Skip to contentSkip to conversation

media: v4l2-h264: Fix memcmp() size in B1 reference list comparison

PATCH Superseded Haotian Zhang 2 participants 2 messages Aug 31 Media (V4L/DVB)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 media: v4l2-h264: Fix memcmp() size in B1 reference list comparison

In v4l2_h264_build_b_ref_lists(), the B0/B1 list equality check passes the entry count builder->num_valid to memcmp() instead of a byte size. Since struct v4l2_h264_reference is two bytes (fields and index), only half of each list is compared, so distinct lists can be wrongly treated as equal and trigger an incorrect swap(b1_reflist[0], b1_reflist[1]).

Change the memcmp() size argument to sizeof(b1_reflist[0]) * builder->num_valid so that the full byte length of both reference lists is compared.

Fixes 624922a2739b ("media: v4l2-core: Add helpers to build the H264 P/B0/B1 reflists")

---
Changes in v2:
- Use sizeof(b1_reflist[0]) * builder->num_valid for the memcmp() size to match the style of the memcpy() calls in the same function.

---
 drivers/media/v4l2-core/v4l2-h264.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/media/v4l2-core/v4l2-h264.c +2 -1
index c00197d095e7..e2b7dee8ea14 100644
@@ -440,7 +440,8 @@ v4l2_h264_build_b_ref_lists(const struct v4l2_h264_reflist_builder *builder,
 	}
 
 	if (builder->num_valid > 1 &&
-	    !memcmp(b1_reflist, b0_reflist, builder->num_valid))
+	    !memcmp(b1_reflist, b0_reflist,
+			sizeof(b1_reflist[0]) * builder->num_valid)))
 		swap(b1_reflist[0], b1_reflist[1]);
 
 	print_ref_list_b(builder, b0_reflist, 0);
Signature · 1 line
2.43.0
#1 lore raw

Hi,

Le lundi 31 août 2026 à 13:54 +0800, Haotian Zhang a écrit :

Quoted 27 lines

In v4l2_h264_build_b_ref_lists(), the B0/B1 list equality check passes the entry count builder->num_valid to memcmp() instead of a byte size. Since struct v4l2_h264_reference is two bytes (fields and index), only half of each list is compared, so distinct lists can be wrongly treated as equal and trigger an incorrect swap(b1_reflist[0], b1_reflist[1]).

Change the memcmp() size argument to sizeof(b1_reflist[0]) * builder->num_valid so that the full byte length of both reference lists is compared.

Fixes 624922a2739b ("media: v4l2-core: Add helpers to build the H264 P/B0/B1 reflists")

---
Changes in v2:
- Use sizeof(b1_reflist[0]) * builder->num_valid for the memcmp() size to match the style of the memcpy() calls in the same function.

---
 drivers/media/v4l2-core/v4l2-h264.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/media/v4l2-core/v4l2-h264.c +0 -0
index c00197d095e7..e2b7dee8ea14 100644

drivers/media/v4l2-core/v4l2-h264.c v4l2_h264_build_b_ref_lists(const struct v4l2_h264_reflist_builder *builder, · line 444

Quoted 6 lines
drivers/media/v4l2-core/v4l2-h264.c +2 -1
@@ -440,7 +440,8 @@ v4l2_h264_build_b_ref_lists(const struct v4l2_h264_reflist_builder *builder,
 	}
 
 	if (builder->num_valid > 1 &&
-	    !memcmp(b1_reflist, b0_reflist, builder->num_valid))
+	    !memcmp(b1_reflist, b0_reflist,
+			sizeof(b1_reflist[0]) * builder->num_valid)))

I don't build test my suggestions, I expect submitter to do that before sending. There is too many closing parathensis here.

Nicolas

Quoted 3 lines
drivers/media/v4l2-core/v4l2-h264.c +0 -0
 		swap(b1_reflist[0], b1_reflist[1]);
 
 	print_ref_list_b(builder, b0_reflist, 0);