Commit c509f15a ("SUNRPC: Split the xdr_buf event class") added
display of the rqst's XID to the svc_xdr_buf_class. However, when
the recvfrom tracepoint fires, rq_xid has yet to be filled in with
the current XID. So it ends up recording the previous XID that was
handled by that svc_rqst.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

The macro is unused, remove it to tame gcc warning:
fs/nfsd/nfs3proc.c:702:0: warning: macro "nfsd3_fhandleres" is not used
[-Wunused-macros]
Signed-off-by: Alex Shi <alex.shi@linux.alibaba.com>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Cc: Chuck Lever <chuck.lever@oracle.com>
Cc: linux-nfs@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

An efficient way to handle multiple Read chunks is to post them all
together and then take a single completion. This is also how the
code is already structured: when the Read completion fires, all
portions of the incoming RPC message are available to be assembled.

The difficult problem is setting up the Read sink buffers so that
the server pulls the client's data into place, making subsequent
pull-up unnecessary. There are several cases:

* No Read chunks. No-op.

* One data item Read chunk. This is the fast case, where the inline
  part of the RPC-over-RDMA message becomes the head and tail, and
  the data item chunk is placed in buf->pages.

* A Position-zero Read chunk. Treated like TCP: the Read chunk is
  pulled into contiguous pages.

+ A Position-zero Read chunk with data item chunks. Treated like
  TCP: all of the Read chunks are pulled into contiguous pages.

+ Multiple data item chunks. Treated like TCP: the inline part is
  copied and the data item chunks are pulled into contiguous pages.

The "*" cases are already supported. This patch adds support for the
"+" cases.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

As a pre-requisite for handling multiple Read chunks in each Read
list, convert svc_rdma_recv_read_chunk() to use the new parsed Read
chunk list.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

I'm about to change the purpose of ri_chunklen: Instead of tracking
the number of bytes in one Read chunk, it will track the total
number of bytes in the Read list. Rename it for clarity.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

We already have trace_svcrdma_decode_rseg(), which records each
ingress Read segment. Instead of reporting those again when they
are about to be posted as RDMA Reads, let's fire one tracepoint
before posting each type of chunk.

So we'll get:

        nfsd-1998  [002]   321.666615: svcrdma_decode_rseg:  cq.id=4 cid=42 segno=0 position=0 192@0x013ca9ebfae14000:0xb0010b05
        nfsd-1998  [002]   321.666615: svcrdma_decode_rseg:  cq.id=4 cid=42 segno=1 position=0 7688@0x013ca9ebf914e000:0xb0010a05
        nfsd-1998  [002]   321.666615: svcrdma_decode_rseg:  cq.id=4 cid=42 segno=2 position=0 28@0x013ca9ebfae15000:0xb0010905
        nfsd-1998  [002]   321.666622: svcrdma_decode_rqst:  cq.id=4 cid=42 xid=0x013ca9eb vers=1 credits=128 proc=RDMA_NOMSG hdrlen=100

        nfsd-1998  [002]   321.666642: svcrdma_post_read_chunk: cq.id=3 cid=112 sqecount=3

kworker/2:1H-221   [002]   321.673949: svcrdma_wc_read:      cq.id=3 cid=112 status=SUCCESS (0/0x0)
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Clean up: These pointers are no longer used.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Refactor svc_rdma_send_reply_chunk() so that it Sends only the parts
of rq_res that do not contain a result payload.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Refactor: svc_rdma_map_reply_msg() is restructured to DMA map only
the parts of rq_res that do not contain a result payload.

This change has been tested to confirm that it does not cause a
regression in the no Write chunk and single Write chunk cases.
Multiple Write chunks have not been tested.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

When counting the number of SGEs needed to construct a Send request,
do not count result payloads. And, when copying the Reply message
into the pull-up buffer, result payloads are not to be copied to the
Send buffer.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Refactor: Instead of re-parsing the ingress RPC Call transport
header when constructing the egress RPC Reply transport header, use
the new parsed Write list and Reply chunk, which are version-
agnostic and already XDR decoded.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Refactor: Instead of re-parsing the ingress RPC Call transport
header when constructing RDMA Writes, use the new parsed chunk lists
for the Write list and Reply chunk, which are version-agnostic and
already XDR-decoded.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Refactor: Don't duplicate header decoding smarts here. Instead, use
the new parsed chunk lists.

Note that the XID sanity test is also removed. The XID is already
looked up by the cb handler, and is rejected if it's not recognized.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Refactor: Don't duplicate header decoding smarts here. Instead, use
the new parsed chunk lists.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

This simple data structure binds the location of each data payload
inside of an RPC message to the chunk that will be used to push it
to or pull it from the client.

There are several benefits to this small additional overhead:

 * It enables support for more than one chunk in incoming Read and
   Write lists.

 * It translates the version-specific on-the-wire format into a
   generic in-memory structure, enabling support for multiple
   versions of the RPC/RDMA transport protocol.

 * It enables the server to re-organize a chunk list if it needs to
   adjust where Read chunk data lands in server memory without
   altering the contents of the XDR-encoded Receive buffer.

Construction of these lists is done while sanity checking each
incoming RPC/RDMA header. Subsequent patches will make use of the
generated data structures.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Refactor: Match the control flow of svc_rdma_encode_write_list().
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

The only RPC/RDMA ordering requirement between RDMA Writes and RDMA
Sends is that the responder must post the Writes on the Send queue
before posting the Send that conveys the RPC Reply for that Write
payload.

The Linux NFS server implementation now has a transport method that
can post result Payload Writes earlier than svc_rdma_sendto:

   ->xpo_result_payload()

This gets RDMA Writes going earlier so they are more likely to be
complete at the remote end before the Send completes.

Some care must be taken with pulled-up Replies. We don't want to
push the Write chunk and then send the same payload data via Send.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Have the NFSD encoders annotate the boundaries of every
direct-data-placement eligible result data payload. Then change
svcrdma to use that annotation instead of the xdr->page_len
when handling Write chunks.

For NFSv4 on RDMA, that enables the ability to recognize multiple
result payloads per compound. This is a pre-requisite for supporting
multiple Write chunks per RPC transaction.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Clean up: "result payload" is a less confusing name for these
payloads. "READ payload" reflects only the NFS usage.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Refactor for subsequent changes.

Constify the xdr_buf argument to ensure the code here does not
modify it, and to enable callers to pass in a
"const struct xdr_buf *".

At the same time, rename the helper functions, which emit RDMA
Writes, not RDMA Sends, and add documenting comments.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Clean up: Ensure the code in rw.c does not modify the argument, and
enable callers to also use "const struct xdr_buf *".
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Clean up: This enables xdr_buf_subsegment()'s callers to pass in a
const pointer to that buffer.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

When space in the Reply chunk runs out in the middle of a segment,
we end up passing a zero-length SGL to rdma_rw_ctx_init(), and it
oopses.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Pull locking fixes from Thomas Gleixner:
 "Two more places which invoke tracing from RCU disabled regions in the
  idle path.

  Similar to the entry path the low level idle functions have to be
  non-instrumentable"

* tag 'locking-urgent-2020-11-29' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  intel_idle: Fix intel_idle() vs tracing
  sched/idle: Fix arch_cpu_idle() vs tracing

Pull irq fixes from Thomas Gleixner:
 "Two fixes for irqchip drivers:

   - Save and restore the GICV3 ITS state unconditionally on
     suspend/resume to handle firmware which fails to do so.

   - Use the correct index into the fwspec parameters to read the irq
     trigger type in the EXIU chip driver"

* tag 'irq-urgent-2020-11-29' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqchip/gic-v3-its: Unconditionally save/restore the ITS state on suspend
  irqchip/exiu: Fix the index of fwspec for IRQ type

Pull EFI fixes from Borislav Petkov:
 "More EFI fixes forwarded from Ard Biesheuvel:

   - revert efivarfs kmemleak fix again - it was a false positive

   - make CONFIG_EFI_EARLYCON depend on CONFIG_EFI explicitly so it does
     not pull in other dependencies unnecessarily if CONFIG_EFI is not
     set

   - defer attempts to load SSDT overrides from EFI vars until after the
     efivar layer is up"

* tag 'efi-urgent-for-v5.10-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  efi: EFI_EARLYCON should depend on EFI
  efivarfs: revert "fix memory leak in efivarfs_create()"
  efi/efivars: Set generic ops before loading SSDT

Pull x86 fixes from Borislav Petkov:
 "A couple of urgent fixes which accumulated this last week:

   - Two resctrl fixes to prevent refcount leaks when manipulating the
     resctrl fs (Xiaochen Shen)

   - Correct prctl(PR_GET_SPECULATION_CTRL) reporting (Anand K Mistry)

   - A fix to not lose already seen MCE severity which determines
     whether the machine can recover (Gabriele Paoloni)"

* tag 'x86_urgent_for_v5.10-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/mce: Do not overwrite no_way_out if mce_end() fails
  x86/speculation: Fix prctl() when spectre_v2_user={seccomp,prctl},ibpb
  x86/resctrl: Add necessary kernfs_put() calls to prevent refcount leak
  x86/resctrl: Remove superfluous kernfs_get() calls to prevent refcount leak

Pull RISC-V fixes from Palmer Dabbelt:
 "I've collected a handful of fixes over the past few weeks:

   - A fix to un-break the build-id argument to the vDSO build, which is
     necessary for the LLVM linker.

   - A fix to initialize the jump label subsystem, without which it (and
     all the stuff that uses it) doesn't actually function.

   - A fix to include <asm/barrier.h> from <vdso/processor.h>, without
     which some drivers won't compile"

* tag 'riscv-for-linus-5.10-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
  RISC-V: fix barrier() use in <vdso/processor.h>
  RISC-V: Add missing jump label initialization
  riscv: Explicitly specify the build id style in vDSO Makefile again

Merge tag 'kbuild-fixes-v5.10' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild

Pull Kbuild fixes from Masahiro Yamada:

 - Remove unused OBJSIZE variable.

 - Fix rootless deb-pkg build in a setgid directory.

* tag 'kbuild-fixes-v5.10' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
  builddeb: Fix rootless build in setuid/setgid directory
  kbuild: remove unused OBJSIZE

Merge tag 'perf-tools-fixes-for-v5.10-2020-11-28' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux

Pull perf tool fixes from Arnaldo Carvalho de Melo:

 - Fix die_entrypc() when DW_AT_ranges DWARF attribute not available

 - Cope with broken DWARF (missing DW_AT_declaration) generated by some
   recent gcc versions

 - Do not generate CGROUP metadata events when not asked to in 'perf
   record'

 - Use proper CPU for shadow stats in 'perf stat'

 - Update copy of libbpf's hashmap.c, silencing tools/perf build warning

 - Fix return value in 'perf diff'

* tag 'perf-tools-fixes-for-v5.10-2020-11-28' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:
  perf probe: Change function definition check due to broken DWARF
  perf probe: Fix to die_entrypc() returns error correctly
  perf stat: Use proper cpu for shadow stats
  perf record: Synthesize cgroup events only if needed
  perf diff: Fix error return value in __cmd_diff()
  perf tools: Update copy of libbpf's hashmap.c

Pull USB / PHY driver fixes from Greg KH:
 "Here are a few small USB and PHY driver fixes for 5.10-rc6. They
  include:

   - small PHY driver fixes to resolve reported issues

   - USB quirks added for "broken" devices

   - typec fixes for reported problems

   - USB gadget fixes for small issues

  Full details are in the shortlog, nothing major in here and all have
  been in linux-next with no reported issues"

* tag 'usb-5.10-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
  usb: typec: stusb160x: fix power-opmode property with typec-power-opmode
  USB: core: Change %pK for __user pointers to %px
  USB: core: Fix regression in Hercules audio card
  usb: gadget: Fix memleak in gadgetfs_fill_super
  usb: gadget: f_midi: Fix memleak in f_midi_alloc
  USB: quirks: Add USB_QUIRK_DISCONNECT_SUSPEND quirk for Lenovo A630Z TIO built-in usb-audio card
  usb: typec: qcom-pmic-typec: fix builtin build errors
  phy: mediatek: fix spelling mistake in Kconfig "veriosn" -> "version"
  phy: qualcomm: Fix 28 nm Hi-Speed USB PHY OF dependency
  phy: qualcomm: usb: Fix SuperSpeed PHY OF dependency
  phy: intel: PHY_INTEL_KEEMBAY_EMMC should depend on ARCH_KEEMBAY
  usb: cdns3: gadget: calculate TD_SIZE based on TD
  usb: cdns3: gadget: initialize link_trb as NULL
  phy: cpcap-usb: Use IRQF_ONESHOT
  phy: qcom-qmp: Initialize another pointer to NULL
  phy: tegra: xusb: Fix dangling pointer on probe failure
  phy: usb: Fix incorrect clearing of tca_drv_sel bit in SETUP reg for 7211

Pull char/misc driver fixes from Greg KH:
 "Here are some small misc driver fixes for 5.10-rc6.  They include:

   - interconnect fixes for reported problems

   - habanalabs bugfix for found issue when doing the switch fallthrough
     patches

   - MAINTAINERS file update for coresight reviewers/maintainers

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

* tag 'char-misc-5.10-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
  MAINTAINERS: Adding help for coresight subsystem
  habanalabs/gaudi: fix missing code in ECC handling
  interconnect: fix memory trashing in of_count_icc_providers()
  interconnect: qcom: qcs404: Remove GPU and display RPM IDs
  interconnect: qcom: msm8916: Remove rpm-ids from non-RPM nodes
  interconnect: qcom: msm8974: Don't boost the NoC rate during boot
  interconnect: qcom: msm8974: Prevent integer overflow in rate

Merge tag 'asm-generic-fixes-5.10-2' of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic

Pull asm-generic fix from Arnd Bergmann:
 "Add correct MAX_POSSIBLE_PHYSMEM_BITS setting to asm-generic.

  This is a single bugfix for a bug that Stefan Agner found on 32-bit
  Arm, but that exists on several other architectures"

* tag 'asm-generic-fixes-5.10-2' of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic:
  arch: pgtable: define MAX_POSSIBLE_PHYSMEM_BITS where needed

Pull ARM SoC fixes from Arnd Bergmann:
 "Another set of patches for devicetree files and Arm SoC specific
  drivers:

   - A fix for OP-TEE shared memory on non-SMP systems

   - multiple code fixes for the OMAP platform, including one regression
     for the CPSW network driver and a few runtime warning fixes

   - Some DT patches for the Rockchip RK3399 platform, in particular
     fixing the MMC device ordering that recently became
     nondeterministic with async probe.

   - Multiple DT fixes for the Tegra platform, including a regression
     fix for suspend/resume on TX2

   - A regression fix for a user-triggered fault in the NXP dpio driver

   - A regression fix for a bug caused by an earlier bug fix in the
     xilinx firmware driver

   - Two more DTC warning fixes

   - Sylvain Lemieux steps down as maintainer for the NXP LPC32xx
     platform"

* tag 'arm-soc-fixes-v5.10-3' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc: (24 commits)
  arm64: tegra: Fix Tegra234 VDK node names
  arm64: tegra: Wrong AON HSP reg property size
  arm64: tegra: Fix USB_VBUS_EN0 regulator on Jetson TX1
  arm64: tegra: Correct the UART for Jetson Xavier NX
  arm64: tegra: Disable the ACONNECT for Jetson TX2
  optee: add writeback to valid memory type
  firmware: xilinx: Use hash-table for api feature check
  firmware: xilinx: Fix SD DLL node reset issue
  soc: fsl: dpio: Get the cpumask through cpumask_of(cpu)
  ARM: dts: dra76x: m_can: fix order of clocks
  bus: ti-sysc: suppress err msg for timers used as clockevent/source
  MAINTAINERS: Remove myself as LPC32xx maintainers
  arm64: dts: qcom: clear the warnings caused by empty dma-ranges
  arm64: dts: broadcom: clear the warnings caused by empty dma-ranges
  ARM: dts: am437x-l4: fix compatible for cpsw switch dt node
  arm64: dts: rockchip: Reorder LED triggers from mmc devices on rk3399-roc-pc.
  arm64: dts: rockchip: Assign a fixed index to mmc devices on rk3399 boards.
  arm64: dts: rockchip: Remove system-power-controller from pmic on Odroid Go Advance
  arm64: dts: rockchip: fix NanoPi R2S GMAC clock name
  ARM: OMAP2+: Manage MPU state properly for omap_enter_idle_coupled()
  ...

Pull networking fixes from Jakub Kicinski:
 "Networking fixes for 5.10-rc6, including fixes from the WiFi driver,
  and CAN subtrees.

  Current release - regressions:

   - gro_cells: reduce number of synchronize_net() calls

   - ch_ktls: release a lock before jumping to an error path

  Current release - always broken:

   - tcp: Allow full IP tos/IPv6 tclass to be reflected in L3 header

  Previous release - regressions:

   - net/tls: fix missing received data after fast remote close

   - vsock/virtio: discard packets only when socket is really closed

   - sock: set sk_err to ee_errno on dequeue from errq

   - cxgb4: fix the panic caused by non smac rewrite

  Previous release - always broken:

   - tcp: fix corner cases around setting ECN with BPF selection of
     congestion control

   - tcp: fix race condition when creating child sockets from syncookies
     on loopback interface

   - usbnet: ipheth: fix connectivity with iOS 14

   - tun: honor IOCB_NOWAIT flag

   - net/packet: fix packet receive on L3 devices without visible hard
     header

   - devlink: Make sure devlink instance and port are in same net
     namespace

   - net: openvswitch: fix TTL decrement action netlink message format

   - bonding: wait for sysfs kobject destruction before freeing struct
     slave

   - net: stmmac: fix upstream patch applied to the wrong context

   - bnxt_en: fix return value and unwind in probe error paths

  Misc:

   - devlink: add extra layer of categorization to the reload stats uAPI
     before it's released"

* tag 'net-5.10-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (68 commits)
  sock: set sk_err to ee_errno on dequeue from errq
  mptcp: fix NULL ptr dereference on bad MPJ
  net: openvswitch: fix TTL decrement action netlink message format
  can: af_can: can_rx_unregister(): remove WARN() statement from list operation sanity check
  can: m_can: m_can_dev_setup(): add support for bosch mcan version 3.3.0
  can: m_can: fix nominal bitiming tseg2 min for version >= 3.1
  can: m_can: m_can_open(): remove IRQF_TRIGGER_FALLING from request_threaded_irq()'s flags
  can: mcp251xfd: mcp251xfd_probe(): bail out if no IRQ was given
  can: gs_usb: fix endianess problem with candleLight firmware
  ch_ktls: lock is not freed
  net/tls: Protect from calling tls_dev_del for TLS RX twice
  devlink: Make sure devlink instance and port are in same net namespace
  devlink: Hold rtnl lock while reading netdev attributes
  ptp: clockmatrix: bug fix for idtcm_strverscmp
  enetc: Let the hardware auto-advance the taprio base-time of 0
  gro_cells: reduce number of synchronize_net() calls
  net: stmmac: fix incorrect merge of patch upstream
  ipv6: addrlabel: fix possible memory leak in ip6addrlbl_net_init
  Documentation: netdev-FAQ: suggest how to post co-dependent series
  ibmvnic: enhance resetting status check during module exit
  ...

Pull SCSI fixes from James Bottomley:
 "Three small fixes in the UFS driver: two are for power management
  issues and the third is to fix a slew of problem in the sysfs code"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: ufs: Fix race between shutdown and runtime resume flow
  scsi: ufs: Make sure clk scaling happens only when HBA is runtime ACTIVE
  scsi: ufs: Fix unexpected values from ufshcd_read_desc_param()

Pull io_uring fixes from Jens Axboe:

 - Out of bounds fix for the cq size cap from earlier this release (Joseph)

 - iov_iter type check fix (Pavel)

 - Files grab + cancelation fix (Pavel)

* tag 'io_uring-5.10-2020-11-27' of git://git.kernel.dk/linux-block:
  io_uring: fix files grab/cancel race
  io_uring: fix ITER_BVEC check
  io_uring: fix shift-out-of-bounds when round up cq size

Pull block fix from Jens Axboe:
 "Just a single fix, for a crash in the keyslot manager"

* tag 'block-5.10-2020-11-27' of git://git.kernel.dk/linux-block:
  block/keyslot-manager: prevent crash when num_slots=1