Re: [PATCH v3 1/2] tools/bootconfig: Fix integer overflow and truncation in size checks
From: Breno Leitao
Date: Thu Sep 10 2026 - 11:15:08 EST
On Thu, Sep 10, 2026 at 10:59:24AM +0900, Masami Hiramatsu (Google) wrote:
> From: Masami Hiramatsu (Google) <mhiramat@xxxxxxxxxx>
>
> Sashiko reported that on 32-bit systems, if an attacker crafts size in
> the bootconfig footer such that adding BOOTCONFIG_FOOTER_SIZE wraps around
> (for instance, if size is 0xFFFFFFFF), the size check in
> load_xbc_from_initrd() can be bypassed:
>
> if (stat.st_size < size + BOOTCONFIG_FOOTER_SIZE) {
> pr_err("bootconfig size is too big\n");
> return -E2BIG;
> }
>
> Furthermore, on 64-bit systems with an initrd > 4.29 GB, comparing a
> corrupted 32-bit size (e.g. 0xFFFFFFFF) against
> stat.st_size - BOOTCONFIG_FOOTER_SIZE can also bypass the check if
> size is not bounded. Similarly, load_xbc_file() passes 64-bit stat.st_size
> directly into the 32-bit int size parameter of load_xbc_fd(), truncating
> large standalone files (>= 2GB).
>
> In both cases, passing 0xFFFFFFFF to load_xbc_fd() truncates to -1,
> resulting in malloc(0), an integer overflow in read(), and an
> out-of-bounds null-byte write.
>
> Fix this by:
> 1. Rejecting size > XBC_DATA_MAX or
> size > stat.st_size - BOOTCONFIG_FOOTER_SIZE in load_xbc_from_initrd().
> 2. Rejecting stat.st_size > XBC_DATA_MAX in load_xbc_file() before passing
> it to load_xbc_fd().
> 3. Checking size < 0 || size > XBC_DATA_MAX defensively in load_xbc_fd().
>
> Fixes: 950313ebf79c ("tools: bootconfig: Add bootconfig command")
> Cc: stable@xxxxxxxxxxxxxxx
> Reported-by: Sashiko <sashiko-bot@xxxxxxxxxx>
> Closes: https://lore.kernel.org/all/20260909161113.16C691F00A3A@xxxxxxxxxxxxxxx/
> Closes: https://lore.kernel.org/all/20260910010137.EE0431F000FF@xxxxxxxxxxxxxxx/
> Assisted-by: Antigravity:gemini-3.8-flash
> Signed-off-by: Masami Hiramatsu (Google) <mhiramat@xxxxxxxxxx>
Reviewed-by: Breno Leitao <leitao@xxxxxxxxxx>