nfs_clvnops: standardize on EOPNOTSUPP for posix_fallocate(2)

POSIX Issue 7 had allowed EINVAL for this case, but issue 8 moves it
to ENOTSUP instead.  ZFS uses the latter and we have some software in
ports already that's wanting to use that to detect the filesystem not
supporting it, so let's standardize on it.

Reviewed by:	imp, kib, rmacklem
Differential Revision:	https://reviews.freebsd.org/D53536

This commit is contained in:

Kyle Evans

2025-11-04 18:30:58 -06:00

parent fa393807c5

commit b87436d8be

1 changed files with 6 additions and 2 deletions

									
										sys/fs/nfsclient/nfs_clvnops.c
									
		+6
		-2
	
												View File
												
				@@ -3896,11 +3896,15 @@ nfs_allocate(struct vop_allocate_args *ap)

							mtx_lock(&nmp->nm_mtx);

							nmp->nm_privflag |= NFSMNTP_NOALLOCATE;

							mtx_unlock(&nmp->nm_mtx);

							error = EINVAL;

							error = EOPNOTSUPP;

						}

					} else {

						/*

						 * Pre-v4.2 NFS server that doesn't support it, or a newer

						 * NFS server that has indicated that it doesn't support it.

						 */

						mtx_unlock(&nmp->nm_mtx);

						error = EINVAL;

						error = EOPNOTSUPP;

					}

					if (attrflag != 0) {

						ret = nfscl_loadattrcache(&vp, &nfsva, NULL, 0, 1);