PyTorch from_file builds oversized tensors after byte-count overflow
An integer overflow lets torch.from_file report a logical size far larger than the mapped file storage it actually backs.
By tensor#integer-overflow
llama.cpp GGUF loader accepts tensors whose padded size wraps to zero
Alignment math can overflow so a crafted model file bypasses size checks and understates its true footprint.
By tensorFour U-Boot filesystem overflows risk pre-boot code execution
Integer overflows in ZFS, SquashFS, EXT4, and a shell move command can under-allocate heap buffers through U-Boot 2026.01-rc4.
By tarpit