On Fri, Feb 14, 2025 at 11:59:33AM +0100, Havard Eidnes wrote: > How that ends up calling __getvfsstat90() I must admit that I do not > (at the time I wrote this sentence ... more below) understand.
My fault - I only check ktrace but not from where the calls happend. The rust libc call ends up in compat_getmntino in NetBSD's libc and there the compat handling is done, but that function expects the argument buffer to be an old struct, not struct vfsstat. It internally calls the __getvfsstat90() syscall and then converts the (correct) struct statvfs (that we would have liked to received) into the old structure. Martin
