#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>
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)
{
struct xen_blkif *blkif = arg;
struct xen_vbd *vbd = &blkif->vbd;
+ int ret;
xen_blkif_get(blkif);
blkif->waiting_reqs = 0;
smp_mb(); /* clear flag *before* checking for work */
- if (do_block_io_op(blkif))
+ ret = do_block_io_op(blkif);
+ if (ret > 0)
blkif->waiting_reqs = 1;
+ if (ret == -EACCES)
+ wait_event_interruptible(blkif->shutdown_wq,
+ kthread_should_stop());
if (log_stats && time_after(jiffies, blkif->st_print))
print_stats(blkif);
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()
*/
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);
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);
}
}
rp = blk_rings->common.sring->req_prod;
rmb(); /* Ensure we see queued requests up to 'rp'. */
+ if (RING_REQUEST_PROD_OVERFLOW(&blk_rings->common, rp)) {
+ rc = blk_rings->common.rsp_prod_pvt;
+ pr_warn(DRV_PFX "Frontend provided bogus ring requests (%d - %d = %d). Halting ring processing on dev=%04x\n",
+ rp, rc, rp - rc, blkif->vbd.pdevice);
+ return -EACCES;
+ }
while (rc != rp) {
if (RING_REQUEST_CONS_OVERFLOW(&blk_rings->common, rc))
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.
int i, nbio = 0;
int operation;
struct blk_plug plug;
+ bool drain = false;
switch (req->operation) {
case BLKIF_OP_READ:
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;
/* 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);
}
}
+ /* 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++) {
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;
+ }
}
- /*
- * We set it one so that the last submit_bio does not have to call
- * atomic_inc.
- */
atomic_set(&pending_req->pendcnt, nbio);
-
- /* Get a reference count for the disk queue and start sending I/O */
blk_start_plug(&plug);
for (i = 0; i < nbio; i++)
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;
fail_put_bio:
for (i = 0; i < nbio; i++)
bio_put(biolist[i]);
+ atomic_set(&pending_req->pendcnt, 1);
__end_block_io_op(pending_req, -EINVAL);
msleep(1); /* back off a bit */
return -EIO;
static void make_response(struct xen_blkif *blkif, u64 id,
unsigned short op, int st)
{
- struct blkif_response resp;
+ struct blkif_response *resp;
unsigned long flags;
union blkif_back_rings *blk_rings = &blkif->blk_rings;
int notify;
- resp.id = id;
- resp.operation = op;
- resp.status = st;
-
spin_lock_irqsave(&blkif->blk_ring_lock, flags);
/* Place on the response ring for the relevant domain. */
switch (blkif->blk_protocol) {
case BLKIF_PROTOCOL_NATIVE:
- memcpy(RING_GET_RESPONSE(&blk_rings->native, blk_rings->native.rsp_prod_pvt),
- &resp, sizeof(resp));
+ resp = RING_GET_RESPONSE(&blk_rings->native,
+ blk_rings->native.rsp_prod_pvt);
break;
case BLKIF_PROTOCOL_X86_32:
- memcpy(RING_GET_RESPONSE(&blk_rings->x86_32, blk_rings->x86_32.rsp_prod_pvt),
- &resp, sizeof(resp));
+ resp = RING_GET_RESPONSE(&blk_rings->x86_32,
+ blk_rings->x86_32.rsp_prod_pvt);
break;
case BLKIF_PROTOCOL_X86_64:
- memcpy(RING_GET_RESPONSE(&blk_rings->x86_64, blk_rings->x86_64.rsp_prod_pvt),
- &resp, sizeof(resp));
+ resp = RING_GET_RESPONSE(&blk_rings->x86_64,
+ blk_rings->x86_64.rsp_prod_pvt);
break;
default:
BUG();
}
+
+ resp->id = id;
+ resp->operation = op;
+ resp->status = st;
+
blk_rings->common.rsp_prod_pvt++;
RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&blk_rings->common, notify);
spin_unlock_irqrestore(&blkif->blk_ring_lock, flags);
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);
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);