Re: [RFC] sysfs_rename_link() and its usage
From: Veaceslav Falico <hidden>
Date: 2014-01-14 19:15:10
Also in:
lkml
On Tue, Jan 14, 2014 at 10:21:35AM -0800, Greg KH wrote:
On Tue, Jan 14, 2014 at 06:17:40PM +0100, Veaceslav Falico wrote:quoted
Hi, I'm hitting a strange issue and/or I'm completely lost in sysfs internals. Consider having two net_device *a, *b; which are registered normally. Now, to create a link from /sys/class/net/a->name/linkname to b, one should use: sysfs_create_link(&(a->dev.kobj), &(b->dev.kobj), linkname); To remove it, even simpler: sysfs_remove_link(&(a->dev.kobj), linkname); This works like a charm. However, if I want to use (obviously, with the symlink present): sysfs_rename_link(&(a->dev.kobj), &(b->dev.kobj), oldname, newname);You forgot the namespace option to this call, what kernel version are you using here?
It's git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next , 3.13-rc6 with some networking patches on top of it. And wrt namespace - there are two functions, one is sysfs_rename_link(), which calls the second one - sysfs_rename_link_ns() with NULL namespace.
quoted
this fails with: "sysfs: ns invalid in 'a->name' for 'oldname'"Looks like the namespace for this link isn't valid.
Yep, though dunno why.
quoted
in 608 struct sysfs_dirent *sysfs_find_dirent(struct sysfs_dirent *parent_sd, ... 615 if (!!sysfs_ns_type(parent_sd) != !!ns) { 616 WARN(1, KERN_WARNING "sysfs: ns %s in '%s' for '%s'\n", 617 sysfs_ns_type(parent_sd) ? "required" : "invalid", 618 parent_sd->s_name, name); 619 return NULL; 620 } Code path: warn_slowpath_fmt+0x46/0x50 sysfs_get_dirent_ns+0x30/0x80 sysfs_find_dirent+0x84/0x110 sysfs_get_dirent_ns+0x3e/0x80 sysfs_rename_link_ns+0x54/0xd0 I have no idea what this code means. Is there any reason for it to fail (i.e. am I doing something wrong?) or I've hit a bug?What exactly are you trying to do here? Care to provide a pointer to your code somewhere?
I've tried to handle the network device renames, ended up doing it with sysfs_remove/add_link() calls. Patches: http://patchwork.ozlabs.org/patch/310798/ http://patchwork.ozlabs.org/patch/310796/ If an upper device (say, vlan, bonding, bridge etc.) enslaves another device, symlinks are created (given bridge0 and eth0): /sys/class/net/eth0/upper_bridge0 -> ../bridge0 /sys/class/net/bridge0/lower_eth0 -> ../eth0 However, in case someone renames eth0/bridge0, these symlinks should also be renamed, and sysfs_rename_link() should fit great here. But it fails.
quoted
I've tested the only user of it (bridge) - and it works fine, however it's not using its own net_device's kobject but rather its own dir.The driver core also uses this function, and it works there, so I'd blame your code :)
Yeah, I'd also like to blame it, I'm afraid of touching anything sysfs-related :).
thanks, greg k-h