avcodec/rawdec: Check bits_per_coded_sample more pedantically for 16bit cases

Fixes: shift exponent -14 is negative
Fixes: 18335/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_RAWVIDEO_fuzzer-5723267192586240
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

diff --git a/libavcodec/rawdec.c b/libavcodec/rawdec.c
index 53f5b76e93e7b61655ea4c3470e8bc4e4d08a56b..0b2d8708e6e4d68683dc7f00edb5f86f0161d2b1 100644 (file)
--- a/libavcodec/rawdec.c
+++ b/libavcodec/rawdec.c
@@ -223,7 +223,7 @@ static int raw_decode(AVCodecContext *avctx, void *data, int *got_frame,
                                                            FFALIGN(avctx->width, 16),
                                                            avctx->height, 1);
     } else {
-        context->is_lt_16bpp = av_get_bits_per_pixel(desc) == 16 && avctx->bits_per_coded_sample && avctx->bits_per_coded_sample < 16;
+        context->is_lt_16bpp = av_get_bits_per_pixel(desc) == 16 && avctx->bits_per_coded_sample > 8 && avctx->bits_per_coded_sample < 16;
         context->frame_size = av_image_get_buffer_size(avctx->pix_fmt, avctx->width,
                                                        avctx->height, 1);
     }