Message ID | 20200510192019.14218-1-michael@niedermayer.cc |
---|---|
State | New |
Headers | show |
Series | [FFmpeg-devel,1/7] avcodec/mpc: Fix multiple numerical overflows in ff_mpc_dequantize_and_synth() | expand |
Context | Check | Description |
---|---|---|
andriy/default | pending | |
andriy/make | success | Make finished |
andriy/make_fate | fail | Make fate failed |
On Sun, May 10, 2020 at 09:20:13PM +0200, Michael Niedermayer wrote: > This uses lrintf() which is not ideal as it still can raise an error > but it is not undefined behavior. So a better solution should ideally be found > > Fixes: -2.4187e+09 is outside the range of representable values of type 'int' > Fixes: signed integer overflow: -14512205 + -2147483648 cannot be represented in type 'int' > Fixes: 20492/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_MPC7_fuzzer-5747263166480384 > > Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg > Signed-off-by: Michael Niedermayer <michael@niedermayer.cc> > --- > libavcodec/mpc.c | 8 ++++---- > 1 file changed, 4 insertions(+), 4 deletions(-) Heres a more correct but slower solution: From 74bdf7ac2ff3b72d8f31e43e41555205b9f1e37f Mon Sep 17 00:00:00 2001 From: Michael Niedermayer <michael@niedermayer.cc> Date: Sun, 10 May 2020 19:04:23 +0200 Subject: [PATCH] avcodec/mpc: Fix multiple numerical overflows in ff_mpc_dequantize_and_synth() Fixes: -2.4187e+09 is outside the range of representable values of type 'int' Fixes: signed integer overflow: -14512205 + -2147483648 cannot be represented in type 'int' Fixes: 20492/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_MPC7_fuzzer-5747263166480384 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <michael@niedermayer.cc> --- libavcodec/mpc.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/libavcodec/mpc.c b/libavcodec/mpc.c index 6cf9b9d520..e56b608d8c 100644 --- a/libavcodec/mpc.c +++ b/libavcodec/mpc.c @@ -75,17 +75,17 @@ void ff_mpc_dequantize_and_synth(MPCContext * c, int maxband, int16_t **out, j = 0; mul = (mpc_CC+1)[bands[i].res[ch]] * mpc_SCF[bands[i].scf_idx[ch][0] & 0xFF]; for(; j < 12; j++) - c->sb_samples[ch][j][i] = mul * c->Q[ch][j + off]; + c->sb_samples[ch][j][i] = av_clipf(mul * c->Q[ch][j + off], INT32_MIN, INT32_MAX); mul = (mpc_CC+1)[bands[i].res[ch]] * mpc_SCF[bands[i].scf_idx[ch][1] & 0xFF]; for(; j < 24; j++) - c->sb_samples[ch][j][i] = mul * c->Q[ch][j + off]; + c->sb_samples[ch][j][i] = av_clipf(mul * c->Q[ch][j + off], INT32_MIN, INT32_MAX); mul = (mpc_CC+1)[bands[i].res[ch]] * mpc_SCF[bands[i].scf_idx[ch][2] & 0xFF]; for(; j < 36; j++) - c->sb_samples[ch][j][i] = mul * c->Q[ch][j + off]; + c->sb_samples[ch][j][i] = av_clipf(mul * c->Q[ch][j + off], INT32_MIN, INT32_MAX); } } if(bands[i].msf){ - int t1, t2; + unsigned t1, t2; for(j = 0; j < SAMPLES_PER_BAND; j++){ t1 = c->sb_samples[0][j][i]; t2 = c->sb_samples[1][j][i];
diff --git a/libavcodec/mpc.c b/libavcodec/mpc.c index 6cf9b9d520..ebd654ba48 100644 --- a/libavcodec/mpc.c +++ b/libavcodec/mpc.c @@ -75,17 +75,17 @@ void ff_mpc_dequantize_and_synth(MPCContext * c, int maxband, int16_t **out, j = 0; mul = (mpc_CC+1)[bands[i].res[ch]] * mpc_SCF[bands[i].scf_idx[ch][0] & 0xFF]; for(; j < 12; j++) - c->sb_samples[ch][j][i] = mul * c->Q[ch][j + off]; + c->sb_samples[ch][j][i] = lrintf(mul * c->Q[ch][j + off]); mul = (mpc_CC+1)[bands[i].res[ch]] * mpc_SCF[bands[i].scf_idx[ch][1] & 0xFF]; for(; j < 24; j++) - c->sb_samples[ch][j][i] = mul * c->Q[ch][j + off]; + c->sb_samples[ch][j][i] = lrintf(mul * c->Q[ch][j + off]); mul = (mpc_CC+1)[bands[i].res[ch]] * mpc_SCF[bands[i].scf_idx[ch][2] & 0xFF]; for(; j < 36; j++) - c->sb_samples[ch][j][i] = mul * c->Q[ch][j + off]; + c->sb_samples[ch][j][i] = lrintf(mul * c->Q[ch][j + off]); } } if(bands[i].msf){ - int t1, t2; + unsigned t1, t2; for(j = 0; j < SAMPLES_PER_BAND; j++){ t1 = c->sb_samples[0][j][i]; t2 = c->sb_samples[1][j][i];
This uses lrintf() which is not ideal as it still can raise an error but it is not undefined behavior. So a better solution should ideally be found Fixes: -2.4187e+09 is outside the range of representable values of type 'int' Fixes: signed integer overflow: -14512205 + -2147483648 cannot be represented in type 'int' Fixes: 20492/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_MPC7_fuzzer-5747263166480384 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <michael@niedermayer.cc> --- libavcodec/mpc.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)