MDEV-25743 added some more use of the printf-style format "%.*s" that expects the length of a string as int and a pointer to the start of the string. Some calls were passing the return value of std::string::size() as the length parameter, although that length is size_t. On 64-bit platforms that follow the LLP64 convention (such as 64-bit Microsoft Windows), sizeof(int)<sizeof(size_t) and incorrect code could be emitted, at least in theory.
I am not aware of any real issue due to this.
Note: the sign mismatch (int is signed and size_t is unsigned) should not be an issue, because we assume that the target architecture uses 2’s complement arithmetics and that sizeof(int)>=4. Any strings that are being output should be much smaller than 2 gigabytes.