Merge branch 'next' into for-linus

25 files changed, 707 insertions, 352 deletions

diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c
index 528f6318ded..321de7b6c44 100644
--- a/drivers/block/aoe/aoeblk.c
+++ b/drivers/block/aoe/aoeblk.c
@@ -15,6 +15,7 @@
 #include <linux/genhd.h>
 #include <linux/netdevice.h>
 #include <linux/mutex.h>
+#include <linux/export.h>
 #include "aoe.h"
 
 static DEFINE_MUTEX(aoeblk_mutex);
@@ -159,7 +160,7 @@ aoeblk_release(struct gendisk *disk, fmode_t mode)
 	return 0;
 }
 
-static int
+static void
 aoeblk_make_request(struct request_queue *q, struct bio *bio)
 {
 	struct sk_buff_head queue;
@@ -172,25 +173,25 @@ aoeblk_make_request(struct request_queue *q, struct bio *bio)
 	if (bio == NULL) {
 		printk(KERN_ERR "aoe: bio is NULL\n");
 		BUG();
-		return 0;
+		return;
 	}
 	d = bio->bi_bdev->bd_disk->private_data;
 	if (d == NULL) {
 		printk(KERN_ERR "aoe: bd_disk->private_data is NULL\n");
 		BUG();
 		bio_endio(bio, -ENXIO);
-		return 0;
+		return;
 	} else if (bio->bi_io_vec == NULL) {
 		printk(KERN_ERR "aoe: bi_io_vec is NULL\n");
 		BUG();
 		bio_endio(bio, -ENXIO);
-		return 0;
+		return;
 	}
 	buf = mempool_alloc(d->bufpool, GFP_NOIO);
 	if (buf == NULL) {
 		printk(KERN_INFO "aoe: buf allocation failure\n");
 		bio_endio(bio, -ENOMEM);
-		return 0;
+		return;
 	}
 	memset(buf, 0, sizeof(*buf));
 	INIT_LIST_HEAD(&buf->bufs);
@@ -211,7 +212,7 @@ aoeblk_make_request(struct request_queue *q, struct bio *bio)
 		spin_unlock_irqrestore(&d->lock, flags);
 		mempool_free(buf, d->bufpool);
 		bio_endio(bio, -ENXIO);
-		return 0;
+		return;
 	}
 
 	list_add_tail(&buf->bufs, &d->bufq);
@@ -222,8 +223,6 @@ aoeblk_make_request(struct request_queue *q, struct bio *bio)
 
 	spin_unlock_irqrestore(&d->lock, flags);
 	aoenet_xmit(&queue);
-
-	return 0;
 }
 
 static int
diff --git a/drivers/block/aoe/aoechr.c b/drivers/block/aoe/aoechr.c
index 146296ca496..5f8e39c43ae 100644
--- a/drivers/block/aoe/aoechr.c
+++ b/drivers/block/aoe/aoechr.c
@@ -11,6 +11,7 @@
 #include <linux/slab.h>
 #include <linux/mutex.h>
 #include <linux/skbuff.h>
+#include <linux/export.h>
 #include "aoe.h"
 
 enum {
diff --git a/drivers/block/brd.c b/drivers/block/brd.c
index dba1c32e1dd..d22119d49e5 100644
--- a/drivers/block/brd.c
+++ b/drivers/block/brd.c
@@ -323,7 +323,7 @@ out:
 	return err;
 }
 
-static int brd_make_request(struct request_queue *q, struct bio *bio)
+static void brd_make_request(struct request_queue *q, struct bio *bio)
 {
 	struct block_device *bdev = bio->bi_bdev;
 	struct brd_device *brd = bdev->bd_disk->private_data;
@@ -359,8 +359,6 @@ static int brd_make_request(struct request_queue *q, struct bio *bio)
 
 out:
 	bio_endio(bio, err);
-
-	return 0;
 }
 
 #ifdef CONFIG_BLK_DEV_XIP
diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c
index 8f4ef656a1a..8004ac30a7a 100644
--- a/drivers/block/cciss.c
+++ b/drivers/block/cciss.c
@@ -24,6 +24,7 @@
 #include <linux/interrupt.h>
 #include <linux/types.h>
 #include <linux/pci.h>
+#include <linux/pci-aspm.h>
 #include <linux/kernel.h>
 #include <linux/slab.h>
 #include <linux/delay.h>
@@ -68,6 +69,10 @@ static int cciss_tape_cmds = 6;
 module_param(cciss_tape_cmds, int, 0644);
 MODULE_PARM_DESC(cciss_tape_cmds,
 	"number of commands to allocate for tape devices (default: 6)");
+static int cciss_simple_mode;
+module_param(cciss_simple_mode, int, S_IRUGO|S_IWUSR);
+MODULE_PARM_DESC(cciss_simple_mode,
+	"Use 'simple mode' rather than 'performant mode'");
 
 static DEFINE_MUTEX(cciss_mutex);
 static struct proc_dir_entry *proc_cciss;
@@ -176,6 +181,7 @@ static void cciss_geometry_inquiry(ctlr_info_t *h, int logvol,
 			unsigned int block_size, InquiryData_struct *inq_buff,
 				   drive_info_struct *drv);
 static void __devinit cciss_interrupt_mode(ctlr_info_t *);
+static int __devinit cciss_enter_simple_mode(struct ctlr_info *h);
 static void start_io(ctlr_info_t *h);
 static int sendcmd_withirq(ctlr_info_t *h, __u8 cmd, void *buff, size_t size,
 			__u8 page_code, unsigned char scsi3addr[],
@@ -388,7 +394,7 @@ static void cciss_seq_show_header(struct seq_file *seq)
 		h->product_name,
 		(unsigned long)h->board_id,
 		h->firm_ver[0], h->firm_ver[1], h->firm_ver[2],
-		h->firm_ver[3], (unsigned int)h->intr[PERF_MODE_INT],
+		h->firm_ver[3], (unsigned int)h->intr[h->intr_mode],
 		h->num_luns,
 		h->Qdepth, h->commands_outstanding,
 		h->maxQsinceinit, h->max_outstanding, h->maxSG);
@@ -636,6 +642,18 @@ static ssize_t host_store_rescan(struct device *dev,
 }
 static DEVICE_ATTR(rescan, S_IWUSR, NULL, host_store_rescan);
 
+static ssize_t host_show_transport_mode(struct device *dev,
+				 struct device_attribute *attr,
+				 char *buf)
+{
+	struct ctlr_info *h = to_hba(dev);
+
+	return snprintf(buf, 20, "%s\n",
+		h->transMethod & CFGTBL_Trans_Performant ?
+			"performant" : "simple");
+}
+static DEVICE_ATTR(transport_mode, S_IRUGO, host_show_transport_mode, NULL);
+
 static ssize_t dev_show_unique_id(struct device *dev,
 				 struct device_attribute *attr,
 				 char *buf)
@@ -808,6 +826,7 @@ static DEVICE_ATTR(usage_count, S_IRUGO, cciss_show_usage_count, NULL);
 static struct attribute *cciss_host_attrs[] = {
 	&dev_attr_rescan.attr,
 	&dev_attr_resettable.attr,
+	&dev_attr_transport_mode.attr,
 	NULL
 };
 
@@ -3984,6 +4003,9 @@ static void __devinit cciss_put_controller_into_performant_mode(ctlr_info_t *h)
 {
 	__u32 trans_support;
 
+	if (cciss_simple_mode)
+		return;
+
 	dev_dbg(&h->pdev->dev, "Trying to put board into Performant mode\n");
 	/* Attempt to put controller into performant mode if supported */
 	/* Does board support performant mode? */
@@ -4081,7 +4103,7 @@ static void __devinit cciss_interrupt_mode(ctlr_info_t *h)
 default_int_mode:
 #endif				/* CONFIG_PCI_MSI */
 	/* if we get here we're going to use the default interrupt mode */
-	h->intr[PERF_MODE_INT] = h->pdev->irq;
+	h->intr[h->intr_mode] = h->pdev->irq;
 	return;
 }
 
@@ -4298,6 +4320,10 @@ static int __devinit cciss_pci_init(ctlr_info_t *h)
 		dev_warn(&h->pdev->dev, "controller appears to be disabled\n");
 		return -ENODEV;
 	}
+
+	pci_disable_link_state(h->pdev, PCIE_LINK_STATE_L0S |
+				PCIE_LINK_STATE_L1 | PCIE_LINK_STATE_CLKPM);
+
 	err = pci_enable_device(h->pdev);
 	if (err) {
 		dev_warn(&h->pdev->dev, "Unable to Enable PCI device\n");
@@ -4341,6 +4367,9 @@ static int __devinit cciss_pci_init(ctlr_info_t *h)
 	}
 	cciss_enable_scsi_prefetch(h);
 	cciss_p600_dma_prefetch_quirk(h);
+	err = cciss_enter_simple_mode(h);
+	if (err)
+		goto err_out_free_res;
 	cciss_put_controller_into_performant_mode(h);
 	return 0;
 
@@ -4533,6 +4562,13 @@ static int cciss_controller_hard_reset(struct pci_dev *pdev,
 		pmcsr &= ~PCI_PM_CTRL_STATE_MASK;
 		pmcsr |= PCI_D0;
 		pci_write_config_word(pdev, pos + PCI_PM_CTRL, pmcsr);
+
+		/*
+		 * The P600 requires a small delay when changing states.
+		 * Otherwise we may think the board did not reset and we bail.
+		 * This for kdump only and is particular to the P600.
+		 */
+		msleep(500);
 	}
 	return 0;
 }
@@ -4843,20 +4879,20 @@ static int cciss_request_irq(ctlr_info_t *h,
 	irqreturn_t (*intxhandler)(int, void *))
 {
 	if (h->msix_vector || h->msi_vector) {
-		if (!request_irq(h->intr[PERF_MODE_INT], msixhandler,
+		if (!request_irq(h->intr[h->intr_mode], msixhandler,
 				IRQF_DISABLED, h->devname, h))
 			return 0;
 		dev_err(&h->pdev->dev, "Unable to get msi irq %d"
-			" for %s\n", h->intr[PERF_MODE_INT],
+			" for %s\n", h->intr[h->intr_mode],
 			h->devname);
 		return -1;
 	}
 
-	if (!request_irq(h->intr[PERF_MODE_INT], intxhandler,
+	if (!request_irq(h->intr[h->intr_mode], intxhandler,
 			IRQF_DISABLED, h->devname, h))
 		return 0;
 	dev_err(&h->pdev->dev, "Unable to get irq %d for %s\n",
-		h->intr[PERF_MODE_INT], h->devname);
+		h->intr[h->intr_mode], h->devname);
 	return -1;
 }
 
@@ -4887,7 +4923,7 @@ static void cciss_undo_allocations_after_kdump_soft_reset(ctlr_info_t *h)
 {
 	int ctlr = h->ctlr;
 
-	free_irq(h->intr[PERF_MODE_INT], h);
+	free_irq(h->intr[h->intr_mode], h);
 #ifdef CONFIG_PCI_MSI
 	if (h->msix_vector)
 		pci_disable_msix(h->pdev);
@@ -4953,6 +4989,7 @@ reinit_after_soft_reset:
 	h = hba[i];
 	h->pdev = pdev;
 	h->busy_initializing = 1;
+	h->intr_mode = cciss_simple_mode ? SIMPLE_MODE_INT : PERF_MODE_INT;
 	INIT_LIST_HEAD(&h->cmpQ);
 	INIT_LIST_HEAD(&h->reqQ);
 	mutex_init(&h->busy_shutting_down);
@@ -5009,7 +5046,7 @@ reinit_after_soft_reset:
 
 	dev_info(&h->pdev->dev, "%s: <0x%x> at PCI %s IRQ %d%s using DAC\n",
 	       h->devname, pdev->device, pci_name(pdev),
-	       h->intr[PERF_MODE_INT], dac ? "" : " not");
+	       h->intr[h->intr_mode], dac ? "" : " not");
 
 	if (cciss_allocate_cmd_pool(h))
 		goto clean4;
@@ -5056,7 +5093,7 @@ reinit_after_soft_reset:
 		spin_lock_irqsave(&h->lock, flags);
 		h->access.set_intr_mask(h, CCISS_INTR_OFF);
 		spin_unlock_irqrestore(&h->lock, flags);
-		free_irq(h->intr[PERF_MODE_INT], h);
+		free_irq(h->intr[h->intr_mode], h);
 		rc = cciss_request_irq(h, cciss_msix_discard_completions,
 					cciss_intx_discard_completions);
 		if (rc) {
@@ -5126,6 +5163,7 @@ reinit_after_soft_reset:
 	h->cciss_max_sectors = 8192;
 
 	rebuild_lun_table(h, 1, 0);
+	cciss_engage_scsi(h);
 	h->busy_initializing = 0;
 	return 1;
 
@@ -5133,7 +5171,7 @@ clean4:
 	cciss_free_cmd_pool(h);
 	cciss_free_scatterlists(h);
 	cciss_free_sg_chain_blocks(h->cmd_sg_list, h->nr_cmds);
-	free_irq(h->intr[PERF_MODE_INT], h);
+	free_irq(h->intr[h->intr_mode], h);
 clean2:
 	unregister_blkdev(h->major, h->devname);
 clean1:
@@ -5172,9 +5210,31 @@ static void cciss_shutdown(struct pci_dev *pdev)
 	if (return_code != IO_OK)
 		dev_warn(&h->pdev->dev, "Error flushing cache\n");
 	h->access.set_intr_mask(h, CCISS_INTR_OFF);
-	free_irq(h->intr[PERF_MODE_INT], h);
+	free_irq(h->intr[h->intr_mode], h);
 }
 
+static int __devinit cciss_enter_simple_mode(struct ctlr_info *h)
+{
+	u32 trans_support;
+
+	trans_support = readl(&(h->cfgtable->TransportSupport));
+	if (!(trans_support & SIMPLE_MODE))
+		return -ENOTSUPP;
+
+	h->max_commands = readl(&(h->cfgtable->CmdsOutMax));
+	writel(CFGTBL_Trans_Simple, &(h->cfgtable->HostWrite.TransportRequest));
+	writel(CFGTBL_ChangeReq, h->vaddr + SA5_DOORBELL);
+	cciss_wait_for_mode_change_ack(h);
+	print_cfg_table(h);
+	if (!(readl(&(h->cfgtable->TransportActive)) & CFGTBL_Trans_Simple)) {
+		dev_warn(&h->pdev->dev, "unable to get board into simple mode\n");
+		return -ENODEV;
+	}
+	h->transMethod = CFGTBL_Trans_Simple;
+	return 0;
+}
+
+
 static void __devexit cciss_remove_one(struct pci_dev *pdev)
 {
 	ctlr_info_t *h;
diff --git a/drivers/block/cciss.h b/drivers/block/cciss.h
index c049548e68b..7fda30e4a24 100644
--- a/drivers/block/cciss.h
+++ b/drivers/block/cciss.h
@@ -92,6 +92,7 @@ struct ctlr_info
 	unsigned int intr[4];
 	unsigned int msix_vector;
 	unsigned int msi_vector;
+	int	intr_mode;
 	int 	cciss_max_sectors;
 	BYTE	cciss_read;
 	BYTE	cciss_write;
diff --git a/drivers/block/cciss_scsi.c b/drivers/block/cciss_scsi.c
index 951a4e33b92..e820b68d2f6 100644
--- a/drivers/block/cciss_scsi.c
+++ b/drivers/block/cciss_scsi.c
@@ -1720,5 +1720,6 @@ static int  cciss_eh_abort_handler(struct scsi_cmnd *scsicmd)
 /* If no tape support, then these become defined out of existence */
 
 #define cciss_scsi_setup(cntl_num)
+#define cciss_engage_scsi(h)
 
 #endif /* CONFIG_CISS_SCSI_TAPE */
diff --git a/drivers/block/cpqarray.c b/drivers/block/cpqarray.c
index b2fceb53e80..9125bbeacd4 100644
--- a/drivers/block/cpqarray.c
+++ b/drivers/block/cpqarray.c
@@ -620,6 +620,7 @@ static int cpqarray_pci_init(ctlr_info_t *c, struct pci_dev *pdev)
 	}
 	vendor_id = pdev->vendor;
 	device_id = pdev->device;
+	revision  = pdev->revision;
 	irq = pdev->irq;
 
 	for(i=0; i<6; i++)
@@ -632,7 +633,6 @@ static int cpqarray_pci_init(ctlr_info_t *c, struct pci_dev *pdev)
 	}
 
 	pci_read_config_word(pdev, PCI_COMMAND, &command);
-	pci_read_config_byte(pdev, PCI_CLASS_REVISION, &revision);
 	pci_read_config_byte(pdev, PCI_CACHE_LINE_SIZE, &cache_line_size);
 	pci_read_config_byte(pdev, PCI_LATENCY_TIMER, &latency_timer);
 
diff --git a/drivers/block/drbd/drbd_bitmap.c b/drivers/block/drbd/drbd_bitmap.c
index 7b976296b56..912f585a760 100644
--- a/drivers/block/drbd/drbd_bitmap.c
+++ b/drivers/block/drbd/drbd_bitmap.c
@@ -378,15 +378,14 @@ static struct page **bm_realloc_pages(struct drbd_bitmap *b, unsigned long want)
 	 * thread.  As we have no disk yet, we are not in the IO path,
 	 * not even the IO path of the peer. */
 	bytes = sizeof(struct page *)*want;
-	new_pages = kmalloc(bytes, GFP_KERNEL);
+	new_pages = kzalloc(bytes, GFP_KERNEL);
 	if (!new_pages) {
-		new_pages = vmalloc(bytes);
+		new_pages = vzalloc(bytes);
 		if (!new_pages)
 			return NULL;
 		vmalloced = 1;
 	}
 
-	memset(new_pages, 0, bytes);
 	if (want >= have) {
 		for (i = 0; i < have; i++)
 			new_pages[i] = old_pages[i];
diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h
index ef2ceed3be4..9cf20355cee 100644
--- a/drivers/block/drbd/drbd_int.h
+++ b/drivers/block/drbd/drbd_int.h
@@ -28,7 +28,6 @@
 
 #include <linux/compiler.h>
 #include <linux/types.h>
-#include <linux/version.h>
 #include <linux/list.h>
 #include <linux/sched.h>
 #include <linux/bitops.h>
@@ -928,7 +927,7 @@ struct drbd_md {
 #define NL_INT64(pn,pr,member) __u64 member;
 #define NL_BIT(pn,pr,member)   unsigned member:1;
 #define NL_STRING(pn,pr,member,len) unsigned char member[len]; int member ## _len;
-#include "linux/drbd_nl.h"
+#include <linux/drbd_nl.h>
 
 struct drbd_backing_dev {
 	struct block_device *backing_bdev;
@@ -1507,7 +1506,7 @@ extern void drbd_free_mdev(struct drbd_conf *mdev);
 extern int proc_details;
 
 /* drbd_req */
-extern int drbd_make_request(struct request_queue *q, struct bio *bio);
+extern void drbd_make_request(struct request_queue *q, struct bio *bio);
 extern int drbd_read_remote(struct drbd_conf *mdev, struct drbd_request *req);
 extern int drbd_merge_bvec(struct request_queue *q, struct bvec_merge_data *bvm, struct bio_vec *bvec);
 extern int is_valid_ar_handle(struct drbd_request *, sector_t);
diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c
index 0feab261e29..af2a25049bc 100644
--- a/drivers/block/drbd/drbd_nl.c
+++ b/drivers/block/drbd/drbd_nl.c
@@ -94,7 +94,7 @@ static int name ## _from_tags(struct drbd_conf *mdev, \
 		 arg->member ## _len = dlen; \
 		 memcpy(arg->member, tags, min_t(size_t, dlen, len)); \
 		 break;
-#include "linux/drbd_nl.h"
+#include <linux/drbd_nl.h>
 
 /* Generate the struct to tag_list functions */
 #define NL_PACKET(name, number, fields) \
@@ -129,7 +129,7 @@ name ## _to_tags(struct drbd_conf *mdev, \
 	put_unaligned(arg->member ## _len, tags++);	\
 	memcpy(tags, arg->member, arg->member ## _len); \
 	tags = (unsigned short *)((char *)tags + arg->member ## _len);
-#include "linux/drbd_nl.h"
+#include <linux/drbd_nl.h>
 
 void drbd_bcast_ev_helper(struct drbd_conf *mdev, char *helper_name);
 void drbd_nl_send_reply(struct cn_msg *, int);
diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c
index 3424d675b76..4a0f314086e 100644
--- a/drivers/block/drbd/drbd_req.c
+++ b/drivers/block/drbd/drbd_req.c
@@ -1073,7 +1073,7 @@ static int drbd_fail_request_early(struct drbd_conf *mdev, int is_write)
 	return 0;
 }
 
-int drbd_make_request(struct request_queue *q, struct bio *bio)
+void drbd_make_request(struct request_queue *q, struct bio *bio)
 {
 	unsigned int s_enr, e_enr;
 	struct drbd_conf *mdev = (struct drbd_conf *) q->queuedata;
@@ -1081,7 +1081,7 @@ int drbd_make_request(struct request_queue *q, struct bio *bio)
 
 	if (drbd_fail_request_early(mdev, bio_data_dir(bio) & WRITE)) {
 		bio_endio(bio, -EPERM);
-		return 0;
+		return;
 	}
 
 	start_time = jiffies;
@@ -1100,7 +1100,8 @@ int drbd_make_request(struct request_queue *q, struct bio *bio)
 
 	if (likely(s_enr == e_enr)) {
 		inc_ap_bio(mdev, 1);
-		return drbd_make_request_common(mdev, bio, start_time);
+		drbd_make_request_common(mdev, bio, start_time);
+		return;
 	}
 
 	/* can this bio be split generically?
@@ -1148,7 +1149,6 @@ int drbd_make_request(struct request_queue *q, struct bio *bio)
 
 		bio_pair_release(bp);
 	}
-	return 0;
 }
 
 /* This is called by bio_add_page().  With this function we reduce
diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c
index 98de8f41867..9955a53733b 100644
--- a/drivers/block/floppy.c
+++ b/drivers/block/floppy.c
@@ -4250,7 +4250,7 @@ static int __init floppy_init(void)
 	use_virtual_dma = can_use_virtual_dma & 1;
 	fdc_state[0].address = FDC1;
 	if (fdc_state[0].address == -1) {
-		del_timer(&fd_timeout);
+		del_timer_sync(&fd_timeout);
 		err = -ENODEV;
 		goto out_unreg_region;
 	}
@@ -4261,7 +4261,7 @@ static int __init floppy_init(void)
 	fdc = 0;		/* reset fdc in case of unexpected interrupt */
 	err = floppy_grab_irq_and_dma();
 	if (err) {
-		del_timer(&fd_timeout);
+		del_timer_sync(&fd_timeout);
 		err = -EBUSY;
 		goto out_unreg_region;
 	}
@@ -4318,7 +4318,7 @@ static int __init floppy_init(void)
 		user_reset_fdc(-1, FD_RESET_ALWAYS, false);
 	}
 	fdc = 0;
-	del_timer(&fd_timeout);
+	del_timer_sync(&fd_timeout);
 	current_drive = 0;
 	initialized = true;
 	if (have_no_fdc) {
@@ -4368,7 +4368,7 @@ out_unreg_blkdev:
 	unregister_blkdev(FLOPPY_MAJOR, "fd");
 out_put_disk:
 	while (dr--) {
-		del_timer(&motor_off_timer[dr]);
+		del_timer_sync(&motor_off_timer[dr]);
 		if (disks[dr]->queue)
 			blk_cleanup_queue(disks[dr]->queue);
 		put_disk(disks[dr]);
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 4720c7ade0a..68b205a9338 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -76,6 +76,8 @@
 #include <linux/splice.h>
 #include <linux/sysfs.h>
 #include <linux/miscdevice.h>
+#include <linux/falloc.h>
+
 #include <asm/uaccess.h>
 
 static DEFINE_IDR(loop_index_idr);
@@ -159,17 +161,19 @@ static struct loop_func_table *xfer_funcs[MAX_LO_CRYPT] = {
 	&xor_funcs
 };
 
-static loff_t get_loop_size(struct loop_device *lo, struct file *file)
+static loff_t get_size(loff_t offset, loff_t sizelimit, struct file *file)
 {
-	loff_t size, offset, loopsize;
+	loff_t size, loopsize;
 
 	/* Compute loopsize in bytes */
 	size = i_size_read(file->f_mapping->host);
-	offset = lo->lo_offset;
 	loopsize = size - offset;
-	if (lo->lo_sizelimit > 0 && lo->lo_sizelimit < loopsize)
-		loopsize = lo->lo_sizelimit;
+	/* offset is beyond i_size, wierd but possible */
+	if (loopsize < 0)
+		return 0;
 
+	if (sizelimit > 0 && sizelimit < loopsize)
+		loopsize = sizelimit;
 	/*
 	 * Unfortunately, if we want to do I/O on the device,
 	 * the number of 512-byte sectors has to fit into a sector_t.
@@ -177,17 +181,25 @@ static loff_t get_loop_size(struct loop_device *lo, struct file *file)
 	return loopsize >> 9;
 }
 
+static loff_t get_loop_size(struct loop_device *lo, struct file *file)
+{
+	return get_size(lo->lo_offset, lo->lo_sizelimit, file);
+}
+
 static int
-figure_loop_size(struct loop_device *lo)
+figure_loop_size(struct loop_device *lo, loff_t offset, loff_t sizelimit)
 {
-	loff_t size = get_loop_size(lo, lo->lo_backing_file);
+	loff_t size = get_size(offset, sizelimit, lo->lo_backing_file);
 	sector_t x = (sector_t)size;
 
 	if (unlikely((loff_t)x != size))
 		return -EFBIG;
-
+	if (lo->lo_offset != offset)
+		lo->lo_offset = offset;
+	if (lo->lo_sizelimit != sizelimit)
+		lo->lo_sizelimit = sizelimit;
 	set_capacity(lo->lo_disk, x);
-	return 0;					
+	return 0;
 }
 
 static inline int
@@ -203,74 +215,6 @@ lo_do_transfer(struct loop_device *lo, int cmd,
 }
 
 /**
- * do_lo_send_aops - helper for writing data to a loop device
- *
- * This is the fast version for backing filesystems which implement the address
- * space operations write_begin and write_end.
- */
-static int do_lo_send_aops(struct loop_device *lo, struct bio_vec *bvec,
-		loff_t pos, struct page *unused)
-{
-	struct file *file = lo->lo_backing_file; /* kudos to NFsckingS */
-	struct address_space *mapping = file->f_mapping;
-	pgoff_t index;
-	unsigned offset, bv_offs;
-	int len, ret;
-
-	mutex_lock(&mapping->host->i_mutex);
-	index = pos >> PAGE_CACHE_SHIFT;
-	offset = pos & ((pgoff_t)PAGE_CACHE_SIZE - 1);
-	bv_offs = bvec->bv_offset;
-	len = bvec->bv_len;
-	while (len > 0) {
-		sector_t IV;
-		unsigned size, copied;
-		int transfer_result;
-		struct page *page;
-		void *fsdata;
-
-		IV = ((sector_t)index << (PAGE_CACHE_SHIFT - 9))+(offset >> 9);
-		size = PAGE_CACHE_SIZE - offset;
-		if (size > len)
-			size = len;
-
-		ret = pagecache_write_begin(file, mapping, pos, size, 0,
-							&page, &fsdata);
-		if (ret)
-			goto fail;
-
-		file_update_time(file);
-
-		transfer_result = lo_do_transfer(lo, WRITE, page, offset,
-				bvec->bv_page, bv_offs, size, IV);
-		copied = size;
-		if (unlikely(transfer_result))
-			copied = 0;
-
-		ret = pagecache_write_end(file, mapping, pos, size, copied,
-							page, fsdata);
-		if (ret < 0 || ret != copied)
-			goto fail;
-
-		if (unlikely(transfer_result))
-			goto fail;
-
-		bv_offs += copied;
-		len -= copied;
-		offset = 0;
-		index++;
-		pos += copied;
-	}
-	ret = 0;
-out:
-	mutex_unlock(&mapping->host->i_mutex);
-	return ret;
-fail:
-	ret = -1;
-	goto out;
-}
-
-/**
  * __do_lo_send_write - helper for writing data to a loop device
  *
  * This helper just factors out common code between do_lo_send_direct_write()
@@ -297,10 +241,8 @@ static int __do_lo_send_write(struct file *file,
 /**
  * do_lo_send_direct_write - helper for writing data to a loop device
  *
- * This is the fast, non-transforming version for backing filesystems which do
- * not implement the address space operations write_begin and write_end.
- * It uses the write file operation which should be present on all writeable
- * filesystems.
+ * This is the fast, non-transforming version that does not need double
+ * buffering.
  */
 static int do_lo_send_direct_write(struct loop_device *lo,
 		struct bio_vec *bvec, loff_t pos, struct page *page)
@@ -316,15 +258,9 @@ static int do_lo_send_direct_write(struct loop_device *lo,
 /**
  * do_lo_send_write - helper for writing data to a loop device
  *
- * This is the slow, transforming version for filesystems which do not
- * implement the address space operations write_begin and write_end.  It
- * uses the write file operation which should be present on all writeable
- * filesystems.
- *
- * Using fops->write is slower than using aops->{prepare,commit}_write in the
- * transforming case because we need to double buffer the data as we cannot do
- * the transformations in place as we do not have direct access to the
- * destination pages of the backing file.
+ * This is the slow, transforming version that needs to double buffer the
+ * data as it cannot do the transformations in place without having direct
+ * access to the destination pages of the backing file.
  */
 static int do_lo_send_write(struct loop_device *lo, struct bio_vec *bvec,
 		loff_t pos, struct page *page)
@@ -350,17 +286,16 @@ static int lo_send(struct loop_device *lo, struct bio *bio, loff_t pos)
 	struct page *page = NULL;
 	int i, ret = 0;
 
-	do_lo_send = do_lo_send_aops;
-	if (!(lo->lo_flags & LO_FLAGS_USE_AOPS)) {
+	if (lo->transfer != transfer_none) {
+		page = alloc_page(GFP_NOIO | __GFP_HIGHMEM);
+		if (unlikely(!page))
+			goto fail;
+		kmap(page);
+		do_lo_send = do_lo_send_write;
+	} else {
 		do_lo_send = do_lo_send_direct_write;
-		if (lo->transfer != transfer_none) {
-			page = alloc_page(GFP_NOIO | __GFP_HIGHMEM);
-			if (unlikely(!page))
-				goto fail;
-			kmap(page);
-			do_lo_send = do_lo_send_write;
-		}
 	}
+
 	bio_for_each_segment(bvec, bio, i) {
 		ret = do_lo_send(lo, bvec, pos, page);
 		if (ret < 0)
@@ -447,7 +382,8 @@ do_lo_receive(struct loop_device *lo,
 
 	if (retval < 0)
 		return retval;
-
+	if (retval != bvec->bv_len)
+		return -EIO;
 	return 0;
 }
 
@@ -484,6 +420,29 @@ static int do_bio_filebacked(struct loop_device *lo, struct bio *bio)
 			}
 		}
 
+		/*
+		 * We use punch hole to reclaim the free space used by the
+		 * image a.k.a. discard. However we do support discard if
+		 * encryption is enabled, because it may give an attacker
+		 * useful information.
+		 */
+		if (bio->bi_rw & REQ_DISCARD) {
+			struct file *file = lo->lo_backing_file;
+			int mode = FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE;
+
+			if ((!file->f_op->fallocate) ||
+			    lo->lo_encrypt_key_size) {
+				ret = -EOPNOTSUPP;
+				goto out;
+			}
+			ret = file->f_op->fallocate(file, mode, pos,
+						    bio->bi_size);
+			if (unlikely(ret && ret != -EINVAL &&
+				     ret != -EOPNOTSUPP))
+				ret = -EIO;
+			goto out;
+		}
+
 		ret = lo_send(lo, bio, pos);
 
 		if ((bio->bi_rw & REQ_FUA) && !ret) {
@@ -514,7 +473,7 @@ static struct bio *loop_get_bio(struct loop_device *lo)
 	return bio_list_pop(&lo->lo_bio_list);
 }
 
-static int loop_make_request(struct request_queue *q, struct bio *old_bio)
+static void loop_make_request(struct request_queue *q, struct bio *old_bio)
 {
 	struct loop_device *lo = q->queuedata;
 	int rw = bio_rw(old_bio);
@@ -532,12 +491,11 @@ static int loop_make_request(struct request_queue *q, struct bio *old_bio)
 	loop_add_bio(lo, old_bio);
 	wake_up(&lo->lo_event);
 	spin_unlock_irq(&lo->lo_lock);
-	return 0;
+	return;
 
 out:
 	spin_unlock_irq(&lo->lo_lock);
 	bio_io_error(old_bio);
-	return 0;
 }
 
 struct switch_request {
@@ -700,7 +658,7 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
 		goto out_putf;
 
 	fput(old_file);
-	if (max_part > 0)
+	if (lo->lo_flags & LO_FLAGS_PARTSCAN)
 		ioctl_by_bdev(bdev, BLKRRPART, 0);
 	return 0;
 
@@ -777,16 +735,25 @@ static ssize_t loop_attr_autoclear_show(struct loop_device *lo, char *buf)
 	return sprintf(buf, "%s\n", autoclear ? "1" : "0");
 }
 
+static ssize_t loop_attr_partscan_show(struct loop_device *lo, char *buf)
+{
+	int partscan = (lo->lo_flags & LO_FLAGS_PARTSCAN);
+
+	return sprintf(buf, "%s\n", partscan ? "1" : "0");
+}
+
 LOOP_ATTR_RO(backing_file);
 LOOP_ATTR_RO(offset);
 LOOP_ATTR_RO(sizelimit);
 LOOP_ATTR_RO(autoclear);
+LOOP_ATTR_RO(partscan);
 
 static struct attribute *loop_attrs[] = {
 	&loop_attr_backing_file.attr,
 	&loop_attr_offset.attr,
 	&loop_attr_sizelimit.attr,
 	&loop_attr_autoclear.attr,
+	&loop_attr_partscan.attr,
 	NULL,
 };
 
@@ -807,6 +774,35 @@ static void loop_sysfs_exit(struct loop_device *lo)
 			   &loop_attribute_group);
 }
 
+static void loop_config_discard(struct loop_device *lo)
+{
+	struct file *file = lo->lo_backing_file;
+	struct inode *inode = file->f_mapping->host;
+	struct request_queue *q = lo->lo_queue;
+
+	/*
+	 * We use punch hole to reclaim the free space used by the
+	 * image a.k.a. discard. However we do support discard if
+	 * encryption is enabled, because it may give an attacker
+	 * useful information.
+	 */
+	if ((!file->f_op->fallocate) ||
+	    lo->lo_encrypt_key_size) {
+		q->limits.discard_granularity = 0;
+		q->limits.discard_alignment = 0;
+		q->limits.max_discard_sectors = 0;
+		q->limits.discard_zeroes_data = 0;
+		queue_flag_clear_unlocked(QUEUE_FLAG_DISCARD, q);
+		return;
+	}
+
+	q->limits.discard_granularity = inode->i_sb->s_blocksize;
+	q->limits.discard_alignment = inode->i_sb->s_blocksize;
+	q->limits.max_discard_sectors = UINT_MAX >> 9;
+	q->limits.discard_zeroes_data = 1;
+	queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, q);
+}
+
 static int loop_set_fd(struct loop_device *lo, fmode_t mode,
 		       struct block_device *bdev, unsigned int arg)
 {
@@ -849,35 +845,23 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode,
 	mapping = file->f_mapping;
 	inode = mapping->host;
 
-	if (!(file->f_mode & FMODE_WRITE))
-		lo_flags |= LO_FLAGS_READ_ONLY;
-
 	error = -EINVAL;
-	if (S_ISREG(inode->i_mode) || S_ISBLK(inode->i_mode)) {
-		const struct address_space_operations *aops = mapping->a_ops;
-
-		if (aops->write_begin)
-			lo_flags |= LO_FLAGS_USE_AOPS;
-		if (!(lo_flags & LO_FLAGS_USE_AOPS) && !file->f_op->write)
-			lo_flags |= LO_FLAGS_READ_ONLY;
+	if (!S_ISREG(inode->i_mode) && !S_ISBLK(inode->i_mode))
+		goto out_putf;
 
-		lo_blocksize = S_ISBLK(inode->i_mode) ?
-			inode->i_bdev->bd_block_size : PAGE_SIZE;
+	if (!(file->f_mode & FMODE_WRITE) || !(mode & FMODE_WRITE) ||
+	    !file->f_op->write)
+		lo_flags |= LO_FLAGS_READ_ONLY;
 
-		error = 0;
-	} else {
-		goto out_putf;
-	}
+	lo_blocksize = S_ISBLK(inode->i_mode) ?
+		inode->i_bdev->bd_block_size : PAGE_SIZE;
 
+	error = -EFBIG;
 	size = get_loop_size(lo, file);
-
-	if ((loff_t)(sector_t)size != size) {
-		error = -EFBIG;
+	if ((loff_t)(sector_t)size != size)
 		goto out_putf;
-	}
 
-	if (!(mode & FMODE_WRITE))
-		lo_flags |= LO_FLAGS_READ_ONLY;
+	error = 0;
 
 	set_device_ro(bdev, (lo_flags & LO_FLAGS_READ_ONLY) != 0);
 
@@ -919,7 +903,9 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode,
 	}
 	lo->lo_state = Lo_bound;
 	wake_up_process(lo->lo_thread);
-	if (max_part > 0)
+	if (part_shift)
+		lo->lo_flags |= LO_FLAGS_PARTSCAN;
+	if (lo->lo_flags & LO_FLAGS_PARTSCAN)
 		ioctl_by_bdev(bdev, BLKRRPART, 0);
 	return 0;
 
@@ -980,10 +966,11 @@ loop_init_xfer(struct loop_device *lo, struct loop_func_table *xfer,
 	return err;
 }
 
-static int loop_clr_fd(struct loop_device *lo, struct block_device *bdev)
+static int loop_clr_fd(struct loop_device *lo)
 {
 	struct file *filp = lo->lo_backing_file;
 	gfp_t gfp = lo->old_gfp_mask;
+	struct block_device *bdev = lo->lo_device;
 
 	if (lo->lo_state != Lo_bound)
 		return -ENXIO;
@@ -1012,7 +999,6 @@ static int loop_clr_fd(struct loop_device *lo, struct block_device *bdev)
 	lo->lo_offset = 0;
 	lo->lo_sizelimit = 0;
 	lo->lo_encrypt_key_size = 0;
-	lo->lo_flags = 0;
 	lo->lo_thread = NULL;
 	memset(lo->lo_encrypt_key, 0, LO_KEY_SIZE);
 	memset(lo->lo_crypt_name, 0, LO_NAME_SIZE);
@@ -1030,8 +1016,11 @@ static int loop_clr_fd(struct loop_device *lo, struct block_device *bdev)
 	lo->lo_state = Lo_unbound;
 	/* This is safe: open() is still holding a reference. */
 	module_put(THIS_MODULE);
-	if (max_part > 0 && bdev)
+	if (lo->lo_flags & LO_FLAGS_PARTSCAN && bdev)
 		ioctl_by_bdev(bdev, BLKRRPART, 0);
+	lo->lo_flags = 0;
+	if (!part_shift)
+		lo->lo_disk->flags |= GENHD_FL_NO_PART_SCAN;
 	mutex_unlock(&lo->lo_ctl_mutex);
 	/*
 	 * Need not hold lo_ctl_mutex to fput backing file.
@@ -1080,11 +1069,10 @@ loop_set_status(struct loop_device *lo, const struct loop_info64 *info)
 
 	if (lo->lo_offset != info->lo_offset ||
 	    lo->lo_sizelimit != info->lo_sizelimit) {
-		lo->lo_offset = info->lo_offset;
-		lo->lo_sizelimit = info->lo_sizelimit;
-		if (figure_loop_size(lo))
+		if (figure_loop_size(lo, info->lo_offset, info->lo_sizelimit))
 			return -EFBIG;
 	}
+	loop_config_discard(lo);
 
 	memcpy(lo->lo_file_name, info->lo_file_name, LO_NAME_SIZE);
 	memcpy(lo->lo_crypt_name, info->lo_crypt_name, LO_NAME_SIZE);
@@ -1100,6 +1088,13 @@ loop_set_status(struct loop_device *lo, const struct loop_info64 *info)
 	     (info->lo_flags & LO_FLAGS_AUTOCLEAR))
 		lo->lo_flags ^= LO_FLAGS_AUTOCLEAR;
 
+	if ((info->lo_flags & LO_FLAGS_PARTSCAN) &&
+	     !(lo->lo_flags & LO_FLAGS_PARTSCAN)) {
+		lo->lo_flags |= LO_FLAGS_PARTSCAN;
+		lo->lo_disk->flags &= ~GENHD_FL_NO_PART_SCAN;
+		ioctl_by_bdev(lo->lo_device, BLKRRPART, 0);
+	}
+
 	lo->lo_encrypt_key_size = info->lo_encrypt_key_size;
 	lo->lo_init[0] = info->lo_init[0];
 	lo->lo_init[1] = info->lo_init[1];
@@ -1260,7 +1255,7 @@ static int loop_set_capacity(struct loop_device *lo, struct block_device *bdev)
 	err = -ENXIO;
 	if (unlikely(lo->lo_state != Lo_bound))
 		goto out;
-	err = figure_loop_size(lo);
+	err = figure_loop_size(lo, lo->lo_offset, lo->lo_sizelimit);
 	if (unlikely(err))
 		goto out;
 	sec = get_capacity(lo->lo_disk);
@@ -1293,18 +1288,24 @@ static int lo_ioctl(struct block_device *bdev, fmode_t mode,
 		break;
 	case LOOP_CLR_FD:
 		/* loop_clr_fd would have unlocked lo_ctl_mutex on success */
-		err = loop_clr_fd(lo, bdev);
+		err = loop_clr_fd(lo);
 		if (!err)
 			goto out_unlocked;
 		break;
 	case LOOP_SET_STATUS:
-		err = loop_set_status_old(lo, (struct loop_info __user *) arg);
+		err = -EPERM;
+		if ((mode & FMODE_WRITE) || capable(CAP_SYS_ADMIN))
+			err = loop_set_status_old(lo,
+					(struct loop_info __user *)arg);
 		break;
 	case LOOP_GET_STATUS:
 		err = loop_get_status_old(lo, (struct loop_info __user *) arg);
 		break;
 	case LOOP_SET_STATUS64:
-		err = loop_set_status64(lo, (struct loop_info64 __user *) arg);
+		err = -EPERM;
+		if ((mode & FMODE_WRITE) || capable(CAP_SYS_ADMIN))
+			err = loop_set_status64(lo,
+					(struct loop_info64 __user *) arg);
 		break;
 	case LOOP_GET_STATUS64:
 		err = loop_get_status64(lo, (struct loop_info64 __user *) arg);
@@ -1513,7 +1514,7 @@ static int lo_release(struct gendisk *disk, fmode_t mode)
 		 * In autoclear mode, stop the loop thread
 		 * and remove configuration after last close.
 		 */
-		err = loop_clr_fd(lo, NULL);
+		err = loop_clr_fd(lo);
 		if (!err)
 			goto out_unlocked;
 	} else {
@@ -1635,6 +1636,27 @@ static int loop_add(struct loop_device **l, int i)
 	if (!disk)
 		goto out_free_queue;
 
+	/*
+	 * Disable partition scanning by default. The in-kernel partition
+	 * scanning can be requested individually per-device during its
+	 * setup. Userspace can always add and remove partitions from all
+	 * devices. The needed partition minors are allocated from the
+	 * extended minor space, the main loop device numbers will continue
+	 * to match the loop minors, regardless of the number of partitions
+	 * used.
+	 *
+	 * If max_part is given, partition scanning is globally enabled for
+	 * all loop devices. The minors for the main loop devices will be
+	 * multiples of max_part.
+	 *
+	 * Note: Global-for-all-devices, set-only-at-init, read-only module
+	 * parameteters like 'max_loop' and 'max_part' make things needlessly
+	 * complicated, are too static, inflexible and may surprise
+	 * userspace tools. Parameters like this in general should be avoided.
+	 */
+	if (!part_shift)
+		disk->flags |= GENHD_FL_NO_PART_SCAN;
+	disk->flags |= GENHD_FL_EXT_DEVT;
 	mutex_init(&lo->lo_ctl_mutex);
 	lo->lo_number		= i;
 	lo->lo_thread		= NULL;
diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index f533f3375e2..c3f0ee16594 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -127,8 +127,7 @@ static void sock_shutdown(struct nbd_device *lo, int lock)
 	if (lock)
 		mutex_lock(&lo->tx_lock);
 	if (lo->sock) {
-		printk(KERN_WARNING "%s: shutting down socket\n",
-			lo->disk->disk_name);
+		dev_warn(disk_to_dev(lo->disk), "shutting down socket\n");
 		kernel_sock_shutdown(lo->sock, SHUT_RDWR);
 		lo->sock = NULL;
 	}
@@ -158,8 +157,9 @@ static int sock_xmit(struct nbd_device *lo, int send, void *buf, int size,
 	sigset_t blocked, oldset;
 
 	if (unlikely(!sock)) {
-		printk(KERN_ERR "%s: Attempted %s on closed socket in sock_xmit\n",
-		       lo->disk->disk_name, (send ? "send" : "recv"));
+		dev_err(disk_to_dev(lo->disk),
+			"Attempted %s on closed socket in sock_xmit\n",
+			(send ? "send" : "recv"));
 		return -EINVAL;
 	}
 
@@ -250,8 +250,8 @@ static int nbd_send_req(struct nbd_device *lo, struct request *req)
 	result = sock_xmit(lo, 1, &request, sizeof(request),
 			(nbd_cmd(req) == NBD_CMD_WRITE) ? MSG_MORE : 0);
 	if (result <= 0) {
-		printk(KERN_ERR "%s: Send control failed (result %d)\n",
-				lo->disk->disk_name, result);
+		dev_err(disk_to_dev(lo->disk),
+			"Send control failed (result %d)\n", result);
 		goto error_out;
 	}
 
@@ -270,8 +270,9 @@ static int nbd_send_req(struct nbd_device *lo, struct request *req)
 					lo->disk->disk_name, req, bvec->bv_len);
 			result = sock_send_bvec(lo, bvec, flags);
 			if (result <= 0) {
-				printk(KERN_ERR "%s: Send data failed (result %d)\n",
-						lo->disk->disk_name, result);
+				dev_err(disk_to_dev(lo->disk),
+					"Send data failed (result %d)\n",
+					result);
 				goto error_out;
 			}
 		}
@@ -328,14 +329,13 @@ static struct request *nbd_read_stat(struct nbd_device *lo)
 	reply.magic = 0;
 	result = sock_xmit(lo, 0, &reply, sizeof(reply), MSG_WAITALL);
 	if (result <= 0) {
-		printk(KERN_ERR "%s: Receive control failed (result %d)\n",
-				lo->disk->disk_name, result);
+		dev_err(disk_to_dev(lo->disk),
+			"Receive control failed (result %d)\n", result);
 		goto harderror;
 	}
 
 	if (ntohl(reply.magic) != NBD_REPLY_MAGIC) {
-		printk(KERN_ERR "%s: Wrong magic (0x%lx)\n",
-				lo->disk->disk_name,
+		dev_err(disk_to_dev(lo->disk), "Wrong magic (0x%lx)\n",
 				(unsigned long)ntohl(reply.magic));
 		result = -EPROTO;
 		goto harderror;
@@ -347,15 +347,15 @@ static struct request *nbd_read_stat(struct nbd_device *lo)
 		if (result != -ENOENT)
 			goto harderror;
 
-		printk(KERN_ERR "%s: Unexpected reply (%p)\n",
-				lo->disk->disk_name, reply.handle);
+		dev_err(disk_to_dev(lo->disk), "Unexpected reply (%p)\n",
+			reply.handle);
 		result = -EBADR;
 		goto harderror;
 	}
 
 	if (ntohl(reply.error)) {
-		printk(KERN_ERR "%s: Other side returned error (%d)\n",
-				lo->disk->disk_name, ntohl(reply.error));
+		dev_err(disk_to_dev(lo->disk), "Other side returned error (%d)\n",
+			ntohl(reply.error));
 		req->errors++;
 		return req;
 	}
@@ -369,8 +369,8 @@ static struct request *nbd_read_stat(struct nbd_device *lo)
 		rq_for_each_segment(bvec, req, iter) {
 			result = sock_recv_bvec(lo, bvec);
 			if (result <= 0) {
-				printk(KERN_ERR "%s: Receive data failed (result %d)\n",
-						lo->disk->disk_name, result);
+				dev_err(disk_to_dev(lo->disk), "Receive data failed (result %d)\n",
+					result);
 				req->errors++;
 				return req;
 			}
@@ -405,10 +405,10 @@ static int nbd_do_it(struct nbd_device *lo)
 
 	BUG_ON(lo->magic != LO_MAGIC);
 
-	lo->pid = current->pid;
-	ret = sysfs_create_file(&disk_to_dev(lo->disk)->kobj, &pid_attr.attr);
+	lo->pid = task_pid_nr(current);
+	ret = device_create_file(disk_to_dev(lo->disk), &pid_attr);
 	if (ret) {
-		printk(KERN_ERR "nbd: sysfs_create_file failed!");
+		dev_err(disk_to_dev(lo->disk), "device_create_file failed!\n");
 		lo->pid = 0;
 		return ret;
 	}
@@ -416,7 +416,7 @@ static int nbd_do_it(struct nbd_device *lo)
 	while ((req = nbd_read_stat(lo)) != NULL)
 		nbd_end_request(req);
 
-	sysfs_remove_file(&disk_to_dev(lo->disk)->kobj, &pid_attr.attr);
+	device_remove_file(disk_to_dev(lo->disk), &pid_attr);
 	lo->pid = 0;
 	return 0;
 }
@@ -457,8 +457,8 @@ static void nbd_handle_req(struct nbd_device *lo, struct request *req)
 	if (rq_data_dir(req) == WRITE) {
 		nbd_cmd(req) = NBD_CMD_WRITE;
 		if (lo->flags & NBD_READ_ONLY) {
-			printk(KERN_ERR "%s: Write on read-only\n",
-					lo->disk->disk_name);
+			dev_err(disk_to_dev(lo->disk),
+				"Write on read-only\n");
 			goto error_out;
 		}
 	}
@@ -468,16 +468,15 @@ static void nbd_handle_req(struct nbd_device *lo, struct request *req)
 	mutex_lock(&lo->tx_lock);
 	if (unlikely(!lo->sock)) {
 		mutex_unlock(&lo->tx_lock);
-		printk(KERN_ERR "%s: Attempted send on closed socket\n",
-		       lo->disk->disk_name);
+		dev_err(disk_to_dev(lo->disk),
+			"Attempted send on closed socket\n");
 		goto error_out;
 	}
 
 	lo->active_req = req;
 
 	if (nbd_send_req(lo, req) != 0) {
-		printk(KERN_ERR "%s: Request send failed\n",
-				lo->disk->disk_name);
+		dev_err(disk_to_dev(lo->disk), "Request send failed\n");
 		req->errors++;
 		nbd_end_request(req);
 	} else {
@@ -549,8 +548,8 @@ static void do_nbd_request(struct request_queue *q)
 		BUG_ON(lo->magic != LO_MAGIC);
 
 		if (unlikely(!lo->sock)) {
-			printk(KERN_ERR "%s: Attempted send on closed socket\n",
-				lo->disk->disk_name);
+			dev_err(disk_to_dev(lo->disk),
+				"Attempted send on closed socket\n");
 			req->errors++;
 			nbd_end_request(req);
 			spin_lock_irq(q->queue_lock);
@@ -576,7 +575,7 @@ static int __nbd_ioctl(struct block_device *bdev, struct nbd_device *lo,
 	case NBD_DISCONNECT: {
 		struct request sreq;
 
-	        printk(KERN_INFO "%s: NBD_DISCONNECT\n", lo->disk->disk_name);
+		dev_info(disk_to_dev(lo->disk), "NBD_DISCONNECT\n");
 
 		blk_rq_init(NULL, &sreq);
 		sreq.cmd_type = REQ_TYPE_SPECIAL;
@@ -674,7 +673,7 @@ static int __nbd_ioctl(struct block_device *bdev, struct nbd_device *lo,
 		file = lo->file;
 		lo->file = NULL;
 		nbd_clear_que(lo);
-		printk(KERN_WARNING "%s: queue cleared\n", lo->disk->disk_name);
+		dev_warn(disk_to_dev(lo->disk), "queue cleared\n");
 		if (file)
 			fput(file);
 		lo->bytesize = 0;
@@ -694,8 +693,8 @@ static int __nbd_ioctl(struct block_device *bdev, struct nbd_device *lo,
 		return 0;
 
 	case NBD_PRINT_DEBUG:
-		printk(KERN_INFO "%s: next = %p, prev = %p, head = %p\n",
-			bdev->bd_disk->disk_name,
+		dev_info(disk_to_dev(lo->disk),
+			"next = %p, prev = %p, head = %p\n",
 			lo->queue_head.next, lo->queue_head.prev,
 			&lo->queue_head);
 		return 0;
@@ -745,7 +744,7 @@ static int __init nbd_init(void)
 	BUILD_BUG_ON(sizeof(struct nbd_request) != 28);
 
 	if (max_part < 0) {
-		printk(KERN_CRIT "nbd: max_part must be >= 0\n");
+		printk(KERN_ERR "nbd: max_part must be >= 0\n");
 		return -EINVAL;
 	}
 
diff --git a/drivers/block/paride/pg.c b/drivers/block/paride/pg.c
index 6b9a2000d56..a79fb4f7ff6 100644
--- a/drivers/block/paride/pg.c
+++ b/drivers/block/paride/pg.c
@@ -630,6 +630,7 @@ static ssize_t pg_read(struct file *filp, char __user *buf, size_t count, loff_t
 		if (dev->status & 0x10)
 			return -ETIME;
 
+	memset(&hdr, 0, sizeof(hdr));
 	hdr.magic = PG_MAGIC;
 	hdr.dlen = dev->dlen;
 	copy = 0;
diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index e133f094ab0..a63b0a2b780 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -2444,7 +2444,7 @@ static void pkt_end_io_read_cloned(struct bio *bio, int err)
 	pkt_bio_finished(pd);
 }
 
-static int pkt_make_request(struct request_queue *q, struct bio *bio)
+static void pkt_make_request(struct request_queue *q, struct bio *bio)
 {
 	struct pktcdvd_device *pd;
 	char b[BDEVNAME_SIZE];
@@ -2473,7 +2473,7 @@ static int pkt_make_request(struct request_queue *q, struct bio *bio)
 		cloned_bio->bi_end_io = pkt_end_io_read_cloned;
 		pd->stats.secs_r += bio->bi_size >> 9;
 		pkt_queue_bio(pd, cloned_bio);
-		return 0;
+		return;
 	}
 
 	if (!test_bit(PACKET_WRITABLE, &pd->flags)) {
@@ -2509,7 +2509,7 @@ static int pkt_make_request(struct request_queue *q, struct bio *bio)
 			pkt_make_request(q, &bp->bio1);
 			pkt_make_request(q, &bp->bio2);
 			bio_pair_release(bp);
-			return 0;
+			return;
 		}
 	}
 
@@ -2533,7 +2533,7 @@ static int pkt_make_request(struct request_queue *q, struct bio *bio)
 				}
 				spin_unlock(&pkt->lock);
 				spin_unlock(&pd->cdrw.active_list_lock);
-				return 0;
+				return;
 			} else {
 				blocked_bio = 1;
 			}
@@ -2584,10 +2584,9 @@ static int pkt_make_request(struct request_queue *q, struct bio *bio)
 		 */
 		wake_up(&pd->wqueue);
 	}
-	return 0;
+	return;
 end_io:
 	bio_io_error(bio);
-	return 0;
 }
 
 
diff --git a/drivers/block/ps3disk.c b/drivers/block/ps3disk.c
index 8e1ce2e2916..da0abc1838c 100644
--- a/drivers/block/ps3disk.c
+++ b/drivers/block/ps3disk.c
@@ -21,6 +21,7 @@
 #include <linux/ata.h>
 #include <linux/blkdev.h>
 #include <linux/slab.h>
+#include <linux/module.h>
 
 #include <asm/lv1call.h>
 #include <asm/ps3stor.h>
diff --git a/drivers/block/ps3vram.c b/drivers/block/ps3vram.c
index b3bdb8af89c..f58cdcfb305 100644
--- a/drivers/block/ps3vram.c
+++ b/drivers/block/ps3vram.c
@@ -10,6 +10,7 @@
 
 #include <linux/blkdev.h>
 #include <linux/delay.h>
+#include <linux/module.h>
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
 #include <linux/slab.h>
@@ -596,7 +597,7 @@ out:
 	return next;
 }
 
-static int ps3vram_make_request(struct request_queue *q, struct bio *bio)
+static void ps3vram_make_request(struct request_queue *q, struct bio *bio)
 {
 	struct ps3_system_bus_device *dev = q->queuedata;
 	struct ps3vram_priv *priv = ps3_system_bus_get_drvdata(dev);
@@ -610,13 +611,11 @@ static int ps3vram_make_request(struct request_queue *q, struct bio *bio)
 	spin_unlock_irq(&priv->lock);
 
 	if (busy)
-		return 0;
+		return;
 
 	do {
 		bio = ps3vram_do_bio(dev, bio);
 	} while (bio);
-
-	return 0;
 }
 
 static int __devinit ps3vram_probe(struct ps3_system_bus_device *dev)
diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
index 15f65b5f3fc..65cc424359b 100644
--- a/drivers/block/rbd.c
+++ b/drivers/block/rbd.c
@@ -192,7 +192,7 @@ static ssize_t rbd_snap_add(struct device *dev,
 			    const char *buf,
 			    size_t count);
 static void __rbd_remove_snap_dev(struct rbd_device *rbd_dev,
-				  struct rbd_snap *snap);;
+				  struct rbd_snap *snap);
 
 
 static struct rbd_device *dev_to_rbd(struct device *dev)
@@ -260,7 +260,7 @@ static struct rbd_client *rbd_client_create(struct ceph_options *opt,
 	kref_init(&rbdc->kref);
 	INIT_LIST_HEAD(&rbdc->node);
 
-	rbdc->client = ceph_create_client(opt, rbdc);
+	rbdc->client = ceph_create_client(opt, rbdc, 0, 0);
 	if (IS_ERR(rbdc->client))
 		goto out_rbdc;
 	opt = NULL; /* Now rbdc->client is responsible for opt */
diff --git a/drivers/block/umem.c b/drivers/block/umem.c
index 031ca720d92..aa2712060bf 100644
--- a/drivers/block/umem.c
+++ b/drivers/block/umem.c
@@ -513,7 +513,7 @@ static void process_page(unsigned long data)
 	}
 }
 
-static int mm_make_request(struct request_queue *q, struct bio *bio)
+static void mm_make_request(struct request_queue *q, struct bio *bio)
 {
 	struct cardinfo *card = q->queuedata;
 	pr_debug("mm_make_request %llu %u\n",
@@ -525,7 +525,7 @@ static int mm_make_request(struct request_queue *q, struct bio *bio)
 	card->biotail = &bio->bi_next;
 	spin_unlock_irq(&card->lock);
 
-	return 0;
+	return;
 }
 
 static irqreturn_t mm_interrupt(int irq, void *__card)
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index 079c08808d8..4d0b70adf5f 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -3,15 +3,19 @@
 #include <linux/slab.h>
 #include <linux/blkdev.h>
 #include <linux/hdreg.h>
+#include <linux/module.h>
 #include <linux/virtio.h>
 #include <linux/virtio_blk.h>
 #include <linux/scatterlist.h>
 #include <linux/string_helpers.h>
 #include <scsi/scsi_cmnd.h>
+#include <linux/idr.h>
 
 #define PART_BITS 4
 
-static int major, index;
+static int major;
+static DEFINE_IDA(vd_index_ida);
+
 struct workqueue_struct *virtblk_wq;
 
 struct virtio_blk
@@ -35,6 +39,9 @@ struct virtio_blk
 	/* What host tells us, plus 2 for header & tailer. */
 	unsigned int sg_elems;
 
+	/* Ida index - used to track minor number allocations. */
+	int index;
+
 	/* Scatterlist: can be too big for stack. */
 	struct scatterlist sg[/*sg_elems*/];
 };
@@ -276,6 +283,11 @@ static int index_to_minor(int index)
 	return index << PART_BITS;
 }
 
+static int minor_to_index(int minor)
+{
+	return minor >> PART_BITS;
+}
+
 static ssize_t virtblk_serial_show(struct device *dev,
 				struct device_attribute *attr, char *buf)
 {
@@ -341,14 +353,17 @@ static int __devinit virtblk_probe(struct virtio_device *vdev)
 {
 	struct virtio_blk *vblk;
 	struct request_queue *q;
-	int err;
+	int err, index;
 	u64 cap;
 	u32 v, blk_size, sg_elems, opt_io_size;
 	u16 min_io_size;
 	u8 physical_block_exp, alignment_offset;
 
-	if (index_to_minor(index) >= 1 << MINORBITS)
-		return -ENOSPC;
+	err = ida_simple_get(&vd_index_ida, 0, minor_to_index(1 << MINORBITS),
+			     GFP_KERNEL);
+	if (err < 0)
+		goto out;
+	index = err;
 
 	/* We need to know how many segments before we allocate. */
 	err = virtio_config_val(vdev, VIRTIO_BLK_F_SEG_MAX,
@@ -365,7 +380,7 @@ static int __devinit virtblk_probe(struct virtio_device *vdev)
 				    sizeof(vblk->sg[0]) * sg_elems, GFP_KERNEL);
 	if (!vblk) {
 		err = -ENOMEM;
-		goto out;
+		goto out_free_index;
 	}
 
 	INIT_LIST_HEAD(&vblk->reqs);
@@ -421,7 +436,7 @@ static int __devinit virtblk_probe(struct virtio_device *vdev)
 	vblk->disk->private_data = vblk;
 	vblk->disk->fops = &virtblk_fops;
 	vblk->disk->driverfs_dev = &vdev->dev;
-	index++;
+	vblk->index = index;
 
 	/* configure queue flush support */
 	if (virtio_has_feature(vdev, VIRTIO_BLK_F_FLUSH))
@@ -516,6 +531,8 @@ out_free_vq:
 	vdev->config->del_vqs(vdev);
 out_free_vblk:
 	kfree(vblk);
+out_free_index:
+	ida_simple_remove(&vd_index_ida, index);
 out:
 	return err;
 }
@@ -523,6 +540,7 @@ out:
 static void __devexit virtblk_remove(struct virtio_device *vdev)
 {
 	struct virtio_blk *vblk = vdev->priv;
+	int index = vblk->index;
 
 	flush_work(&vblk->config_work);
 
@@ -538,6 +556,7 @@ static void __devexit virtblk_remove(struct virtio_device *vdev)
 	mempool_destroy(vblk->pool);
 	vdev->config->del_vqs(vdev);
 	kfree(vblk);
+	ida_simple_remove(&vd_index_ida, index);
 }
 
 static const struct virtio_device_id id_table[] = {
diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c
index 2330a9ad5e9..15ec4db194d 100644
--- a/drivers/block/xen-blkback/blkback.c
+++ b/drivers/block/xen-blkback/blkback.c
@@ -39,6 +39,9 @@
 #include <linux/list.h>
 #include <linux/delay.h>
 #include <linux/freezer.h>
+#include <linux/loop.h>
+#include <linux/falloc.h>
+#include <linux/fs.h>
 
 #include <xen/events.h>
 #include <xen/page.h>
@@ -258,13 +261,16 @@ irqreturn_t xen_blkif_be_int(int irq, void *dev_id)
 
 static void print_stats(struct xen_blkif *blkif)
 {
-	pr_info("xen-blkback (%s): oo %3d  |  rd %4d  |  wr %4d  |  f %4d\n",
+	pr_info("xen-blkback (%s): oo %3d  |  rd %4d  |  wr %4d  |  f %4d"
+		 "  |  ds %4d\n",
 		 current->comm, blkif->st_oo_req,
-		 blkif->st_rd_req, blkif->st_wr_req, blkif->st_f_req);
+		 blkif->st_rd_req, blkif->st_wr_req,
+		 blkif->st_f_req, blkif->st_ds_req);
 	blkif->st_print = jiffies + msecs_to_jiffies(10 * 1000);
 	blkif->st_rd_req = 0;
 	blkif->st_wr_req = 0;
 	blkif->st_oo_req = 0;
+	blkif->st_ds_req = 0;
 }
 
 int xen_blkif_schedule(void *arg)
@@ -396,7 +402,7 @@ static int xen_blkbk_map(struct blkif_request *req,
 			continue;
 
 		ret = m2p_add_override(PFN_DOWN(map[i].dev_bus_addr),
-			blkbk->pending_page(pending_req, i), false);
+			blkbk->pending_page(pending_req, i), NULL);
 		if (ret) {
 			pr_alert(DRV_PFX "Failed to install M2P override for %lx (ret: %d)\n",
 				 (unsigned long)map[i].dev_bus_addr, ret);
@@ -410,6 +416,59 @@ static int xen_blkbk_map(struct blkif_request *req,
 	return ret;
 }
 
+static void xen_blk_discard(struct xen_blkif *blkif, struct blkif_request *req)
+{
+	int err = 0;
+	int status = BLKIF_RSP_OKAY;
+	struct block_device *bdev = blkif->vbd.bdev;
+
+	if (blkif->blk_backend_type == BLKIF_BACKEND_PHY)
+		/* just forward the discard request */
+		err = blkdev_issue_discard(bdev,
+				req->u.discard.sector_number,
+				req->u.discard.nr_sectors,
+				GFP_KERNEL, 0);
+	else if (blkif->blk_backend_type == BLKIF_BACKEND_FILE) {
+		/* punch a hole in the backing file */
+		struct loop_device *lo = bdev->bd_disk->private_data;
+		struct file *file = lo->lo_backing_file;
+
+		if (file->f_op->fallocate)
+			err = file->f_op->fallocate(file,
+				FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE,
+				req->u.discard.sector_number << 9,
+				req->u.discard.nr_sectors << 9);
+		else
+			err = -EOPNOTSUPP;
+	} else
+		err = -EOPNOTSUPP;
+
+	if (err == -EOPNOTSUPP) {
+		pr_debug(DRV_PFX "discard op failed, not supported\n");
+		status = BLKIF_RSP_EOPNOTSUPP;
+	} else if (err)
+		status = BLKIF_RSP_ERROR;
+
+	make_response(blkif, req->id, req->operation, status);
+}
+
+static void xen_blk_drain_io(struct xen_blkif *blkif)
+{
+	atomic_set(&blkif->drain, 1);
+	do {
+		/* The initial value is one, and one refcnt taken at the
+		 * start of the xen_blkif_schedule thread. */
+		if (atomic_read(&blkif->refcnt) <= 2)
+			break;
+		wait_for_completion_interruptible_timeout(
+				&blkif->drain_complete, HZ);
+
+		if (!atomic_read(&blkif->drain))
+			break;
+	} while (!kthread_should_stop());
+	atomic_set(&blkif->drain, 0);
+}
+
 /*
  * Completion callback on the bio's. Called as bh->b_end_io()
  */
@@ -422,6 +481,11 @@ static void __end_block_io_op(struct pending_req *pending_req, int error)
 		pr_debug(DRV_PFX "flush diskcache op failed, not supported\n");
 		xen_blkbk_flush_diskcache(XBT_NIL, pending_req->blkif->be, 0);
 		pending_req->status = BLKIF_RSP_EOPNOTSUPP;
+	} else if ((pending_req->operation == BLKIF_OP_WRITE_BARRIER) &&
+		    (error == -EOPNOTSUPP)) {
+		pr_debug(DRV_PFX "write barrier op failed, not supported\n");
+		xen_blkbk_barrier(XBT_NIL, pending_req->blkif->be, 0);
+		pending_req->status = BLKIF_RSP_EOPNOTSUPP;
 	} else if (error) {
 		pr_debug(DRV_PFX "Buffer not up-to-date at end of operation,"
 			 " error=%d\n", error);
@@ -438,6 +502,10 @@ static void __end_block_io_op(struct pending_req *pending_req, int error)
 		make_response(pending_req->blkif, pending_req->id,
 			      pending_req->operation, pending_req->status);
 		xen_blkif_put(pending_req->blkif);
+		if (atomic_read(&pending_req->blkif->refcnt) <= 2) {
+			if (atomic_read(&pending_req->blkif->drain))
+				complete(&pending_req->blkif->drain_complete);
+		}
 		free_req(pending_req);
 	}
 }
@@ -532,7 +600,6 @@ do_block_io_op(struct xen_blkif *blkif)
 
 	return more_to_do;
 }
-
 /*
  * Transmutation of the 'struct blkif_request' to a proper 'struct bio'
  * and call the 'submit_bio' to pass it to the underlying storage.
@@ -549,6 +616,7 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif,
 	int i, nbio = 0;
 	int operation;
 	struct blk_plug plug;
+	bool drain = false;
 
 	switch (req->operation) {
 	case BLKIF_OP_READ:
@@ -559,11 +627,16 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif,
 		blkif->st_wr_req++;
 		operation = WRITE_ODIRECT;
 		break;
+	case BLKIF_OP_WRITE_BARRIER:
+		drain = true;
 	case BLKIF_OP_FLUSH_DISKCACHE:
 		blkif->st_f_req++;
 		operation = WRITE_FLUSH;
 		break;
-	case BLKIF_OP_WRITE_BARRIER:
+	case BLKIF_OP_DISCARD:
+		blkif->st_ds_req++;
+		operation = REQ_DISCARD;
+		break;
 	default:
 		operation = 0; /* make gcc happy */
 		goto fail_response;
@@ -572,7 +645,8 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif,
 
 	/* Check that the number of segments is sane. */
 	nseg = req->nr_segments;
-	if (unlikely(nseg == 0 && operation != WRITE_FLUSH) ||
+	if (unlikely(nseg == 0 && operation != WRITE_FLUSH &&
+				operation != REQ_DISCARD) ||
 	    unlikely(nseg > BLKIF_MAX_SEGMENTS_PER_REQUEST)) {
 		pr_debug(DRV_PFX "Bad number of segments in request (%d)\n",
 			 nseg);
@@ -621,16 +695,25 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif,
 		}
 	}
 
+	/* Wait on all outstanding I/O's and once that has been completed
+	 * issue the WRITE_FLUSH.
+	 */
+	if (drain)
+		xen_blk_drain_io(pending_req->blkif);
+
 	/*
 	 * If we have failed at this point, we need to undo the M2P override,
 	 * set gnttab_set_unmap_op on all of the grant references and perform
 	 * the hypercall to unmap the grants - that is all done in
 	 * xen_blkbk_unmap.
 	 */
-	if (xen_blkbk_map(req, pending_req, seg))
+	if (operation != REQ_DISCARD && xen_blkbk_map(req, pending_req, seg))
 		goto fail_flush;
 
-	/* This corresponding xen_blkif_put is done in __end_block_io_op */
+	/*
+	 * This corresponding xen_blkif_put is done in __end_block_io_op, or
+	 * below (in "!bio") if we are handling a BLKIF_OP_DISCARD.
+	 */
 	xen_blkif_get(blkif);
 
 	for (i = 0; i < nseg; i++) {
@@ -654,18 +737,25 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif,
 		preq.sector_number += seg[i].nsec;
 	}
 
-	/* This will be hit if the operation was a flush. */
+	/* This will be hit if the operation was a flush or discard. */
 	if (!bio) {
-		BUG_ON(operation != WRITE_FLUSH);
+		BUG_ON(operation != WRITE_FLUSH && operation != REQ_DISCARD);
 
-		bio = bio_alloc(GFP_KERNEL, 0);
-		if (unlikely(bio == NULL))
-			goto fail_put_bio;
+		if (operation == WRITE_FLUSH) {
+			bio = bio_alloc(GFP_KERNEL, 0);
+			if (unlikely(bio == NULL))
+				goto fail_put_bio;
 
-		biolist[nbio++] = bio;
-		bio->bi_bdev    = preq.bdev;
-		bio->bi_private = pending_req;
-		bio->bi_end_io  = end_block_io_op;
+			biolist[nbio++] = bio;
+			bio->bi_bdev    = preq.bdev;
+			bio->bi_private = pending_req;
+			bio->bi_end_io  = end_block_io_op;
+		} else if (operation == REQ_DISCARD) {
+			xen_blk_discard(blkif, req);
+			xen_blkif_put(blkif);
+			free_req(pending_req);
+			return 0;
+		}
 	}
 
 	/*
@@ -685,7 +775,7 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif,
 
 	if (operation == READ)
 		blkif->st_rd_sect += preq.nr_sects;
-	else if (operation == WRITE || operation == WRITE_FLUSH)
+	else if (operation & WRITE)
 		blkif->st_wr_sect += preq.nr_sects;
 
 	return 0;
@@ -765,9 +855,9 @@ static int __init xen_blkif_init(void)
 
 	mmap_pages = xen_blkif_reqs * BLKIF_MAX_SEGMENTS_PER_REQUEST;
 
-	blkbk->pending_reqs          = kmalloc(sizeof(blkbk->pending_reqs[0]) *
+	blkbk->pending_reqs          = kzalloc(sizeof(blkbk->pending_reqs[0]) *
 					xen_blkif_reqs, GFP_KERNEL);
-	blkbk->pending_grant_handles = kzalloc(sizeof(blkbk->pending_grant_handles[0]) *
+	blkbk->pending_grant_handles = kmalloc(sizeof(blkbk->pending_grant_handles[0]) *
 					mmap_pages, GFP_KERNEL);
 	blkbk->pending_pages         = kzalloc(sizeof(blkbk->pending_pages[0]) *
 					mmap_pages, GFP_KERNEL);
@@ -790,8 +880,6 @@ static int __init xen_blkif_init(void)
 	if (rc)
 		goto failed_init;
 
-	memset(blkbk->pending_reqs, 0, sizeof(blkbk->pending_reqs));
-
 	INIT_LIST_HEAD(&blkbk->pending_free);
 	spin_lock_init(&blkbk->pending_free_lock);
 	init_waitqueue_head(&blkbk->pending_free_wq);
diff --git a/drivers/block/xen-blkback/common.h b/drivers/block/xen-blkback/common.h
index 9e40b283a46..dfb1b3a43a5 100644
--- a/drivers/block/xen-blkback/common.h
+++ b/drivers/block/xen-blkback/common.h
@@ -27,7 +27,6 @@
 #ifndef __XEN_BLKIF__BACKEND__COMMON_H__
 #define __XEN_BLKIF__BACKEND__COMMON_H__
 
-#include <linux/version.h>
 #include <linux/module.h>
 #include <linux/interrupt.h>
 #include <linux/slab.h>
@@ -46,7 +45,7 @@
 
 #define DRV_PFX "xen-blkback:"
 #define DPRINTK(fmt, args...)				\
-	pr_debug(DRV_PFX "(%s:%d) " fmt ".\n",	\
+	pr_debug(DRV_PFX "(%s:%d) " fmt ".\n",		\
 		 __func__, __LINE__, ##args)
 
 
@@ -63,13 +62,26 @@ struct blkif_common_response {
 
 /* i386 protocol version */
 #pragma pack(push, 4)
+
+struct blkif_x86_32_request_rw {
+	blkif_sector_t sector_number;/* start sector idx on disk (r/w only)  */
+	struct blkif_request_segment seg[BLKIF_MAX_SEGMENTS_PER_REQUEST];
+};
+
+struct blkif_x86_32_request_discard {
+	blkif_sector_t sector_number;/* start sector idx on disk (r/w only)  */
+	uint64_t nr_sectors;
+};
+
 struct blkif_x86_32_request {
 	uint8_t        operation;    /* BLKIF_OP_???                         */
 	uint8_t        nr_segments;  /* number of segments                   */
 	blkif_vdev_t   handle;       /* only for read/write requests         */
 	uint64_t       id;           /* private guest value, echoed in resp  */
-	blkif_sector_t sector_number;/* start sector idx on disk (r/w only)  */
-	struct blkif_request_segment seg[BLKIF_MAX_SEGMENTS_PER_REQUEST];
+	union {
+		struct blkif_x86_32_request_rw rw;
+		struct blkif_x86_32_request_discard discard;
+	} u;
 };
 struct blkif_x86_32_response {
 	uint64_t        id;              /* copied from request */
@@ -79,13 +91,26 @@ struct blkif_x86_32_response {
 #pragma pack(pop)
 
 /* x86_64 protocol version */
+
+struct blkif_x86_64_request_rw {
+	blkif_sector_t sector_number;/* start sector idx on disk (r/w only)  */
+	struct blkif_request_segment seg[BLKIF_MAX_SEGMENTS_PER_REQUEST];
+};
+
+struct blkif_x86_64_request_discard {
+	blkif_sector_t sector_number;/* start sector idx on disk (r/w only)  */
+	uint64_t nr_sectors;
+};
+
 struct blkif_x86_64_request {
 	uint8_t        operation;    /* BLKIF_OP_???                         */
 	uint8_t        nr_segments;  /* number of segments                   */
 	blkif_vdev_t   handle;       /* only for read/write requests         */
 	uint64_t       __attribute__((__aligned__(8))) id;
-	blkif_sector_t sector_number;/* start sector idx on disk (r/w only)  */
-	struct blkif_request_segment seg[BLKIF_MAX_SEGMENTS_PER_REQUEST];
+	union {
+		struct blkif_x86_64_request_rw rw;
+		struct blkif_x86_64_request_discard discard;
+	} u;
 };
 struct blkif_x86_64_response {
 	uint64_t       __attribute__((__aligned__(8))) id;
@@ -113,6 +138,11 @@ enum blkif_protocol {
 	BLKIF_PROTOCOL_X86_64 = 3,
 };
 
+enum blkif_backend_type {
+	BLKIF_BACKEND_PHY  = 1,
+	BLKIF_BACKEND_FILE = 2,
+};
+
 struct xen_vbd {
 	/* What the domain refers to this vbd as. */
 	blkif_vdev_t		handle;
@@ -138,8 +168,9 @@ struct xen_blkif {
 	unsigned int		irq;
 	/* Comms information. */
 	enum blkif_protocol	blk_protocol;
+	enum blkif_backend_type blk_backend_type;
 	union blkif_back_rings	blk_rings;
-	struct vm_struct	*blk_ring_area;
+	void			*blk_ring;
 	/* The VBD attached to this interface. */
 	struct xen_vbd		vbd;
 	/* Back pointer to the backend_info. */
@@ -149,6 +180,9 @@ struct xen_blkif {
 	atomic_t		refcnt;
 
 	wait_queue_head_t	wq;
+	/* for barrier (drain) requests */
+	struct completion	drain_complete;
+	atomic_t		drain;
 	/* One thread per one blkif. */
 	struct task_struct	*xenblkd;
 	unsigned int		waiting_reqs;
@@ -159,13 +193,11 @@ struct xen_blkif {
 	int			st_wr_req;
 	int			st_oo_req;
 	int			st_f_req;
+	int			st_ds_req;
 	int			st_rd_sect;
 	int			st_wr_sect;
 
 	wait_queue_head_t	waiting_to_free;
-
-	grant_handle_t		shmem_handle;
-	grant_ref_t		shmem_ref;
 };
 
 
@@ -182,7 +214,7 @@ struct xen_blkif {
 
 struct phys_req {
 	unsigned short		dev;
-	unsigned short		nr_sects;
+	blkif_sector_t		nr_sects;
 	struct block_device	*bdev;
 	blkif_sector_t		sector_number;
 };
@@ -196,6 +228,8 @@ int xen_blkif_schedule(void *arg);
 int xen_blkbk_flush_diskcache(struct xenbus_transaction xbt,
 			      struct backend_info *be, int state);
 
+int xen_blkbk_barrier(struct xenbus_transaction xbt,
+		      struct backend_info *be, int state);
 struct xenbus_device *xen_blkbk_xenbus(struct backend_info *be);
 
 static inline void blkif_get_x86_32_req(struct blkif_request *dst,
@@ -206,12 +240,25 @@ static inline void blkif_get_x86_32_req(struct blkif_request *dst,
 	dst->nr_segments = src->nr_segments;
 	dst->handle = src->handle;
 	dst->id = src->id;
-	dst->u.rw.sector_number = src->sector_number;
-	barrier();
-	if (n > dst->nr_segments)
-		n = dst->nr_segments;
-	for (i = 0; i < n; i++)
-		dst->u.rw.seg[i] = src->seg[i];
+	switch (src->operation) {
+	case BLKIF_OP_READ:
+	case BLKIF_OP_WRITE:
+	case BLKIF_OP_WRITE_BARRIER:
+	case BLKIF_OP_FLUSH_DISKCACHE:
+		dst->u.rw.sector_number = src->u.rw.sector_number;
+		barrier();
+		if (n > dst->nr_segments)
+			n = dst->nr_segments;
+		for (i = 0; i < n; i++)
+			dst->u.rw.seg[i] = src->u.rw.seg[i];
+		break;
+	case BLKIF_OP_DISCARD:
+		dst->u.discard.sector_number = src->u.discard.sector_number;
+		dst->u.discard.nr_sectors = src->u.discard.nr_sectors;
+		break;
+	default:
+		break;
+	}
 }
 
 static inline void blkif_get_x86_64_req(struct blkif_request *dst,
@@ -222,12 +269,25 @@ static inline void blkif_get_x86_64_req(struct blkif_request *dst,
 	dst->nr_segments = src->nr_segments;
 	dst->handle = src->handle;
 	dst->id = src->id;
-	dst->u.rw.sector_number = src->sector_number;
-	barrier();
-	if (n > dst->nr_segments)
-		n = dst->nr_segments;
-	for (i = 0; i < n; i++)
-		dst->u.rw.seg[i] = src->seg[i];
+	switch (src->operation) {
+	case BLKIF_OP_READ:
+	case BLKIF_OP_WRITE:
+	case BLKIF_OP_WRITE_BARRIER:
+	case BLKIF_OP_FLUSH_DISKCACHE:
+		dst->u.rw.sector_number = src->u.rw.sector_number;
+		barrier();
+		if (n > dst->nr_segments)
+			n = dst->nr_segments;
+		for (i = 0; i < n; i++)
+			dst->u.rw.seg[i] = src->u.rw.seg[i];
+		break;
+	case BLKIF_OP_DISCARD:
+		dst->u.discard.sector_number = src->u.discard.sector_number;
+		dst->u.discard.nr_sectors = src->u.discard.nr_sectors;
+		break;
+	default:
+		break;
+	}
 }
 
 #endif /* __XEN_BLKIF__BACKEND__COMMON_H__ */
diff --git a/drivers/block/xen-blkback/xenbus.c b/drivers/block/xen-blkback/xenbus.c
index 3f129b45451..f759ad4584c 100644
--- a/drivers/block/xen-blkback/xenbus.c
+++ b/drivers/block/xen-blkback/xenbus.c
@@ -114,44 +114,14 @@ static struct xen_blkif *xen_blkif_alloc(domid_t domid)
 	spin_lock_init(&blkif->blk_ring_lock);
 	atomic_set(&blkif->refcnt, 1);
 	init_waitqueue_head(&blkif->wq);
+	init_completion(&blkif->drain_complete);
+	atomic_set(&blkif->drain, 0);
 	blkif->st_print = jiffies;
 	init_waitqueue_head(&blkif->waiting_to_free);
 
 	return blkif;
 }
 
-static int map_frontend_page(struct xen_blkif *blkif, unsigned long shared_page)
-{
-	struct gnttab_map_grant_ref op;
-
-	gnttab_set_map_op(&op, (unsigned long)blkif->blk_ring_area->addr,
-			  GNTMAP_host_map, shared_page, blkif->domid);
-
-	if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1))
-		BUG();
-
-	if (op.status) {
-		DPRINTK("Grant table operation failure !\n");
-		return op.status;
-	}
-
-	blkif->shmem_ref = shared_page;
-	blkif->shmem_handle = op.handle;
-
-	return 0;
-}
-
-static void unmap_frontend_page(struct xen_blkif *blkif)
-{
-	struct gnttab_unmap_grant_ref op;
-
-	gnttab_set_unmap_op(&op, (unsigned long)blkif->blk_ring_area->addr,
-			    GNTMAP_host_map, blkif->shmem_handle);
-
-	if (HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1))
-		BUG();
-}
-
 static int xen_blkif_map(struct xen_blkif *blkif, unsigned long shared_page,
 			 unsigned int evtchn)
 {
@@ -161,35 +131,29 @@ static int xen_blkif_map(struct xen_blkif *blkif, unsigned long shared_page,
 	if (blkif->irq)
 		return 0;
 
-	blkif->blk_ring_area = alloc_vm_area(PAGE_SIZE);
-	if (!blkif->blk_ring_area)
-		return -ENOMEM;
-
-	err = map_frontend_page(blkif, shared_page);
-	if (err) {
-		free_vm_area(blkif->blk_ring_area);
+	err = xenbus_map_ring_valloc(blkif->be->dev, shared_page, &blkif->blk_ring);
+	if (err < 0)
 		return err;
-	}
 
 	switch (blkif->blk_protocol) {
 	case BLKIF_PROTOCOL_NATIVE:
 	{
 		struct blkif_sring *sring;
-		sring = (struct blkif_sring *)blkif->blk_ring_area->addr;
+		sring = (struct blkif_sring *)blkif->blk_ring;
 		BACK_RING_INIT(&blkif->blk_rings.native, sring, PAGE_SIZE);
 		break;
 	}
 	case BLKIF_PROTOCOL_X86_32:
 	{
 		struct blkif_x86_32_sring *sring_x86_32;
-		sring_x86_32 = (struct blkif_x86_32_sring *)blkif->blk_ring_area->addr;
+		sring_x86_32 = (struct blkif_x86_32_sring *)blkif->blk_ring;
 		BACK_RING_INIT(&blkif->blk_rings.x86_32, sring_x86_32, PAGE_SIZE);
 		break;
 	}
 	case BLKIF_PROTOCOL_X86_64:
 	{
 		struct blkif_x86_64_sring *sring_x86_64;
-		sring_x86_64 = (struct blkif_x86_64_sring *)blkif->blk_ring_area->addr;
+		sring_x86_64 = (struct blkif_x86_64_sring *)blkif->blk_ring;
 		BACK_RING_INIT(&blkif->blk_rings.x86_64, sring_x86_64, PAGE_SIZE);
 		break;
 	}
@@ -201,8 +165,7 @@ static int xen_blkif_map(struct xen_blkif *blkif, unsigned long shared_page,
 						    xen_blkif_be_int, 0,
 						    "blkif-backend", blkif);
 	if (err < 0) {
-		unmap_frontend_page(blkif);
-		free_vm_area(blkif->blk_ring_area);
+		xenbus_unmap_ring_vfree(blkif->be->dev, blkif->blk_ring);
 		blkif->blk_rings.common.sring = NULL;
 		return err;
 	}
@@ -228,8 +191,7 @@ static void xen_blkif_disconnect(struct xen_blkif *blkif)
 	}
 
 	if (blkif->blk_rings.common.sring) {
-		unmap_frontend_page(blkif);
-		free_vm_area(blkif->blk_ring_area);
+		xenbus_unmap_ring_vfree(blkif->be->dev, blkif->blk_ring);
 		blkif->blk_rings.common.sring = NULL;
 	}
 }
@@ -272,6 +234,7 @@ VBD_SHOW(oo_req,  "%d\n", be->blkif->st_oo_req);
 VBD_SHOW(rd_req,  "%d\n", be->blkif->st_rd_req);
 VBD_SHOW(wr_req,  "%d\n", be->blkif->st_wr_req);
 VBD_SHOW(f_req,  "%d\n", be->blkif->st_f_req);
+VBD_SHOW(ds_req,  "%d\n", be->blkif->st_ds_req);
 VBD_SHOW(rd_sect, "%d\n", be->blkif->st_rd_sect);
 VBD_SHOW(wr_sect, "%d\n", be->blkif->st_wr_sect);
 
@@ -280,6 +243,7 @@ static struct attribute *xen_vbdstat_attrs[] = {
 	&dev_attr_rd_req.attr,
 	&dev_attr_wr_req.attr,
 	&dev_attr_f_req.attr,
+	&dev_attr_ds_req.attr,
 	&dev_attr_rd_sect.attr,
 	&dev_attr_wr_sect.attr,
 	NULL
@@ -419,6 +383,73 @@ int xen_blkbk_flush_diskcache(struct xenbus_transaction xbt,
 	return err;
 }
 
+int xen_blkbk_discard(struct xenbus_transaction xbt, struct backend_info *be)
+{
+	struct xenbus_device *dev = be->dev;
+	struct xen_blkif *blkif = be->blkif;
+	char *type;
+	int err;
+	int state = 0;
+
+	type = xenbus_read(XBT_NIL, dev->nodename, "type", NULL);
+	if (!IS_ERR(type)) {
+		if (strncmp(type, "file", 4) == 0) {
+			state = 1;
+			blkif->blk_backend_type = BLKIF_BACKEND_FILE;
+		}
+		if (strncmp(type, "phy", 3) == 0) {
+			struct block_device *bdev = be->blkif->vbd.bdev;
+			struct request_queue *q = bdev_get_queue(bdev);
+			if (blk_queue_discard(q)) {
+				err = xenbus_printf(xbt, dev->nodename,
+					"discard-granularity", "%u",
+					q->limits.discard_granularity);
+				if (err) {
+					xenbus_dev_fatal(dev, err,
+						"writing discard-granularity");
+					goto kfree;
+				}
+				err = xenbus_printf(xbt, dev->nodename,
+					"discard-alignment", "%u",
+					q->limits.discard_alignment);
+				if (err) {
+					xenbus_dev_fatal(dev, err,
+						"writing discard-alignment");
+					goto kfree;
+				}
+				state = 1;
+				blkif->blk_backend_type = BLKIF_BACKEND_PHY;
+			}
+		}
+	} else {
+		err = PTR_ERR(type);
+		xenbus_dev_fatal(dev, err, "reading type");
+		goto out;
+	}
+
+	err = xenbus_printf(xbt, dev->nodename, "feature-discard",
+			    "%d", state);
+	if (err)
+		xenbus_dev_fatal(dev, err, "writing feature-discard");
+kfree:
+	kfree(type);
+out:
+	return err;
+}
+int xen_blkbk_barrier(struct xenbus_transaction xbt,
+		      struct backend_info *be, int state)
+{
+	struct xenbus_device *dev = be->dev;
+	int err;
+
+	err = xenbus_printf(xbt, dev->nodename, "feature-barrier",
+			    "%d", state);
+	if (err)
+		xenbus_dev_fatal(dev, err, "writing feature-barrier");
+
+	return err;
+}
+
 /*
  * Entry point to this code when a new device is created.  Allocate the basic
  * structures, and watch the store waiting for the hotplug scripts to tell us
@@ -590,7 +621,7 @@ static void frontend_changed(struct xenbus_device *dev,
 
 		/*
 		 * Enforce precondition before potential leak point.
-		 * blkif_disconnect() is idempotent.
+		 * xen_blkif_disconnect() is idempotent.
 		 */
 		xen_blkif_disconnect(be->blkif);
 
@@ -601,17 +632,17 @@ static void frontend_changed(struct xenbus_device *dev,
 		break;
 
 	case XenbusStateClosing:
-		xen_blkif_disconnect(be->blkif);
 		xenbus_switch_state(dev, XenbusStateClosing);
 		break;
 
 	case XenbusStateClosed:
+		xen_blkif_disconnect(be->blkif);
 		xenbus_switch_state(dev, XenbusStateClosed);
 		if (xenbus_dev_is_online(dev))
 			break;
 		/* fall through if not online */
 	case XenbusStateUnknown:
-		/* implies blkif_disconnect() via blkback_remove() */
+		/* implies xen_blkif_disconnect() via xen_blkbk_remove() */
 		device_unregister(&dev->dev);
 		break;
 
@@ -650,6 +681,11 @@ again:
 	if (err)
 		goto abort;
 
+	err = xen_blkbk_discard(xbt, be);
+
+	/* If we can't advertise it is OK. */
+	err = xen_blkbk_barrier(xbt, be, be->blkif->vbd.flush_support);
+
 	err = xenbus_printf(xbt, dev->nodename, "sectors", "%llu",
 			    (unsigned long long)vbd_sz(&be->blkif->vbd));
 	if (err) {
diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c
index 9ea8c2576c7..7b2ec590841 100644
--- a/drivers/block/xen-blkfront.c
+++ b/drivers/block/xen-blkfront.c
@@ -98,6 +98,9 @@ struct blkfront_info
 	unsigned long shadow_free;
 	unsigned int feature_flush;
 	unsigned int flush_op;
+	unsigned int feature_discard;
+	unsigned int discard_granularity;
+	unsigned int discard_alignment;
 	int is_ready;
 };
 
@@ -302,29 +305,36 @@ static int blkif_queue_request(struct request *req)
 		ring_req->operation = info->flush_op;
 	}
 
-	ring_req->nr_segments = blk_rq_map_sg(req->q, req, info->sg);
-	BUG_ON(ring_req->nr_segments > BLKIF_MAX_SEGMENTS_PER_REQUEST);
+	if (unlikely(req->cmd_flags & REQ_DISCARD)) {
+		/* id, sector_number and handle are set above. */
+		ring_req->operation = BLKIF_OP_DISCARD;
+		ring_req->nr_segments = 0;
+		ring_req->u.discard.nr_sectors = blk_rq_sectors(req);
+	} else {
+		ring_req->nr_segments = blk_rq_map_sg(req->q, req, info->sg);
+		BUG_ON(ring_req->nr_segments > BLKIF_MAX_SEGMENTS_PER_REQUEST);
 
-	for_each_sg(info->sg, sg, ring_req->nr_segments, i) {
-		buffer_mfn = pfn_to_mfn(page_to_pfn(sg_page(sg)));
-		fsect = sg->offset >> 9;
-		lsect = fsect + (sg->length >> 9) - 1;
-		/* install a grant reference. */
-		ref = gnttab_claim_grant_reference(&gref_head);
-		BUG_ON(ref == -ENOSPC);
+		for_each_sg(info->sg, sg, ring_req->nr_segments, i) {
+			buffer_mfn = pfn_to_mfn(page_to_pfn(sg_page(sg)));
+			fsect = sg->offset >> 9;
+			lsect = fsect + (sg->length >> 9) - 1;
+			/* install a grant reference. */
+			ref = gnttab_claim_grant_reference(&gref_head);
+			BUG_ON(ref == -ENOSPC);
 
-		gnttab_grant_foreign_access_ref(
-				ref,
-				info->xbdev->otherend_id,
-				buffer_mfn,
-				rq_data_dir(req) );
-
-		info->shadow[id].frame[i] = mfn_to_pfn(buffer_mfn);
-		ring_req->u.rw.seg[i] =
-				(struct blkif_request_segment) {
-					.gref       = ref,
-					.first_sect = fsect,
-					.last_sect  = lsect };
+			gnttab_grant_foreign_access_ref(
+					ref,
+					info->xbdev->otherend_id,
+					buffer_mfn,
+					rq_data_dir(req));
+
+			info->shadow[id].frame[i] = mfn_to_pfn(buffer_mfn);
+			ring_req->u.rw.seg[i] =
+					(struct blkif_request_segment) {
+						.gref       = ref,
+						.first_sect = fsect,
+						.last_sect  = lsect };
+		}
 	}
 
 	info->ring.req_prod_pvt++;
@@ -370,7 +380,9 @@ static void do_blkif_request(struct request_queue *rq)
 
 		blk_start_request(req);
 
-		if (req->cmd_type != REQ_TYPE_FS) {
+		if ((req->cmd_type != REQ_TYPE_FS) ||
+		    ((req->cmd_flags & (REQ_FLUSH | REQ_FUA)) &&
+		    !info->flush_op)) {
 			__blk_end_request_all(req, -EIO);
 			continue;
 		}
@@ -399,6 +411,7 @@ wait:
 static int xlvbd_init_blk_queue(struct gendisk *gd, u16 sector_size)
 {
 	struct request_queue *rq;
+	struct blkfront_info *info = gd->private_data;
 
 	rq = blk_init_queue(do_blkif_request, &blkif_io_lock);
 	if (rq == NULL)
@@ -406,6 +419,13 @@ static int xlvbd_init_blk_queue(struct gendisk *gd, u16 sector_size)
 
 	queue_flag_set_unlocked(QUEUE_FLAG_VIRT, rq);
 
+	if (info->feature_discard) {
+		queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, rq);
+		blk_queue_max_discard_sectors(rq, get_capacity(gd));
+		rq->limits.discard_granularity = info->discard_granularity;
+		rq->limits.discard_alignment = info->discard_alignment;
+	}
+
 	/* Hard sector size and max sectors impersonate the equiv. hardware. */
 	blk_queue_logical_block_size(rq, sector_size);
 	blk_queue_max_hw_sectors(rq, 512);
@@ -722,6 +742,17 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id)
 
 		error = (bret->status == BLKIF_RSP_OKAY) ? 0 : -EIO;
 		switch (bret->operation) {
+		case BLKIF_OP_DISCARD:
+			if (unlikely(bret->status == BLKIF_RSP_EOPNOTSUPP)) {
+				struct request_queue *rq = info->rq;
+				printk(KERN_WARNING "blkfront: %s: discard op failed\n",
+					   info->gd->disk_name);
+				error = -EOPNOTSUPP;
+				info->feature_discard = 0;
+				queue_flag_clear(QUEUE_FLAG_DISCARD, rq);
+			}
+			__blk_end_request_all(req, error);
+			break;
 		case BLKIF_OP_FLUSH_DISKCACHE:
 		case BLKIF_OP_WRITE_BARRIER:
 			if (unlikely(bret->status == BLKIF_RSP_EOPNOTSUPP)) {
@@ -1098,6 +1129,33 @@ blkfront_closing(struct blkfront_info *info)
 	bdput(bdev);
 }
 
+static void blkfront_setup_discard(struct blkfront_info *info)
+{
+	int err;
+	char *type;
+	unsigned int discard_granularity;
+	unsigned int discard_alignment;
+
+	type = xenbus_read(XBT_NIL, info->xbdev->otherend, "type", NULL);
+	if (IS_ERR(type))
+		return;
+
+	if (strncmp(type, "phy", 3) == 0) {
+		err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
+			"discard-granularity", "%u", &discard_granularity,
+			"discard-alignment", "%u", &discard_alignment,
+			NULL);
+		if (!err) {
+			info->feature_discard = 1;
+			info->discard_granularity = discard_granularity;
+			info->discard_alignment = discard_alignment;
+		}
+	} else if (strncmp(type, "file", 4) == 0)
+		info->feature_discard = 1;
+
+	kfree(type);
+}
+
 /*
  * Invoked when the backend is finally 'ready' (and has told produced
  * the details about the physical device - #sectors, size, etc).
@@ -1108,7 +1166,7 @@ static void blkfront_connect(struct blkfront_info *info)
 	unsigned long sector_size;
 	unsigned int binfo;
 	int err;
-	int barrier, flush;
+	int barrier, flush, discard;
 
 	switch (info->connected) {
 	case BLKIF_STATE_CONNECTED:
@@ -1178,7 +1236,14 @@ static void blkfront_connect(struct blkfront_info *info)
 		info->feature_flush = REQ_FLUSH;
 		info->flush_op = BLKIF_OP_FLUSH_DISKCACHE;
 	}
-		
+
+	err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
+			    "feature-discard", "%d", &discard,
+			    NULL);
+
+	if (!err && discard)
+		blkfront_setup_discard(info);
+
 	err = xlvbd_alloc_gendisk(sectors, info, binfo, sector_size);
 	if (err) {
 		xenbus_dev_fatal(info->xbdev, err, "xlvbd_add at %s",
@@ -1385,6 +1450,8 @@ static struct xenbus_driver blkfront = {
 
 static int __init xlblk_init(void)
 {
+	int ret;
+
 	if (!xen_domain())
 		return -ENODEV;
 
@@ -1394,7 +1461,13 @@ static int __init xlblk_init(void)
 		return -ENODEV;
 	}
 
-	return xenbus_register_frontend(&blkfront);
+	ret = xenbus_register_frontend(&blkfront);
+	if (ret) {
+		unregister_blkdev(XENVBD_MAJOR, DEV_NAME);
+		return ret;
+	}
+
+	return 0;
 }
 module_init(xlblk_init);