diff mbox series

[FFmpeg-devel,3/6] avcodec/rka: Fix some integer anomalies

Message ID 20230220192929.4493-3-michael@niedermayer.cc
State Accepted
Commit d495747a9fc5fb8dcac4b23c5bdca3c5c16dd152
Headers show
Series [FFmpeg-devel,1/6] avformat/mov: Check samplesize and offset to avoid integer overflow | expand

Checks

Context Check Description
yinshiyou/make_loongarch64 success Make finished
yinshiyou/make_fate_loongarch64 success Make fate finished
andriy/make_x86 success Make finished
andriy/make_fate_x86 success Make fate finished

Commit Message

Michael Niedermayer Feb. 20, 2023, 7:29 p.m. UTC 
Fixes: left shift of negative value -3201
Fixes: integer overflow: -76470276 * -25608 cannot be represented in type 'int'
Fixes: 56052/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_RKA_fuzzer-5236218750435328

Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
---
 libavcodec/rka.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
diff mbox series

Patch

diff --git a/libavcodec/rka.c b/libavcodec/rka.c
index 461baf1e1f..994c563ffd 100644
--- a/libavcodec/rka.c
+++ b/libavcodec/rka.c
@@ -723,16 +723,16 @@  static int decode_filter(RKAContext *s, ChContext *ctx, ACoder *ac, int off, uns
             last_val = val;
             src = &ctx->buf1[off + -1];
             for (int i = 0; i < filt.size && i < 15; i++)
-                sum += filt.coeffs[i] * src[-i];
+                sum += filt.coeffs[i] * (unsigned)src[-i];
             sum = sum * 2;
             for (int i = 15; i < filt.size; i++)
-                sum += filt.coeffs[i] * src[-i];
+                sum += filt.coeffs[i] * (unsigned)src[-i];
             sum = sum >> 6;
             if (ctx->cmode == 0) {
                 if (bits == 0) {
                     ctx->buf1[off] = sum + val;
                 } else {
-                    ctx->buf1[off] = (val + (sum >> bits) << bits) +
+                    ctx->buf1[off] = (val + (sum >> bits)) * (1 << bits) +
                         (((1U << bits) - 1U) & ctx->buf1[off + -1]);
                 }
                 ctx->buf0[off] = ctx->buf1[off] + ctx->buf0[off + -1];