From 553b328664030eee678b7c3f4cbe7b2e67a8285d Mon Sep 17 00:00:00 2001 From: Sam Lantinga Date: Mon, 12 Jun 2017 16:39:15 -0700 Subject: [PATCH] Fixed bug 3668 - Overflow of SDL_AudioCVT.filters with some downmixes Simon Hug There's a chance that an audio conversion from many channels to a few can use more than 9 audio filters. SDL_AudioCVT has 10 SDL_AudioFilter pointers of which one has to be the terminating NULL pointer. The SDL code has no checks for this limit. If it overflows there can be stack or heap corruption or a call to 0xa. Attached patch adds a function that checks for this limit and throws an error if it is reached. Also adds some documentation. Test parameters that trigger this issue: AUDIO_U16MSB with 224 channels at 46359 Hz V AUDIO_S16MSB with 6 channels at 27463 Hz The fuzzer program I uploaded in bug 3667 has more of them. --- include/SDL_audio.h | 14 ++++++-- src/audio/SDL_audiocvt.c | 69 ++++++++++++++++++++++++++++++++-------- 2 files changed, 67 insertions(+), 16 deletions(-) diff --git a/include/SDL_audio.h b/include/SDL_audio.h index 677525a698b0d..ada6058e9d181 100644 --- a/include/SDL_audio.h +++ b/include/SDL_audio.h @@ -184,7 +184,17 @@ typedef void (SDLCALL * SDL_AudioFilter) (struct SDL_AudioCVT * cvt, SDL_AudioFormat format); /** - * A structure to hold a set of audio conversion filters and buffers. + * \brief Upper limit of filters in SDL_AudioCVT + * + * The maximum number of SDL_AudioFilter functions in SDL_AudioCVT is + * currently limited to 9. The SDL_AudioCVT.filters array has 10 pointers, + * one of which is the terminating NULL pointer. + */ +#define SDL_AUDIOCVT_MAX_FILTERS 9 + +/** + * \struct SDL_AudioCVT + * \brief A structure to hold a set of audio conversion filters and buffers. * * Note that various parts of the conversion pipeline can take advantage * of SIMD operations (like SSE2, for example). SDL_AudioCVT doesn't require @@ -214,7 +224,7 @@ typedef struct SDL_AudioCVT int len_cvt; /**< Length of converted audio buffer */ int len_mult; /**< buffer must be len*len_mult big */ double len_ratio; /**< Given len, final size is len*len_ratio */ - SDL_AudioFilter filters[10]; /**< Filter list */ + SDL_AudioFilter filters[SDL_AUDIOCVT_MAX_FILTERS + 1]; /**< NULL-terminated list of filter functions */ int filter_index; /**< Current audio conversion function */ } SDL_AUDIOCVT_PACKED SDL_AudioCVT; diff --git a/src/audio/SDL_audiocvt.c b/src/audio/SDL_audiocvt.c index aa4501f680a5e..0e1ee63e86039 100644 --- a/src/audio/SDL_audiocvt.c +++ b/src/audio/SDL_audiocvt.c @@ -514,6 +514,19 @@ SDL_Convert_Byteswap(SDL_AudioCVT *cvt, SDL_AudioFormat format) } } +static int +SDL_AddAudioCVTFilter(SDL_AudioCVT *cvt, const SDL_AudioFilter filter) +{ + if (cvt->filter_index >= SDL_AUDIOCVT_MAX_FILTERS) { + return SDL_SetError("Too many filters needed for conversion, exceeded maximum of %d", SDL_AUDIOCVT_MAX_FILTERS); + } + if (filter == NULL) { + return SDL_SetError("Audio filter pointer is NULL"); + } + cvt->filters[cvt->filter_index++] = filter; + cvt->filters[cvt->filter_index] = NULL; /* Moving terminator */ + return 0; +} static int SDL_BuildAudioTypeCVTToFloat(SDL_AudioCVT *cvt, const SDL_AudioFormat src_fmt) @@ -521,7 +534,9 @@ SDL_BuildAudioTypeCVTToFloat(SDL_AudioCVT *cvt, const SDL_AudioFormat src_fmt) int retval = 0; /* 0 == no conversion necessary. */ if ((SDL_AUDIO_ISBIGENDIAN(src_fmt) != 0) == (SDL_BYTEORDER == SDL_LIL_ENDIAN)) { - cvt->filters[cvt->filter_index++] = SDL_Convert_Byteswap; + if (SDL_AddAudioCVTFilter(cvt, SDL_Convert_Byteswap) < 0) { + return -1; + } retval = 1; /* added a converter. */ } @@ -543,7 +558,9 @@ SDL_BuildAudioTypeCVTToFloat(SDL_AudioCVT *cvt, const SDL_AudioFormat src_fmt) return SDL_SetError("No conversion available for these formats"); } - cvt->filters[cvt->filter_index++] = filter; + if (SDL_AddAudioCVTFilter(cvt, filter) < 0) { + return -1; + } if (src_bitsize < dst_bitsize) { const int mult = (dst_bitsize / src_bitsize); cvt->len_mult *= mult; @@ -580,7 +597,9 @@ SDL_BuildAudioTypeCVTFromFloat(SDL_AudioCVT *cvt, const SDL_AudioFormat dst_fmt) return SDL_SetError("No conversion available for these formats"); } - cvt->filters[cvt->filter_index++] = filter; + if (SDL_AddAudioCVTFilter(cvt, filter) < 0) { + return -1; + } if (src_bitsize < dst_bitsize) { const int mult = (dst_bitsize / src_bitsize); cvt->len_mult *= mult; @@ -592,7 +611,9 @@ SDL_BuildAudioTypeCVTFromFloat(SDL_AudioCVT *cvt, const SDL_AudioFormat dst_fmt) } if ((SDL_AUDIO_ISBIGENDIAN(dst_fmt) != 0) == (SDL_BYTEORDER == SDL_LIL_ENDIAN)) { - cvt->filters[cvt->filter_index++] = SDL_Convert_Byteswap; + if (SDL_AddAudioCVTFilter(cvt, SDL_Convert_Byteswap) < 0) { + return -1; + } retval = 1; /* added a converter. */ } @@ -665,7 +686,9 @@ SDL_BuildAudioResampleCVT(SDL_AudioCVT * cvt, const int dst_channels, } /* Update (cvt) with filter details... */ - cvt->filters[cvt->filter_index++] = filter; + if (SDL_AddAudioCVTFilter(cvt, filter) < 0) { + return -1; + } if (src_rate < dst_rate) { const double mult = ((double) dst_rate) / ((double) src_rate); cvt->len_mult *= (int) SDL_ceil(mult); @@ -739,7 +762,9 @@ SDL_BuildAudioCVT(SDL_AudioCVT * cvt, format as well. */ if ((src_channels == 2) && (dst_channels == 2) && (src_fmt == AUDIO_S16SYS) && (dst_fmt == AUDIO_S16SYS) && (src_rate != dst_rate)) { cvt->needed = 1; - cvt->filters[cvt->filter_index++] = SDL_ResampleCVT_si16_c2; + if (SDL_AddAudioCVTFilter(cvt, SDL_ResampleCVT_si16_c2) < 0) { + return -1; + } if (src_rate < dst_rate) { const double mult = ((double) dst_rate) / ((double) src_rate); cvt->len_mult *= (int) SDL_ceil(mult); @@ -772,7 +797,9 @@ SDL_BuildAudioCVT(SDL_AudioCVT * cvt, /* just a byteswap needed? */ if ((src_fmt & ~SDL_AUDIO_MASK_ENDIAN) == (dst_fmt & ~SDL_AUDIO_MASK_ENDIAN)) { - cvt->filters[cvt->filter_index++] = SDL_Convert_Byteswap; + if (SDL_AddAudioCVTFilter(cvt, SDL_Convert_Byteswap) < 0) { + return -1; + } cvt->needed = 1; return 1; } @@ -786,36 +813,48 @@ SDL_BuildAudioCVT(SDL_AudioCVT * cvt, /* Channel conversion */ if (src_channels != dst_channels) { if ((src_channels == 1) && (dst_channels > 1)) { - cvt->filters[cvt->filter_index++] = SDL_ConvertMonoToStereo; + if (SDL_AddAudioCVTFilter(cvt, SDL_ConvertMonoToStereo) < 0) { + return -1; + } cvt->len_mult *= 2; src_channels = 2; cvt->len_ratio *= 2; } if ((src_channels == 2) && (dst_channels == 6)) { - cvt->filters[cvt->filter_index++] = SDL_ConvertStereoTo51; + if (SDL_AddAudioCVTFilter(cvt, SDL_ConvertStereoTo51) < 0) { + return -1; + } src_channels = 6; cvt->len_mult *= 3; cvt->len_ratio *= 3; } if ((src_channels == 2) && (dst_channels == 4)) { - cvt->filters[cvt->filter_index++] = SDL_ConvertStereoToQuad; + if (SDL_AddAudioCVTFilter(cvt, SDL_ConvertStereoToQuad) < 0) { + return -1; + } src_channels = 4; cvt->len_mult *= 2; cvt->len_ratio *= 2; } while ((src_channels * 2) <= dst_channels) { - cvt->filters[cvt->filter_index++] = SDL_ConvertMonoToStereo; + if (SDL_AddAudioCVTFilter(cvt, SDL_ConvertMonoToStereo) < 0) { + return -1; + } cvt->len_mult *= 2; src_channels *= 2; cvt->len_ratio *= 2; } if ((src_channels == 6) && (dst_channels <= 2)) { - cvt->filters[cvt->filter_index++] = SDL_Convert51ToStereo; + if (SDL_AddAudioCVTFilter(cvt, SDL_Convert51ToStereo) < 0) { + return -1; + } src_channels = 2; cvt->len_ratio /= 3; } if ((src_channels == 6) && (dst_channels == 4)) { - cvt->filters[cvt->filter_index++] = SDL_Convert51ToQuad; + if (SDL_AddAudioCVTFilter(cvt, SDL_Convert51ToQuad) < 0) { + return -1; + } src_channels = 4; cvt->len_ratio /= 2; } @@ -837,7 +876,9 @@ SDL_BuildAudioCVT(SDL_AudioCVT * cvt, filter = SDL_ConvertStereoToMono; } - cvt->filters[cvt->filter_index++] = filter; + if (SDL_AddAudioCVTFilter(cvt, filter) < 0) { + return -1; + } src_channels /= 2; cvt->len_ratio /= 2;