Steve French wrote: > Is your change to cifs_get_share_name code intentional? You allow the > \\server\share\path string to not be terminated by a path (ie be > \\server\share) Yes, there was error in previous version, I've stumbled upon it while mounting referrals without a path part in it. (the function still do what it did but now it correctly handles ref unc that consists only from treeName) cifs_get_share_name should return unc from ref UNC as if it was handled by mount.cifs helper (i.e. cut off the path part from unc and leave only a treeName, the path part would be specified as prefixpath option later in compose_mount_options). > On Sat, Apr 26, 2008 at 12:41 PM, Igor Mammedov <niallain@xxxxxxxxx> wrote: >> Steve French wrote: >> > I merged the first patch in your remaining cifs dfs series patch >> > 0001-Adds-to-dns_resolver-checking-if-the-server-name-is.patch into >> > the cifs tree, but isn't this section of the 2nd patch of three ie >> > 0002-fixed-compatibility-issue-with-samba-a-refferal-req.patch >> > incorrect? It is converting the slashes in the whole path (not in the >> > UNC prefix) - so escape sequences in real posix directory and file >> > names will be converted to separators. This seems wrong. We should >> > be converting just the treename part here. >> > >> > ... >> >> Fixed, now it converts only treeName part of the path: >> 0003-fixed-compatibility-issue-with-samba-a-refferal-requ.patch >> >> In my test setup it works. >> >> -- >> >> Best regards, >> >> ------------------------- >> Igor Mammedov, >> niallain "at" gmail.com >> >> >> >> >> > > > -- Best regards, ------------------------- Igor Mammedov, niallain "at" gmail.com -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
