The image size validation computed iw * ih * 4 in a 32-bit long, which overflows on MS-Windows and can wrap to match a short blob, leading to an out-of-bounds read when the pixels are encoded. Compute the size in 64-bit.
https://github.com/vim/vim/pull/20463
(1 file)
—
Reply to this email directly, view it on GitHub, or unsubscribe.
Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS and Android. Download it today!
You are receiving this because you are subscribed to this thread.![]()