Re: [PATCH 0/2] MR7579: ntdll: Return STATUS_ACCESS_VIOLATION if ret_len is not writable

[Paul Gofman (＠gofman)]

On Fri Mar 14 17:58:21 2025 +0000, Paul Gofman wrote:
> So the right way would probably be to add the test case for this class
> with such conditions and then set *ret_len before checking length and
> returning STATUS_INFO_LENGTH_MISMATCH or STATUS_BUFFER_TOO_SMALL.

Maybe it doesn't necessarily need to be tested with invalid ret_len address / access violation specifically, it looks reasonable to set *ret_len before returning STATUS_INFO_LENGTH_MISMATCH status, not doing that is probably an oversight.