linux-stable.git/include/linux/sunrpc/svc.h, branch v4.12.2

nfsd: Revert "nfsd: check for oversized NFSv2/v3 arguments"

2017-05-16T20:16:30+00:00

This reverts commit 51f567777799 "nfsd: check for oversized NFSv2/v3
arguments", which breaks support for NFSv3 ACLs.

That patch was actually an earlier draft of a fix for the problem that
was eventually fixed by e6838a29ecb "nfsd: check for oversized NFSv2/v3
arguments".  But somehow I accidentally left this earlier draft in the
branch that was part of my 2.12 pull request.

Reported-by: Eryu Guan 
Cc: stable@vger.kernel.org
Signed-off-by: J. Bruce Fields

NFSv4: Fix callback server shutdown

2017-04-27T22:00:16+00:00

We want to use kthread_stop() in order to ensure the threads are
shut down before we tear down the nfs_callback_info in nfs_callback_down.

Tested-and-reviewed-by: Kinglong Mee 
Reported-by: Kinglong Mee 
Fixes: bb6aeba736ba9 ("NFSv4.x: Switch to using svc_set_num_threads()...")
Signed-off-by: Trond Myklebust 
Signed-off-by: J. Bruce Fields

nfsd: check for oversized NFSv2/v3 arguments

2017-04-25T21:25:53+00:00

A client can append random data to the end of an NFSv2 or NFSv3 RPC call
without our complaining; we'll just stop parsing at the end of the
expected data and ignore the rest.

Encoded arguments and replies are stored together in an array of pages,
and if a call is too large it could leave inadequate space for the
reply.  This is normally OK because NFS RPC's typically have either
short arguments and long replies (like READ) or long arguments and short
replies (like WRITE).  But a client that sends an incorrectly long reply
can violate those assumptions.  This was observed to cause crashes.

So, insist that the argument not be any longer than we expect.

Also, several operations increment rq_next_page in the decode routine
before checking the argument size, which can leave rq_next_page pointing
well past the end of the page array, causing trouble later in
svc_free_pages.

As followup we may also want to rewrite the encoding routines to check
more carefully that they aren't running off the end of the page array.

Reported-by: Tuomas Haanpää 
Reported-by: Ari Kauppi 
Cc: stable@vger.kernel.org
Signed-off-by: J. Bruce Fields

nfs/nfsd/sunrpc: enforce transport requirements for NFSv4

2017-02-24T22:03:34+00:00

NFSv4 requires a transport "that is specified to avoid network
congestion" (RFC 7530, section 3.1, paragraph 2).  In practical terms,
that means that you should not run NFSv4 over UDP. The server has never
enforced that requirement, however.

This patchset fixes this by adding a new flag to the svc_version that
states that it has these transport requirements. With that, we can check
that the transport has XPT_CONG_CTRL set before processing an RPC. If it
doesn't we reject it with RPC_PROG_MISMATCH.

Signed-off-by: Jeff Layton 
Signed-off-by: J. Bruce Fields

sunrpc: turn bitfield flags in svc_version into bools

2017-02-24T20:50:08+00:00

It's just simpler to read this way, IMO. Also, no need to explicitly
set vs_hidden to false in the nfsacl ones.

Signed-off-by: Jeff Layton 
Signed-off-by: J. Bruce Fields

SUNRPC: Add a server side per-connection limit

2016-07-13T19:53:48+00:00

Allow the user to limit the number of requests serviced through a single
connection, to help prevent faster clients from starving slower clients.

Signed-off-by: Trond Myklebust 
Signed-off-by: J. Bruce Fields

mm, fs: remove remaining PAGE_CACHE_* and page_cache_{get,release} usage

2016-04-04T17:41:08+00:00

Mostly direct substitution with occasional adjustment or removing
outdated comments.

Signed-off-by: Kirill A. Shutemov 
Acked-by: Michal Hocko 
Signed-off-by: Linus Torvalds

nfsd/sunrpc: factor svc_rqst allocation and freeing from sv_nrthreads refcounting

2015-08-10T20:05:44+00:00

In later patches, we'll want to be able to allocate and free svc_rqst
structures without monkeying with the serv->sv_nrthreads refcount.

Factor those pieces out of their respective functions.

Signed-off-by: Shirley Ma 
Acked-by: Jeff Layton 
Tested-by: Shirley Ma 
Signed-off-by: J. Bruce Fields

nfsd/sunrpc: move pool_mode definitions into svc.h

2015-08-10T20:05:43+00:00

In later patches, we're going to need to allow code external to svc.c
to figure out what pool_mode is in use. Move these definitions into
svc.h to prepare for that.

Also, make the svc_pool_map object available and exported so that other
modules can peek in there to get insight into what pool mode is in use.
Likewise, export svc_pool_map_get/put function to make it safe to do so.

Signed-off-by: Shirley Ma 
Acked-by: Jeff Layton 
Tested-by: Shirley Ma 
Signed-off-by: J. Bruce Fields

nfsd/sunrpc: abstract out svc_set_num_threads to sv_ops

2015-08-10T20:05:43+00:00

Add an operation that will do setup of the service. In the case of a
classic thread-based service that means starting up threads. In the case
of a workqueue-based service, the setup will do something different.

Signed-off-by: Shirley Ma 
Acked-by: Jeff Layton 
Tested-by: Shirley Ma 
Signed-off-by: J. Bruce Fields