linux-stable.git/drivers/block/rbd.c, branch linux-4.5.y

rbd: use GFP_NOIO consistently for request allocations

2016-04-20T06:45:21+00:00

commit 2224d879c7c0f85c14183ef82eb48bd875ceb599 upstream.

As of 5a60e87603c4c533492c515b7f62578189b03c9c, RBD object request
allocations are made via rbd_obj_request_create() with GFP_NOIO.
However, subsequent OSD request allocations in rbd_osd_req_create*()
use GFP_ATOMIC.

With heavy page cache usage (e.g. OSDs running on same host as krbd
client), rbd_osd_req_create() order-1 GFP_ATOMIC allocations have been
observed to fail, where direct reclaim would have allowed GFP_NOIO
allocations to succeed.

Suggested-by: Vlastimil Babka 
Suggested-by: Neil Brown 
Signed-off-by: David Disseldorp 
Signed-off-by: Ilya Dryomov 
Signed-off-by: Greg Kroah-Hartman

rbd: delete an unnecessary check before rbd_dev_destroy()

2016-01-21T18:36:07+00:00

The rbd_dev_destroy() function tests whether its argument is NULL
and then returns immediately. Thus the test around the call is not needed.

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring 
Signed-off-by: Ilya Dryomov

rbd: don't put snap_context twice in rbd_queue_workfn()

2015-12-04T13:29:18+00:00

Commit 4e752f0ab0e8 ("rbd: access snapshot context and mapping size
safely") moved ceph_get_snap_context() out of rbd_img_request_create()
and into rbd_queue_workfn(), adding a ceph_put_snap_context() to the
error path in rbd_queue_workfn().  However, rbd_img_request_create()
consumes a ref on snapc, so calling ceph_put_snap_context() after
a successful rbd_img_request_create() leads to an extra put.  Fix it.

Cc: stable@vger.kernel.org # 3.18+
Signed-off-by: Ilya Dryomov 
Reviewed-by: Josh Durgin

rbd: remove duplicate calls to rbd_dev_mapping_clear()

2015-11-02T22:36:48+00:00

Commit d1cf5788450e ("rbd: set mapping info earlier") defined
rbd_dev_mapping_clear(), but, just a few days after, commit
f35a4dee14c3 ("rbd: set the mapping size and features later") moved
rbd_dev_mapping_set() calls and added another rbd_dev_mapping_clear()
call instead of moving the old one.  Around the same time, another
duplicate was introduced in rbd_dev_device_release() - kill both.

Signed-off-by: Ilya Dryomov

rbd: set device_type::release instead of device::release

2015-11-02T22:36:48+00:00

No point in providing an empty device_type::release callback and then
setting device::release for each rbd_dev dynamically.

Signed-off-by: Ilya Dryomov

rbd: don't free rbd_dev outside of the release callback

2015-11-02T22:36:48+00:00

struct rbd_device has struct device embedded in it, which means it's
part of kobject universe and has an unpredictable life cycle.  Freeing
its memory outside of the release callback is flawed, yet commits
200a6a8be5db ("rbd: don't destroy rbd_dev in device release function")
and 8ad42cd0c002 ("rbd: don't have device release destroy rbd_dev")
moved rbd_dev_destroy() out to rbd_dev_image_release().

This commit reverts most of that, the key points are:

- rbd_dev->dev is initialized in rbd_dev_create(), making it possible
  to use rbd_dev_destroy() - which is just a put_device() - both before
  we register with device core and after.

- rbd_dev_release() (the release callback) is the only place we
  kfree(rbd_dev).  It's also where we do module_put(), keeping the
  module unload race window as small as possible.

- We pin the module in rbd_dev_create(), but only for mapping
  rbd_dev-s.  Moving image related stuff out of struct rbd_device into
  another struct which isn't tied with sysfs and device core is long
  overdue, but until that happens, this will keep rbd module refcount
  (which users can observe with lsmod) sane.

Fixes: http://tracker.ceph.com/issues/12697

Cc: Alex Elder 
Signed-off-by: Ilya Dryomov

rbd: return -ENOMEM instead of pool id if rbd_dev_create() fails

2015-11-02T22:36:48+00:00

Returning pool id (i.e. >= 0) from a sysfs ->store() callback makes
userspace think it needs to retry the write.  Fix it - it's a leftover
from the times when the equivalent of rbd_dev_create() was the first
action in rbd_add().

Signed-off-by: Ilya Dryomov

rbd: drop null test before destroy functions

2015-11-02T22:36:47+00:00

Remove unneeded NULL test.

The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)

// 
@@ expression x; @@
-if (x != NULL) {
  \(kmem_cache_destroy\|mempool_destroy\|dma_pool_destroy\)(x);
  x = NULL;
-}
// 

Signed-off-by: Julia Lawall 
Signed-off-by: Ilya Dryomov

rbd: require stable pages if message data CRCs are enabled

2015-10-30T18:25:02+00:00

rbd requires stable pages, as it performs a crc of the page data before
they are send to the OSDs.

But since kernel 3.9 (patch 1d1d1a767206fbe5d4c69493b7e6d2a8d08cc0a0
"mm: only enforce stable page writes if the backing device requires
it") it is not assumed anymore that block devices require stable pages.

This patch sets the necessary flag to get stable pages back for rbd.

In a ceph installation that provides multiple ext4 formatted rbd
devices "bad crc" messages appeared regularly (ca 1 message every 1-2
minutes on every OSD that provided the data for the rbd) in the
OSD-logs before this patch. After this patch this messages are pretty
much gone (only ca 1-2 / month / OSD).

Cc: stable@vger.kernel.org # 3.9+, needs backporting
Signed-off-by: Ronny Hegewald 
[idryomov@gmail.com: require stable pages only in crc case, changelog]
Signed-off-by: Ilya Dryomov

rbd: prevent kernel stack blow up on rbd map

2015-10-23T16:37:24+00:00

Mapping an image with a long parent chain (e.g. image foo, whose parent
is bar, whose parent is baz, etc) currently leads to a kernel stack
overflow, due to the following recursion in the reply path:

  rbd_osd_req_callback()
    rbd_obj_request_complete()
      rbd_img_obj_callback()
        rbd_img_parent_read_callback()
          rbd_obj_request_complete()
            ...

Limit the parent chain to 16 images, which is ~5K worth of stack.  When
the above recursion is eliminated, this limit can be lifted.

Fixes: http://tracker.ceph.com/issues/12538

Cc: stable@vger.kernel.org # 3.10+, needs backporting for < 4.2
Signed-off-by: Ilya Dryomov 
Reviewed-by: Josh Durgin