9pfs: local: clarify fchmodat_nofollow() implementation

Since fchmodat(2) on Linux doesn't support AT_SYMLINK_NOFOLLOW, we have to
implement it using workarounds. There are two different ways, depending on
whether the system supports O_PATH or not.

In the case O_PATH is supported, we rely on the behavhior of openat(2)
when passing O_NOFOLLOW | O_PATH and the file is a symbolic link. Even
if openat_file() already adds O_NOFOLLOW to the flags, this patch makes
it explicit that we need both creation flags to obtain the expected
behavior.

This is only cleanup, no functional change.

Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>

hw/9pfs/9p-local.c

@@ -349,11 +349,11 @@ static int fchmodat_nofollow(int dirfd, const char *name, mode_t mode)
         return -1;
     }
 
-    /* Access modes are ignored when O_PATH is supported. We try O_RDONLY and
-     * O_WRONLY for old-systems that don't support O_PATH.
-     */
-    fd = openat_file(dirfd, name, O_RDONLY | O_PATH_9P_UTIL, 0);
+    fd = openat_file(dirfd, name, O_RDONLY | O_PATH_9P_UTIL | O_NOFOLLOW, 0);
 #if O_PATH_9P_UTIL == 0
+    /* Fallback for systems that don't support O_PATH: we depend on the file
+     * being readable or writable.
+     */
     if (fd == -1) {
         /* In case the file is writable-only and isn't a directory. */
         if (errno == EACCES) {
@@ -368,6 +368,10 @@ static int fchmodat_nofollow(int dirfd, const char *name, mode_t mode)
     }
     ret = fchmod(fd, mode);
 #else
+    /* Access modes are ignored when O_PATH is supported. If name is a symbolic
+     * link, O_PATH | O_NOFOLLOW causes openat(2) to return a file descriptor
+     * referring to the symbolic link.
+     */
     if (fd == -1) {
         return -1;
     }