Merge tag 'for-6.14/block-20250118' of git://git.kernel.dk/linux

Pull block updates from Jens Axboe:

- NVMe pull requests via Keith:
- Target support for PCI-Endpoint transport (Damien)
- TCP IO queue spreading fixes (Sagi, Chaitanya)
- Target handling for "limited retry" flags (Guixen)
- Poll type fix (Yongsoo)
- Xarray storage error handling (Keisuke)
- Host memory buffer free size fix on error (Francis)

- MD pull requests via Song:
- Reintroduce md-linear (Yu Kuai)
- md-bitmap refactor and fix (Yu Kuai)
- Replace kmap_atomic with kmap_local_page (David Reaver)

- Quite a few queue freeze and debugfs deadlock fixes

Ming introduced lockdep support for this in the 6.13 kernel, and it
has (unsurprisingly) uncovered quite a few issues

- Use const attributes for IO schedulers

- Remove bio ioprio wrappers

- Fixes for stacked device atomic write support

- Refactor queue affinity helpers, in preparation for better supporting
isolated CPUs

- Cleanups of loop O_DIRECT handling

- Cleanup of BLK_MQ_F_* flags

- Add rotational support for null_blk

- Various fixes and cleanups

* tag 'for-6.14/block-20250118' of git://git.kernel.dk/linux: (106 commits)
block: Don't trim an atomic write
block: Add common atomic writes enable flag
md/md-linear: Fix a NULL vs IS_ERR() bug in linear_add()
block: limit disk max sectors to (LLONG_MAX >> 9)
block: Change blk_stack_atomic_writes_limits() unit_min check
block: Ensure start sector is aligned for stacking atomic writes
blk-mq: Move more error handling into blk_mq_submit_bio()
block: Reorder the request allocation code in blk_mq_submit_bio()
nvme: fix bogus kzalloc() return check in nvme_init_effects_log()
md/md-bitmap: move bitmap_{start, end}write to md upper layer
md/raid5: implement pers->bitmap_sector()
md: add a new callback pers->bitmap_sector()
md/md-bitmap: remove the last parameter for bimtap_ops->endwrite()
md/md-bitmap: factor behind write counters out from bitmap_{start/end}write()
md: Replace deprecated kmap_atomic() with kmap_local_page()
md: reintroduce md-linear
partitions: ldm: remove the initial kernel-doc notation
blk-cgroup: rwstat: fix kernel-doc warnings in header file
blk-cgroup: fix kernel-doc warnings in header file
nbd: fix partial sending
...

Documentation/PCI/endpoint/index.rst

···
    pci-ntb-howto
    pci-vntb-function
    pci-vntb-howto
+   pci-nvme-function
 
    function/binding/pci-test
    function/binding/pci-ntb

Documentation/PCI/endpoint/pci-nvme-function.rst

···
+.. SPDX-License-Identifier: GPL-2.0
+
+=================
+PCI NVMe Function
+=================
+
+:Author: Damien Le Moal <[e-mail]>
+
+The PCI NVMe endpoint function implements a PCI NVMe controller using the NVMe
+subsystem target core code. The driver for this function resides with the NVMe
+subsystem as drivers/nvme/target/nvmet-pciep.c.
+
+See Documentation/nvme/nvme-pci-endpoint-target.rst for more details.

Documentation/nvme/index.rst

···
+.. SPDX-License-Identifier: GPL-2.0
+
+==============
+NVMe Subsystem
+==============
+
+.. toctree::
+   :maxdepth: 2
+   :numbered:
+
+   feature-and-quirk-policy
+   nvme-pci-endpoint-target

Documentation/nvme/nvme-pci-endpoint-target.rst

···
+.. SPDX-License-Identifier: GPL-2.0
+
+=================================
+NVMe PCI Endpoint Function Target
+=================================
+
+:Author: Damien Le Moal <[e-mail]>
+
+The NVMe PCI endpoint function target driver implements a NVMe PCIe controller
+using a NVMe fabrics target controller configured with the PCI transport type.
+
+Overview
+========
+
+The NVMe PCI endpoint function target driver allows exposing a NVMe target
+controller over a PCIe link, thus implementing an NVMe PCIe device similar to a
+regular M.2 SSD. The target controller is created in the same manner as when
+using NVMe over fabrics: the controller represents the interface to an NVMe
+subsystem using a port. The port transfer type must be configured to be
+"pci". The subsystem can be configured to have namespaces backed by regular
+files or block devices, or can use NVMe passthrough to expose to the PCI host an
+existing physical NVMe device or a NVMe fabrics host controller (e.g. a NVMe TCP
+host controller).
+
+The NVMe PCI endpoint function target driver relies as much as possible on the
+NVMe target core code to parse and execute NVMe commands submitted by the PCIe
+host. However, using the PCI endpoint framework API and DMA API, the driver is
+also responsible for managing all data transfers over the PCIe link. This
+implies that the NVMe PCI endpoint function target driver implements several
+NVMe data structure management and some NVMe command parsing.
+
+1) The driver manages retrieval of NVMe commands in submission queues using DMA
+   if supported, or MMIO otherwise. Each command retrieved is then executed
+   using a work item to maximize performance with the parallel execution of
+   multiple commands on different CPUs. The driver uses a work item to
+   constantly poll the doorbell of all submission queues to detect command
+   submissions from the PCIe host.
+
+2) The driver transfers completion queues entries of completed commands to the
+   PCIe host using MMIO copy of the entries in the host completion queue.
+   After posting completion entries in a completion queue, the driver uses the
+   PCI endpoint framework API to raise an interrupt to the host to signal the
+   commands completion.
+
+3) For any command that has a data buffer, the NVMe PCI endpoint target driver
+   parses the command PRPs or SGLs lists to create a list of PCI address
+   segments representing the mapping of the command data buffer on the host.
+   The command data buffer is transferred over the PCIe link using this list of
+   PCI address segments using DMA, if supported. If DMA is not supported, MMIO
+   is used, which results in poor performance. For write commands, the command
+   data buffer is transferred from the host into a local memory buffer before
+   executing the command using the target core code. For read commands, a local
+   memory buffer is allocated to execute the command and the content of that
+   buffer is transferred to the host once the command completes.
+
+Controller Capabilities
+-----------------------
+
+The NVMe capabilities exposed to the PCIe host through the BAR 0 registers
+are almost identical to the capabilities of the NVMe target controller
+implemented by the target core code. There are some exceptions.
+
+1) The NVMe PCI endpoint target driver always sets the controller capability
+   CQR bit to request "Contiguous Queues Required". This is to facilitate the
+   mapping of a queue PCI address range to the local CPU address space.
+
+2) The doorbell stride (DSTRB) is always set to be 4B
+
+3) Since the PCI endpoint framework does not provide a way to handle PCI level
+   resets, the controller capability NSSR bit (NVM Subsystem Reset Supported)
+   is always cleared.
+
+4) The boot partition support (BPS), Persistent Memory Region Supported (PMRS)
+   and Controller Memory Buffer Supported (CMBS) capabilities are never
+   reported.
+
+Supported Features
+------------------
+
+The NVMe PCI endpoint target driver implements support for both PRPs and SGLs.
+The driver also implements IRQ vector coalescing and submission queue
+arbitration burst.
+
+The maximum number of queues and the maximum data transfer size (MDTS) are
+configurable through configfs before starting the controller. To avoid issues
+with excessive local memory usage for executing commands, MDTS defaults to 512
+KB and is limited to a maximum of 2 MB (arbitrary limit).
+
+Mimimum number of PCI Address Mapping Windows Required
+------------------------------------------------------
+
+Most PCI endpoint controllers provide a limited number of mapping windows for
+mapping a PCI address range to local CPU memory addresses. The NVMe PCI
+endpoint target controllers uses mapping windows for the following.
+
+1) One memory window for raising MSI or MSI-X interrupts
+2) One memory window for MMIO transfers
+3) One memory window for each completion queue
+
+Given the highly asynchronous nature of the NVMe PCI endpoint target driver
+operation, the memory windows as described above will generally not be used
+simultaneously, but that may happen. So a safe maximum number of completion
+queues that can be supported is equal to the total number of memory mapping
+windows of the PCI endpoint controller minus two. E.g. for an endpoint PCI
+controller with 32 outbound memory windows available, up to 30 completion
+queues can be safely operated without any risk of getting PCI address mapping
+errors due to the lack of memory windows.
+
+Maximum Number of Queue Pairs
+-----------------------------
+
+Upon binding of the NVMe PCI endpoint target driver to the PCI endpoint
+controller, BAR 0 is allocated with enough space to accommodate the admin queue
+and multiple I/O queues. The maximum of number of I/O queues pairs that can be
+supported is limited by several factors.
+
+1) The NVMe target core code limits the maximum number of I/O queues to the
+   number of online CPUs.
+2) The total number of queue pairs, including the admin queue, cannot exceed
+   the number of MSI-X or MSI vectors available.
+3) The total number of completion queues must not exceed the total number of
+   PCI mapping windows minus 2 (see above).
+
+The NVMe endpoint function driver allows configuring the maximum number of
+queue pairs through configfs.
+
+Limitations and NVMe Specification Non-Compliance
+-------------------------------------------------
+
+Similar to the NVMe target core code, the NVMe PCI endpoint target driver does
+not support multiple submission queues using the same completion queue. All
+submission queues must specify a unique completion queue.
+
+
+User Guide
+==========
+
+This section describes the hardware requirements and how to setup an NVMe PCI
+endpoint target device.
+
+Kernel Requirements
+-------------------
+
+The kernel must be compiled with the configuration options CONFIG_PCI_ENDPOINT,
+CONFIG_PCI_ENDPOINT_CONFIGFS, and CONFIG_NVME_TARGET_PCI_EPF enabled.
+CONFIG_PCI, CONFIG_BLK_DEV_NVME and CONFIG_NVME_TARGET must also be enabled
+(obviously).
+
+In addition to this, at least one PCI endpoint controller driver should be
+available for the endpoint hardware used.
+
+To facilitate testing, enabling the null-blk driver (CONFIG_BLK_DEV_NULL_BLK)
+is also recommended. With this, a simple setup using a null_blk block device
+as a subsystem namespace can be used.
+
+Hardware Requirements
+---------------------
+
+To use the NVMe PCI endpoint target driver, at least one endpoint controller
+device is required.
+
+To find the list of endpoint controller devices in the system::
+
+       # ls /sys/class/pci_epc/
+        a40000000.pcie-ep
+
+If PCI_ENDPOINT_CONFIGFS is enabled::
+
+       # ls /sys/kernel/config/pci_ep/controllers
+        a40000000.pcie-ep
+
+The endpoint board must of course also be connected to a host with a PCI cable
+with RX-TX signal swapped. If the host PCI slot used does not have
+plug-and-play capabilities, the host should be powered off when the NVMe PCI
+endpoint device is configured.
+
+NVMe Endpoint Device
+--------------------
+
+Creating an NVMe endpoint device is a two step process. First, an NVMe target
+subsystem and port must be defined. Second, the NVMe PCI endpoint device must
+be setup and bound to the subsystem and port created.
+
+Creating a NVMe Subsystem and Port
+----------------------------------
+
+Details about how to configure a NVMe target subsystem and port are outside the
+scope of this document. The following only provides a simple example of a port
+and subsystem with a single namespace backed by a null_blk device.
+
+First, make sure that configfs is enabled::
+
+       # mount -t configfs none /sys/kernel/config
+
+Next, create a null_blk device (default settings give a 250 GB device without
+memory backing). The block device created will be /dev/nullb0 by default::
+
+        # modprobe null_blk
+        # ls /dev/nullb0
+        /dev/nullb0
+
+The NVMe PCI endpoint function target driver must be loaded::
+
+        # modprobe nvmet_pci_epf
+        # lsmod | grep nvmet
+        nvmet_pci_epf          32768  0
+        nvmet                 118784  1 nvmet_pci_epf
+        nvme_core             131072  2 nvmet_pci_epf,nvmet
+
+Now, create a subsystem and a port that we will use to create a PCI target
+controller when setting up the NVMe PCI endpoint target device. In this
+example, the port is created with a maximum of 4 I/O queue pairs::
+
+        # cd /sys/kernel/config/nvmet/subsystems
+        # mkdir nvmepf.0.nqn
+        # echo -n "Linux-pci-epf" > nvmepf.0.nqn/attr_model
+        # echo "0x1b96" > nvmepf.0.nqn/attr_vendor_id
+        # echo "0x1b96" > nvmepf.0.nqn/attr_subsys_vendor_id
+        # echo 1 > nvmepf.0.nqn/attr_allow_any_host
+        # echo 4 > nvmepf.0.nqn/attr_qid_max
+
+Next, create and enable the subsystem namespace using the null_blk block
+device::
+
+        # mkdir nvmepf.0.nqn/namespaces/1
+        # echo -n "/dev/nullb0" > nvmepf.0.nqn/namespaces/1/device_path
+        # echo 1 > "nvmepf.0.nqn/namespaces/1/enable"
+
+Finally, create the target port and link it to the subsystem::
+
+        # cd /sys/kernel/config/nvmet/ports
+        # mkdir 1
+        # echo -n "pci" > 1/addr_trtype
+        # ln -s /sys/kernel/config/nvmet/subsystems/nvmepf.0.nqn \
+                /sys/kernel/config/nvmet/ports/1/subsystems/nvmepf.0.nqn
+
+Creating a NVMe PCI Endpoint Device
+-----------------------------------
+
+With the NVMe target subsystem and port ready for use, the NVMe PCI endpoint
+device can now be created and enabled. The NVMe PCI endpoint target driver
+should already be loaded (that is done automatically when the port is created)::
+
+        # ls /sys/kernel/config/pci_ep/functions
+        nvmet_pci_epf
+
+Next, create function 0::
+
+        # cd /sys/kernel/config/pci_ep/functions/nvmet_pci_epf
+        # mkdir nvmepf.0
+        # ls nvmepf.0/
+        baseclass_code    msix_interrupts   secondary
+        cache_line_size   nvme              subclass_code
+        deviceid          primary           subsys_id
+        interrupt_pin     progif_code       subsys_vendor_id
+        msi_interrupts    revid             vendorid
+
+Configure the function using any device ID (the vendor ID for the device will
+be automatically set to the same value as the NVMe target subsystem vendor
+ID)::
+
+        # cd /sys/kernel/config/pci_ep/functions/nvmet_pci_epf
+        # echo 0xBEEF > nvmepf.0/deviceid
+        # echo 32 > nvmepf.0/msix_interrupts
+
+If the PCI endpoint controller used does not support MSI-X, MSI can be
+configured instead::
+
+        # echo 32 > nvmepf.0/msi_interrupts
+
+Next, let's bind our endpoint device with the target subsystem and port that we
+created::
+
+        # echo 1 > nvmepf.0/nvme/portid
+        # echo "nvmepf.0.nqn" > nvmepf.0/nvme/subsysnqn
+
+The endpoint function can then be bound to the endpoint controller and the
+controller started::
+
+        # cd /sys/kernel/config/pci_ep
+        # ln -s functions/nvmet_pci_epf/nvmepf.0 controllers/a40000000.pcie-ep/
+        # echo 1 > controllers/a40000000.pcie-ep/start
+
+On the endpoint machine, kernel messages will show information as the NVMe
+target device and endpoint device are created and connected.
+
+.. code-block:: text
+
+        null_blk: disk nullb0 created
+        null_blk: module loaded
+        nvmet: adding nsid 1 to subsystem nvmepf.0.nqn
+        nvmet_pci_epf nvmet_pci_epf.0: PCI endpoint controller supports MSI-X, 32 vectors
+        nvmet: Created nvm controller 1 for subsystem nvmepf.0.nqn for NQN nqn.2014-08.org.nvmexpress:uuid:2ab90791-2246-4fbb-961d-4c3d5a5a0176.
+        nvmet_pci_epf nvmet_pci_epf.0: New PCI ctrl "nvmepf.0.nqn", 4 I/O queues, mdts 524288 B
+
+PCI Root-Complex Host
+---------------------
+
+Booting the PCI host will result in the initialization of the PCIe link (this
+may be signaled by the PCI endpoint driver with a kernel message). A kernel
+message on the endpoint will also signal when the host NVMe driver enables the
+device controller::
+
+        nvmet_pci_epf nvmet_pci_epf.0: Enabling controller
+
+On the host side, the NVMe PCI endpoint function target device will is
+discoverable as a PCI device, with the vendor ID and device ID as configured::
+
+        # lspci -n
+        0000:01:00.0 0108: 1b96:beef
+
+An this device will be recognized as an NVMe device with a single namespace::
+
+        # lsblk
+        NAME        MAJ:MIN RM   SIZE RO TYPE MOUNTPOINTS
+        nvme0n1     259:0    0   250G  0 disk
+
+The NVMe endpoint block device can then be used as any other regular NVMe
+namespace block device. The *nvme* command line utility can be used to get more
+detailed information about the endpoint device::
+
+        # nvme id-ctrl /dev/nvme0
+        NVME Identify Controller:
+        vid       : 0x1b96
+        ssvid     : 0x1b96
+        sn        : 94993c85650ef7bcd625
+        mn        : Linux-pci-epf
+        fr        : 6.13.0-r
+        rab       : 6
+        ieee      : 000000
+        cmic      : 0xb
+        mdts      : 7
+        cntlid    : 0x1
+        ver       : 0x20100
+        ...
+
+
+Endpoint Bindings
+=================
+
+The NVMe PCI endpoint target driver uses the PCI endpoint configfs device
+attributes as follows.
+
+================   ===========================================================
+vendorid           Ignored (the vendor id of the NVMe target subsystem is used)
+deviceid           Anything is OK (e.g. PCI_ANY_ID)
+revid              Do not care
+progif_code        Must be 0x02 (NVM Express)
+baseclass_code     Must be 0x01 (PCI_BASE_CLASS_STORAGE)
+subclass_code      Must be 0x08 (Non-Volatile Memory controller)
+cache_line_size    Do not care
+subsys_vendor_id   Ignored (the subsystem vendor id of the NVMe target subsystem
+		   is used)
+subsys_id          Anything is OK (e.g. PCI_ANY_ID)
+msi_interrupts     At least equal to the number of queue pairs desired
+msix_interrupts    At least equal to the number of queue pairs desired
+interrupt_pin      Interrupt PIN to use if MSI and MSI-X are not supported
+================   ===========================================================
+
+The NVMe PCI endpoint target function also has some specific configurable
+fields defined in the *nvme* subdirectory of the function directory. These
+fields are as follows.
+
+================   ===========================================================
+mdts_kb            Maximum data transfer size in KiB (default: 512)
+portid             The ID of the target port to use
+subsysnqn          The NQN of the target subsystem to use
+================   ===========================================================

Documentation/subsystem-apis.rst

···
    cdrom/index
    scsi/index
    target/index
+   nvme/index
 
 Other subsystems
 ----------------

arch/um/drivers/ubd_kern.c

···
 	ubd_dev->tag_set.ops = &ubd_mq_ops;
 	ubd_dev->tag_set.queue_depth = 64;
 	ubd_dev->tag_set.numa_node = NUMA_NO_NODE;
-	ubd_dev->tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
 	ubd_dev->tag_set.driver_data = ubd_dev;
 	ubd_dev->tag_set.nr_hw_queues = 1;
 

block/Makefile

···
 obj-$(CONFIG_IOSCHED_BFQ)	+= bfq.o
 
 obj-$(CONFIG_BLK_DEV_INTEGRITY) += bio-integrity.o blk-integrity.o t10-pi.o
-obj-$(CONFIG_BLK_MQ_PCI)	+= blk-mq-pci.o
-obj-$(CONFIG_BLK_MQ_VIRTIO)	+= blk-mq-virtio.o
 obj-$(CONFIG_BLK_DEV_ZONED)	+= blk-zoned.o
 obj-$(CONFIG_BLK_WBT)		+= blk-wbt.o
 obj-$(CONFIG_BLK_DEBUG_FS)	+= blk-mq-debugfs.o

block/bfq-iosched.c

···
 #define BFQ_ATTR(name) \
 	__ATTR(name, 0644, bfq_##name##_show, bfq_##name##_store)
 
-static struct elv_fs_entry bfq_attrs[] = {
+static const struct elv_fs_entry bfq_attrs[] = {
 	BFQ_ATTR(fifo_expire_sync),
 	BFQ_ATTR(fifo_expire_async),
 	BFQ_ATTR(back_seek_max),

block/bio.c

···
 
 /*
  * Try to merge a page into a segment, while obeying the hardware segment
- * size limit.  This is not for normal read/write bios, but for passthrough
- * or Zone Append operations that we can't split.
+ * size limit.
+ *
+ * This is kept around for the integrity metadata, which is still tries
+ * to build the initial bio to the hardware limit and doesn't have proper
+ * helpers to split.  Hopefully this will go away soon.
  */
 bool bvec_try_merge_hw_page(struct request_queue *q, struct bio_vec *bv,
 		struct page *page, unsigned len, unsigned offset,
···
 		return false;
 	return bvec_try_merge_page(bv, page, len, offset, same_page);
 }
-
-/**
- * bio_add_hw_page - attempt to add a page to a bio with hw constraints
- * @q: the target queue
- * @bio: destination bio
- * @page: page to add
- * @len: vec entry length
- * @offset: vec entry offset
- * @max_sectors: maximum number of sectors that can be added
- * @same_page: return if the segment has been merged inside the same page
- *
- * Add a page to a bio while respecting the hardware max_sectors, max_segment
- * and gap limitations.
- */
-int bio_add_hw_page(struct request_queue *q, struct bio *bio,
-		struct page *page, unsigned int len, unsigned int offset,
-		unsigned int max_sectors, bool *same_page)
-{
-	unsigned int max_size = max_sectors << SECTOR_SHIFT;
-
-	if (WARN_ON_ONCE(bio_flagged(bio, BIO_CLONED)))
-		return 0;
-
-	len = min3(len, max_size, queue_max_segment_size(q));
-	if (len > max_size - bio->bi_iter.bi_size)
-		return 0;
-
-	if (bio->bi_vcnt > 0) {
-		struct bio_vec *bv = &bio->bi_io_vec[bio->bi_vcnt - 1];
-
-		if (bvec_try_merge_hw_page(q, bv, page, len, offset,
-				same_page)) {
-			bio->bi_iter.bi_size += len;
-			return len;
-		}
-
-		if (bio->bi_vcnt >=
-		    min(bio->bi_max_vecs, queue_max_segments(q)))
-			return 0;
-
-		/*
-		 * If the queue doesn't support SG gaps and adding this segment
-		 * would create a gap, disallow it.
-		 */
-		if (bvec_gap_to_prev(&q->limits, bv, offset))
-			return 0;
-	}
-
-	bvec_set_page(&bio->bi_io_vec[bio->bi_vcnt], page, len, offset);
-	bio->bi_vcnt++;
-	bio->bi_iter.bi_size += len;
-	return len;
-}
-
-/**
- * bio_add_hw_folio - attempt to add a folio to a bio with hw constraints
- * @q: the target queue
- * @bio: destination bio
- * @folio: folio to add
- * @len: vec entry length
- * @offset: vec entry offset in the folio
- * @max_sectors: maximum number of sectors that can be added
- * @same_page: return if the segment has been merged inside the same folio
- *
- * Add a folio to a bio while respecting the hardware max_sectors, max_segment
- * and gap limitations.
- */
-int bio_add_hw_folio(struct request_queue *q, struct bio *bio,
-		struct folio *folio, size_t len, size_t offset,
-		unsigned int max_sectors, bool *same_page)
-{
-	if (len > UINT_MAX || offset > UINT_MAX)
-		return 0;
-	return bio_add_hw_page(q, bio, folio_page(folio, 0), len, offset,
-			       max_sectors, same_page);
-}
-
-/**
- * bio_add_pc_page	- attempt to add page to passthrough bio
- * @q: the target queue
- * @bio: destination bio
- * @page: page to add
- * @len: vec entry length
- * @offset: vec entry offset
- *
- * Attempt to add a page to the bio_vec maplist. This can fail for a
- * number of reasons, such as the bio being full or target block device
- * limitations. The target block device must allow bio's up to PAGE_SIZE,
- * so it is always possible to add a single page to an empty bio.
- *
- * This should only be used by passthrough bios.
- */
-int bio_add_pc_page(struct request_queue *q, struct bio *bio,
-		struct page *page, unsigned int len, unsigned int offset)
-{
-	bool same_page = false;
-	return bio_add_hw_page(q, bio, page, len, offset,
-			queue_max_hw_sectors(q), &same_page);
-}
-EXPORT_SYMBOL(bio_add_pc_page);
 
 /**
  * __bio_add_page - add page(s) to a bio in a new segment
···
  */
 void bio_trim(struct bio *bio, sector_t offset, sector_t size)
 {
+	/* We should never trim an atomic write */
+	if (WARN_ON_ONCE(bio->bi_opf & REQ_ATOMIC && size))
+		return;
+
 	if (WARN_ON_ONCE(offset > BIO_MAX_SECTORS || size > BIO_MAX_SECTORS ||
 			 offset + size > bio_sectors(bio)))
 		return;

block/blk-cgroup-rwstat.h

···
 /**
  * blkg_rwstat_add - add a value to a blkg_rwstat
  * @rwstat: target blkg_rwstat
- * @op: REQ_OP and flags
+ * @opf: REQ_OP and flags
  * @val: value to add
  *
  * Add @val to @rwstat.  The counters are chosen according to @rw.  The
···
 /**
  * blkg_rwstat_read - read the current values of a blkg_rwstat
  * @rwstat: blkg_rwstat to read
+ * @result: where to put the current values
  *
- * Read the current snapshot of @rwstat and return it in the aux counts.
+ * Read the current snapshot of @rwstat and return it in the @result counts.
  */
 static inline void blkg_rwstat_read(struct blkg_rwstat *rwstat,
 		struct blkg_rwstat_sample *result)

block/blk-cgroup.h

···
 
 /**
  * bio_issue_as_root_blkg - see if this bio needs to be issued as root blkg
- * @return: true if this bio needs to be submitted with the root blkg context.
+ * @bio: the target &bio
+ *
+ * Return: true if this bio needs to be submitted with the root blkg context.
  *
  * In order to avoid priority inversions we sometimes need to issue a bio as if
  * it were attached to the root blkg, and then backcharge to the actual owning
···
  * @q: request_queue of interest
  *
  * Lookup blkg for the @blkcg - @q pair.
-
+ *
  * Must be called in a RCU critical section.
  */
 static inline struct blkcg_gq *blkg_lookup(struct blkcg *blkcg,
···
 }
 
 /**
- * blkg_to_pdata - get policy private data
+ * blkg_to_pd - get policy private data
  * @blkg: blkg of interest
  * @pol: policy of interest
  *
···
 }
 
 /**
- * pdata_to_blkg - get blkg associated with policy private data
+ * pd_to_blkg - get blkg associated with policy private data
  * @pd: policy private data of interest
  *
  * @pd is policy private data.  Determine the blkg it's associated with.

block/blk-core.c

···
 		blk_mq_submit_bio(bio);
 	} else if (likely(bio_queue_enter(bio) == 0)) {
 		struct gendisk *disk = bio->bi_bdev->bd_disk;
-
-		disk->fops->submit_bio(bio);
+	
+		if ((bio->bi_opf & REQ_POLLED) &&
+		    !(disk->queue->limits.features & BLK_FEAT_POLL)) {
+			bio->bi_status = BLK_STS_NOTSUPP;
+			bio_endio(bio);
+		} else {
+			disk->fops->submit_bio(bio);
+		}
 		blk_queue_exit(disk->queue);
 	}
 
···
 		}
 	}
 
-	if (!(q->limits.features & BLK_FEAT_POLL) &&
-			(bio->bi_opf & REQ_POLLED)) {
-		bio_clear_polled(bio);
-		goto not_supported;
-	}
-
 	switch (bio_op(bio)) {
 	case REQ_OP_READ:
 		break;
···
 		return 0;
 
 	q = bdev_get_queue(bdev);
-	if (cookie == BLK_QC_T_NONE || !(q->limits.features & BLK_FEAT_POLL))
+	if (cookie == BLK_QC_T_NONE)
 		return 0;
 
 	blk_flush_plug(current->plug, false);
···
 	} else {
 		struct gendisk *disk = q->disk;
 
-		if (disk && disk->fops->poll_bio)
+		if ((q->limits.features & BLK_FEAT_POLL) && disk &&
+		    disk->fops->poll_bio)
 			ret = disk->fops->poll_bio(bio, iob, flags);
 	}
 	blk_queue_exit(q);

block/blk-integrity.c

···
 	else
 		lim.integrity.flags |= flag;
 
-	blk_mq_freeze_queue(q);
-	err = queue_limits_commit_update(q, &lim);
-	blk_mq_unfreeze_queue(q);
+	err = queue_limits_commit_update_frozen(q, &lim);
 	if (err)
 		return err;
 	return count;

block/blk-map.c

···
 			}
 		}
 
-		if (bio_add_pc_page(rq->q, bio, page, bytes, offset) < bytes) {
+		if (bio_add_page(bio, page, bytes, offset) < bytes) {
 			if (!map_data)
 				__free_page(page);
 			break;
···
 static int bio_map_user_iov(struct request *rq, struct iov_iter *iter,
 		gfp_t gfp_mask)
 {
-	iov_iter_extraction_t extraction_flags = 0;
-	unsigned int max_sectors = queue_max_hw_sectors(rq->q);
 	unsigned int nr_vecs = iov_iter_npages(iter, BIO_MAX_VECS);
 	struct bio *bio;
 	int ret;
-	int j;
 
 	if (!iov_iter_count(iter))
 		return -EINVAL;
 
 	bio = blk_rq_map_bio_alloc(rq, nr_vecs, gfp_mask);
-	if (bio == NULL)
+	if (!bio)
 		return -ENOMEM;
-
-	if (blk_queue_pci_p2pdma(rq->q))
-		extraction_flags |= ITER_ALLOW_P2PDMA;
-	if (iov_iter_extract_will_pin(iter))
-		bio_set_flag(bio, BIO_PAGE_PINNED);
-
-	while (iov_iter_count(iter)) {
-		struct page *stack_pages[UIO_FASTIOV];
-		struct page **pages = stack_pages;
-		ssize_t bytes;
-		size_t offs;
-		int npages;
-
-		if (nr_vecs > ARRAY_SIZE(stack_pages))
-			pages = NULL;
-
-		bytes = iov_iter_extract_pages(iter, &pages, LONG_MAX,
-					       nr_vecs, extraction_flags, &offs);
-		if (unlikely(bytes <= 0)) {
-			ret = bytes ? bytes : -EFAULT;
-			goto out_unmap;
-		}
-
-		npages = DIV_ROUND_UP(offs + bytes, PAGE_SIZE);
-
-		if (unlikely(offs & queue_dma_alignment(rq->q)))
-			j = 0;
-		else {
-			for (j = 0; j < npages; j++) {
-				struct page *page = pages[j];
-				unsigned int n = PAGE_SIZE - offs;
-				bool same_page = false;
-
-				if (n > bytes)
-					n = bytes;
-
-				if (!bio_add_hw_page(rq->q, bio, page, n, offs,
-						     max_sectors, &same_page))
-					break;
-
-				if (same_page)
-					bio_release_page(bio, page);
-				bytes -= n;
-				offs = 0;
-			}
-		}
-		/*
-		 * release the pages we didn't map into the bio, if any
-		 */
-		while (j < npages)
-			bio_release_page(bio, pages[j++]);
-		if (pages != stack_pages)
-			kvfree(pages);
-		/* couldn't stuff something into bio? */
-		if (bytes) {
-			iov_iter_revert(iter, bytes);
-			break;
-		}
-	}
-
+	ret = bio_iov_iter_get_pages(bio, iter);
+	if (ret)
+		goto out_put;
 	ret = blk_rq_append_bio(rq, bio);
 	if (ret)
-		goto out_unmap;
+		goto out_release;
 	return 0;
 
- out_unmap:
+out_release:
 	bio_release_pages(bio, false);
+out_put:
 	blk_mq_map_bio_put(bio);
 	return ret;
 }
···
 			page = virt_to_page(data);
 		else
 			page = vmalloc_to_page(data);
-		if (bio_add_pc_page(q, bio, page, bytes,
-				    offset) < bytes) {
+		if (bio_add_page(bio, page, bytes, offset) < bytes) {
 			/* we don't support partial mappings */
 			bio_uninit(bio);
 			kfree(bio);
···
 		if (!reading)
 			memcpy(page_address(page), p, bytes);
 
-		if (bio_add_pc_page(q, bio, page, bytes, 0) < bytes)
+		if (bio_add_page(bio, page, bytes, 0) < bytes)
 			break;
 
 		len -= bytes;
···
  */
 int blk_rq_append_bio(struct request *rq, struct bio *bio)
 {
-	struct bvec_iter iter;
-	struct bio_vec bv;
+	const struct queue_limits *lim = &rq->q->limits;
+	unsigned int max_bytes = lim->max_hw_sectors << SECTOR_SHIFT;
 	unsigned int nr_segs = 0;
+	int ret;
 
-	bio_for_each_bvec(bv, bio, iter)
-		nr_segs++;
+	/* check that the data layout matches the hardware restrictions */
+	ret = bio_split_rw_at(bio, lim, &nr_segs, max_bytes);
+	if (ret) {
+		/* if we would have to split the bio, copy instead */
+		if (ret > 0)
+			ret = -EREMOTEIO;
+		return ret;
+	}
 
-	if (!rq->bio) {
-		blk_rq_bio_prep(rq, bio, nr_segs);
-	} else {
+	if (rq->bio) {
 		if (!ll_back_merge_fn(rq, bio, nr_segs))
 			return -EINVAL;
 		rq->biotail->bi_next = bio;
 		rq->biotail = bio;
-		rq->__data_len += (bio)->bi_iter.bi_size;
+		rq->__data_len += bio->bi_iter.bi_size;
 		bio_crypt_free_ctx(bio);
+		return 0;
 	}
 
+	rq->nr_phys_segments = nr_segs;
+	rq->bio = rq->biotail = bio;
+	rq->__data_len = bio->bi_iter.bi_size;
 	return 0;
 }
 EXPORT_SYMBOL(blk_rq_append_bio);
···
 /* Prepare bio for passthrough IO given ITER_BVEC iter */
 static int blk_rq_map_user_bvec(struct request *rq, const struct iov_iter *iter)
 {
-	const struct queue_limits *lim = &rq->q->limits;
-	unsigned int max_bytes = lim->max_hw_sectors << SECTOR_SHIFT;
-	unsigned int nsegs;
+	unsigned int max_bytes = rq->q->limits.max_hw_sectors << SECTOR_SHIFT;
 	struct bio *bio;
 	int ret;
 
···
 		return -ENOMEM;
 	bio_iov_bvec_set(bio, iter);
 
-	/* check that the data layout matches the hardware restrictions */
-	ret = bio_split_rw_at(bio, lim, &nsegs, max_bytes);
-	if (ret) {
-		/* if we would have to split the bio, copy instead */
-		if (ret > 0)
-			ret = -EREMOTEIO;
+	ret = blk_rq_append_bio(rq, bio);
+	if (ret)
 		blk_mq_map_bio_put(bio);
-		return ret;
-	}
-
-	blk_rq_bio_prep(rq, bio, nsegs);
-	return 0;
+	return ret;
 }
 
 /**
···
 			ret = bio_copy_user_iov(rq, map_data, &i, gfp_mask);
 		else
 			ret = bio_map_user_iov(rq, &i, gfp_mask);
-		if (ret)
+		if (ret) {
+			if (ret == -EREMOTEIO)
+				ret = -EINVAL;
 			goto unmap_rq;
+		}
 		if (!bio)
 			bio = rq->bio;
 	} while (iov_iter_count(&i));

block/blk-merge.c

···
 	return nr_phys_segs;
 }
 
+struct phys_vec {
+	phys_addr_t	paddr;
+	u32		len;
+};
+
+static bool blk_map_iter_next(struct request *req,
+		struct req_iterator *iter, struct phys_vec *vec)
+{
+	unsigned int max_size;
+	struct bio_vec bv;
+
+	if (req->rq_flags & RQF_SPECIAL_PAYLOAD) {
+		if (!iter->bio)
+			return false;
+		vec->paddr = bvec_phys(&req->special_vec);
+		vec->len = req->special_vec.bv_len;
+		iter->bio = NULL;
+		return true;
+	}
+
+	if (!iter->iter.bi_size)
+		return false;
+
+	bv = mp_bvec_iter_bvec(iter->bio->bi_io_vec, iter->iter);
+	vec->paddr = bvec_phys(&bv);
+	max_size = get_max_segment_size(&req->q->limits, vec->paddr, UINT_MAX);
+	bv.bv_len = min(bv.bv_len, max_size);
+	bio_advance_iter_single(iter->bio, &iter->iter, bv.bv_len);
+
+	/*
+	 * If we are entirely done with this bi_io_vec entry, check if the next
+	 * one could be merged into it.  This typically happens when moving to
+	 * the next bio, but some callers also don't pack bvecs tight.
+	 */
+	while (!iter->iter.bi_size || !iter->iter.bi_bvec_done) {
+		struct bio_vec next;
+
+		if (!iter->iter.bi_size) {
+			if (!iter->bio->bi_next)
+				break;
+			iter->bio = iter->bio->bi_next;
+			iter->iter = iter->bio->bi_iter;
+		}
+
+		next = mp_bvec_iter_bvec(iter->bio->bi_io_vec, iter->iter);
+		if (bv.bv_len + next.bv_len > max_size ||
+		    !biovec_phys_mergeable(req->q, &bv, &next))
+			break;
+
+		bv.bv_len += next.bv_len;
+		bio_advance_iter_single(iter->bio, &iter->iter, next.bv_len);
+	}
+
+	vec->len = bv.bv_len;
+	return true;
+}
+
 static inline struct scatterlist *blk_next_sg(struct scatterlist **sg,
 		struct scatterlist *sglist)
 {
···
 	return sg_next(*sg);
 }
 
-static unsigned blk_bvec_map_sg(struct request_queue *q,
-		struct bio_vec *bvec, struct scatterlist *sglist,
-		struct scatterlist **sg)
-{
-	unsigned nbytes = bvec->bv_len;
-	unsigned nsegs = 0, total = 0;
-
-	while (nbytes > 0) {
-		unsigned offset = bvec->bv_offset + total;
-		unsigned len = get_max_segment_size(&q->limits,
-				bvec_phys(bvec) + total, nbytes);
-		struct page *page = bvec->bv_page;
-
-		/*
-		 * Unfortunately a fair number of drivers barf on scatterlists
-		 * that have an offset larger than PAGE_SIZE, despite other
-		 * subsystems dealing with that invariant just fine.  For now
-		 * stick to the legacy format where we never present those from
-		 * the block layer, but the code below should be removed once
-		 * these offenders (mostly MMC/SD drivers) are fixed.
-		 */
-		page += (offset >> PAGE_SHIFT);
-		offset &= ~PAGE_MASK;
-
-		*sg = blk_next_sg(sg, sglist);
-		sg_set_page(*sg, page, len, offset);
-
-		total += len;
-		nbytes -= len;
-		nsegs++;
-	}
-
-	return nsegs;
-}
-
-static inline int __blk_bvec_map_sg(struct bio_vec bv,
-		struct scatterlist *sglist, struct scatterlist **sg)
-{
-	*sg = blk_next_sg(sg, sglist);
-	sg_set_page(*sg, bv.bv_page, bv.bv_len, bv.bv_offset);
-	return 1;
-}
-
-/* only try to merge bvecs into one sg if they are from two bios */
-static inline bool
-__blk_segment_map_sg_merge(struct request_queue *q, struct bio_vec *bvec,
-			   struct bio_vec *bvprv, struct scatterlist **sg)
-{
-
-	int nbytes = bvec->bv_len;
-
-	if (!*sg)
-		return false;
-
-	if ((*sg)->length + nbytes > queue_max_segment_size(q))
-		return false;
-
-	if (!biovec_phys_mergeable(q, bvprv, bvec))
-		return false;
-
-	(*sg)->length += nbytes;
-
-	return true;
-}
-
-static int __blk_bios_map_sg(struct request_queue *q, struct bio *bio,
-			     struct scatterlist *sglist,
-			     struct scatterlist **sg)
-{
-	struct bio_vec bvec, bvprv = { NULL };
-	struct bvec_iter iter;
-	int nsegs = 0;
-	bool new_bio = false;
-
-	for_each_bio(bio) {
-		bio_for_each_bvec(bvec, bio, iter) {
-			/*
-			 * Only try to merge bvecs from two bios given we
-			 * have done bio internal merge when adding pages
-			 * to bio
-			 */
-			if (new_bio &&
-			    __blk_segment_map_sg_merge(q, &bvec, &bvprv, sg))
-				goto next_bvec;
-
-			if (bvec.bv_offset + bvec.bv_len <= PAGE_SIZE)
-				nsegs += __blk_bvec_map_sg(bvec, sglist, sg);
-			else
-				nsegs += blk_bvec_map_sg(q, &bvec, sglist, sg);
- next_bvec:
-			new_bio = false;
-		}
-		if (likely(bio->bi_iter.bi_size)) {
-			bvprv = bvec;
-			new_bio = true;
-		}
-	}
-
-	return nsegs;
-}
-
 /*
- * map a request to scatterlist, return number of sg entries setup. Caller
- * must make sure sg can hold rq->nr_phys_segments entries
+ * Map a request to scatterlist, return number of sg entries setup. Caller
+ * must make sure sg can hold rq->nr_phys_segments entries.
  */
 int __blk_rq_map_sg(struct request_queue *q, struct request *rq,
 		struct scatterlist *sglist, struct scatterlist **last_sg)
 {
+	struct req_iterator iter = {
+		.bio	= rq->bio,
+		.iter	= rq->bio->bi_iter,
+	};
+	struct phys_vec vec;
 	int nsegs = 0;
 
-	if (rq->rq_flags & RQF_SPECIAL_PAYLOAD)
-		nsegs = __blk_bvec_map_sg(rq->special_vec, sglist, last_sg);
-	else if (rq->bio)
-		nsegs = __blk_bios_map_sg(q, rq->bio, sglist, last_sg);
+	while (blk_map_iter_next(rq, &iter, &vec)) {
+		*last_sg = blk_next_sg(last_sg, sglist);
+		sg_set_page(*last_sg, phys_to_page(vec.paddr), vec.len,
+				offset_in_page(vec.paddr));
+		nsegs++;
+	}
 
 	if (*last_sg)
 		sg_mark_end(*last_sg);

block/blk-mq-cpumap.c

···
 #include <linux/smp.h>
 #include <linux/cpu.h>
 #include <linux/group_cpus.h>
+#include <linux/device/bus.h>
 
 #include "blk.h"
 #include "blk-mq.h"
···
 
 	return NUMA_NO_NODE;
 }
+
+/**
+ * blk_mq_map_hw_queues - Create CPU to hardware queue mapping
+ * @qmap:	CPU to hardware queue map
+ * @dev:	The device to map queues
+ * @offset:	Queue offset to use for the device
+ *
+ * Create a CPU to hardware queue mapping in @qmap. The struct bus_type
+ * irq_get_affinity callback will be used to retrieve the affinity.
+ */
+void blk_mq_map_hw_queues(struct blk_mq_queue_map *qmap,
+			  struct device *dev, unsigned int offset)
+
+{
+	const struct cpumask *mask;
+	unsigned int queue, cpu;
+
+	if (!dev->bus->irq_get_affinity)
+		goto fallback;
+
+	for (queue = 0; queue < qmap->nr_queues; queue++) {
+		mask = dev->bus->irq_get_affinity(dev, queue + offset);
+		if (!mask)
+			goto fallback;
+
+		for_each_cpu(cpu, mask)
+			qmap->mq_map[cpu] = qmap->queue_offset + queue;
+	}
+
+	return;
+
+fallback:
+	WARN_ON_ONCE(qmap->nr_queues > 1);
+	blk_mq_clear_mq_map(qmap);
+}
+EXPORT_SYMBOL_GPL(blk_mq_map_hw_queues);

block/blk-mq-debugfs.c

···
 	return 0;
 }
 
-#define BLK_TAG_ALLOC_NAME(name) [BLK_TAG_ALLOC_##name] = #name
-static const char *const alloc_policy_name[] = {
-	BLK_TAG_ALLOC_NAME(FIFO),
-	BLK_TAG_ALLOC_NAME(RR),
-};
-#undef BLK_TAG_ALLOC_NAME
-
 #define HCTX_FLAG_NAME(name) [ilog2(BLK_MQ_F_##name)] = #name
 static const char *const hctx_flag_name[] = {
-	HCTX_FLAG_NAME(SHOULD_MERGE),
 	HCTX_FLAG_NAME(TAG_QUEUE_SHARED),
 	HCTX_FLAG_NAME(STACKING),
 	HCTX_FLAG_NAME(TAG_HCTX_SHARED),
 	HCTX_FLAG_NAME(BLOCKING),
-	HCTX_FLAG_NAME(NO_SCHED),
+	HCTX_FLAG_NAME(TAG_RR),
 	HCTX_FLAG_NAME(NO_SCHED_BY_DEFAULT),
 };
 #undef HCTX_FLAG_NAME
···
 static int hctx_flags_show(void *data, struct seq_file *m)
 {
 	struct blk_mq_hw_ctx *hctx = data;
-	const int alloc_policy = BLK_MQ_FLAG_TO_ALLOC_POLICY(hctx->flags);
 
-	BUILD_BUG_ON(ARRAY_SIZE(hctx_flag_name) !=
-			BLK_MQ_F_ALLOC_POLICY_START_BIT);
-	BUILD_BUG_ON(ARRAY_SIZE(alloc_policy_name) != BLK_TAG_ALLOC_MAX);
+	BUILD_BUG_ON(ARRAY_SIZE(hctx_flag_name) != ilog2(BLK_MQ_F_MAX));
 
-	seq_puts(m, "alloc_policy=");
-	if (alloc_policy < ARRAY_SIZE(alloc_policy_name) &&
-	    alloc_policy_name[alloc_policy])
-		seq_puts(m, alloc_policy_name[alloc_policy]);
-	else
-		seq_printf(m, "%d", alloc_policy);
-	seq_puts(m, " ");
-	blk_flags_show(m,
-		       hctx->flags ^ BLK_ALLOC_POLICY_TO_MQ_FLAG(alloc_policy),
-		       hctx_flag_name, ARRAY_SIZE(hctx_flag_name));
+	blk_flags_show(m, hctx->flags, hctx_flag_name,
+			ARRAY_SIZE(hctx_flag_name));
 	seq_puts(m, "\n");
 	return 0;
 }

block/blk-mq-pci.c

···
-// SPDX-License-Identifier: GPL-2.0
-/*
- * Copyright (c) 2016 Christoph Hellwig.
- */
-#include <linux/kobject.h>
-#include <linux/blkdev.h>
-#include <linux/blk-mq-pci.h>
-#include <linux/pci.h>
-#include <linux/module.h>
-
-#include "blk-mq.h"
-
-/**
- * blk_mq_pci_map_queues - provide a default queue mapping for PCI device
- * @qmap:	CPU to hardware queue map.
- * @pdev:	PCI device associated with @set.
- * @offset:	Offset to use for the pci irq vector
- *
- * This function assumes the PCI device @pdev has at least as many available
- * interrupt vectors as @set has queues.  It will then query the vector
- * corresponding to each queue for it's affinity mask and built queue mapping
- * that maps a queue to the CPUs that have irq affinity for the corresponding
- * vector.
- */
-void blk_mq_pci_map_queues(struct blk_mq_queue_map *qmap, struct pci_dev *pdev,
-			   int offset)
-{
-	const struct cpumask *mask;
-	unsigned int queue, cpu;
-
-	for (queue = 0; queue < qmap->nr_queues; queue++) {
-		mask = pci_irq_get_affinity(pdev, queue + offset);
-		if (!mask)
-			goto fallback;
-
-		for_each_cpu(cpu, mask)
-			qmap->mq_map[cpu] = qmap->queue_offset + queue;
-	}
-
-	return;
-
-fallback:
-	WARN_ON_ONCE(qmap->nr_queues > 1);
-	blk_mq_clear_mq_map(qmap);
-}
-EXPORT_SYMBOL_GPL(blk_mq_pci_map_queues);

block/blk-mq-sched.c

···
 	ctx = blk_mq_get_ctx(q);
 	hctx = blk_mq_map_queue(q, bio->bi_opf, ctx);
 	type = hctx->type;
-	if (!(hctx->flags & BLK_MQ_F_SHOULD_MERGE) ||
-	    list_empty_careful(&ctx->rq_lists[type]))
+	if (list_empty_careful(&ctx->rq_lists[type]))
 		goto out_put;
 
 	/* default per sw-queue merge */

block/blk-mq-tag.c

···
 				       node);
 }
 
-int blk_mq_init_bitmaps(struct sbitmap_queue *bitmap_tags,
-			struct sbitmap_queue *breserved_tags,
-			unsigned int queue_depth, unsigned int reserved,
-			int node, int alloc_policy)
-{
-	unsigned int depth = queue_depth - reserved;
-	bool round_robin = alloc_policy == BLK_TAG_ALLOC_RR;
-
-	if (bt_alloc(bitmap_tags, depth, round_robin, node))
-		return -ENOMEM;
-	if (bt_alloc(breserved_tags, reserved, round_robin, node))
-		goto free_bitmap_tags;
-
-	return 0;
-
-free_bitmap_tags:
-	sbitmap_queue_free(bitmap_tags);
-	return -ENOMEM;
-}
-
 struct blk_mq_tags *blk_mq_init_tags(unsigned int total_tags,
-				     unsigned int reserved_tags,
-				     int node, int alloc_policy)
+		unsigned int reserved_tags, unsigned int flags, int node)
 {
+	unsigned int depth = total_tags - reserved_tags;
+	bool round_robin = flags & BLK_MQ_F_TAG_RR;
 	struct blk_mq_tags *tags;
 
 	if (total_tags > BLK_MQ_TAG_MAX) {
···
 	tags->nr_tags = total_tags;
 	tags->nr_reserved_tags = reserved_tags;
 	spin_lock_init(&tags->lock);
+	if (bt_alloc(&tags->bitmap_tags, depth, round_robin, node))
+		goto out_free_tags;
+	if (bt_alloc(&tags->breserved_tags, reserved_tags, round_robin, node))
+		goto out_free_bitmap_tags;
 
-	if (blk_mq_init_bitmaps(&tags->bitmap_tags, &tags->breserved_tags,
-				total_tags, reserved_tags, node,
-				alloc_policy) < 0) {
-		kfree(tags);
-		return NULL;
-	}
 	return tags;
+
+out_free_bitmap_tags:
+	sbitmap_queue_free(&tags->bitmap_tags);
+out_free_tags:
+	kfree(tags);
+	return NULL;
 }
 
 void blk_mq_free_tags(struct blk_mq_tags *tags)

block/blk-mq-virtio.c

···
-// SPDX-License-Identifier: GPL-2.0
-/*
- * Copyright (c) 2016 Christoph Hellwig.
- */
-#include <linux/device.h>
-#include <linux/blk-mq-virtio.h>
-#include <linux/virtio_config.h>
-#include <linux/module.h>
-#include "blk-mq.h"
-
-/**
- * blk_mq_virtio_map_queues - provide a default queue mapping for virtio device
- * @qmap:	CPU to hardware queue map.
- * @vdev:	virtio device to provide a mapping for.
- * @first_vec:	first interrupt vectors to use for queues (usually 0)
- *
- * This function assumes the virtio device @vdev has at least as many available
- * interrupt vectors as @set has queues.  It will then query the vector
- * corresponding to each queue for it's affinity mask and built queue mapping
- * that maps a queue to the CPUs that have irq affinity for the corresponding
- * vector.
- */
-void blk_mq_virtio_map_queues(struct blk_mq_queue_map *qmap,
-		struct virtio_device *vdev, int first_vec)
-{
-	const struct cpumask *mask;
-	unsigned int queue, cpu;
-
-	if (!vdev->config->get_vq_affinity)
-		goto fallback;
-
-	for (queue = 0; queue < qmap->nr_queues; queue++) {
-		mask = vdev->config->get_vq_affinity(vdev, first_vec + queue);
-		if (!mask)
-			goto fallback;
-
-		for_each_cpu(cpu, mask)
-			qmap->mq_map[cpu] = qmap->queue_offset + queue;
-	}
-
-	return;
-
-fallback:
-	blk_mq_map_queues(qmap);
-}
-EXPORT_SYMBOL_GPL(blk_mq_virtio_map_queues);

block/blk-mq.c

···
 	if (!q->mq_freeze_depth) {
 		q->mq_freeze_owner = owner;
 		q->mq_freeze_owner_depth = 1;
+		q->mq_freeze_disk_dead = !q->disk ||
+			test_bit(GD_DEAD, &q->disk->state) ||
+			!blk_queue_registered(q);
+		q->mq_freeze_queue_dying = blk_queue_dying(q);
 		return true;
 	}
 
···
 /* verify the last unfreeze in owner context */
 static bool blk_unfreeze_check_owner(struct request_queue *q)
 {
-	if (!q->mq_freeze_owner)
-		return false;
 	if (q->mq_freeze_owner != current)
 		return false;
 	if (--q->mq_freeze_owner_depth == 0) {
···
 void blk_freeze_queue_start(struct request_queue *q)
 {
 	if (__blk_freeze_queue_start(q, current))
-		blk_freeze_acquire_lock(q, false, false);
+		blk_freeze_acquire_lock(q);
 }
 EXPORT_SYMBOL_GPL(blk_freeze_queue_start);
 
···
 void blk_mq_unfreeze_queue(struct request_queue *q)
 {
 	if (__blk_mq_unfreeze_queue(q, false))
-		blk_unfreeze_release_lock(q, false, false);
+		blk_unfreeze_release_lock(q);
 }
 EXPORT_SYMBOL_GPL(blk_mq_unfreeze_queue);
 
···
 	if (bio->bi_opf & REQ_RAHEAD)
 		rq->cmd_flags |= REQ_FAILFAST_MASK;
 
+	rq->bio = rq->biotail = bio;
 	rq->__sector = bio->bi_iter.bi_sector;
-	blk_rq_bio_prep(rq, bio, nr_segs);
+	rq->__data_len = bio->bi_iter.bi_size;
+	rq->nr_phys_segments = nr_segs;
 	if (bio_integrity(bio))
 		rq->nr_integrity_segments = blk_rq_count_integrity_sg(rq->q,
 								      bio);
···
 	}
 
 	rq = __blk_mq_alloc_requests(&data);
-	if (rq)
-		return rq;
-	rq_qos_cleanup(q, bio);
-	if (bio->bi_opf & REQ_NOWAIT)
-		bio_wouldblock_error(bio);
-	return NULL;
+	if (unlikely(!rq))
+		rq_qos_cleanup(q, bio);
+	return rq;
 }
 
 /*
···
 	}
 
 	/*
-	 * Device reconfiguration may change logical block size, so alignment
-	 * check has to be done with queue usage counter held
+	 * Device reconfiguration may change logical block size or reduce the
+	 * number of poll queues, so the checks for alignment and poll support
+	 * have to be done with queue usage counter held.
 	 */
 	if (unlikely(bio_unaligned(bio, q))) {
 		bio_io_error(bio);
+		goto queue_exit;
+	}
+
+	if ((bio->bi_opf & REQ_POLLED) && !blk_mq_can_poll(q)) {
+		bio->bi_status = BLK_STS_NOTSUPP;
+		bio_endio(bio);
 		goto queue_exit;
 	}
 
···
 		goto queue_exit;
 
 new_request:
-	if (!rq) {
-		rq = blk_mq_get_new_requests(q, plug, bio, nr_segs);
-		if (unlikely(!rq))
-			goto queue_exit;
-	} else {
+	if (rq) {
 		blk_mq_use_cached_rq(rq, plug, bio);
+	} else {
+		rq = blk_mq_get_new_requests(q, plug, bio, nr_segs);
+		if (unlikely(!rq)) {
+			if (bio->bi_opf & REQ_NOWAIT)
+				bio_wouldblock_error(bio);
+			goto queue_exit;
+		}
 	}
 
 	trace_block_getrq(bio);
···
 	if (node == NUMA_NO_NODE)
 		node = set->numa_node;
 
-	tags = blk_mq_init_tags(nr_tags, reserved_tags, node,
-				BLK_MQ_FLAG_TO_ALLOC_POLICY(set->flags));
+	tags = blk_mq_init_tags(nr_tags, reserved_tags, set->flags, node);
 	if (!tags)
 		return NULL;
 
···
 	blk_mq_sysfs_deinit(q);
 }
 
-static bool blk_mq_can_poll(struct blk_mq_tag_set *set)
-{
-	return set->nr_maps > HCTX_TYPE_POLL &&
-		set->map[HCTX_TYPE_POLL].nr_queues;
-}
-
 struct request_queue *blk_mq_alloc_queue(struct blk_mq_tag_set *set,
 		struct queue_limits *lim, void *queuedata)
 {
···
 	if (!lim)
 		lim = &default_lim;
 	lim->features |= BLK_FEAT_IO_STAT | BLK_FEAT_NOWAIT;
-	if (blk_mq_can_poll(set))
+	if (set->nr_maps > HCTX_TYPE_POLL)
 		lim->features |= BLK_FEAT_POLL;
 
 	q = blk_alloc_queue(lim, set->numa_node);
···
 fallback:
 	blk_mq_update_queue_map(set);
 	list_for_each_entry(q, &set->tag_list, tag_set_list) {
-		struct queue_limits lim;
-
 		blk_mq_realloc_hw_ctxs(set, q);
 
 		if (q->nr_hw_queues != set->nr_hw_queues) {
···
 			set->nr_hw_queues = prev_nr_hw_queues;
 			goto fallback;
 		}
-		lim = queue_limits_start_update(q);
-		if (blk_mq_can_poll(set))
-			lim.features |= BLK_FEAT_POLL;
-		else
-			lim.features &= ~BLK_FEAT_POLL;
-		if (queue_limits_commit_update(q, &lim) < 0)
-			pr_warn("updating the poll flag failed\n");
 		blk_mq_map_swqueue(q);
 	}
 
···
 int blk_mq_poll(struct request_queue *q, blk_qc_t cookie,
 		struct io_comp_batch *iob, unsigned int flags)
 {
-	struct blk_mq_hw_ctx *hctx = xa_load(&q->hctx_table, cookie);
-
-	return blk_hctx_poll(q, hctx, iob, flags);
+	if (!blk_mq_can_poll(q))
+		return 0;
+	return blk_hctx_poll(q, xa_load(&q->hctx_table, cookie), iob, flags);
 }
 
 int blk_rq_poll(struct request *rq, struct io_comp_batch *iob,

block/blk-mq.h

···
 };
 
 struct blk_mq_tags *blk_mq_init_tags(unsigned int nr_tags,
-		unsigned int reserved_tags, int node, int alloc_policy);
+		unsigned int reserved_tags, unsigned int flags, int node);
 void blk_mq_free_tags(struct blk_mq_tags *tags);
-int blk_mq_init_bitmaps(struct sbitmap_queue *bitmap_tags,
-		struct sbitmap_queue *breserved_tags, unsigned int queue_depth,
-		unsigned int reserved, int node, int alloc_policy);
 
 unsigned int blk_mq_get_tag(struct blk_mq_alloc_data *data);
 unsigned long blk_mq_get_tags(struct blk_mq_alloc_data *data, int nr_tags,
···
 
 #define blk_mq_run_dispatch_ops(q, dispatch_ops)		\
 	__blk_mq_run_dispatch_ops(q, true, dispatch_ops)	\
+
+static inline bool blk_mq_can_poll(struct request_queue *q)
+{
+	return (q->limits.features & BLK_FEAT_POLL) &&
+		q->tag_set->map[HCTX_TYPE_POLL].nr_queues;
+}
 
 #endif

block/blk-settings.c

···
 {
 	unsigned int boundary_sectors;
 
+	if (!(lim->features & BLK_FEAT_ATOMIC_WRITES))
+		goto unsupported;
+
 	if (!lim->atomic_write_hw_max)
 		goto unsupported;
 
···
  * @lim:	limits to apply
  *
  * Apply the limits in @lim that were obtained from queue_limits_start_update()
- * and updated by the caller to @q.
+ * and updated by the caller to @q.  The caller must have frozen the queue or
+ * ensure that there are no outstanding I/Os by other means.
  *
  * Returns 0 if successful, else a negative error code.
  */
···
 	return error;
 }
 EXPORT_SYMBOL_GPL(queue_limits_commit_update);
+
+/**
+ * queue_limits_commit_update_frozen - commit an atomic update of queue limits
+ * @q:		queue to update
+ * @lim:	limits to apply
+ *
+ * Apply the limits in @lim that were obtained from queue_limits_start_update()
+ * and updated with the new values by the caller to @q.  Freezes the queue
+ * before the update and unfreezes it after.
+ *
+ * Returns 0 if successful, else a negative error code.
+ */
+int queue_limits_commit_update_frozen(struct request_queue *q,
+		struct queue_limits *lim)
+{
+	int ret;
+
+	blk_mq_freeze_queue(q);
+	ret = queue_limits_commit_update(q, lim);
+	blk_mq_unfreeze_queue(q);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(queue_limits_commit_update_frozen);
 
 /**
  * queue_limits_set - apply queue limits to queue
···
 }
 
 static void blk_stack_atomic_writes_limits(struct queue_limits *t,
-				struct queue_limits *b)
+				struct queue_limits *b, sector_t start)
 {
-	if (!(t->features & BLK_FEAT_ATOMIC_WRITES_STACKED))
+	if (!(b->features & BLK_FEAT_ATOMIC_WRITES))
 		goto unsupported;
 
-	if (!b->atomic_write_unit_min)
+	if (!b->atomic_write_hw_unit_min)
+		goto unsupported;
+
+	if (!blk_atomic_write_start_sect_aligned(start, b))
 		goto unsupported;
 
 	/*
···
 	t->atomic_write_hw_unit_max = 0;
 	t->atomic_write_hw_unit_min = 0;
 	t->atomic_write_hw_boundary = 0;
-	t->features &= ~BLK_FEAT_ATOMIC_WRITES_STACKED;
 }
 
 /**
···
 		t->zone_write_granularity = 0;
 		t->max_zone_append_sectors = 0;
 	}
-	blk_stack_atomic_writes_limits(t, b);
+	blk_stack_atomic_writes_limits(t, b, start);
 
 	return ret;
 }

block/blk-sysfs.c

···
 	struct attribute attr;
 	ssize_t (*show)(struct gendisk *disk, char *page);
 	ssize_t (*store)(struct gendisk *disk, const char *page, size_t count);
+	int (*store_limit)(struct gendisk *disk, const char *page,
+			size_t count, struct queue_limits *lim);
 	void (*load_module)(struct gendisk *disk, const char *page, size_t count);
 };
 
···
 QUEUE_SYSFS_SHOW_CONST(write_same_max, 0)
 QUEUE_SYSFS_SHOW_CONST(poll_delay, -1)
 
-static ssize_t queue_max_discard_sectors_store(struct gendisk *disk,
-		const char *page, size_t count)
+static int queue_max_discard_sectors_store(struct gendisk *disk,
+		const char *page, size_t count, struct queue_limits *lim)
 {
 	unsigned long max_discard_bytes;
-	struct queue_limits lim;
 	ssize_t ret;
-	int err;
 
 	ret = queue_var_store(&max_discard_bytes, page, count);
 	if (ret < 0)
···
 	if ((max_discard_bytes >> SECTOR_SHIFT) > UINT_MAX)
 		return -EINVAL;
 
-	lim = queue_limits_start_update(disk->queue);
-	lim.max_user_discard_sectors = max_discard_bytes >> SECTOR_SHIFT;
-	err = queue_limits_commit_update(disk->queue, &lim);
-	if (err)
-		return err;
-	return ret;
+	lim->max_user_discard_sectors = max_discard_bytes >> SECTOR_SHIFT;
+	return 0;
 }
 
-static ssize_t
-queue_max_sectors_store(struct gendisk *disk, const char *page, size_t count)
+static int
+queue_max_sectors_store(struct gendisk *disk, const char *page, size_t count,
+		struct queue_limits *lim)
 {
 	unsigned long max_sectors_kb;
-	struct queue_limits lim;
 	ssize_t ret;
-	int err;
 
 	ret = queue_var_store(&max_sectors_kb, page, count);
 	if (ret < 0)
 		return ret;
 
-	lim = queue_limits_start_update(disk->queue);
-	lim.max_user_sectors = max_sectors_kb << 1;
-	err = queue_limits_commit_update(disk->queue, &lim);
-	if (err)
-		return err;
-	return ret;
+	lim->max_user_sectors = max_sectors_kb << 1;
+	return 0;
 }
 
 static ssize_t queue_feature_store(struct gendisk *disk, const char *page,
-		size_t count, blk_features_t feature)
+		size_t count, struct queue_limits *lim, blk_features_t feature)
 {
-	struct queue_limits lim;
 	unsigned long val;
 	ssize_t ret;
 
···
 	if (ret < 0)
 		return ret;
 
-	lim = queue_limits_start_update(disk->queue);
 	if (val)
-		lim.features |= feature;
+		lim->features |= feature;
 	else
-		lim.features &= ~feature;
-	ret = queue_limits_commit_update(disk->queue, &lim);
-	if (ret)
-		return ret;
-	return count;
+		lim->features &= ~feature;
+	return 0;
 }
 
 #define QUEUE_SYSFS_FEATURE(_name, _feature)				\
···
 	return sysfs_emit(page, "%u\n",					\
 		!!(disk->queue->limits.features & _feature));		\
 }									\
-static ssize_t queue_##_name##_store(struct gendisk *disk,		\
-		const char *page, size_t count)				\
+static int queue_##_name##_store(struct gendisk *disk,			\
+		const char *page, size_t count, struct queue_limits *lim) \
 {									\
-	return queue_feature_store(disk, page, count, _feature);	\
+	return queue_feature_store(disk, page, count, lim, _feature);	\
 }
 
 QUEUE_SYSFS_FEATURE(rotational, BLK_FEAT_ROTATIONAL)
···
 		!!(disk->queue->limits.features & _feature));		\
 }
 
-QUEUE_SYSFS_FEATURE_SHOW(poll, BLK_FEAT_POLL);
 QUEUE_SYSFS_FEATURE_SHOW(fua, BLK_FEAT_FUA);
 QUEUE_SYSFS_FEATURE_SHOW(dax, BLK_FEAT_DAX);
+
+static ssize_t queue_poll_show(struct gendisk *disk, char *page)
+{
+	if (queue_is_mq(disk->queue))
+		return sysfs_emit(page, "%u\n", blk_mq_can_poll(disk->queue));
+	return sysfs_emit(page, "%u\n",
+		!!(disk->queue->limits.features & BLK_FEAT_POLL));
+}
 
 static ssize_t queue_zoned_show(struct gendisk *disk, char *page)
 {
···
 	return queue_var_show(!!blk_queue_passthrough_stat(disk->queue), page);
 }
 
-static ssize_t queue_iostats_passthrough_store(struct gendisk *disk,
-					       const char *page, size_t count)
+static int queue_iostats_passthrough_store(struct gendisk *disk,
+		const char *page, size_t count, struct queue_limits *lim)
 {
-	struct queue_limits lim;
 	unsigned long ios;
 	ssize_t ret;
 
···
 	if (ret < 0)
 		return ret;
 
-	lim = queue_limits_start_update(disk->queue);
 	if (ios)
-		lim.flags |= BLK_FLAG_IOSTATS_PASSTHROUGH;
+		lim->flags |= BLK_FLAG_IOSTATS_PASSTHROUGH;
 	else
-		lim.flags &= ~BLK_FLAG_IOSTATS_PASSTHROUGH;
-
-	ret = queue_limits_commit_update(disk->queue, &lim);
-	if (ret)
-		return ret;
-
-	return count;
+		lim->flags &= ~BLK_FLAG_IOSTATS_PASSTHROUGH;
+	return 0;
 }
+
 static ssize_t queue_nomerges_show(struct gendisk *disk, char *page)
 {
 	return queue_var_show((blk_queue_nomerges(disk->queue) << 1) |
···
 	return sysfs_emit(page, "write through\n");
 }
 
-static ssize_t queue_wc_store(struct gendisk *disk, const char *page,
-			      size_t count)
+static int queue_wc_store(struct gendisk *disk, const char *page,
+		size_t count, struct queue_limits *lim)
 {
-	struct queue_limits lim;
 	bool disable;
-	int err;
 
 	if (!strncmp(page, "write back", 10)) {
 		disable = false;
···
 		return -EINVAL;
 	}
 
-	lim = queue_limits_start_update(disk->queue);
 	if (disable)
-		lim.flags |= BLK_FLAG_WRITE_CACHE_DISABLED;
+		lim->flags |= BLK_FLAG_WRITE_CACHE_DISABLED;
 	else
-		lim.flags &= ~BLK_FLAG_WRITE_CACHE_DISABLED;
-	err = queue_limits_commit_update(disk->queue, &lim);
-	if (err)
-		return err;
-	return count;
+		lim->flags &= ~BLK_FLAG_WRITE_CACHE_DISABLED;
+	return 0;
 }
 
 #define QUEUE_RO_ENTRY(_prefix, _name)			\
···
 	.store	= _prefix##_store,			\
 };
 
+#define QUEUE_LIM_RW_ENTRY(_prefix, _name)			\
+static struct queue_sysfs_entry _prefix##_entry = {	\
+	.attr		= { .name = _name, .mode = 0644 },	\
+	.show		= _prefix##_show,			\
+	.store_limit	= _prefix##_store,			\
+}
+
 #define QUEUE_RW_LOAD_MODULE_ENTRY(_prefix, _name)		\
 static struct queue_sysfs_entry _prefix##_entry = {		\
 	.attr		= { .name = _name, .mode = 0644 },	\
···
 
 QUEUE_RW_ENTRY(queue_requests, "nr_requests");
 QUEUE_RW_ENTRY(queue_ra, "read_ahead_kb");
-QUEUE_RW_ENTRY(queue_max_sectors, "max_sectors_kb");
+QUEUE_LIM_RW_ENTRY(queue_max_sectors, "max_sectors_kb");
 QUEUE_RO_ENTRY(queue_max_hw_sectors, "max_hw_sectors_kb");
 QUEUE_RO_ENTRY(queue_max_segments, "max_segments");
 QUEUE_RO_ENTRY(queue_max_integrity_segments, "max_integrity_segments");
···
 QUEUE_RO_ENTRY(queue_max_discard_segments, "max_discard_segments");
 QUEUE_RO_ENTRY(queue_discard_granularity, "discard_granularity");
 QUEUE_RO_ENTRY(queue_max_hw_discard_sectors, "discard_max_hw_bytes");
-QUEUE_RW_ENTRY(queue_max_discard_sectors, "discard_max_bytes");
+QUEUE_LIM_RW_ENTRY(queue_max_discard_sectors, "discard_max_bytes");
 QUEUE_RO_ENTRY(queue_discard_zeroes_data, "discard_zeroes_data");
 
 QUEUE_RO_ENTRY(queue_atomic_write_max_sectors, "atomic_write_max_bytes");
···
 QUEUE_RO_ENTRY(queue_max_active_zones, "max_active_zones");
 
 QUEUE_RW_ENTRY(queue_nomerges, "nomerges");
-QUEUE_RW_ENTRY(queue_iostats_passthrough, "iostats_passthrough");
+QUEUE_LIM_RW_ENTRY(queue_iostats_passthrough, "iostats_passthrough");
 QUEUE_RW_ENTRY(queue_rq_affinity, "rq_affinity");
 QUEUE_RW_ENTRY(queue_poll, "io_poll");
 QUEUE_RW_ENTRY(queue_poll_delay, "io_poll_delay");
-QUEUE_RW_ENTRY(queue_wc, "write_cache");
+QUEUE_LIM_RW_ENTRY(queue_wc, "write_cache");
 QUEUE_RO_ENTRY(queue_fua, "fua");
 QUEUE_RO_ENTRY(queue_dax, "dax");
 QUEUE_RW_ENTRY(queue_io_timeout, "io_timeout");
···
 	.show = queue_logical_block_size_show,
 };
 
-QUEUE_RW_ENTRY(queue_rotational, "rotational");
-QUEUE_RW_ENTRY(queue_iostats, "iostats");
-QUEUE_RW_ENTRY(queue_add_random, "add_random");
-QUEUE_RW_ENTRY(queue_stable_writes, "stable_writes");
+QUEUE_LIM_RW_ENTRY(queue_rotational, "rotational");
+QUEUE_LIM_RW_ENTRY(queue_iostats, "iostats");
+QUEUE_LIM_RW_ENTRY(queue_add_random, "add_random");
+QUEUE_LIM_RW_ENTRY(queue_stable_writes, "stable_writes");
 
 #ifdef CONFIG_BLK_WBT
 static ssize_t queue_var_store64(s64 *var, const char *page)
···
 	struct queue_sysfs_entry *entry = to_queue(attr);
 	struct gendisk *disk = container_of(kobj, struct gendisk, queue_kobj);
 	struct request_queue *q = disk->queue;
+	unsigned int noio_flag;
 	ssize_t res;
 
-	if (!entry->store)
+	if (!entry->store_limit && !entry->store)
 		return -EIO;
 
 	/*
···
 	if (entry->load_module)
 		entry->load_module(disk, page, length);
 
-	blk_mq_freeze_queue(q);
+	if (entry->store_limit) {
+		struct queue_limits lim = queue_limits_start_update(q);
+
+		res = entry->store_limit(disk, page, length, &lim);
+		if (res < 0) {
+			queue_limits_cancel_update(q);
+			return res;
+		}
+
+		res = queue_limits_commit_update_frozen(q, &lim);
+		if (res)
+			return res;
+		return length;
+	}
+
 	mutex_lock(&q->sysfs_lock);
+	blk_mq_freeze_queue(q);
+	noio_flag = memalloc_noio_save();
 	res = entry->store(disk, page, length);
-	mutex_unlock(&q->sysfs_lock);
+	memalloc_noio_restore(noio_flag);
 	blk_mq_unfreeze_queue(q);
+	mutex_unlock(&q->sysfs_lock);
 	return res;
 }
 

block/blk-zoned.c

···
  */
 
 #include <linux/kernel.h>
-#include <linux/module.h>
 #include <linux/blkdev.h>
 #include <linux/blk-mq.h>
-#include <linux/mm.h>
-#include <linux/vmalloc.h>
-#include <linux/sched/mm.h>
 #include <linux/spinlock.h>
 #include <linux/refcount.h>
 #include <linux/mempool.h>
···
 static inline bool disk_should_remove_zone_wplug(struct gendisk *disk,
 						 struct blk_zone_wplug *zwplug)
 {
+	lockdep_assert_held(&zwplug->lock);
+
 	/* If the zone write plug was already removed, we are done. */
 	if (zwplug->flags & BLK_ZONE_WPLUG_UNHASHED)
 		return false;
···
 	bio_clear_flag(bio, BIO_ZONE_WRITE_PLUGGING);
 	bio_io_error(bio);
 	disk_put_zone_wplug(zwplug);
+	/* Drop the reference taken by disk_zone_wplug_add_bio(() */
 	blk_queue_exit(q);
 }
 
···
 			break;
 		}
 
-		/*
-		 * Drop the extra reference on the queue usage we got when
-		 * plugging the BIO and advance the write pointer offset.
-		 */
+		/* Drop the reference taken by disk_zone_wplug_add_bio(). */
 		blk_queue_exit(q);
 		zwplug->wp_offset += bio_sectors(bio);
 
···
 				       struct bio *bio)
 {
 	struct gendisk *disk = bio->bi_bdev->bd_disk;
+
+	lockdep_assert_held(&zwplug->lock);
 
 	/*
 	 * If we lost track of the zone write pointer due to a write error,
···
 	unsigned int nr_seq_zones, nr_conv_zones;
 	unsigned int pool_size;
 	struct queue_limits lim;
-	int ret;
 
 	disk->nr_zones = args->nr_zones;
 	disk->zone_capacity = args->zone_capacity;
···
 	}
 
 commit:
-	blk_mq_freeze_queue(q);
-	ret = queue_limits_commit_update(q, &lim);
-	blk_mq_unfreeze_queue(q);
-
-	return ret;
+	return queue_limits_commit_update_frozen(q, &lim);
 }
 
 static int blk_revalidate_conv_zone(struct blk_zone *zone, unsigned int idx,
···
 EXPORT_SYMBOL_GPL(blk_zone_issue_zeroout);
 
 #ifdef CONFIG_BLK_DEBUG_FS
+static void queue_zone_wplug_show(struct blk_zone_wplug *zwplug,
+				  struct seq_file *m)
+{
+	unsigned int zwp_wp_offset, zwp_flags;
+	unsigned int zwp_zone_no, zwp_ref;
+	unsigned int zwp_bio_list_size;
+	unsigned long flags;
+
+	spin_lock_irqsave(&zwplug->lock, flags);
+	zwp_zone_no = zwplug->zone_no;
+	zwp_flags = zwplug->flags;
+	zwp_ref = refcount_read(&zwplug->ref);
+	zwp_wp_offset = zwplug->wp_offset;
+	zwp_bio_list_size = bio_list_size(&zwplug->bio_list);
+	spin_unlock_irqrestore(&zwplug->lock, flags);
+
+	seq_printf(m, "%u 0x%x %u %u %u\n", zwp_zone_no, zwp_flags, zwp_ref,
+		   zwp_wp_offset, zwp_bio_list_size);
+}
 
 int queue_zone_wplugs_show(void *data, struct seq_file *m)
 {
 	struct request_queue *q = data;
 	struct gendisk *disk = q->disk;
 	struct blk_zone_wplug *zwplug;
-	unsigned int zwp_wp_offset, zwp_flags;
-	unsigned int zwp_zone_no, zwp_ref;
-	unsigned int zwp_bio_list_size, i;
-	unsigned long flags;
+	unsigned int i;
 
 	if (!disk->zone_wplugs_hash)
 		return 0;
 
 	rcu_read_lock();
-	for (i = 0; i < disk_zone_wplugs_hash_size(disk); i++) {
-		hlist_for_each_entry_rcu(zwplug,
-					 &disk->zone_wplugs_hash[i], node) {
-			spin_lock_irqsave(&zwplug->lock, flags);
-			zwp_zone_no = zwplug->zone_no;
-			zwp_flags = zwplug->flags;
-			zwp_ref = refcount_read(&zwplug->ref);
-			zwp_wp_offset = zwplug->wp_offset;
-			zwp_bio_list_size = bio_list_size(&zwplug->bio_list);
-			spin_unlock_irqrestore(&zwplug->lock, flags);
-
-			seq_printf(m, "%u 0x%x %u %u %u\n",
-				   zwp_zone_no, zwp_flags, zwp_ref,
-				   zwp_wp_offset, zwp_bio_list_size);
-		}
-	}
+	for (i = 0; i < disk_zone_wplugs_hash_size(disk); i++)
+		hlist_for_each_entry_rcu(zwplug, &disk->zone_wplugs_hash[i],
+					 node)
+			queue_zone_wplug_show(zwplug, m);
 	rcu_read_unlock();
 
 	return 0;

block/blk.h

···
 
 struct elevator_type;
 
+#define	BLK_DEV_MAX_SECTORS	(LLONG_MAX >> 9)
+
 /* Max future timer expiry for timeouts */
 #define BLK_MAX_TIMEOUT		(5 * HZ)
 
···
 struct gendisk *__alloc_disk_node(struct request_queue *q, int node_id,
 		struct lock_class_key *lkclass);
 
-int bio_add_hw_page(struct request_queue *q, struct bio *bio,
-		struct page *page, unsigned int len, unsigned int offset,
-		unsigned int max_sectors, bool *same_page);
-
-int bio_add_hw_folio(struct request_queue *q, struct bio *bio,
-		struct folio *folio, size_t len, size_t offset,
-		unsigned int max_sectors, bool *same_page);
-
 /*
  * Clean up a page appropriately, where the page may be pinned, may have a
  * ref taken on it or neither.
···
 void blk_integrity_prepare(struct request *rq);
 void blk_integrity_complete(struct request *rq, unsigned int nr_bytes);
 
-static inline void blk_freeze_acquire_lock(struct request_queue *q, bool
-		disk_dead, bool queue_dying)
+#ifdef CONFIG_LOCKDEP
+static inline void blk_freeze_acquire_lock(struct request_queue *q)
 {
-	if (!disk_dead)
+	if (!q->mq_freeze_disk_dead)
 		rwsem_acquire(&q->io_lockdep_map, 0, 1, _RET_IP_);
-	if (!queue_dying)
+	if (!q->mq_freeze_queue_dying)
 		rwsem_acquire(&q->q_lockdep_map, 0, 1, _RET_IP_);
 }
 
-static inline void blk_unfreeze_release_lock(struct request_queue *q, bool
-		disk_dead, bool queue_dying)
+static inline void blk_unfreeze_release_lock(struct request_queue *q)
 {
-	if (!queue_dying)
+	if (!q->mq_freeze_queue_dying)
 		rwsem_release(&q->q_lockdep_map, _RET_IP_);
-	if (!disk_dead)
+	if (!q->mq_freeze_disk_dead)
 		rwsem_release(&q->io_lockdep_map, _RET_IP_);
 }
+#else
+static inline void blk_freeze_acquire_lock(struct request_queue *q)
+{
+}
+static inline void blk_unfreeze_release_lock(struct request_queue *q)
+{
+}
+#endif
 
 #endif /* BLK_INTERNAL_H */

block/bsg-lib.c

···
 	set->queue_depth = 128;
 	set->numa_node = NUMA_NO_NODE;
 	set->cmd_size = sizeof(struct bsg_job) + dd_job_size;
-	set->flags = BLK_MQ_F_NO_SCHED | BLK_MQ_F_BLOCKING;
+	set->flags = BLK_MQ_F_BLOCKING;
 	if (blk_mq_alloc_tag_set(set))
 		goto out_tag_set;
 

block/elevator.c

···
 	return NULL;
 }
 
-#define to_elv(atr) container_of((atr), struct elv_fs_entry, attr)
+#define to_elv(atr) container_of_const((atr), struct elv_fs_entry, attr)
 
 static ssize_t
 elv_attr_show(struct kobject *kobj, struct attribute *attr, char *page)
 {
-	struct elv_fs_entry *entry = to_elv(attr);
+	const struct elv_fs_entry *entry = to_elv(attr);
 	struct elevator_queue *e;
 	ssize_t error;
 
···
 elv_attr_store(struct kobject *kobj, struct attribute *attr,
 	       const char *page, size_t length)
 {
-	struct elv_fs_entry *entry = to_elv(attr);
+	const struct elv_fs_entry *entry = to_elv(attr);
 	struct elevator_queue *e;
 	ssize_t error;
 
···
 
 	error = kobject_add(&e->kobj, &q->disk->queue_kobj, "iosched");
 	if (!error) {
-		struct elv_fs_entry *attr = e->type->elevator_attrs;
+		const struct elv_fs_entry *attr = e->type->elevator_attrs;
 		if (attr) {
 			while (attr->attr.name) {
 				if (sysfs_create_file(&e->kobj, &attr->attr))
···
 }
 EXPORT_SYMBOL_GPL(elv_unregister);
 
-static inline bool elv_support_iosched(struct request_queue *q)
-{
-	if (!queue_is_mq(q) ||
-	    (q->tag_set->flags & BLK_MQ_F_NO_SCHED))
-		return false;
-	return true;
-}
-
 /*
  * For single queue devices, default to using mq-deadline. If we have multiple
  * queues or mq-deadline is not available, default to "none".
···
 	struct elevator_type *e;
 	int err;
 
-	if (!elv_support_iosched(q))
-		return;
-
 	WARN_ON_ONCE(blk_queue_registered(q));
 
 	if (unlikely(q->elevator))
···
 	 *
 	 * Disk isn't added yet, so verifying queue lock only manually.
 	 */
-	blk_freeze_queue_start_non_owner(q);
-	blk_freeze_acquire_lock(q, true, false);
-	blk_mq_freeze_queue_wait(q);
+	blk_mq_freeze_queue(q);
 
 	blk_mq_cancel_work_sync(q);
 
 	err = blk_mq_init_sched(q, e);
 
-	blk_unfreeze_release_lock(q, true, false);
-	blk_mq_unfreeze_queue_non_owner(q);
+	blk_mq_unfreeze_queue(q);
 
 	if (err) {
 		pr_warn("\"%s\" elevator initialization failed, "
···
 	struct elevator_type *found;
 	const char *name;
 
-	if (!elv_support_iosched(disk->queue))
-		return;
-
 	strscpy(elevator_name, buf, sizeof(elevator_name));
 	name = strstrip(elevator_name);
 
···
 	char elevator_name[ELV_NAME_MAX];
 	int ret;
 
-	if (!elv_support_iosched(disk->queue))
-		return count;
-
 	strscpy(elevator_name, buf, sizeof(elevator_name));
 	ret = elevator_change(disk->queue, strstrip(elevator_name));
 	if (!ret)
···
 	struct elevator_queue *eq = q->elevator;
 	struct elevator_type *cur = NULL, *e;
 	int len = 0;
-
-	if (!elv_support_iosched(q))
-		return sprintf(name, "none\n");
 
 	if (!q->elevator) {
 		len += sprintf(name+len, "[none] ");

block/elevator.h

···
 
 	size_t icq_size;	/* see iocontext.h */
 	size_t icq_align;	/* ditto */
-	struct elv_fs_entry *elevator_attrs;
+	const struct elv_fs_entry *elevator_attrs;
 	const char *elevator_name;
 	const char *elevator_alias;
 	struct module *elevator_owner;

block/genhd.c

···
 
 void set_capacity(struct gendisk *disk, sector_t sectors)
 {
+	if (sectors > BLK_DEV_MAX_SECTORS) {
+		pr_warn_once("%s: truncate capacity from %lld to %lld\n",
+				disk->disk_name, sectors,
+				BLK_DEV_MAX_SECTORS);
+		sectors = BLK_DEV_MAX_SECTORS;
+	}
+
 	bdev_set_nr_sectors(disk->part0, sectors);
 }
 EXPORT_SYMBOL(set_capacity);
···
 	struct device *ddev = disk_to_dev(disk);
 	int ret;
 
-	/* Only makes sense for bio-based to set ->poll_bio */
-	if (queue_is_mq(disk->queue) && disk->fops->poll_bio)
+	if (WARN_ON_ONCE(bdev_nr_sectors(disk->part0) > BLK_DEV_MAX_SECTORS))
 		return -EINVAL;
 
-	/*
-	 * The disk queue should now be all set with enough information about
-	 * the device for the elevator code to pick an adequate default
-	 * elevator if one is needed, that is, for devices requesting queue
-	 * registration.
-	 */
-	elevator_init_mq(disk->queue);
+	if (queue_is_mq(disk->queue)) {
+		/*
+		 * ->submit_bio and ->poll_bio are bypassed for blk-mq drivers.
+		 */
+		if (disk->fops->submit_bio || disk->fops->poll_bio)
+			return -EINVAL;
 
-	/* Mark bdev as having a submit_bio, if needed */
-	if (disk->fops->submit_bio)
+		/*
+		 * Initialize the I/O scheduler code and pick a default one if
+		 * needed.
+		 */
+		elevator_init_mq(disk->queue);
+	} else {
+		if (!disk->fops->submit_bio)
+			return -EINVAL;
 		bdev_set_flag(disk->part0, BD_HAS_SUBMIT_BIO);
+	}
 
 	/*
 	 * If the driver provides an explicit major number it also must provide
···
 	struct request_queue *q = disk->queue;
 	struct block_device *part;
 	unsigned long idx;
-	bool start_drain, queue_dying;
+	bool start_drain;
 
 	might_sleep();
 
···
 	 */
 	mutex_lock(&disk->open_mutex);
 	start_drain = __blk_mark_disk_dead(disk);
-	queue_dying = blk_queue_dying(q);
 	if (start_drain)
-		blk_freeze_acquire_lock(q, true, queue_dying);
+		blk_freeze_acquire_lock(q);
 	xa_for_each_start(&disk->part_tbl, idx, part, 1)
 		drop_partition(part);
 	mutex_unlock(&disk->open_mutex);
···
 		blk_mq_exit_queue(q);
 
 	if (start_drain)
-		blk_unfreeze_release_lock(q, true, queue_dying);
+		blk_unfreeze_release_lock(q);
 }
 EXPORT_SYMBOL(del_gendisk);
 
···
 }
 
 #ifdef CONFIG_BLOCK_LEGACY_AUTOLOAD
-void blk_request_module(dev_t devt)
+static bool blk_probe_dev(dev_t devt)
 {
 	unsigned int major = MAJOR(devt);
 	struct blk_major_name **n;
···
 		if ((*n)->major == major && (*n)->probe) {
 			(*n)->probe(devt);
 			mutex_unlock(&major_names_lock);
-			return;
+			return true;
 		}
 	}
 	mutex_unlock(&major_names_lock);
+	return false;
+}
 
-	if (request_module("block-major-%d-%d", MAJOR(devt), MINOR(devt)) > 0)
-		/* Make old-style 2.4 aliases work */
-		request_module("block-major-%d", MAJOR(devt));
+void blk_request_module(dev_t devt)
+{
+	int error;
+
+	if (blk_probe_dev(devt))
+		return;
+
+	error = request_module("block-major-%d-%d", MAJOR(devt), MINOR(devt));
+	/* Make old-style 2.4 aliases work */
+	if (error > 0)
+		error = request_module("block-major-%d", MAJOR(devt));
+	if (!error)
+		blk_probe_dev(devt);
 }
 #endif /* CONFIG_BLOCK_LEGACY_AUTOLOAD */
 

block/kyber-iosched.c

···
 #undef KYBER_LAT_SHOW_STORE
 
 #define KYBER_LAT_ATTR(op) __ATTR(op##_lat_nsec, 0644, kyber_##op##_lat_show, kyber_##op##_lat_store)
-static struct elv_fs_entry kyber_sched_attrs[] = {
+static const struct elv_fs_entry kyber_sched_attrs[] = {
 	KYBER_LAT_ATTR(read),
 	KYBER_LAT_ATTR(write),
 	__ATTR_NULL

block/mq-deadline.c

···
 #define DD_ATTR(name) \
 	__ATTR(name, 0644, deadline_##name##_show, deadline_##name##_store)
 
-static struct elv_fs_entry deadline_attrs[] = {
+static const struct elv_fs_entry deadline_attrs[] = {
 	DD_ATTR(read_expire),
 	DD_ATTR(write_expire),
 	DD_ATTR(writes_starved),

block/partitions/ldm.h

···
 // SPDX-License-Identifier: GPL-2.0-or-later
-/**
+/*
  * ldm - Part of the Linux-NTFS project.
  *
  * Copyright (C) 2001,2002 Richard Russon <[e-mail]>

drivers/ata/ahci.h

···
 	.shost_groups		= ahci_shost_groups,			\
 	.sdev_groups		= ahci_sdev_groups,			\
 	.change_queue_depth     = ata_scsi_change_queue_depth,		\
-	.tag_alloc_policy       = BLK_TAG_ALLOC_RR,             	\
+	.tag_alloc_policy_rr	= true,					\
 	.device_configure	= ata_scsi_device_configure
 
 extern struct ata_port_operations ahci_ops;

drivers/ata/pata_macio.c

···
 	.device_configure	= pata_macio_device_configure,
 	.sdev_groups		= ata_common_sdev_groups,
 	.can_queue		= ATA_DEF_QUEUE,
-	.tag_alloc_policy	= BLK_TAG_ALLOC_RR,
+	.tag_alloc_policy_rr	= true,
 };
 
 static struct ata_port_operations pata_macio_ops = {

drivers/ata/sata_mv.c

···
 	.dma_boundary		= MV_DMA_BOUNDARY,
 	.sdev_groups		= ata_ncq_sdev_groups,
 	.change_queue_depth	= ata_scsi_change_queue_depth,
-	.tag_alloc_policy	= BLK_TAG_ALLOC_RR,
+	.tag_alloc_policy_rr	= true,
 	.device_configure	= ata_scsi_device_configure
 };
 

drivers/ata/sata_nv.c

···
 	.device_configure	= nv_adma_device_configure,
 	.sdev_groups		= ata_ncq_sdev_groups,
 	.change_queue_depth     = ata_scsi_change_queue_depth,
-	.tag_alloc_policy	= BLK_TAG_ALLOC_RR,
+	.tag_alloc_policy_rr	= true,
 };
 
 static const struct scsi_host_template nv_swncq_sht = {
···
 	.device_configure	= nv_swncq_device_configure,
 	.sdev_groups		= ata_ncq_sdev_groups,
 	.change_queue_depth     = ata_scsi_change_queue_depth,
-	.tag_alloc_policy	= BLK_TAG_ALLOC_RR,
+	.tag_alloc_policy_rr	= true,
 };
 
 /*

drivers/ata/sata_sil24.c

···
 	.can_queue		= SIL24_MAX_CMDS,
 	.sg_tablesize		= SIL24_MAX_SGE,
 	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.tag_alloc_policy	= BLK_TAG_ALLOC_FIFO,
 	.sdev_groups		= ata_ncq_sdev_groups,
 	.change_queue_depth	= ata_scsi_change_queue_depth,
 	.device_configure	= ata_scsi_device_configure

drivers/block/amiflop.c

···
 	unit[drive].tag_set.nr_maps = 1;
 	unit[drive].tag_set.queue_depth = 2;
 	unit[drive].tag_set.numa_node = NUMA_NO_NODE;
-	unit[drive].tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
 	if (blk_mq_alloc_tag_set(&unit[drive].tag_set))
 		goto out_cleanup_trackbuf;
 

drivers/block/aoe/aoeblk.c

···
 	set->nr_hw_queues = 1;
 	set->queue_depth = 128;
 	set->numa_node = NUMA_NO_NODE;
-	set->flags = BLK_MQ_F_SHOULD_MERGE;
 	err = blk_mq_alloc_tag_set(set);
 	if (err) {
 		pr_err("aoe: cannot allocate tag set for %ld.%d\n",

drivers/block/ataflop.c

···
 		unit[i].tag_set.nr_maps = 1;
 		unit[i].tag_set.queue_depth = 2;
 		unit[i].tag_set.numa_node = NUMA_NO_NODE;
-		unit[i].tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
 		ret = blk_mq_alloc_tag_set(&unit[i].tag_set);
 		if (ret)
 			goto err;

drivers/block/floppy.c

···
 		tag_sets[drive].nr_maps = 1;
 		tag_sets[drive].queue_depth = 2;
 		tag_sets[drive].numa_node = NUMA_NO_NODE;
-		tag_sets[drive].flags = BLK_MQ_F_SHOULD_MERGE;
 		err = blk_mq_alloc_tag_set(&tag_sets[drive]);
 		if (err)
 			goto out_put_disk;

drivers/block/loop.c

···
 	struct list_head        idle_worker_list;
 	struct rb_root          worker_tree;
 	struct timer_list       timer;
-	bool			use_dio;
 	bool			sysfs_inited;
 
 	struct request_queue	*lo_queue;
···
 	return true;
 }
 
-static void __loop_update_dio(struct loop_device *lo, bool dio)
+static bool lo_can_use_dio(struct loop_device *lo)
 {
-	struct file *file = lo->lo_backing_file;
-	struct inode *inode = file->f_mapping->host;
-	struct block_device *backing_bdev = NULL;
-	bool use_dio;
+	struct inode *inode = lo->lo_backing_file->f_mapping->host;
+
+	if (!(lo->lo_backing_file->f_mode & FMODE_CAN_ODIRECT))
+		return false;
 
 	if (S_ISBLK(inode->i_mode))
-		backing_bdev = I_BDEV(inode);
-	else if (inode->i_sb->s_bdev)
-		backing_bdev = inode->i_sb->s_bdev;
+		return lo_bdev_can_use_dio(lo, I_BDEV(inode));
+	if (inode->i_sb->s_bdev)
+		return lo_bdev_can_use_dio(lo, inode->i_sb->s_bdev);
+	return true;
+}
 
-	use_dio = dio && (file->f_mode & FMODE_CAN_ODIRECT) &&
-		(!backing_bdev || lo_bdev_can_use_dio(lo, backing_bdev));
+/*
+ * Direct I/O can be enabled either by using an O_DIRECT file descriptor, or by
+ * passing in the LO_FLAGS_DIRECT_IO flag from userspace.  It will be silently
+ * disabled when the device block size is too small or the offset is unaligned.
+ *
+ * loop_get_status will always report the effective LO_FLAGS_DIRECT_IO flag and
+ * not the originally passed in one.
+ */
+static inline void loop_update_dio(struct loop_device *lo)
+{
+	bool dio_in_use = lo->lo_flags & LO_FLAGS_DIRECT_IO;
 
-	if (lo->use_dio == use_dio)
-		return;
+	lockdep_assert_held(&lo->lo_mutex);
+	WARN_ON_ONCE(lo->lo_state == Lo_bound &&
+		     lo->lo_queue->mq_freeze_depth == 0);
 
-	/* flush dirty pages before changing direct IO */
-	vfs_fsync(file, 0);
-
-	/*
-	 * The flag of LO_FLAGS_DIRECT_IO is handled similarly with
-	 * LO_FLAGS_READ_ONLY, both are set from kernel, and losetup
-	 * will get updated by ioctl(LOOP_GET_STATUS)
-	 */
-	if (lo->lo_state == Lo_bound)
-		blk_mq_freeze_queue(lo->lo_queue);
-	lo->use_dio = use_dio;
-	if (use_dio)
+	if (lo->lo_backing_file->f_flags & O_DIRECT)
 		lo->lo_flags |= LO_FLAGS_DIRECT_IO;
-	else
+	if ((lo->lo_flags & LO_FLAGS_DIRECT_IO) && !lo_can_use_dio(lo))
 		lo->lo_flags &= ~LO_FLAGS_DIRECT_IO;
-	if (lo->lo_state == Lo_bound)
-		blk_mq_unfreeze_queue(lo->lo_queue);
+
+	/* flush dirty pages before starting to issue direct I/O */
+	if ((lo->lo_flags & LO_FLAGS_DIRECT_IO) && !dio_in_use)
+		vfs_fsync(lo->lo_backing_file, 0);
 }
 
 /**
···
 		lim.discard_granularity = 0;
 	}
 
+	/*
+	 * XXX: this updates the queue limits without freezing the queue, which
+	 * is against the locking protocol and dangerous.  But we can't just
+	 * freeze the queue as we're inside the ->queue_rq method here.  So this
+	 * should move out into a workqueue unless we get the file operations to
+	 * advertise if they support specific fallocate operations.
+	 */
 	queue_limits_commit_update(lo->lo_queue, &lim);
 }
 
···
 		WARN_ON_ONCE(1);
 		return -EIO;
 	}
-}
-
-static inline void loop_update_dio(struct loop_device *lo)
-{
-	__loop_update_dio(lo, (lo->lo_backing_file->f_flags & O_DIRECT) |
-				lo->use_dio);
 }
 
 static void loop_reread_partitions(struct loop_device *lo)
···
 
 	memcpy(lo->lo_file_name, info->lo_file_name, LO_NAME_SIZE);
 	lo->lo_file_name[LO_NAME_SIZE-1] = 0;
-	lo->lo_flags = info->lo_flags;
 	return 0;
 }
 
···
 	return SECTOR_SIZE;
 }
 
-static int loop_reconfigure_limits(struct loop_device *lo, unsigned int bsize)
+static void loop_update_limits(struct loop_device *lo, struct queue_limits *lim,
+		unsigned int bsize)
 {
 	struct file *file = lo->lo_backing_file;
 	struct inode *inode = file->f_mapping->host;
 	struct block_device *backing_bdev = NULL;
-	struct queue_limits lim;
 	u32 granularity = 0, max_discard_sectors = 0;
 
 	if (S_ISBLK(inode->i_mode))
···
 
 	loop_get_discard_config(lo, &granularity, &max_discard_sectors);
 
-	lim = queue_limits_start_update(lo->lo_queue);
-	lim.logical_block_size = bsize;
-	lim.physical_block_size = bsize;
-	lim.io_min = bsize;
-	lim.features &= ~(BLK_FEAT_WRITE_CACHE | BLK_FEAT_ROTATIONAL);
+	lim->logical_block_size = bsize;
+	lim->physical_block_size = bsize;
+	lim->io_min = bsize;
+	lim->features &= ~(BLK_FEAT_WRITE_CACHE | BLK_FEAT_ROTATIONAL);
 	if (file->f_op->fsync && !(lo->lo_flags & LO_FLAGS_READ_ONLY))
-		lim.features |= BLK_FEAT_WRITE_CACHE;
+		lim->features |= BLK_FEAT_WRITE_CACHE;
 	if (backing_bdev && !bdev_nonrot(backing_bdev))
-		lim.features |= BLK_FEAT_ROTATIONAL;
-	lim.max_hw_discard_sectors = max_discard_sectors;
-	lim.max_write_zeroes_sectors = max_discard_sectors;
+		lim->features |= BLK_FEAT_ROTATIONAL;
+	lim->max_hw_discard_sectors = max_discard_sectors;
+	lim->max_write_zeroes_sectors = max_discard_sectors;
 	if (max_discard_sectors)
-		lim.discard_granularity = granularity;
+		lim->discard_granularity = granularity;
 	else
-		lim.discard_granularity = 0;
-	return queue_limits_commit_update(lo->lo_queue, &lim);
+		lim->discard_granularity = 0;
 }
 
 static int loop_configure(struct loop_device *lo, blk_mode_t mode,
···
 {
 	struct file *file = fget(config->fd);
 	struct address_space *mapping;
+	struct queue_limits lim;
 	int error;
 	loff_t size;
 	bool partscan;
···
 	error = loop_set_status_from_info(lo, &config->info);
 	if (error)
 		goto out_unlock;
+	lo->lo_flags = config->info.lo_flags;
 
 	if (!(file->f_mode & FMODE_WRITE) || !(mode & BLK_OPEN_WRITE) ||
 	    !file->f_op->write_iter)
···
 	disk_force_media_change(lo->lo_disk);
 	set_disk_ro(lo->lo_disk, (lo->lo_flags & LO_FLAGS_READ_ONLY) != 0);
 
-	lo->use_dio = lo->lo_flags & LO_FLAGS_DIRECT_IO;
 	lo->lo_device = bdev;
 	lo->lo_backing_file = file;
 	lo->old_gfp_mask = mapping_gfp_mask(mapping);
 	mapping_set_gfp_mask(mapping, lo->old_gfp_mask & ~(__GFP_IO|__GFP_FS));
 
-	error = loop_reconfigure_limits(lo, config->block_size);
+	lim = queue_limits_start_update(lo->lo_queue);
+	loop_update_limits(lo, &lim, config->block_size);
+	/* No need to freeze the queue as the device isn't bound yet. */
+	error = queue_limits_commit_update(lo->lo_queue, &lim);
 	if (error)
 		goto out_unlock;
 
···
 	lo->lo_sizelimit = 0;
 	memset(lo->lo_file_name, 0, LO_NAME_SIZE);
 
-	/* reset the block size to the default */
+	/*
+	 * Reset the block size to the default.
+	 *
+	 * No queue freezing needed because this is called from the final
+	 * ->release call only, so there can't be any outstanding I/O.
+	 */
 	lim = queue_limits_start_update(lo->lo_queue);
 	lim.logical_block_size = SECTOR_SIZE;
 	lim.physical_block_size = SECTOR_SIZE;
···
 loop_set_status(struct loop_device *lo, const struct loop_info64 *info)
 {
 	int err;
-	int prev_lo_flags;
 	bool partscan = false;
 	bool size_changed = false;
 
···
 		invalidate_bdev(lo->lo_device);
 	}
 
-	/* I/O need to be drained during transfer transition */
+	/* I/O needs to be drained before changing lo_offset or lo_sizelimit */
 	blk_mq_freeze_queue(lo->lo_queue);
-
-	prev_lo_flags = lo->lo_flags;
 
 	err = loop_set_status_from_info(lo, info);
 	if (err)
 		goto out_unfreeze;
 
-	/* Mask out flags that can't be set using LOOP_SET_STATUS. */
-	lo->lo_flags &= LOOP_SET_STATUS_SETTABLE_FLAGS;
-	/* For those flags, use the previous values instead */
-	lo->lo_flags |= prev_lo_flags & ~LOOP_SET_STATUS_SETTABLE_FLAGS;
-	/* For flags that can't be cleared, use previous values too */
-	lo->lo_flags |= prev_lo_flags & ~LOOP_SET_STATUS_CLEARABLE_FLAGS;
+	partscan = !(lo->lo_flags & LO_FLAGS_PARTSCAN) &&
+		(info->lo_flags & LO_FLAGS_PARTSCAN);
+
+	lo->lo_flags &= ~(LOOP_SET_STATUS_SETTABLE_FLAGS |
+			  LOOP_SET_STATUS_CLEARABLE_FLAGS);
+	lo->lo_flags |= (info->lo_flags & LOOP_SET_STATUS_SETTABLE_FLAGS);
 
 	if (size_changed) {
 		loff_t new_size = get_size(lo->lo_offset, lo->lo_sizelimit,
···
 		loop_set_size(lo, new_size);
 	}
 
-	/* update dio if lo_offset or transfer is changed */
-	__loop_update_dio(lo, lo->use_dio);
+	/* update the direct I/O flag if lo_offset changed */
+	loop_update_dio(lo);
 
 out_unfreeze:
 	blk_mq_unfreeze_queue(lo->lo_queue);
-
-	if (!err && (lo->lo_flags & LO_FLAGS_PARTSCAN) &&
-	     !(prev_lo_flags & LO_FLAGS_PARTSCAN)) {
+	if (partscan)
 		clear_bit(GD_SUPPRESS_PART_SCAN, &lo->lo_disk->state);
-		partscan = true;
-	}
 out_unlock:
 	mutex_unlock(&lo->lo_mutex);
 	if (partscan)
···
 
 static int loop_set_dio(struct loop_device *lo, unsigned long arg)
 {
-	int error = -ENXIO;
-	if (lo->lo_state != Lo_bound)
-		goto out;
+	bool use_dio = !!arg;
 
-	__loop_update_dio(lo, !!arg);
-	if (lo->use_dio == !!arg)
+	if (lo->lo_state != Lo_bound)
+		return -ENXIO;
+	if (use_dio == !!(lo->lo_flags & LO_FLAGS_DIRECT_IO))
 		return 0;
-	error = -EINVAL;
- out:
-	return error;
+
+	if (use_dio) {
+		if (!lo_can_use_dio(lo))
+			return -EINVAL;
+		/* flush dirty pages before starting to use direct I/O */
+		vfs_fsync(lo->lo_backing_file, 0);
+	}
+
+	blk_mq_freeze_queue(lo->lo_queue);
+	if (use_dio)
+		lo->lo_flags |= LO_FLAGS_DIRECT_IO;
+	else
+		lo->lo_flags &= ~LO_FLAGS_DIRECT_IO;
+	blk_mq_unfreeze_queue(lo->lo_queue);
+	return 0;
 }
 
 static int loop_set_block_size(struct loop_device *lo, unsigned long arg)
 {
+	struct queue_limits lim;
 	int err = 0;
 
 	if (lo->lo_state != Lo_bound)
···
 	sync_blockdev(lo->lo_device);
 	invalidate_bdev(lo->lo_device);
 
+	lim = queue_limits_start_update(lo->lo_queue);
+	loop_update_limits(lo, &lim, arg);
+
 	blk_mq_freeze_queue(lo->lo_queue);
-	err = loop_reconfigure_limits(lo, arg);
+	err = queue_limits_commit_update(lo->lo_queue, &lim);
 	loop_update_dio(lo);
 	blk_mq_unfreeze_queue(lo->lo_queue);
 
···
 		cmd->use_aio = false;
 		break;
 	default:
-		cmd->use_aio = lo->use_dio;
+		cmd->use_aio = lo->lo_flags & LO_FLAGS_DIRECT_IO;
 		break;
 	}
 
···
 	lo->tag_set.queue_depth = hw_queue_depth;
 	lo->tag_set.numa_node = NUMA_NO_NODE;
 	lo->tag_set.cmd_size = sizeof(struct loop_cmd);
-	lo->tag_set.flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_STACKING |
-		BLK_MQ_F_NO_SCHED_BY_DEFAULT;
+	lo->tag_set.flags = BLK_MQ_F_STACKING | BLK_MQ_F_NO_SCHED_BY_DEFAULT;
 	lo->tag_set.driver_data = lo;
 
 	err = blk_mq_alloc_tag_set(&lo->tag_set);

drivers/block/mtip32xx/mtip32xx.c

···
 	dd->tags.reserved_tags = 1;
 	dd->tags.cmd_size = sizeof(struct mtip_cmd);
 	dd->tags.numa_node = dd->numa_node;
-	dd->tags.flags = BLK_MQ_F_SHOULD_MERGE;
 	dd->tags.driver_data = dd;
 	dd->tags.timeout = MTIP_NCQ_CMD_TIMEOUT_MS;
 

drivers/block/nbd.c

···
 	bool dead;
 	int fallback_index;
 	int cookie;
+	struct work_struct work;
 };
 
 struct recv_thread_args {
···
  * by cmd->lock.
  */
 #define NBD_CMD_INFLIGHT	2
+
+/* Just part of request header or data payload is sent successfully */
+#define NBD_CMD_PARTIAL_SEND	3
 
 struct nbd_cmd {
 	struct nbd_device *nbd;
···
 	nsock->sent = 0;
 }
 
-static int __nbd_set_size(struct nbd_device *nbd, loff_t bytesize,
-		loff_t blksize)
+static int nbd_set_size(struct nbd_device *nbd, loff_t bytesize, loff_t blksize)
 {
 	struct queue_limits lim;
 	int error;
···
 
 	lim.logical_block_size = blksize;
 	lim.physical_block_size = blksize;
-	error = queue_limits_commit_update(nbd->disk->queue, &lim);
+	error = queue_limits_commit_update_frozen(nbd->disk->queue, &lim);
 	if (error)
 		return error;
 
···
 	if (!set_capacity_and_notify(nbd->disk, bytesize >> 9))
 		kobject_uevent(&nbd_to_dev(nbd)->kobj, KOBJ_CHANGE);
 	return 0;
-}
-
-static int nbd_set_size(struct nbd_device *nbd, loff_t bytesize,
-		loff_t blksize)
-{
-	int error;
-
-	blk_mq_freeze_queue(nbd->disk->queue);
-	error = __nbd_set_size(nbd, bytesize, blksize);
-	blk_mq_unfreeze_queue(nbd->disk->queue);
-
-	return error;
 }
 
 static void nbd_complete_rq(struct request *req)
···
 
 	if (!mutex_trylock(&cmd->lock))
 		return BLK_EH_RESET_TIMER;
+
+	/* partial send is handled in nbd_sock's work function */
+	if (test_bit(NBD_CMD_PARTIAL_SEND, &cmd->flags)) {
+		mutex_unlock(&cmd->lock);
+		return BLK_EH_RESET_TIMER;
+	}
 
 	if (!test_bit(NBD_CMD_INFLIGHT, &cmd->flags)) {
 		mutex_unlock(&cmd->lock);
···
 }
 
 /*
+ * We've already sent header or part of data payload, have no choice but
+ * to set pending and schedule it in work.
+ *
+ * And we have to return BLK_STS_OK to block core, otherwise this same
+ * request may be re-dispatched with different tag, but our header has
+ * been sent out with old tag, and this way does confuse reply handling.
+ */
+static void nbd_sched_pending_work(struct nbd_device *nbd,
+				   struct nbd_sock *nsock,
+				   struct nbd_cmd *cmd, int sent)
+{
+	struct request *req = blk_mq_rq_from_pdu(cmd);
+
+	/* pending work should be scheduled only once */
+	WARN_ON_ONCE(test_bit(NBD_CMD_PARTIAL_SEND, &cmd->flags));
+
+	nsock->pending = req;
+	nsock->sent = sent;
+	set_bit(NBD_CMD_PARTIAL_SEND, &cmd->flags);
+	refcount_inc(&nbd->config_refs);
+	schedule_work(&nsock->work);
+}
+
+/*
  * Returns BLK_STS_RESOURCE if the caller should retry after a delay.
  * Returns BLK_STS_IOERR if sending failed.
  */
···
 			 * completely done.
 			 */
 			if (sent) {
-				nsock->pending = req;
-				nsock->sent = sent;
+				nbd_sched_pending_work(nbd, nsock, cmd, sent);
+				return BLK_STS_OK;
 			}
 			set_bit(NBD_CMD_REQUEUED, &cmd->flags);
 			return BLK_STS_RESOURCE;
···
 			result = sock_xmit(nbd, index, 1, &from, flags, &sent);
 			if (result < 0) {
 				if (was_interrupted(result)) {
-					/* We've already sent the header, we
-					 * have no choice but to set pending and
-					 * return BUSY.
-					 */
-					nsock->pending = req;
-					nsock->sent = sent;
-					set_bit(NBD_CMD_REQUEUED, &cmd->flags);
-					return BLK_STS_RESOURCE;
+					nbd_sched_pending_work(nbd, nsock, cmd, sent);
+					return BLK_STS_OK;
 				}
 				dev_err(disk_to_dev(nbd->disk),
 					"Send data failed (result %d)\n",
···
 	return BLK_STS_OK;
 
 requeue:
+	/*
+	 * Can't requeue in case we are dealing with partial send
+	 *
+	 * We must run from pending work function.
+	 * */
+	if (test_bit(NBD_CMD_PARTIAL_SEND, &cmd->flags))
+		return BLK_STS_OK;
+
 	/* retry on a different socket */
 	dev_err_ratelimited(disk_to_dev(nbd->disk),
 			    "Request send failed, requeueing\n");
 	nbd_mark_nsock_dead(nbd, nsock, 1);
 	nbd_requeue_cmd(cmd);
 	return BLK_STS_OK;
+}
+
+/* handle partial sending */
+static void nbd_pending_cmd_work(struct work_struct *work)
+{
+	struct nbd_sock *nsock = container_of(work, struct nbd_sock, work);
+	struct request *req = nsock->pending;
+	struct nbd_cmd *cmd = blk_mq_rq_to_pdu(req);
+	struct nbd_device *nbd = cmd->nbd;
+	unsigned long deadline = READ_ONCE(req->deadline);
+	unsigned int wait_ms = 2;
+
+	mutex_lock(&cmd->lock);
+
+	WARN_ON_ONCE(test_bit(NBD_CMD_REQUEUED, &cmd->flags));
+	if (WARN_ON_ONCE(!test_bit(NBD_CMD_PARTIAL_SEND, &cmd->flags)))
+		goto out;
+
+	mutex_lock(&nsock->tx_lock);
+	while (true) {
+		nbd_send_cmd(nbd, cmd, cmd->index);
+		if (!nsock->pending)
+			break;
+
+		/* don't bother timeout handler for partial sending */
+		if (READ_ONCE(jiffies) + msecs_to_jiffies(wait_ms) >= deadline) {
+			cmd->status = BLK_STS_IOERR;
+			blk_mq_complete_request(req);
+			break;
+		}
+		msleep(wait_ms);
+		wait_ms *= 2;
+	}
+	mutex_unlock(&nsock->tx_lock);
+	clear_bit(NBD_CMD_PARTIAL_SEND, &cmd->flags);
+out:
+	mutex_unlock(&cmd->lock);
+	nbd_config_put(nbd);
 }
 
 static int nbd_read_reply(struct nbd_device *nbd, struct socket *sock,
···
 	nsock->pending = NULL;
 	nsock->sent = 0;
 	nsock->cookie = 0;
+	INIT_WORK(&nsock->work, nbd_pending_cmd_work);
 	socks[config->num_connections++] = nsock;
 	atomic_inc(&config->live_connections);
 	blk_mq_unfreeze_queue(nbd->disk->queue);
···
 	nbd->tag_set.queue_depth = 128;
 	nbd->tag_set.numa_node = NUMA_NO_NODE;
 	nbd->tag_set.cmd_size = sizeof(struct nbd_cmd);
-	nbd->tag_set.flags = BLK_MQ_F_SHOULD_MERGE |
-		BLK_MQ_F_BLOCKING;
+	nbd->tag_set.flags = BLK_MQ_F_BLOCKING;
 	nbd->tag_set.driver_data = nbd;
 	INIT_WORK(&nbd->remove_work, nbd_dev_remove_work);
 	nbd->backend = NULL;
···
 	flush_workqueue(nbd->recv_workq);
 	nbd_clear_que(nbd);
 	nbd->task_setup = NULL;
+	clear_bit(NBD_RT_BOUND, &nbd->config->runtime_flags);
 	mutex_unlock(&nbd->config_lock);
 
 	if (test_and_clear_bit(NBD_RT_HAS_CONFIG_REF,

drivers/block/null_blk/main.c

···
 module_param_named(zone_full, g_zone_full, bool, S_IRUGO);
 MODULE_PARM_DESC(zone_full, "Initialize the sequential write required zones of a zoned device to be full. Default: false");
 
+static bool g_rotational;
+module_param_named(rotational, g_rotational, bool, S_IRUGO);
+MODULE_PARM_DESC(rotational, "Set the rotational feature for the device. Default: false");
+
 static struct nullb_device *null_alloc_dev(void);
 static void null_free_dev(struct nullb_device *dev);
 static void null_del_dev(struct nullb *nullb);
···
 NULLB_DEVICE_ATTR(shared_tags, bool, NULL);
 NULLB_DEVICE_ATTR(shared_tag_bitmap, bool, NULL);
 NULLB_DEVICE_ATTR(fua, bool, NULL);
+NULLB_DEVICE_ATTR(rotational, bool, NULL);
 
 static ssize_t nullb_device_power_show(struct config_item *item, char *page)
 {
···
 	&nullb_device_attr_shared_tags,
 	&nullb_device_attr_shared_tag_bitmap,
 	&nullb_device_attr_fua,
+	&nullb_device_attr_rotational,
 	NULL,
 };
 
···
 			"shared_tags,size,submit_queues,use_per_node_hctx,"
 			"virt_boundary,zoned,zone_capacity,zone_max_active,"
 			"zone_max_open,zone_nr_conv,zone_offline,zone_readonly,"
-			"zone_size,zone_append_max_sectors,zone_full\n");
+			"zone_size,zone_append_max_sectors,zone_full,"
+			"rotational\n");
 }
 
 CONFIGFS_ATTR_RO(memb_group_, features);
···
 	dev->shared_tags = g_shared_tags;
 	dev->shared_tag_bitmap = g_shared_tag_bitmap;
 	dev->fua = g_fua;
+	dev->rotational = g_rotational;
 
 	return dev;
 }
···
 	if (radix_tree_insert(root, idx, t_page)) {
 		null_free_page(t_page);
 		t_page = radix_tree_lookup(root, idx);
-		WARN_ON(!t_page || t_page->page->index != idx);
+		WARN_ON(!t_page || t_page->page->private != idx);
 	} else if (is_cache)
 		nullb->dev->curr_cache += PAGE_SIZE;
 
···
 				(void **)t_pages, pos, FREE_BATCH);
 
 		for (i = 0; i < nr_pages; i++) {
-			pos = t_pages[i]->page->index;
+			pos = t_pages[i]->page->private;
 			ret = radix_tree_delete_item(root, pos, t_pages[i]);
 			WARN_ON(ret != t_pages[i]);
 			null_free_page(ret);
···
 
 	root = is_cache ? &nullb->dev->cache : &nullb->dev->data;
 	t_page = radix_tree_lookup(root, idx);
-	WARN_ON(t_page && t_page->page->index != idx);
+	WARN_ON(t_page && t_page->page->private != idx);
 
 	if (t_page && (for_write || test_bit(sector_bit, t_page->bitmap)))
 		return t_page;
···
 
 	spin_lock_irq(&nullb->lock);
 	idx = sector >> PAGE_SECTORS_SHIFT;
-	t_page->page->index = idx;
+	t_page->page->private = idx;
 	t_page = null_radix_tree_insert(nullb, idx, t_page, !ignore_cache);
 	radix_tree_preload_end();
 
···
 	struct nullb_page *t_page, *ret;
 	void *dst, *src;
 
-	idx = c_page->page->index;
+	idx = c_page->page->private;
 
 	t_page = null_insert_page(nullb, idx << PAGE_SECTORS_SHIFT, true);
 
···
 	 * avoid race, we don't allow page free
 	 */
 	for (i = 0; i < nr_pages; i++) {
-		nullb->cache_flush_pos = c_pages[i]->page->index;
+		nullb->cache_flush_pos = c_pages[i]->page->private;
 		/*
 		 * We found the page which is being flushed to disk by other
 		 * threads
···
 	tag_set.nr_hw_queues = g_submit_queues;
 	tag_set.queue_depth = g_hw_queue_depth;
 	tag_set.numa_node = g_home_node;
-	tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
 	if (g_no_sched)
-		tag_set.flags |= BLK_MQ_F_NO_SCHED;
+		tag_set.flags |= BLK_MQ_F_NO_SCHED_BY_DEFAULT;
 	if (g_shared_tag_bitmap)
 		tag_set.flags |= BLK_MQ_F_TAG_HCTX_SHARED;
 	if (g_blocking)
···
 	nullb->tag_set->nr_hw_queues = nullb->dev->submit_queues;
 	nullb->tag_set->queue_depth = nullb->dev->hw_queue_depth;
 	nullb->tag_set->numa_node = nullb->dev->home_node;
-	nullb->tag_set->flags = BLK_MQ_F_SHOULD_MERGE;
 	if (nullb->dev->no_sched)
-		nullb->tag_set->flags |= BLK_MQ_F_NO_SCHED;
+		nullb->tag_set->flags |= BLK_MQ_F_NO_SCHED_BY_DEFAULT;
 	if (nullb->dev->shared_tag_bitmap)
 		nullb->tag_set->flags |= BLK_MQ_F_TAG_HCTX_SHARED;
 	if (nullb->dev->blocking)
···
 		if (dev->fua)
 			lim.features |= BLK_FEAT_FUA;
 	}
+
+	if (dev->rotational)
+		lim.features |= BLK_FEAT_ROTATIONAL;
 
 	nullb->disk = blk_mq_alloc_disk(nullb->tag_set, &lim, nullb);
 	if (IS_ERR(nullb->disk)) {

drivers/block/null_blk/null_blk.h

···
 	bool shared_tags; /* share tag set between devices for blk-mq */
 	bool shared_tag_bitmap; /* use hostwide shared tags */
 	bool fua; /* Support FUA */
+	bool rotational; /* Fake rotational device */
 };
 
 struct nullb {

drivers/block/ps3disk.c

···
 	unsigned int devidx;
 	struct queue_limits lim = {
 		.logical_block_size	= dev->blk_size,
-		.max_hw_sectors		= dev->bounce_size >> 9,
+		.max_hw_sectors		= BOUNCE_SIZE >> 9,
 		.max_segments		= -1,
-		.max_segment_size	= dev->bounce_size,
+		.max_segment_size	= BOUNCE_SIZE,
 		.dma_alignment		= dev->blk_size - 1,
 		.features		= BLK_FEAT_WRITE_CACHE |
 					  BLK_FEAT_ROTATIONAL,
···
 
 	ps3disk_identify(dev);
 
-	error = blk_mq_alloc_sq_tag_set(&priv->tag_set, &ps3disk_mq_ops, 1,
-					BLK_MQ_F_SHOULD_MERGE);
+	error = blk_mq_alloc_sq_tag_set(&priv->tag_set, &ps3disk_mq_ops, 1, 0);
 	if (error)
 		goto fail_teardown;
 

drivers/block/rbd.c

···
 	rbd_dev->tag_set.ops = &rbd_mq_ops;
 	rbd_dev->tag_set.queue_depth = rbd_dev->opts->queue_depth;
 	rbd_dev->tag_set.numa_node = NUMA_NO_NODE;
-	rbd_dev->tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
 	rbd_dev->tag_set.nr_hw_queues = num_present_cpus();
 	rbd_dev->tag_set.cmd_size = sizeof(struct rbd_img_request);
 

drivers/block/rnbd/rnbd-clt.c

···
 	tag_set->ops		= &rnbd_mq_ops;
 	tag_set->queue_depth	= sess->queue_depth;
 	tag_set->numa_node		= NUMA_NO_NODE;
-	tag_set->flags		= BLK_MQ_F_SHOULD_MERGE |
-				  BLK_MQ_F_TAG_QUEUE_SHARED;
+	tag_set->flags		= BLK_MQ_F_TAG_QUEUE_SHARED;
 	tag_set->cmd_size	= sizeof(struct rnbd_iu) + RNBD_RDMA_SGL_SIZE;
 
 	/* for HCTX_TYPE_DEFAULT, HCTX_TYPE_READ, HCTX_TYPE_POLL */

drivers/block/rnbd/rnbd-srv.c

···
 	bio->bi_iter.bi_sector = le64_to_cpu(msg->sector);
 	prio = srv_sess->ver < RNBD_PROTO_VER_MAJOR ||
 	       usrlen < sizeof(*msg) ? 0 : le16_to_cpu(msg->prio);
-	bio_set_prio(bio, prio);
+	bio->bi_ioprio = prio;
 
 	submit_bio(bio);
 

drivers/block/rnull.rs

···
     license: "GPL v2",
 }
 
+#[pin_data]
 struct NullBlkModule {
-    _disk: Pin<KBox<Mutex<GenDisk<NullBlkDevice>>>>,
+    #[pin]
+    _disk: Mutex<GenDisk<NullBlkDevice>>,
 }
 
-impl kernel::Module for NullBlkModule {
-    fn init(_module: &'static ThisModule) -> Result<Self> {
+impl kernel::InPlaceModule for NullBlkModule {
+    fn init(_module: &'static ThisModule) -> impl PinInit<Self, Error> {
         pr_info!("Rust null_blk loaded\n");
-        let tagset = Arc::pin_init(TagSet::new(1, 256, 1), flags::GFP_KERNEL)?;
 
-        let disk = gen_disk::GenDiskBuilder::new()
-            .capacity_sectors(4096 << 11)
-            .logical_block_size(4096)?
-            .physical_block_size(4096)?
-            .rotational(false)
-            .build(format_args!("rnullb{}", 0), tagset)?;
+        // Use a immediately-called closure as a stable `try` block
+        let disk = /* try */ (|| {
+            let tagset = Arc::pin_init(TagSet::new(1, 256, 1), flags::GFP_KERNEL)?;
 
-        let disk = KBox::pin_init(new_mutex!(disk, "nullb:disk"), flags::GFP_KERNEL)?;
+            gen_disk::GenDiskBuilder::new()
+                .capacity_sectors(4096 << 11)
+                .logical_block_size(4096)?
+                .physical_block_size(4096)?
+                .rotational(false)
+                .build(format_args!("rnullb{}", 0), tagset)
+        })();
 
-        Ok(Self { _disk: disk })
+        try_pin_init!(Self {
+            _disk <- new_mutex!(disk?, "nullb:disk"),
+        })
     }
 }
 

drivers/block/sunvdc.c

···
 	}
 
 	err = blk_mq_alloc_sq_tag_set(&port->tag_set, &vdc_mq_ops,
-			VDC_TX_RING_SIZE, BLK_MQ_F_SHOULD_MERGE);
+			VDC_TX_RING_SIZE, 0);
 	if (err)
 		return err;
 

drivers/block/swim.c

···
 
 	for (drive = 0; drive < swd->floppy_count; drive++) {
 		err = blk_mq_alloc_sq_tag_set(&swd->unit[drive].tag_set,
-				&swim_mq_ops, 2, BLK_MQ_F_SHOULD_MERGE);
+				&swim_mq_ops, 2, 0);
 		if (err)
 			goto exit_put_disks;
 

drivers/block/swim3.c

···
 	fs = &floppy_states[floppy_count];
 	memset(fs, 0, sizeof(*fs));
 
-	rc = blk_mq_alloc_sq_tag_set(&fs->tag_set, &swim3_mq_ops, 2,
-			BLK_MQ_F_SHOULD_MERGE);
+	rc = blk_mq_alloc_sq_tag_set(&fs->tag_set, &swim3_mq_ops, 2, 0);
 	if (rc)
 		goto out_unregister;
 

drivers/block/ublk_drv.c

···
 	ub->tag_set.queue_depth = ub->dev_info.queue_depth;
 	ub->tag_set.numa_node = NUMA_NO_NODE;
 	ub->tag_set.cmd_size = sizeof(struct ublk_rq_data);
-	ub->tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
 	ub->tag_set.driver_data = ub;
 	return blk_mq_alloc_tag_set(&ub->tag_set);
 }

drivers/block/virtio_blk.c

···
 #include <linux/string_helpers.h>
 #include <linux/idr.h>
 #include <linux/blk-mq.h>
-#include <linux/blk-mq-virtio.h>
 #include <linux/numa.h>
 #include <linux/vmalloc.h>
 #include <uapi/linux/virtio_ring.h>
···
 		lim.features |= BLK_FEAT_WRITE_CACHE;
 	else
 		lim.features &= ~BLK_FEAT_WRITE_CACHE;
-	blk_mq_freeze_queue(disk->queue);
-	i = queue_limits_commit_update(disk->queue, &lim);
-	blk_mq_unfreeze_queue(disk->queue);
+	i = queue_limits_commit_update_frozen(disk->queue, &lim);
 	if (i)
 		return i;
 	return count;
···
 		if (i == HCTX_TYPE_POLL)
 			blk_mq_map_queues(&set->map[i]);
 		else
-			blk_mq_virtio_map_queues(&set->map[i], vblk->vdev, 0);
+			blk_mq_map_hw_queues(&set->map[i],
+					     &vblk->vdev->dev, 0);
 	}
 }
 
···
 	vblk->tag_set.ops = &virtio_mq_ops;
 	vblk->tag_set.queue_depth = queue_depth;
 	vblk->tag_set.numa_node = NUMA_NO_NODE;
-	vblk->tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
 	vblk->tag_set.cmd_size =
 		sizeof(struct virtblk_req) +
 		sizeof(struct scatterlist) * VIRTIO_BLK_INLINE_SG_CNT;

drivers/block/xen-blkfront.c

···
 	} else
 		info->tag_set.queue_depth = BLK_RING_SIZE(info);
 	info->tag_set.numa_node = NUMA_NO_NODE;
-	info->tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
 	info->tag_set.cmd_size = sizeof(struct blkif_req);
 	info->tag_set.driver_data = info;
 

drivers/block/z2ram.c

···
 	tag_set.nr_maps = 1;
 	tag_set.queue_depth = 16;
 	tag_set.numa_node = NUMA_NO_NODE;
-	tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
 	ret = blk_mq_alloc_tag_set(&tag_set);
 	if (ret)
 		goto out_unregister_blkdev;

drivers/cdrom/gdrom.c

···
 	probe_gdrom_setupcd();
 
 	err = blk_mq_alloc_sq_tag_set(&gd.tag_set, &gdrom_mq_ops, 1,
-				BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_BLOCKING);
+				BLK_MQ_F_BLOCKING);
 	if (err)
 		goto probe_fail_free_cd_info;
 

drivers/md/Kconfig

···
 	  various kernel APIs and can only work with files on a file system not
 	  actually sitting on the MD device.
 
+config MD_LINEAR
+	tristate "Linear (append) mode"
+	depends on BLK_DEV_MD
+	help
+	  If you say Y here, then your multiple devices driver will be able to
+	  use the so-called linear mode, i.e. it will combine the hard disk
+	  partitions by simply appending one to the other.
+
+	  To compile this as a module, choose M here: the module
+	  will be called linear.
+
+	  If unsure, say Y.
+
 config MD_RAID0
 	tristate "RAID-0 (striping) mode"
 	depends on BLK_DEV_MD

drivers/md/Makefile

···
 
 md-mod-y	+= md.o md-bitmap.o
 raid456-y	+= raid5.o raid5-cache.o raid5-ppl.o
+linear-y       += md-linear.o
 
 # Note: link order is important.  All raid personalities
 # and must come before md.o, as they each initialise
 # themselves, and md.o may use the personalities when it
 # auto-initialised.
 
+obj-$(CONFIG_MD_LINEAR)		+= linear.o
 obj-$(CONFIG_MD_RAID0)		+= raid0.o
 obj-$(CONFIG_MD_RAID1)		+= raid1.o
 obj-$(CONFIG_MD_RAID10)		+= raid10.o

drivers/md/bcache/movinggc.c

···
 	bio_init(bio, NULL, bio->bi_inline_vecs,
 		 DIV_ROUND_UP(KEY_SIZE(&io->w->key), PAGE_SECTORS), 0);
 	bio_get(bio);
-	bio_set_prio(bio, IOPRIO_PRIO_VALUE(IOPRIO_CLASS_IDLE, 0));
+	bio->bi_ioprio = IOPRIO_PRIO_VALUE(IOPRIO_CLASS_IDLE, 0);
 
 	bio->bi_iter.bi_size	= KEY_SIZE(&io->w->key) << 9;
 	bio->bi_private		= &io->cl;

drivers/md/bcache/writeback.c

···
 	bio_init(bio, NULL, bio->bi_inline_vecs,
 		 DIV_ROUND_UP(KEY_SIZE(&w->key), PAGE_SECTORS), 0);
 	if (!io->dc->writeback_percent)
-		bio_set_prio(bio, IOPRIO_PRIO_VALUE(IOPRIO_CLASS_IDLE, 0));
+		bio->bi_ioprio = IOPRIO_PRIO_VALUE(IOPRIO_CLASS_IDLE, 0);
 
 	bio->bi_iter.bi_size	= KEY_SIZE(&w->key) << 9;
 	bio->bi_private		= w;

drivers/md/dm-rq.c

···
 	md->tag_set->ops = &dm_mq_ops;
 	md->tag_set->queue_depth = dm_get_blk_mq_queue_depth();
 	md->tag_set->numa_node = md->numa_node_id;
-	md->tag_set->flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_STACKING;
+	md->tag_set->flags = BLK_MQ_F_STACKING;
 	md->tag_set->nr_hw_queues = dm_get_blk_mq_nr_hw_queues();
 	md->tag_set->driver_data = md;
 

drivers/md/dm-verity-fec.c

···
 	struct bio *bio = dm_bio_from_per_bio_data(io, v->ti->per_io_data_size);
 
 	par = fec_read_parity(v, rsb, block_offset, &offset,
-			      par_buf_offset, &buf, bio_prio(bio));
+			      par_buf_offset, &buf, bio->bi_ioprio);
 	if (IS_ERR(par))
 		return PTR_ERR(par);
 
···
 			dm_bufio_release(buf);
 
 			par = fec_read_parity(v, rsb, block_offset, &offset,
-					      par_buf_offset, &buf, bio_prio(bio));
+					      par_buf_offset, &buf, bio->bi_ioprio);
 			if (IS_ERR(par))
 				return PTR_ERR(par);
 		}
···
 			bufio = v->bufio;
 		}
 
-		bbuf = dm_bufio_read_with_ioprio(bufio, block, &buf, bio_prio(bio));
+		bbuf = dm_bufio_read_with_ioprio(bufio, block, &buf, bio->bi_ioprio);
 		if (IS_ERR(bbuf)) {
 			DMWARN_LIMIT("%s: FEC %llu: read failed (%llu): %ld",
 				     v->data_dev->name,

drivers/md/dm-verity-target.c

···
 		}
 	} else {
 		data = dm_bufio_read_with_ioprio(v->bufio, hash_block,
-						&buf, bio_prio(bio));
+						&buf, bio->bi_ioprio);
 	}
 
 	if (IS_ERR(data))
···
 
 	verity_fec_init_io(io);
 
-	verity_submit_prefetch(v, io, bio_prio(bio));
+	verity_submit_prefetch(v, io, bio->bi_ioprio);
 
 	submit_bio_noacct(bio);
 

drivers/md/md-autodetect.c

···
  *             instead of just one.  -- KTK
  * 18May2000: Added support for persistent-superblock arrays:
  *             md=n,0,factor,fault,device-list   uses RAID0 for device n
+ *             md=n,-1,factor,fault,device-list  uses LINEAR for device n
  *             md=n,device-list      reads a RAID superblock from the devices
  *             elements in device-list are read by name_to_kdev_t so can be
  *             a hex number or something like /dev/hda1 /dev/sdb
···
 		md_setup_ents++;
 	switch (get_option(&str, &level)) {	/* RAID level */
 	case 2: /* could be 0 or -1.. */
-		if (level == 0) {
+		if (level == 0 || level == LEVEL_LINEAR) {
 			if (get_option(&str, &factor) != 2 ||	/* Chunk Size */
 					get_option(&str, &fault) != 2) {
 				printk(KERN_WARNING "md: Too few arguments supplied to md=.\n");
···
 			}
 			md_setup_args[ent].level = level;
 			md_setup_args[ent].chunk = 1 << (factor+12);
-			pername = "raid0";
+			if (level ==  LEVEL_LINEAR)
+				pername = "linear";
+			else
+				pername = "raid0";
 			break;
 		}
 		fallthrough;

drivers/md/md-bitmap.c

···
 		return;
 	if (!bitmap->storage.sb_page) /* no superblock */
 		return;
-	sb = kmap_atomic(bitmap->storage.sb_page);
+	sb = kmap_local_page(bitmap->storage.sb_page);
 	sb->events = cpu_to_le64(bitmap->mddev->events);
 	if (bitmap->mddev->events < bitmap->events_cleared)
 		/* rocking back to read-only */
···
 	sb->nodes = cpu_to_le32(bitmap->mddev->bitmap_info.nodes);
 	sb->sectors_reserved = cpu_to_le32(bitmap->mddev->
 					   bitmap_info.space);
-	kunmap_atomic(sb);
+	kunmap_local(sb);
 
 	if (bitmap->storage.file)
 		write_file_page(bitmap, bitmap->storage.sb_page, 1);
···
 
 	if (!bitmap || !bitmap->storage.sb_page)
 		return;
-	sb = kmap_atomic(bitmap->storage.sb_page);
+	sb = kmap_local_page(bitmap->storage.sb_page);
 	pr_debug("%s: bitmap file superblock:\n", bmname(bitmap));
 	pr_debug("         magic: %08x\n", le32_to_cpu(sb->magic));
 	pr_debug("       version: %u\n", le32_to_cpu(sb->version));
···
 	pr_debug("     sync size: %llu KB\n",
 		 (unsigned long long)le64_to_cpu(sb->sync_size)/2);
 	pr_debug("max write behind: %u\n", le32_to_cpu(sb->write_behind));
-	kunmap_atomic(sb);
+	kunmap_local(sb);
 }
 
 /*
···
 		return -ENOMEM;
 	bitmap->storage.sb_index = 0;
 
-	sb = kmap_atomic(bitmap->storage.sb_page);
+	sb = kmap_local_page(bitmap->storage.sb_page);
 
 	sb->magic = cpu_to_le32(BITMAP_MAGIC);
 	sb->version = cpu_to_le32(BITMAP_MAJOR_HI);
···
 	chunksize = bitmap->mddev->bitmap_info.chunksize;
 	BUG_ON(!chunksize);
 	if (!is_power_of_2(chunksize)) {
-		kunmap_atomic(sb);
+		kunmap_local(sb);
 		pr_warn("bitmap chunksize not a power of 2\n");
 		return -EINVAL;
 	}
···
 	sb->events_cleared = cpu_to_le64(bitmap->mddev->events);
 	bitmap->mddev->bitmap_info.nodes = 0;
 
-	kunmap_atomic(sb);
+	kunmap_local(sb);
 
 	return 0;
 }
···
 		return err;
 
 	err = -EINVAL;
-	sb = kmap_atomic(sb_page);
+	sb = kmap_local_page(sb_page);
 
 	chunksize = le32_to_cpu(sb->chunksize);
 	daemon_sleep = le32_to_cpu(sb->daemon_sleep) * HZ;
···
 	err = 0;
 
 out:
-	kunmap_atomic(sb);
+	kunmap_local(sb);
 	if (err == 0 && nodes && (bitmap->cluster_slot < 0)) {
 		/* Assigning chunksize is required for "re_read" */
 		bitmap->mddev->bitmap_info.chunksize = chunksize;
···
 	bit = file_page_offset(&bitmap->storage, chunk);
 
 	/* set the bit */
-	kaddr = kmap_atomic(page);
+	kaddr = kmap_local_page(page);
 	if (test_bit(BITMAP_HOSTENDIAN, &bitmap->flags))
 		set_bit(bit, kaddr);
 	else
 		set_bit_le(bit, kaddr);
-	kunmap_atomic(kaddr);
+	kunmap_local(kaddr);
 	pr_debug("set file bit %lu page %lu\n", bit, index);
 	/* record page number so it gets flushed to disk when unplug occurs */
 	set_page_attr(bitmap, index - node_offset, BITMAP_PAGE_DIRTY);
···
 	if (!page)
 		return;
 	bit = file_page_offset(&bitmap->storage, chunk);
-	paddr = kmap_atomic(page);
+	paddr = kmap_local_page(page);
 	if (test_bit(BITMAP_HOSTENDIAN, &bitmap->flags))
 		clear_bit(bit, paddr);
 	else
 		clear_bit_le(bit, paddr);
-	kunmap_atomic(paddr);
+	kunmap_local(paddr);
 	if (!test_page_attr(bitmap, index - node_offset, BITMAP_PAGE_NEEDWRITE)) {
 		set_page_attr(bitmap, index - node_offset, BITMAP_PAGE_PENDING);
 		bitmap->allclean = 0;
···
 	if (!page)
 		return -EINVAL;
 	bit = file_page_offset(&bitmap->storage, chunk);
-	paddr = kmap_atomic(page);
+	paddr = kmap_local_page(page);
 	if (test_bit(BITMAP_HOSTENDIAN, &bitmap->flags))
 		set = test_bit(bit, paddr);
 	else
 		set = test_bit_le(bit, paddr);
-	kunmap_atomic(paddr);
+	kunmap_local(paddr);
 	return set;
 }
 
···
 			 * If the bitmap is out of date, dirty the whole page
 			 * and write it out
 			 */
-			paddr = kmap_atomic(page);
+			paddr = kmap_local_page(page);
 			memset(paddr + offset, 0xff, PAGE_SIZE - offset);
-			kunmap_atomic(paddr);
+			kunmap_local(paddr);
 
 			filemap_write_page(bitmap, i, true);
 			if (test_bit(BITMAP_WRITE_ERROR, &bitmap->flags)) {
···
 		void *paddr;
 		bool was_set;
 
-		paddr = kmap_atomic(page);
+		paddr = kmap_local_page(page);
 		if (test_bit(BITMAP_HOSTENDIAN, &bitmap->flags))
 			was_set = test_bit(bit, paddr);
 		else
 			was_set = test_bit_le(bit, paddr);
-		kunmap_atomic(paddr);
+		kunmap_local(paddr);
 
 		if (was_set) {
 			/* if the disk bit is set, set the memory bit */
···
 		bitmap_super_t *sb;
 		bitmap->need_sync = 0;
 		if (bitmap->storage.filemap) {
-			sb = kmap_atomic(bitmap->storage.sb_page);
+			sb = kmap_local_page(bitmap->storage.sb_page);
 			sb->events_cleared =
 				cpu_to_le64(bitmap->events_cleared);
-			kunmap_atomic(sb);
+			kunmap_local(sb);
 			set_page_attr(bitmap, 0,
 				      BITMAP_PAGE_NEEDWRITE);
 		}
···
 }
 
 static int bitmap_startwrite(struct mddev *mddev, sector_t offset,
-			     unsigned long sectors, bool behind)
+			     unsigned long sectors)
 {
 	struct bitmap *bitmap = mddev->bitmap;
 
 	if (!bitmap)
 		return 0;
-
-	if (behind) {
-		int bw;
-		atomic_inc(&bitmap->behind_writes);
-		bw = atomic_read(&bitmap->behind_writes);
-		if (bw > bitmap->behind_writes_used)
-			bitmap->behind_writes_used = bw;
-
-		pr_debug("inc write-behind count %d/%lu\n",
-			 bw, bitmap->mddev->bitmap_info.max_write_behind);
-	}
 
 	while (sectors) {
 		sector_t blocks;
···
 }
 
 static void bitmap_endwrite(struct mddev *mddev, sector_t offset,
-			    unsigned long sectors, bool success, bool behind)
+			    unsigned long sectors)
 {
 	struct bitmap *bitmap = mddev->bitmap;
 
 	if (!bitmap)
 		return;
-
-	if (behind) {
-		if (atomic_dec_and_test(&bitmap->behind_writes))
-			wake_up(&bitmap->behind_wait);
-		pr_debug("dec write-behind count %d/%lu\n",
-			 atomic_read(&bitmap->behind_writes),
-			 bitmap->mddev->bitmap_info.max_write_behind);
-	}
 
 	while (sectors) {
 		sector_t blocks;
···
 			return;
 		}
 
-		if (success && !bitmap->mddev->degraded &&
-		    bitmap->events_cleared < bitmap->mddev->events) {
-			bitmap->events_cleared = bitmap->mddev->events;
-			bitmap->need_sync = 1;
-			sysfs_notify_dirent_safe(bitmap->sysfs_can_clear);
-		}
-
-		if (!success && !NEEDED(*bmc))
+		if (!bitmap->mddev->degraded) {
+			if (bitmap->events_cleared < bitmap->mddev->events) {
+				bitmap->events_cleared = bitmap->mddev->events;
+				bitmap->need_sync = 1;
+				sysfs_notify_dirent_safe(
+						bitmap->sysfs_can_clear);
+			}
+		} else if (!NEEDED(*bmc)) {
 			*bmc |= NEEDED_MASK;
+		}
 
 		if (COUNTER(*bmc) == COUNTER_MAX)
 			wake_up(&bitmap->overflow_wait);
···
 				kfree(bp[k].map);
 	kfree(bp);
 	kfree(bitmap);
+}
+
+static void bitmap_start_behind_write(struct mddev *mddev)
+{
+	struct bitmap *bitmap = mddev->bitmap;
+	int bw;
+
+	if (!bitmap)
+		return;
+
+	atomic_inc(&bitmap->behind_writes);
+	bw = atomic_read(&bitmap->behind_writes);
+	if (bw > bitmap->behind_writes_used)
+		bitmap->behind_writes_used = bw;
+
+	pr_debug("inc write-behind count %d/%lu\n",
+		 bw, bitmap->mddev->bitmap_info.max_write_behind);
+}
+
+static void bitmap_end_behind_write(struct mddev *mddev)
+{
+	struct bitmap *bitmap = mddev->bitmap;
+
+	if (!bitmap)
+		return;
+
+	if (atomic_dec_and_test(&bitmap->behind_writes))
+		wake_up(&bitmap->behind_wait);
+	pr_debug("dec write-behind count %d/%lu\n",
+		 atomic_read(&bitmap->behind_writes),
+		 bitmap->mddev->bitmap_info.max_write_behind);
 }
 
 static void bitmap_wait_behind_writes(struct mddev *mddev)
···
 	.dirty_bits		= bitmap_dirty_bits,
 	.unplug			= bitmap_unplug,
 	.daemon_work		= bitmap_daemon_work,
+
+	.start_behind_write	= bitmap_start_behind_write,
+	.end_behind_write	= bitmap_end_behind_write,
 	.wait_behind_writes	= bitmap_wait_behind_writes,
 
 	.startwrite		= bitmap_startwrite,

drivers/md/md-bitmap.h

···
 			   unsigned long e);
 	void (*unplug)(struct mddev *mddev, bool sync);
 	void (*daemon_work)(struct mddev *mddev);
+
+	void (*start_behind_write)(struct mddev *mddev);
+	void (*end_behind_write)(struct mddev *mddev);
 	void (*wait_behind_writes)(struct mddev *mddev);
 
 	int (*startwrite)(struct mddev *mddev, sector_t offset,
-			  unsigned long sectors, bool behind);
+			  unsigned long sectors);
 	void (*endwrite)(struct mddev *mddev, sector_t offset,
-			 unsigned long sectors, bool success, bool behind);
+			 unsigned long sectors);
 	bool (*start_sync)(struct mddev *mddev, sector_t offset,
 			   sector_t *blocks, bool degraded);
 	void (*end_sync)(struct mddev *mddev, sector_t offset, sector_t *blocks);

drivers/md/md-linear.c

···
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * linear.c : Multiple Devices driver for Linux Copyright (C) 1994-96 Marc
+ * ZYNGIER <[e-mail]> or <[e-mail]>
+ */
+
+#include <linux/blkdev.h>
+#include <linux/raid/md_u.h>
+#include <linux/seq_file.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <trace/events/block.h>
+#include "md.h"
+
+struct dev_info {
+	struct md_rdev	*rdev;
+	sector_t	end_sector;
+};
+
+struct linear_conf {
+	struct rcu_head         rcu;
+	sector_t                array_sectors;
+	/* a copy of mddev->raid_disks */
+	int                     raid_disks;
+	struct dev_info         disks[] __counted_by(raid_disks);
+};
+
+/*
+ * find which device holds a particular offset
+ */
+static inline struct dev_info *which_dev(struct mddev *mddev, sector_t sector)
+{
+	int lo, mid, hi;
+	struct linear_conf *conf;
+
+	lo = 0;
+	hi = mddev->raid_disks - 1;
+	conf = mddev->private;
+
+	/*
+	 * Binary Search
+	 */
+
+	while (hi > lo) {
+
+		mid = (hi + lo) / 2;
+		if (sector < conf->disks[mid].end_sector)
+			hi = mid;
+		else
+			lo = mid + 1;
+	}
+
+	return conf->disks + lo;
+}
+
+static sector_t linear_size(struct mddev *mddev, sector_t sectors, int raid_disks)
+{
+	struct linear_conf *conf;
+	sector_t array_sectors;
+
+	conf = mddev->private;
+	WARN_ONCE(sectors || raid_disks,
+		  "%s does not support generic reshape\n", __func__);
+	array_sectors = conf->array_sectors;
+
+	return array_sectors;
+}
+
+static int linear_set_limits(struct mddev *mddev)
+{
+	struct queue_limits lim;
+	int err;
+
+	md_init_stacking_limits(&lim);
+	lim.max_hw_sectors = mddev->chunk_sectors;
+	lim.max_write_zeroes_sectors = mddev->chunk_sectors;
+	lim.io_min = mddev->chunk_sectors << 9;
+	err = mddev_stack_rdev_limits(mddev, &lim, MDDEV_STACK_INTEGRITY);
+	if (err) {
+		queue_limits_cancel_update(mddev->gendisk->queue);
+		return err;
+	}
+
+	return queue_limits_set(mddev->gendisk->queue, &lim);
+}
+
+static struct linear_conf *linear_conf(struct mddev *mddev, int raid_disks)
+{
+	struct linear_conf *conf;
+	struct md_rdev *rdev;
+	int ret = -EINVAL;
+	int cnt;
+	int i;
+
+	conf = kzalloc(struct_size(conf, disks, raid_disks), GFP_KERNEL);
+	if (!conf)
+		return ERR_PTR(-ENOMEM);
+
+	/*
+	 * conf->raid_disks is copy of mddev->raid_disks. The reason to
+	 * keep a copy of mddev->raid_disks in struct linear_conf is,
+	 * mddev->raid_disks may not be consistent with pointers number of
+	 * conf->disks[] when it is updated in linear_add() and used to
+	 * iterate old conf->disks[] earray in linear_congested().
+	 * Here conf->raid_disks is always consitent with number of
+	 * pointers in conf->disks[] array, and mddev->private is updated
+	 * with rcu_assign_pointer() in linear_addr(), such race can be
+	 * avoided.
+	 */
+	conf->raid_disks = raid_disks;
+
+	cnt = 0;
+	conf->array_sectors = 0;
+
+	rdev_for_each(rdev, mddev) {
+		int j = rdev->raid_disk;
+		struct dev_info *disk = conf->disks + j;
+		sector_t sectors;
+
+		if (j < 0 || j >= raid_disks || disk->rdev) {
+			pr_warn("md/linear:%s: disk numbering problem. Aborting!\n",
+				mdname(mddev));
+			goto out;
+		}
+
+		disk->rdev = rdev;
+		if (mddev->chunk_sectors) {
+			sectors = rdev->sectors;
+			sector_div(sectors, mddev->chunk_sectors);
+			rdev->sectors = sectors * mddev->chunk_sectors;
+		}
+
+		conf->array_sectors += rdev->sectors;
+		cnt++;
+	}
+	if (cnt != raid_disks) {
+		pr_warn("md/linear:%s: not enough drives present. Aborting!\n",
+			mdname(mddev));
+		goto out;
+	}
+
+	/*
+	 * Here we calculate the device offsets.
+	 */
+	conf->disks[0].end_sector = conf->disks[0].rdev->sectors;
+
+	for (i = 1; i < raid_disks; i++)
+		conf->disks[i].end_sector =
+			conf->disks[i-1].end_sector +
+			conf->disks[i].rdev->sectors;
+
+	if (!mddev_is_dm(mddev)) {
+		ret = linear_set_limits(mddev);
+		if (ret)
+			goto out;
+	}
+
+	return conf;
+
+out:
+	kfree(conf);
+	return ERR_PTR(ret);
+}
+
+static int linear_run(struct mddev *mddev)
+{
+	struct linear_conf *conf;
+	int ret;
+
+	if (md_check_no_bitmap(mddev))
+		return -EINVAL;
+
+	conf = linear_conf(mddev, mddev->raid_disks);
+	if (IS_ERR(conf))
+		return PTR_ERR(conf);
+
+	mddev->private = conf;
+	md_set_array_sectors(mddev, linear_size(mddev, 0, 0));
+
+	ret =  md_integrity_register(mddev);
+	if (ret) {
+		kfree(conf);
+		mddev->private = NULL;
+	}
+	return ret;
+}
+
+static int linear_add(struct mddev *mddev, struct md_rdev *rdev)
+{
+	/* Adding a drive to a linear array allows the array to grow.
+	 * It is permitted if the new drive has a matching superblock
+	 * already on it, with raid_disk equal to raid_disks.
+	 * It is achieved by creating a new linear_private_data structure
+	 * and swapping it in in-place of the current one.
+	 * The current one is never freed until the array is stopped.
+	 * This avoids races.
+	 */
+	struct linear_conf *newconf, *oldconf;
+
+	if (rdev->saved_raid_disk != mddev->raid_disks)
+		return -EINVAL;
+
+	rdev->raid_disk = rdev->saved_raid_disk;
+	rdev->saved_raid_disk = -1;
+
+	newconf = linear_conf(mddev, mddev->raid_disks + 1);
+	if (IS_ERR(newconf))
+		return PTR_ERR(newconf);
+
+	/* newconf->raid_disks already keeps a copy of * the increased
+	 * value of mddev->raid_disks, WARN_ONCE() is just used to make
+	 * sure of this. It is possible that oldconf is still referenced
+	 * in linear_congested(), therefore kfree_rcu() is used to free
+	 * oldconf until no one uses it anymore.
+	 */
+	oldconf = rcu_dereference_protected(mddev->private,
+			lockdep_is_held(&mddev->reconfig_mutex));
+	mddev->raid_disks++;
+	WARN_ONCE(mddev->raid_disks != newconf->raid_disks,
+		"copied raid_disks doesn't match mddev->raid_disks");
+	rcu_assign_pointer(mddev->private, newconf);
+	md_set_array_sectors(mddev, linear_size(mddev, 0, 0));
+	set_capacity_and_notify(mddev->gendisk, mddev->array_sectors);
+	kfree_rcu(oldconf, rcu);
+	return 0;
+}
+
+static void linear_free(struct mddev *mddev, void *priv)
+{
+	struct linear_conf *conf = priv;
+
+	kfree(conf);
+}
+
+static bool linear_make_request(struct mddev *mddev, struct bio *bio)
+{
+	struct dev_info *tmp_dev;
+	sector_t start_sector, end_sector, data_offset;
+	sector_t bio_sector = bio->bi_iter.bi_sector;
+
+	if (unlikely(bio->bi_opf & REQ_PREFLUSH)
+	    && md_flush_request(mddev, bio))
+		return true;
+
+	tmp_dev = which_dev(mddev, bio_sector);
+	start_sector = tmp_dev->end_sector - tmp_dev->rdev->sectors;
+	end_sector = tmp_dev->end_sector;
+	data_offset = tmp_dev->rdev->data_offset;
+
+	if (unlikely(bio_sector >= end_sector ||
+		     bio_sector < start_sector))
+		goto out_of_bounds;
+
+	if (unlikely(is_rdev_broken(tmp_dev->rdev))) {
+		md_error(mddev, tmp_dev->rdev);
+		bio_io_error(bio);
+		return true;
+	}
+
+	if (unlikely(bio_end_sector(bio) > end_sector)) {
+		/* This bio crosses a device boundary, so we have to split it */
+		struct bio *split = bio_split(bio, end_sector - bio_sector,
+					      GFP_NOIO, &mddev->bio_set);
+
+		if (IS_ERR(split)) {
+			bio->bi_status = errno_to_blk_status(PTR_ERR(split));
+			bio_endio(bio);
+			return true;
+		}
+
+		bio_chain(split, bio);
+		submit_bio_noacct(bio);
+		bio = split;
+	}
+
+	md_account_bio(mddev, &bio);
+	bio_set_dev(bio, tmp_dev->rdev->bdev);
+	bio->bi_iter.bi_sector = bio->bi_iter.bi_sector -
+		start_sector + data_offset;
+
+	if (unlikely((bio_op(bio) == REQ_OP_DISCARD) &&
+		     !bdev_max_discard_sectors(bio->bi_bdev))) {
+		/* Just ignore it */
+		bio_endio(bio);
+	} else {
+		if (mddev->gendisk)
+			trace_block_bio_remap(bio, disk_devt(mddev->gendisk),
+					      bio_sector);
+		mddev_check_write_zeroes(mddev, bio);
+		submit_bio_noacct(bio);
+	}
+	return true;
+
+out_of_bounds:
+	pr_err("md/linear:%s: make_request: Sector %llu out of bounds on dev %pg: %llu sectors, offset %llu\n",
+	       mdname(mddev),
+	       (unsigned long long)bio->bi_iter.bi_sector,
+	       tmp_dev->rdev->bdev,
+	       (unsigned long long)tmp_dev->rdev->sectors,
+	       (unsigned long long)start_sector);
+	bio_io_error(bio);
+	return true;
+}
+
+static void linear_status(struct seq_file *seq, struct mddev *mddev)
+{
+	seq_printf(seq, " %dk rounding", mddev->chunk_sectors / 2);
+}
+
+static void linear_error(struct mddev *mddev, struct md_rdev *rdev)
+{
+	if (!test_and_set_bit(MD_BROKEN, &mddev->flags)) {
+		char *md_name = mdname(mddev);
+
+		pr_crit("md/linear%s: Disk failure on %pg detected, failing array.\n",
+			md_name, rdev->bdev);
+	}
+}
+
+static void linear_quiesce(struct mddev *mddev, int state)
+{
+}
+
+static struct md_personality linear_personality = {
+	.name		= "linear",
+	.level		= LEVEL_LINEAR,
+	.owner		= THIS_MODULE,
+	.make_request	= linear_make_request,
+	.run		= linear_run,
+	.free		= linear_free,
+	.status		= linear_status,
+	.hot_add_disk	= linear_add,
+	.size		= linear_size,
+	.quiesce	= linear_quiesce,
+	.error_handler	= linear_error,
+};
+
+static int __init linear_init(void)
+{
+	return register_md_personality(&linear_personality);
+}
+
+static void linear_exit(void)
+{
+	unregister_md_personality(&linear_personality);
+}
+
+module_init(linear_init);
+module_exit(linear_exit);
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Linear device concatenation personality for MD (deprecated)");
+MODULE_ALIAS("md-personality-1"); /* LINEAR - deprecated*/
+MODULE_ALIAS("md-linear");
+MODULE_ALIAS("md-level--1");

drivers/md/md.c

···
 		return;
 	mddev->pers->error_handler(mddev, rdev);
 
-	if (mddev->pers->level == 0)
+	if (mddev->pers->level == 0 || mddev->pers->level == LEVEL_LINEAR)
 		return;
 
 	if (mddev->degraded && !test_bit(MD_BROKEN, &mddev->flags))
···
 }
 EXPORT_SYMBOL_GPL(md_submit_discard_bio);
 
+static void md_bitmap_start(struct mddev *mddev,
+			    struct md_io_clone *md_io_clone)
+{
+	if (mddev->pers->bitmap_sector)
+		mddev->pers->bitmap_sector(mddev, &md_io_clone->offset,
+					   &md_io_clone->sectors);
+
+	mddev->bitmap_ops->startwrite(mddev, md_io_clone->offset,
+				      md_io_clone->sectors);
+}
+
+static void md_bitmap_end(struct mddev *mddev, struct md_io_clone *md_io_clone)
+{
+	mddev->bitmap_ops->endwrite(mddev, md_io_clone->offset,
+				    md_io_clone->sectors);
+}
+
 static void md_end_clone_io(struct bio *bio)
 {
 	struct md_io_clone *md_io_clone = bio->bi_private;
 	struct bio *orig_bio = md_io_clone->orig_bio;
 	struct mddev *mddev = md_io_clone->mddev;
+
+	if (bio_data_dir(orig_bio) == WRITE && mddev->bitmap)
+		md_bitmap_end(mddev, md_io_clone);
 
 	if (bio->bi_status && !orig_bio->bi_status)
 		orig_bio->bi_status = bio->bi_status;
···
 	if (blk_queue_io_stat(bdev->bd_disk->queue))
 		md_io_clone->start_time = bio_start_io_acct(*bio);
 
+	if (bio_data_dir(*bio) == WRITE && mddev->bitmap) {
+		md_io_clone->offset = (*bio)->bi_iter.bi_sector;
+		md_io_clone->sectors = bio_sectors(*bio);
+		md_bitmap_start(mddev, md_io_clone);
+	}
+
 	clone->bi_end_io = md_end_clone_io;
 	clone->bi_private = md_io_clone;
 	*bio = clone;
···
 	struct md_io_clone *md_io_clone = bio->bi_private;
 	struct bio *orig_bio = md_io_clone->orig_bio;
 	struct mddev *mddev = md_io_clone->mddev;
+
+	if (bio_data_dir(orig_bio) == WRITE && mddev->bitmap)
+		md_bitmap_end(mddev, md_io_clone);
 
 	if (bio->bi_status && !orig_bio->bi_status)
 		orig_bio->bi_status = bio->bi_status;

drivers/md/md.h

···
 	void *(*takeover) (struct mddev *mddev);
 	/* Changes the consistency policy of an active array. */
 	int (*change_consistency_policy)(struct mddev *mddev, const char *buf);
+	/* convert io ranges from array to bitmap */
+	void (*bitmap_sector)(struct mddev *mddev, sector_t *offset,
+			      unsigned long *sectors);
 };
 
 struct md_sysfs_entry {
···
 	struct mddev	*mddev;
 	struct bio	*orig_bio;
 	unsigned long	start_time;
+	sector_t	offset;
+	unsigned long	sectors;
 	struct bio	bio_clone;
 };
 

drivers/md/raid0.c

···
 	lim.max_write_zeroes_sectors = mddev->chunk_sectors;
 	lim.io_min = mddev->chunk_sectors << 9;
 	lim.io_opt = lim.io_min * mddev->raid_disks;
-	lim.features |= BLK_FEAT_ATOMIC_WRITES_STACKED;
+	lim.features |= BLK_FEAT_ATOMIC_WRITES;
 	err = mddev_stack_rdev_limits(mddev, &lim, MDDEV_STACK_INTEGRITY);
 	if (err) {
 		queue_limits_cancel_update(mddev->gendisk->queue);

drivers/md/raid1.c

···
 		r1_bio->behind_master_bio = NULL;
 	}
 
-	/* clear the bitmap if all writes complete successfully */
-	mddev->bitmap_ops->endwrite(mddev, r1_bio->sector, r1_bio->sectors,
-				    !test_bit(R1BIO_Degraded, &r1_bio->state),
-				    test_bit(R1BIO_BehindIO, &r1_bio->state));
+	if (test_bit(R1BIO_BehindIO, &r1_bio->state))
+		mddev->bitmap_ops->end_behind_write(mddev);
 	md_write_end(mddev);
 }
 
···
 		if (!test_bit(Faulty, &rdev->flags))
 			set_bit(R1BIO_WriteError, &r1_bio->state);
 		else {
-			/* Fail the request */
-			set_bit(R1BIO_Degraded, &r1_bio->state);
 			/* Finished with this branch */
 			r1_bio->bios[mirror] = NULL;
 			to_put = bio;
···
 			write_behind = true;
 
 		r1_bio->bios[i] = NULL;
-		if (!rdev || test_bit(Faulty, &rdev->flags)) {
-			if (i < conf->raid_disks)
-				set_bit(R1BIO_Degraded, &r1_bio->state);
+		if (!rdev || test_bit(Faulty, &rdev->flags))
 			continue;
-		}
 
 		atomic_inc(&rdev->nr_pending);
 		if (test_bit(WriteErrorSeen, &rdev->flags)) {
···
 					 */
 					max_sectors = bad_sectors;
 				rdev_dec_pending(rdev, mddev);
-				/* We don't set R1BIO_Degraded as that
-				 * only applies if the disk is
-				 * missing, so it might be re-added,
-				 * and we want to know to recover this
-				 * chunk.
-				 * In this case the device is here,
-				 * and the fact that this chunk is not
-				 * in-sync is recorded in the bad
-				 * block log
-				 */
 				continue;
 			}
 			if (is_bad) {
···
 			    stats.behind_writes < max_write_behind)
 				alloc_behind_master_bio(r1_bio, bio);
 
-			mddev->bitmap_ops->startwrite(
-				mddev, r1_bio->sector, r1_bio->sectors,
-				test_bit(R1BIO_BehindIO, &r1_bio->state));
+			if (test_bit(R1BIO_BehindIO, &r1_bio->state))
+				mddev->bitmap_ops->start_behind_write(mddev);
 			first_clone = 0;
 		}
 
···
 			 * errors.
 			 */
 			fail = true;
-			if (!narrow_write_error(r1_bio, m)) {
+			if (!narrow_write_error(r1_bio, m))
 				md_error(conf->mddev,
 					 conf->mirrors[m].rdev);
 				/* an I/O failed, we can't clear the bitmap */
-				set_bit(R1BIO_Degraded, &r1_bio->state);
-			}
 			rdev_dec_pending(conf->mirrors[m].rdev,
 					 conf->mddev);
 		}
···
 			list_del(&r1_bio->retry_list);
 			idx = sector_to_idx(r1_bio->sector);
 			atomic_dec(&conf->nr_queued[idx]);
-			if (mddev->degraded)
-				set_bit(R1BIO_Degraded, &r1_bio->state);
 			if (test_bit(R1BIO_WriteError, &r1_bio->state))
 				close_write(r1_bio);
 			raid_end_bio_io(r1_bio);
···
 
 	md_init_stacking_limits(&lim);
 	lim.max_write_zeroes_sectors = 0;
-	lim.features |= BLK_FEAT_ATOMIC_WRITES_STACKED;
+	lim.features |= BLK_FEAT_ATOMIC_WRITES;
 	err = mddev_stack_rdev_limits(mddev, &lim, MDDEV_STACK_INTEGRITY);
 	if (err) {
 		queue_limits_cancel_update(mddev->gendisk->queue);

drivers/md/raid1.h

···
 enum r1bio_state {
 	R1BIO_Uptodate,
 	R1BIO_IsSync,
-	R1BIO_Degraded,
 	R1BIO_BehindIO,
 /* Set ReadError on bios that experience a readerror so that
  * raid1d knows what to do with them.

drivers/md/raid10.c

···
 {
 	struct mddev *mddev = r10_bio->mddev;
 
-	/* clear the bitmap if all writes complete successfully */
-	mddev->bitmap_ops->endwrite(mddev, r10_bio->sector, r10_bio->sectors,
-				    !test_bit(R10BIO_Degraded, &r10_bio->state),
-				    false);
 	md_write_end(mddev);
 }
 
···
 				set_bit(R10BIO_WriteError, &r10_bio->state);
 			else {
 				/* Fail the request */
-				set_bit(R10BIO_Degraded, &r10_bio->state);
 				r10_bio->devs[slot].bio = NULL;
 				to_put = bio;
 				dec_rdev = 1;
···
 		r10_bio->devs[i].bio = NULL;
 		r10_bio->devs[i].repl_bio = NULL;
 
-		if (!rdev && !rrdev) {
-			set_bit(R10BIO_Degraded, &r10_bio->state);
+		if (!rdev && !rrdev)
 			continue;
-		}
 		if (rdev && test_bit(WriteErrorSeen, &rdev->flags)) {
 			sector_t first_bad;
 			sector_t dev_sector = r10_bio->devs[i].addr;
···
 					 * to other devices yet
 					 */
 					max_sectors = bad_sectors;
-				/* We don't set R10BIO_Degraded as that
-				 * only applies if the disk is missing,
-				 * so it might be re-added, and we want to
-				 * know to recover this chunk.
-				 * In this case the device is here, and the
-				 * fact that this chunk is not in-sync is
-				 * recorded in the bad block log.
-				 */
 				continue;
 			}
 			if (is_bad) {
···
 	md_account_bio(mddev, &bio);
 	r10_bio->master_bio = bio;
 	atomic_set(&r10_bio->remaining, 1);
-	mddev->bitmap_ops->startwrite(mddev, r10_bio->sector, r10_bio->sectors,
-				      false);
 
 	for (i = 0; i < conf->copies; i++) {
 		if (r10_bio->devs[i].bio)
···
 				rdev_dec_pending(rdev, conf->mddev);
 			} else if (bio != NULL && bio->bi_status) {
 				fail = true;
-				if (!narrow_write_error(r10_bio, m)) {
+				if (!narrow_write_error(r10_bio, m))
 					md_error(conf->mddev, rdev);
-					set_bit(R10BIO_Degraded,
-						&r10_bio->state);
-				}
 				rdev_dec_pending(rdev, conf->mddev);
 			}
 			bio = r10_bio->devs[m].repl_bio;
···
 			r10_bio = list_first_entry(&tmp, struct r10bio,
 						   retry_list);
 			list_del(&r10_bio->retry_list);
-			if (mddev->degraded)
-				set_bit(R10BIO_Degraded, &r10_bio->state);
 
 			if (test_bit(R10BIO_WriteError,
 				     &r10_bio->state))
···
 	lim.max_write_zeroes_sectors = 0;
 	lim.io_min = mddev->chunk_sectors << 9;
 	lim.io_opt = lim.io_min * raid10_nr_stripes(conf);
-	lim.features |= BLK_FEAT_ATOMIC_WRITES_STACKED;
+	lim.features |= BLK_FEAT_ATOMIC_WRITES;
 	err = mddev_stack_rdev_limits(mddev, &lim, MDDEV_STACK_INTEGRITY);
 	if (err) {
 		queue_limits_cancel_update(mddev->gendisk->queue);

drivers/md/raid10.h

···
 	R10BIO_IsSync,
 	R10BIO_IsRecover,
 	R10BIO_IsReshape,
-	R10BIO_Degraded,
 /* Set ReadError on bios that experience a read error
  * so that raid10d knows what to do with them.
  */

drivers/md/raid5-cache.c

···
 		if (sh->dev[i].written) {
 			set_bit(R5_UPTODATE, &sh->dev[i].flags);
 			r5c_return_dev_pending_writes(conf, &sh->dev[i]);
-			conf->mddev->bitmap_ops->endwrite(conf->mddev,
-					sh->sector, RAID5_STRIPE_SECTORS(conf),
-					!test_bit(STRIPE_DEGRADED, &sh->state),
-					false);
 		}
 	}
 }
···
 		/* checksum is already calculated in last run */
 		if (test_bit(STRIPE_LOG_TRAPPED, &sh->state))
 			continue;
-		addr = kmap_atomic(sh->dev[i].page);
+		addr = kmap_local_page(sh->dev[i].page);
 		sh->dev[i].log_checksum = crc32c_le(log->uuid_checksum,
 						    addr, PAGE_SIZE);
-		kunmap_atomic(addr);
+		kunmap_local(addr);
 	}
 	parity_pages = 1 + !!(sh->qd_idx >= 0);
 	data_pages = write_disks - parity_pages;
···
 	u32 checksum;
 
 	r5l_recovery_read_page(log, ctx, page, log_offset);
-	addr = kmap_atomic(page);
+	addr = kmap_local_page(page);
 	checksum = crc32c_le(log->uuid_checksum, addr, PAGE_SIZE);
-	kunmap_atomic(addr);
+	kunmap_local(addr);
 	return (le32_to_cpu(log_checksum) == checksum) ? 0 : -EINVAL;
 }
 
···
 				payload->size = cpu_to_le32(BLOCK_SECTORS);
 				payload->location = cpu_to_le64(
 					raid5_compute_blocknr(sh, i, 0));
-				addr = kmap_atomic(dev->page);
+				addr = kmap_local_page(dev->page);
 				payload->checksum[0] = cpu_to_le32(
 					crc32c_le(log->uuid_checksum, addr,
 						  PAGE_SIZE));
-				kunmap_atomic(addr);
+				kunmap_local(addr);
 				sync_page_io(log->rdev, write_pos, PAGE_SIZE,
 					     dev->page, REQ_OP_WRITE, false);
 				write_pos = r5l_ring_add(log, write_pos,
···
 
 		if (!test_bit(R5_Wantwrite, &sh->dev[i].flags))
 			continue;
-		addr = kmap_atomic(sh->dev[i].page);
+		addr = kmap_local_page(sh->dev[i].page);
 		sh->dev[i].log_checksum = crc32c_le(log->uuid_checksum,
 						    addr, PAGE_SIZE);
-		kunmap_atomic(addr);
+		kunmap_local(addr);
 		pages++;
 	}
 	WARN_ON(pages == 0);

drivers/md/raid5.c

···
 	if (raid5_has_log(conf) || raid5_has_ppl(conf))
 		return false;
 	return test_bit(STRIPE_BATCH_READY, &sh->state) &&
-		!test_bit(STRIPE_BITMAP_PENDING, &sh->state) &&
-		is_full_stripe_write(sh);
+	       is_full_stripe_write(sh);
 }
 
 /* we only do back search */
···
 				submit_bio_noacct(rbi);
 		}
 		if (!rdev && !rrdev) {
-			if (op_is_write(op))
-				set_bit(STRIPE_DEGRADED, &sh->state);
 			pr_debug("skip op %d on disc %d for sector %llu\n",
 				bi->bi_opf, i, (unsigned long long)sh->sector);
 			clear_bit(R5_LOCKED, &sh->dev[i].flags);
···
 			set_bit(R5_MadeGoodRepl, &sh->dev[i].flags);
 	} else {
 		if (bi->bi_status) {
-			set_bit(STRIPE_DEGRADED, &sh->state);
 			set_bit(WriteErrorSeen, &rdev->flags);
 			set_bit(R5_WriteError, &sh->dev[i].flags);
 			if (!test_and_set_bit(WantReplacement, &rdev->flags))
···
 		 (*bip)->bi_iter.bi_sector, sh->sector, dd_idx,
 		 sh->dev[dd_idx].sector);
 
-	if (conf->mddev->bitmap && firstwrite) {
-		/* Cannot hold spinlock over bitmap_startwrite,
-		 * but must ensure this isn't added to a batch until
-		 * we have added to the bitmap and set bm_seq.
-		 * So set STRIPE_BITMAP_PENDING to prevent
-		 * batching.
-		 * If multiple __add_stripe_bio() calls race here they
-		 * much all set STRIPE_BITMAP_PENDING.  So only the first one
-		 * to complete "bitmap_startwrite" gets to set
-		 * STRIPE_BIT_DELAY.  This is important as once a stripe
-		 * is added to a batch, STRIPE_BIT_DELAY cannot be changed
-		 * any more.
-		 */
-		set_bit(STRIPE_BITMAP_PENDING, &sh->state);
-		spin_unlock_irq(&sh->stripe_lock);
-		conf->mddev->bitmap_ops->startwrite(conf->mddev, sh->sector,
-					RAID5_STRIPE_SECTORS(conf), false);
-		spin_lock_irq(&sh->stripe_lock);
-		clear_bit(STRIPE_BITMAP_PENDING, &sh->state);
-		if (!sh->batch_head) {
-			sh->bm_seq = conf->seq_flush+1;
-			set_bit(STRIPE_BIT_DELAY, &sh->state);
-		}
+	if (conf->mddev->bitmap && firstwrite && !sh->batch_head) {
+		sh->bm_seq = conf->seq_flush+1;
+		set_bit(STRIPE_BIT_DELAY, &sh->state);
 	}
 }
 
···
 	BUG_ON(sh->batch_head);
 	for (i = disks; i--; ) {
 		struct bio *bi;
-		int bitmap_end = 0;
 
 		if (test_bit(R5_ReadError, &sh->dev[i].flags)) {
 			struct md_rdev *rdev = conf->disks[i].rdev;
···
 		sh->dev[i].towrite = NULL;
 		sh->overwrite_disks = 0;
 		spin_unlock_irq(&sh->stripe_lock);
-		if (bi)
-			bitmap_end = 1;
 
 		log_stripe_write_finished(sh);
 
···
 			bio_io_error(bi);
 			bi = nextbi;
 		}
-		if (bitmap_end)
-			conf->mddev->bitmap_ops->endwrite(conf->mddev,
-					sh->sector, RAID5_STRIPE_SECTORS(conf),
-					false, false);
-		bitmap_end = 0;
 		/* and fail all 'written' */
 		bi = sh->dev[i].written;
 		sh->dev[i].written = NULL;
···
 			sh->dev[i].page = sh->dev[i].orig_page;
 		}
 
-		if (bi) bitmap_end = 1;
 		while (bi && bi->bi_iter.bi_sector <
 		       sh->dev[i].sector + RAID5_STRIPE_SECTORS(conf)) {
 			struct bio *bi2 = r5_next_bio(conf, bi, sh->dev[i].sector);
···
 				bi = nextbi;
 			}
 		}
-		if (bitmap_end)
-			conf->mddev->bitmap_ops->endwrite(conf->mddev,
-					sh->sector, RAID5_STRIPE_SECTORS(conf),
-					false, false);
 		/* If we were in the middle of a write the parity block might
 		 * still be locked - so just clear all R5_LOCKED flags
 		 */
···
 					bio_endio(wbi);
 					wbi = wbi2;
 				}
-				conf->mddev->bitmap_ops->endwrite(conf->mddev,
-					sh->sector, RAID5_STRIPE_SECTORS(conf),
-					!test_bit(STRIPE_DEGRADED, &sh->state),
-					false);
+
 				if (head_sh->batch_head) {
 					sh = list_first_entry(&sh->batch_list,
 							      struct stripe_head,
···
 		s->locked++;
 		set_bit(R5_Wantwrite, &dev->flags);
 
-		clear_bit(STRIPE_DEGRADED, &sh->state);
 		set_bit(STRIPE_INSYNC, &sh->state);
 		break;
 	case check_state_run:
···
 			clear_bit(R5_Wantwrite, &dev->flags);
 			s->locked--;
 		}
-		clear_bit(STRIPE_DEGRADED, &sh->state);
 
 		set_bit(STRIPE_INSYNC, &sh->state);
 		break;
···
 					  (1 << STRIPE_COMPUTE_RUN)  |
 					  (1 << STRIPE_DISCARD) |
 					  (1 << STRIPE_BATCH_READY) |
-					  (1 << STRIPE_BATCH_ERR) |
-					  (1 << STRIPE_BITMAP_PENDING)),
+					  (1 << STRIPE_BATCH_ERR)),
 			"stripe state: %lx\n", sh->state);
 		WARN_ONCE(head_sh->state & ((1 << STRIPE_DISCARD) |
 					      (1 << STRIPE_REPLACED)),
···
 
 		set_mask_bits(&sh->state, ~(STRIPE_EXPAND_SYNC_FLAGS |
 					    (1 << STRIPE_PREREAD_ACTIVE) |
-					    (1 << STRIPE_DEGRADED) |
 					    (1 << STRIPE_ON_UNPLUG_LIST)),
 			      head_sh->state & (1 << STRIPE_INSYNC));
 
···
 		}
 		spin_unlock_irq(&sh->stripe_lock);
 		if (conf->mddev->bitmap) {
-			for (d = 0; d < conf->raid_disks - conf->max_degraded;
-			     d++)
-				mddev->bitmap_ops->startwrite(mddev, sh->sector,
-					RAID5_STRIPE_SECTORS(conf), false);
 			sh->bm_seq = conf->seq_flush + 1;
 			set_bit(STRIPE_BIT_DELAY, &sh->state);
 		}
···
 	if (ahead_of_reshape(mddev, logical_sector, reshape_safe))
 		return LOC_INSIDE_RESHAPE;
 	return LOC_BEHIND_RESHAPE;
+}
+
+static void raid5_bitmap_sector(struct mddev *mddev, sector_t *offset,
+				unsigned long *sectors)
+{
+	struct r5conf *conf = mddev->private;
+	sector_t start = *offset;
+	sector_t end = start + *sectors;
+	sector_t prev_start = start;
+	sector_t prev_end = end;
+	int sectors_per_chunk;
+	enum reshape_loc loc;
+	int dd_idx;
+
+	sectors_per_chunk = conf->chunk_sectors *
+		(conf->raid_disks - conf->max_degraded);
+	start = round_down(start, sectors_per_chunk);
+	end = round_up(end, sectors_per_chunk);
+
+	start = raid5_compute_sector(conf, start, 0, &dd_idx, NULL);
+	end = raid5_compute_sector(conf, end, 0, &dd_idx, NULL);
+
+	/*
+	 * For LOC_INSIDE_RESHAPE, this IO will wait for reshape to make
+	 * progress, hence it's the same as LOC_BEHIND_RESHAPE.
+	 */
+	loc = get_reshape_loc(mddev, conf, prev_start);
+	if (likely(loc != LOC_AHEAD_OF_RESHAPE)) {
+		*offset = start;
+		*sectors = end - start;
+		return;
+	}
+
+	sectors_per_chunk = conf->prev_chunk_sectors *
+		(conf->previous_raid_disks - conf->max_degraded);
+	prev_start = round_down(prev_start, sectors_per_chunk);
+	prev_end = round_down(prev_end, sectors_per_chunk);
+
+	prev_start = raid5_compute_sector(conf, prev_start, 1, &dd_idx, NULL);
+	prev_end = raid5_compute_sector(conf, prev_end, 1, &dd_idx, NULL);
+
+	/*
+	 * for LOC_AHEAD_OF_RESHAPE, reshape can make progress before this IO
+	 * is handled in make_stripe_request(), we can't know this here hence
+	 * we set bits for both.
+	 */
+	*offset = min(start, prev_start);
+	*sectors = max(end, prev_end) - *offset;
 }
 
 static enum stripe_result make_stripe_request(struct mddev *mddev,
···
 	.takeover	= raid6_takeover,
 	.change_consistency_policy = raid5_change_consistency_policy,
 	.prepare_suspend = raid5_prepare_suspend,
+	.bitmap_sector	= raid5_bitmap_sector,
 };
 static struct md_personality raid5_personality =
 {
···
 	.takeover	= raid5_takeover,
 	.change_consistency_policy = raid5_change_consistency_policy,
 	.prepare_suspend = raid5_prepare_suspend,
+	.bitmap_sector	= raid5_bitmap_sector,
 };
 
 static struct md_personality raid4_personality =
···
 	.takeover	= raid4_takeover,
 	.change_consistency_policy = raid5_change_consistency_policy,
 	.prepare_suspend = raid5_prepare_suspend,
+	.bitmap_sector	= raid5_bitmap_sector,
 };
 
 static int __init raid5_init(void)

drivers/md/raid5.h

···
 	STRIPE_REPLACED,
 	STRIPE_PREREAD_ACTIVE,
 	STRIPE_DELAYED,
-	STRIPE_DEGRADED,
 	STRIPE_BIT_DELAY,
 	STRIPE_EXPANDING,
 	STRIPE_EXPAND_SOURCE,
···
 	STRIPE_ON_RELEASE_LIST,
 	STRIPE_BATCH_READY,
 	STRIPE_BATCH_ERR,
-	STRIPE_BITMAP_PENDING,	/* Being added to bitmap, don't add
-				 * to batch yet.
-				 */
 	STRIPE_LOG_TRAPPED,	/* trapped into log (see raid5-cache.c)
 				 * this bit is used in two scenarios:
 				 *

drivers/memstick/core/ms_block.c

···
 	if (msb->disk_id  < 0)
 		return msb->disk_id;
 
-	rc = blk_mq_alloc_sq_tag_set(&msb->tag_set, &msb_mq_ops, 2,
-				     BLK_MQ_F_SHOULD_MERGE);
+	rc = blk_mq_alloc_sq_tag_set(&msb->tag_set, &msb_mq_ops, 2, 0);
 	if (rc)
 		goto out_release_id;
 

drivers/memstick/core/mspro_block.c

···
 	if (disk_id < 0)
 		return disk_id;
 
-	rc = blk_mq_alloc_sq_tag_set(&msb->tag_set, &mspro_mq_ops, 2,
-				     BLK_MQ_F_SHOULD_MERGE);
+	rc = blk_mq_alloc_sq_tag_set(&msb->tag_set, &mspro_mq_ops, 2, 0);
 	if (rc)
 		goto out_release_id;
 

drivers/mmc/core/queue.c

···
 	else
 		mq->tag_set.queue_depth = MMC_QUEUE_DEPTH;
 	mq->tag_set.numa_node = NUMA_NO_NODE;
-	mq->tag_set.flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_BLOCKING;
+	mq->tag_set.flags = BLK_MQ_F_BLOCKING;
 	mq->tag_set.nr_hw_queues = 1;
 	mq->tag_set.cmd_size = sizeof(struct mmc_queue_req);
 	mq->tag_set.driver_data = mq;

drivers/mtd/mtd_blkdevs.c

···
 		goto out_list_del;
 
 	ret = blk_mq_alloc_sq_tag_set(new->tag_set, &mtd_mq_ops, 2,
-			BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_BLOCKING);
+			BLK_MQ_F_BLOCKING);
 	if (ret)
 		goto out_kfree_tag_set;
 	

drivers/mtd/ubi/block.c

···
 	dev->tag_set.ops = &ubiblock_mq_ops;
 	dev->tag_set.queue_depth = 64;
 	dev->tag_set.numa_node = NUMA_NO_NODE;
-	dev->tag_set.flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_BLOCKING;
+	dev->tag_set.flags = BLK_MQ_F_BLOCKING;
 	dev->tag_set.cmd_size = sizeof(struct ubiblock_pdu);
 	dev->tag_set.driver_data = dev;
 	dev->tag_set.nr_hw_queues = 1;

drivers/nvme/host/apple.c

···
 	anv->admin_tagset.timeout = NVME_ADMIN_TIMEOUT;
 	anv->admin_tagset.numa_node = NUMA_NO_NODE;
 	anv->admin_tagset.cmd_size = sizeof(struct apple_nvme_iod);
-	anv->admin_tagset.flags = BLK_MQ_F_NO_SCHED;
 	anv->admin_tagset.driver_data = &anv->adminq;
 
 	ret = blk_mq_alloc_tag_set(&anv->admin_tagset);
···
 	anv->tagset.timeout = NVME_IO_TIMEOUT;
 	anv->tagset.numa_node = NUMA_NO_NODE;
 	anv->tagset.cmd_size = sizeof(struct apple_nvme_iod);
-	anv->tagset.flags = BLK_MQ_F_SHOULD_MERGE;
 	anv->tagset.driver_data = &anv->ioq;
 
 	ret = blk_mq_alloc_tag_set(&anv->tagset);

drivers/nvme/host/core.c

···
 	lim->atomic_write_hw_boundary = boundary;
 	lim->atomic_write_hw_unit_min = bs;
 	lim->atomic_write_hw_unit_max = rounddown_pow_of_two(atomic_bs);
+	lim->features |= BLK_FEAT_ATOMIC_WRITES;
 }
 
 static u32 nvme_max_drv_segments(struct nvme_ctrl *ctrl)
···
 	struct queue_limits lim;
 	int ret;
 
-	blk_mq_freeze_queue(ns->disk->queue);
 	lim = queue_limits_start_update(ns->disk->queue);
 	nvme_set_ctrl_limits(ns->ctrl, &lim);
+
+	blk_mq_freeze_queue(ns->disk->queue);
 	ret = queue_limits_commit_update(ns->disk->queue, &lim);
 	set_disk_ro(ns->disk, nvme_ns_is_readonly(ns, info));
 	blk_mq_unfreeze_queue(ns->disk->queue);
···
 			goto out;
 	}
 
+	lim = queue_limits_start_update(ns->disk->queue);
+
 	blk_mq_freeze_queue(ns->disk->queue);
 	ns->head->lba_shift = id->lbaf[lbaf].ds;
 	ns->head->nuse = le64_to_cpu(id->nuse);
 	capacity = nvme_lba_to_sect(ns->head, le64_to_cpu(id->nsze));
-
-	lim = queue_limits_start_update(ns->disk->queue);
 	nvme_set_ctrl_limits(ns->ctrl, &lim);
 	nvme_configure_metadata(ns->ctrl, ns->head, id, nvm, info);
 	nvme_set_chunk_sectors(ns, id, &lim);
···
 		struct queue_limits *ns_lim = &ns->disk->queue->limits;
 		struct queue_limits lim;
 
+		lim = queue_limits_start_update(ns->head->disk->queue);
 		blk_mq_freeze_queue(ns->head->disk->queue);
 		/*
 		 * queue_limits mixes values that are the hardware limitations
···
 		 * the splitting limits in to make sure we still obey possibly
 		 * lower limitations of other controllers.
 		 */
-		lim = queue_limits_start_update(ns->head->disk->queue);
 		lim.logical_block_size = ns_lim->logical_block_size;
 		lim.physical_block_size = ns_lim->physical_block_size;
 		lim.io_min = ns_lim->io_min;
···
 static int nvme_get_effects_log(struct nvme_ctrl *ctrl, u8 csi,
 				struct nvme_effects_log **log)
 {
-	struct nvme_effects_log	*cel = xa_load(&ctrl->cels, csi);
+	struct nvme_effects_log *old, *cel = xa_load(&ctrl->cels, csi);
 	int ret;
 
 	if (cel)
···
 		return ret;
 	}
 
-	xa_store(&ctrl->cels, csi, cel, GFP_KERNEL);
+	old = xa_store(&ctrl->cels, csi, cel, GFP_KERNEL);
+	if (xa_is_err(old)) {
+		kfree(cel);
+		return xa_err(old);
+	}
 out:
 	*log = cel;
 	return 0;
···
 	return ret;
 }
 
+static int nvme_init_effects_log(struct nvme_ctrl *ctrl,
+		u8 csi, struct nvme_effects_log **log)
+{
+	struct nvme_effects_log *effects, *old;
+
+	effects = kzalloc(sizeof(*effects), GFP_KERNEL);
+	if (!effects)
+		return -ENOMEM;
+
+	old = xa_store(&ctrl->cels, csi, effects, GFP_KERNEL);
+	if (xa_is_err(old)) {
+		kfree(effects);
+		return xa_err(old);
+	}
+
+	*log = effects;
+	return 0;
+}
+
 static void nvme_init_known_nvm_effects(struct nvme_ctrl *ctrl)
 {
 	struct nvme_effects_log	*log = ctrl->effects;
···
 	}
 
 	if (!ctrl->effects) {
-		ctrl->effects = kzalloc(sizeof(*ctrl->effects), GFP_KERNEL);
-		if (!ctrl->effects)
-			return -ENOMEM;
-		xa_store(&ctrl->cels, NVME_CSI_NVM, ctrl->effects, GFP_KERNEL);
+		ret = nvme_init_effects_log(ctrl, NVME_CSI_NVM, &ctrl->effects);
+		if (ret < 0)
+			return ret;
 	}
 
 	nvme_init_known_nvm_effects(ctrl);
···
 		/* Reserved for fabric connect and keep alive */
 		set->reserved_tags = 2;
 	set->numa_node = ctrl->numa_node;
-	set->flags = BLK_MQ_F_NO_SCHED;
 	if (ctrl->ops->flags & NVME_F_BLOCKING)
 		set->flags |= BLK_MQ_F_BLOCKING;
 	set->cmd_size = cmd_size;
···
 		/* Reserved for fabric connect */
 		set->reserved_tags = 1;
 	set->numa_node = ctrl->numa_node;
-	set->flags = BLK_MQ_F_SHOULD_MERGE;
 	if (ctrl->ops->flags & NVME_F_BLOCKING)
 		set->flags |= BLK_MQ_F_BLOCKING;
 	set->cmd_size = cmd_size;

drivers/nvme/host/fc.c

···
 #include <linux/nvme-fc.h>
 #include "fc.h"
 #include <scsi/scsi_transport_fc.h>
-#include <linux/blk-mq-pci.h>
 
 /* *************************** Data Structures/Defines ****************** */
 

drivers/nvme/host/nvme.h

···
 	return (ctrl->ctrl_config & NVME_CC_CSS_MASK) == NVME_CC_CSS_CSI;
 }
 
-#ifdef CONFIG_NVME_VERBOSE_ERRORS
-const char *nvme_get_error_status_str(u16 status);
-const char *nvme_get_opcode_str(u8 opcode);
-const char *nvme_get_admin_opcode_str(u8 opcode);
-const char *nvme_get_fabrics_opcode_str(u8 opcode);
-#else /* CONFIG_NVME_VERBOSE_ERRORS */
-static inline const char *nvme_get_error_status_str(u16 status)
-{
-	return "I/O Error";
-}
-static inline const char *nvme_get_opcode_str(u8 opcode)
-{
-	return "I/O Cmd";
-}
-static inline const char *nvme_get_admin_opcode_str(u8 opcode)
-{
-	return "Admin Cmd";
-}
-
-static inline const char *nvme_get_fabrics_opcode_str(u8 opcode)
-{
-	return "Fabrics Cmd";
-}
-#endif /* CONFIG_NVME_VERBOSE_ERRORS */
-
-static inline const char *nvme_opcode_str(int qid, u8 opcode)
-{
-	return qid ? nvme_get_opcode_str(opcode) :
-		nvme_get_admin_opcode_str(opcode);
-}
-
-static inline const char *nvme_fabrics_opcode_str(
-		int qid, const struct nvme_command *cmd)
-{
-	if (nvme_is_fabrics(cmd))
-		return nvme_get_fabrics_opcode_str(cmd->fabrics.fctype);
-
-	return nvme_opcode_str(qid, cmd->common.opcode);
-}
 #endif /* _NVME_H */

drivers/nvme/host/pci.c

···
 #include <linux/async.h>
 #include <linux/blkdev.h>
 #include <linux/blk-mq.h>
-#include <linux/blk-mq-pci.h>
 #include <linux/blk-integrity.h>
 #include <linux/dmi.h>
 #include <linux/init.h>
···
 		/*
 		 * Ensure that the doorbell is updated before reading the event
 		 * index from memory.  The controller needs to provide similar
-		 * ordering to ensure the envent index is updated before reading
+		 * ordering to ensure the event index is updated before reading
 		 * the doorbell.
 		 */
 		mb();
···
 		 */
 		map->queue_offset = qoff;
 		if (i != HCTX_TYPE_POLL && offset)
-			blk_mq_pci_map_queues(map, to_pci_dev(dev->dev), offset);
+			blk_mq_map_hw_queues(map, dev->dev, offset);
 		else
 			blk_mq_map_queues(map);
 		qoff += map->nr_queues;
···
 	}
 }
 
-static inline int nvme_poll_cq(struct nvme_queue *nvmeq,
-			       struct io_comp_batch *iob)
+static inline bool nvme_poll_cq(struct nvme_queue *nvmeq,
+			        struct io_comp_batch *iob)
 {
-	int found = 0;
+	bool found = false;
 
 	while (nvme_cqe_pending(nvmeq)) {
-		found++;
+		found = true;
 		/*
 		 * load-load control dependency between phase and the rest of
 		 * the cqe requires a full read memory barrier
···
 			sizeof(*dev->host_mem_descs), &dev->host_mem_descs_dma,
 			GFP_KERNEL);
 	if (!dev->host_mem_descs) {
-		dma_free_noncontiguous(dev->dev, dev->host_mem_size,
-				dev->hmb_sgt, DMA_BIDIRECTIONAL);
+		dma_free_noncontiguous(dev->dev, size, dev->hmb_sgt,
+				DMA_BIDIRECTIONAL);
 		dev->hmb_sgt = NULL;
 		return -ENOMEM;
 	}

drivers/nvme/host/tcp.c

···
 		 "nvme TLS handshake timeout in seconds (default 10)");
 #endif
 
+static atomic_t nvme_tcp_cpu_queues[NR_CPUS];
+
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 /* lockdep can detect a circular dependency of the form
  *   sk_lock -> mmap_lock (page fault) -> fs locks -> sk_lock
···
 	NVME_TCP_Q_ALLOCATED	= 0,
 	NVME_TCP_Q_LIVE		= 1,
 	NVME_TCP_Q_POLLING	= 2,
+	NVME_TCP_Q_IO_CPU_SET	= 3,
 };
 
 enum nvme_tcp_recv_state {
···
 			  ctrl->io_queues[HCTX_TYPE_POLL];
 }
 
+/**
+ * Track the number of queues assigned to each cpu using a global per-cpu
+ * counter and select the least used cpu from the mq_map. Our goal is to spread
+ * different controllers I/O threads across different cpu cores.
+ *
+ * Note that the accounting is not 100% perfect, but we don't need to be, we're
+ * simply putting our best effort to select the best candidate cpu core that we
+ * find at any given point.
+ */
 static void nvme_tcp_set_queue_io_cpu(struct nvme_tcp_queue *queue)
 {
 	struct nvme_tcp_ctrl *ctrl = queue->ctrl;
-	int qid = nvme_tcp_queue_id(queue);
-	int n = 0;
+	struct blk_mq_tag_set *set = &ctrl->tag_set;
+	int qid = nvme_tcp_queue_id(queue) - 1;
+	unsigned int *mq_map = NULL;
+	int cpu, min_queues = INT_MAX, io_cpu;
+
+	if (wq_unbound)
+		goto out;
 
 	if (nvme_tcp_default_queue(queue))
-		n = qid - 1;
+		mq_map = set->map[HCTX_TYPE_DEFAULT].mq_map;
 	else if (nvme_tcp_read_queue(queue))
-		n = qid - ctrl->io_queues[HCTX_TYPE_DEFAULT] - 1;
+		mq_map = set->map[HCTX_TYPE_READ].mq_map;
 	else if (nvme_tcp_poll_queue(queue))
-		n = qid - ctrl->io_queues[HCTX_TYPE_DEFAULT] -
-				ctrl->io_queues[HCTX_TYPE_READ] - 1;
-	if (wq_unbound)
-		queue->io_cpu = WORK_CPU_UNBOUND;
-	else
-		queue->io_cpu = cpumask_next_wrap(n - 1, cpu_online_mask, -1, false);
+		mq_map = set->map[HCTX_TYPE_POLL].mq_map;
+
+	if (WARN_ON(!mq_map))
+		goto out;
+
+	/* Search for the least used cpu from the mq_map */
+	io_cpu = WORK_CPU_UNBOUND;
+	for_each_online_cpu(cpu) {
+		int num_queues = atomic_read(&nvme_tcp_cpu_queues[cpu]);
+
+		if (mq_map[cpu] != qid)
+			continue;
+		if (num_queues < min_queues) {
+			io_cpu = cpu;
+			min_queues = num_queues;
+		}
+	}
+	if (io_cpu != WORK_CPU_UNBOUND) {
+		queue->io_cpu = io_cpu;
+		atomic_inc(&nvme_tcp_cpu_queues[io_cpu]);
+		set_bit(NVME_TCP_Q_IO_CPU_SET, &queue->flags);
+	}
+out:
+	dev_dbg(ctrl->ctrl.device, "queue %d: using cpu %d\n",
+		qid, queue->io_cpu);
 }
 
 static void nvme_tcp_tls_done(void *data, int status, key_serial_t pskid)
···
 
 	queue->sock->sk->sk_allocation = GFP_ATOMIC;
 	queue->sock->sk->sk_use_task_frag = false;
-	nvme_tcp_set_queue_io_cpu(queue);
+	queue->io_cpu = WORK_CPU_UNBOUND;
 	queue->request = NULL;
 	queue->data_remaining = 0;
 	queue->ddgst_remaining = 0;
···
 	if (!test_bit(NVME_TCP_Q_ALLOCATED, &queue->flags))
 		return;
 
+	if (test_and_clear_bit(NVME_TCP_Q_IO_CPU_SET, &queue->flags))
+		atomic_dec(&nvme_tcp_cpu_queues[queue->io_cpu]);
+
 	mutex_lock(&queue->queue_lock);
 	if (test_and_clear_bit(NVME_TCP_Q_LIVE, &queue->flags))
 		__nvme_tcp_stop_queue(queue);
···
 	nvme_tcp_init_recv_ctx(queue);
 	nvme_tcp_setup_sock_ops(queue);
 
-	if (idx)
+	if (idx) {
+		nvme_tcp_set_queue_io_cpu(queue);
 		ret = nvmf_connect_io_queue(nctrl, idx);
-	else
+	} else
 		ret = nvmf_connect_admin_queue(nctrl);
 
 	if (!ret) {
···
 static int __init nvme_tcp_init_module(void)
 {
 	unsigned int wq_flags = WQ_MEM_RECLAIM | WQ_HIGHPRI | WQ_SYSFS;
+	int cpu;
 
 	BUILD_BUG_ON(sizeof(struct nvme_tcp_hdr) != 8);
 	BUILD_BUG_ON(sizeof(struct nvme_tcp_cmd_pdu) != 72);
···
 	nvme_tcp_wq = alloc_workqueue("nvme_tcp_wq", wq_flags, 0);
 	if (!nvme_tcp_wq)
 		return -ENOMEM;
+
+	for_each_possible_cpu(cpu)
+		atomic_set(&nvme_tcp_cpu_queues[cpu], 0);
 
 	nvmf_register_transport(&nvme_tcp_transport);
 	return 0;

drivers/nvme/target/Kconfig

···
 	  target side.
 
 	  If unsure, say N.
+
+config NVME_TARGET_PCI_EPF
+	tristate "NVMe PCI Endpoint Function target support"
+	depends on NVME_TARGET && PCI_ENDPOINT
+	depends on NVME_CORE=y || NVME_CORE=NVME_TARGET
+	help
+	  This enables the NVMe PCI Endpoint Function target driver support,
+	  which allows creating a NVMe PCI controller using an endpoint mode
+	  capable PCI controller.
+
+	  If unsure, say N.

drivers/nvme/target/Makefile

···
 obj-$(CONFIG_NVME_TARGET_FC)		+= nvmet-fc.o
 obj-$(CONFIG_NVME_TARGET_FCLOOP)	+= nvme-fcloop.o
 obj-$(CONFIG_NVME_TARGET_TCP)		+= nvmet-tcp.o
+obj-$(CONFIG_NVME_TARGET_PCI_EPF)	+= nvmet-pci-epf.o
 
 nvmet-y		+= core.o configfs.o admin-cmd.o fabrics-cmd.o \
 			discovery.o io-cmd-file.o io-cmd-bdev.o pr.o
···
 nvmet-fc-y	+= fc.o
 nvme-fcloop-y	+= fcloop.o
 nvmet-tcp-y	+= tcp.o
+nvmet-pci-epf-y	+= pci-epf.o
 nvmet-$(CONFIG_TRACING)	+= trace.o

drivers/nvme/target/admin-cmd.c

···
 #include <linux/unaligned.h>
 #include "nvmet.h"
 
+static void nvmet_execute_delete_sq(struct nvmet_req *req)
+{
+	struct nvmet_ctrl *ctrl = req->sq->ctrl;
+	u16 sqid = le16_to_cpu(req->cmd->delete_queue.qid);
+	u16 status;
+
+	if (!nvmet_is_pci_ctrl(ctrl)) {
+		status = nvmet_report_invalid_opcode(req);
+		goto complete;
+	}
+
+	if (!sqid) {
+		status = NVME_SC_QID_INVALID | NVME_STATUS_DNR;
+		goto complete;
+	}
+
+	status = nvmet_check_sqid(ctrl, sqid, false);
+	if (status != NVME_SC_SUCCESS)
+		goto complete;
+
+	status = ctrl->ops->delete_sq(ctrl, sqid);
+
+complete:
+	nvmet_req_complete(req, status);
+}
+
+static void nvmet_execute_create_sq(struct nvmet_req *req)
+{
+	struct nvmet_ctrl *ctrl = req->sq->ctrl;
+	struct nvme_command *cmd = req->cmd;
+	u16 sqid = le16_to_cpu(cmd->create_sq.sqid);
+	u16 cqid = le16_to_cpu(cmd->create_sq.cqid);
+	u16 sq_flags = le16_to_cpu(cmd->create_sq.sq_flags);
+	u16 qsize = le16_to_cpu(cmd->create_sq.qsize);
+	u64 prp1 = le64_to_cpu(cmd->create_sq.prp1);
+	u16 status;
+
+	if (!nvmet_is_pci_ctrl(ctrl)) {
+		status = nvmet_report_invalid_opcode(req);
+		goto complete;
+	}
+
+	if (!sqid) {
+		status = NVME_SC_QID_INVALID | NVME_STATUS_DNR;
+		goto complete;
+	}
+
+	status = nvmet_check_sqid(ctrl, sqid, true);
+	if (status != NVME_SC_SUCCESS)
+		goto complete;
+
+	/*
+	 * Note: The NVMe specification allows multiple SQs to use the same CQ.
+	 * However, the target code does not really support that. So for now,
+	 * prevent this and fail the command if sqid and cqid are different.
+	 */
+	if (!cqid || cqid != sqid) {
+		pr_err("SQ %u: Unsupported CQID %u\n", sqid, cqid);
+		status = NVME_SC_CQ_INVALID | NVME_STATUS_DNR;
+		goto complete;
+	}
+
+	if (!qsize || qsize > NVME_CAP_MQES(ctrl->cap)) {
+		status = NVME_SC_QUEUE_SIZE | NVME_STATUS_DNR;
+		goto complete;
+	}
+
+	status = ctrl->ops->create_sq(ctrl, sqid, sq_flags, qsize, prp1);
+
+complete:
+	nvmet_req_complete(req, status);
+}
+
+static void nvmet_execute_delete_cq(struct nvmet_req *req)
+{
+	struct nvmet_ctrl *ctrl = req->sq->ctrl;
+	u16 cqid = le16_to_cpu(req->cmd->delete_queue.qid);
+	u16 status;
+
+	if (!nvmet_is_pci_ctrl(ctrl)) {
+		status = nvmet_report_invalid_opcode(req);
+		goto complete;
+	}
+
+	if (!cqid) {
+		status = NVME_SC_QID_INVALID | NVME_STATUS_DNR;
+		goto complete;
+	}
+
+	status = nvmet_check_cqid(ctrl, cqid);
+	if (status != NVME_SC_SUCCESS)
+		goto complete;
+
+	status = ctrl->ops->delete_cq(ctrl, cqid);
+
+complete:
+	nvmet_req_complete(req, status);
+}
+
+static void nvmet_execute_create_cq(struct nvmet_req *req)
+{
+	struct nvmet_ctrl *ctrl = req->sq->ctrl;
+	struct nvme_command *cmd = req->cmd;
+	u16 cqid = le16_to_cpu(cmd->create_cq.cqid);
+	u16 cq_flags = le16_to_cpu(cmd->create_cq.cq_flags);
+	u16 qsize = le16_to_cpu(cmd->create_cq.qsize);
+	u16 irq_vector = le16_to_cpu(cmd->create_cq.irq_vector);
+	u64 prp1 = le64_to_cpu(cmd->create_cq.prp1);
+	u16 status;
+
+	if (!nvmet_is_pci_ctrl(ctrl)) {
+		status = nvmet_report_invalid_opcode(req);
+		goto complete;
+	}
+
+	if (!cqid) {
+		status = NVME_SC_QID_INVALID | NVME_STATUS_DNR;
+		goto complete;
+	}
+
+	status = nvmet_check_cqid(ctrl, cqid);
+	if (status != NVME_SC_SUCCESS)
+		goto complete;
+
+	if (!qsize || qsize > NVME_CAP_MQES(ctrl->cap)) {
+		status = NVME_SC_QUEUE_SIZE | NVME_STATUS_DNR;
+		goto complete;
+	}
+
+	status = ctrl->ops->create_cq(ctrl, cqid, cq_flags, qsize,
+				      prp1, irq_vector);
+
+complete:
+	nvmet_req_complete(req, status);
+}
+
 u32 nvmet_get_log_page_len(struct nvme_command *cmd)
 {
 	u32 len = le16_to_cpu(cmd->get_log_page.numdu);
···
 	nvmet_req_complete(req, status);
 }
 
-static void nvmet_get_cmd_effects_nvm(struct nvme_effects_log *log)
+static void nvmet_get_cmd_effects_admin(struct nvmet_ctrl *ctrl,
+					struct nvme_effects_log *log)
 {
+	/* For a PCI target controller, advertize support for the . */
+	if (nvmet_is_pci_ctrl(ctrl)) {
+		log->acs[nvme_admin_delete_sq] =
+		log->acs[nvme_admin_create_sq] =
+		log->acs[nvme_admin_delete_cq] =
+		log->acs[nvme_admin_create_cq] =
+			cpu_to_le32(NVME_CMD_EFFECTS_CSUPP);
+	}
+
 	log->acs[nvme_admin_get_log_page] =
 	log->acs[nvme_admin_identify] =
 	log->acs[nvme_admin_abort_cmd] =
···
 	log->acs[nvme_admin_async_event] =
 	log->acs[nvme_admin_keep_alive] =
 		cpu_to_le32(NVME_CMD_EFFECTS_CSUPP);
+}
 
+static void nvmet_get_cmd_effects_nvm(struct nvme_effects_log *log)
+{
 	log->iocs[nvme_cmd_read] =
 	log->iocs[nvme_cmd_flush] =
 	log->iocs[nvme_cmd_dsm]	=
···
 
 static void nvmet_execute_get_log_cmd_effects_ns(struct nvmet_req *req)
 {
+	struct nvmet_ctrl *ctrl = req->sq->ctrl;
 	struct nvme_effects_log *log;
 	u16 status = NVME_SC_SUCCESS;
 
···
 
 	switch (req->cmd->get_log_page.csi) {
 	case NVME_CSI_NVM:
+		nvmet_get_cmd_effects_admin(ctrl, log);
 		nvmet_get_cmd_effects_nvm(log);
 		break;
 	case NVME_CSI_ZNS:
···
 			status = NVME_SC_INVALID_IO_CMD_SET;
 			goto free;
 		}
+		nvmet_get_cmd_effects_admin(ctrl, log);
 		nvmet_get_cmd_effects_nvm(log);
 		nvmet_get_cmd_effects_zns(log);
 		break;
···
 	struct nvmet_ctrl *ctrl = req->sq->ctrl;
 	struct nvmet_subsys *subsys = ctrl->subsys;
 	struct nvme_id_ctrl *id;
-	u32 cmd_capsule_size;
+	u32 cmd_capsule_size, ctratt;
 	u16 status = 0;
 
 	if (!subsys->subsys_discovered) {
···
 		goto out;
 	}
 
-	/* XXX: figure out how to assign real vendors IDs. */
-	id->vid = 0;
-	id->ssvid = 0;
+	id->vid = cpu_to_le16(subsys->vendor_id);
+	id->ssvid = cpu_to_le16(subsys->subsys_vendor_id);
 
 	memcpy(id->sn, ctrl->subsys->serial, NVMET_SN_MAX_SIZE);
 	memcpy_and_pad(id->mn, sizeof(id->mn), subsys->model_number,
···
 
 	/* XXX: figure out what to do about RTD3R/RTD3 */
 	id->oaes = cpu_to_le32(NVMET_AEN_CFG_OPTIONAL);
-	id->ctratt = cpu_to_le32(NVME_CTRL_ATTR_HID_128_BIT |
-		NVME_CTRL_ATTR_TBKAS);
+	ctratt = NVME_CTRL_ATTR_HID_128_BIT | NVME_CTRL_ATTR_TBKAS;
+	if (nvmet_is_pci_ctrl(ctrl))
+		ctratt |= NVME_CTRL_ATTR_RHII;
+	id->ctratt = cpu_to_le32(ctratt);
 
 	id->oacs = 0;
 
···
 	return 0;
 }
 
+static u16 nvmet_set_feat_host_id(struct nvmet_req *req)
+{
+	struct nvmet_ctrl *ctrl = req->sq->ctrl;
+
+	if (!nvmet_is_pci_ctrl(ctrl))
+		return NVME_SC_CMD_SEQ_ERROR | NVME_STATUS_DNR;
+
+	/*
+	 * The NVMe base specifications v2.1 recommends supporting 128-bits host
+	 * IDs (section 5.1.25.1.28.1). However, that same section also says
+	 * that "The controller may support a 64-bit Host Identifier and/or an
+	 * extended 128-bit Host Identifier". So simplify this support and do
+	 * not support 64-bits host IDs to avoid needing to check that all
+	 * controllers associated with the same subsystem all use the same host
+	 * ID size.
+	 */
+	if (!(req->cmd->common.cdw11 & cpu_to_le32(1 << 0))) {
+		req->error_loc = offsetof(struct nvme_common_command, cdw11);
+		return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+	}
+
+	return nvmet_copy_from_sgl(req, 0, &req->sq->ctrl->hostid,
+				   sizeof(req->sq->ctrl->hostid));
+}
+
+static u16 nvmet_set_feat_irq_coalesce(struct nvmet_req *req)
+{
+	struct nvmet_ctrl *ctrl = req->sq->ctrl;
+	u32 cdw11 = le32_to_cpu(req->cmd->common.cdw11);
+	struct nvmet_feat_irq_coalesce irqc = {
+		.time = (cdw11 >> 8) & 0xff,
+		.thr = cdw11 & 0xff,
+	};
+
+	/*
+	 * This feature is not supported for fabrics controllers and mandatory
+	 * for PCI controllers.
+	 */
+	if (!nvmet_is_pci_ctrl(ctrl)) {
+		req->error_loc = offsetof(struct nvme_common_command, cdw10);
+		return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+	}
+
+	return ctrl->ops->set_feature(ctrl, NVME_FEAT_IRQ_COALESCE, &irqc);
+}
+
+static u16 nvmet_set_feat_irq_config(struct nvmet_req *req)
+{
+	struct nvmet_ctrl *ctrl = req->sq->ctrl;
+	u32 cdw11 = le32_to_cpu(req->cmd->common.cdw11);
+	struct nvmet_feat_irq_config irqcfg = {
+		.iv = cdw11 & 0xffff,
+		.cd = (cdw11 >> 16) & 0x1,
+	};
+
+	/*
+	 * This feature is not supported for fabrics controllers and mandatory
+	 * for PCI controllers.
+	 */
+	if (!nvmet_is_pci_ctrl(ctrl)) {
+		req->error_loc = offsetof(struct nvme_common_command, cdw10);
+		return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+	}
+
+	return ctrl->ops->set_feature(ctrl, NVME_FEAT_IRQ_CONFIG, &irqcfg);
+}
+
+static u16 nvmet_set_feat_arbitration(struct nvmet_req *req)
+{
+	struct nvmet_ctrl *ctrl = req->sq->ctrl;
+	u32 cdw11 = le32_to_cpu(req->cmd->common.cdw11);
+	struct nvmet_feat_arbitration arb = {
+		.hpw = (cdw11 >> 24) & 0xff,
+		.mpw = (cdw11 >> 16) & 0xff,
+		.lpw = (cdw11 >> 8) & 0xff,
+		.ab = cdw11 & 0x3,
+	};
+
+	if (!ctrl->ops->set_feature) {
+		req->error_loc = offsetof(struct nvme_common_command, cdw10);
+		return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+	}
+
+	return ctrl->ops->set_feature(ctrl, NVME_FEAT_ARBITRATION, &arb);
+}
+
 void nvmet_execute_set_features(struct nvmet_req *req)
 {
 	struct nvmet_subsys *subsys = nvmet_req_subsys(req);
···
 		return;
 
 	switch (cdw10 & 0xff) {
+	case NVME_FEAT_ARBITRATION:
+		status = nvmet_set_feat_arbitration(req);
+		break;
 	case NVME_FEAT_NUM_QUEUES:
 		ncqr = (cdw11 >> 16) & 0xffff;
 		nsqr = cdw11 & 0xffff;
···
 		nvmet_set_result(req,
 			(subsys->max_qid - 1) | ((subsys->max_qid - 1) << 16));
 		break;
+	case NVME_FEAT_IRQ_COALESCE:
+		status = nvmet_set_feat_irq_coalesce(req);
+		break;
+	case NVME_FEAT_IRQ_CONFIG:
+		status = nvmet_set_feat_irq_config(req);
+		break;
 	case NVME_FEAT_KATO:
 		status = nvmet_set_feat_kato(req);
 		break;
···
 		status = nvmet_set_feat_async_event(req, NVMET_AEN_CFG_ALL);
 		break;
 	case NVME_FEAT_HOST_ID:
-		status = NVME_SC_CMD_SEQ_ERROR | NVME_STATUS_DNR;
+		status = nvmet_set_feat_host_id(req);
 		break;
 	case NVME_FEAT_WRITE_PROTECT:
 		status = nvmet_set_feat_write_protect(req);
···
 	return 0;
 }
 
+static u16 nvmet_get_feat_irq_coalesce(struct nvmet_req *req)
+{
+	struct nvmet_ctrl *ctrl = req->sq->ctrl;
+	struct nvmet_feat_irq_coalesce irqc = { };
+	u16 status;
+
+	/*
+	 * This feature is not supported for fabrics controllers and mandatory
+	 * for PCI controllers.
+	 */
+	if (!nvmet_is_pci_ctrl(ctrl)) {
+		req->error_loc = offsetof(struct nvme_common_command, cdw10);
+		return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+	}
+
+	status = ctrl->ops->get_feature(ctrl, NVME_FEAT_IRQ_COALESCE, &irqc);
+	if (status != NVME_SC_SUCCESS)
+		return status;
+
+	nvmet_set_result(req, ((u32)irqc.time << 8) | (u32)irqc.thr);
+
+	return NVME_SC_SUCCESS;
+}
+
+static u16 nvmet_get_feat_irq_config(struct nvmet_req *req)
+{
+	struct nvmet_ctrl *ctrl = req->sq->ctrl;
+	u32 iv = le32_to_cpu(req->cmd->common.cdw11) & 0xffff;
+	struct nvmet_feat_irq_config irqcfg = { .iv = iv };
+	u16 status;
+
+	/*
+	 * This feature is not supported for fabrics controllers and mandatory
+	 * for PCI controllers.
+	 */
+	if (!nvmet_is_pci_ctrl(ctrl)) {
+		req->error_loc = offsetof(struct nvme_common_command, cdw10);
+		return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+	}
+
+	status = ctrl->ops->get_feature(ctrl, NVME_FEAT_IRQ_CONFIG, &irqcfg);
+	if (status != NVME_SC_SUCCESS)
+		return status;
+
+	nvmet_set_result(req, ((u32)irqcfg.cd << 16) | iv);
+
+	return NVME_SC_SUCCESS;
+}
+
+static u16 nvmet_get_feat_arbitration(struct nvmet_req *req)
+{
+	struct nvmet_ctrl *ctrl = req->sq->ctrl;
+	struct nvmet_feat_arbitration arb = { };
+	u16 status;
+
+	if (!ctrl->ops->get_feature) {
+		req->error_loc = offsetof(struct nvme_common_command, cdw10);
+		return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+	}
+
+	status = ctrl->ops->get_feature(ctrl, NVME_FEAT_ARBITRATION, &arb);
+	if (status != NVME_SC_SUCCESS)
+		return status;
+
+	nvmet_set_result(req,
+			 ((u32)arb.hpw << 24) |
+			 ((u32)arb.mpw << 16) |
+			 ((u32)arb.lpw << 8) |
+			 (arb.ab & 0x3));
+
+	return NVME_SC_SUCCESS;
+}
+
 void nvmet_get_feat_kato(struct nvmet_req *req)
 {
 	nvmet_set_result(req, req->sq->ctrl->kato * 1000);
···
 	 * need to come up with some fake values for these.
 	 */
 #if 0
-	case NVME_FEAT_ARBITRATION:
-		break;
 	case NVME_FEAT_POWER_MGMT:
 		break;
 	case NVME_FEAT_TEMP_THRESH:
 		break;
 	case NVME_FEAT_ERR_RECOVERY:
 		break;
-	case NVME_FEAT_IRQ_COALESCE:
-		break;
-	case NVME_FEAT_IRQ_CONFIG:
-		break;
 	case NVME_FEAT_WRITE_ATOMIC:
 		break;
 #endif
+	case NVME_FEAT_ARBITRATION:
+		status = nvmet_get_feat_arbitration(req);
+		break;
+	case NVME_FEAT_IRQ_COALESCE:
+		status = nvmet_get_feat_irq_coalesce(req);
+		break;
+	case NVME_FEAT_IRQ_CONFIG:
+		status = nvmet_get_feat_irq_config(req);
+		break;
 	case NVME_FEAT_ASYNC_EVENT:
 		nvmet_get_feat_async_event(req);
 		break;
···
 	nvmet_req_complete(req, status);
 }
 
+u32 nvmet_admin_cmd_data_len(struct nvmet_req *req)
+{
+	struct nvme_command *cmd = req->cmd;
+
+	if (nvme_is_fabrics(cmd))
+		return nvmet_fabrics_admin_cmd_data_len(req);
+	if (nvmet_is_disc_subsys(nvmet_req_subsys(req)))
+		return nvmet_discovery_cmd_data_len(req);
+
+	switch (cmd->common.opcode) {
+	case nvme_admin_get_log_page:
+		return nvmet_get_log_page_len(cmd);
+	case nvme_admin_identify:
+		return NVME_IDENTIFY_DATA_SIZE;
+	case nvme_admin_get_features:
+		return nvmet_feat_data_len(req, le32_to_cpu(cmd->common.cdw10));
+	default:
+		return 0;
+	}
+}
+
 u16 nvmet_parse_admin_cmd(struct nvmet_req *req)
 {
 	struct nvme_command *cmd = req->cmd;
···
 	if (unlikely(ret))
 		return ret;
 
+	/* For PCI controllers, admin commands shall not use SGL. */
+	if (nvmet_is_pci_ctrl(req->sq->ctrl) && !req->sq->qid &&
+	    cmd->common.flags & NVME_CMD_SGL_ALL)
+		return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+
 	if (nvmet_is_passthru_req(req))
 		return nvmet_parse_passthru_admin_cmd(req);
 
 	switch (cmd->common.opcode) {
+	case nvme_admin_delete_sq:
+		req->execute = nvmet_execute_delete_sq;
+		return 0;
+	case nvme_admin_create_sq:
+		req->execute = nvmet_execute_create_sq;
+		return 0;
 	case nvme_admin_get_log_page:
 		req->execute = nvmet_execute_get_log_page;
+		return 0;
+	case nvme_admin_delete_cq:
+		req->execute = nvmet_execute_delete_cq;
+		return 0;
+	case nvme_admin_create_cq:
+		req->execute = nvmet_execute_create_cq;
 		return 0;
 	case nvme_admin_identify:
 		req->execute = nvmet_execute_identify;

drivers/nvme/target/configfs.c

···
 	{ NVMF_TRTYPE_RDMA,	"rdma" },
 	{ NVMF_TRTYPE_FC,	"fc" },
 	{ NVMF_TRTYPE_TCP,	"tcp" },
+	{ NVMF_TRTYPE_PCI,	"pci" },
 	{ NVMF_TRTYPE_LOOP,	"loop" },
 };
 
···
 	{ NVMF_ADDR_FAMILY_IP6,		"ipv6" },
 	{ NVMF_ADDR_FAMILY_IB,		"ib" },
 	{ NVMF_ADDR_FAMILY_FC,		"fc" },
+	{ NVMF_ADDR_FAMILY_PCI,		"pci" },
 	{ NVMF_ADDR_FAMILY_LOOP,	"loop" },
 };
 
···
 }
 CONFIGFS_ATTR(nvmet_subsys_, attr_cntlid_max);
 
+static ssize_t nvmet_subsys_attr_vendor_id_show(struct config_item *item,
+		char *page)
+{
+	return snprintf(page, PAGE_SIZE, "0x%x\n", to_subsys(item)->vendor_id);
+}
+
+static ssize_t nvmet_subsys_attr_vendor_id_store(struct config_item *item,
+		const char *page, size_t count)
+{
+	u16 vid;
+
+	if (kstrtou16(page, 0, &vid))
+		return -EINVAL;
+
+	down_write(&nvmet_config_sem);
+	to_subsys(item)->vendor_id = vid;
+	up_write(&nvmet_config_sem);
+	return count;
+}
+CONFIGFS_ATTR(nvmet_subsys_, attr_vendor_id);
+
+static ssize_t nvmet_subsys_attr_subsys_vendor_id_show(struct config_item *item,
+		char *page)
+{
+	return snprintf(page, PAGE_SIZE, "0x%x\n",
+			to_subsys(item)->subsys_vendor_id);
+}
+
+static ssize_t nvmet_subsys_attr_subsys_vendor_id_store(struct config_item *item,
+		const char *page, size_t count)
+{
+	u16 ssvid;
+
+	if (kstrtou16(page, 0, &ssvid))
+		return -EINVAL;
+
+	down_write(&nvmet_config_sem);
+	to_subsys(item)->subsys_vendor_id = ssvid;
+	up_write(&nvmet_config_sem);
+	return count;
+}
+CONFIGFS_ATTR(nvmet_subsys_, attr_subsys_vendor_id);
+
 static ssize_t nvmet_subsys_attr_model_show(struct config_item *item,
 					    char *page)
 {
···
 	&nvmet_subsys_attr_attr_serial,
 	&nvmet_subsys_attr_attr_cntlid_min,
 	&nvmet_subsys_attr_attr_cntlid_max,
+	&nvmet_subsys_attr_attr_vendor_id,
+	&nvmet_subsys_attr_attr_subsys_vendor_id,
 	&nvmet_subsys_attr_attr_model,
 	&nvmet_subsys_attr_attr_qid_max,
 	&nvmet_subsys_attr_attr_ieee_oui,
···
 		return ERR_PTR(-ENOMEM);
 
 	INIT_LIST_HEAD(&port->entry);
+	port->disc_addr.trtype = NVMF_TRTYPE_MAX;
 	config_group_init_type_name(&port->group, name, &nvmet_referral_type);
 
 	return &port->group;
···
 	port->inline_data_size = -1;	/* < 0 == let the transport choose */
 	port->max_queue_size = -1;	/* < 0 == let the transport choose */
 
+	port->disc_addr.trtype = NVMF_TRTYPE_MAX;
 	port->disc_addr.portid = cpu_to_le16(portid);
 	port->disc_addr.adrfam = NVMF_ADDR_FAMILY_MAX;
 	port->disc_addr.treq = NVMF_TREQ_DISABLE_SQFLOW;

drivers/nvme/target/core.c

···
 	complete(&sq->confirm_done);
 }
 
+u16 nvmet_check_cqid(struct nvmet_ctrl *ctrl, u16 cqid)
+{
+	if (!ctrl->sqs)
+		return NVME_SC_INTERNAL | NVME_STATUS_DNR;
+
+	if (cqid > ctrl->subsys->max_qid)
+		return NVME_SC_QID_INVALID | NVME_STATUS_DNR;
+
+	/*
+	 * Note: For PCI controllers, the NVMe specifications allows multiple
+	 * SQs to share a single CQ. However, we do not support this yet, so
+	 * check that there is no SQ defined for a CQ. If one exist, then the
+	 * CQ ID is invalid for creation as well as when the CQ is being
+	 * deleted (as that would mean that the SQ was not deleted before the
+	 * CQ).
+	 */
+	if (ctrl->sqs[cqid])
+		return NVME_SC_QID_INVALID | NVME_STATUS_DNR;
+
+	return NVME_SC_SUCCESS;
+}
+
+u16 nvmet_cq_create(struct nvmet_ctrl *ctrl, struct nvmet_cq *cq,
+		    u16 qid, u16 size)
+{
+	u16 status;
+
+	status = nvmet_check_cqid(ctrl, qid);
+	if (status != NVME_SC_SUCCESS)
+		return status;
+
+	nvmet_cq_setup(ctrl, cq, qid, size);
+
+	return NVME_SC_SUCCESS;
+}
+EXPORT_SYMBOL_GPL(nvmet_cq_create);
+
+u16 nvmet_check_sqid(struct nvmet_ctrl *ctrl, u16 sqid,
+		     bool create)
+{
+	if (!ctrl->sqs)
+		return NVME_SC_INTERNAL | NVME_STATUS_DNR;
+
+	if (sqid > ctrl->subsys->max_qid)
+		return NVME_SC_QID_INVALID | NVME_STATUS_DNR;
+
+	if ((create && ctrl->sqs[sqid]) ||
+	    (!create && !ctrl->sqs[sqid]))
+		return NVME_SC_QID_INVALID | NVME_STATUS_DNR;
+
+	return NVME_SC_SUCCESS;
+}
+
+u16 nvmet_sq_create(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq,
+		    u16 sqid, u16 size)
+{
+	u16 status;
+	int ret;
+
+	if (!kref_get_unless_zero(&ctrl->ref))
+		return NVME_SC_INTERNAL | NVME_STATUS_DNR;
+
+	status = nvmet_check_sqid(ctrl, sqid, true);
+	if (status != NVME_SC_SUCCESS)
+		return status;
+
+	ret = nvmet_sq_init(sq);
+	if (ret) {
+		status = NVME_SC_INTERNAL | NVME_STATUS_DNR;
+		goto ctrl_put;
+	}
+
+	nvmet_sq_setup(ctrl, sq, sqid, size);
+	sq->ctrl = ctrl;
+
+	return NVME_SC_SUCCESS;
+
+ctrl_put:
+	nvmet_ctrl_put(ctrl);
+	return status;
+}
+EXPORT_SYMBOL_GPL(nvmet_sq_create);
+
 void nvmet_sq_destroy(struct nvmet_sq *sq)
 {
 	struct nvmet_ctrl *ctrl = sq->ctrl;
···
 	}
 
 	return 0;
+}
+
+static u32 nvmet_io_cmd_transfer_len(struct nvmet_req *req)
+{
+	struct nvme_command *cmd = req->cmd;
+	u32 metadata_len = 0;
+
+	if (nvme_is_fabrics(cmd))
+		return nvmet_fabrics_io_cmd_data_len(req);
+
+	if (!req->ns)
+		return 0;
+
+	switch (req->cmd->common.opcode) {
+	case nvme_cmd_read:
+	case nvme_cmd_write:
+	case nvme_cmd_zone_append:
+		if (req->sq->ctrl->pi_support && nvmet_ns_has_pi(req->ns))
+			metadata_len = nvmet_rw_metadata_len(req);
+		return nvmet_rw_data_len(req) + metadata_len;
+	case nvme_cmd_dsm:
+		return nvmet_dsm_len(req);
+	case nvme_cmd_zone_mgmt_recv:
+		return (le32_to_cpu(req->cmd->zmr.numd) + 1) << 2;
+	default:
+		return 0;
+	}
 }
 
 static u16 nvmet_parse_io_cmd(struct nvmet_req *req)
···
 	/*
 	 * For fabrics, PSDT field shall describe metadata pointer (MPTR) that
 	 * contains an address of a single contiguous physical buffer that is
-	 * byte aligned.
+	 * byte aligned. For PCI controllers, this is optional so not enforced.
 	 */
 	if (unlikely((flags & NVME_CMD_SGL_ALL) != NVME_CMD_SGL_METABUF)) {
-		req->error_loc = offsetof(struct nvme_common_command, flags);
-		status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
-		goto fail;
+		if (!req->sq->ctrl || !nvmet_is_pci_ctrl(req->sq->ctrl)) {
+			req->error_loc =
+				offsetof(struct nvme_common_command, flags);
+			status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+			goto fail;
+		}
 	}
 
 	if (unlikely(!req->sq->ctrl))
···
 }
 EXPORT_SYMBOL_GPL(nvmet_req_uninit);
 
+size_t nvmet_req_transfer_len(struct nvmet_req *req)
+{
+	if (likely(req->sq->qid != 0))
+		return nvmet_io_cmd_transfer_len(req);
+	if (unlikely(!req->sq->ctrl))
+		return nvmet_connect_cmd_data_len(req);
+	return nvmet_admin_cmd_data_len(req);
+}
+EXPORT_SYMBOL_GPL(nvmet_req_transfer_len);
+
 bool nvmet_check_transfer_len(struct nvmet_req *req, size_t len)
 {
 	if (unlikely(len != req->transfer_len)) {
+		u16 status;
+
 		req->error_loc = offsetof(struct nvme_common_command, dptr);
-		nvmet_req_complete(req, NVME_SC_SGL_INVALID_DATA | NVME_STATUS_DNR);
+		if (req->cmd->common.flags & NVME_CMD_SGL_ALL)
+			status = NVME_SC_SGL_INVALID_DATA;
+		else
+			status = NVME_SC_INVALID_FIELD;
+		nvmet_req_complete(req, status | NVME_STATUS_DNR);
 		return false;
 	}
 
···
 bool nvmet_check_data_len_lte(struct nvmet_req *req, size_t data_len)
 {
 	if (unlikely(data_len > req->transfer_len)) {
+		u16 status;
+
 		req->error_loc = offsetof(struct nvme_common_command, dptr);
-		nvmet_req_complete(req, NVME_SC_SGL_INVALID_DATA | NVME_STATUS_DNR);
+		if (req->cmd->common.flags & NVME_CMD_SGL_ALL)
+			status = NVME_SC_SGL_INVALID_DATA;
+		else
+			status = NVME_SC_INVALID_FIELD;
+		nvmet_req_complete(req, status | NVME_STATUS_DNR);
 		return false;
 	}
 
···
 }
 EXPORT_SYMBOL_GPL(nvmet_req_free_sgls);
 
-static inline bool nvmet_cc_en(u32 cc)
-{
-	return (cc >> NVME_CC_EN_SHIFT) & 0x1;
-}
-
-static inline u8 nvmet_cc_css(u32 cc)
-{
-	return (cc >> NVME_CC_CSS_SHIFT) & 0x7;
-}
-
-static inline u8 nvmet_cc_mps(u32 cc)
-{
-	return (cc >> NVME_CC_MPS_SHIFT) & 0xf;
-}
-
-static inline u8 nvmet_cc_ams(u32 cc)
-{
-	return (cc >> NVME_CC_AMS_SHIFT) & 0x7;
-}
-
-static inline u8 nvmet_cc_shn(u32 cc)
-{
-	return (cc >> NVME_CC_SHN_SHIFT) & 0x3;
-}
-
-static inline u8 nvmet_cc_iosqes(u32 cc)
-{
-	return (cc >> NVME_CC_IOSQES_SHIFT) & 0xf;
-}
-
-static inline u8 nvmet_cc_iocqes(u32 cc)
-{
-	return (cc >> NVME_CC_IOCQES_SHIFT) & 0xf;
-}
-
 static inline bool nvmet_css_supported(u8 cc_css)
 {
 	switch (cc_css << NVME_CC_CSS_SHIFT) {
···
 		ctrl->csts &= ~NVME_CSTS_SHST_CMPLT;
 	mutex_unlock(&ctrl->lock);
 }
+EXPORT_SYMBOL_GPL(nvmet_update_cc);
 
 static void nvmet_init_cap(struct nvmet_ctrl *ctrl)
 {
···
  * Note: ctrl->subsys->lock should be held when calling this function
  */
 static void nvmet_setup_p2p_ns_map(struct nvmet_ctrl *ctrl,
-		struct nvmet_req *req)
+		struct device *p2p_client)
 {
 	struct nvmet_ns *ns;
 	unsigned long idx;
 
-	if (!req->p2p_client)
+	if (!p2p_client)
 		return;
 
-	ctrl->p2p_client = get_device(req->p2p_client);
+	ctrl->p2p_client = get_device(p2p_client);
 
 	nvmet_for_each_enabled_ns(&ctrl->subsys->namespaces, idx, ns)
 		nvmet_p2pmem_ns_add_p2p(ctrl, ns);
···
 	ctrl->ops->delete_ctrl(ctrl);
 }
 
-u16 nvmet_alloc_ctrl(const char *subsysnqn, const char *hostnqn,
-		struct nvmet_req *req, u32 kato, struct nvmet_ctrl **ctrlp,
-		uuid_t *hostid)
+struct nvmet_ctrl *nvmet_alloc_ctrl(struct nvmet_alloc_ctrl_args *args)
 {
 	struct nvmet_subsys *subsys;
 	struct nvmet_ctrl *ctrl;
+	u32 kato = args->kato;
+	u8 dhchap_status;
 	int ret;
-	u16 status;
 
-	status = NVME_SC_CONNECT_INVALID_PARAM | NVME_STATUS_DNR;
-	subsys = nvmet_find_get_subsys(req->port, subsysnqn);
+	args->status = NVME_SC_CONNECT_INVALID_PARAM | NVME_STATUS_DNR;
+	subsys = nvmet_find_get_subsys(args->port, args->subsysnqn);
 	if (!subsys) {
 		pr_warn("connect request for invalid subsystem %s!\n",
-			subsysnqn);
-		req->cqe->result.u32 = IPO_IATTR_CONNECT_DATA(subsysnqn);
-		req->error_loc = offsetof(struct nvme_common_command, dptr);
-		goto out;
+			args->subsysnqn);
+		args->result = IPO_IATTR_CONNECT_DATA(subsysnqn);
+		args->error_loc = offsetof(struct nvme_common_command, dptr);
+		return NULL;
 	}
 
 	down_read(&nvmet_config_sem);
-	if (!nvmet_host_allowed(subsys, hostnqn)) {
+	if (!nvmet_host_allowed(subsys, args->hostnqn)) {
 		pr_info("connect by host %s for subsystem %s not allowed\n",
-			hostnqn, subsysnqn);
-		req->cqe->result.u32 = IPO_IATTR_CONNECT_DATA(hostnqn);
+			args->hostnqn, args->subsysnqn);
+		args->result = IPO_IATTR_CONNECT_DATA(hostnqn);
 		up_read(&nvmet_config_sem);
-		status = NVME_SC_CONNECT_INVALID_HOST | NVME_STATUS_DNR;
-		req->error_loc = offsetof(struct nvme_common_command, dptr);
+		args->status = NVME_SC_CONNECT_INVALID_HOST | NVME_STATUS_DNR;
+		args->error_loc = offsetof(struct nvme_common_command, dptr);
 		goto out_put_subsystem;
 	}
 	up_read(&nvmet_config_sem);
 
-	status = NVME_SC_INTERNAL;
+	args->status = NVME_SC_INTERNAL;
 	ctrl = kzalloc(sizeof(*ctrl), GFP_KERNEL);
 	if (!ctrl)
 		goto out_put_subsystem;
 	mutex_init(&ctrl->lock);
 
-	ctrl->port = req->port;
-	ctrl->ops = req->ops;
+	ctrl->port = args->port;
+	ctrl->ops = args->ops;
 
 #ifdef CONFIG_NVME_TARGET_PASSTHRU
 	/* By default, set loop targets to clear IDS by default */
···
 	INIT_WORK(&ctrl->fatal_err_work, nvmet_fatal_error_handler);
 	INIT_DELAYED_WORK(&ctrl->ka_work, nvmet_keep_alive_timer);
 
-	memcpy(ctrl->subsysnqn, subsysnqn, NVMF_NQN_SIZE);
-	memcpy(ctrl->hostnqn, hostnqn, NVMF_NQN_SIZE);
+	memcpy(ctrl->subsysnqn, args->subsysnqn, NVMF_NQN_SIZE);
+	memcpy(ctrl->hostnqn, args->hostnqn, NVMF_NQN_SIZE);
 
 	kref_init(&ctrl->ref);
 	ctrl->subsys = subsys;
···
 			     subsys->cntlid_min, subsys->cntlid_max,
 			     GFP_KERNEL);
 	if (ret < 0) {
-		status = NVME_SC_CONNECT_CTRL_BUSY | NVME_STATUS_DNR;
+		args->status = NVME_SC_CONNECT_CTRL_BUSY | NVME_STATUS_DNR;
 		goto out_free_sqs;
 	}
 	ctrl->cntlid = ret;
 
-	uuid_copy(&ctrl->hostid, hostid);
+	uuid_copy(&ctrl->hostid, args->hostid);
 
 	/*
 	 * Discovery controllers may use some arbitrary high value
···
 	if (ret)
 		goto init_pr_fail;
 	list_add_tail(&ctrl->subsys_entry, &subsys->ctrls);
-	nvmet_setup_p2p_ns_map(ctrl, req);
+	nvmet_setup_p2p_ns_map(ctrl, args->p2p_client);
 	nvmet_debugfs_ctrl_setup(ctrl);
 	mutex_unlock(&subsys->lock);
 
-	*ctrlp = ctrl;
-	return 0;
+	if (args->hostid)
+		uuid_copy(&ctrl->hostid, args->hostid);
+
+	dhchap_status = nvmet_setup_auth(ctrl);
+	if (dhchap_status) {
+		pr_err("Failed to setup authentication, dhchap status %u\n",
+		       dhchap_status);
+		nvmet_ctrl_put(ctrl);
+		if (dhchap_status == NVME_AUTH_DHCHAP_FAILURE_FAILED)
+			args->status =
+				NVME_SC_CONNECT_INVALID_HOST | NVME_STATUS_DNR;
+		else
+			args->status = NVME_SC_INTERNAL;
+		return NULL;
+	}
+
+	args->status = NVME_SC_SUCCESS;
+
+	pr_info("Created %s controller %d for subsystem %s for NQN %s%s%s.\n",
+		nvmet_is_disc_subsys(ctrl->subsys) ? "discovery" : "nvm",
+		ctrl->cntlid, ctrl->subsys->subsysnqn, ctrl->hostnqn,
+		ctrl->pi_support ? " T10-PI is enabled" : "",
+		nvmet_has_auth(ctrl) ? " with DH-HMAC-CHAP" : "");
+
+	return ctrl;
 
 init_pr_fail:
 	mutex_unlock(&subsys->lock);
···
 	kfree(ctrl);
 out_put_subsystem:
 	nvmet_subsys_put(subsys);
-out:
-	return status;
+	return NULL;
 }
+EXPORT_SYMBOL_GPL(nvmet_alloc_ctrl);
 
 static void nvmet_ctrl_free(struct kref *ref)
 {
···
 {
 	kref_put(&ctrl->ref, nvmet_ctrl_free);
 }
+EXPORT_SYMBOL_GPL(nvmet_ctrl_put);
 
 void nvmet_ctrl_fatal_error(struct nvmet_ctrl *ctrl)
 {

drivers/nvme/target/discovery.c

···
 	}
 
 	list_for_each_entry(r, &req->port->referrals, entry) {
+		if (r->disc_addr.trtype == NVMF_TRTYPE_PCI)
+			continue;
+
 		nvmet_format_discovery_entry(hdr, r,
 				NVME_DISC_SUBSYS_NAME,
 				r->disc_addr.traddr,
···
 	}
 
 	nvmet_req_complete(req, stat);
+}
+
+u32 nvmet_discovery_cmd_data_len(struct nvmet_req *req)
+{
+	struct nvme_command *cmd = req->cmd;
+
+	switch (cmd->common.opcode) {
+	case nvme_admin_get_log_page:
+		return nvmet_get_log_page_len(req->cmd);
+	case nvme_admin_identify:
+		return NVME_IDENTIFY_DATA_SIZE;
+	default:
+		return 0;
+	}
 }
 
 u16 nvmet_parse_discovery_cmd(struct nvmet_req *req)

drivers/nvme/target/fabrics-cmd-auth.c

···
 	return data->rescode_exp;
 }
 
+u32 nvmet_auth_send_data_len(struct nvmet_req *req)
+{
+	return le32_to_cpu(req->cmd->auth_send.tl);
+}
+
 void nvmet_execute_auth_send(struct nvmet_req *req)
 {
 	struct nvmet_ctrl *ctrl = req->sq->ctrl;
···
 			offsetof(struct nvmf_auth_send_command, spsp1);
 		goto done;
 	}
-	tl = le32_to_cpu(req->cmd->auth_send.tl);
+	tl = nvmet_auth_send_data_len(req);
 	if (!tl) {
 		status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
 		req->error_loc =
···
 	data->rescode_exp = req->sq->dhchap_status;
 }
 
+u32 nvmet_auth_receive_data_len(struct nvmet_req *req)
+{
+	return le32_to_cpu(req->cmd->auth_receive.al);
+}
+
 void nvmet_execute_auth_receive(struct nvmet_req *req)
 {
 	struct nvmet_ctrl *ctrl = req->sq->ctrl;
···
 			offsetof(struct nvmf_auth_receive_command, spsp1);
 		goto done;
 	}
-	al = le32_to_cpu(req->cmd->auth_receive.al);
+	al = nvmet_auth_receive_data_len(req);
 	if (!al) {
 		status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
 		req->error_loc =

drivers/nvme/target/fabrics-cmd.c

···
 	nvmet_req_complete(req, status);
 }
 
+u32 nvmet_fabrics_admin_cmd_data_len(struct nvmet_req *req)
+{
+	struct nvme_command *cmd = req->cmd;
+
+	switch (cmd->fabrics.fctype) {
+#ifdef CONFIG_NVME_TARGET_AUTH
+	case nvme_fabrics_type_auth_send:
+		return nvmet_auth_send_data_len(req);
+	case nvme_fabrics_type_auth_receive:
+		return nvmet_auth_receive_data_len(req);
+#endif
+	default:
+		return 0;
+	}
+}
+
 u16 nvmet_parse_fabrics_admin_cmd(struct nvmet_req *req)
 {
 	struct nvme_command *cmd = req->cmd;
···
 	}
 
 	return 0;
+}
+
+u32 nvmet_fabrics_io_cmd_data_len(struct nvmet_req *req)
+{
+	struct nvme_command *cmd = req->cmd;
+
+	switch (cmd->fabrics.fctype) {
+#ifdef CONFIG_NVME_TARGET_AUTH
+	case nvme_fabrics_type_auth_send:
+		return nvmet_auth_send_data_len(req);
+	case nvme_fabrics_type_auth_receive:
+		return nvmet_auth_receive_data_len(req);
+#endif
+	default:
+		return 0;
+	}
 }
 
 u16 nvmet_parse_fabrics_io_cmd(struct nvmet_req *req)
···
 	struct nvmf_connect_command *c = &req->cmd->connect;
 	struct nvmf_connect_data *d;
 	struct nvmet_ctrl *ctrl = NULL;
-	u16 status;
-	u8 dhchap_status;
+	struct nvmet_alloc_ctrl_args args = {
+		.port = req->port,
+		.ops = req->ops,
+		.p2p_client = req->p2p_client,
+		.kato = le32_to_cpu(c->kato),
+	};
 
 	if (!nvmet_check_transfer_len(req, sizeof(struct nvmf_connect_data)))
 		return;
 
 	d = kmalloc(sizeof(*d), GFP_KERNEL);
 	if (!d) {
-		status = NVME_SC_INTERNAL;
+		args.status = NVME_SC_INTERNAL;
 		goto complete;
 	}
 
-	status = nvmet_copy_from_sgl(req, 0, d, sizeof(*d));
-	if (status)
+	args.status = nvmet_copy_from_sgl(req, 0, d, sizeof(*d));
+	if (args.status)
 		goto out;
 
 	if (c->recfmt != 0) {
 		pr_warn("invalid connect version (%d).\n",
 			le16_to_cpu(c->recfmt));
-		req->error_loc = offsetof(struct nvmf_connect_command, recfmt);
-		status = NVME_SC_CONNECT_FORMAT | NVME_STATUS_DNR;
+		args.error_loc = offsetof(struct nvmf_connect_command, recfmt);
+		args.status = NVME_SC_CONNECT_FORMAT | NVME_STATUS_DNR;
 		goto out;
 	}
 
 	if (unlikely(d->cntlid != cpu_to_le16(0xffff))) {
 		pr_warn("connect attempt for invalid controller ID %#x\n",
 			d->cntlid);
-		status = NVME_SC_CONNECT_INVALID_PARAM | NVME_STATUS_DNR;
-		req->cqe->result.u32 = IPO_IATTR_CONNECT_DATA(cntlid);
+		args.status = NVME_SC_CONNECT_INVALID_PARAM | NVME_STATUS_DNR;
+		args.result = IPO_IATTR_CONNECT_DATA(cntlid);
 		goto out;
 	}
 
 	d->subsysnqn[NVMF_NQN_FIELD_LEN - 1] = '\0';
 	d->hostnqn[NVMF_NQN_FIELD_LEN - 1] = '\0';
-	status = nvmet_alloc_ctrl(d->subsysnqn, d->hostnqn, req,
-				  le32_to_cpu(c->kato), &ctrl, &d->hostid);
-	if (status)
+
+	args.subsysnqn = d->subsysnqn;
+	args.hostnqn = d->hostnqn;
+	args.hostid = &d->hostid;
+	args.kato = c->kato;
+
+	ctrl = nvmet_alloc_ctrl(&args);
+	if (!ctrl)
 		goto out;
 
-	dhchap_status = nvmet_setup_auth(ctrl);
-	if (dhchap_status) {
-		pr_err("Failed to setup authentication, dhchap status %u\n",
-		       dhchap_status);
-		nvmet_ctrl_put(ctrl);
-		if (dhchap_status == NVME_AUTH_DHCHAP_FAILURE_FAILED)
-			status = (NVME_SC_CONNECT_INVALID_HOST | NVME_STATUS_DNR);
-		else
-			status = NVME_SC_INTERNAL;
-		goto out;
-	}
-
-	status = nvmet_install_queue(ctrl, req);
-	if (status) {
+	args.status = nvmet_install_queue(ctrl, req);
+	if (args.status) {
 		nvmet_ctrl_put(ctrl);
 		goto out;
 	}
 
-	pr_info("creating %s controller %d for subsystem %s for NQN %s%s%s.\n",
-		nvmet_is_disc_subsys(ctrl->subsys) ? "discovery" : "nvm",
-		ctrl->cntlid, ctrl->subsys->subsysnqn, ctrl->hostnqn,
-		ctrl->pi_support ? " T10-PI is enabled" : "",
-		nvmet_has_auth(ctrl) ? " with DH-HMAC-CHAP" : "");
-	req->cqe->result.u32 = cpu_to_le32(nvmet_connect_result(ctrl));
+	args.result = cpu_to_le32(nvmet_connect_result(ctrl));
 out:
 	kfree(d);
 complete:
-	nvmet_req_complete(req, status);
+	req->error_loc = args.error_loc;
+	req->cqe->result.u32 = args.result;
+	nvmet_req_complete(req, args.status);
 }
 
 static void nvmet_execute_io_connect(struct nvmet_req *req)
···
 out_ctrl_put:
 	nvmet_ctrl_put(ctrl);
 	goto out;
+}
+
+u32 nvmet_connect_cmd_data_len(struct nvmet_req *req)
+{
+	struct nvme_command *cmd = req->cmd;
+
+	if (!nvme_is_fabrics(cmd) ||
+	    cmd->fabrics.fctype != nvme_fabrics_type_connect)
+		return 0;
+
+	return sizeof(struct nvmf_connect_data);
 }
 
 u16 nvmet_parse_connect_cmd(struct nvmet_req *req)

drivers/nvme/target/io-cmd-bdev.c

···
 		iter_flags = SG_MITER_FROM_SG;
 	}
 
+	if (req->cmd->rw.control & NVME_RW_LR)
+		opf |= REQ_FAILFAST_DEV;
+
 	if (is_pci_p2pdma_page(sg_page(req->sg)))
 		opf |= REQ_NOMERGE;
 

drivers/nvme/target/nvmet.h

···
 	struct nvmet_subsys	*subsys;
 	struct nvmet_sq		**sqs;
 
+	void			*drvdata;
+
 	bool			reset_tbkas;
 
 	struct mutex		lock;
···
 	struct config_group	namespaces_group;
 	struct config_group	allowed_hosts_group;
 
+	u16			vendor_id;
+	u16			subsys_vendor_id;
 	char			*model_number;
 	u32			ieee_oui;
 	char			*firmware_rev;
···
 	void (*discovery_chg)(struct nvmet_port *port);
 	u8 (*get_mdts)(const struct nvmet_ctrl *ctrl);
 	u16 (*get_max_queue_size)(const struct nvmet_ctrl *ctrl);
+
+	/* Operations mandatory for PCI target controllers */
+	u16 (*create_sq)(struct nvmet_ctrl *ctrl, u16 sqid, u16 flags,
+			 u16 qsize, u64 prp1);
+	u16 (*delete_sq)(struct nvmet_ctrl *ctrl, u16 sqid);
+	u16 (*create_cq)(struct nvmet_ctrl *ctrl, u16 cqid, u16 flags,
+			 u16 qsize, u64 prp1, u16 irq_vector);
+	u16 (*delete_cq)(struct nvmet_ctrl *ctrl, u16 cqid);
+	u16 (*set_feature)(const struct nvmet_ctrl *ctrl, u8 feat,
+			   void *feat_data);
+	u16 (*get_feature)(const struct nvmet_ctrl *ctrl, u8 feat,
+			   void *feat_data);
 };
 
 #define NVMET_MAX_INLINE_BIOVEC	8
···
 void nvmet_stop_keep_alive_timer(struct nvmet_ctrl *ctrl);
 
 u16 nvmet_parse_connect_cmd(struct nvmet_req *req);
+u32 nvmet_connect_cmd_data_len(struct nvmet_req *req);
 void nvmet_bdev_set_limits(struct block_device *bdev, struct nvme_id_ns *id);
 u16 nvmet_bdev_parse_io_cmd(struct nvmet_req *req);
 u16 nvmet_file_parse_io_cmd(struct nvmet_req *req);
 u16 nvmet_bdev_zns_parse_io_cmd(struct nvmet_req *req);
+u32 nvmet_admin_cmd_data_len(struct nvmet_req *req);
 u16 nvmet_parse_admin_cmd(struct nvmet_req *req);
+u32 nvmet_discovery_cmd_data_len(struct nvmet_req *req);
 u16 nvmet_parse_discovery_cmd(struct nvmet_req *req);
 u16 nvmet_parse_fabrics_admin_cmd(struct nvmet_req *req);
+u32 nvmet_fabrics_admin_cmd_data_len(struct nvmet_req *req);
 u16 nvmet_parse_fabrics_io_cmd(struct nvmet_req *req);
+u32 nvmet_fabrics_io_cmd_data_len(struct nvmet_req *req);
 
 bool nvmet_req_init(struct nvmet_req *req, struct nvmet_cq *cq,
 		struct nvmet_sq *sq, const struct nvmet_fabrics_ops *ops);
 void nvmet_req_uninit(struct nvmet_req *req);
+size_t nvmet_req_transfer_len(struct nvmet_req *req);
 bool nvmet_check_transfer_len(struct nvmet_req *req, size_t len);
 bool nvmet_check_data_len_lte(struct nvmet_req *req, size_t data_len);
 void nvmet_req_complete(struct nvmet_req *req, u16 status);
···
 void nvmet_execute_get_features(struct nvmet_req *req);
 void nvmet_execute_keep_alive(struct nvmet_req *req);
 
+u16 nvmet_check_cqid(struct nvmet_ctrl *ctrl, u16 cqid);
 void nvmet_cq_setup(struct nvmet_ctrl *ctrl, struct nvmet_cq *cq, u16 qid,
 		u16 size);
+u16 nvmet_cq_create(struct nvmet_ctrl *ctrl, struct nvmet_cq *cq, u16 qid,
+		u16 size);
+u16 nvmet_check_sqid(struct nvmet_ctrl *ctrl, u16 sqid, bool create);
 void nvmet_sq_setup(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq, u16 qid,
+		u16 size);
+u16 nvmet_sq_create(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq, u16 qid,
 		u16 size);
 void nvmet_sq_destroy(struct nvmet_sq *sq);
 int nvmet_sq_init(struct nvmet_sq *sq);
···
 void nvmet_ctrl_fatal_error(struct nvmet_ctrl *ctrl);
 
 void nvmet_update_cc(struct nvmet_ctrl *ctrl, u32 new);
-u16 nvmet_alloc_ctrl(const char *subsysnqn, const char *hostnqn,
-		struct nvmet_req *req, u32 kato, struct nvmet_ctrl **ctrlp,
-		uuid_t *hostid);
+
+struct nvmet_alloc_ctrl_args {
+	struct nvmet_port	*port;
+	char			*subsysnqn;
+	char			*hostnqn;
+	uuid_t			*hostid;
+	const struct nvmet_fabrics_ops *ops;
+	struct device		*p2p_client;
+	u32			kato;
+	u32			result;
+	u16			error_loc;
+	u16			status;
+};
+
+struct nvmet_ctrl *nvmet_alloc_ctrl(struct nvmet_alloc_ctrl_args *args);
 struct nvmet_ctrl *nvmet_ctrl_find_get(const char *subsysnqn,
 				       const char *hostnqn, u16 cntlid,
 				       struct nvmet_req *req);
···
     return subsys->type != NVME_NQN_NVME;
 }
 
+static inline bool nvmet_is_pci_ctrl(struct nvmet_ctrl *ctrl)
+{
+	return ctrl->port->disc_addr.trtype == NVMF_TRTYPE_PCI;
+}
+
 #ifdef CONFIG_NVME_TARGET_PASSTHRU
 void nvmet_passthru_subsys_free(struct nvmet_subsys *subsys);
 int nvmet_passthru_ctrl_enable(struct nvmet_subsys *subsys);
···
 u16 errno_to_nvme_status(struct nvmet_req *req, int errno);
 u16 nvmet_report_invalid_opcode(struct nvmet_req *req);
 
+static inline bool nvmet_cc_en(u32 cc)
+{
+	return (cc >> NVME_CC_EN_SHIFT) & 0x1;
+}
+
+static inline u8 nvmet_cc_css(u32 cc)
+{
+	return (cc >> NVME_CC_CSS_SHIFT) & 0x7;
+}
+
+static inline u8 nvmet_cc_mps(u32 cc)
+{
+	return (cc >> NVME_CC_MPS_SHIFT) & 0xf;
+}
+
+static inline u8 nvmet_cc_ams(u32 cc)
+{
+	return (cc >> NVME_CC_AMS_SHIFT) & 0x7;
+}
+
+static inline u8 nvmet_cc_shn(u32 cc)
+{
+	return (cc >> NVME_CC_SHN_SHIFT) & 0x3;
+}
+
+static inline u8 nvmet_cc_iosqes(u32 cc)
+{
+	return (cc >> NVME_CC_IOSQES_SHIFT) & 0xf;
+}
+
+static inline u8 nvmet_cc_iocqes(u32 cc)
+{
+	return (cc >> NVME_CC_IOCQES_SHIFT) & 0xf;
+}
+
 /* Convert a 32-bit number to a 16-bit 0's based number */
 static inline __le16 to0based(u32 a)
 {
···
 }
 
 #ifdef CONFIG_NVME_TARGET_AUTH
+u32 nvmet_auth_send_data_len(struct nvmet_req *req);
 void nvmet_execute_auth_send(struct nvmet_req *req);
+u32 nvmet_auth_receive_data_len(struct nvmet_req *req);
 void nvmet_execute_auth_receive(struct nvmet_req *req);
 int nvmet_auth_set_key(struct nvmet_host *host, const char *secret,
 		       bool set_ctrl);
···
 {
 	percpu_ref_put(&pc_ref->ref);
 }
+
+/*
+ * Data for the get_feature() and set_feature() operations of PCI target
+ * controllers.
+ */
+struct nvmet_feat_irq_coalesce {
+	u8		thr;
+	u8		time;
+};
+
+struct nvmet_feat_irq_config {
+	u16		iv;
+	bool		cd;
+};
+
+struct nvmet_feat_arbitration {
+	u8		hpw;
+	u8		mpw;
+	u8		lpw;
+	u8		ab;
+};
+
 #endif /* _NVMET_H */

drivers/nvme/target/passthru.c

···
 {
 	struct scatterlist *sg;
 	struct bio *bio;
+	int ret = -EINVAL;
 	int i;
 
 	if (req->sg_cnt > BIO_MAX_VECS)
···
 	}
 
 	for_each_sg(req->sg, sg, req->sg_cnt, i) {
-		if (bio_add_pc_page(rq->q, bio, sg_page(sg), sg->length,
-				    sg->offset) < sg->length) {
-			nvmet_req_bio_put(req, bio);
-			return -EINVAL;
-		}
+		if (bio_add_page(bio, sg_page(sg), sg->length, sg->offset) <
+				sg->length)
+			goto out_bio_put;
 	}
 
-	blk_rq_bio_prep(rq, bio, req->sg_cnt);
-
+	ret = blk_rq_append_bio(rq, bio);
+	if (ret)
+		goto out_bio_put;
 	return 0;
+
+out_bio_put:
+	nvmet_req_bio_put(req, bio);
+	return ret;
 }
 
 static void nvmet_passthru_execute_cmd(struct nvmet_req *req)

drivers/nvme/target/pci-epf.c

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * NVMe PCI Endpoint Function target driver.
+ *
+ * Copyright (c) 2024, Western Digital Corporation or its affiliates.
+ * Copyright (c) 2024, Rick Wertenbroek <[e-mail]>
+ *                     REDS Institute, HEIG-VD, HES-SO, Switzerland
+ */
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/delay.h>
+#include <linux/dmaengine.h>
+#include <linux/io.h>
+#include <linux/mempool.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/nvme.h>
+#include <linux/pci_ids.h>
+#include <linux/pci-epc.h>
+#include <linux/pci-epf.h>
+#include <linux/pci_regs.h>
+#include <linux/slab.h>
+
+#include "nvmet.h"
+
+static LIST_HEAD(nvmet_pci_epf_ports);
+static DEFINE_MUTEX(nvmet_pci_epf_ports_mutex);
+
+/*
+ * Default and maximum allowed data transfer size. For the default,
+ * allow up to 128 page-sized segments. For the maximum allowed,
+ * use 4 times the default (which is completely arbitrary).
+ */
+#define NVMET_PCI_EPF_MAX_SEGS		128
+#define NVMET_PCI_EPF_MDTS_KB		\
+	(NVMET_PCI_EPF_MAX_SEGS << (PAGE_SHIFT - 10))
+#define NVMET_PCI_EPF_MAX_MDTS_KB	(NVMET_PCI_EPF_MDTS_KB * 4)
+
+/*
+ * IRQ vector coalescing threshold: by default, post 8 CQEs before raising an
+ * interrupt vector to the host. This default 8 is completely arbitrary and can
+ * be changed by the host with a nvme_set_features command.
+ */
+#define NVMET_PCI_EPF_IV_THRESHOLD	8
+
+/*
+ * BAR CC register and SQ polling intervals.
+ */
+#define NVMET_PCI_EPF_CC_POLL_INTERVAL	msecs_to_jiffies(5)
+#define NVMET_PCI_EPF_SQ_POLL_INTERVAL	msecs_to_jiffies(5)
+#define NVMET_PCI_EPF_SQ_POLL_IDLE	msecs_to_jiffies(5000)
+
+/*
+ * SQ arbitration burst default: fetch at most 8 commands at a time from an SQ.
+ */
+#define NVMET_PCI_EPF_SQ_AB		8
+
+/*
+ * Handling of CQs is normally immediate, unless we fail to map a CQ or the CQ
+ * is full, in which case we retry the CQ processing after this interval.
+ */
+#define NVMET_PCI_EPF_CQ_RETRY_INTERVAL	msecs_to_jiffies(1)
+
+enum nvmet_pci_epf_queue_flags {
+	NVMET_PCI_EPF_Q_IS_SQ = 0,	/* The queue is a submission queue */
+	NVMET_PCI_EPF_Q_LIVE,		/* The queue is live */
+	NVMET_PCI_EPF_Q_IRQ_ENABLED,	/* IRQ is enabled for this queue */
+};
+
+/*
+ * IRQ vector descriptor.
+ */
+struct nvmet_pci_epf_irq_vector {
+	unsigned int	vector;
+	unsigned int	ref;
+	bool		cd;
+	int		nr_irqs;
+};
+
+struct nvmet_pci_epf_queue {
+	union {
+		struct nvmet_sq		nvme_sq;
+		struct nvmet_cq		nvme_cq;
+	};
+	struct nvmet_pci_epf_ctrl	*ctrl;
+	unsigned long			flags;
+
+	u64				pci_addr;
+	size_t				pci_size;
+	struct pci_epc_map		pci_map;
+
+	u16				qid;
+	u16				depth;
+	u16				vector;
+	u16				head;
+	u16				tail;
+	u16				phase;
+	u32				db;
+
+	size_t				qes;
+
+	struct nvmet_pci_epf_irq_vector	*iv;
+	struct workqueue_struct		*iod_wq;
+	struct delayed_work		work;
+	spinlock_t			lock;
+	struct list_head		list;
+};
+
+/*
+ * PCI Root Complex (RC) address data segment for mapping an admin or
+ * I/O command buffer @buf of @length bytes to the PCI address @pci_addr.
+ */
+struct nvmet_pci_epf_segment {
+	void				*buf;
+	u64				pci_addr;
+	u32				length;
+};
+
+/*
+ * Command descriptors.
+ */
+struct nvmet_pci_epf_iod {
+	struct list_head		link;
+
+	struct nvmet_req		req;
+	struct nvme_command		cmd;
+	struct nvme_completion		cqe;
+	unsigned int			status;
+
+	struct nvmet_pci_epf_ctrl	*ctrl;
+
+	struct nvmet_pci_epf_queue	*sq;
+	struct nvmet_pci_epf_queue	*cq;
+
+	/* Data transfer size and direction for the command. */
+	size_t				data_len;
+	enum dma_data_direction		dma_dir;
+
+	/*
+	 * PCI Root Complex (RC) address data segments: if nr_data_segs is 1, we
+	 * use only @data_seg. Otherwise, the array of segments @data_segs is
+	 * allocated to manage multiple PCI address data segments. @data_sgl and
+	 * @data_sgt are used to setup the command request for execution by the
+	 * target core.
+	 */
+	unsigned int			nr_data_segs;
+	struct nvmet_pci_epf_segment	data_seg;
+	struct nvmet_pci_epf_segment	*data_segs;
+	struct scatterlist		data_sgl;
+	struct sg_table			data_sgt;
+
+	struct work_struct		work;
+	struct completion		done;
+};
+
+/*
+ * PCI target controller private data.
+ */
+struct nvmet_pci_epf_ctrl {
+	struct nvmet_pci_epf		*nvme_epf;
+	struct nvmet_port		*port;
+	struct nvmet_ctrl		*tctrl;
+	struct device			*dev;
+
+	unsigned int			nr_queues;
+	struct nvmet_pci_epf_queue	*sq;
+	struct nvmet_pci_epf_queue	*cq;
+	unsigned int			sq_ab;
+
+	mempool_t			iod_pool;
+	void				*bar;
+	u64				cap;
+	u32				cc;
+	u32				csts;
+
+	size_t				io_sqes;
+	size_t				io_cqes;
+
+	size_t				mps_shift;
+	size_t				mps;
+	size_t				mps_mask;
+
+	unsigned int			mdts;
+
+	struct delayed_work		poll_cc;
+	struct delayed_work		poll_sqs;
+
+	struct mutex			irq_lock;
+	struct nvmet_pci_epf_irq_vector	*irq_vectors;
+	unsigned int			irq_vector_threshold;
+
+	bool				link_up;
+	bool				enabled;
+};
+
+/*
+ * PCI EPF driver private data.
+ */
+struct nvmet_pci_epf {
+	struct pci_epf			*epf;
+
+	const struct pci_epc_features	*epc_features;
+
+	void				*reg_bar;
+	size_t				msix_table_offset;
+
+	unsigned int			irq_type;
+	unsigned int			nr_vectors;
+
+	struct nvmet_pci_epf_ctrl	ctrl;
+
+	bool				dma_enabled;
+	struct dma_chan			*dma_tx_chan;
+	struct mutex			dma_tx_lock;
+	struct dma_chan			*dma_rx_chan;
+	struct mutex			dma_rx_lock;
+
+	struct mutex			mmio_lock;
+
+	/* PCI endpoint function configfs attributes. */
+	struct config_group		group;
+	__le16				portid;
+	char				subsysnqn[NVMF_NQN_SIZE];
+	unsigned int			mdts_kb;
+};
+
+static inline u32 nvmet_pci_epf_bar_read32(struct nvmet_pci_epf_ctrl *ctrl,
+					   u32 off)
+{
+	__le32 *bar_reg = ctrl->bar + off;
+
+	return le32_to_cpu(READ_ONCE(*bar_reg));
+}
+
+static inline void nvmet_pci_epf_bar_write32(struct nvmet_pci_epf_ctrl *ctrl,
+					     u32 off, u32 val)
+{
+	__le32 *bar_reg = ctrl->bar + off;
+
+	WRITE_ONCE(*bar_reg, cpu_to_le32(val));
+}
+
+static inline u64 nvmet_pci_epf_bar_read64(struct nvmet_pci_epf_ctrl *ctrl,
+					   u32 off)
+{
+	return (u64)nvmet_pci_epf_bar_read32(ctrl, off) |
+		((u64)nvmet_pci_epf_bar_read32(ctrl, off + 4) << 32);
+}
+
+static inline void nvmet_pci_epf_bar_write64(struct nvmet_pci_epf_ctrl *ctrl,
+					     u32 off, u64 val)
+{
+	nvmet_pci_epf_bar_write32(ctrl, off, val & 0xFFFFFFFF);
+	nvmet_pci_epf_bar_write32(ctrl, off + 4, (val >> 32) & 0xFFFFFFFF);
+}
+
+static inline int nvmet_pci_epf_mem_map(struct nvmet_pci_epf *nvme_epf,
+		u64 pci_addr, size_t size, struct pci_epc_map *map)
+{
+	struct pci_epf *epf = nvme_epf->epf;
+
+	return pci_epc_mem_map(epf->epc, epf->func_no, epf->vfunc_no,
+			       pci_addr, size, map);
+}
+
+static inline void nvmet_pci_epf_mem_unmap(struct nvmet_pci_epf *nvme_epf,
+					   struct pci_epc_map *map)
+{
+	struct pci_epf *epf = nvme_epf->epf;
+
+	pci_epc_mem_unmap(epf->epc, epf->func_no, epf->vfunc_no, map);
+}
+
+struct nvmet_pci_epf_dma_filter {
+	struct device *dev;
+	u32 dma_mask;
+};
+
+static bool nvmet_pci_epf_dma_filter(struct dma_chan *chan, void *arg)
+{
+	struct nvmet_pci_epf_dma_filter *filter = arg;
+	struct dma_slave_caps caps;
+
+	memset(&caps, 0, sizeof(caps));
+	dma_get_slave_caps(chan, &caps);
+
+	return chan->device->dev == filter->dev &&
+		(filter->dma_mask & caps.directions);
+}
+
+static void nvmet_pci_epf_init_dma(struct nvmet_pci_epf *nvme_epf)
+{
+	struct pci_epf *epf = nvme_epf->epf;
+	struct device *dev = &epf->dev;
+	struct nvmet_pci_epf_dma_filter filter;
+	struct dma_chan *chan;
+	dma_cap_mask_t mask;
+
+	mutex_init(&nvme_epf->dma_rx_lock);
+	mutex_init(&nvme_epf->dma_tx_lock);
+
+	dma_cap_zero(mask);
+	dma_cap_set(DMA_SLAVE, mask);
+
+	filter.dev = epf->epc->dev.parent;
+	filter.dma_mask = BIT(DMA_DEV_TO_MEM);
+
+	chan = dma_request_channel(mask, nvmet_pci_epf_dma_filter, &filter);
+	if (!chan)
+		goto out_dma_no_rx;
+
+	nvme_epf->dma_rx_chan = chan;
+
+	filter.dma_mask = BIT(DMA_MEM_TO_DEV);
+	chan = dma_request_channel(mask, nvmet_pci_epf_dma_filter, &filter);
+	if (!chan)
+		goto out_dma_no_tx;
+
+	nvme_epf->dma_tx_chan = chan;
+
+	nvme_epf->dma_enabled = true;
+
+	dev_dbg(dev, "Using DMA RX channel %s, maximum segment size %u B\n",
+		dma_chan_name(chan),
+		dma_get_max_seg_size(dmaengine_get_dma_device(chan)));
+
+	dev_dbg(dev, "Using DMA TX channel %s, maximum segment size %u B\n",
+		dma_chan_name(chan),
+		dma_get_max_seg_size(dmaengine_get_dma_device(chan)));
+
+	return;
+
+out_dma_no_tx:
+	dma_release_channel(nvme_epf->dma_rx_chan);
+	nvme_epf->dma_rx_chan = NULL;
+
+out_dma_no_rx:
+	mutex_destroy(&nvme_epf->dma_rx_lock);
+	mutex_destroy(&nvme_epf->dma_tx_lock);
+	nvme_epf->dma_enabled = false;
+
+	dev_info(&epf->dev, "DMA not supported, falling back to MMIO\n");
+}
+
+static void nvmet_pci_epf_deinit_dma(struct nvmet_pci_epf *nvme_epf)
+{
+	if (!nvme_epf->dma_enabled)
+		return;
+
+	dma_release_channel(nvme_epf->dma_tx_chan);
+	nvme_epf->dma_tx_chan = NULL;
+	dma_release_channel(nvme_epf->dma_rx_chan);
+	nvme_epf->dma_rx_chan = NULL;
+	mutex_destroy(&nvme_epf->dma_rx_lock);
+	mutex_destroy(&nvme_epf->dma_tx_lock);
+	nvme_epf->dma_enabled = false;
+}
+
+static int nvmet_pci_epf_dma_transfer(struct nvmet_pci_epf *nvme_epf,
+		struct nvmet_pci_epf_segment *seg, enum dma_data_direction dir)
+{
+	struct pci_epf *epf = nvme_epf->epf;
+	struct dma_async_tx_descriptor *desc;
+	struct dma_slave_config sconf = {};
+	struct device *dev = &epf->dev;
+	struct device *dma_dev;
+	struct dma_chan *chan;
+	dma_cookie_t cookie;
+	dma_addr_t dma_addr;
+	struct mutex *lock;
+	int ret;
+
+	switch (dir) {
+	case DMA_FROM_DEVICE:
+		lock = &nvme_epf->dma_rx_lock;
+		chan = nvme_epf->dma_rx_chan;
+		sconf.direction = DMA_DEV_TO_MEM;
+		sconf.src_addr = seg->pci_addr;
+		break;
+	case DMA_TO_DEVICE:
+		lock = &nvme_epf->dma_tx_lock;
+		chan = nvme_epf->dma_tx_chan;
+		sconf.direction = DMA_MEM_TO_DEV;
+		sconf.dst_addr = seg->pci_addr;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	mutex_lock(lock);
+
+	dma_dev = dmaengine_get_dma_device(chan);
+	dma_addr = dma_map_single(dma_dev, seg->buf, seg->length, dir);
+	ret = dma_mapping_error(dma_dev, dma_addr);
+	if (ret)
+		goto unlock;
+
+	ret = dmaengine_slave_config(chan, &sconf);
+	if (ret) {
+		dev_err(dev, "Failed to configure DMA channel\n");
+		goto unmap;
+	}
+
+	desc = dmaengine_prep_slave_single(chan, dma_addr, seg->length,
+					   sconf.direction, DMA_CTRL_ACK);
+	if (!desc) {
+		dev_err(dev, "Failed to prepare DMA\n");
+		ret = -EIO;
+		goto unmap;
+	}
+
+	cookie = dmaengine_submit(desc);
+	ret = dma_submit_error(cookie);
+	if (ret) {
+		dev_err(dev, "Failed to do DMA submit (err=%d)\n", ret);
+		goto unmap;
+	}
+
+	if (dma_sync_wait(chan, cookie) != DMA_COMPLETE) {
+		dev_err(dev, "DMA transfer failed\n");
+		ret = -EIO;
+	}
+
+	dmaengine_terminate_sync(chan);
+
+unmap:
+	dma_unmap_single(dma_dev, dma_addr, seg->length, dir);
+
+unlock:
+	mutex_unlock(lock);
+
+	return ret;
+}
+
+static int nvmet_pci_epf_mmio_transfer(struct nvmet_pci_epf *nvme_epf,
+		struct nvmet_pci_epf_segment *seg, enum dma_data_direction dir)
+{
+	u64 pci_addr = seg->pci_addr;
+	u32 length = seg->length;
+	void *buf = seg->buf;
+	struct pci_epc_map map;
+	int ret = -EINVAL;
+
+	/*
+	 * Note: MMIO transfers do not need serialization but this is a
+	 * simple way to avoid using too many mapping windows.
+	 */
+	mutex_lock(&nvme_epf->mmio_lock);
+
+	while (length) {
+		ret = nvmet_pci_epf_mem_map(nvme_epf, pci_addr, length, &map);
+		if (ret)
+			break;
+
+		switch (dir) {
+		case DMA_FROM_DEVICE:
+			memcpy_fromio(buf, map.virt_addr, map.pci_size);
+			break;
+		case DMA_TO_DEVICE:
+			memcpy_toio(map.virt_addr, buf, map.pci_size);
+			break;
+		default:
+			ret = -EINVAL;
+			goto unlock;
+		}
+
+		pci_addr += map.pci_size;
+		buf += map.pci_size;
+		length -= map.pci_size;
+
+		nvmet_pci_epf_mem_unmap(nvme_epf, &map);
+	}
+
+unlock:
+	mutex_unlock(&nvme_epf->mmio_lock);
+
+	return ret;
+}
+
+static inline int nvmet_pci_epf_transfer_seg(struct nvmet_pci_epf *nvme_epf,
+		struct nvmet_pci_epf_segment *seg, enum dma_data_direction dir)
+{
+	if (nvme_epf->dma_enabled)
+		return nvmet_pci_epf_dma_transfer(nvme_epf, seg, dir);
+
+	return nvmet_pci_epf_mmio_transfer(nvme_epf, seg, dir);
+}
+
+static inline int nvmet_pci_epf_transfer(struct nvmet_pci_epf_ctrl *ctrl,
+					 void *buf, u64 pci_addr, u32 length,
+					 enum dma_data_direction dir)
+{
+	struct nvmet_pci_epf_segment seg = {
+		.buf = buf,
+		.pci_addr = pci_addr,
+		.length = length,
+	};
+
+	return nvmet_pci_epf_transfer_seg(ctrl->nvme_epf, &seg, dir);
+}
+
+static int nvmet_pci_epf_alloc_irq_vectors(struct nvmet_pci_epf_ctrl *ctrl)
+{
+	ctrl->irq_vectors = kcalloc(ctrl->nr_queues,
+				    sizeof(struct nvmet_pci_epf_irq_vector),
+				    GFP_KERNEL);
+	if (!ctrl->irq_vectors)
+		return -ENOMEM;
+
+	mutex_init(&ctrl->irq_lock);
+
+	return 0;
+}
+
+static void nvmet_pci_epf_free_irq_vectors(struct nvmet_pci_epf_ctrl *ctrl)
+{
+	if (ctrl->irq_vectors) {
+		mutex_destroy(&ctrl->irq_lock);
+		kfree(ctrl->irq_vectors);
+		ctrl->irq_vectors = NULL;
+	}
+}
+
+static struct nvmet_pci_epf_irq_vector *
+nvmet_pci_epf_find_irq_vector(struct nvmet_pci_epf_ctrl *ctrl, u16 vector)
+{
+	struct nvmet_pci_epf_irq_vector *iv;
+	int i;
+
+	lockdep_assert_held(&ctrl->irq_lock);
+
+	for (i = 0; i < ctrl->nr_queues; i++) {
+		iv = &ctrl->irq_vectors[i];
+		if (iv->ref && iv->vector == vector)
+			return iv;
+	}
+
+	return NULL;
+}
+
+static struct nvmet_pci_epf_irq_vector *
+nvmet_pci_epf_add_irq_vector(struct nvmet_pci_epf_ctrl *ctrl, u16 vector)
+{
+	struct nvmet_pci_epf_irq_vector *iv;
+	int i;
+
+	mutex_lock(&ctrl->irq_lock);
+
+	iv = nvmet_pci_epf_find_irq_vector(ctrl, vector);
+	if (iv) {
+		iv->ref++;
+		goto unlock;
+	}
+
+	for (i = 0; i < ctrl->nr_queues; i++) {
+		iv = &ctrl->irq_vectors[i];
+		if (!iv->ref)
+			break;
+	}
+
+	if (WARN_ON_ONCE(!iv))
+		goto unlock;
+
+	iv->ref = 1;
+	iv->vector = vector;
+	iv->nr_irqs = 0;
+
+unlock:
+	mutex_unlock(&ctrl->irq_lock);
+
+	return iv;
+}
+
+static void nvmet_pci_epf_remove_irq_vector(struct nvmet_pci_epf_ctrl *ctrl,
+					    u16 vector)
+{
+	struct nvmet_pci_epf_irq_vector *iv;
+
+	mutex_lock(&ctrl->irq_lock);
+
+	iv = nvmet_pci_epf_find_irq_vector(ctrl, vector);
+	if (iv) {
+		iv->ref--;
+		if (!iv->ref) {
+			iv->vector = 0;
+			iv->nr_irqs = 0;
+		}
+	}
+
+	mutex_unlock(&ctrl->irq_lock);
+}
+
+static bool nvmet_pci_epf_should_raise_irq(struct nvmet_pci_epf_ctrl *ctrl,
+		struct nvmet_pci_epf_queue *cq, bool force)
+{
+	struct nvmet_pci_epf_irq_vector *iv = cq->iv;
+	bool ret;
+
+	if (!test_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags))
+		return false;
+
+	/* IRQ coalescing for the admin queue is not allowed. */
+	if (!cq->qid)
+		return true;
+
+	if (iv->cd)
+		return true;
+
+	if (force) {
+		ret = iv->nr_irqs > 0;
+	} else {
+		iv->nr_irqs++;
+		ret = iv->nr_irqs >= ctrl->irq_vector_threshold;
+	}
+	if (ret)
+		iv->nr_irqs = 0;
+
+	return ret;
+}
+
+static void nvmet_pci_epf_raise_irq(struct nvmet_pci_epf_ctrl *ctrl,
+		struct nvmet_pci_epf_queue *cq, bool force)
+{
+	struct nvmet_pci_epf *nvme_epf = ctrl->nvme_epf;
+	struct pci_epf *epf = nvme_epf->epf;
+	int ret = 0;
+
+	if (!test_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags))
+		return;
+
+	mutex_lock(&ctrl->irq_lock);
+
+	if (!nvmet_pci_epf_should_raise_irq(ctrl, cq, force))
+		goto unlock;
+
+	switch (nvme_epf->irq_type) {
+	case PCI_IRQ_MSIX:
+	case PCI_IRQ_MSI:
+		ret = pci_epc_raise_irq(epf->epc, epf->func_no, epf->vfunc_no,
+					nvme_epf->irq_type, cq->vector + 1);
+		if (!ret)
+			break;
+		/*
+		 * If we got an error, it is likely because the host is using
+		 * legacy IRQs (e.g. BIOS, grub).
+		 */
+		fallthrough;
+	case PCI_IRQ_INTX:
+		ret = pci_epc_raise_irq(epf->epc, epf->func_no, epf->vfunc_no,
+					PCI_IRQ_INTX, 0);
+		break;
+	default:
+		WARN_ON_ONCE(1);
+		ret = -EINVAL;
+		break;
+	}
+
+	if (ret)
+		dev_err(ctrl->dev, "Failed to raise IRQ (err=%d)\n", ret);
+
+unlock:
+	mutex_unlock(&ctrl->irq_lock);
+}
+
+static inline const char *nvmet_pci_epf_iod_name(struct nvmet_pci_epf_iod *iod)
+{
+	return nvme_opcode_str(iod->sq->qid, iod->cmd.common.opcode);
+}
+
+static void nvmet_pci_epf_exec_iod_work(struct work_struct *work);
+
+static struct nvmet_pci_epf_iod *
+nvmet_pci_epf_alloc_iod(struct nvmet_pci_epf_queue *sq)
+{
+	struct nvmet_pci_epf_ctrl *ctrl = sq->ctrl;
+	struct nvmet_pci_epf_iod *iod;
+
+	iod = mempool_alloc(&ctrl->iod_pool, GFP_KERNEL);
+	if (unlikely(!iod))
+		return NULL;
+
+	memset(iod, 0, sizeof(*iod));
+	iod->req.cmd = &iod->cmd;
+	iod->req.cqe = &iod->cqe;
+	iod->req.port = ctrl->port;
+	iod->ctrl = ctrl;
+	iod->sq = sq;
+	iod->cq = &ctrl->cq[sq->qid];
+	INIT_LIST_HEAD(&iod->link);
+	iod->dma_dir = DMA_NONE;
+	INIT_WORK(&iod->work, nvmet_pci_epf_exec_iod_work);
+	init_completion(&iod->done);
+
+	return iod;
+}
+
+/*
+ * Allocate or grow a command table of PCI segments.
+ */
+static int nvmet_pci_epf_alloc_iod_data_segs(struct nvmet_pci_epf_iod *iod,
+					     int nsegs)
+{
+	struct nvmet_pci_epf_segment *segs;
+	int nr_segs = iod->nr_data_segs + nsegs;
+
+	segs = krealloc(iod->data_segs,
+			nr_segs * sizeof(struct nvmet_pci_epf_segment),
+			GFP_KERNEL | __GFP_ZERO);
+	if (!segs)
+		return -ENOMEM;
+
+	iod->nr_data_segs = nr_segs;
+	iod->data_segs = segs;
+
+	return 0;
+}
+
+static void nvmet_pci_epf_free_iod(struct nvmet_pci_epf_iod *iod)
+{
+	int i;
+
+	if (iod->data_segs) {
+		for (i = 0; i < iod->nr_data_segs; i++)
+			kfree(iod->data_segs[i].buf);
+		if (iod->data_segs != &iod->data_seg)
+			kfree(iod->data_segs);
+	}
+	if (iod->data_sgt.nents > 1)
+		sg_free_table(&iod->data_sgt);
+	mempool_free(iod, &iod->ctrl->iod_pool);
+}
+
+static int nvmet_pci_epf_transfer_iod_data(struct nvmet_pci_epf_iod *iod)
+{
+	struct nvmet_pci_epf *nvme_epf = iod->ctrl->nvme_epf;
+	struct nvmet_pci_epf_segment *seg = &iod->data_segs[0];
+	int i, ret;
+
+	/* Split the data transfer according to the PCI segments. */
+	for (i = 0; i < iod->nr_data_segs; i++, seg++) {
+		ret = nvmet_pci_epf_transfer_seg(nvme_epf, seg, iod->dma_dir);
+		if (ret) {
+			iod->status = NVME_SC_DATA_XFER_ERROR | NVME_STATUS_DNR;
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static inline u32 nvmet_pci_epf_prp_ofst(struct nvmet_pci_epf_ctrl *ctrl,
+					 u64 prp)
+{
+	return prp & ctrl->mps_mask;
+}
+
+static inline size_t nvmet_pci_epf_prp_size(struct nvmet_pci_epf_ctrl *ctrl,
+					    u64 prp)
+{
+	return ctrl->mps - nvmet_pci_epf_prp_ofst(ctrl, prp);
+}
+
+/*
+ * Transfer a PRP list from the host and return the number of prps.
+ */
+static int nvmet_pci_epf_get_prp_list(struct nvmet_pci_epf_ctrl *ctrl, u64 prp,
+				      size_t xfer_len, __le64 *prps)
+{
+	size_t nr_prps = (xfer_len + ctrl->mps_mask) >> ctrl->mps_shift;
+	u32 length;
+	int ret;
+
+	/*
+	 * Compute the number of PRPs required for the number of bytes to
+	 * transfer (xfer_len). If this number overflows the memory page size
+	 * with the PRP list pointer specified, only return the space available
+	 * in the memory page, the last PRP in there will be a PRP list pointer
+	 * to the remaining PRPs.
+	 */
+	length = min(nvmet_pci_epf_prp_size(ctrl, prp), nr_prps << 3);
+	ret = nvmet_pci_epf_transfer(ctrl, prps, prp, length, DMA_FROM_DEVICE);
+	if (ret)
+		return ret;
+
+	return length >> 3;
+}
+
+static int nvmet_pci_epf_iod_parse_prp_list(struct nvmet_pci_epf_ctrl *ctrl,
+					    struct nvmet_pci_epf_iod *iod)
+{
+	struct nvme_command *cmd = &iod->cmd;
+	struct nvmet_pci_epf_segment *seg;
+	size_t size = 0, ofst, prp_size, xfer_len;
+	size_t transfer_len = iod->data_len;
+	int nr_segs, nr_prps = 0;
+	u64 pci_addr, prp;
+	int i = 0, ret;
+	__le64 *prps;
+
+	prps = kzalloc(ctrl->mps, GFP_KERNEL);
+	if (!prps)
+		goto err_internal;
+
+	/*
+	 * Allocate PCI segments for the command: this considers the worst case
+	 * scenario where all prps are discontiguous, so get as many segments
+	 * as we can have prps. In practice, most of the time, we will have
+	 * far less PCI segments than prps.
+	 */
+	prp = le64_to_cpu(cmd->common.dptr.prp1);
+	if (!prp)
+		goto err_invalid_field;
+
+	ofst = nvmet_pci_epf_prp_ofst(ctrl, prp);
+	nr_segs = (transfer_len + ofst + ctrl->mps - 1) >> ctrl->mps_shift;
+
+	ret = nvmet_pci_epf_alloc_iod_data_segs(iod, nr_segs);
+	if (ret)
+		goto err_internal;
+
+	/* Set the first segment using prp1. */
+	seg = &iod->data_segs[0];
+	seg->pci_addr = prp;
+	seg->length = nvmet_pci_epf_prp_size(ctrl, prp);
+
+	size = seg->length;
+	pci_addr = prp + size;
+	nr_segs = 1;
+
+	/*
+	 * Now build the PCI address segments using the PRP lists, starting
+	 * from prp2.
+	 */
+	prp = le64_to_cpu(cmd->common.dptr.prp2);
+	if (!prp)
+		goto err_invalid_field;
+
+	while (size < transfer_len) {
+		xfer_len = transfer_len - size;
+
+		if (!nr_prps) {
+			nr_prps = nvmet_pci_epf_get_prp_list(ctrl, prp,
+							     xfer_len, prps);
+			if (nr_prps < 0)
+				goto err_internal;
+
+			i = 0;
+			ofst = 0;
+		}
+
+		/* Current entry */
+		prp = le64_to_cpu(prps[i]);
+		if (!prp)
+			goto err_invalid_field;
+
+		/* Did we reach the last PRP entry of the list? */
+		if (xfer_len > ctrl->mps && i == nr_prps - 1) {
+			/* We need more PRPs: PRP is a list pointer. */
+			nr_prps = 0;
+			continue;
+		}
+
+		/* Only the first PRP is allowed to have an offset. */
+		if (nvmet_pci_epf_prp_ofst(ctrl, prp))
+			goto err_invalid_offset;
+
+		if (prp != pci_addr) {
+			/* Discontiguous prp: new segment. */
+			nr_segs++;
+			if (WARN_ON_ONCE(nr_segs > iod->nr_data_segs))
+				goto err_internal;
+
+			seg++;
+			seg->pci_addr = prp;
+			seg->length = 0;
+			pci_addr = prp;
+		}
+
+		prp_size = min_t(size_t, ctrl->mps, xfer_len);
+		seg->length += prp_size;
+		pci_addr += prp_size;
+		size += prp_size;
+
+		i++;
+	}
+
+	iod->nr_data_segs = nr_segs;
+	ret = 0;
+
+	if (size != transfer_len) {
+		dev_err(ctrl->dev,
+			"PRPs transfer length mismatch: got %zu B, need %zu B\n",
+			size, transfer_len);
+		goto err_internal;
+	}
+
+	kfree(prps);
+
+	return 0;
+
+err_invalid_offset:
+	dev_err(ctrl->dev, "PRPs list invalid offset\n");
+	iod->status = NVME_SC_PRP_INVALID_OFFSET | NVME_STATUS_DNR;
+	goto err;
+
+err_invalid_field:
+	dev_err(ctrl->dev, "PRPs list invalid field\n");
+	iod->status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+	goto err;
+
+err_internal:
+	dev_err(ctrl->dev, "PRPs list internal error\n");
+	iod->status = NVME_SC_INTERNAL | NVME_STATUS_DNR;
+
+err:
+	kfree(prps);
+	return -EINVAL;
+}
+
+static int nvmet_pci_epf_iod_parse_prp_simple(struct nvmet_pci_epf_ctrl *ctrl,
+					      struct nvmet_pci_epf_iod *iod)
+{
+	struct nvme_command *cmd = &iod->cmd;
+	size_t transfer_len = iod->data_len;
+	int ret, nr_segs = 1;
+	u64 prp1, prp2 = 0;
+	size_t prp1_size;
+
+	prp1 = le64_to_cpu(cmd->common.dptr.prp1);
+	prp1_size = nvmet_pci_epf_prp_size(ctrl, prp1);
+
+	/* For commands crossing a page boundary, we should have prp2. */
+	if (transfer_len > prp1_size) {
+		prp2 = le64_to_cpu(cmd->common.dptr.prp2);
+		if (!prp2) {
+			iod->status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+			return -EINVAL;
+		}
+		if (nvmet_pci_epf_prp_ofst(ctrl, prp2)) {
+			iod->status =
+				NVME_SC_PRP_INVALID_OFFSET | NVME_STATUS_DNR;
+			return -EINVAL;
+		}
+		if (prp2 != prp1 + prp1_size)
+			nr_segs = 2;
+	}
+
+	if (nr_segs == 1) {
+		iod->nr_data_segs = 1;
+		iod->data_segs = &iod->data_seg;
+		iod->data_segs[0].pci_addr = prp1;
+		iod->data_segs[0].length = transfer_len;
+		return 0;
+	}
+
+	ret = nvmet_pci_epf_alloc_iod_data_segs(iod, nr_segs);
+	if (ret) {
+		iod->status = NVME_SC_INTERNAL | NVME_STATUS_DNR;
+		return ret;
+	}
+
+	iod->data_segs[0].pci_addr = prp1;
+	iod->data_segs[0].length = prp1_size;
+	iod->data_segs[1].pci_addr = prp2;
+	iod->data_segs[1].length = transfer_len - prp1_size;
+
+	return 0;
+}
+
+static int nvmet_pci_epf_iod_parse_prps(struct nvmet_pci_epf_iod *iod)
+{
+	struct nvmet_pci_epf_ctrl *ctrl = iod->ctrl;
+	u64 prp1 = le64_to_cpu(iod->cmd.common.dptr.prp1);
+	size_t ofst;
+
+	/* Get the PCI address segments for the command using its PRPs. */
+	ofst = nvmet_pci_epf_prp_ofst(ctrl, prp1);
+	if (ofst & 0x3) {
+		iod->status = NVME_SC_PRP_INVALID_OFFSET | NVME_STATUS_DNR;
+		return -EINVAL;
+	}
+
+	if (iod->data_len + ofst <= ctrl->mps * 2)
+		return nvmet_pci_epf_iod_parse_prp_simple(ctrl, iod);
+
+	return nvmet_pci_epf_iod_parse_prp_list(ctrl, iod);
+}
+
+/*
+ * Transfer an SGL segment from the host and return the number of data
+ * descriptors and the next segment descriptor, if any.
+ */
+static struct nvme_sgl_desc *
+nvmet_pci_epf_get_sgl_segment(struct nvmet_pci_epf_ctrl *ctrl,
+			      struct nvme_sgl_desc *desc, unsigned int *nr_sgls)
+{
+	struct nvme_sgl_desc *sgls;
+	u32 length = le32_to_cpu(desc->length);
+	int nr_descs, ret;
+	void *buf;
+
+	buf = kmalloc(length, GFP_KERNEL);
+	if (!buf)
+		return NULL;
+
+	ret = nvmet_pci_epf_transfer(ctrl, buf, le64_to_cpu(desc->addr), length,
+				     DMA_FROM_DEVICE);
+	if (ret) {
+		kfree(buf);
+		return NULL;
+	}
+
+	sgls = buf;
+	nr_descs = length / sizeof(struct nvme_sgl_desc);
+	if (sgls[nr_descs - 1].type == (NVME_SGL_FMT_SEG_DESC << 4) ||
+	    sgls[nr_descs - 1].type == (NVME_SGL_FMT_LAST_SEG_DESC << 4)) {
+		/*
+		 * We have another SGL segment following this one: do not count
+		 * it as a regular data SGL descriptor and return it to the
+		 * caller.
+		 */
+		*desc = sgls[nr_descs - 1];
+		nr_descs--;
+	} else {
+		/* We do not have another SGL segment after this one. */
+		desc->length = 0;
+	}
+
+	*nr_sgls = nr_descs;
+
+	return sgls;
+}
+
+static int nvmet_pci_epf_iod_parse_sgl_segments(struct nvmet_pci_epf_ctrl *ctrl,
+						struct nvmet_pci_epf_iod *iod)
+{
+	struct nvme_command *cmd = &iod->cmd;
+	struct nvme_sgl_desc seg = cmd->common.dptr.sgl;
+	struct nvme_sgl_desc *sgls = NULL;
+	int n = 0, i, nr_sgls;
+	int ret;
+
+	/*
+	 * We do not support inline data nor keyed SGLs, so we should be seeing
+	 * only segment descriptors.
+	 */
+	if (seg.type != (NVME_SGL_FMT_SEG_DESC << 4) &&
+	    seg.type != (NVME_SGL_FMT_LAST_SEG_DESC << 4)) {
+		iod->status = NVME_SC_SGL_INVALID_TYPE | NVME_STATUS_DNR;
+		return -EIO;
+	}
+
+	while (seg.length) {
+		sgls = nvmet_pci_epf_get_sgl_segment(ctrl, &seg, &nr_sgls);
+		if (!sgls) {
+			iod->status = NVME_SC_INTERNAL | NVME_STATUS_DNR;
+			return -EIO;
+		}
+
+		/* Grow the PCI segment table as needed. */
+		ret = nvmet_pci_epf_alloc_iod_data_segs(iod, nr_sgls);
+		if (ret) {
+			iod->status = NVME_SC_INTERNAL | NVME_STATUS_DNR;
+			goto out;
+		}
+
+		/*
+		 * Parse the SGL descriptors to build the PCI segment table,
+		 * checking the descriptor type as we go.
+		 */
+		for (i = 0; i < nr_sgls; i++) {
+			if (sgls[i].type != (NVME_SGL_FMT_DATA_DESC << 4)) {
+				iod->status = NVME_SC_SGL_INVALID_TYPE |
+					NVME_STATUS_DNR;
+				goto out;
+			}
+			iod->data_segs[n].pci_addr = le64_to_cpu(sgls[i].addr);
+			iod->data_segs[n].length = le32_to_cpu(sgls[i].length);
+			n++;
+		}
+
+		kfree(sgls);
+	}
+
+ out:
+	if (iod->status != NVME_SC_SUCCESS) {
+		kfree(sgls);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int nvmet_pci_epf_iod_parse_sgls(struct nvmet_pci_epf_iod *iod)
+{
+	struct nvmet_pci_epf_ctrl *ctrl = iod->ctrl;
+	struct nvme_sgl_desc *sgl = &iod->cmd.common.dptr.sgl;
+
+	if (sgl->type == (NVME_SGL_FMT_DATA_DESC << 4)) {
+		/* Single data descriptor case. */
+		iod->nr_data_segs = 1;
+		iod->data_segs = &iod->data_seg;
+		iod->data_seg.pci_addr = le64_to_cpu(sgl->addr);
+		iod->data_seg.length = le32_to_cpu(sgl->length);
+		return 0;
+	}
+
+	return nvmet_pci_epf_iod_parse_sgl_segments(ctrl, iod);
+}
+
+static int nvmet_pci_epf_alloc_iod_data_buf(struct nvmet_pci_epf_iod *iod)
+{
+	struct nvmet_pci_epf_ctrl *ctrl = iod->ctrl;
+	struct nvmet_req *req = &iod->req;
+	struct nvmet_pci_epf_segment *seg;
+	struct scatterlist *sg;
+	int ret, i;
+
+	if (iod->data_len > ctrl->mdts) {
+		iod->status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+		return -EINVAL;
+	}
+
+	/*
+	 * Get the PCI address segments for the command data buffer using either
+	 * its SGLs or PRPs.
+	 */
+	if (iod->cmd.common.flags & NVME_CMD_SGL_ALL)
+		ret = nvmet_pci_epf_iod_parse_sgls(iod);
+	else
+		ret = nvmet_pci_epf_iod_parse_prps(iod);
+	if (ret)
+		return ret;
+
+	/* Get a command buffer using SGLs matching the PCI segments. */
+	if (iod->nr_data_segs == 1) {
+		sg_init_table(&iod->data_sgl, 1);
+		iod->data_sgt.sgl = &iod->data_sgl;
+		iod->data_sgt.nents = 1;
+		iod->data_sgt.orig_nents = 1;
+	} else {
+		ret = sg_alloc_table(&iod->data_sgt, iod->nr_data_segs,
+				     GFP_KERNEL);
+		if (ret)
+			goto err_nomem;
+	}
+
+	for_each_sgtable_sg(&iod->data_sgt, sg, i) {
+		seg = &iod->data_segs[i];
+		seg->buf = kmalloc(seg->length, GFP_KERNEL);
+		if (!seg->buf)
+			goto err_nomem;
+		sg_set_buf(sg, seg->buf, seg->length);
+	}
+
+	req->transfer_len = iod->data_len;
+	req->sg = iod->data_sgt.sgl;
+	req->sg_cnt = iod->data_sgt.nents;
+
+	return 0;
+
+err_nomem:
+	iod->status = NVME_SC_INTERNAL | NVME_STATUS_DNR;
+	return -ENOMEM;
+}
+
+static void nvmet_pci_epf_complete_iod(struct nvmet_pci_epf_iod *iod)
+{
+	struct nvmet_pci_epf_queue *cq = iod->cq;
+	unsigned long flags;
+
+	/* Print an error message for failed commands, except AENs. */
+	iod->status = le16_to_cpu(iod->cqe.status) >> 1;
+	if (iod->status && iod->cmd.common.opcode != nvme_admin_async_event)
+		dev_err(iod->ctrl->dev,
+			"CQ[%d]: Command %s (0x%x) status 0x%0x\n",
+			iod->sq->qid, nvmet_pci_epf_iod_name(iod),
+			iod->cmd.common.opcode, iod->status);
+
+	/*
+	 * Add the command to the list of completed commands and schedule the
+	 * CQ work.
+	 */
+	spin_lock_irqsave(&cq->lock, flags);
+	list_add_tail(&iod->link, &cq->list);
+	queue_delayed_work(system_highpri_wq, &cq->work, 0);
+	spin_unlock_irqrestore(&cq->lock, flags);
+}
+
+static void nvmet_pci_epf_drain_queue(struct nvmet_pci_epf_queue *queue)
+{
+	struct nvmet_pci_epf_iod *iod;
+	unsigned long flags;
+
+	spin_lock_irqsave(&queue->lock, flags);
+	while (!list_empty(&queue->list)) {
+		iod = list_first_entry(&queue->list, struct nvmet_pci_epf_iod,
+				       link);
+		list_del_init(&iod->link);
+		nvmet_pci_epf_free_iod(iod);
+	}
+	spin_unlock_irqrestore(&queue->lock, flags);
+}
+
+static int nvmet_pci_epf_add_port(struct nvmet_port *port)
+{
+	mutex_lock(&nvmet_pci_epf_ports_mutex);
+	list_add_tail(&port->entry, &nvmet_pci_epf_ports);
+	mutex_unlock(&nvmet_pci_epf_ports_mutex);
+	return 0;
+}
+
+static void nvmet_pci_epf_remove_port(struct nvmet_port *port)
+{
+	mutex_lock(&nvmet_pci_epf_ports_mutex);
+	list_del_init(&port->entry);
+	mutex_unlock(&nvmet_pci_epf_ports_mutex);
+}
+
+static struct nvmet_port *
+nvmet_pci_epf_find_port(struct nvmet_pci_epf_ctrl *ctrl, __le16 portid)
+{
+	struct nvmet_port *p, *port = NULL;
+
+	mutex_lock(&nvmet_pci_epf_ports_mutex);
+	list_for_each_entry(p, &nvmet_pci_epf_ports, entry) {
+		if (p->disc_addr.portid == portid) {
+			port = p;
+			break;
+		}
+	}
+	mutex_unlock(&nvmet_pci_epf_ports_mutex);
+
+	return port;
+}
+
+static void nvmet_pci_epf_queue_response(struct nvmet_req *req)
+{
+	struct nvmet_pci_epf_iod *iod =
+		container_of(req, struct nvmet_pci_epf_iod, req);
+
+	iod->status = le16_to_cpu(req->cqe->status) >> 1;
+
+	/* If we have no data to transfer, directly complete the command. */
+	if (!iod->data_len || iod->dma_dir != DMA_TO_DEVICE) {
+		nvmet_pci_epf_complete_iod(iod);
+		return;
+	}
+
+	complete(&iod->done);
+}
+
+static u8 nvmet_pci_epf_get_mdts(const struct nvmet_ctrl *tctrl)
+{
+	struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata;
+	int page_shift = NVME_CAP_MPSMIN(tctrl->cap) + 12;
+
+	return ilog2(ctrl->mdts) - page_shift;
+}
+
+static u16 nvmet_pci_epf_create_cq(struct nvmet_ctrl *tctrl,
+		u16 cqid, u16 flags, u16 qsize, u64 pci_addr, u16 vector)
+{
+	struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata;
+	struct nvmet_pci_epf_queue *cq = &ctrl->cq[cqid];
+	u16 status;
+
+	if (test_and_set_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags))
+		return NVME_SC_QID_INVALID | NVME_STATUS_DNR;
+
+	if (!(flags & NVME_QUEUE_PHYS_CONTIG))
+		return NVME_SC_INVALID_QUEUE | NVME_STATUS_DNR;
+
+	if (flags & NVME_CQ_IRQ_ENABLED)
+		set_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags);
+
+	cq->pci_addr = pci_addr;
+	cq->qid = cqid;
+	cq->depth = qsize + 1;
+	cq->vector = vector;
+	cq->head = 0;
+	cq->tail = 0;
+	cq->phase = 1;
+	cq->db = NVME_REG_DBS + (((cqid * 2) + 1) * sizeof(u32));
+	nvmet_pci_epf_bar_write32(ctrl, cq->db, 0);
+
+	if (!cqid)
+		cq->qes = sizeof(struct nvme_completion);
+	else
+		cq->qes = ctrl->io_cqes;
+	cq->pci_size = cq->qes * cq->depth;
+
+	cq->iv = nvmet_pci_epf_add_irq_vector(ctrl, vector);
+	if (!cq->iv) {
+		status = NVME_SC_INTERNAL | NVME_STATUS_DNR;
+		goto err;
+	}
+
+	status = nvmet_cq_create(tctrl, &cq->nvme_cq, cqid, cq->depth);
+	if (status != NVME_SC_SUCCESS)
+		goto err;
+
+	dev_dbg(ctrl->dev, "CQ[%u]: %u entries of %zu B, IRQ vector %u\n",
+		cqid, qsize, cq->qes, cq->vector);
+
+	return NVME_SC_SUCCESS;
+
+err:
+	clear_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags);
+	clear_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags);
+	return status;
+}
+
+static u16 nvmet_pci_epf_delete_cq(struct nvmet_ctrl *tctrl, u16 cqid)
+{
+	struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata;
+	struct nvmet_pci_epf_queue *cq = &ctrl->cq[cqid];
+
+	if (!test_and_clear_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags))
+		return NVME_SC_QID_INVALID | NVME_STATUS_DNR;
+
+	cancel_delayed_work_sync(&cq->work);
+	nvmet_pci_epf_drain_queue(cq);
+	nvmet_pci_epf_remove_irq_vector(ctrl, cq->vector);
+
+	return NVME_SC_SUCCESS;
+}
+
+static u16 nvmet_pci_epf_create_sq(struct nvmet_ctrl *tctrl,
+		u16 sqid, u16 flags, u16 qsize, u64 pci_addr)
+{
+	struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata;
+	struct nvmet_pci_epf_queue *sq = &ctrl->sq[sqid];
+	u16 status;
+
+	if (test_and_set_bit(NVMET_PCI_EPF_Q_LIVE, &sq->flags))
+		return NVME_SC_QID_INVALID | NVME_STATUS_DNR;
+
+	if (!(flags & NVME_QUEUE_PHYS_CONTIG))
+		return NVME_SC_INVALID_QUEUE | NVME_STATUS_DNR;
+
+	sq->pci_addr = pci_addr;
+	sq->qid = sqid;
+	sq->depth = qsize + 1;
+	sq->head = 0;
+	sq->tail = 0;
+	sq->phase = 0;
+	sq->db = NVME_REG_DBS + (sqid * 2 * sizeof(u32));
+	nvmet_pci_epf_bar_write32(ctrl, sq->db, 0);
+	if (!sqid)
+		sq->qes = 1UL << NVME_ADM_SQES;
+	else
+		sq->qes = ctrl->io_sqes;
+	sq->pci_size = sq->qes * sq->depth;
+
+	status = nvmet_sq_create(tctrl, &sq->nvme_sq, sqid, sq->depth);
+	if (status != NVME_SC_SUCCESS)
+		goto out_clear_bit;
+
+	sq->iod_wq = alloc_workqueue("sq%d_wq", WQ_UNBOUND,
+				min_t(int, sq->depth, WQ_MAX_ACTIVE), sqid);
+	if (!sq->iod_wq) {
+		dev_err(ctrl->dev, "Failed to create SQ %d work queue\n", sqid);
+		status = NVME_SC_INTERNAL | NVME_STATUS_DNR;
+		goto out_destroy_sq;
+	}
+
+	dev_dbg(ctrl->dev, "SQ[%u]: %u entries of %zu B\n",
+		sqid, qsize, sq->qes);
+
+	return NVME_SC_SUCCESS;
+
+out_destroy_sq:
+	nvmet_sq_destroy(&sq->nvme_sq);
+out_clear_bit:
+	clear_bit(NVMET_PCI_EPF_Q_LIVE, &sq->flags);
+	return status;
+}
+
+static u16 nvmet_pci_epf_delete_sq(struct nvmet_ctrl *tctrl, u16 sqid)
+{
+	struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata;
+	struct nvmet_pci_epf_queue *sq = &ctrl->sq[sqid];
+
+	if (!test_and_clear_bit(NVMET_PCI_EPF_Q_LIVE, &sq->flags))
+		return NVME_SC_QID_INVALID | NVME_STATUS_DNR;
+
+	flush_workqueue(sq->iod_wq);
+	destroy_workqueue(sq->iod_wq);
+	sq->iod_wq = NULL;
+
+	nvmet_pci_epf_drain_queue(sq);
+
+	if (sq->nvme_sq.ctrl)
+		nvmet_sq_destroy(&sq->nvme_sq);
+
+	return NVME_SC_SUCCESS;
+}
+
+static u16 nvmet_pci_epf_get_feat(const struct nvmet_ctrl *tctrl,
+				  u8 feat, void *data)
+{
+	struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata;
+	struct nvmet_feat_arbitration *arb;
+	struct nvmet_feat_irq_coalesce *irqc;
+	struct nvmet_feat_irq_config *irqcfg;
+	struct nvmet_pci_epf_irq_vector *iv;
+	u16 status;
+
+	switch (feat) {
+	case NVME_FEAT_ARBITRATION:
+		arb = data;
+		if (!ctrl->sq_ab)
+			arb->ab = 0x7;
+		else
+			arb->ab = ilog2(ctrl->sq_ab);
+		return NVME_SC_SUCCESS;
+
+	case NVME_FEAT_IRQ_COALESCE:
+		irqc = data;
+		irqc->thr = ctrl->irq_vector_threshold;
+		irqc->time = 0;
+		return NVME_SC_SUCCESS;
+
+	case NVME_FEAT_IRQ_CONFIG:
+		irqcfg = data;
+		mutex_lock(&ctrl->irq_lock);
+		iv = nvmet_pci_epf_find_irq_vector(ctrl, irqcfg->iv);
+		if (iv) {
+			irqcfg->cd = iv->cd;
+			status = NVME_SC_SUCCESS;
+		} else {
+			status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+		}
+		mutex_unlock(&ctrl->irq_lock);
+		return status;
+
+	default:
+		return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+	}
+}
+
+static u16 nvmet_pci_epf_set_feat(const struct nvmet_ctrl *tctrl,
+				  u8 feat, void *data)
+{
+	struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata;
+	struct nvmet_feat_arbitration *arb;
+	struct nvmet_feat_irq_coalesce *irqc;
+	struct nvmet_feat_irq_config *irqcfg;
+	struct nvmet_pci_epf_irq_vector *iv;
+	u16 status;
+
+	switch (feat) {
+	case NVME_FEAT_ARBITRATION:
+		arb = data;
+		if (arb->ab == 0x7)
+			ctrl->sq_ab = 0;
+		else
+			ctrl->sq_ab = 1 << arb->ab;
+		return NVME_SC_SUCCESS;
+
+	case NVME_FEAT_IRQ_COALESCE:
+		/*
+		 * Since we do not implement precise IRQ coalescing timing,
+		 * ignore the time field.
+		 */
+		irqc = data;
+		ctrl->irq_vector_threshold = irqc->thr + 1;
+		return NVME_SC_SUCCESS;
+
+	case NVME_FEAT_IRQ_CONFIG:
+		irqcfg = data;
+		mutex_lock(&ctrl->irq_lock);
+		iv = nvmet_pci_epf_find_irq_vector(ctrl, irqcfg->iv);
+		if (iv) {
+			iv->cd = irqcfg->cd;
+			status = NVME_SC_SUCCESS;
+		} else {
+			status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+		}
+		mutex_unlock(&ctrl->irq_lock);
+		return status;
+
+	default:
+		return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR;
+	}
+}
+
+static const struct nvmet_fabrics_ops nvmet_pci_epf_fabrics_ops = {
+	.owner		= THIS_MODULE,
+	.type		= NVMF_TRTYPE_PCI,
+	.add_port	= nvmet_pci_epf_add_port,
+	.remove_port	= nvmet_pci_epf_remove_port,
+	.queue_response = nvmet_pci_epf_queue_response,
+	.get_mdts	= nvmet_pci_epf_get_mdts,
+	.create_cq	= nvmet_pci_epf_create_cq,
+	.delete_cq	= nvmet_pci_epf_delete_cq,
+	.create_sq	= nvmet_pci_epf_create_sq,
+	.delete_sq	= nvmet_pci_epf_delete_sq,
+	.get_feature	= nvmet_pci_epf_get_feat,
+	.set_feature	= nvmet_pci_epf_set_feat,
+};
+
+static void nvmet_pci_epf_cq_work(struct work_struct *work);
+
+static void nvmet_pci_epf_init_queue(struct nvmet_pci_epf_ctrl *ctrl,
+				     unsigned int qid, bool sq)
+{
+	struct nvmet_pci_epf_queue *queue;
+
+	if (sq) {
+		queue = &ctrl->sq[qid];
+		set_bit(NVMET_PCI_EPF_Q_IS_SQ, &queue->flags);
+	} else {
+		queue = &ctrl->cq[qid];
+		INIT_DELAYED_WORK(&queue->work, nvmet_pci_epf_cq_work);
+	}
+	queue->ctrl = ctrl;
+	queue->qid = qid;
+	spin_lock_init(&queue->lock);
+	INIT_LIST_HEAD(&queue->list);
+}
+
+static int nvmet_pci_epf_alloc_queues(struct nvmet_pci_epf_ctrl *ctrl)
+{
+	unsigned int qid;
+
+	ctrl->sq = kcalloc(ctrl->nr_queues,
+			   sizeof(struct nvmet_pci_epf_queue), GFP_KERNEL);
+	if (!ctrl->sq)
+		return -ENOMEM;
+
+	ctrl->cq = kcalloc(ctrl->nr_queues,
+			   sizeof(struct nvmet_pci_epf_queue), GFP_KERNEL);
+	if (!ctrl->cq) {
+		kfree(ctrl->sq);
+		ctrl->sq = NULL;
+		return -ENOMEM;
+	}
+
+	for (qid = 0; qid < ctrl->nr_queues; qid++) {
+		nvmet_pci_epf_init_queue(ctrl, qid, true);
+		nvmet_pci_epf_init_queue(ctrl, qid, false);
+	}
+
+	return 0;
+}
+
+static void nvmet_pci_epf_free_queues(struct nvmet_pci_epf_ctrl *ctrl)
+{
+	kfree(ctrl->sq);
+	ctrl->sq = NULL;
+	kfree(ctrl->cq);
+	ctrl->cq = NULL;
+}
+
+static int nvmet_pci_epf_map_queue(struct nvmet_pci_epf_ctrl *ctrl,
+				   struct nvmet_pci_epf_queue *queue)
+{
+	struct nvmet_pci_epf *nvme_epf = ctrl->nvme_epf;
+	int ret;
+
+	ret = nvmet_pci_epf_mem_map(nvme_epf, queue->pci_addr,
+				      queue->pci_size, &queue->pci_map);
+	if (ret) {
+		dev_err(ctrl->dev, "Failed to map queue %u (err=%d)\n",
+			queue->qid, ret);
+		return ret;
+	}
+
+	if (queue->pci_map.pci_size < queue->pci_size) {
+		dev_err(ctrl->dev, "Invalid partial mapping of queue %u\n",
+			queue->qid);
+		nvmet_pci_epf_mem_unmap(nvme_epf, &queue->pci_map);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static inline void nvmet_pci_epf_unmap_queue(struct nvmet_pci_epf_ctrl *ctrl,
+					     struct nvmet_pci_epf_queue *queue)
+{
+	nvmet_pci_epf_mem_unmap(ctrl->nvme_epf, &queue->pci_map);
+}
+
+static void nvmet_pci_epf_exec_iod_work(struct work_struct *work)
+{
+	struct nvmet_pci_epf_iod *iod =
+		container_of(work, struct nvmet_pci_epf_iod, work);
+	struct nvmet_req *req = &iod->req;
+	int ret;
+
+	if (!iod->ctrl->link_up) {
+		nvmet_pci_epf_free_iod(iod);
+		return;
+	}
+
+	if (!test_bit(NVMET_PCI_EPF_Q_LIVE, &iod->sq->flags)) {
+		iod->status = NVME_SC_QID_INVALID | NVME_STATUS_DNR;
+		goto complete;
+	}
+
+	if (!nvmet_req_init(req, &iod->cq->nvme_cq, &iod->sq->nvme_sq,
+			    &nvmet_pci_epf_fabrics_ops))
+		goto complete;
+
+	iod->data_len = nvmet_req_transfer_len(req);
+	if (iod->data_len) {
+		/*
+		 * Get the data DMA transfer direction. Here "device" means the
+		 * PCI root-complex host.
+		 */
+		if (nvme_is_write(&iod->cmd))
+			iod->dma_dir = DMA_FROM_DEVICE;
+		else
+			iod->dma_dir = DMA_TO_DEVICE;
+
+		/*
+		 * Setup the command data buffer and get the command data from
+		 * the host if needed.
+		 */
+		ret = nvmet_pci_epf_alloc_iod_data_buf(iod);
+		if (!ret && iod->dma_dir == DMA_FROM_DEVICE)
+			ret = nvmet_pci_epf_transfer_iod_data(iod);
+		if (ret) {
+			nvmet_req_uninit(req);
+			goto complete;
+		}
+	}
+
+	req->execute(req);
+
+	/*
+	 * If we do not have data to transfer after the command execution
+	 * finishes, nvmet_pci_epf_queue_response() will complete the command
+	 * directly. No need to wait for the completion in this case.
+	 */
+	if (!iod->data_len || iod->dma_dir != DMA_TO_DEVICE)
+		return;
+
+	wait_for_completion(&iod->done);
+
+	if (iod->status == NVME_SC_SUCCESS) {
+		WARN_ON_ONCE(!iod->data_len || iod->dma_dir != DMA_TO_DEVICE);
+		nvmet_pci_epf_transfer_iod_data(iod);
+	}
+
+complete:
+	nvmet_pci_epf_complete_iod(iod);
+}
+
+static int nvmet_pci_epf_process_sq(struct nvmet_pci_epf_ctrl *ctrl,
+				    struct nvmet_pci_epf_queue *sq)
+{
+	struct nvmet_pci_epf_iod *iod;
+	int ret, n = 0;
+
+	sq->tail = nvmet_pci_epf_bar_read32(ctrl, sq->db);
+	while (sq->head != sq->tail && (!ctrl->sq_ab || n < ctrl->sq_ab)) {
+		iod = nvmet_pci_epf_alloc_iod(sq);
+		if (!iod)
+			break;
+
+		/* Get the NVMe command submitted by the host. */
+		ret = nvmet_pci_epf_transfer(ctrl, &iod->cmd,
+					     sq->pci_addr + sq->head * sq->qes,
+					     sq->qes, DMA_FROM_DEVICE);
+		if (ret) {
+			/* Not much we can do... */
+			nvmet_pci_epf_free_iod(iod);
+			break;
+		}
+
+		dev_dbg(ctrl->dev, "SQ[%u]: head %u, tail %u, command %s\n",
+			sq->qid, sq->head, sq->tail,
+			nvmet_pci_epf_iod_name(iod));
+
+		sq->head++;
+		if (sq->head == sq->depth)
+			sq->head = 0;
+		n++;
+
+		queue_work_on(WORK_CPU_UNBOUND, sq->iod_wq, &iod->work);
+
+		sq->tail = nvmet_pci_epf_bar_read32(ctrl, sq->db);
+	}
+
+	return n;
+}
+
+static void nvmet_pci_epf_poll_sqs_work(struct work_struct *work)
+{
+	struct nvmet_pci_epf_ctrl *ctrl =
+		container_of(work, struct nvmet_pci_epf_ctrl, poll_sqs.work);
+	struct nvmet_pci_epf_queue *sq;
+	unsigned long last = 0;
+	int i, nr_sqs;
+
+	while (ctrl->link_up && ctrl->enabled) {
+		nr_sqs = 0;
+		/* Do round-robin arbitration. */
+		for (i = 0; i < ctrl->nr_queues; i++) {
+			sq = &ctrl->sq[i];
+			if (!test_bit(NVMET_PCI_EPF_Q_LIVE, &sq->flags))
+				continue;
+			if (nvmet_pci_epf_process_sq(ctrl, sq))
+				nr_sqs++;
+		}
+
+		if (nr_sqs) {
+			last = jiffies;
+			continue;
+		}
+
+		/*
+		 * If we have not received any command on any queue for more
+		 * than NVMET_PCI_EPF_SQ_POLL_IDLE, assume we are idle and
+		 * reschedule. This avoids "burning" a CPU when the controller
+		 * is idle for a long time.
+		 */
+		if (time_is_before_jiffies(last + NVMET_PCI_EPF_SQ_POLL_IDLE))
+			break;
+
+		cpu_relax();
+	}
+
+	schedule_delayed_work(&ctrl->poll_sqs, NVMET_PCI_EPF_SQ_POLL_INTERVAL);
+}
+
+static void nvmet_pci_epf_cq_work(struct work_struct *work)
+{
+	struct nvmet_pci_epf_queue *cq =
+		container_of(work, struct nvmet_pci_epf_queue, work.work);
+	struct nvmet_pci_epf_ctrl *ctrl = cq->ctrl;
+	struct nvme_completion *cqe;
+	struct nvmet_pci_epf_iod *iod;
+	unsigned long flags;
+	int ret, n = 0;
+
+	ret = nvmet_pci_epf_map_queue(ctrl, cq);
+	if (ret)
+		goto again;
+
+	while (test_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags) && ctrl->link_up) {
+
+		/* Check that the CQ is not full. */
+		cq->head = nvmet_pci_epf_bar_read32(ctrl, cq->db);
+		if (cq->head == cq->tail + 1) {
+			ret = -EAGAIN;
+			break;
+		}
+
+		spin_lock_irqsave(&cq->lock, flags);
+		iod = list_first_entry_or_null(&cq->list,
+					       struct nvmet_pci_epf_iod, link);
+		if (iod)
+			list_del_init(&iod->link);
+		spin_unlock_irqrestore(&cq->lock, flags);
+
+		if (!iod)
+			break;
+
+		/* Post the IOD completion entry. */
+		cqe = &iod->cqe;
+		cqe->status = cpu_to_le16((iod->status << 1) | cq->phase);
+
+		dev_dbg(ctrl->dev,
+			"CQ[%u]: %s status 0x%x, result 0x%llx, head %u, tail %u, phase %u\n",
+			cq->qid, nvmet_pci_epf_iod_name(iod), iod->status,
+			le64_to_cpu(cqe->result.u64), cq->head, cq->tail,
+			cq->phase);
+
+		memcpy_toio(cq->pci_map.virt_addr + cq->tail * cq->qes,
+			    cqe, cq->qes);
+
+		cq->tail++;
+		if (cq->tail >= cq->depth) {
+			cq->tail = 0;
+			cq->phase ^= 1;
+		}
+
+		nvmet_pci_epf_free_iod(iod);
+
+		/* Signal the host. */
+		nvmet_pci_epf_raise_irq(ctrl, cq, false);
+		n++;
+	}
+
+	nvmet_pci_epf_unmap_queue(ctrl, cq);
+
+	/*
+	 * We do not support precise IRQ coalescing time (100ns units as per
+	 * NVMe specifications). So if we have posted completion entries without
+	 * reaching the interrupt coalescing threshold, raise an interrupt.
+	 */
+	if (n)
+		nvmet_pci_epf_raise_irq(ctrl, cq, true);
+
+again:
+	if (ret < 0)
+		queue_delayed_work(system_highpri_wq, &cq->work,
+				   NVMET_PCI_EPF_CQ_RETRY_INTERVAL);
+}
+
+static int nvmet_pci_epf_enable_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
+{
+	u64 pci_addr, asq, acq;
+	u32 aqa;
+	u16 status, qsize;
+
+	if (ctrl->enabled)
+		return 0;
+
+	dev_info(ctrl->dev, "Enabling controller\n");
+
+	ctrl->mps_shift = nvmet_cc_mps(ctrl->cc) + 12;
+	ctrl->mps = 1UL << ctrl->mps_shift;
+	ctrl->mps_mask = ctrl->mps - 1;
+
+	ctrl->io_sqes = 1UL << nvmet_cc_iosqes(ctrl->cc);
+	if (ctrl->io_sqes < sizeof(struct nvme_command)) {
+		dev_err(ctrl->dev, "Unsupported I/O SQES %zu (need %zu)\n",
+			ctrl->io_sqes, sizeof(struct nvme_command));
+		return -EINVAL;
+	}
+
+	ctrl->io_cqes = 1UL << nvmet_cc_iocqes(ctrl->cc);
+	if (ctrl->io_cqes < sizeof(struct nvme_completion)) {
+		dev_err(ctrl->dev, "Unsupported I/O CQES %zu (need %zu)\n",
+			ctrl->io_sqes, sizeof(struct nvme_completion));
+		return -EINVAL;
+	}
+
+	/* Create the admin queue. */
+	aqa = nvmet_pci_epf_bar_read32(ctrl, NVME_REG_AQA);
+	asq = nvmet_pci_epf_bar_read64(ctrl, NVME_REG_ASQ);
+	acq = nvmet_pci_epf_bar_read64(ctrl, NVME_REG_ACQ);
+
+	qsize = (aqa & 0x0fff0000) >> 16;
+	pci_addr = acq & GENMASK_ULL(63, 12);
+	status = nvmet_pci_epf_create_cq(ctrl->tctrl, 0,
+				NVME_CQ_IRQ_ENABLED | NVME_QUEUE_PHYS_CONTIG,
+				qsize, pci_addr, 0);
+	if (status != NVME_SC_SUCCESS) {
+		dev_err(ctrl->dev, "Failed to create admin completion queue\n");
+		return -EINVAL;
+	}
+
+	qsize = aqa & 0x00000fff;
+	pci_addr = asq & GENMASK_ULL(63, 12);
+	status = nvmet_pci_epf_create_sq(ctrl->tctrl, 0, NVME_QUEUE_PHYS_CONTIG,
+					 qsize, pci_addr);
+	if (status != NVME_SC_SUCCESS) {
+		dev_err(ctrl->dev, "Failed to create admin submission queue\n");
+		nvmet_pci_epf_delete_cq(ctrl->tctrl, 0);
+		return -EINVAL;
+	}
+
+	ctrl->sq_ab = NVMET_PCI_EPF_SQ_AB;
+	ctrl->irq_vector_threshold = NVMET_PCI_EPF_IV_THRESHOLD;
+	ctrl->enabled = true;
+
+	/* Start polling the controller SQs. */
+	schedule_delayed_work(&ctrl->poll_sqs, 0);
+
+	return 0;
+}
+
+static void nvmet_pci_epf_disable_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
+{
+	int qid;
+
+	if (!ctrl->enabled)
+		return;
+
+	dev_info(ctrl->dev, "Disabling controller\n");
+
+	ctrl->enabled = false;
+	cancel_delayed_work_sync(&ctrl->poll_sqs);
+
+	/* Delete all I/O queues first. */
+	for (qid = 1; qid < ctrl->nr_queues; qid++)
+		nvmet_pci_epf_delete_sq(ctrl->tctrl, qid);
+
+	for (qid = 1; qid < ctrl->nr_queues; qid++)
+		nvmet_pci_epf_delete_cq(ctrl->tctrl, qid);
+
+	/* Delete the admin queue last. */
+	nvmet_pci_epf_delete_sq(ctrl->tctrl, 0);
+	nvmet_pci_epf_delete_cq(ctrl->tctrl, 0);
+}
+
+static void nvmet_pci_epf_poll_cc_work(struct work_struct *work)
+{
+	struct nvmet_pci_epf_ctrl *ctrl =
+		container_of(work, struct nvmet_pci_epf_ctrl, poll_cc.work);
+	u32 old_cc, new_cc;
+	int ret;
+
+	if (!ctrl->tctrl)
+		return;
+
+	old_cc = ctrl->cc;
+	new_cc = nvmet_pci_epf_bar_read32(ctrl, NVME_REG_CC);
+	ctrl->cc = new_cc;
+
+	if (nvmet_cc_en(new_cc) && !nvmet_cc_en(old_cc)) {
+		ret = nvmet_pci_epf_enable_ctrl(ctrl);
+		if (ret)
+			return;
+		ctrl->csts |= NVME_CSTS_RDY;
+	}
+
+	if (!nvmet_cc_en(new_cc) && nvmet_cc_en(old_cc)) {
+		nvmet_pci_epf_disable_ctrl(ctrl);
+		ctrl->csts &= ~NVME_CSTS_RDY;
+	}
+
+	if (nvmet_cc_shn(new_cc) && !nvmet_cc_shn(old_cc)) {
+		nvmet_pci_epf_disable_ctrl(ctrl);
+		ctrl->csts |= NVME_CSTS_SHST_CMPLT;
+	}
+
+	if (!nvmet_cc_shn(new_cc) && nvmet_cc_shn(old_cc))
+		ctrl->csts &= ~NVME_CSTS_SHST_CMPLT;
+
+	nvmet_update_cc(ctrl->tctrl, ctrl->cc);
+	nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CSTS, ctrl->csts);
+
+	schedule_delayed_work(&ctrl->poll_cc, NVMET_PCI_EPF_CC_POLL_INTERVAL);
+}
+
+static void nvmet_pci_epf_init_bar(struct nvmet_pci_epf_ctrl *ctrl)
+{
+	struct nvmet_ctrl *tctrl = ctrl->tctrl;
+
+	ctrl->bar = ctrl->nvme_epf->reg_bar;
+
+	/* Copy the target controller capabilities as a base. */
+	ctrl->cap = tctrl->cap;
+
+	/* Contiguous Queues Required (CQR). */
+	ctrl->cap |= 0x1ULL << 16;
+
+	/* Set Doorbell stride to 4B (DSTRB). */
+	ctrl->cap &= ~GENMASK_ULL(35, 32);
+
+	/* Clear NVM Subsystem Reset Supported (NSSRS). */
+	ctrl->cap &= ~(0x1ULL << 36);
+
+	/* Clear Boot Partition Support (BPS). */
+	ctrl->cap &= ~(0x1ULL << 45);
+
+	/* Clear Persistent Memory Region Supported (PMRS). */
+	ctrl->cap &= ~(0x1ULL << 56);
+
+	/* Clear Controller Memory Buffer Supported (CMBS). */
+	ctrl->cap &= ~(0x1ULL << 57);
+
+	/* Controller configuration. */
+	ctrl->cc = tctrl->cc & (~NVME_CC_ENABLE);
+
+	/* Controller status. */
+	ctrl->csts = ctrl->tctrl->csts;
+
+	nvmet_pci_epf_bar_write64(ctrl, NVME_REG_CAP, ctrl->cap);
+	nvmet_pci_epf_bar_write32(ctrl, NVME_REG_VS, tctrl->subsys->ver);
+	nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CSTS, ctrl->csts);
+	nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CC, ctrl->cc);
+}
+
+static int nvmet_pci_epf_create_ctrl(struct nvmet_pci_epf *nvme_epf,
+				     unsigned int max_nr_queues)
+{
+	struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl;
+	struct nvmet_alloc_ctrl_args args = {};
+	char hostnqn[NVMF_NQN_SIZE];
+	uuid_t id;
+	int ret;
+
+	memset(ctrl, 0, sizeof(*ctrl));
+	ctrl->dev = &nvme_epf->epf->dev;
+	mutex_init(&ctrl->irq_lock);
+	ctrl->nvme_epf = nvme_epf;
+	ctrl->mdts = nvme_epf->mdts_kb * SZ_1K;
+	INIT_DELAYED_WORK(&ctrl->poll_cc, nvmet_pci_epf_poll_cc_work);
+	INIT_DELAYED_WORK(&ctrl->poll_sqs, nvmet_pci_epf_poll_sqs_work);
+
+	ret = mempool_init_kmalloc_pool(&ctrl->iod_pool,
+					max_nr_queues * NVMET_MAX_QUEUE_SIZE,
+					sizeof(struct nvmet_pci_epf_iod));
+	if (ret) {
+		dev_err(ctrl->dev, "Failed to initialize IOD mempool\n");
+		return ret;
+	}
+
+	ctrl->port = nvmet_pci_epf_find_port(ctrl, nvme_epf->portid);
+	if (!ctrl->port) {
+		dev_err(ctrl->dev, "Port not found\n");
+		ret = -EINVAL;
+		goto out_mempool_exit;
+	}
+
+	/* Create the target controller. */
+	uuid_gen(&id);
+	snprintf(hostnqn, NVMF_NQN_SIZE,
+		 "nqn.2014-08.org.nvmexpress:uuid:%pUb", &id);
+	args.port = ctrl->port;
+	args.subsysnqn = nvme_epf->subsysnqn;
+	memset(&id, 0, sizeof(uuid_t));
+	args.hostid = &id;
+	args.hostnqn = hostnqn;
+	args.ops = &nvmet_pci_epf_fabrics_ops;
+
+	ctrl->tctrl = nvmet_alloc_ctrl(&args);
+	if (!ctrl->tctrl) {
+		dev_err(ctrl->dev, "Failed to create target controller\n");
+		ret = -ENOMEM;
+		goto out_mempool_exit;
+	}
+	ctrl->tctrl->drvdata = ctrl;
+
+	/* We do not support protection information for now. */
+	if (ctrl->tctrl->pi_support) {
+		dev_err(ctrl->dev,
+			"Protection information (PI) is not supported\n");
+		ret = -ENOTSUPP;
+		goto out_put_ctrl;
+	}
+
+	/* Allocate our queues, up to the maximum number. */
+	ctrl->nr_queues = min(ctrl->tctrl->subsys->max_qid + 1, max_nr_queues);
+	ret = nvmet_pci_epf_alloc_queues(ctrl);
+	if (ret)
+		goto out_put_ctrl;
+
+	/*
+	 * Allocate the IRQ vectors descriptors. We cannot have more than the
+	 * maximum number of queues.
+	 */
+	ret = nvmet_pci_epf_alloc_irq_vectors(ctrl);
+	if (ret)
+		goto out_free_queues;
+
+	dev_info(ctrl->dev,
+		 "New PCI ctrl \"%s\", %u I/O queues, mdts %u B\n",
+		 ctrl->tctrl->subsys->subsysnqn, ctrl->nr_queues - 1,
+		 ctrl->mdts);
+
+	/* Initialize BAR 0 using the target controller CAP. */
+	nvmet_pci_epf_init_bar(ctrl);
+
+	return 0;
+
+out_free_queues:
+	nvmet_pci_epf_free_queues(ctrl);
+out_put_ctrl:
+	nvmet_ctrl_put(ctrl->tctrl);
+	ctrl->tctrl = NULL;
+out_mempool_exit:
+	mempool_exit(&ctrl->iod_pool);
+	return ret;
+}
+
+static void nvmet_pci_epf_start_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
+{
+	schedule_delayed_work(&ctrl->poll_cc, NVMET_PCI_EPF_CC_POLL_INTERVAL);
+}
+
+static void nvmet_pci_epf_stop_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
+{
+	cancel_delayed_work_sync(&ctrl->poll_cc);
+
+	nvmet_pci_epf_disable_ctrl(ctrl);
+}
+
+static void nvmet_pci_epf_destroy_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
+{
+	if (!ctrl->tctrl)
+		return;
+
+	dev_info(ctrl->dev, "Destroying PCI ctrl \"%s\"\n",
+		 ctrl->tctrl->subsys->subsysnqn);
+
+	nvmet_pci_epf_stop_ctrl(ctrl);
+
+	nvmet_pci_epf_free_queues(ctrl);
+	nvmet_pci_epf_free_irq_vectors(ctrl);
+
+	nvmet_ctrl_put(ctrl->tctrl);
+	ctrl->tctrl = NULL;
+
+	mempool_exit(&ctrl->iod_pool);
+}
+
+static int nvmet_pci_epf_configure_bar(struct nvmet_pci_epf *nvme_epf)
+{
+	struct pci_epf *epf = nvme_epf->epf;
+	const struct pci_epc_features *epc_features = nvme_epf->epc_features;
+	size_t reg_size, reg_bar_size;
+	size_t msix_table_size = 0;
+
+	/*
+	 * The first free BAR will be our register BAR and per NVMe
+	 * specifications, it must be BAR 0.
+	 */
+	if (pci_epc_get_first_free_bar(epc_features) != BAR_0) {
+		dev_err(&epf->dev, "BAR 0 is not free\n");
+		return -ENODEV;
+	}
+
+	if (epc_features->bar[BAR_0].only_64bit)
+		epf->bar[BAR_0].flags |= PCI_BASE_ADDRESS_MEM_TYPE_64;
+
+	/*
+	 * Calculate the size of the register bar: NVMe registers first with
+	 * enough space for the doorbells, followed by the MSI-X table
+	 * if supported.
+	 */
+	reg_size = NVME_REG_DBS + (NVMET_NR_QUEUES * 2 * sizeof(u32));
+	reg_size = ALIGN(reg_size, 8);
+
+	if (epc_features->msix_capable) {
+		size_t pba_size;
+
+		msix_table_size = PCI_MSIX_ENTRY_SIZE * epf->msix_interrupts;
+		nvme_epf->msix_table_offset = reg_size;
+		pba_size = ALIGN(DIV_ROUND_UP(epf->msix_interrupts, 8), 8);
+
+		reg_size += msix_table_size + pba_size;
+	}
+
+	if (epc_features->bar[BAR_0].type == BAR_FIXED) {
+		if (reg_size > epc_features->bar[BAR_0].fixed_size) {
+			dev_err(&epf->dev,
+				"BAR 0 size %llu B too small, need %zu B\n",
+				epc_features->bar[BAR_0].fixed_size,
+				reg_size);
+			return -ENOMEM;
+		}
+		reg_bar_size = epc_features->bar[BAR_0].fixed_size;
+	} else {
+		reg_bar_size = ALIGN(reg_size, max(epc_features->align, 4096));
+	}
+
+	nvme_epf->reg_bar = pci_epf_alloc_space(epf, reg_bar_size, BAR_0,
+						epc_features, PRIMARY_INTERFACE);
+	if (!nvme_epf->reg_bar) {
+		dev_err(&epf->dev, "Failed to allocate BAR 0\n");
+		return -ENOMEM;
+	}
+	memset(nvme_epf->reg_bar, 0, reg_bar_size);
+
+	return 0;
+}
+
+static void nvmet_pci_epf_free_bar(struct nvmet_pci_epf *nvme_epf)
+{
+	struct pci_epf *epf = nvme_epf->epf;
+
+	if (!nvme_epf->reg_bar)
+		return;
+
+	pci_epf_free_space(epf, nvme_epf->reg_bar, BAR_0, PRIMARY_INTERFACE);
+	nvme_epf->reg_bar = NULL;
+}
+
+static void nvmet_pci_epf_clear_bar(struct nvmet_pci_epf *nvme_epf)
+{
+	struct pci_epf *epf = nvme_epf->epf;
+
+	pci_epc_clear_bar(epf->epc, epf->func_no, epf->vfunc_no,
+			  &epf->bar[BAR_0]);
+}
+
+static int nvmet_pci_epf_init_irq(struct nvmet_pci_epf *nvme_epf)
+{
+	const struct pci_epc_features *epc_features = nvme_epf->epc_features;
+	struct pci_epf *epf = nvme_epf->epf;
+	int ret;
+
+	/* Enable MSI-X if supported, otherwise, use MSI. */
+	if (epc_features->msix_capable && epf->msix_interrupts) {
+		ret = pci_epc_set_msix(epf->epc, epf->func_no, epf->vfunc_no,
+				       epf->msix_interrupts, BAR_0,
+				       nvme_epf->msix_table_offset);
+		if (ret) {
+			dev_err(&epf->dev, "Failed to configure MSI-X\n");
+			return ret;
+		}
+
+		nvme_epf->nr_vectors = epf->msix_interrupts;
+		nvme_epf->irq_type = PCI_IRQ_MSIX;
+
+		return 0;
+	}
+
+	if (epc_features->msi_capable && epf->msi_interrupts) {
+		ret = pci_epc_set_msi(epf->epc, epf->func_no, epf->vfunc_no,
+				      epf->msi_interrupts);
+		if (ret) {
+			dev_err(&epf->dev, "Failed to configure MSI\n");
+			return ret;
+		}
+
+		nvme_epf->nr_vectors = epf->msi_interrupts;
+		nvme_epf->irq_type = PCI_IRQ_MSI;
+
+		return 0;
+	}
+
+	/* MSI and MSI-X are not supported: fall back to INTx. */
+	nvme_epf->nr_vectors = 1;
+	nvme_epf->irq_type = PCI_IRQ_INTX;
+
+	return 0;
+}
+
+static int nvmet_pci_epf_epc_init(struct pci_epf *epf)
+{
+	struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf);
+	const struct pci_epc_features *epc_features = nvme_epf->epc_features;
+	struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl;
+	unsigned int max_nr_queues = NVMET_NR_QUEUES;
+	int ret;
+
+	/* For now, do not support virtual functions. */
+	if (epf->vfunc_no > 0) {
+		dev_err(&epf->dev, "Virtual functions are not supported\n");
+		return -EINVAL;
+	}
+
+	/*
+	 * Cap the maximum number of queues we can support on the controller
+	 * with the number of IRQs we can use.
+	 */
+	if (epc_features->msix_capable && epf->msix_interrupts) {
+		dev_info(&epf->dev,
+			 "PCI endpoint controller supports MSI-X, %u vectors\n",
+			 epf->msix_interrupts);
+		max_nr_queues = min(max_nr_queues, epf->msix_interrupts);
+	} else if (epc_features->msi_capable && epf->msi_interrupts) {
+		dev_info(&epf->dev,
+			 "PCI endpoint controller supports MSI, %u vectors\n",
+			 epf->msi_interrupts);
+		max_nr_queues = min(max_nr_queues, epf->msi_interrupts);
+	}
+
+	if (max_nr_queues < 2) {
+		dev_err(&epf->dev, "Invalid maximum number of queues %u\n",
+			max_nr_queues);
+		return -EINVAL;
+	}
+
+	/* Create the target controller. */
+	ret = nvmet_pci_epf_create_ctrl(nvme_epf, max_nr_queues);
+	if (ret) {
+		dev_err(&epf->dev,
+			"Failed to create NVMe PCI target controller (err=%d)\n",
+			ret);
+		return ret;
+	}
+
+	/* Set device ID, class, etc. */
+	epf->header->vendorid = ctrl->tctrl->subsys->vendor_id;
+	epf->header->subsys_vendor_id = ctrl->tctrl->subsys->subsys_vendor_id;
+	ret = pci_epc_write_header(epf->epc, epf->func_no, epf->vfunc_no,
+				   epf->header);
+	if (ret) {
+		dev_err(&epf->dev,
+			"Failed to write configuration header (err=%d)\n", ret);
+		goto out_destroy_ctrl;
+	}
+
+	ret = pci_epc_set_bar(epf->epc, epf->func_no, epf->vfunc_no,
+			      &epf->bar[BAR_0]);
+	if (ret) {
+		dev_err(&epf->dev, "Failed to set BAR 0 (err=%d)\n", ret);
+		goto out_destroy_ctrl;
+	}
+
+	/*
+	 * Enable interrupts and start polling the controller BAR if we do not
+	 * have a link up notifier.
+	 */
+	ret = nvmet_pci_epf_init_irq(nvme_epf);
+	if (ret)
+		goto out_clear_bar;
+
+	if (!epc_features->linkup_notifier) {
+		ctrl->link_up = true;
+		nvmet_pci_epf_start_ctrl(&nvme_epf->ctrl);
+	}
+
+	return 0;
+
+out_clear_bar:
+	nvmet_pci_epf_clear_bar(nvme_epf);
+out_destroy_ctrl:
+	nvmet_pci_epf_destroy_ctrl(&nvme_epf->ctrl);
+	return ret;
+}
+
+static void nvmet_pci_epf_epc_deinit(struct pci_epf *epf)
+{
+	struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf);
+	struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl;
+
+	ctrl->link_up = false;
+	nvmet_pci_epf_destroy_ctrl(ctrl);
+
+	nvmet_pci_epf_deinit_dma(nvme_epf);
+	nvmet_pci_epf_clear_bar(nvme_epf);
+}
+
+static int nvmet_pci_epf_link_up(struct pci_epf *epf)
+{
+	struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf);
+	struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl;
+
+	ctrl->link_up = true;
+	nvmet_pci_epf_start_ctrl(ctrl);
+
+	return 0;
+}
+
+static int nvmet_pci_epf_link_down(struct pci_epf *epf)
+{
+	struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf);
+	struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl;
+
+	ctrl->link_up = false;
+	nvmet_pci_epf_stop_ctrl(ctrl);
+
+	return 0;
+}
+
+static const struct pci_epc_event_ops nvmet_pci_epf_event_ops = {
+	.epc_init = nvmet_pci_epf_epc_init,
+	.epc_deinit = nvmet_pci_epf_epc_deinit,
+	.link_up = nvmet_pci_epf_link_up,
+	.link_down = nvmet_pci_epf_link_down,
+};
+
+static int nvmet_pci_epf_bind(struct pci_epf *epf)
+{
+	struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf);
+	const struct pci_epc_features *epc_features;
+	struct pci_epc *epc = epf->epc;
+	int ret;
+
+	if (WARN_ON_ONCE(!epc))
+		return -EINVAL;
+
+	epc_features = pci_epc_get_features(epc, epf->func_no, epf->vfunc_no);
+	if (!epc_features) {
+		dev_err(&epf->dev, "epc_features not implemented\n");
+		return -EOPNOTSUPP;
+	}
+	nvme_epf->epc_features = epc_features;
+
+	ret = nvmet_pci_epf_configure_bar(nvme_epf);
+	if (ret)
+		return ret;
+
+	nvmet_pci_epf_init_dma(nvme_epf);
+
+	return 0;
+}
+
+static void nvmet_pci_epf_unbind(struct pci_epf *epf)
+{
+	struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf);
+	struct pci_epc *epc = epf->epc;
+
+	nvmet_pci_epf_destroy_ctrl(&nvme_epf->ctrl);
+
+	if (epc->init_complete) {
+		nvmet_pci_epf_deinit_dma(nvme_epf);
+		nvmet_pci_epf_clear_bar(nvme_epf);
+	}
+
+	nvmet_pci_epf_free_bar(nvme_epf);
+}
+
+static struct pci_epf_header nvme_epf_pci_header = {
+	.vendorid	= PCI_ANY_ID,
+	.deviceid	= PCI_ANY_ID,
+	.progif_code	= 0x02, /* NVM Express */
+	.baseclass_code = PCI_BASE_CLASS_STORAGE,
+	.subclass_code	= 0x08, /* Non-Volatile Memory controller */
+	.interrupt_pin	= PCI_INTERRUPT_INTA,
+};
+
+static int nvmet_pci_epf_probe(struct pci_epf *epf,
+			       const struct pci_epf_device_id *id)
+{
+	struct nvmet_pci_epf *nvme_epf;
+	int ret;
+
+	nvme_epf = devm_kzalloc(&epf->dev, sizeof(*nvme_epf), GFP_KERNEL);
+	if (!nvme_epf)
+		return -ENOMEM;
+
+	ret = devm_mutex_init(&epf->dev, &nvme_epf->mmio_lock);
+	if (ret)
+		return ret;
+
+	nvme_epf->epf = epf;
+	nvme_epf->mdts_kb = NVMET_PCI_EPF_MDTS_KB;
+
+	epf->event_ops = &nvmet_pci_epf_event_ops;
+	epf->header = &nvme_epf_pci_header;
+	epf_set_drvdata(epf, nvme_epf);
+
+	return 0;
+}
+
+#define to_nvme_epf(epf_group)	\
+	container_of(epf_group, struct nvmet_pci_epf, group)
+
+static ssize_t nvmet_pci_epf_portid_show(struct config_item *item, char *page)
+{
+	struct config_group *group = to_config_group(item);
+	struct nvmet_pci_epf *nvme_epf = to_nvme_epf(group);
+
+	return sysfs_emit(page, "%u\n", le16_to_cpu(nvme_epf->portid));
+}
+
+static ssize_t nvmet_pci_epf_portid_store(struct config_item *item,
+					  const char *page, size_t len)
+{
+	struct config_group *group = to_config_group(item);
+	struct nvmet_pci_epf *nvme_epf = to_nvme_epf(group);
+	u16 portid;
+
+	/* Do not allow setting this when the function is already started. */
+	if (nvme_epf->ctrl.tctrl)
+		return -EBUSY;
+
+	if (!len)
+		return -EINVAL;
+
+	if (kstrtou16(page, 0, &portid))
+		return -EINVAL;
+
+	nvme_epf->portid = cpu_to_le16(portid);
+
+	return len;
+}
+
+CONFIGFS_ATTR(nvmet_pci_epf_, portid);
+
+static ssize_t nvmet_pci_epf_subsysnqn_show(struct config_item *item,
+					    char *page)
+{
+	struct config_group *group = to_config_group(item);
+	struct nvmet_pci_epf *nvme_epf = to_nvme_epf(group);
+
+	return sysfs_emit(page, "%s\n", nvme_epf->subsysnqn);
+}
+
+static ssize_t nvmet_pci_epf_subsysnqn_store(struct config_item *item,
+					     const char *page, size_t len)
+{
+	struct config_group *group = to_config_group(item);
+	struct nvmet_pci_epf *nvme_epf = to_nvme_epf(group);
+
+	/* Do not allow setting this when the function is already started. */
+	if (nvme_epf->ctrl.tctrl)
+		return -EBUSY;
+
+	if (!len)
+		return -EINVAL;
+
+	strscpy(nvme_epf->subsysnqn, page, len);
+
+	return len;
+}
+
+CONFIGFS_ATTR(nvmet_pci_epf_, subsysnqn);
+
+static ssize_t nvmet_pci_epf_mdts_kb_show(struct config_item *item, char *page)
+{
+	struct config_group *group = to_config_group(item);
+	struct nvmet_pci_epf *nvme_epf = to_nvme_epf(group);
+
+	return sysfs_emit(page, "%u\n", nvme_epf->mdts_kb);
+}
+
+static ssize_t nvmet_pci_epf_mdts_kb_store(struct config_item *item,
+					   const char *page, size_t len)
+{
+	struct config_group *group = to_config_group(item);
+	struct nvmet_pci_epf *nvme_epf = to_nvme_epf(group);
+	unsigned long mdts_kb;
+	int ret;
+
+	if (nvme_epf->ctrl.tctrl)
+		return -EBUSY;
+
+	ret = kstrtoul(page, 0, &mdts_kb);
+	if (ret)
+		return ret;
+	if (!mdts_kb)
+		mdts_kb = NVMET_PCI_EPF_MDTS_KB;
+	else if (mdts_kb > NVMET_PCI_EPF_MAX_MDTS_KB)
+		mdts_kb = NVMET_PCI_EPF_MAX_MDTS_KB;
+
+	if (!is_power_of_2(mdts_kb))
+		return -EINVAL;
+
+	nvme_epf->mdts_kb = mdts_kb;
+
+	return len;
+}
+
+CONFIGFS_ATTR(nvmet_pci_epf_, mdts_kb);
+
+static struct configfs_attribute *nvmet_pci_epf_attrs[] = {
+	&nvmet_pci_epf_attr_portid,
+	&nvmet_pci_epf_attr_subsysnqn,
+	&nvmet_pci_epf_attr_mdts_kb,
+	NULL,
+};
+
+static const struct config_item_type nvmet_pci_epf_group_type = {
+	.ct_attrs	= nvmet_pci_epf_attrs,
+	.ct_owner	= THIS_MODULE,
+};
+
+static struct config_group *nvmet_pci_epf_add_cfs(struct pci_epf *epf,
+						  struct config_group *group)
+{
+	struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf);
+
+	config_group_init_type_name(&nvme_epf->group, "nvme",
+				    &nvmet_pci_epf_group_type);
+
+	return &nvme_epf->group;
+}
+
+static const struct pci_epf_device_id nvmet_pci_epf_ids[] = {
+	{ .name = "nvmet_pci_epf" },
+	{},
+};
+
+static struct pci_epf_ops nvmet_pci_epf_ops = {
+	.bind	= nvmet_pci_epf_bind,
+	.unbind	= nvmet_pci_epf_unbind,
+	.add_cfs = nvmet_pci_epf_add_cfs,
+};
+
+static struct pci_epf_driver nvmet_pci_epf_driver = {
+	.driver.name	= "nvmet_pci_epf",
+	.probe		= nvmet_pci_epf_probe,
+	.id_table	= nvmet_pci_epf_ids,
+	.ops		= &nvmet_pci_epf_ops,
+	.owner		= THIS_MODULE,
+};
+
+static int __init nvmet_pci_epf_init_module(void)
+{
+	int ret;
+
+	ret = pci_epf_register_driver(&nvmet_pci_epf_driver);
+	if (ret)
+		return ret;
+
+	ret = nvmet_register_transport(&nvmet_pci_epf_fabrics_ops);
+	if (ret) {
+		pci_epf_unregister_driver(&nvmet_pci_epf_driver);
+		return ret;
+	}
+
+	return 0;
+}
+
+static void __exit nvmet_pci_epf_cleanup_module(void)
+{
+	nvmet_unregister_transport(&nvmet_pci_epf_fabrics_ops);
+	pci_epf_unregister_driver(&nvmet_pci_epf_driver);
+}
+
+module_init(nvmet_pci_epf_init_module);
+module_exit(nvmet_pci_epf_cleanup_module);
+
+MODULE_DESCRIPTION("NVMe PCI Endpoint Function target driver");
+MODULE_AUTHOR("Damien Le Moal <[e-mail]>");
+MODULE_LICENSE("GPL");

drivers/nvme/target/zns.c

···
 	for_each_sg(req->sg, sg, req->sg_cnt, sg_cnt) {
 		unsigned int len = sg->length;
 
-		if (bio_add_pc_page(bdev_get_queue(bio->bi_bdev), bio,
-				sg_page(sg), len, sg->offset) != len) {
+		if (bio_add_page(bio, sg_page(sg), len, sg->offset) != len) {
 			status = NVME_SC_INTERNAL;
 			goto out_put_bio;
 		}

drivers/pci/pci-driver.c

···
 		iommu_device_unuse_default_domain(dev);
 }
 
+/*
+ * pci_device_irq_get_affinity - get IRQ affinity mask for device
+ * @dev: ptr to dev structure
+ * @irq_vec: interrupt vector number
+ *
+ * Return the CPU affinity mask for @dev and @irq_vec.
+ */
+static const struct cpumask *pci_device_irq_get_affinity(struct device *dev,
+					unsigned int irq_vec)
+{
+	return pci_irq_get_affinity(to_pci_dev(dev), irq_vec);
+}
+
 const struct bus_type pci_bus_type = {
 	.name		= "pci",
 	.match		= pci_bus_match,
···
 	.probe		= pci_device_probe,
 	.remove		= pci_device_remove,
 	.shutdown	= pci_device_shutdown,
+	.irq_get_affinity = pci_device_irq_get_affinity,
 	.dev_groups	= pci_dev_groups,
 	.bus_groups	= pci_bus_groups,
 	.drv_groups	= pci_drv_groups,

drivers/s390/block/dasd_genhd.c

···
 	block->tag_set.cmd_size = sizeof(struct dasd_ccw_req);
 	block->tag_set.nr_hw_queues = nr_hw_queues;
 	block->tag_set.queue_depth = queue_depth;
-	block->tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
 	block->tag_set.numa_node = NUMA_NO_NODE;
 	rc = blk_mq_alloc_tag_set(&block->tag_set);
 	if (rc)

drivers/s390/block/scm_blk.c

···
 	bdev->tag_set.cmd_size = sizeof(blk_status_t);
 	bdev->tag_set.nr_hw_queues = nr_requests;
 	bdev->tag_set.queue_depth = nr_requests_per_io * nr_requests;
-	bdev->tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
 	bdev->tag_set.numa_node = NUMA_NO_NODE;
 
 	ret = blk_mq_alloc_tag_set(&bdev->tag_set);

drivers/scsi/fnic/fnic_main.c

···
 #include <linux/spinlock.h>
 #include <linux/workqueue.h>
 #include <linux/if_ether.h>
-#include <linux/blk-mq-pci.h>
 #include <scsi/fc/fc_fip.h>
 #include <scsi/scsi_host.h>
 #include <scsi/scsi_transport.h>
···
 		return;
 	}
 
-	blk_mq_pci_map_queues(qmap, l_pdev, FNIC_PCI_OFFSET);
+	blk_mq_map_hw_queues(qmap, &l_pdev->dev, FNIC_PCI_OFFSET);
 }
 
 static int fnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)

drivers/scsi/hisi_sas/hisi_sas.h

···
 
 #include <linux/acpi.h>
 #include <linux/blk-mq.h>
-#include <linux/blk-mq-pci.h>
 #include <linux/clk.h>
 #include <linux/debugfs.h>
 #include <linux/dmapool.h>

drivers/scsi/hisi_sas/hisi_sas_v3_hw.c

···
 		if (i == HCTX_TYPE_POLL)
 			blk_mq_map_queues(qmap);
 		else
-			blk_mq_pci_map_queues(qmap, hisi_hba->pci_dev,
-					      BASE_VECTORS_V3_HW);
+			blk_mq_map_hw_queues(qmap, hisi_hba->dev,
+					     BASE_VECTORS_V3_HW);
 		qoff += qmap->nr_queues;
 	}
 }
···
 	.slave_alloc		= hisi_sas_slave_alloc,
 	.shost_groups		= host_v3_hw_groups,
 	.sdev_groups		= sdev_groups_v3_hw,
-	.tag_alloc_policy	= BLK_TAG_ALLOC_RR,
+	.tag_alloc_policy_rr	= true,
 	.host_reset             = hisi_sas_host_reset,
 	.host_tagset		= 1,
 	.mq_poll		= queue_complete_v3_hw,

drivers/scsi/megaraid/megaraid_sas_base.c

···
 #include <linux/poll.h>
 #include <linux/vmalloc.h>
 #include <linux/irq_poll.h>
-#include <linux/blk-mq-pci.h>
 
 #include <scsi/scsi.h>
 #include <scsi/scsi_cmnd.h>
···
 	map = &shost->tag_set.map[HCTX_TYPE_DEFAULT];
 	map->nr_queues = instance->msix_vectors - offset;
 	map->queue_offset = 0;
-	blk_mq_pci_map_queues(map, instance->pdev, offset);
+	blk_mq_map_hw_queues(map, &instance->pdev->dev, offset);
 	qoff += map->nr_queues;
 	offset += map->nr_queues;
 

drivers/scsi/mpi3mr/mpi3mr.h

···
 
 #include <linux/blkdev.h>
 #include <linux/blk-mq.h>
-#include <linux/blk-mq-pci.h>
 #include <linux/delay.h>
 #include <linux/dmapool.h>
 #include <linux/errno.h>

drivers/scsi/mpi3mr/mpi3mr_os.c

···
 		 */
 		map->queue_offset = qoff;
 		if (i != HCTX_TYPE_POLL)
-			blk_mq_pci_map_queues(map, mrioc->pdev, offset);
+			blk_mq_map_hw_queues(map, &mrioc->pdev->dev, offset);
 		else
 			blk_mq_map_queues(map);
 

drivers/scsi/mpt3sas/mpt3sas_scsih.c

···
 #include <linux/pci.h>
 #include <linux/interrupt.h>
 #include <linux/raid_class.h>
-#include <linux/blk-mq-pci.h>
 #include <linux/unaligned.h>
 
 #include "mpt3sas_base.h"
···
 		 */
 		map->queue_offset = qoff;
 		if (i != HCTX_TYPE_POLL)
-			blk_mq_pci_map_queues(map, ioc->pdev, offset);
+			blk_mq_map_hw_queues(map, &ioc->pdev->dev, offset);
 		else
 			blk_mq_map_queues(map);
 

drivers/scsi/pm8001/pm8001_init.c

···
 	struct blk_mq_queue_map *qmap = &shost->tag_set.map[HCTX_TYPE_DEFAULT];
 
 	if (pm8001_ha->number_of_intr > 1) {
-		blk_mq_pci_map_queues(qmap, pm8001_ha->pdev, 1);
+		blk_mq_map_hw_queues(qmap, &pm8001_ha->pdev->dev, 1);
 		return;
 	}
 

drivers/scsi/pm8001/pm8001_sas.h

···
 #include <scsi/sas_ata.h>
 #include <linux/atomic.h>
 #include <linux/blk-mq.h>
-#include <linux/blk-mq-pci.h>
 #include "pm8001_defs.h"
 
 #define DRV_NAME		"pm80xx"

drivers/scsi/qla2xxx/qla_nvme.c

···
 #include <linux/delay.h>
 #include <linux/nvme.h>
 #include <linux/nvme-fc.h>
-#include <linux/blk-mq-pci.h>
 #include <linux/blk-mq.h>
 
 static struct nvme_fc_port_template qla_nvme_fc_transport;
···
 {
 	struct scsi_qla_host *vha = lport->private;
 
-	blk_mq_pci_map_queues(map, vha->hw->pdev, vha->irq_offset);
+	blk_mq_map_hw_queues(map, &vha->hw->pdev->dev, vha->irq_offset);
 }
 
 static void qla_nvme_localport_delete(struct nvme_fc_local_port *lport)

drivers/scsi/qla2xxx/qla_os.c

···
 #include <linux/mutex.h>
 #include <linux/kobject.h>
 #include <linux/slab.h>
-#include <linux/blk-mq-pci.h>
 #include <linux/refcount.h>
 #include <linux/crash_dump.h>
 #include <linux/trace_events.h>
···
 	if (USER_CTRL_IRQ(vha->hw) || !vha->hw->mqiobase)
 		blk_mq_map_queues(qmap);
 	else
-		blk_mq_pci_map_queues(qmap, vha->hw->pdev, vha->irq_offset);
+		blk_mq_map_hw_queues(qmap, &vha->hw->pdev->dev,
+				       vha->irq_offset);
 }
 
 struct scsi_host_template qla2xxx_driver_template = {

drivers/scsi/scsi_lib.c

···
 	tag_set->queue_depth = shost->can_queue;
 	tag_set->cmd_size = cmd_size;
 	tag_set->numa_node = dev_to_node(shost->dma_dev);
-	tag_set->flags = BLK_MQ_F_SHOULD_MERGE;
-	tag_set->flags |=
-		BLK_ALLOC_POLICY_TO_MQ_FLAG(shost->hostt->tag_alloc_policy);
+	if (shost->hostt->tag_alloc_policy_rr)
+		tag_set->flags |= BLK_MQ_F_TAG_RR;
 	if (shost->queuecommand_may_block)
 		tag_set->flags |= BLK_MQ_F_BLOCKING;
 	tag_set->driver_data = shost;

drivers/scsi/sd.c

···
 
 		lim = queue_limits_start_update(sdkp->disk->queue);
 		sd_set_flush_flag(sdkp, &lim);
-		blk_mq_freeze_queue(sdkp->disk->queue);
-		ret = queue_limits_commit_update(sdkp->disk->queue, &lim);
-		blk_mq_unfreeze_queue(sdkp->disk->queue);
+		ret = queue_limits_commit_update_frozen(sdkp->disk->queue,
+				&lim);
 		if (ret)
 			return ret;
 		return count;
···
 
 	lim = queue_limits_start_update(sdkp->disk->queue);
 	sd_config_discard(sdkp, &lim, mode);
-	blk_mq_freeze_queue(sdkp->disk->queue);
-	err = queue_limits_commit_update(sdkp->disk->queue, &lim);
-	blk_mq_unfreeze_queue(sdkp->disk->queue);
+	err = queue_limits_commit_update_frozen(sdkp->disk->queue, &lim);
 	if (err)
 		return err;
 	return count;
···
 
 	lim = queue_limits_start_update(sdkp->disk->queue);
 	sd_config_write_same(sdkp, &lim);
-	blk_mq_freeze_queue(sdkp->disk->queue);
-	err = queue_limits_commit_update(sdkp->disk->queue, &lim);
-	blk_mq_unfreeze_queue(sdkp->disk->queue);
+	err = queue_limits_commit_update_frozen(sdkp->disk->queue, &lim);
 	if (err)
 		return err;
 	return count;
···
 	lim->atomic_write_hw_boundary = 0;
 	lim->atomic_write_hw_unit_min = unit_min * logical_block_size;
 	lim->atomic_write_hw_unit_max = unit_max * logical_block_size;
+	lim->features |= BLK_FEAT_ATOMIC_WRITES;
 }
 
 static blk_status_t sd_setup_write_same16_cmnd(struct scsi_cmnd *cmd,
···
 	sd_config_write_same(sdkp, &lim);
 	kfree(buffer);
 
-	blk_mq_freeze_queue(sdkp->disk->queue);
-	err = queue_limits_commit_update(sdkp->disk->queue, &lim);
-	blk_mq_unfreeze_queue(sdkp->disk->queue);
+	err = queue_limits_commit_update_frozen(sdkp->disk->queue, &lim);
 	if (err)
 		return err;
 

drivers/scsi/smartpqi/smartpqi_init.c

···
 #include <linux/bcd.h>
 #include <linux/reboot.h>
 #include <linux/cciss_ioctl.h>
-#include <linux/blk-mq-pci.h>
 #include <scsi/scsi_host.h>
 #include <scsi/scsi_cmnd.h>
 #include <scsi/scsi_device.h>
···
 	struct pqi_ctrl_info *ctrl_info = shost_to_hba(shost);
 
 	if (!ctrl_info->disable_managed_interrupts)
-		return blk_mq_pci_map_queues(&shost->tag_set.map[HCTX_TYPE_DEFAULT],
-			      ctrl_info->pci_dev, 0);
+		blk_mq_map_hw_queues(&shost->tag_set.map[HCTX_TYPE_DEFAULT],
+				       &ctrl_info->pci_dev->dev, 0);
 	else
-		return blk_mq_map_queues(&shost->tag_set.map[HCTX_TYPE_DEFAULT]);
+		blk_mq_map_queues(&shost->tag_set.map[HCTX_TYPE_DEFAULT]);
 }
 
 static inline bool pqi_is_tape_changer_device(struct pqi_scsi_dev *device)

drivers/scsi/sr.c

···
 
 	lim = queue_limits_start_update(q);
 	lim.logical_block_size = sector_size;
-	blk_mq_freeze_queue(q);
-	err = queue_limits_commit_update(q, &lim);
-	blk_mq_unfreeze_queue(q);
-	return err;
+	return queue_limits_commit_update_frozen(q, &lim);
 }
 
 static int get_capabilities(struct scsi_cd *cd)

drivers/scsi/virtio_scsi.c

···
 #include <scsi/scsi_tcq.h>
 #include <scsi/scsi_devinfo.h>
 #include <linux/seqlock.h>
-#include <linux/blk-mq-virtio.h>
 
 #include "sd.h"
 
···
 		if (i == HCTX_TYPE_POLL)
 			blk_mq_map_queues(map);
 		else
-			blk_mq_virtio_map_queues(map, vscsi->vdev, 2);
+			blk_mq_map_hw_queues(map, &vscsi->vdev->dev, 2);
 	}
 }
 

drivers/target/target_core_pscsi.c

···
 pscsi_map_sg(struct se_cmd *cmd, struct scatterlist *sgl, u32 sgl_nents,
 		struct request *req)
 {
-	struct pscsi_dev_virt *pdv = PSCSI_DEV(cmd->se_dev);
 	struct bio *bio = NULL;
 	struct page *page;
 	struct scatterlist *sg;
···
 					(rw) ? "rw" : "r", nr_vecs);
 			}
 
-			pr_debug("PSCSI: Calling bio_add_pc_page() i: %d"
+			pr_debug("PSCSI: Calling bio_add_page() i: %d"
 				" bio: %p page: %p len: %d off: %d\n", i, bio,
 				page, len, off);
 
-			rc = bio_add_pc_page(pdv->pdv_sd->request_queue,
-					bio, page, bytes, off);
+			rc = bio_add_page(bio, page, bytes, off);
 			pr_debug("PSCSI: bio->bi_vcnt: %d nr_vecs: %d\n",
 				bio_segments(bio), nr_vecs);
 			if (rc != bytes) {

drivers/ufs/core/ufshcd.c

···
 		.nr_hw_queues	= 1,
 		.queue_depth	= hba->nutmrs,
 		.ops		= &ufshcd_tmf_ops,
-		.flags		= BLK_MQ_F_NO_SCHED,
 	};
 	err = blk_mq_alloc_tag_set(&hba->tmf_tag_set);
 	if (err < 0)

drivers/usb/storage/scsiglue.c

···
 	if (sscanf(buf, "%hu", &ms) <= 0)
 		return -EINVAL;
 
-	blk_mq_freeze_queue(sdev->request_queue);
 	lim = queue_limits_start_update(sdev->request_queue);
 	lim.max_hw_sectors = ms;
-	ret = queue_limits_commit_update(sdev->request_queue, &lim);
-	blk_mq_unfreeze_queue(sdev->request_queue);
-
+	ret = queue_limits_commit_update_frozen(sdev->request_queue, &lim);
 	if (ret)
 		return ret;
 	return count;

drivers/virtio/virtio.c

···
 	of_node_put(dev->dev.of_node);
 }
 
+/*
+ * virtio_irq_get_affinity - get IRQ affinity mask for device
+ * @_d: ptr to dev structure
+ * @irq_vec: interrupt vector number
+ *
+ * Return the CPU affinity mask for @_d and @irq_vec.
+ */
+static const struct cpumask *virtio_irq_get_affinity(struct device *_d,
+						     unsigned int irq_vec)
+{
+	struct virtio_device *dev = dev_to_virtio(_d);
+
+	if (!dev->config->get_vq_affinity)
+		return NULL;
+
+	return dev->config->get_vq_affinity(dev, irq_vec);
+}
+
 static const struct bus_type virtio_bus = {
 	.name  = "virtio",
 	.match = virtio_dev_match,
···
 	.uevent = virtio_uevent,
 	.probe = virtio_dev_probe,
 	.remove = virtio_dev_remove,
+	.irq_get_affinity = virtio_irq_get_affinity,
 };
 
 int __register_virtio_driver(struct virtio_driver *driver, struct module *owner)

fs/bcachefs/move.c

···
 	io->write_sectors	= k.k->size;
 
 	bio_init(&io->write.op.wbio.bio, NULL, io->bi_inline_vecs, pages, 0);
-	bio_set_prio(&io->write.op.wbio.bio,
-		     IOPRIO_PRIO_VALUE(IOPRIO_CLASS_IDLE, 0));
+	io->write.op.wbio.bio.bi_ioprio =
+		     IOPRIO_PRIO_VALUE(IOPRIO_CLASS_IDLE, 0);
 
 	if (bch2_bio_alloc_pages(&io->write.op.wbio.bio, sectors << 9,
 				 GFP_KERNEL))
···
 	io->rbio.opts		= io_opts;
 	bio_init(&io->rbio.bio, NULL, io->bi_inline_vecs, pages, 0);
 	io->rbio.bio.bi_vcnt = pages;
-	bio_set_prio(&io->rbio.bio, IOPRIO_PRIO_VALUE(IOPRIO_CLASS_IDLE, 0));
+	io->rbio.bio.bi_ioprio = IOPRIO_PRIO_VALUE(IOPRIO_CLASS_IDLE, 0);
 	io->rbio.bio.bi_iter.bi_size = sectors << 9;
 
 	io->rbio.bio.bi_opf		= REQ_OP_READ;

include/linux/bio.h

···
 	return min(nr_segs, BIO_MAX_VECS);
 }
 
-#define bio_prio(bio)			(bio)->bi_ioprio
-#define bio_set_prio(bio, prio)		((bio)->bi_ioprio = prio)
-
 #define bio_iter_iovec(bio, iter)				\
 	bvec_iter_bvec((bio)->bi_io_vec, (iter))
 
···
 			      unsigned off);
 bool __must_check bio_add_folio(struct bio *bio, struct folio *folio,
 				size_t len, size_t off);
-extern int bio_add_pc_page(struct request_queue *, struct bio *, struct page *,
-			   unsigned int, unsigned int);
 void __bio_add_page(struct bio *bio, struct page *page,
 		unsigned int len, unsigned int off);
 void bio_add_folio_nofail(struct bio *bio, struct folio *folio, size_t len,

include/linux/blk-mq-pci.h

···
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _LINUX_BLK_MQ_PCI_H
-#define _LINUX_BLK_MQ_PCI_H
-
-struct blk_mq_queue_map;
-struct pci_dev;
-
-void blk_mq_pci_map_queues(struct blk_mq_queue_map *qmap, struct pci_dev *pdev,
-			   int offset);
-
-#endif /* _LINUX_BLK_MQ_PCI_H */

include/linux/blk-mq-virtio.h

···
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _LINUX_BLK_MQ_VIRTIO_H
-#define _LINUX_BLK_MQ_VIRTIO_H
-
-struct blk_mq_queue_map;
-struct virtio_device;
-
-void blk_mq_virtio_map_queues(struct blk_mq_queue_map *qmap,
-		struct virtio_device *vdev, int first_vec);
-
-#endif /* _LINUX_BLK_MQ_VIRTIO_H */

include/linux/blk-mq.h

···
 	BLK_EH_RESET_TIMER,
 };
 
-/* Keep alloc_policy_name[] in sync with the definitions below */
-enum {
-	BLK_TAG_ALLOC_FIFO,	/* allocate starting from 0 */
-	BLK_TAG_ALLOC_RR,	/* allocate starting from last allocated tag */
-	BLK_TAG_ALLOC_MAX
-};
-
 /**
  * struct blk_mq_hw_ctx - State for a hardware queue facing the hardware
  * block device
···
 
 /* Keep hctx_flag_name[] in sync with the definitions below */
 enum {
-	BLK_MQ_F_SHOULD_MERGE	= 1 << 0,
 	BLK_MQ_F_TAG_QUEUE_SHARED = 1 << 1,
 	/*
 	 * Set when this device requires underlying blk-mq device for
···
 	BLK_MQ_F_STACKING	= 1 << 2,
 	BLK_MQ_F_TAG_HCTX_SHARED = 1 << 3,
 	BLK_MQ_F_BLOCKING	= 1 << 4,
-	/* Do not allow an I/O scheduler to be configured. */
-	BLK_MQ_F_NO_SCHED	= 1 << 5,
+
+	/*
+	 * Alloc tags on a round-robin base instead of the first available one.
+	 */
+	BLK_MQ_F_TAG_RR		= 1 << 5,
 
 	/*
 	 * Select 'none' during queue registration in case of a single hwq
 	 * or shared hwqs instead of 'mq-deadline'.
 	 */
 	BLK_MQ_F_NO_SCHED_BY_DEFAULT	= 1 << 6,
-	BLK_MQ_F_ALLOC_POLICY_START_BIT = 7,
-	BLK_MQ_F_ALLOC_POLICY_BITS = 1,
+
+	BLK_MQ_F_MAX = 1 << 7,
 };
-#define BLK_MQ_FLAG_TO_ALLOC_POLICY(flags) \
-	((flags >> BLK_MQ_F_ALLOC_POLICY_START_BIT) & \
-		((1 << BLK_MQ_F_ALLOC_POLICY_BITS) - 1))
-#define BLK_ALLOC_POLICY_TO_MQ_FLAG(policy) \
-	((policy & ((1 << BLK_MQ_F_ALLOC_POLICY_BITS) - 1)) \
-		<< BLK_MQ_F_ALLOC_POLICY_START_BIT)
 
 #define BLK_MQ_MAX_DEPTH	(10240)
 #define BLK_MQ_NO_HCTX_IDX	(-1U)
···
 void blk_freeze_queue_start_non_owner(struct request_queue *q);
 
 void blk_mq_map_queues(struct blk_mq_queue_map *qmap);
+void blk_mq_map_hw_queues(struct blk_mq_queue_map *qmap,
+			  struct device *dev, unsigned int offset);
 void blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set, int nr_hw_queues);
 
 void blk_mq_quiesce_queue_nowait(struct request_queue *q);
···
 {
 	if (rq->q->mq_ops->cleanup_rq)
 		rq->q->mq_ops->cleanup_rq(rq);
-}
-
-static inline void blk_rq_bio_prep(struct request *rq, struct bio *bio,
-		unsigned int nr_segs)
-{
-	rq->nr_phys_segments = nr_segs;
-	rq->__data_len = bio->bi_iter.bi_size;
-	rq->bio = rq->biotail = bio;
 }
 
 void blk_mq_hctx_set_fq_lock_class(struct blk_mq_hw_ctx *hctx,

include/linux/blkdev.h

···
 #define BLK_FEAT_RAID_PARTIAL_STRIPES_EXPENSIVE \
 	((__force blk_features_t)(1u << 15))
 
-/* stacked device can/does support atomic writes */
-#define BLK_FEAT_ATOMIC_WRITES_STACKED \
+/* atomic writes enabled */
+#define BLK_FEAT_ATOMIC_WRITES \
 	((__force blk_features_t)(1u << 16))
 
 /*
···
 #ifdef CONFIG_LOCKDEP
 	struct task_struct	*mq_freeze_owner;
 	int			mq_freeze_owner_depth;
+	/*
+	 * Records disk & queue state in current context, used in unfreeze
+	 * queue
+	 */
+	bool			mq_freeze_disk_dead;
+	bool			mq_freeze_queue_dying;
 #endif
 	wait_queue_head_t	mq_freeze_wq;
 	/*
···
  * the caller can modify.  The caller must call queue_limits_commit_update()
  * to finish the update.
  *
- * Context: process context.  The caller must have frozen the queue or ensured
- * that there is outstanding I/O by other means.
+ * Context: process context.
  */
 static inline struct queue_limits
 queue_limits_start_update(struct request_queue *q)
···
 	mutex_lock(&q->limits_lock);
 	return q->limits;
 }
+int queue_limits_commit_update_frozen(struct request_queue *q,
+		struct queue_limits *lim);
 int queue_limits_commit_update(struct request_queue *q,
 		struct queue_limits *lim);
 int queue_limits_set(struct request_queue *q, struct queue_limits *lim);
···
 	void (*complete)(struct io_comp_batch *);
 };
 
+static inline bool blk_atomic_write_start_sect_aligned(sector_t sector,
+						struct queue_limits *limits)
+{
+	unsigned int alignment = max(limits->atomic_write_hw_unit_min,
+				limits->atomic_write_hw_boundary);
+
+	return IS_ALIGNED(sector, alignment >> SECTOR_SHIFT);
+}
+
 static inline bool bdev_can_atomic_write(struct block_device *bdev)
 {
 	struct request_queue *bd_queue = bdev->bd_queue;
···
 	if (!limits->atomic_write_unit_min)
 		return false;
 
-	if (bdev_is_partition(bdev)) {
-		sector_t bd_start_sect = bdev->bd_start_sect;
-		unsigned int alignment =
-			max(limits->atomic_write_unit_min,
-			    limits->atomic_write_hw_boundary);
-
-		if (!IS_ALIGNED(bd_start_sect, alignment >> SECTOR_SHIFT))
-			return false;
-	}
+	if (bdev_is_partition(bdev))
+		return blk_atomic_write_start_sect_aligned(bdev->bd_start_sect,
+							limits);
 
 	return true;
 }

include/linux/bvec.h

···
  */
 static inline phys_addr_t bvec_phys(const struct bio_vec *bvec)
 {
-	/*
-	 * Note this open codes page_to_phys because page_to_phys is defined in
-	 * <asm/io.h>, which we don't want to pull in here.  If it ever moves to
-	 * a sensible place we should start using it.
-	 */
-	return PFN_PHYS(page_to_pfn(bvec->bv_page)) + bvec->bv_offset;
+	return page_to_phys(bvec->bv_page) + bvec->bv_offset;
 }
 
 #endif /* __LINUX_BVEC_H */

include/linux/device/bus.h

···
  *		will never get called until they do.
  * @remove:	Called when a device removed from this bus.
  * @shutdown:	Called at shut-down time to quiesce the device.
+ * @irq_get_affinity:	Get IRQ affinity mask for the device on this bus.
  *
  * @online:	Called to put the device back online (after offlining it).
  * @offline:	Called to put the device offline for hot-removal. May fail.
···
 	void (*sync_state)(struct device *dev);
 	void (*remove)(struct device *dev);
 	void (*shutdown)(struct device *dev);
+	const struct cpumask *(*irq_get_affinity)(struct device *dev,
+			unsigned int irq_vec);
 
 	int (*online)(struct device *dev);
 	int (*offline)(struct device *dev);

include/linux/libata.h

···
 #define ATA_SUBBASE_SHT(drv_name)				\
 	__ATA_BASE_SHT(drv_name),				\
 	.can_queue		= ATA_DEF_QUEUE,		\
-	.tag_alloc_policy	= BLK_TAG_ALLOC_RR,		\
+	.tag_alloc_policy_rr	= true,				\
 	.device_configure	= ata_scsi_device_configure
 
 #define ATA_SUBBASE_SHT_QD(drv_name, drv_qd)			\
 	__ATA_BASE_SHT(drv_name),				\
 	.can_queue		= drv_qd,			\
-	.tag_alloc_policy	= BLK_TAG_ALLOC_RR,		\
+	.tag_alloc_policy_rr	= true,				\
 	.device_configure	= ata_scsi_device_configure
 
 #define ATA_BASE_SHT(drv_name)					\

include/linux/nvme.h

···
 
 /* Transport Type codes for Discovery Log Page entry TRTYPE field */
 enum {
+	NVMF_TRTYPE_PCI		= 0,	/* PCI */
 	NVMF_TRTYPE_RDMA	= 1,	/* RDMA */
 	NVMF_TRTYPE_FC		= 2,	/* Fibre Channel */
 	NVMF_TRTYPE_TCP		= 3,	/* TCP/IP */
···
 	NVME_CTRL_ATTR_HID_128_BIT	= (1 << 0),
 	NVME_CTRL_ATTR_TBKAS		= (1 << 6),
 	NVME_CTRL_ATTR_ELBAS		= (1 << 15),
+	NVME_CTRL_ATTR_RHII		= (1 << 18),
 };
 
 struct nvme_id_ctrl {
···
 static inline bool nvme_is_fabrics(const struct nvme_command *cmd)
 {
 	return cmd->common.opcode == nvme_fabrics_command;
+}
+
+#ifdef CONFIG_NVME_VERBOSE_ERRORS
+const char *nvme_get_error_status_str(u16 status);
+const char *nvme_get_opcode_str(u8 opcode);
+const char *nvme_get_admin_opcode_str(u8 opcode);
+const char *nvme_get_fabrics_opcode_str(u8 opcode);
+#else /* CONFIG_NVME_VERBOSE_ERRORS */
+static inline const char *nvme_get_error_status_str(u16 status)
+{
+	return "I/O Error";
+}
+static inline const char *nvme_get_opcode_str(u8 opcode)
+{
+	return "I/O Cmd";
+}
+static inline const char *nvme_get_admin_opcode_str(u8 opcode)
+{
+	return "Admin Cmd";
+}
+
+static inline const char *nvme_get_fabrics_opcode_str(u8 opcode)
+{
+	return "Fabrics Cmd";
+}
+#endif /* CONFIG_NVME_VERBOSE_ERRORS */
+
+static inline const char *nvme_opcode_str(int qid, u8 opcode)
+{
+	return qid ? nvme_get_opcode_str(opcode) :
+		nvme_get_admin_opcode_str(opcode);
+}
+
+static inline const char *nvme_fabrics_opcode_str(
+		int qid, const struct nvme_command *cmd)
+{
+	if (nvme_is_fabrics(cmd))
+		return nvme_get_fabrics_opcode_str(cmd->fabrics.fctype);
+
+	return nvme_opcode_str(qid, cmd->common.opcode);
 }
 
 struct nvme_error_slot {

include/scsi/scsi_host.h

···
 	 */
 	short cmd_per_lun;
 
-	/* If use block layer to manage tags, this is tag allocation policy */
-	int tag_alloc_policy;
+	/*
+	 * Allocate tags starting from last allocated tag.
+	 */
+	bool tag_alloc_policy_rr : 1;
 
 	/*
 	 * Track QUEUE_FULL events and reduce queue depth on demand.

include/uapi/linux/raid/md_p.h

···
 	char	set_name[32];	/* set and interpreted by user-space */
 
 	__le64	ctime;		/* lo 40 bits are seconds, top 24 are microseconds or 0*/
-	__le32	level;		/* 0,1,4,5 */
+	__le32	level;		/* 0,1,4,5, -1 (linear) */
 	__le32	layout;		/* only for raid5 and raid10 currently */
 	__le64	size;		/* used size of component devices, in 512byte sectors */
 

include/uapi/linux/raid/md_u.h

···
 
 } mdu_array_info_t;
 
+#define LEVEL_LINEAR		(-1)
+
 /* we need a value for 'no level specified' and 0
  * means 'raid0', so we need something else.  This is
  * for internal use only

kernel/trace/blktrace.c

···
 	return ret;
 }
 
-static int __blk_trace_setup(struct request_queue *q, char *name, dev_t dev,
-			     struct block_device *bdev, char __user *arg)
+int blk_trace_setup(struct request_queue *q, char *name, dev_t dev,
+		    struct block_device *bdev,
+		    char __user *arg)
 {
 	struct blk_user_trace_setup buts;
 	int ret;
···
 	if (ret)
 		return -EFAULT;
 
+	mutex_lock(&q->debugfs_mutex);
 	ret = do_blk_trace_setup(q, name, dev, bdev, &buts);
+	mutex_unlock(&q->debugfs_mutex);
 	if (ret)
 		return ret;
 
 	if (copy_to_user(arg, &buts, sizeof(buts))) {
-		__blk_trace_remove(q);
+		blk_trace_remove(q);
 		return -EFAULT;
 	}
 	return 0;
-}
-
-int blk_trace_setup(struct request_queue *q, char *name, dev_t dev,
-		    struct block_device *bdev,
-		    char __user *arg)
-{
-	int ret;
-
-	mutex_lock(&q->debugfs_mutex);
-	ret = __blk_trace_setup(q, name, dev, bdev, arg);
-	mutex_unlock(&q->debugfs_mutex);
-
-	return ret;
 }
 EXPORT_SYMBOL_GPL(blk_trace_setup);
 
···
 		.pid = cbuts.pid,
 	};
 
+	mutex_lock(&q->debugfs_mutex);
 	ret = do_blk_trace_setup(q, name, dev, bdev, &buts);
+	mutex_unlock(&q->debugfs_mutex);
 	if (ret)
 		return ret;
 
 	if (copy_to_user(arg, &buts.name, ARRAY_SIZE(buts.name))) {
-		__blk_trace_remove(q);
+		blk_trace_remove(q);
 		return -EFAULT;
 	}
 
···
 	int ret, start = 0;
 	char b[BDEVNAME_SIZE];
 
-	mutex_lock(&q->debugfs_mutex);
-
 	switch (cmd) {
 	case BLKTRACESETUP:
 		snprintf(b, sizeof(b), "%pg", bdev);
-		ret = __blk_trace_setup(q, b, bdev->bd_dev, bdev, arg);
+		ret = blk_trace_setup(q, b, bdev->bd_dev, bdev, arg);
 		break;
 #if defined(CONFIG_COMPAT) && defined(CONFIG_X86_64)
 	case BLKTRACESETUP32:
···
 		start = 1;
 		fallthrough;
 	case BLKTRACESTOP:
-		ret = __blk_trace_startstop(q, start);
+		ret = blk_trace_startstop(q, start);
 		break;
 	case BLKTRACETEARDOWN:
-		ret = __blk_trace_remove(q);
+		ret = blk_trace_remove(q);
 		break;
 	default:
 		ret = -ENOTTY;
 		break;
 	}
-
-	mutex_unlock(&q->debugfs_mutex);
 	return ret;
 }
 

rust/kernel/block/mq/tag_set.rs

···
                     numa_node: bindings::NUMA_NO_NODE,
                     queue_depth: num_tags,
                     cmd_size,
-                    flags: bindings::BLK_MQ_F_SHOULD_MERGE,
+                    flags: 0,
                     driver_data: core::ptr::null_mut::<crate::ffi::c_void>(),
                     nr_maps: num_maps,
                     ..tag_set