PR #21768 opened by AnandajithS
URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/21768
Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/21768.patch

Currently, for **yuv411p** scaling format, when destination width is 8 and 
**SWS_FAST_BILINEAR** is enabled, the width of the destination chroma plane 
(**chrDstW**) will be calculated to be 2. This will cause a division by zero 
when calculating the horizontal increment for chroma plane (**chrXInc**).

This PR updates the small width check condition to include width 8 to ensure 
that fast bilinear mode is not used in this case and bilinear mode is used 
instead.

Fixes #21585


>From c9bfa6a8728cd50a7996365f68966fd71ecfb36b Mon Sep 17 00:00:00 2001
From: AnandajithS <[email protected]>
Date: Mon, 16 Feb 2026 02:08:36 +0530
Subject: [PATCH] libswscale/utils: prevent division by zero for chroma width 8

---
 libswscale/utils.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/libswscale/utils.c b/libswscale/utils.c
index 52095ab2c7..acea46d2e7 100644
--- a/libswscale/utils.c
+++ b/libswscale/utils.c
@@ -1208,7 +1208,7 @@ av_cold int ff_sws_init_single_context(SwsContext *sws, 
SwsFilter *srcFilter,
         return AVERROR(EINVAL);
     }
     if (flags & SWS_FAST_BILINEAR) {
-        if (srcW < 8 || dstW < 8) {
+        if (srcW < 8 || dstW <= 8) {
             flags ^= SWS_FAST_BILINEAR | SWS_BILINEAR;
             sws->flags = flags;
         }
-- 
2.52.0

_______________________________________________
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to