block: Revert "[SCSI] genhd: add a new attribute "alias" in gendisk"
authorTejun Heo <tj@kernel.org>
Thu, 10 Nov 2011 08:03:55 +0000 (09:03 +0100)
committerJens Axboe <axboe@kernel.dk>
Thu, 10 Nov 2011 08:03:55 +0000 (09:03 +0100)
commitd0985394e7fee6b25a7cc8335d45bc1c1a8ab2d3
tree44f19c0500580fd0c3ea39f355c9636bec59d677
parent1ea6b8f48918282bdca0b32a34095504ee65bab5
block: Revert "[SCSI] genhd: add a new attribute "alias" in gendisk"

This reverts commit a72c5e5eb738033938ab30d6a634b74d1d060f10.

The commit introduced alias for block devices which is intended to be
used during logging although actual usage hasn't been committed yet.
This approach adds very limited benefit (raw log might be easier to
follow) which can be trivially implemented in userland but has a lot
of problems.

It is much worse than netif renames because it doesn't rename the
actual device but just adds conveninence name which isn't used
universally or enforced.  Everything internal including device lookup
and sysfs still uses the internal name and nothing prevents two
devices from using conflicting alias - ie. sda can have sdb as its
alias.

This has been nacked by people working on device driver core, block
layer and kernel-userland interface and shouldn't have been
upstreamed.  Revert it.

 http://thread.gmane.org/gmane.linux.kernel/1155104
 http://thread.gmane.org/gmane.linux.scsi/68632
 http://thread.gmane.org/gmane.linux.scsi/69776

Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Acked-by: Kay Sievers <kay.sievers@vrfy.org>
Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: Nao Nishijima <nao.nishijima.xt@hitachi.com>
Cc: Alan Cox <alan@linux.intel.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Documentation/ABI/testing/sysfs-block
block/genhd.c
include/linux/genhd.h