Remove including <linux/version.h> that don't need it.
Signed-off-by: Tian Tao <tiantao6@hisilicon.com>
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>

"*" is missed  in size determination as we are passing register set
rather than a pointer.

Fixes: dcad7854 ("sky: switch to ->regset_get()")
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@gmail.com>
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>

Reconstruct vdso framework to support future vsyscall,
vgettimeofday features. These are very important features to reduce
system calls into the kernel for performance improvement.

The patch is reference RISC-V's
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Cc: Palmer Dabbelt <palmerdabbelt@google.com>

Pick up the patch from the 'Link' made by Mark Rutland. Keep the
same with x86, arm, arm64, arc, sh, power.

Link: https://lore.kernel.org/linux-arm-kernel/1499782763-31418-1-git-send-email-mark.rutland@arm.com/Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Cc: Mark Rutland <mark.rutland@arm.com>

Sync arch/riscv/mm/fault.c into arch/csky for easy maintenance.
Here are the patches related to the modification:

cac4d1dc "riscv/mm/fault: Move no context handling to no_context()"
ac416a72 "riscv/mm/fault: Move vmalloc fault handling to vmalloc_fault()"
6c11ffbf "riscv/mm/fault: Move fault error handling to mm_fault_error()"
afb8c6fe "riscv/mm/fault: Move access error check to function"
bda281d5 "riscv/mm/fault: Simplify fault error handling"
a51271d9 "riscv/mm/fault: Move bad area handling to bad_area()"
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Palmer Dabbelt <palmerdabbelt@google.com>
Cc: Arnd Bergmann <arnd@arndb.de>

We must succeed parent's context irq status in page fault handler.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>

Similar to other architectures:
In addition to in_atomic, we also need pagefault_disabled() to
check.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>

The function update_mmu_cache could be called by user-io mapping.
There is no space of struct page in mem_map for the pte. Just
ignore the user-io mmaping in update_mmu_cache.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>

The past code only passes the FAULT_FLAG_WRITE into
handle_mm_fault and missing USER & DEFAULT & RETRY.

The patch references to arch/riscv/mm/fault.c, but there is no
FAULT_FLAG_INSTRUCTION in csky hw.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>

Print all 1024 jtlb entries and 16 iutlb entries and 16 dutlb
entries in show_regs.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>

TLB invalidate didn't contain a barrier operation in csky cpu and
we need to prevent previous PTW response after TLB invalidation
instruction. Of cause, the ASID changing also needs to take care
of the issue.

CPU0                    CPU1
===============         ===============
set_pte
sync_is()        ->     See the previous set_pte for all harts
tlbi.vas         ->     Invalidate all harts TLB entry & flush pipeline
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>

Here is the log after enabled:

[    1.798972] kmemleak: Kernel memory leak detector initialized (mem pool available: 15851)
[    1.798983] kmemleak: Automatic memory scanning thread started
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>

There is a prologue on page fault handler which marking pages dirty
and/or accessed in page attributes, but all of these have been
handled in handle_pte_fault.

 - Add flush_tlb_one in vmalloc page fault instead of prologue.
 - Using cmxchg_fixup C codes in do_page_fault instead of ASM one.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>

Fixup commit c2d1adfa9a24 "csky: Add memory layout 2.5G(user):1.5G
(kernel)". That patch broke the global bit in PTE.

C-SKY TLB's entry contain two pages:
vpn, vpn + 1 -> ppn0, ppn1

All PPN's attributes contain global bit and final global is PPN0.G
& PPN1.G. So we must keep PPN0.G and PPN1.G same in one TLB's
entry.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>

There are two implementation of spinlock in arch/csky:
 - simple one (NR_CPU = 1,2)
 - tick's one (NR_CPU = 3,4)
Remove the simple one.

There is already smp_mb in spinlock, so remove the definition of
smp_mb__after_spinlock.

Link: https://lore.kernel.org/linux-csky/20200807081253.GD2674@hirez.programming.kicks-ass.net/#tSigned-off-by: Guo Ren <guoren@linux.alibaba.com>
Cc: Peter Zijlstra <peterz@infradead.org>k
Cc: Arnd Bergmann <arnd@arndb.de>

Optimize the performance of cmpxchg by using more fine-grained
acquire/release barriers.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Paul E. McKenney <paulmck@kernel.org>

Arnd said:
I would guess that for csky, this is a mistake, as the architecture
is fairly new and should be able to implement it.

Guo reply:
The c610, c807, c810 don't support SMP, so futex_cmpxchg_enabled = 1
with asm-generic's implementation.
For c860, there is no HAVE_FUTEX_CMPXCHG and cmpxchg_inatomic/inuser
implementation, so futex_cmpxchg_enabled = 0.

Thx for point it out, we'll implement cmpxchg_inatomic/inuser for
C860 and still use asm-generic for non-smp CPUs.

LTP test:
futex_wait01    1  TPASS  :  futex_wait(): errno=ETIMEDOUT(110): Connection timed out
futex_wait01    2  TPASS  :  futex_wait(): errno=EAGAIN/EWOULDBLOCK(11): Resource temporarily unavailable
futex_wait01    3  TPASS  :  futex_wait(): errno=ETIMEDOUT(110): Connection timed out
futex_wait01    4  TPASS  :  futex_wait(): errno=EAGAIN/EWOULDBLOCK(11): Resource temporarily unavailable
futex_wait02    1  TPASS  :  futex_wait() woken up
futex_wait03    1  TPASS  :  futex_wait() woken up
futex_wait04    1  TPASS  :  futex_wait() returned -1: errno=EAGAIN/EWOULDBLOCK(11): Resource temporarily unavailable
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Cc: Arnd Bergmann <arnd@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Paul E. McKenney <paulmck@kernel.org>
Link: https://lore.kernel.org/lkml/CAK8P3a3+WaQNyJ6Za2qfu6=0mBgU1hApnRXrdp1b1=P7wwyRUg@mail.gmail.com/

Remove shareable bit for ordering barrier, just keep ordering
in current hart is enough for SMP. Using three continuous
sync.is as PTW barrier to prevent speculative PTW in 860
microarchitecture.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>

Use generic atomic implementation based on cmpxchg. So remove csky
asm/atomic.h.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Arnd Bergmann <arnd@kernel.org>
Cc: Paul E. McKenney <paulmck@kernel.org>

Current show_regs didn't display regs->usp and it confused debug.
So fixup wrong SP display and add PT_REGS.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>

Current perf init will failed with:
[    1.452433] csky-pmu: probe of soc:pmu failed with error -16

This patch fix it up with adding CPUHP_AP_PERF_CSKY_ONLINE in
cpuhotplug.h.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>

There are two ways for translating va to pa for csky:
 - Use TLB(Translate Lookup Buffer) and PTW (Page Table Walk)
 - Use SSEG0/1 (Simple Segment Mapping)

We use tlb mapping 0-2G and 3G-4G virtual address area and SSEG0/1
are for 2G-2.5G and 2.5G-3G translation. We could disable SSEG0
to use 2G-2.5G as TLB user mapping.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>

Merge tag 'kbuild-fixes-v5.11' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild

Pull Kbuild fixes from Masahiro Yamada:

 - Search for <ncurses.h> in the default header path of HOSTCC

 - Tweak the option order to be kind to old BSD awk

 - Remove 'kvmconfig' and 'xenconfig' shorthands

 - Fix documentation

* tag 'kbuild-fixes-v5.11' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
  Documentation: kbuild: Fix section reference
  kconfig: remove 'kvmconfig' and 'xenconfig' shorthands
  lib/raid6: Let $(UNROLL) rules work with macOS userland
  kconfig: Support building mconf with vendor sysroot ncurses
  kconfig: config script: add a little user help
  MAINTAINERS: adjust GCC PLUGINS after gcc-plugin.sh removal

Pull SCSI fixes from James Bottomley:
 "This is two driver fixes (megaraid_sas and hisi_sas).

  The megaraid one is a revert of a previous revert of a cpu hotplug fix
  which exposed a bug in the block layer which has been fixed in this
  merge window.

  The hisi_sas performance enhancement comes from switching to interrupt
  managed completion queues, which depended on the addition of
  devm_platform_get_irqs_affinity() which is now upstream via the irq
  tree in the last merge window"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: hisi_sas: Expose HW queues for v2 hw
  Revert "Revert "scsi: megaraid_sas: Added support for shared host tagset for cpuhotplug""

Pull block fixes from Jens Axboe:

 - Missing CRC32 selections (Arnd)

 - Fix for a merge window regression with bdev inode init (Christoph)

 - bcache fixes

 - rnbd fixes

 - NVMe pull request from Christoph:
    - fix a race in the nvme-tcp send code (Sagi Grimberg)
    - fix a list corruption in an nvme-rdma error path (Israel Rukshin)
    - avoid a possible double fetch in nvme-pci (Lalithambika Krishnakumar)
    - add the susystem NQN quirk for a Samsung driver (Gopal Tiwari)
    - fix two compiler warnings in nvme-fcloop (James Smart)
    - don't call sleeping functions from irq context in nvme-fc (James Smart)
    - remove an unused argument (Max Gurtovoy)
    - remove unused exports (Minwoo Im)

 - Use-after-free fix for partition iteration (Ming)

 - Missing blk-mq debugfs flag annotation (John)

 - Bdev freeze regression fix (Satya)

 - blk-iocost NULL pointer deref fix (Tejun)

* tag 'block-5.11-2021-01-10' of git://git.kernel.dk/linux-block: (26 commits)
  bcache: set bcache device into read-only mode for BCH_FEATURE_INCOMPAT_OBSO_LARGE_BUCKET
  bcache: introduce BCH_FEATURE_INCOMPAT_LOG_LARGE_BUCKET_SIZE for large bucket
  bcache: check unsupported feature sets for bcache register
  bcache: fix typo from SUUP to SUPP in features.h
  bcache: set pdev_set_uuid before scond loop iteration
  blk-mq-debugfs: Add decode for BLK_MQ_F_TAG_HCTX_SHARED
  block/rnbd-clt: avoid module unload race with close confirmation
  block/rnbd: Adding name to the Contributors List
  block/rnbd-clt: Fix sg table use after free
  block/rnbd-srv: Fix use after free in rnbd_srv_sess_dev_force_close
  block/rnbd: Select SG_POOL for RNBD_CLIENT
  block: pre-initialize struct block_device in bdev_alloc_inode
  fs: Fix freeze_bdev()/thaw_bdev() accounting of bd_fsfreeze_sb
  nvme: remove the unused status argument from nvme_trace_bio_complete
  nvmet-rdma: Fix list_del corruption on queue establishment failure
  nvme: unexport functions with no external caller
  nvme: avoid possible double fetch in handling CQE
  nvme-tcp: Fix possible race of io_work and direct send
  nvme-pci: mark Samsung PM1725a as IGNORE_DEV_SUBNQN
  nvme-fcloop: Fix sscanf type and list_first_entry_or_null warnings
  ...

Pull io_uring fixes from Jens Axboe:
 "A bit larger than I had hoped at this point, but it's all changes that
  will be directed towards stable anyway. In detail:

   - Fix a merge window regression on error return (Matthew)

   - Remove useless variable declaration/assignment (Ye Bin)

   - IOPOLL fixes (Pavel)

   - Exit and cancelation fixes (Pavel)

   - fasync lockdep complaint fix (Pavel)

   - Ensure SQPOLL is synchronized with creator life time (Pavel)"

* tag 'io_uring-5.11-2021-01-10' of git://git.kernel.dk/linux-block:
  io_uring: stop SQPOLL submit on creator's death
  io_uring: add warn_once for io_uring_flush()
  io_uring: inline io_uring_attempt_task_drop()
  io_uring: io_rw_reissue lockdep annotations
  io_uring: synchronise ev_posted() with waitqueues
  io_uring: dont kill fasync under completion_lock
  io_uring: trigger eventfd for IOPOLL
  io_uring: Fix return value from alloc_fixed_file_ref_node
  io_uring: Delete useless variable ‘id’ in io_prep_async_work
  io_uring: cancel more aggressively in exit_work
  io_uring: drop file refs after task cancel
  io_uring: patch up IOPOLL overflow_flush sync
  io_uring: synchronise IOPOLL on task_submit fail

Pull USB fixes from Greg KH:
 "Here are a number of small USB driver fixes for 5.11-rc3.

  Include in here are:

   - USB gadget driver fixes for reported issues

   - new usb-serial driver ids

   - dma from stack bugfixes

   - typec bugfixes

   - dwc3 bugfixes

   - xhci driver bugfixes

   - other small misc usb driver bugfixes

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

* tag 'usb-5.11-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (35 commits)
  usb: dwc3: gadget: Clear wait flag on dequeue
  usb: typec: Send uevent for num_altmodes update
  usb: typec: Fix copy paste error for NVIDIA alt-mode description
  usb: gadget: enable super speed plus
  kcov, usb: hide in_serving_softirq checks in __usb_hcd_giveback_urb
  usb: uas: Add PNY USB Portable SSD to unusual_uas
  usb: gadget: configfs: Preserve function ordering after bind failure
  usb: gadget: select CONFIG_CRC32
  usb: gadget: core: change the comment for usb_gadget_connect
  usb: gadget: configfs: Fix use-after-free issue with udc_name
  usb: dwc3: gadget: Restart DWC3 gadget when enabling pullup
  usb: usbip: vhci_hcd: protect shift size
  USB: usblp: fix DMA to stack
  USB: serial: iuu_phoenix: fix DMA from stack
  USB: serial: option: add LongSung M5710 module support
  USB: serial: option: add Quectel EM160R-GL
  USB: Gadget: dummy-hcd: Fix shift-out-of-bounds bug
  usb: gadget: f_uac2: reset wMaxPacketSize
  usb: dwc3: ulpi: Fix USB2.0 HS/FS/LS PHY suspend regression
  usb: dwc3: ulpi: Replace CPU-based busyloop with Protocol-based one
  ...

Pull staging driver fixes from Greg KH:
 "Here are some small staging driver fixes for 5.11-rc3. Nothing major,
  just resolving some reported issues:

   - cleanup some remaining mentions of the ION drivers that were
     removed in 5.11-rc1

   - comedi driver bugfix

   - two error path memory leak fixes

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

* tag 'staging-5.11-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
  staging: ION: remove some references to CONFIG_ION
  staging: mt7621-dma: Fix a resource leak in an error handling path
  Staging: comedi: Return -EFAULT if copy_to_user() fails
  staging: spmi: hisi-spmi-controller: Fix some error handling paths

Pull char/misc driver fixes from Greg KH:
 "Here are some small char and misc driver fixes for 5.11-rc3.

  The majority here are fixes for the habanalabs drivers, but also in
  here are:

   - crypto driver fix

   - pvpanic driver fix

   - updated font file

   - interconnect driver fixes

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

* tag 'char-misc-5.11-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: (26 commits)
  Fonts: font_ter16x32: Update font with new upstream Terminus release
  misc: pvpanic: Check devm_ioport_map() for NULL
  speakup: Add github repository URL and bug tracker
  MAINTAINERS: Update Georgi's email address
  crypto: asym_tpm: correct zero out potential secrets
  habanalabs: Fix memleak in hl_device_reset
  interconnect: imx8mq: Use icc_sync_state
  interconnect: imx: Remove a useless test
  interconnect: imx: Add a missing of_node_put after of_device_is_available
  interconnect: qcom: fix rpmh link failures
  habanalabs: fix order of status check
  habanalabs: register to pci shutdown callback
  habanalabs: add validation cs counter, fix misplaced counters
  habanalabs/gaudi: retry loading TPC f/w on -EINTR
  habanalabs: adjust pci controller init to new firmware
  habanalabs: update comment in hl_boot_if.h
  habanalabs/gaudi: enhance reset message
  habanalabs: full FW hard reset support
  habanalabs/gaudi: disable CGM at HW initialization
  habanalabs: Revise comment to align with mirror list name
  ...

Section 3.11 was incorrectly called 3.9, fix it.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Pull ARC fixes from Vineet Gupta:

 - Address the 2nd boot failure due to snafu in signal handling code
   (first was generic console ttynull issue)

 - misc other fixes

* tag 'arc-5.11-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc:
  ARC: [hsdk]: Enable FPU_SAVE_RESTORE
  ARC: unbork 5.11 bootup: fix snafu in _TIF_NOTIFY_SIGNAL handling
  include/soc: remove headers for EZChip NPS
  arch/arc: add copy_user_page() to <asm/page.h> to fix build error on ARC

Pull powerpc fixes from Michael Ellerman:

 - A fix for machine check handling with VMAP stack on 32-bit.

 - A clang build fix.

Thanks to Christophe Leroy and Nathan Chancellor.

* tag 'powerpc-5.11-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  powerpc: Handle .text.{hot,unlikely}.* in linker script
  powerpc/32s: Fix RTAS machine check with VMAP stack

Pull x86 fixes from Borislav Petkov:
 "As expected, fixes started trickling in after the holidays so here is
  the accumulated pile of x86 fixes for 5.11:

   - A fix for fanotify_mark() missing the conversion of x86_32 native
     syscalls which take 64-bit arguments to the compat handlers due to
     former having a general compat handler. (Brian Gerst)

   - Add a forgotten pmd page destructor call to pud_free_pmd_page()
     where a pmd page is freed. (Dan Williams)

   - Make IN/OUT insns with an u8 immediate port operand handling for
     SEV-ES guests more precise by using only the single port byte and
     not the whole s32 value of the insn decoder. (Peter Gonda)

   - Correct a straddling end range check before returning the proper
     MTRR type, when the end address is the same as top of memory.
     (Ying-Tsun Huang)

   - Change PQR_ASSOC MSR update scheme when moving a task to a resctrl
     resource group to avoid significant performance overhead with some
     resctrl workloads. (Fenghua Yu)

   - Avoid the actual task move overhead when the task is already in the
     resource group. (Fenghua Yu)"

* tag 'x86_urgent_for_v5.11_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/resctrl: Don't move a task to the same resource group
  x86/resctrl: Use an IPI instead of task_work_add() to update PQR_ASSOC MSR
  x86/mtrr: Correct the range check before performing MTRR type lookups
  x86/sev-es: Fix SEV-ES OUT/IN immediate opcode vc handling
  x86/mm: Fix leak of pmd ptlock
  fanotify: Fix sys_fanotify_mark() on native x86-32

Merge tag 'hwmon-for-v5.11-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging

Pull hwmon fixes from Guenter Roeck:

 - Fix possible KASAN issue in amd_energy driver

 - Avoid configuration problem in pwm-fan driver

 - Fix kernel-doc warning in sbtsi_temp documentation

* tag 'hwmon-for-v5.11-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
  hwmon: (amd_energy) fix allocation of hwmon_channel_info config
  hwmon: (pwm-fan) Ensure that calculation doesn't discard big period values
  hwmon: (sbtsi_temp) Fix Documenation kernel-doc warning

Pull dmaengine fixes from Vinod Koul:
 "A bunch of dmaengine driver fixes for:

   - coverity discovered issues for xilinx driver

   - qcom, gpi driver fix for undefined bhaviour and one off cleanup

   - update Peter's email for TI DMA drivers

   - one-off for idxd driver

   - resource leak fix for mediatek and milbeaut drivers"

* tag 'dmaengine-fix-5.11' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine:
  dmaengine: stm32-mdma: fix STM32_MDMA_VERY_HIGH_PRIORITY value
  dmaengine: xilinx_dma: fix mixed_enum_type coverity warning
  dmaengine: xilinx_dma: fix incompatible param warning in _child_probe()
  dmaengine: xilinx_dma: check dma_async_device_register return value
  dmaengine: qcom: fix gpi undefined behavior
  dt-bindings: dma: ti: Update maintainer and author information
  MAINTAINERS: Add entry for Texas Instruments DMA drivers
  qcom: bam_dma: Delete useless kfree code
  dmaengine: dw-edma: Fix use after free in dw_edma_alloc_chunk()
  dmaengine: milbeaut-xdmac: Fix a resource leak in the error handling path of the probe function
  dmaengine: mediatek: mtk-hsdma: Fix a resource leak in the error handling path of the probe function
  dmaengine: qcom: gpi: Fixes a format mismatch
  dmaengine: idxd: off by one in cleanup code
  dmaengine: ti: k3-udma: Fix pktdma rchan TPL level setup

Pull i2c fixes from Wolfram Sang:
 "Three driver bugfixes for I2C. Buisness as usual"

* 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
  i2c: mediatek: Fix apdma and i2c hand-shake timeout
  i2c: i801: Fix the i2c-mux gpiod_lookup_table not being properly terminated
  i2c: sprd: use a specific timeout to avoid system hang up issue

Change my email contact ahead of a likely painful eleven-month migration
to a certain cobalt enteprisey groupware cloud product that will totally
break my workflow.  Some day I may get used to having to email being
sequestered behind both claret and cerulean oath2+sms 2fa layers, but
for now I'll stick with keying in one password to receive an email vs.
the required four.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

When the creator of SQPOLL io_uring dies (i.e. sqo_task), we don't want
its internals like ->files and ->mm to be poked by the SQPOLL task, it
have never been nice and recently got racy. That can happen when the
owner undergoes destruction and SQPOLL tasks tries to submit new
requests in parallel, and so calls io_sq_thread_acquire*().

That patch halts SQPOLL submissions when sqo_task dies by introducing
sqo_dead flag. Once set, the SQPOLL task must not do any submission,
which is synchronised by uring_lock as well as the new flag.

The tricky part is to make sure that disabling always happens, that
means either the ring is discovered by creator's do_exit() -> cancel,
or if the final close() happens before it's done by the creator. The
last is guaranteed by the fact that for SQPOLL the creator task and only
it holds exactly one file note, so either it pins up to do_exit() or
removed by the creator on the final put in flush. (see comments in
uring_flush() around file->f_count == 2).

One more place that can trigger io_sq_thread_acquire_*() is
__io_req_task_submit(). Shoot off requests on sqo_dead there, even
though actually we don't need to. That's because cancellation of
sqo_task should wait for the request before going any further.

note 1: io_disable_sqo_submit() does io_ring_set_wakeup_flag() so the
caller would enter the ring to get an error, but it still doesn't
guarantee that the flag won't be cleared.

note 2: if final __userspace__ close happens not from the creator
task, the file note will pin the ring until the task dies.

Fixed: b1b6b5a3 ("kernel/io_uring: cancel io_uring before task works")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

files_cancel() should cancel all relevant requests and drop file notes,
so we should never have file notes after that, including on-exit fput
and flush. Add a WARN_ONCE to be sure.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>