
workbuf);
if (wuffs_base__status__is_ok(&status)) {
status = wuffs_brotli__decoder__add_dictionary(
&dec, WUFFS_BROTLI__DICTIONARY_TYPE_RAW,
wuffs_base__make_slice_u8((uint8_t*)g_raw_dictionary,
sizeof g_raw_dictionary - 1),
workbuf);
}
break;
}
if (!wuffs_base__status__is_ok(&status)) {
return wuffs_base__status__message(&status);
}
wuffs_base__range_ii_u64 workbuf_len =
wuffs_brotli__decoder__workbuf_len(&dec);
if (workbuf.len > workbuf_len.max_incl) {
workbuf.len = (size_t)workbuf_len.max_incl;
}
static uint8_t dst_buffer[DST_BUFFER_ARRAY_SIZE];
wuffs_base__io_buffer dst =
wuffs_base__ptr_u8__writer(&dst_buffer[0], DST_BUFFER_ARRAY_SIZE);
// Use random-ish I/O limits for some of the inputs.
bool limited = ((hash >> 5) & 3) == 0;
size_t wlimit = 1 + ((hash >> 8) & 0xFFFF);
size_t rlimit = 1 + ((hash >> 24) & 0xFFF);
while (true) {
dst.meta.wi = 0;
wuffs_base__io_buffer limited_dst = dst;
wuffs_base__io_buffer limited_src = *src;
if (limited) {
if (limited_dst.data.len > wlimit) {
limited_dst.data.len = wlimit;
}
if ((limited_src.meta.wi - limited_src.meta.ri) > rlimit) {
limited_src.meta.wi = limited_src.meta.ri + rlimit;
limited_src.meta.closed = false;
}
}
status = wuffs_brotli__decoder__transform_io(&dec, &limited_dst,
&limited_src, workbuf);
src->meta.ri = limited_src.meta.ri;
if (status.repr == wuffs_base__suspension__short_read) {
if (limited && (src->meta.ri < src->meta.wi)) {
continue;
}
break;
} else if (status.repr != wuffs_base__suspension__short_write) {
break;
}
if (limited_dst.meta.wi == 0) {
fprintf(stderr, "wuffs_brotli__decoder__transform_io made no progress\n");
intentional_segfault();
}
}
return wuffs_base__status__message(&status);
}