Upon nfsd shutdown any pending DRC cache is freed. DRC cache use is
tracked via a percpu counter. In the current code the percpu counter
is destroyed before. If any pending cache is still present,
percpu_counter_add is called with a percpu counter==NULL. This causes
a kernel crash.
The solution is to destroy the percpu counter after the cache is freed.

Fixes: e567b98c (“nfsd: protect concurrent access to nfsd stats counters”)
Signed-off-by: Julian Schroeder <jumaco@amazon.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

KASAN report null-ptr-deref as follows:

  BUG: KASAN: null-ptr-deref in nfsd_fill_super+0xc6/0xe0 [nfsd]
  Write of size 8 at addr 000000000000005d by task a.out/852

  CPU: 7 PID: 852 Comm: a.out Not tainted 5.18.0-rc7-dirty #66
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.14.0-1.fc33 04/01/2014
  Call Trace:
   <TASK>
   dump_stack_lvl+0x34/0x44
   kasan_report+0xab/0x120
   ? nfsd_mkdir+0x71/0x1c0 [nfsd]
   ? nfsd_fill_super+0xc6/0xe0 [nfsd]
   nfsd_fill_super+0xc6/0xe0 [nfsd]
   ? nfsd_mkdir+0x1c0/0x1c0 [nfsd]
   get_tree_keyed+0x8e/0x100
   vfs_get_tree+0x41/0xf0
   __do_sys_fsconfig+0x590/0x670
   ? fscontext_read+0x180/0x180
   ? anon_inode_getfd+0x4f/0x70
   do_syscall_64+0x35/0x80
   entry_SYSCALL_64_after_hwframe+0x44/0xae

This can be reproduce by concurrent operations:
	1. fsopen(nfsd)/fsconfig
	2. insmod/rmmod nfsd

Since the nfsd file system is registered before than nfsd_net allocated,
the caller may get the file_system_type and use the nfsd_net before it
allocated, then null-ptr-deref occurred.

So init_nfsd() should call register_filesystem() last.

Fixes: bd5ae928 ("nfsd: register pernet ops last, unregister first")
Signed-off-by: Zhang Xiaoxu <zhangxiaoxu5@huawei.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

If laundry_wq create failed, the cld notifier should be unregistered.
Signed-off-by: Zhang Xiaoxu <zhangxiaoxu5@huawei.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

I noticed CPU pipeline stalls while using perf.

Once an svc thread is scheduled and executing an RPC, no other
processes will touch svc_rqst::rq_flags. Thus bus-locked atomics are
not needed outside the svc thread scheduler.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

The flags are defined using C macros, so TRACE_DEFINE_ENUM is
unnecessary.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Instrument calls to nfsd_open_verified() to get a sense of the
filecache hit rate.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Clean up nfsd4_open() by converting a large comment at the only
call site for nfsd4_process_open2() to a kerneldoc comment in
front of that function.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Clean up: Pull case arms back one tab stop to conform every other
switch statement in fs/nfsd/nfs4proc.c.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Clean up: These relics are not likely to benefit server
administrators.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

There have been reports of races that cause NFSv4 OPEN(CREATE) to
return an error even though the requested file was created. NFSv4
does not provide a status code for this case.

To mitigate some of these problems, reorganize the NFSv4
OPEN(CREATE) logic to allocate resources before the file is actually
created, and open the new file while the parent directory is still
locked.

Two new APIs are added:

+ Add an API that works like nfsd_file_acquire() but does not open
the underlying file. The OPEN(CREATE) path can use this API when it
already has an open file.

+ Add an API that is kin to dentry_open(). NFSD needs to create a
file and grab an open "struct file *" atomically. The
alloc_empty_file() has to be done before the inode create. If it
fails (for example, because the NFS server has exceeded its
max_files limit), we avoid creating the file and can still return
an error to the NFS client.

BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=382Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: JianHong Yin <jiyin@redhat.com>

Its only caller always passes S_IFREG as the @type parameter. As an
additional clean-up, add a kerneldoc comment.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Now that its two callers have their own version-specific instance of
this function, do_nfsd_create() is no longer used.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

The NFSv3 CREATE and NFSv4 OPEN(CREATE) use cases are about to
diverge such that it makes sense to split do_nfsd_create() into one
version for NFSv3 and one for NFSv4.

As a first step, copy do_nfsd_create() to nfs3proc.c and remove
NFSv4-specific logic.

One immediate legibility benefit is that the logic for handling
NFSv3 createhow is now quite straightforward. NFSv4 createhow
has some subtleties that IMO do not belong in generic code.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

I'd like to move do_nfsd_create() out of vfs.c. Therefore
nfsd_create_setattr() needs to be made publicly visible.

Note that both call sites in vfs.c commit both the new object and
its parent directory, so just combine those common metadata commits
into nfsd_create_setattr().
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Clean up: The "out" label already invokes fh_drop_write().

Note that fh_drop_write() is already careful not to invoke
mnt_drop_write() if either it has already been done or there is
nothing to drop. Therefore no change in behavior is expected.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

As near as I can tell, mode bit masking and setting S_IFREG is
already done by do_nfsd_create() and vfs_create(). The NFSv4 path
(do_open_lookup), for example, does not bother with this special
processing.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Clean up: There is one caller. The @cpu argument can be made
implicit now that a get_cpu/put_cpu pair is no longer needed.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

svc_xprt_enqueue() disables preemption via get_cpu() and then asks
for a pool of a specific CPU (current) via svc_pool_for_cpu().
While preemption is disabled, svc_xprt_enqueue() acquires
svc_pool::sp_lock with bottom-halfs disabled, which can sleep on
PREEMPT_RT.

Disabling preemption is not required here. The pool is protected with a
lock so the following list access is safe even cross-CPU. The following
iteration through svc_pool::sp_all_threads is under RCU-readlock and
remaining operations within the loop are atomic and do not rely on
disabled-preemption.

Use raw_smp_processor_id() as the argument for the requested CPU in
svc_pool_for_cpu().
Reported-by: Mike Galbraith <umgwanakikbuti@gmail.com>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Update client_info_show to show state of courtesy client
and seconds since last renew.
Reviewed-by: J. Bruce Fields <bfields@fieldses.org>
Signed-off-by: Dai Ngo <dai.ngo@oracle.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

This patch allows expired client with lock state to be in COURTESY
state. Lock conflict with COURTESY client is resolved by the fs/lock
code using the lm_lock_expirable and lm_expire_lock callback in the
struct lock_manager_operations.

If conflict client is in COURTESY state, set it to EXPIRABLE and
schedule the laundromat to run immediately to expire the client. The
callback lm_expire_lock waits for the laundromat to flush its work
queue before returning to caller.
Reviewed-by: J. Bruce Fields <bfields@fieldses.org>
Signed-off-by: Dai Ngo <dai.ngo@oracle.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Add 2 new callbacks, lm_lock_expirable and lm_expire_lock, to
lock_manager_operations to allow the lock manager to take appropriate
action to resolve the lock conflict if possible.

A new field, lm_mod_owner, is also added to lock_manager_operations.
The lm_mod_owner is used by the fs/lock code to make sure the lock
manager module such as nfsd, is not freed while lock conflict is being
resolved.

lm_lock_expirable checks and returns true to indicate that the lock
conflict can be resolved else return false. This callback must be
called with the flc_lock held so it can not block.

lm_expire_lock is called to resolve the lock conflict if the returned
value from lm_lock_expirable is true. This callback is called without
the flc_lock held since it's allowed to block. Upon returning from
this callback, the lock conflict should be resolved and the caller is
expected to restart the conflict check from the beginnning of the list.

Lock manager, such as NFSv4 courteous server, uses this callback to
resolve conflict by destroying lock owner, or the NFSv4 courtesy client
(client that has expired but allowed to maintains its states) that owns
the lock.
Reviewed-by: J. Bruce Fields <bfields@fieldses.org>
Signed-off-by: Dai Ngo <dai.ngo@oracle.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>

Add helper locks_owner_has_blockers to check if there is any blockers
for a given lockowner.
Reviewed-by: J. Bruce Fields <bfields@fieldses.org>
Signed-off-by: Dai Ngo <dai.ngo@oracle.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>

This patch moves create/destroy of laundry_wq from nfs4_state_start
and nfs4_state_shutdown_net to init_nfsd and exit_nfsd to prevent
the laundromat from being freed while a thread is processing a
conflicting lock.
Reviewed-by: J. Bruce Fields <bfields@fieldses.org>
Signed-off-by: Dai Ngo <dai.ngo@oracle.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

This patch allows expired client with open state to be in COURTESY
state. Share/access conflict with COURTESY client is resolved by
setting COURTESY client to EXPIRABLE state, schedule laundromat
to run and returning nfserr_jukebox to the request client.
Reviewed-by: J. Bruce Fields <bfields@fieldses.org>
Signed-off-by: Dai Ngo <dai.ngo@oracle.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

This patch provides courteous server support for delegation only.
Only expired client with delegation but no conflict and no open
or lock state is allowed to be in COURTESY state.

Delegation conflict with COURTESY/EXPIRABLE client is resolved by
setting it to EXPIRABLE, queue work for the laundromat and return
delay to the caller. Conflict is resolved when the laudromat runs
and expires the EXIRABLE client while the NFS client retries the
OPEN request. Local thread request that gets conflict is doing the
retry in _break_lease.

Client in COURTESY or EXPIRABLE state is allowed to reconnect and
continues to have access to its state. Access to the nfs4_client by
the reconnecting thread and the laundromat is serialized via the
client_lock.
Reviewed-by: J. Bruce Fields <bfields@fieldses.org>
Signed-off-by: Dai Ngo <dai.ngo@oracle.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Clean up: This field is now always set to zero.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Clean up: svc_tcp_sendto() always sets rq_xprt_ctxt to NULL.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

The second parameter of wait_for_completion_interruptible_timeout()
is a jiffies value whose type is "unsigned long". Avoid an
unnecessary and potentially fraught implicit type conversion at the
wait_for_completion_interruptible_timeout() call site in
cache_wait_req().
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Cache deferral injection stress-tests the cache deferral logic as
well as upper layer protocol deferred request handlers. This
facility is for developers and professional testers to ensure
coverage of the rqst deferral code paths. To date, we haven't
had an adequate way to ensure these code paths are covered
during testing, short of temporary code changes to force their
use.

A file called /sys/kernel/debug/fail_sunrpc/ignore-cache-wait
enables administrators to disable cache deferral injection while
allowing other types of sunrpc errors to be injected. The default
setting is that cache deferral injection is enabled (ignore=false).

To enable support for cache deferral injection,
CONFIG_FAULT_INJECTION, CONFIG_FAULT_INJECTION_DEBUG_FS, and
CONFIG_SUNRPC_DEBUG must all be set to "Y".
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Replace the temporary fix from commit 4d500445 ("SUNRPC: Fix the
svc_deferred_event trace class") with the use of __sockaddr and
friends, which is the preferred solution (but only available in 5.18
and newer).
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

nfsd_splice_actor() checks that the page being spliced does not
match the previous element in the svc_rqst::rq_pages array. We
believe this is to prevent a double put_page() in cases where the
READ payload is partially contained in the xdr_buf's head buffer.

However, the NFSD READ proc functions no longer place any part of
the READ payload in the head buffer, in order to properly support
NFS/RDMA READ with Write chunks. Therefore, simplify the logic in
nfsd_splice_actor() to remove this unnecessary check.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Merge tag 'driver-core-5.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core

Pull driver core fixes from Greg KH:
 "Here is one fix, and three documentation updates for 5.18-rc7.

  The fix is for the firmware loader which resolves a long-reported
  problem where the credentials of the firmware loader could be set to a
  userspace process without enough permissions to actually load the
  firmware image. Many Android vendors have been reporting this for
  quite some time.

  The documentation updates are for the embargoed-hardware-issues.rst
  file to add a new entry, change an existing one, and sort the list to
  make changes easier in the future.

  All of these have been in linux-next for a while with no reported
  issues"

* tag 'driver-core-5.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
  Documentation/process: Update ARM contact for embargoed hardware issues
  Documentation/process: Add embargoed HW contact for Ampere Computing
  Documentation/process: Make groups alphabetical and use tabs consistently
  firmware_loader: use kernel credentials when reading firmware

Pull char/misc driver fixes from Greg KH:
 "Here are two small driver fixes for 5.18-rc7 that resolve reported
  problems:

   - slimbus driver irq bugfix

   - interconnect sync state bugfix

  Both of these have been in linux-next with no reported problems"

* tag 'char-misc-5.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
  slimbus: qcom: Fix IRQ check in qcom_slim_probe
  interconnect: Restore sync state by ignoring ipa-virt in provider count

Pull tty/serial driver fixes from Greg KH:
 "Here are some small tty n_gsm and serial driver fixes for 5.18-rc7
  that resolve reported problems. They include:

   - n_gsm fixes for reported issues

   - 8250_mtk driver fixes for some platforms

   - fsl_lpuart driver fix for reported problem.

   - digicolor driver fix for reported problem.

  All have been in linux-next for a while with no reported problems"

* tag 'tty-5.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
  fsl_lpuart: Don't enable interrupts too early
  tty: n_gsm: fix invalid gsmtty_write_room() result
  tty: n_gsm: fix mux activation issues in gsm_config()
  tty: n_gsm: fix buffer over-read in gsm_dlci_data()
  serial: 8250_mtk: Fix register address for XON/XOFF character
  serial: 8250_mtk: Make sure to select the right FEATURE_SEL
  serial: 8250_mtk: Fix UART_EFR register address
  tty/serial: digicolor: fix possible null-ptr-deref in digicolor_uart_probe()

Pull USB fixes from Greg KH:
 "Here are some small fixes for reported issues with some USB drivers.
  They include:

   - xhci fixes for xhci-mtk platform driver

   - typec driver fixes for reported problems.

   - cdc-wdm read-stuck fix

   - gadget driver fix for reported race condition

   - new usb-serial driver ids

  All of these have been in linux-next with no reported problems"

* tag 'usb-5.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
  usb: xhci-mtk: remove bandwidth budget table
  usb: xhci-mtk: fix fs isoc's transfer error
  usb: gadget: fix race when gadget driver register via ioctl
  usb: typec: tcpci_mt6360: Update for BMC PHY setting
  usb: gadget: uvc: allow for application to cleanly shutdown
  usb: typec: tcpci: Don't skip cleanup in .remove() on error
  usb: cdc-wdm: fix reading stuck on device close
  USB: serial: qcserial: add support for Sierra Wireless EM7590
  USB: serial: option: add Fibocom MA510 modem
  USB: serial: option: add Fibocom L610 modem
  USB: serial: pl2303: add device id for HP LM930 Display

Pull powerpc fix from Michael Ellerman:

 - Fix KVM PR on 32-bit, which was broken by some MMU code refactoring.

Thanks to: Alexander Graf, and Matt Evans.

* tag 'powerpc-5.18-5' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  KVM: PPC: Book3S PR: Enable MSR_DR for switch_mmu_context()

Pull x86 fix from Thomas Gleixner:
 "A single fix for the handling of unpopulated sub-pmd spaces.

  The copy & pasta from the corresponding s390 code screwed up the
  address calculation for marking the sub-pmd ranges via memset by
  omitting the ALIGN_DOWN() to calculate the proper start address.

  It's a mystery why this code is not generic and shared because there
  is nothing architecture specific in there, but that's too intrusive
  for a backportable fix"

* tag 'x86-urgent-2022-05-15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/mm: Fix marking of unused sub-pmd ranges

Pull scheduler fix from Thomas Gleixner:
 "The recent expansion of the sched switch tracepoint inserted a new
  argument in the middle of the arguments. This reordering broke BPF
  programs which relied on the old argument list.

  While tracepoints are not considered stable ABI, it's not trivial to
  make BPF cope with such a change, but it's being worked on. For now
  restore the original argument order and move the new argument to the
  end of the argument list"

* tag 'sched-urgent-2022-05-15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched/tracing: Append prev_state to tp args instead