Merge tag 'for-5.7/block-2020-03-29' of git://git.kernel.dk/linux-block

10f36b1e · Linus Torvalds <[email protected]> · 2020-03-30 11:20:13 -07:00

Pull block updates from Jens Axboe:

 - Online capacity resizing (Balbir)

 - Number of hardware queue change fixes (Bart)

 - null_blk fault injection addition (Bart)

 - Cleanup of queue allocation, unifying the node/no-node API
   (Christoph)

 - Cleanup of genhd, moving code to where it makes sense (Christoph)

 - Cleanup of the partition handling code (Christoph)

 - disk stat fixes/improvements (Konstantin)

 - BFQ improvements (Paolo)

 - Various fixes and improvements

* tag 'for-5.7/block-2020-03-29' of git://git.kernel.dk/linux-block: (72 commits)
  block: return NULL in blk_alloc_queue() on error
  block: move bio_map_* to blk-map.c
  Revert "blkdev: check for valid request queue before issuing flush"
  block: simplify queue allocation
  bcache: pass the make_request methods to blk_queue_make_request
  null_blk: use blk_mq_init_queue_data
  block: add a blk_mq_init_queue_data helper
  block: move the ->devnode callback to struct block_device_operations
  block: move the part_stat* helpers from genhd.h to a new header
  block: move block layer internals out of include/linux/genhd.h
  block: move guard_bio_eod to bio.c
  block: unexport get_gendisk
  block: unexport disk_map_sector_rcu
  block: unexport disk_get_part
  block: mark part_in_flight and part_in_flight_rw static
  block: mark block_depr static
  block: factor out requeue handling from dispatch code
  block/diskstats: replace time_in_queue with sum of request times
  block/diskstats: accumulate all per-cpu counters in one pass
  block/diskstats: more accurate approximation of io_ticks for slow disks
  ...

Commit: 10f36b1e80a9f7afdaefe6f0b06dcdf89715eed7
Parents: 3a0eb192 · 654a3667

Merge commit. Changes below are compared with the first parent: 3a0eb192.

Changes

111 files changed · +2430 −2519

File	Status	Add	Del	Content
Documentation/admin-guide/iostats.rst	modified	+4	−1	View file
Documentation/block/capability.rst	modified	+4	−12	View file
Documentation/scsi/scsi_mid_low_api.txt	modified	+0	−21	View file
arch/m68k/emu/nfblock.c	modified	+1	−2	View file
arch/xtensa/platforms/iss/simdisk.c	modified	+1	−2	View file
block/Makefile	modified	+1	−2	View file
block/bfq-cgroup.c	modified	+52	−35	View file
block/bfq-iosched.c	modified	+12	−6	View file
block/bfq-iosched.h	modified	+1	−0	View file
block/bio.c	modified	+65	−515	View file
block/blk-cgroup.c	modified	+1	−1	View file
block/blk-core.c	modified	+38	−44	View file
block/blk-flush.c	modified	+2	−14	View file
block/blk-ioc.c	modified	+7	−0	View file
block/blk-iocost.c	modified	+0	−3	View file
block/blk-map.c	modified	+508	−0	View file
block/blk-mq.c	modified	+38	−21	View file
block/blk-settings.c	modified	+0	−36	View file
block/blk-zoned.c	modified	+1	−1	View file
block/blk.h	modified	+136	−2	View file
block/genhd.c	modified	+192	−27	View file
block/ioctl.c	modified	+1	−0	View file
block/opal_proto.h	modified	+1	−0	View file
block/partition-generic.c	deleted	+0	−623	View before deletion
block/partitions/Makefile	modified	+1	−2	View file
block/partitions/acorn.c	modified	+0	−1	View file
block/partitions/acorn.h	deleted	+0	−15	View before deletion
block/partitions/aix.c	modified	+0	−1	View file
block/partitions/amiga.c	modified	+4	−7	View file
block/partitions/amiga.h	deleted	+0	−7	View before deletion
block/partitions/atari.h	modified	+0	−1	View file
block/partitions/check.c	deleted	+0	−198	View before deletion
block/partitions/check.h	modified	+28	−13	View file
block/partitions/cmdline.c	modified	+0	−1	View file
block/partitions/cmdline.h	deleted	+0	−3	View before deletion
block/partitions/core.c	added	+658	−0	View file
block/partitions/efi.h	modified	+0	−3	View file
block/partitions/ibm.c	modified	+0	−1	View file
block/partitions/ibm.h	deleted	+0	−2	View before deletion
block/partitions/karma.c	modified	+2	−1	View file
block/partitions/karma.h	deleted	+0	−9	View before deletion
block/partitions/ldm.c	modified	+3	−3	View file
block/partitions/ldm.h	modified	+0	−2	View file
block/partitions/mac.h	modified	+0	−1	View file
block/partitions/msdos.c	modified	+151	−21	View file
block/partitions/msdos.h	deleted	+0	−9	View before deletion
block/partitions/osf.c	modified	+1	−1	View file
block/partitions/osf.h	deleted	+0	−8	View before deletion
block/partitions/sgi.c	modified	+6	−1	View file
block/partitions/sgi.h	deleted	+0	−9	View before deletion
block/partitions/sun.c	modified	+8	−1	View file
block/partitions/sun.h	deleted	+0	−9	View before deletion
block/partitions/sysv68.c	modified	+0	−1	View file
block/partitions/sysv68.h	deleted	+0	−2	View before deletion
block/partitions/ultrix.c	modified	+0	−1	View file
block/partitions/ultrix.h	deleted	+0	−6	View before deletion
block/sed-opal.c	modified	+1	−1	View file
drivers/block/brd.c	modified	+1	−3	View file
drivers/block/drbd/drbd_main.c	modified	+1	−2	View file
drivers/block/drbd/drbd_receiver.c	modified	+1	−0	View file
drivers/block/drbd/drbd_worker.c	modified	+1	−0	View file
drivers/block/null_blk_main.c	modified	+65	−41	View file
drivers/block/pktcdvd.c	modified	+7	−8	View file
drivers/block/ps3vram.c	modified	+1	−2	View file
drivers/block/rsxx/dev.c	modified	+1	−2	View file
drivers/block/umem.c	modified	+1	−3	View file
drivers/block/virtio_blk.c	modified	+1	−4	View file
drivers/block/xen-blkfront.c	modified	+1	−5	View file
drivers/block/zram/zram_drv.c	modified	+2	−3	View file
drivers/lightnvm/core.c	modified	+1	−2	View file
drivers/md/bcache/request.c	modified	+2	−5	View file
drivers/md/bcache/request.h	modified	+3	−0	View file
drivers/md/bcache/super.c	modified	+6	−5	View file
drivers/md/dm.c	modified	+5	−5	View file
drivers/md/md.c	modified	+5	−4	View file
drivers/nvdimm/blk.c	modified	+1	−2	View file
drivers/nvdimm/btt.c	modified	+1	−2	View file
drivers/nvdimm/pmem.c	modified	+1	−2	View file
drivers/nvme/host/core.c	modified	+1	−1	View file
drivers/nvme/host/multipath.c	modified	+1	−2	View file
drivers/nvme/target/admin-cmd.c	modified	+1	−0	View file
drivers/s390/block/dcssblk.c	modified	+2	−2	View file
drivers/s390/block/xpram.c	modified	+2	−2	View file
drivers/scsi/BusLogic.c	modified	+5	−3	View file
drivers/scsi/aacraid/linit.c	modified	+4	−3	View file
drivers/scsi/aic7xxx/aic79xx_osm.c	modified	+3	−10	View file
drivers/scsi/aic7xxx/aic7xxx_osm.c	modified	+3	−10	View file
drivers/scsi/arcmsr/arcmsr_hba.c	modified	+4	−9	View file
drivers/scsi/megaraid.c	modified	+2	−11	View file
drivers/scsi/scsi_debug.c	modified	+3	−2	View file
drivers/scsi/scsicam.c	modified	+92	−94	View file
drivers/scsi/sd.c	modified	+2	−1	View file
fs/block_dev.c	modified	+16	−4	View file
fs/buffer.c	modified	+0	−43	View file
fs/ext4/super.c	modified	+4	−4	View file
fs/ext4/sysfs.c	modified	+1	−0	View file
fs/f2fs/f2fs.h	modified	+1	−0	View file
fs/f2fs/super.c	modified	+1	−0	View file
fs/internal.h	modified	+0	−1	View file
fs/reiserfs/journal.c	modified	+2	−3	View file
include/linux/bio.h	modified	+1	−14	View file
include/linux/blk-mq.h	modified	+6	−1	View file
include/linux/blkdev.h	modified	+2	−12	View file
include/linux/fs.h	modified	+0	−1	View file
include/linux/genhd.h	modified	+61	−462	View file
include/linux/iocontext.h	modified	+1	−0	View file
include/linux/msdos_partition.h	added	+50	−0	View file
include/linux/part_stat.h	added	+115	−0	View file
block/partitions/aix.h → include/linux/raid/detect.h	renamed	+2	−1	View file
include/scsi/scsicam.h	modified	+3	−4	View file
init/do_mounts.c	modified	+2	−10	View file

Only the first 600 diff lines are shown. Download the complete patch.

Documentation/admin-guide/iostats.rst

View file

diff --git a/Documentation/admin-guide/iostats.rst b/Documentation/admin-guide/iostats.rst
index df5b834..9b14b0c 100644
--- a/Documentation/admin-guide/iostats.rst
+++ b/Documentation/admin-guide/iostats.rst
@@ -100,7 +100,7 @@ Field 10 -- # of milliseconds spent doing I/Os (unsigned int)
 
     Since 5.0 this field counts jiffies when at least one request was
     started or completed. If request runs more than 2 jiffies then some
-    I/O time will not be accounted unless there are other requests.
+    I/O time might be not accounted in case of concurrent requests.
 
 Field 11 -- weighted # of milliseconds spent doing I/Os (unsigned int)
     This field is incremented at each I/O start, I/O completion, I/O
@@ -143,6 +143,9 @@ are summed (possibly overflowing the unsigned long variable they are
 summed to) and the result given to the user.  There is no convenient
 user interface for accessing the per-CPU counters themselves.
 
+Since 4.19 request times are measured with nanoseconds precision and
+truncated to milliseconds before showing in this interface.
+
 Disks vs Partitions
 -------------------
 

Documentation/block/capability.rst

View file

diff --git a/Documentation/block/capability.rst b/Documentation/block/capability.rst
index 2cf258d..160a514 100644
--- a/Documentation/block/capability.rst
+++ b/Documentation/block/capability.rst
@@ -2,17 +2,9 @@
 Generic Block Device Capability
 ===============================
 
-This file documents the sysfs file block/<disk>/capability
+This file documents the sysfs file ``block/<disk>/capability``.
 
-capability is a hex word indicating which capabilities a specific disk
-supports.  For more information on bits not listed here, see
-include/linux/genhd.h
+``capability`` is a bitfield, printed in hexadecimal, indicating which
+capabilities a specific block device supports:
 
-GENHD_FL_MEDIA_CHANGE_NOTIFY
-----------------------------
-
-Value: 4
-
-When this bit is set, the disk supports Asynchronous Notification
-of media change events.  These events will be broadcast to user
-space via kernel uevent.
+.. kernel-doc:: include/linux/genhd.h

Documentation/scsi/scsi_mid_low_api.txt

View file

diff --git a/Documentation/scsi/scsi_mid_low_api.txt b/Documentation/scsi/scsi_mid_low_api.txt
index 2a4be1c..537f047 100644
--- a/Documentation/scsi/scsi_mid_low_api.txt
+++ b/Documentation/scsi/scsi_mid_low_api.txt
@@ -299,7 +299,6 @@ Summary:
    scsi_host_alloc - return a new scsi_host instance whose refcount==1
    scsi_host_get - increments Scsi_Host instance's refcount
    scsi_host_put - decrements Scsi_Host instance's refcount (free if 0)
-   scsi_partsize - parse partition table into cylinders, heads + sectors
    scsi_register - create and register a scsi host adapter instance.
    scsi_remove_device - detach and remove a SCSI device
    scsi_remove_host - detach and remove all SCSI devices owned by host
@@ -473,26 +472,6 @@ void scsi_host_put(struct Scsi_Host *shost)
 
 
 /**
- * scsi_partsize - parse partition table into cylinders, heads + sectors
- * @buf: pointer to partition table
- * @capacity: size of (total) disk in 512 byte sectors
- * @cyls: outputs number of cylinders calculated via this pointer
- * @hds: outputs number of heads calculated via this pointer
- * @secs: outputs number of sectors calculated via this pointer
- *
- *      Returns 0 on success, -1 on failure
- *
- *      Might block: no
- *
- *      Notes: Caller owns memory returned (free with kfree() )
- *
- *      Defined in: drivers/scsi/scsicam.c
- **/
-int scsi_partsize(unsigned char *buf, unsigned long capacity,
-                  unsigned int *cyls, unsigned int *hds, unsigned int *secs)
-
-
-/**
  * scsi_register - create and register a scsi host adapter instance.
  * @sht:        pointer to scsi host template
  * @privsize:   extra bytes to allocate in hostdata array (which is the

arch/m68k/emu/nfblock.c

View file

diff --git a/arch/m68k/emu/nfblock.c b/arch/m68k/emu/nfblock.c
index 40712e4..c3a6304 100644
--- a/arch/m68k/emu/nfblock.c
+++ b/arch/m68k/emu/nfblock.c
@@ -118,12 +118,11 @@ static int __init nfhd_init_one(int id, u32 blocks, u32 bsize)
 	dev->bsize = bsize;
 	dev->bshift = ffs(bsize) - 10;
 
-	dev->queue = blk_alloc_queue(GFP_KERNEL);
+	dev->queue = blk_alloc_queue(nfhd_make_request, NUMA_NO_NODE);
 	if (dev->queue == NULL)
 		goto free_dev;
 
 	dev->queue->queuedata = dev;
-	blk_queue_make_request(dev->queue, nfhd_make_request);
 	blk_queue_logical_block_size(dev->queue, bsize);
 
 	dev->disk = alloc_disk(16);

arch/xtensa/platforms/iss/simdisk.c

View file

diff --git a/arch/xtensa/platforms/iss/simdisk.c b/arch/xtensa/platforms/iss/simdisk.c
index 8331098..49322b6 100644
--- a/arch/xtensa/platforms/iss/simdisk.c
+++ b/arch/xtensa/platforms/iss/simdisk.c
@@ -267,13 +267,12 @@ static int __init simdisk_setup(struct simdisk *dev, int which,
 	spin_lock_init(&dev->lock);
 	dev->users = 0;
 
-	dev->queue = blk_alloc_queue(GFP_KERNEL);
+	dev->queue = blk_alloc_queue(simdisk_make_request, NUMA_NO_NODE);
 	if (dev->queue == NULL) {
 		pr_err("blk_alloc_queue failed\n");
 		goto out_alloc_queue;
 	}
 
-	blk_queue_make_request(dev->queue, simdisk_make_request);
 	dev->queue->queuedata = dev;
 
 	dev->gd = alloc_disk(SIMDISK_MINORS);

block/Makefile

View file

diff --git a/block/Makefile b/block/Makefile
index 1a43750..206b96e 100644
--- a/block/Makefile
+++ b/block/Makefile
@@ -8,8 +8,7 @@ obj-$(CONFIG_BLOCK) := bio.o elevator.o blk-core.o blk-sysfs.o \
 			blk-exec.o blk-merge.o blk-softirq.o blk-timeout.o \
 			blk-lib.o blk-mq.o blk-mq-tag.o blk-stat.o \
 			blk-mq-sysfs.o blk-mq-cpumap.o blk-mq-sched.o ioctl.o \
-			genhd.o partition-generic.o ioprio.o \
-			badblocks.o partitions/ blk-rq-qos.o
+			genhd.o ioprio.o badblocks.o partitions/ blk-rq-qos.o
 
 obj-$(CONFIG_BOUNCE)		+= bounce.o
 obj-$(CONFIG_BLK_SCSI_REQUEST)	+= scsi_ioctl.o

block/bfq-cgroup.c

View file

diff --git a/block/bfq-cgroup.c b/block/bfq-cgroup.c
index f0ff665..68882b9 100644
--- a/block/bfq-cgroup.c
+++ b/block/bfq-cgroup.c
@@ -642,6 +642,12 @@ void bfq_bfqq_move(struct bfq_data *bfqd, struct bfq_queue *bfqq,
 {
 	struct bfq_entity *entity = &bfqq->entity;
 
+	/*
+	 * Get extra reference to prevent bfqq from being freed in
+	 * next possible expire or deactivate.
+	 */
+	bfqq->ref++;
+
 	/* If bfqq is empty, then bfq_bfqq_expire also invokes
 	 * bfq_del_bfqq_busy, thereby removing bfqq and its entity
 	 * from data structures related to current group. Otherwise we
@@ -652,12 +658,6 @@ void bfq_bfqq_move(struct bfq_data *bfqd, struct bfq_queue *bfqq,
 		bfq_bfqq_expire(bfqd, bfqd->in_service_queue,
 				false, BFQQE_PREEMPTED);
 
-	/*
-	 * get extra reference to prevent bfqq from being freed in
-	 * next possible deactivate
-	 */
-	bfqq->ref++;
-
 	if (bfq_bfqq_busy(bfqq))
 		bfq_deactivate_bfqq(bfqd, bfqq, false, false);
 	else if (entity->on_st_or_in_serv)
@@ -677,7 +677,7 @@ void bfq_bfqq_move(struct bfq_data *bfqd, struct bfq_queue *bfqq,
 
 	if (!bfqd->in_service_queue && !bfqd->rq_in_driver)
 		bfq_schedule_dispatch(bfqd);
-	/* release extra ref taken above */
+	/* release extra ref taken above, bfqq may happen to be freed now */
 	bfq_put_queue(bfqq);
 }
 
@@ -714,10 +714,7 @@ static struct bfq_group *__bfq_bic_change_cgroup(struct bfq_data *bfqd,
 
 		if (entity->sched_data != &bfqg->sched_data) {
 			bic_set_bfqq(bic, NULL, 0);
-			bfq_log_bfqq(bfqd, async_bfqq,
-				     "bic_change_group: %p %d",
-				     async_bfqq, async_bfqq->ref);
-			bfq_put_queue(async_bfqq);
+			bfq_release_process_ref(bfqd, async_bfqq);
 		}
 	}
 
@@ -818,39 +815,53 @@ static void bfq_flush_idle_tree(struct bfq_service_tree *st)
 /**
  * bfq_reparent_leaf_entity - move leaf entity to the root_group.
  * @bfqd: the device data structure with the root group.
- * @entity: the entity to move.
+ * @entity: the entity to move, if entity is a leaf; or the parent entity
+ *	    of an active leaf entity to move, if entity is not a leaf.
  */
 static void bfq_reparent_leaf_entity(struct bfq_data *bfqd,
-				     struct bfq_entity *entity)
+				     struct bfq_entity *entity,
+				     int ioprio_class)
 {
-	struct bfq_queue *bfqq = bfq_entity_to_bfqq(entity);
+	struct bfq_queue *bfqq;
+	struct bfq_entity *child_entity = entity;
 
+	while (child_entity->my_sched_data) { /* leaf not reached yet */
+		struct bfq_sched_data *child_sd = child_entity->my_sched_data;
+		struct bfq_service_tree *child_st = child_sd->service_tree +
+			ioprio_class;
+		struct rb_root *child_active = &child_st->active;
+
+		child_entity = bfq_entity_of(rb_first(child_active));
+
+		if (!child_entity)
+			child_entity = child_sd->in_service_entity;
+	}
+
+	bfqq = bfq_entity_to_bfqq(child_entity);
 	bfq_bfqq_move(bfqd, bfqq, bfqd->root_group);
 }
 
 /**
- * bfq_reparent_active_entities - move to the root group all active
- *                                entities.
+ * bfq_reparent_active_queues - move to the root group all active queues.
  * @bfqd: the device data structure with the root group.
  * @bfqg: the group to move from.
- * @st: the service tree with the entities.
+ * @st: the service tree to start the search from.
  */
-static void bfq_reparent_active_entities(struct bfq_data *bfqd,
-					 struct bfq_group *bfqg,
-					 struct bfq_service_tree *st)
+static void bfq_reparent_active_queues(struct bfq_data *bfqd,
+				       struct bfq_group *bfqg,
+				       struct bfq_service_tree *st,
+				       int ioprio_class)
 {
 	struct rb_root *active = &st->active;
-	struct bfq_entity *entity = NULL;
-
-	if (!RB_EMPTY_ROOT(&st->active))
-		entity = bfq_entity_of(rb_first(active));
+	struct bfq_entity *entity;
 
-	for (; entity ; entity = bfq_entity_of(rb_first(active)))
-		bfq_reparent_leaf_entity(bfqd, entity);
+	while ((entity = bfq_entity_of(rb_first(active))))
+		bfq_reparent_leaf_entity(bfqd, entity, ioprio_class);
 
 	if (bfqg->sched_data.in_service_entity)
 		bfq_reparent_leaf_entity(bfqd,
-			bfqg->sched_data.in_service_entity);
+					 bfqg->sched_data.in_service_entity,
+					 ioprio_class);
 }
 
 /**
@@ -883,13 +894,6 @@ static void bfq_pd_offline(struct blkg_policy_data *pd)
 		st = bfqg->sched_data.service_tree + i;
 
 		/*
-		 * The idle tree may still contain bfq_queues belonging
-		 * to exited task because they never migrated to a different
-		 * cgroup from the one being destroyed now.
-		 */
-		bfq_flush_idle_tree(st);
-
-		/*
 		 * It may happen that some queues are still active
 		 * (busy) upon group destruction (if the corresponding
 		 * processes have been forced to terminate). We move
@@ -901,7 +905,20 @@ static void bfq_pd_offline(struct blkg_policy_data *pd)
 		 * There is no need to put the sync queues, as the
 		 * scheduler has taken no reference.
 		 */
-		bfq_reparent_active_entities(bfqd, bfqg, st);
+		bfq_reparent_active_queues(bfqd, bfqg, st, i);
+
+		/*
+		 * The idle tree may still contain bfq_queues
+		 * belonging to exited task because they never
+		 * migrated to a different cgroup from the one being
+		 * destroyed now. In addition, even
+		 * bfq_reparent_active_queues() may happen to add some
+		 * entities to the idle tree. It happens if, in some
+		 * of the calls to bfq_bfqq_move() performed by
+		 * bfq_reparent_active_queues(), the queue to move is
+		 * empty and gets expired.
+		 */
+		bfq_flush_idle_tree(st);
 	}
 
 	__bfq_deactivate_entity(entity, false);

block/bfq-iosched.c

View file

diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index 8c436ab..78ba57e 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -2716,8 +2716,6 @@ static void bfq_bfqq_save_state(struct bfq_queue *bfqq)
 	}
 }
 
-
-static
 void bfq_release_process_ref(struct bfq_data *bfqd, struct bfq_queue *bfqq)
 {
 	/*
@@ -6215,20 +6213,28 @@ static struct bfq_queue *bfq_init_rq(struct request *rq)
 	return bfqq;
 }
 
-static void bfq_idle_slice_timer_body(struct bfq_queue *bfqq)
+static void
+bfq_idle_slice_timer_body(struct bfq_data *bfqd, struct bfq_queue *bfqq)
 {
-	struct bfq_data *bfqd = bfqq->bfqd;
 	enum bfqq_expiration reason;
 	unsigned long flags;
 
 	spin_lock_irqsave(&bfqd->lock, flags);
-	bfq_clear_bfqq_wait_request(bfqq);
 
+	/*
+	 * Considering that bfqq may be in race, we should firstly check
+	 * whether bfqq is in service before doing something on it. If
+	 * the bfqq in race is not in service, it has already been expired
+	 * through __bfq_bfqq_expire func and its wait_request flags has
+	 * been cleared in __bfq_bfqd_reset_in_service func.
+	 */
 	if (bfqq != bfqd->in_service_queue) {
 		spin_unlock_irqrestore(&bfqd->lock, flags);
 		return;
 	}
 
+	bfq_clear_bfqq_wait_request(bfqq);
+
 	if (bfq_bfqq_budget_timeout(bfqq))
 		/*
 		 * Also here the queue can be safely expired
@@ -6273,7 +6279,7 @@ static enum hrtimer_restart bfq_idle_slice_timer(struct hrtimer *timer)
 	 * early.
 	 */
 	if (bfqq)
-		bfq_idle_slice_timer_body(bfqq);
+		bfq_idle_slice_timer_body(bfqd, bfqq);
 
 	return HRTIMER_NORESTART;
 }

block/bfq-iosched.h

View file

diff --git a/block/bfq-iosched.h b/block/bfq-iosched.h
index d1233af..cd224aa 100644
--- a/block/bfq-iosched.h
+++ b/block/bfq-iosched.h
@@ -955,6 +955,7 @@ void bfq_bfqq_expire(struct bfq_data *bfqd, struct bfq_queue *bfqq,
 		     bool compensate, enum bfqq_expiration reason);
 void bfq_put_queue(struct bfq_queue *bfqq);
 void bfq_end_wr_async_queues(struct bfq_data *bfqd, struct bfq_group *bfqg);
+void bfq_release_process_ref(struct bfq_data *bfqd, struct bfq_queue *bfqq);
 void bfq_schedule_dispatch(struct bfq_data *bfqd);
 void bfq_put_async_queues(struct bfq_data *bfqd, struct bfq_group *bfqg);
 

block/bio.c

View file

diff --git a/block/bio.c b/block/bio.c
index 94d6972..21cbaa6 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -17,6 +17,7 @@
 #include <linux/cgroup.h>
 #include <linux/blk-cgroup.h>
 #include <linux/highmem.h>
+#include <linux/sched/sysctl.h>
 
 #include <trace/events/block.h>
 #include "blk.h"
@@ -588,6 +589,49 @@ void bio_truncate(struct bio *bio, unsigned new_size)
 }
 
 /**
+ * guard_bio_eod - truncate a BIO to fit the block device
+ * @bio:	bio to truncate
+ *
+ * This allows us to do IO even on the odd last sectors of a device, even if the
+ * block size is some multiple of the physical sector size.
+ *
+ * We'll just truncate the bio to the size of the device, and clear the end of
+ * the buffer head manually.  Truly out-of-range accesses will turn into actual
+ * I/O errors, this only handles the "we need to be able to do I/O at the final
+ * sector" case.
+ */
+void guard_bio_eod(struct bio *bio)
+{
+	sector_t maxsector;
+	struct hd_struct *part;
+
+	rcu_read_lock();
+	part = __disk_get_part(bio->bi_disk, bio->bi_partno);
+	if (part)
+		maxsector = part_nr_sects_read(part);
+	else
+		maxsector = get_capacity(bio->bi_disk);
+	rcu_read_unlock();
+
+	if (!maxsector)
+		return;
+
+	/*
+	 * If the *whole* IO is past the end of the device,
+	 * let it through, and the IO layer will turn it into
+	 * an EIO.
+	 */
+	if (unlikely(bio->bi_iter.bi_sector >= maxsector))
+		return;
+
+	maxsector -= bio->bi_iter.bi_sector;
+	if (likely((bio->bi_iter.bi_size >> 9) <= maxsector))
+		return;
+
+	bio_truncate(bio, maxsector << 9);
+}
+
+/**
  * bio_put - release a reference to a bio
  * @bio:   bio to release reference to
  *
@@ -679,6 +723,12 @@ struct bio *bio_clone_fast(struct bio *bio, gfp_t gfp_mask, struct bio_set *bs)
 }
 EXPORT_SYMBOL(bio_clone_fast);
 
+const char *bio_devname(struct bio *bio, char *buf)
+{
+	return disk_name(bio->bi_disk, bio->bi_partno, buf);
+}
+EXPORT_SYMBOL(bio_devname);
+
 static inline bool page_is_mergeable(const struct bio_vec *bv,
 		struct page *page, unsigned int len, unsigned int off,
 		bool *same_page)
@@ -730,7 +780,7 @@ static bool bio_try_merge_pc_page(struct request_queue *q, struct bio *bio,
  *
  *	This should only be used by passthrough bios.
  */
-static int __bio_add_pc_page(struct request_queue *q, struct bio *bio,
+int __bio_add_pc_page(struct request_queue *q, struct bio *bio,
 		struct page *page, unsigned int len, unsigned int offset,
 		bool *same_page)
 {
@@ -1019,12 +1069,21 @@ static void submit_bio_wait_endio(struct bio *bio)
 int submit_bio_wait(struct bio *bio)
 {
 	DECLARE_COMPLETION_ONSTACK_MAP(done, bio->bi_disk->lockdep_map);
+	unsigned long hang_check;
 
 	bio->bi_private = &done;
 	bio->bi_end_io = submit_bio_wait_endio;
 	bio->bi_opf |= REQ_SYNC;
 	submit_bio(bio);
-	wait_for_completion_io(&done);
+
+	/* Prevent hang_check timer from firing at us during very long I/O */
+	hang_check = sysctl_hung_task_timeout_secs;
+	if (hang_check)
+		while (!wait_for_completion_io_timeout(&done,
+					hang_check * (HZ/2)))
+			;
+	else
+		wait_for_completion_io(&done);
 
 	return blk_status_to_errno(bio->bi_status);
 }
@@ -1135,90 +1194,6 @@ void bio_list_copy_data(struct bio *dst, struct bio *src)
 }
 EXPORT_SYMBOL(bio_list_copy_data);
 
-struct bio_map_data {
-	int is_our_pages;
-	struct iov_iter iter;
-	struct iovec iov[];
-};
-
-static struct bio_map_data *bio_alloc_map_data(struct iov_iter *data,
-					       gfp_t gfp_mask)
-{
-	struct bio_map_data *bmd;
-	if (data->nr_segs > UIO_MAXIOV)
-		return NULL;
-
-	bmd = kmalloc(struct_size(bmd, iov, data->nr_segs), gfp_mask);
-	if (!bmd)
-		return NULL;
-	memcpy(bmd->iov, data->iov, sizeof(struct iovec) * data->nr_segs);
-	bmd->iter = *data;
-	bmd->iter.iov = bmd->iov;
-	return bmd;
-}
-
-/**
- * bio_copy_from_iter - copy all pages from iov_iter to bio
- * @bio: The &struct bio which describes the I/O as destination
- * @iter: iov_iter as source
- *
- * Copy all pages from iov_iter to bio.
- * Returns 0 on success, or error on failure.
- */
-static int bio_copy_from_iter(struct bio *bio, struct iov_iter *iter)
-{
-	struct bio_vec *bvec;
-	struct bvec_iter_all iter_all;
-
-	bio_for_each_segment_all(bvec, bio, iter_all) {
-		ssize_t ret;
-
-		ret = copy_page_from_iter(bvec->bv_page,
-					  bvec->bv_offset,
-					  bvec->bv_len,
-					  iter);
-
-		if (!iov_iter_count(iter))
-			break;
-
-		if (ret < bvec->bv_len)
-			return -EFAULT;
-	}
-
-	return 0;
-}
-
-/**
- * bio_copy_to_iter - copy all pages from bio to iov_iter
- * @bio: The &struct bio which describes the I/O as source
- * @iter: iov_iter as destination
- *
- * Copy all pages from bio to iov_iter.
- * Returns 0 on success, or error on failure.
- */
-static int bio_copy_to_iter(struct bio *bio, struct iov_iter iter)
-{
-	struct bio_vec *bvec;
-	struct bvec_iter_all iter_all;
-
-	bio_for_each_segment_all(bvec, bio, iter_all) {
-		ssize_t ret;
-
-		ret = copy_page_to_iter(bvec->bv_page,
-					bvec->bv_offset,
-					bvec->bv_len,
-					&iter);
-
-		if (!iov_iter_count(&iter))
-			break;
-
-		if (ret < bvec->bv_len)
-			return -EFAULT;
-	}
-
-	return 0;
-}
-
 void bio_free_pages(struct bio *bio)
 {
 	struct bio_vec *bvec;
@@ -1229,430 +1204,6 @@ void bio_free_pages(struct bio *bio)
 }
 EXPORT_SYMBOL(bio_free_pages);
 
-/**
- *	bio_uncopy_user	-	finish previously mapped bio
- *	@bio: bio being terminated
- *
- *	Free pages allocated from bio_copy_user_iov() and write back data
- *	to user space in case of a read.
- */
-int bio_uncopy_user(struct bio *bio)
-{
-	struct bio_map_data *bmd = bio->bi_private;
-	int ret = 0;
-
-	if (!bio_flagged(bio, BIO_NULL_MAPPED)) {
-		/*
-		 * if we're in a workqueue, the request is orphaned, so
-		 * don't copy into a random user address space, just free
-		 * and return -EINTR so user space doesn't expect any data.
-		 */
-		if (!current->mm)
-			ret = -EINTR;
-		else if (bio_data_dir(bio) == READ)
-			ret = bio_copy_to_iter(bio, bmd->iter);
-		if (bmd->is_our_pages)
-			bio_free_pages(bio);
-	}
-	kfree(bmd);
-	bio_put(bio);
-	return ret;
-}
-
-/**
- *	bio_copy_user_iov	-	copy user data to bio
- *	@q:		destination block queue
- *	@map_data:	pointer to the rq_map_data holding pages (if necessary)
- *	@iter:		iovec iterator
- *	@gfp_mask:	memory allocation flags