jorgecarleitao commented on pull request #11280: URL: https://github.com/apache/arrow/pull/11280#issuecomment-931963412
Let's say that this was the other way around, i.e. that we were converting our exporters to present buffer 1 as `i8` when before was `i32`. For an array of length 10, an older importer (e.g. `pyarrow==3`) would be reading up to ` 10 * 4` bytes from that buffer (10 * size of `i32`), when the buffer only had `10` bytes (10 * size of `i8`). This is an out of bound read, which is a security vulnerability. I was just pointing out that this change does not introduce such a vulnerability because we are moving the export from a smaller buffer (`i8`) to a larger buffer (`i32`), which makes older importers to break, but not to expose any out of bound reads. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
